United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-78-115
June 1978
Research and Development
EPA Evaluation of
Bahco Industrial
Boiler Scrubber
System at
Rickenbacker AFB
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-600/7-78-115
1S78
1©
by
E.L Biedell, R.J. Ferb, G.W. Malamud,
C.D. Ruff, and N.J. Stevens
Research-Cottrell, Inc.
P.O. Box 750
Bound Brook, New Jersey 08805
Interagency Agreement No. D5-0718
Program Element No. EHE624A
EPA Project Officer: John E. Williams
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
-------
CES 272 ABSTRACT
A comprehensive test program which characterized the particulate
removal, sulfur dioxide removal, operating costs, maintenance
costs, waste product properties operating experience and capa-
city of a size 50 R-C/Bahco scrubbing system installed to
treat flue gas from coal fired boilers at Rickenbacker AFB,
Columbus, Ohio, was completed. Tests were conducted over an
18 month period during which 27,216 tons of coal were burned.
The results from this program demonstrate that this system is
capable of controlling both particulate and sulfur dioxide
emissions from the combustion of high sulfur (2-4%) mid-
Western coal at firing rates from 20 MM to 200 MM Btu/hr.
Particulate emissions were reduced to as low as 0.15 Ibs./MM
Btu. Sulfur dioxide emissions were reduced to as low as 0.1
Ibs./MM Btu with lime and 0.6 Ibs./MM Btu with limestone.
Operating costs were $5.28 per ton of coal burned including
$0.21 maintenance costs when using lime. A cost of $4.27 per
ton for optimum operation with limestone was projected. Waste
product properties relative to dewatering handling and disposal
were found to be similar to those measured for other FGD waste
products.
Operation of the system required less than 2,000 man hours per
year. During the test program, the system experienced downtime
due to auxiliary equipment defects, an inadequate spare parts
inventory and minor system modifications. There was no process
related downtime. Based on the observations made during the
test programs, future system availability above 95% is projected,
-------
CONTENTS
Abstract 2
List of Figures 5
List of Tables 7
List of Abbreviations and Symbols 9
Executive Summary 11
1. Introduction 19
2. R-C/Bahco Scrubbing System 22
Process Description 23
Major Equipment 23
3. Test Program 33
Test Description 33
Process Variable Test Design 36
4. Operability/Material Balance Test Results .... 43
R-C Bahco Scrubber Operating Limits .... 43
Material Balance 46
5. Sulfur Dioxide Removal Tests 56
Preliminary S02 Removal Tests 56
Lime Screening Tests 56
Lime Verification Tests 64
Limestone Process Variable Tests 68
Lime vs. Limestone 73
Conclusions 76
6. Particulate Removal Tests 77
Initial Particulate Tests 77
Fractional Efficiency Tests 81
Particulate Performance Models 86
Conclusions 100
7. Scrubber Sludge Characterization Tests 101
Slurry bewatering 102
Transportability 112
Physical/Structural Properties . 113
Environmental Acceptability 114
Conclusions 116
8. Operating Experience 117
Monthly Operating Summaries 121
Evaluation of Downtime 121
Scrubber Inspections 127
Conclusions 133
-------
CONTENTS" CCONT.)
Page
9-. Operating and Maintenance Costs' ....... 136
Operating Costs: ........... 136
Maintenance Costs- ........... 139
Recommended Operating Conditions ...
Appendices ...................... 143
A. Conversion Factors: British, to ST Units . . 143
B. Analytical and Tes;ting Methods ....... 145
C. Material Balance Test Data ......... 153
D. Lime Test Data ............... 162
E. Reagent and Coal Data ............ 173
F. Lime Verification Test Data ......... 177
G. Limestone Test Data ............. 185
S. Particulate Test Data Summary ........ 195
J. R-C/Bahco Scrubber Operating Log ...... 211
K. Sulfur Dioxide Performance Test Results. . . 215
-------
LIST. OF FIGURES
Number Page
Ul The R-C/Bah.co Scrubber at RAFB ........... 20
2-1 The R-C/Bahco System Process Flow Diagram ...... 24
2-2 Lined Storage Pond at RAFB ... .......... 28
3-1 The Characterization Program Test Schedule ..... 34
4-1 Verification of Inlet and Outlet S02 Concentration
Data ....................... 50
4-2 The Relationship between Slurry Solids Concentration
and Specific Gravity ............... 51
4-3 A Comparison of Gas Flow Rate Derived from Fan
Performance Data and Pi tot Tube Measurements ... 55
5-1 The Relationships between SO2 Emission Rates and
Lime Stoichiometry ... ..... ........ 58
5-2 SO2 Removal Efficiency as a Function of Lime
Stoichiometry ................ . . gg
5-3 S02 Verification Test Results
5-4 S02 Removal Efficiency as a Function of Limestone/
SO2 Stoichiometry and Slurry Pumping Rate ..... 70
5-5 Comparison of Predicted and Observed S02 Removal
Efficiency for Limestone .............
5-6 The Effect of Lime and Limestone Stoichiometry on
Dissolver pH
6-1 The Effect of Scrubber Pressure Drop on Parti culate
Emission Rates
6-2 A Comparison of Various Particle Size Distributions
for RAFB ..................... 80
5
-------
LIST OF FIGURES (CONT)
Number Page
6-3 The Effect of Total Scrubber Pressure Drop on the
Cut Off Diameter for 90% Collection Efficiency ... 83
6-4 The Effect of Total Scrubber Pressure Drop and Gas
Rate on the Cut Off Diameter for 50% Collection
Efficiency 84
6-5 Particulate Collection Efficiency as Related to
Particle Size at RAFB 87
6-6 Examples of Reentrainment from the R-C/Bahco Scrubber
at RAFB 88
6-7 An Example of Bypassing from the R-C/Bahco Scrubber
at RAFB 90
6-8 The Effect of Venturi Pressure Drop on Liquid Pickup . 92
6-9 A Comparison of Observed and Predicted Particulate
Penetration 97
6-10 A Comparison of Observed and Predicted Particulate
Collection Efficiencies for the R-C/Bahco Scrubber
at RAFB 98
7-1 A Comparison of Lime and Limestone Slurry Settling
Rates 103
7-2 The Effect of Operating Time and Slurry Feed
Concentration on Centrifuge Cake Density 105
7-3 Filtration Rate as a Function of Form Time and
Slurry Concentration 108
7-4 A Photomicrograph of Lime Sludge Showing Calcium
Sulfite Crystals 110
7-5 A Photomicrograph of Limestone Sludge Showing Gypsum
Crystals Ill
8-1 A Summary of the Operation of the R-C/Bahco Scrubber
at RAFB from March 1976 to May 1977 118
8-2 Bahco Scrubber Module ' 128
8-3 R-C/Bahco Scrubber Internals Showing the Venturi, Pan
and Spray Manifold in the Second Stage 130
8-4 Build up on the Vanes in the Stack Area After Six
Months of Operation 135
-------
LIST OF TABLES
Number pa?e
2.1 Typical Syst.em Operating Conditions 12
3.1 Screening Test Matrix for the R-C/Bahco Test Program. 38
3.2 Target Levels for Lime Tests 39
3.3 Target Levels for Limestone Tests 40
3.4 Analytical Tests 42
4.1 R-C/Bahco Scrubber .^Operating Limits - « 44
4.2 Material Balance-Results . . v-.' . ^ .......... 48
4.3 Slurry Solids Concentration and Specific Gravity . . 52
4.4 Comparison of Gas Flow Measurement.Techniques .... 54
5.1 S02 Removal Efficiency Data 57
5.2 Lime Screening Test Variable Levels 59
5.3 SO2 Variable Screening Test Results" 61
5.4 Lime Screening Test Data Summary 62
5.5 Lime Verification Test Results. . 65
5.6 Limestone Screening Test Data Summary 69
5.7 Lime and Limestone Slurry Analyses 74
6.1 Particulate Emission Rates 73
6.2 Inlet Particle Size Distributions 7g
6".3 Fractional Efficiency Test .Results 82
-------
LIST OF TABLES (CONT)
NumBex Page.
6.4 Venturi Liquid Pickup Rate 91
6.5 Penetration Model Coefficients 96
g.5 A Comparison of One and Two Stage Models 99
7.1 Settling Test Results 104
7.2 Centrifuge Test Results 106
7.3 Filter Leaf Test Results -109
7.4 Slurry Solids Composition 109
7.5 Transportability Test Results -113
7.6 CBR Test Summary 114
7.7 Sludge Leachate Analyses 115
8.1 Down Time Related to Auxiliary Equipment 119
8.2 Down Time From Other Sources 120
9.1 Operating Cost Summary ..... .135
9.2 Equipment Power Reauirements 137
9.3 Optimum Operating Costs 139
9.4 Maintenance Labor and Material Costs 140
9.5 Summary of Recommended Operating Conditions 142
8
-------
LIST OF ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
GBR California bearing ratio
COD chemical oxygen demand
gr grains
in.
W.C. inches water column
Ib-m pound moles
LOI loss on ignition
mg milligrams
MM million
ppb parts per billion
ppm parts per million
SCFB " standard cubic foot, dry
SCFMD/ standard cubic feet per
SCFMW minute, dry/wet
S.G. specific gravity
St. stoichiometry
TDS total dissolved solids
SYMBOLS(Cont.)
SYMBOLS
Sl'a2
C1
Cin
''out
D
dp/DP
°P50
D.
P90
G
Hc
H
coefficients for
(L/G)^, (L/G) 2
Cunningham correction
factor
inlet particle
concentration
between 1st and 2nd
stages
particle concentration
leaving 2nd stage
collector droplet size
particle diameter
diameter at which
scrubber removes 50%
of particulate
diameter at which
scrubber removes 90%
of particulate
initial feed concen-
tration (% solids)
gas flow rate
height of liquid in
cylinder
initial height of
sludge layer
H
u
L1'L2
L/G
P1'P
PPMf
SGL
SG.
m
SGe
SG,
SG
u
av
UF
V C
vga
X°
APlfAP2
V
underflow sludge
layer height
slurry flow rate
first, second stage
slurry flow rates
liquid/gas ratio
total penetration
penetration through
1st, 2nd stages
flocculant concen-
tration, ppm
liquid specific
gravity
mixture specific
gravity
solids specific
gravity
feed specific
gravity
underflow sludge
specific gravity
settling time for
solids with concen-
tration UFC
average gas
temperature
underflow sludge concentrate
average gas velocity
undisturbed upstream velocity
mean concentration
first, second stage
pressure drops
overall efficiency
microns
gas viscosity
gas density
particle density
standard deviation
inertial impaction
parameter
-------
ACKNOWLEDGMENTS
The cooperation of the base civil engineering staff of
Rickenbacker AFB, Colonel Earl L. Krueger, Base Civil Engineer,
James B. Rasor, Associate Base Civil Engineer, and Herbert
Robinson, Scrubber Technician, is gratefully.acknowledged.
The U. S. Air Force made the R-C/Bahco flue gas desulfur-
ization system at RAFB available for this test program and
contributed in every possible way to the success of the program.
10
-------
EXECUTIVE SUMMARY
INTRODUCTION
This report describes the results of a program sponsored by
the Environmental Protection Agency and performed by Research-
Cottrell (R-C) under a contract with the United States Air Force
(USAF) at the R-C/Bahco sulfur dioxide and particulate scrubbing
system installed at Rickenbacker Air Force Base (RAFB) in Columbus,
Ohio. The program was conducted to characterize the performance
of an R-C/Bahco scrubber handling flue gas from the combustion of
midwestern high sulfur coal.
The R-C/Bahco system was designed, erected and installed by
R-C under a contract funded by the USAF as part of their program
to demonstrate the viability of air pollution control technology.
The system installed at RAFB in the first application of this
technology in the U.S. and its first application anywhere on a
coal fired boiler. The results of the R-C/Bahco characterization
program are presented in this report.
The scrubbing facility at RAFB, which consists of a size 50
R-C/Bahco scrubber and auxiliary equipment, was designed to treat
108,000 ACFM1 of flue gas at 475 F containing 1500 to 2000 ppm of
sulfur dioxide and 0.2 to 2.0 grains per SCF of particulates.
The R-C/Bahco system is a two stage venturi scrubber manu-
factured by Research-Cottrell under license from A. B. Bahco of
Sweden. The reagent storage and feed system was designed to
handle both lime and limestone. The system was designed to handle
up to 108,000 ACFM of flue gas at 475°F, which is equilavent to a
coal firing rate of 200,000,000 Btu/hr, as well as summer load
conditions when the coal firing rate is under 20,000,000 Btu/hr.
The system was guaranteed to meet Ohio emissions standards in
force in 1974. These standards required an SO- removal of 83% at
maximum load for 3.5% sulfur coal, i.e. a maximum emission of 1.0
Ibs. of SO2 per million Btu of coal fired. The standards also
limited particulate emissions to 0.16 Ibs. per million Btu at
max imum 1oad.
Maximum requirements for power (600 KW average), water {45
gpm), lime (2010 tons per year a 1.1 lime-S02 stoichlometry), and
operating labor (2 man hours per shift plus supervison) were
guaranteed by R-C.
(1) Although it is the policy of the EPA to use the S.I. system for
quantitative descriptions, the British System is used in this
report. Readers who are more accustomed to S.I. units are re-
ferred to the conversion table in Appendix A.
11
-------
Typical scrubber operating conditions are listed in Table
2.1.
TABLE 2.1 TYPICAL SCRUBBER OPERATING CONDITIONS
Flue Gas 64,000 ACFM @380°F
SO- Concentration 1390 ppm
1st Stage AP 10. in. W.C.
2nd Stage AP 8 in. W.C.
Lime-S02 Stoichiometry 0.876
S02 Removal 87.6% (0.615 #/MM Btu outlet)
Lime Utilization 100%
Particulate Emissions 0.16 #/MM Btu
There were no significant problems in the scrubbing equipment
related to scale build-up or fouling. Minor build-ups were
observed in the gas inlet area and at the base of the stack.
These deposits posed no threat to long term operation.
The overall cost of power, reagents and operating manpower
was $5.07 per ton of coal. This figure was below the estimated
cost of $5.92 per ton and substantially less than the guaranteed
cost of $7.56 per ton.
THE R-C/BAHCO SYSTEM TEST PROGRAM
The major goals of this program were:
o To demonstrate operability of the R-C/Bahco system at
RAFB.
o To study system variables to determine optimum and
limiting conditions of the system for sulfur dioxide
and particulate removal.
o To evaluate and monitor the system over an extended
period to obtain maintenance and operating cost data.
Information regarding system operability included the deter-
mination of desirable operating conditions and the measurement of
system performance in terms of S02 and particulate removal.
In addition, studies including complete material balances, system
stability, and operating requirements were completed. SO2 and
particulate removal efficiencies and the effects of several
important system variables including reagent type were determined.
A comprehensive evaluation of waste product or "sludge" properties
which included dewatering, transport, and disposal characteristics
was completed. A complete evaluation of system reliability
including an analysis of all downtime was performed. In addition,
detailed cost data was obtained for utilities, reagents, operating
and maintenance labor and waste disposal.
12
-------
Preliminary Test Program
The preliminary test program which began shortly after
start-up was undertaken to determine the capacity limits of the
system, to run preliminary SO- and particulate removal tests and
to check data acquisition and analytical techniques.
These tests indicated that the gas handling capacity of the
scrubber was substantially higher than the design capacity of
55,000 SCFM. Gas flows over 60,000 SCFM were possible without
exceeding the capacity of the final mist eliminator. The system
operated at pressure drops as low as 12 in. W. C. for both stages
to over 30 in. W. C. The pumping rate to the scrubber had to be
limited to 3000 gpm, i.e., 15% above the design flow of 2600 gpm
to avoid flooding the flue gas inlet manifold when operating at
low load conditions. The system operated satisfactorily at
slurry solids from 2% to 25% by weight. Particulate emissions
were above the required level of 0.16 #/mm Btu when substantial
quantities of soot were being formed in the heat plant. Sulfur
dioxide removal-efficiency was well above the required 83%;
levels in the 95% plus range were readily attainable. Data
acquisition and analytical techniques were verified during the
completion of two overall material balances.
Lime Tests
The results of the lime tests indicate that SO2 removal is
controlled only by the lime-sulfur dioxide stoichiometry. Other
operating variables such as gas flow, pressure drops, liquid rates
and solids concentration had essentially no effect on SO- removal
efficiency. The lime tests covered a wide range of flue gas flow
rates, 30,000 to 55,000 SCFM, slurry solids concentrations, 2.0 to
15.0%, stoichiometries, 0.6 to 1.08 moles of available CaO per mole
of S02 and inlet SO2 concentrations of 400 to 1800 ppm.
Virtually any removal efficiency up to 98 + % can be achieved
by adjusting the lime-sulfur dioxide stoichiometry. Virtually no
excess lime is required for S02 removal efficiencies up to 90%,
i.e. S02 removal is equal to lime-S02 stoichiometry up to this'
point and with an excess of approximately 10%, 98 + % removal
can be achieved.
The fact that SO2 removal when using lime is not adversely
affected by changes in operating variables illustrates the fact
that good gas liquid contact is maintained over the entire opera-
ting range of the system. This also demonstrates that precise
control of these variables is not critical to successful operation
13
-------
Limestone Tests
The limestone tests were essentially a duplication of the
lime tests. The only major differences were the range of
stoichiometries (0.6 to 1.6 moles of available CaCO^ or MgCCU
per mole of SO-) and inlet S02 levels (250 to 600 ppm) investi-
gated. The relatively low S02 levels resulted from the low
load on the heat plant during the limestone test program.
The results of these tests indicate that slurry circula-
tion rate, in addition to stoichiometry, is important in deter-
ming the SO2 efficiency. The following performance model based
on limestone stoichimetry and slurry circulation rate was
developed to predict SO2 removal efficiency.
% S02 removal = (St)0'52 x(L)°'55 eq. (5.1)
where: St = limestone-inlet S02 stoichiometry
L = second stage slurry circulation rate
The results of these tests indicate that SCU removal with
limestone above 90% in the R-C/Bahco system is not practical
since limestone utilization drops to approximately 60%. This
results from the high stoichiometry required to exceed this
level of SO2 removal. However, most industrial boiler appli-
cations require SO2 removal well below the 90% level. For
such requirements the use of .limestone will accomplish adequate
SO2 removal at a reagent cost substantially below that for
lime.
Particulate Tests
Particulate tests were run using both Andersen Impactors
and ASME thimbles simultaneously at the inlet and outlet of the
scrubber. Collection efficiency as a function of particle size
was determined over a wide range of flue gas flow rates and
venturi pressure drops. Collection efficiency in the 0.3 to
0.5 micron range was 15.2 to 66.7%, in the 0.5 to 1.0 micron
range 27.1 to 86.7%, in the 1.0 to 2.0 micron range 85.3 to
98.9% and in the 2.0 to 5.0 micron range 95.0 to 99.9%. There
were substantial variations in observed particulate collection
efficiences under similar operation conditions. These varia-
tions were caused by fluctuations in the operation of the heat
plant which resulted in significant emissions of soot at vari-
ous times during the particulate test program.
Particulate performance models based on inertial impaction
and penetration were developed using these results.
14
-------
Filtration Rate Tests
Filtration rates varied from approximately 40 to 200 Ibs/hr
ft . Filtration rates for limestone sludges were generally lower
than those observed for lime. Lime sludge filter cakes contained
approximately 57 to 58% solids and limestone cakes 72 to 76%
solids.
Centrifuge Tests
In the centrifugation tests, limestone sludge dewatered to
approximately 65 to 67% solids at conditions which produced a
lime sludge of 55 to 57%.
Physical/Structural Property Tests
The physical/structural property tests indicated that both
sludges when dewatered exhibit similar cohesive and adhesive pro-
perties making them somewhat difficult to handle. Limestone
sludge samples exhibited higher bearing capacities and higher
confined compressive strengths than lime sludge samples.
Leachate Tests
Leachate from both lime and limestone sludges were very
similar and were characterized by TDS of 2500 to 3000 ppm, COD
values of 6 - 8 mg/1, and contained heavy metals from 5 to 50
ppb.
Overall Systems Monitoring
The monitoring program provided detailed information on
operating costs, maintenance costs and operating experience.
This information coupled with the results of the test program
aided in the determination of optimum operating conditions for
the RAFB system.
Operating and Maintenance Costs
During the test period, the gas from the combustion of
27,216 tons of 2.5% sulfur (av.) coal was treated. The total
cost for utilities, reagent, supplies and operating labor was
$5.07 per ton of coal burned. These costs were projected to be
$5.92 per ton of coal burned and a cost ceiling of $7.56 per ton,
based on current reagent and power cost, was guaranteed in the
contract. Maintenance, labor and materials added $0 21 to the
cost for a total of $5.28 per ton of coal burned. An operating
cost of $4.06 per ton can be achieved if limestone is used as
the scrubbing reagent and fan settings and makeup water consump-
tion are optimized. ^
15
-------
The first stage venturi average gas temperature (Tav) and slurry
flow rate (L,) and the second stage slurry flow rate (L2) were
related to tne overall penetration for a given size range of
particles (d ) by the following equation:
PQ = exp(a-LL1d Tav) exp (a2L2d ) eq. (6-10)
where a^ and a2 are the model's correlation coefficients for the
first and second stages respectively.
Analysis of the test data indicated that collection effi-
ciency from 0.3 to 1.0 microns was controlled primarily by condi-
tions in the upper (second stage) venturi and collection efficiency
above 1 micron was primarily controlled by conditions in the
lower (first stage) venturi.
Slurry Entrainment and Gas Bypassing
During the particulate tests, two phenomena were observed
when the system was operated near its capacity limits. The
first, called entrainment, which occurs at very low venturi
pressure drops, i.e. under 6 in. W.C., involves small droplets of
slurry carrying through the second stage mist eliminator and out
the stack as indicated by the collection efficiency data and coated
sampling probes. The second, called bypassing, is characterized
by pulsations in the gas flow through the scrubber and results in
low collection efficiency in all particle size ranges. This
phenomenon occurs when relatively high pressure drops, i.e. 12 +
in. W.C. in either venturi coupled with a slurry flow rate in the
scrubber which is less than the minimum required for liquid
pickup.
Sludge Characterization Tests
The characterization tests produced data on dewatering
rates, physical/structural properties and leachates from lime and
limestone based sludges from the R-C/Bahco system at RAFB. A
complete description of the tests and results are presented in
the report.
Settling Rate Tests
Settling rate tests were conducted on both lime and limestone
sludge with and without flocculent. Limestone sludge settled
much more rapidly than lime sludge and flocculent addition improved
limestone settling rates appreciably. Limestone settling rates
were 20-22 Ibs/ft^/day and those for limestone ranged from 164
Ibs/ftVday without flocculent and 578 Ibs/ft2/day with floc-
culent.
16
-------
Operating Experience
As indicated in this report, the system performs satisfac-
torily in six important areas:
o S02 removal efficiency
o Particulate removal efficiency
o Scrubber reliability
o Ease of operation
o Minimal routine maintenance requirements
o Moderate operating costs
However, the system suffered from a considerable amount of
downtime. During the test period which included 11,024 hours,
there were 4,830 hours of downtime. Most of this downtime,
2,766 hours, resulted from booster fan problems which have been
rectified. An additional 1,088 hours were lost due to other
auxiliary equipment problems. However, of the total of 3,854
hours lost due to equipment problems, 1,035 hours were due to
delays resulting from a lack of spare parts. Heat plant outages
and minor system modifications resulted in an additional 507
hours of downtime. The table below summarizes the downtime
during the test period.
SUMMARY OF TABLES 8.1 AND 8.2
DOWNTIME DURING THE TEST PERIOD
Category Downtime Hrs.
Fan 2,766
Auxiliary Equipment 1,088
Heat Plant Outages 388
Modifications 119
Other Items 469
4,830
The other items which constituted less than 5% of the total
period included routine maintenance and freezeup problems which
occurred in water and instrument air lines.
17
-------
Now that the major problems with auxiliary equipment have
been cleared up and the freezeup problems have been eliminated,
the system should operate 95% or more of the time as it did
during the 1976-77 winter including December, January and February.
Optimium Operating Conditions
At the present time local air pollution control requirements
limit SO2 and particulate emissions to 2.2 and 0.16 Ibs/MM Btu,
respectively. These requirements can be met most economically by
the R-C/Bahco system by operating at the conditions listed in
Table 9.5 below:
TABLE 9.5 OPTIMUM OPERATING CONDITIONS
Lime Limestone
Stoichiometry 0.70 0.75
Second Stage Slurry
Rate 2,200 to 2,400 gpm
Flue gas rates 35,000 to 55,000 SCFM
First and Second Stage
Pressure Drops 7 to 10 in. W.C.
Present emissions standards will be met at the above condi-
tions. In addition, reagent consumption and power requirements
will be minimized with a minimum of operator attention.
18
-------
SECTION 1
INTRODUCTION
In September 1974, the United States Air Force took a major
step in demonstrating the applicability of flue gas scrubbing
technology to industrial coal-fired plants. A contract was
awarded to Research-Cottrell for an SO- and particulate emission
control system for the Central Heat Plant at Rickenbacker Air
Force Base (RAFB) near Columbus, Ohio. An R-C/Bahco scrubber
was selected for this project. (See Figure 1-1) The scrubber,
which accomplishes both SO- and particulate removal, is based on
technology developed by AB Bahco in Sweden and tested world-wide
on oil-fired boilers, incinerators and other applications.
In April 1975, when engineering of the system was in its
final stages, a second contract sponsored by the USEPA was
awarded to Research-Cottrell. The overall objective of this
program (Contract No. F33617-75-90100) was to characterize the
R-C/Bahco scrubbing system installed at RAFB in terms of its
performance, reliability, and economics for SO- and particulate
control on small coal-fired plants.
The R-C/Bahco system was put into operation in March, 1976,
and the characterization program was initiated in April of that
year. The first phase involved preliminary testing of the
system's operating limits and performance capabilities. The
next phase consisted of the following screening and process
variable test programs: lime scrubbing of SO2, limestone
scrubbing of SO2, and particulate removal.
These tests were designed to determine the effect of
process variables on SO- and particulate removal efficiencies
and to provide data for developing mathematical relationships to
describe the performance of the R-C/Bahco system.
The final phase of the test program focused on a compre-
hensive monitoring program to characterize the operation of
the system and to develop operating and maintenance cost data
In addition, a program to characterize the sludge produced by*
the scrubber was carried out.
19
-------
ho
O
Figure H The R-C/Bahco Scrubbing System at RAFB
-------
The test program also included the design, procurement and
installation of equipment and instrumentation necessary to
execute various phases of the test program. The results of this
test program are presented in subsequent sections of this report.
21
-------
SECTION 2
R-C/BAHCO SCRUBBING SYSTEM
R-C BAHCO SCRUBBING SYSTEM
This FGD test program was carried out at the R-C/Bahco
scrubbing system installed at the Central Heat Plant at Ricken-
backer Air Force Base near Columbus, Ohio. The heat plant houses
eight coal-fired hot water generators with a total fuel burning
capacity of approximately 330 mm BTU/hr. These stoker-fired
generators burn 2.5 to 3.5% sulfur Ohio coal with an average
heating value of 11,300 Btu/lb.
The R-C/Bahco system was designed to treat up to 108,000
ACFM of flue gas generated at the peak winter load of approxi-
mately 200 MM Btu/hr. The system which must operate over a
relatively narrow range of gas flow, 35,000 to 50,000 SCFM has an
essentially unlimited turndown capability for handling flue gas
by mixing air with the flue gas at low boiler loads. This allows
the system to handle seasonal load variations from 20 to 200 MM
Btu/hr, S02 concentrations from 200 to 2000 ppm and particulate
loadings up to 2 gr/SCFD., Ifi addition, the scrubbing system
copes with 100% load increases occurring in as little as an
hour's time. Table 2.1 shows scrubber operating conditions on
March 30, 1976, these were approximate average load conditions
for RAFB.
TABLE 2.1 TYPICAL SYSTEM OPERATING CONDITIONS
Flue Gas
SO2 Concentration
1st Stage AP
2nd Stage AP
Lime-S02 Stoichiometry
SO2 Removal
Lime Utilization
Particulate Emission
64,000 ACFM@380°F(37,500 SCFM)
1390 ppm
10 in. W.C.
8 in. W.C.
0.876
87.6% (0.615 #/MM Btu outlet)
100%
0.16 #/MM Btu
22
-------
PROCESS DESCRIPTION
Hot flue gas from each of the Heat Plant generators is
passed into a common flue which contains a by-pass stack. This
stack allows makeup air to be drawn into the system at low load
to maintain efficient operation of the mechanical collector and
scrubber. Flue gas, with or without makeup air, is passed
through a mechanical collector to remove coarse particulate
matter before entering the booster fan.
This fan forces flue gas into the first stage of the scrubber
where it is vigorously mixed with scrubbing slurry in an inverted
venturi. In this stage, flue gas is cooled to its adiabatic
saturation temperature and SO, and particulate are scrubbed from
the gas. This partially scrubbed gas rises to the second stage
where it is contacted with slurry containing fresh lime to complete
the required S0~ and particulate removal. Gas from the second
stage enters a cyclonic mist eliminator where entrained slurry
droplets are separated from the gas by centrifugal force to
produce an essentially droplet-free effluent.
Pebble lime from a storage silo is slaked and added directly
to the slurry in the lime dissolving tank. The resulting fresh
lime mixture is pumped to the second stage (upper) venturi to
treat the flue gas stream. The slurry flows by gravity from the
second stage to the first stage where it contacts hot flue gas
entering the scrubber. This countercurrent flow arrangement
results in high S02 removal and efficient reagent usage.
Spent slurry flows by gravity from the first stage of the
scrubber to the dissolving tank. Part of the spent stream
leaving this stage is diverted to the thickener where the slurry
is concentrated to 35 to 40% solids. Overflow from the thickener
returns to the dissolving tank and the underflow is pumped to a
Hypalon-lined sludge pond near the Heat Plant.
The slurry and gas streams described above are illustrated
in Figure 2-1.
MAJOR EQUIPMENT
The following items, shown in Figure 2-1, constitute the
major equipment installed in the R-C/Bahco System at RAFB. A
brief description for each item, including its role in the opera-
tion of the system, is provided below.
Flue Svstem
The flue system includes individual tie-ins to each of
eight boilers. Manual diversion dampers allow for gas flow into
the flue system or bypassing through individual stacks. In
addition, a stack in the main flue upstream from the mechanical
23
-------
10
REAGENT SYSTEM
MODULE
n
LIME OR
LIMESTONE
TRUCK
^
-^--^-^s^
J.--
\/
V,
\y
REAGENT
STORAGE
THICKENER
OVERFLOW
REAGENT
FEEDER
& SLAKER
1 '
^ rC\"
STACK-
MAKEUP
WATER-
t
THICKENER
OVERFLOW
TO
LIME
DISSOLVING
TANK
-R-C/BAHCO
SCRUBBER
UNLOADING^
STATION
SLUDGE
TO POND
BY-PASS
/MAKE UP STACK
r^t
1=3'
FLUE GAS
FROM HEAT
PLANT
MECHANICAL
COLLECTOR
LIME
DISSOLVING TANK
LEVEL
TANK
2nd STAGE PUMP
FLY ASH
DISPOSAL
MILL PUMP
Figure 2-1: R-C/Bahco Scrubber System Flow Diagram
-------
collector allows for the addition of makeup air to the gas or
bypassing of the scrubber. The flues are insulated for personnel
protection and to avoid corrosion when the system is operated with
makeup air at low load conditions. A multiport pitot tube is
located in the main flue upstream from the mechanical collector to
measure gas flow rates.
Mechanical Collector
A Flex-Kleen mechanical collector handles 108,000 ACFM of
flue gas at 475°F at dust loads up to 2 gr. per SCF. This col-
lector is located in the main flue upstream from the booster fan.
The Flex-Kleen collector operates at approximately 5 in.
w.c. pressure drop at full load and 1.5 in. w.c. at minimum load.
The overall efficiency for a combination of the Flex-Kleen collector
and the individual collectors is 85 to 95%. A common vacuum type
ash handling system is used to convey bottom ash from the stokers
as well as the ash from the mechanical collectors to a silo. Ash
stored in the silo is removed by truck to a disposal site.
Booster Fan
The booster fan draws flue gas and air mixtures through the
mechanical collector and forces them into the R-C/Bahco scrubber.
The fan was oversized by 200 H.P., for a total of 700 H.P., to
allow for high gas flow rates at pressures up to 30 in. W.C. for
the EPA test program. The scrubber normally operates at 15 to 18
in. W.C. and the mechanical collector and flues require an addi-
tional 5 in. W.C. at full load.
The gas flow rate is varied to compensate for seasonal load
changes by adjusting the fan inlet dampers. In the winter very
little.air is mixed with the flue gas and in the summer up to two
thirds of the flow is makeup air.
In addition to the pitot tube, the fan current draw is used
to check the flue gas flow rate.
Reagent System
The reagent system, shown in Figure 2-1, consists of a reagent
unloading system, storage bin, feeder, and lime slaker.
The pneumatic unloading system handles 3/4" pebble lime or
limestone at a rate of 25 tons per hr. The 120-ton capacity
storage bin is equipped with a dust collector, pressure-vacuum
relief, level indicator, high and low level alarms, outlet slide
gate, and a motor driven live bottom which is activiated by the
lime slaker.
25
-------
The weigh belt feeder is equipped with manual and automatic
controls and a totalizer. Either the S02 mass flow rate or
dissolving tank pH can be used to control the reagent feed rate.
The lime slaker includes a water totalizer, grit removal circuit,
high temperature alarm, and dust and vapor venting system. The
slaker overflows into the lime dissolving tank located directly
below.
When limestone is being used as the reagent for SO,, removal,
the slaker is used to wet the limestone prior to its entering the
dissolver tank.
Lime Dissolving Tank
The lime dissolving tank, which serves as the surge tank for
the entire system, is made of 316L SS. In this tank slaked lime
or limestone is blended with spent slurry for recirculation to
the scrubber.
R-C/Bahco Scrubber
The scrubber module is an R-C/Bahco size 50 fabricated from
316L SS. This module is approximately twelve feet in diameter by
sixty feet high and has a nominal gas handling capacity of
50,000 SCFM.
A fiberglass reinforced polyester (FRP) stack, 5.5 feet in
diameter and 20 feet high, is mounted on top of the scrubber. The
scrubber module incorporates two inverted fixed-diameter Venturis.
Each has a corresponding level tank located outside the shell of
the scrubber with a manually adjusted weir. The pressure drop
for each venturi can be varied over a range of 5 to 15 in. W.C.
by adjusting the position of the weir in the appropriate level
tank. A fluid mill, which grinds coarse particles in the slurry,
is located in the bottom of the scrubber module.
Second Stage Slurry Recycle Pump
The second stage recycle pump circulates slurry through the
entire scrubbing system. This pump is rubber lined and is rated
at 2600 gpm at 20 psig. A 316 SS shaft sleeve and a water purge
in the stuffing box are used to minimize wear and corrosion.
Mill Pump
The mill pump is identical to the second stage slurry pump
but it operates at 2000 gpm at 25 psig. This pump is the prime
mover for the fluid mill at the base of the scrubber module.
26
-------
Thickener
A thickener which is 25 feet in diameter and 8 feet in
height is used for solids surge capacity, slurry density control,
and thickening sludge for disposal.
The tank is Douglas Fir and the rake mechanism is rubber-
covered carbon steel. The rake mechanism has a lifting device
and a torque sensor with a high torque alarm and cutoff for its
protection.
The maximum thickener feed rate is approximately 100 GPM at
10 wt.% solids and the maximum underflow rate is 35 GPM at 25
wt.% solids. The minimum underflow rate is approximately
3.5 GPM at 40 wt.% solids without recvcle of solids to the
scrubber. Thickener overflow is returned to the lime dissolving
tank by gravity.
Sludge concentrated adequately for disposal, 35-40 wt.%
solids, is continuously pumped from the bottom of the thickener.
A slurry density control provides for recirculation of the
sludge to the scrubbing system if the scrubber slurry is below
10% solids. Sludge flow is diverted to the pond for disposal
when scrubber slurry solids reach approximately 12%.
Sludge Transfer System
The sludge transfer system includes two thickener underflow
pumps and two transfer lines to provide 100% standby capacity.
The two underflow pumps are air-operated with replaceable neoprene
diaphragms and 316 SS wetted parts. The pumps are capable of
pumping up to 40 GPM of sludge at 75 psig.
Quick disconnects are used to hook up these pumps to the
thickener and the Ih in. I.D. polybutylene sludge transfer
lines.
Sludge line velocities are maintained between 4 and 6 feet
per second by setting the pumping rate at 20 to 30 GPM. The
sludge lines run underground inside a sleeve to permit easy
removal for cleaning or replacement in the event that it is
required.
Sludge Pond
The sludge disposal pond shown in Figure 2-2 is located
approximately 700 feet from the scrubbing system. The pond is
lined with Hypalon, (chloro-sulfonated polyethylene) and is
approximately 450 feet long, 250 feet wide, and 12 feet deep An
27
-------
K)
CO
Figure 2-2 Lined Storage Pond at RAFB
-------
underdrain system which allows ground water to be removed from
beneath the liner also serves as a means of detecting any leaks
which may occur. The pond was designed to hold sludge produced
by scrubbing flue gas from the combustion of 200,000 tons of 5%
sulfur coal. The life of this pond is well over five years at
present levels of coal consumption.
PROCESS CONTROL
The ability of a flue gas scrubber to accommodate changes in
load or other operating parameters is a very important factor in
determining its suitability for a given application. For small
industrial applications this ability is essential from three
points of view. First, the scrubber must be able to reduce
emissions of pollutants to acceptable levels over the entire
range of flue gas rates generated. Second, accommodations to
load changes must be accomplished rapidly without special attention
from operating personnel. Third, variations in the scrubbing
system's internal processes must not interfere with its ability
to perform as required.
Manual Controls
A combination of manual and automatic controls are used to
adjust and regulate system variables. The manual mode is used to
control variables which do not have to be adjusted frequently or
are essentially constant over the entire operating range of the
system. Gas flow rate, slurry circulation rate and first and
second stage pressure drops are controlled manually.
Gas Flow Rate
At RAFB the gas rate to the scrubber is set manually by
adjusting the booster fan inlet dampers. Variations in flue gas
rates are accommodated by mixing makeup air with the flue gas to
maintain the desired total gas flow rate.
The makeup air system includes an open stack which also
serves as a system bypass. No control dampers or other devices
are used to regulate makeup air rates. Merely positioning the
booster fan damper to obtain the desired total gas flow rate is
sufficient for control purposes. Control is maintained, irrespective
of the flue gas rate, as long as the flue gas volume is less than
the total flow that the booster fan is able to accommodate for
the damper position selected. If the flue gas flow exceeds this
total rate, gas bypasses through the makeup air stack and activates
a temperature alarm to alert the heat plant operators.
29
-------
Slurry Recirculation Rates
All slurry circulation rates are manually adjusted and are
set to maintain line velocities between 4 and 8 ft/sec. The
system is designed to accommodate all loads and load changes
without adjusting these slurry circulation rates.
The following loops use this constant flow principle:
o Scrubber mill or first stage slurry recycle loop
o Second stage slurry feed
o Thickener feed
o Thickener underflow (sludge disposal)
An alumina flow nozzle located in the mill in the lower
section of the scrubber module is used to control slurry flow in
the first stage recycle loop. This nozzle was designed to
restrict slurry line velocities to 6-8 ft/sec and to provide the
necessary agitation and fluid grinding in the mill.
The main slurry flow from the dissolver tank to the scrubber
via the second stage slurry pump was set by checking the pumping
rate and setting the pump speed to deliver the desired flow rate.
Minor flow adjustments can be made if necessary by adjusting a
rubber pinch valve on the second stage slurry pump discharge.
The thickener feed rate is set at 60 to 80 gpm. This flow
rate is automatically adjusted via a magnetic flow meter -
pneumatic pinch valve flow control loop.
The thickener underflow or sludge flow is adjusted by re-
gulating the air flow to the diaphragm type sludge pump. This
rate is set to achieve a thickener under flow concentration of
approximately 40% solids and a flow rate of approximately 20 gpm.
Plugging and abrasion of lines was avoided by using slurry
lines with maximum pitch, minimum length, a minimum of bends as
well as the conservative design velocities (4-8 ft/sec). Traps
provided at low points in the level tanks and liquid seals
allowed for periodic blow-down of accumulated coarse solids.
These coarse solids account for most of the abrasive wear experienced
in this type of system.
30
-------
Pressure Drop
Pressure drops in both the upper and lower Venturis are
manually adjusted by raising or lowering weirs in level tanks
outside the scrubber. Each stage can be adjusted independently
to produce a pressure drop from 5 to 15 in. W-C. Raising or
lowering the weir causes the slurry level in the scrubber, near
the lower edge of the venturif to rise or fall. The pressure
drop in the venturi is directly proportional to the slurry level
in the vicinity of the venturi; therefore, weir adjustments pro-
duce proportional changes in venturi pressure drop. The pressure
drop once set is virtually unaffected by changes in gas flow
rate. This insensitivity of pressure drop to gas flow results
from the self-adjusting action of the slurry level in each venturi.
When an increase in gas flow tends to increase the pressure drop,
the slurry level tends to drop because of increased pickup of
slurry by the gas stream. This drop in slurry level causes a
decrease in pressure drop. When gas flow rate decreases, the
slurry level rises, increasing the pressure drop. Thus, the
pressure drop is essentially self-compensating as the gas flow
varies and tends to stabilize at a value near the inital setting.
This insensitivity of pressure drop to gas flow eliminates
the necessity for making adjustments to accommodate the frequent
heat plant load changes imposed by the daily needs of the Base.
Automatic Controls
There are three essential automatic controls in the R-C/
Bahco scrubber: reagent feed, slurry density, and makeup water
or system level.
Reagent Feed
Reagent-S02 Stoichiometry is the key variable for controlling
SO, removal efficiency in this FGD system. Allowable emission
rates are exceeded if too little reagent is supplied to the
system. If too much is added, excessive reagent costs will be
incurred. The reagent feed system at RAFB is designed to main-
tain a preselected reagent-S02 Stoichiometry for any load con-
dition and any coal sulfur content. Both the gas rate and the
SO2 concentration to the scrubber are measured continuously.
These measurements are combined in a reagent feed rate ratio
controller which can regulate feed rate over a range of 20 to 1
to maintain the desired reagent S02 Stoichiometry.
Slurry Density
Maintaining slurry density in the system is important for
two reasons: first, the proper operation of the sludqe disooLl
system requires a relatively constant feed composition and
second, a minimum inventory of solids in the system is'required
31
-------
to eliminate scale formation. The slurry density control system
at RAFB operates between set points of 10 and 12% solids, 67 and
69 lbs/ftj respectively. A sensor monitors slurry density and
a controller is activated to allow thickened sludge at 40% solids
to3flow to the pond when the density reaches approximately 69 Ibs/
ft . Sludge flows continuously to the pond until the density in
the system drops to 67 lbs/ft3. When this point is reached, sludge
is recycled to the scrubber and the line to the pond is flushed
with water. This switching process is repeated as necessary to
maintain system density in the desired range.
Makeup Water
The total water requirement for the system varies directly
with load. Evaporative cooling of the flue gas consumes the
bulk of the water used.
Water is added to the system at several locations including
slurry pump seals and the lime slaker. These items require a
relatively constant amount of water irrespective of system load.
The balance of the makeup water is added through six spray
manifolds located inside the scrubber module. The amount of
water added through these sprays is regulated by a level sensor
located in the lime dissolving tank. This level sensor activates
a programmed controller which adds water in a preselected sequence
for a preset time period through each of the six spray manifolds.
This controller is designed to start water additions when the
level in the dissolver tank reaches the low set point and stop it
when the upper set point is reached. Subsequent evaporation and
sludge removal cause the level in the dissolver to drop initi-
ating the water addition cycle as necessary to maintain the
desired level.
There were a number of problems associated with the elec-
tronic instruments installed at RAFB which were all related to
improper grounding of milliamp current signal loops. Rewiring
these control .loops rectified the problems.
32
-------
SECTION 3
TEST PROGRAM
This section outlines the R-C/Bahco test program from March
1976 through June 1977. The following tests were conducted on
the R-C/Bahco FGD system at Rickenbacker AFB:
o Operability/Material Balance
o Lime-reagent process variable
o Lime-reagent verification
o Particulate collection efficiency
o Limestone reagent process variable
o Sludge Characterization
o Scrubber reliability monitoring
TEST DESCRIPTIONS
Operability/Material Balance Tests (May-Sept., 1976)
These tests were conducted to establish the range of opera-
ting conditions over which the R-C/Bahco scrubber could be
operated and to verify performance at design conditions by com-
pleting material balances. The testing schedule is shown in
Figure 3-1.
Maximum and minimum gas flow rates, pressure drops, and
slurry circulation rates were determined. In addition, pre-
liminary SO2 and particulate performance data at the limits of
the system's capabilities was obtained.
The system was operated at the design gas rate of 50,000
SCFM and complete material balances on calcium, sulfur and total
solids were performed.
33
-------
U)
Start-up
Operability/
Material
Balance
Lime Process
Variable
Lime
Verification
Paniculate
Limestone
Process
Variable
Sludge
Characterization
Reliability
Monitoring
-c
Tiiir
Tiiiiir
I 1 II
J -
j i
J I
i I I I
Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. April May June
1976 1977
FIGURE 3-1 The Characterization Program Test Schedule
-------
Lime Process Variable Tests (Dec. 1976, Feb. 1977)
These statistically designed tests helped to establish the
quantitative effect on S02 removal of the following process
variables: gas flow rate, first and second stage pressure
drops, mill and second stage slurry rates, lime/S02 stoichiometric
ratio, slurry inventory, and slurry solids concentration.
Lime-Reagent Verification Tests (March 1977)
These tests were run to verify the results obtained in the
lime process variable tests and to determine the effect of very
dilute scrubber slurry (2% solids) on system performance.
Particulate Collection Efficiency Tests (April 1977)
These tests were a continuation of the particulate tests
run in combination with the SO2 efficiency tests in December
and February. Relationships between system variables, including
particle size distribution and particulate removal efficiency,
were determined.
Limestone Process Variable Tests (May 1977)
The limestone process variable tests were run utilizing
the same statistically designed test plan used for lime. The
effect of system variables on S02 removal efficiency and reagent
utilization were determined.
Sludge Characterization Tests (April-May 1977)
Sludge samples generated at RAFB were tested to determine
dewatering, transport and disposal characteristics. Samples of
sludge from lime as well as limestone scrubbing were tested.
Reliability Monitoring (March 1976 - June 1977)
The R-C/Bahco system was monitored from March, 1976,
to June, 1977, to document the system's operating and maintenance
history and to obtain data for a cost analysis. Data was
gathered on reagent, coal, water and power consumption as well
as on operating and maintenance labor.
The results of the various test programs are presented in
subsequent sections of this report: The Operability/Material
Balance Test results are presented in section 4, the Lime and
Limestone Process Variable Test results and the Verification
Test results are presented in section 5, the Particulate Test
results are presented in section 6, the Sludge Characterization
test results are presented in section 7, and the results obtained
during the Reliability Monitoring are presented in sections 8
35
-------
DESIGN OF PROCESS VARIABLES TESTS
Identification of Significant Control Variables
S02 scrubbing with a slurry containing either lime or
limestone may be influenced by many factors including interactions
between variables. In order to study the performance of the
R-C/Bahco scrubber efficiently, a series of statistically
designed experiments was employed. This type of experimental
plan maximizes the amount of information obtained from a relatively
small number of tests.
The overall system was examined in the first step in this
experimental plan to identify the controllable independent
variables. Eight variables were selected and incorporated into
a statistical screening design. These screening tests were
used to determine the significance of the process variables in
relation to several dependent (response) variables.
The controlled independent variables selected were:
X,: Scrubber inlet gas flow rate (SCFM)
X~: Scrubber first stage pressure drop (in. W.C.)
X3: Scrubber second stage pressure drop (in.W.C.)
X.: Slurry rate to the second stage (GPM)
X5: Slurry rate to the mill (GPM)
Xfi: Stoichiometric ratio (moles CaO or CaC03/mole S02
in the inlet gas stream)
X_: System slurry volume (gals.)
X0: Slurry concentration (Wt. % solids)
8
The dependent (response) variables monitored were:
1. Exit gas temperature
2. Outlet SO2 concentration
3. 1st stage level tank slurry pH
4. 1st stage drop collector slurry pH
5. 2nd stage slurry feed pH (dissolver pH)
6. 2nd stage level tank slurry pH
36
-------
7. 2nd stage drop collector slurry pH
Component concentrations (Wt.% CaSO, 2H20,
Wt.% CaSO., 1/2 H00, Wt.% alkali {CaO, CaCO.
8.
M+. * r'!"cr/3 - -L-/ < "2"
9. 1st stage liquid pick-up
10. 2nd stage liquid pick-up
In addition to the above the inlet SG>2 concentration was moni-
tored for all tests.
Lime Reagent Tests
A test program consisting of twenty-one runs was undertaken
to study the SO, removal efficiency of the R-C/Bahco scrubber
using lime as tne scrubbing reagent. The screening test program
was a 1/16 fractional factorial of a 28"4 full factorial stati-
stical design.1 The program was comprised of sixteen combina-
tions of high and low levels of the eight controllable independent
variables shown in Table 3.1. Five tests at centerpoint condi-
tions were run to estimate the experimental error. Tests were
performed in a random order to minimize the effects of extraneous
factors on the experiments. However, in certain cases, randomi-
zation required major system changes which would have resulted in
substantial testing delays. The test sequence was modified to
minimize these delays. Table 3.2 shows the target levels for
each of the eight independent variables used in the lime tests.
Limestone Reagent Tests
The use of limestone as an alternate scrubbing reagent for
SO2 removal was investigated in a second series of screening tests.
The same independent variables and the same svstem responses used
in the lime tests were studied. Adjustments were made"in the
levels selected for the independent controllable variables to
accommodate system operating limits encountered in the lime test-
series and to provide for a wider range of stoichiometry to re-
flect the lower reagent utilization expected for limestone.
The adjusted variable levels used for limestone are listed in
Table 3.3.
Process Variable Test Procedures
After test conditions were set for each run and had stabi-
lized, usually within an hour, data for all control and response
variables was taken. Approximately one hour after the first set
of data was recorded a second set was taken. The averacre values
of these results were used in subsequent analyses. Load vari-
ations at RAFB cycled in a time span similar to the system's 12
to 24-hour residence time. This fact rendered steady-state
-K-P
(1) See Fractional Factorial Designs, Part I G E P Bov
and J.S. Hunter, Technometrics, August, 1961, Vol."3, NO. 3,
37
-------
TABLE 3.1 SCREENING TEST MATRIX FOR THE R-C/BAHCO TEST PROGRAM
Run Variable
No. No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Xl
+
+
+
+
+
+
+
+
-
-
-
-
-
-
-
-
0
0
0
0
0
x2
+
f
+
+
-
-
-
-
+
+
+
+
-
-
-
-
0
0
0
0
0
X3
+
+
+
+
-
-
+
+
-
-
+
+
-
-
0
0
0
0
0
X4
+
_
+
-
+
-
+
-
+
-
+
-
+
-
0
0
0
0
0
X5
+
_
+
-
+
+
-
-
+
+
-
+
-
-
+
0
0
0
0
0
X6
+
+
-
-
-
+
+
-
-
+
+
+
+
-
-
0
0
0
0
0
X7
+
_
+
-
+
+
-
+
-
-
+
-
+
-f-
-
0
0
0
0
0
X8
+
_
+
+
-
-
+
-
+
+
-
-
+
+
-
0
0
0
0
0
Note: High target levels are designated by (+) symbols, low
levels by (-) symbols and center points by (0) symbols,
38
-------
TABLE 3.2 TARGET LEVELS FOR LIME TESTS
2nd Stage
Gas Rate 1st Stage 2nd Stage Slurry Rate,
Lime/SO,
Mill Slurry Stoichi- Slurry Volume, Slurry Cone.,
Target
Levels
High
Level (+)
Center
Point (0)
Low
Level (-)
(SCFM)
Xl
50,000
42, 500
35,000
AP, (in. w.c.)
12
9
6
AP, (in w.c. )
X3
12
9
6
(GPM)
X4 '
2500
1750
1000
Rate (GPM)
X5
2500
1750
1000
ometry
X6
1.0
0.85
0.7
(Gal)
X?
17,000
15,000
13,000
(Wt.%)
X8
15
10
5
-------
TABLE 3.3 TARGET LEVELS FOR LIMESTONE TESTS
2nd Stage
Limestone/
SO-
Target
Levels
High
Level (+)
Center
Point (0)
Low
Level (-)
Gas Rate
(SCFM)
50,000
42, 500
35,000
1st Stage
AP, (in w.c. )
X2
12
9
6
2nd Stage
AP, (in. w.c. )
X3
12
9
6
Slurry Rate,
(GPM)
2500
2000
1500
Mill Slurry
Rate (GPM)
X5
2500
1750
1000
StOichi-
ometrv
X6
1.3
1.0
0.8
Slurry Volume, Slurry Cone. ,
(Gal) (Wt.%)
v y
*7 8
17,000 10
15,000 6
13,000 2
-------
operation impossible. The effect of this situation was minimized
by running tests in as short, a time as practical and by sub-
sequently analyzing measurements and samples to determine the
actual levels of controlled variables during the tests.
These tests were run by setting the appropriate controlled
variables and allowing the system to stabilize. This usually
occurred within an hour. An initial set of data was taken at
this time and a second set was taken thirty to sixty minutes
later. These sets of data were then compared to see if there
were any substantial changes. If these were, additional time
was allowed for stabilization and new data sets were collected.
Basically employing non-steady state testing facilitated
testing at the expense of precise regulation of certain con-
trolled variables, especially limestone/SO^ stoichiometry. A
full range of variable levels was achieved in spite of the in-
ability to preselect them with precision. Analytical tests used
during the test program are listed in Table 3.4.
41
-------
TABLE 3.4 ANALYTICAL TESTS
Method'
Sample
Components
Analyzer
TGA
Impaction
pH Meter
Slurrv
Photometric Gas
Wet Chemical Lime
TGA Lime
Wet Chemical Gas
Wet Chemical Gas
Gas
ASME & EPA Gas
TGA Limestone
Wet Chemical Limestone
Wet Chemical Slurry
Combustion, Coal
etc.
Slurry
.JjH-O, CaSO4.2H_0,
, MgCO,, CaTOH)*
Loss on Ignition '
SO.,
Alkalinity as CaO, Acid
Insolubles, Calcium,
Magnesium, SO., Cl
CaCO.,
co2, o2, co
so2
Fly ash particle size
Fly ash grain loading
CaC03, MgC03
Ca, Mg, total alkalinity,
acid insolubles, SO,
Ca, Mg, TDS, SO4, SO,, Cl
Acid Insolubles, S.G7, %
solids
% S, Heating value, ultimate
analysis
pH
duPont SO,
Analyzer
Orsat
Analyzer
EPA Method
6 SR-C Method
Andersen
Impactor
Great Lakes
pH Probes,
Corning
Battery-powered
Meter
(2) See Appendix B for a description of thermogravimetric analysis and
a list of the wet chemical and other tests used.
42
-------
SECTION 4
OPERABILITY/MATERIAL BALANCE TEST RESULTS
R-C/BAHCO SCRUBBER OPERATING LIMITS
The performance of an R-C/Bahco scrubber is measured by its
ability to handle variations in flue gas flow rate and temperature,
while achieving reductions in SO2 concentrations and particulate
loadings sufficient to meet applicalbe emissions standards. The
following scrubber system variables must be evaluated to deter-
mine levels which must be maintained to insure compliance with
emissions standards.
o Gas Rate
o Slurry Circulation Rate
o First and Second Stage Venturi Pressure Drops
o Slurry Solids Concentration
o Reagent/S02 Stoichiometry
A test program was carried out to establish ranges of the
above variables for the R-C/Bahco System at RAFB which would pro-
duce SO- and particulate emissions within the limits allowed by
the local regulation. The results of these tests, which are
summarized in Table 4.1, were used to select operating conditions
in the subsequent process variable test programs. In addition,
results from other portions of the overall characterization pro-
gram which defined system limits are included in Table 4.1.
Gas Rate
The total gas rate to the scrubber must be maintained between
35,000 and 55,000 SCFM. In this range of gas flow, liquid pickup
in the venturi results in liquid-to-gas ratios high enough to
saturate the gas stream and produce good S02 and particulate
removal. In addition to inadequate liquid pickup at gas rates
below 35,000 SCFM, unstable operation affecting the first stage
venturi can occur. At first stage pressure drops over 12 in.
w.c., gas flow tends to surge, producing intermittent gas flow
through the scrubber, when this occurs, SO- and particulate
emissions exceed allowable levels.
43
-------
At gas rates approaching 60,000 SCFM, the capacity of the 2nd
stage mist eliminator is exceeded and droplets of slurry entrained
by the gas stream are emitted from the scrubber.
TABLE 4.1 R-C/BAHCO SCRUBBER OPERATING LIMITS
Variable Minimum Maximum
Gas Rate, ACFM 35,000 55,000
Slurry Circulation Rate, GPM 1,500 3,000
Venturi Pressure Drops/Stage
for each stage in. W.C. 6 12
Slurry
Concentration, wt.%
Solids 2 25
Reagent/SO- Stoichiometry
Moles Reagent/Moles S02
Based on Inlet SO- levels
Lime 0.45 1.05
Limestone 0.55 1.2
Slurry Circulation Rate
The second stage pump circulates slurry through the scrubbing
system. A slurry circulation rate of 1500 gpm is required to
maintain agitation in tanks, line velocities sufficient to keep
solids suspended and adequate flow to the Venturis. A slurry
rate in excess of 3000 gpm can result in slurry overflowing
from the first stage into the inlet ductwork when liquid pickup
in the first stage venturi is low.
Slurry flow through various sections of the scrubber is
essentially self regulating once the adjustable weirs in the
first and second stage level tanks have been set for the desired
pressure drops.
Since high slurry circulation rates promote S02 absorption
with limestone, minimize the potential for solids accumulations
and minimize the need to throttle the flow of slurry to the
scrubber, the circulation rate should be maintained between
2200 and 2400 gpm.
44
-------
Venturi Pressure Drop
In each stage, adjustable weirs are used to regulate ^
pressure drop in the Venturis. A minimum pressure drop of 5 in
w.c. in the first stage is recommended to maintain sufficient
liquid pickup to provide for efficient particulate and S02 removal,
cooling of the flue gas, agitation in the drop collector and
adequate flow velocity in the lines from the drop collector to
keep solids suspended. A similar pressure drop limitation applies
to the second stage.
If the recommended pressure drop of 12 in. w.c. is exceeded
in the first stage, unstable operation can occur. Excessive pres-
sure drop in the second stage, especially at low gas flow rates,
results in the emission of slurry droplets from the scrubber. A
normal slurry pumping rate of 2400 gpm and weir levels set to
maintain 8 - 9 in w.c. pressure drop in each venturi allows the
scrubber to accommodate day-to-day load changes without adjustment
while maintaining acceptable levels of particulate and SO2 removal.
Slurry Concentration
The R-C/Bahco system is not sensitive to variation in slurry
concentration. The scrubber operated successfully for several
months at 25% solids without problems. Concentrations as low as
2% solids did not affect S02 removal or scrubber operation.
However, when the scrubber is operated at or above average load
conditions and the solids concentration drops below 5% the thickener
cannot produce a sludge suitable for disposal, i.e. 35 to 45%
solids.
Reagent/S02 Stoichiometry
The range of the reagent/SCU Stoichiometry ignoring SCU
emission requirements is limited by the capacity of the reagent
feed system, the SO2 rate to the scrubbing system and the ability
of the- scrubber components to withstand corrosion at low pH levels.
During the process variable tests, lime stoichiometries varied
from 0.36 to 1.08 and from 0.59 to 1.55 during the limestone
tests. Minimum levels of 0.45 for lime and 0.55 for limestone are
suggested to prevent pH levels from dropping below 4.0 . Maximum
levels of 1.05 for lime and 1.2 for limestone are suggested.
These levels represent the point at which very little additional
S02 removal can be obtained by increasing the amount of reagent.
In other words, these stoichiometries, when running at optimum
scrubber operating conditions, represent the upper limit of SO
removal for the system. 2
Maintaining an optimum stoichiometric ratio is critical if
SO2 emissions are to be kept below the required level of 2.2#/million
Btu of coal fired while minimizing reagent costs. in the ranqe of
45
-------
of SO2 removal required, i.e. 70%, for a 3.3% sulfur coal, when
lime is used as the scrubbing reagent, the lime/SO- ratio is set
at approximately 5% above the minimum ratio of 0.70. This safety
factor reduces the possibility that lime feed variations will pro-
duce excessive S02 emissions but also increases overall operating
costs by approximately 2%.
When limestone is used as the scrubbing reagent, larger
reagent feed rate variations can be tolerated without substan-
tially altering SO2 removal efficiency since limestone is not
fully utilized in the system. Approximately 5 to 10% of the
limestone is unreacted and is recirculated in the slurry. This
recirculated limestone acts as a buffer and compensates for sub-
stantial short term feed rate variations while maintainina the
desired SO, removal efficiency. The required S02 emission rate
of 2.2 Ibs/MM Btu can be achieved at a 0.75 limestone stoichiometry,
MATERIAL BALANCES
As part of the preliminary test program, measurements were
made to perform material balances on the R-C/Bahco scrubbing
system. Overall balances for calcium, sulfur and total solids
were completed during a two month period following the startup in
March 1976. These balances verified the fact that the R-C/Bahco
scrubber was performing satisfactorily and that the methods used
to measure critical parameters were accurate. Data from these
tests are tabulated in Appendix C.
There are many measurements which must be made to complete an
accurate material balance. At the RAFB facility, the accuracy of
the following key measurements determined the overall accuracy of
the balances performed. An analysis of these measurements resul-
ted in an expected uncertainty of approximately 16% in the material
balance..
o SO- concentration in the flue gas
o Slurry solids concentrations
o Sludge flow rate to the pond
o Gas flow rate to the scrubber
o Lime feed rate and composition
46
-------
Material Balance Results
Ideally, material balance tests should be run at steady state
conditions for relatively long periods of time to minimize the
effects of short term system variations and inherent inaccuracies
in the measurements taken. The following procedure was selected
since steady state conditions could not be achieved at RAFB due
to the outside constraints on the heat plant output and exten-
ded testing is economically impractical. Each of the material
balances was run over a period of approximately 43 hrs. and appro-
ximately twenty measurements of each important variable were taken.
The results of these balances, shown as time averages or
rates for the test periods, are presented in Table 4.2. The
calcium rates indicated a deficit of 16.2% in the first case
and an excess of 4.5% in the second, the sulfur rates in both
cases indicated deficits of 16.6 and 13.2% and the solids a
deficit of 2.4% and an excess of 12.0% in the second.
Many measurements in the first case were combined to obtain
these results. Some of the more important measurements are dis-
cussed in detail below. Each measurement has an inherent amount
of uncertainty or error associated with it. When these measure-
ments are combined as in the above balances, the total uncertainty
is greater than any individual value. The inherent error involved
in the material balance measurements were 14.9%1 for the total solids
and calcium, and 16.5% for sulfur. The variations observed in the
balances were within the estimated ranges of uncertainty. This
indicates that there were no errors which were unaccounted for
and the balances were as accurate as could be expected.
SO2 Concentration
Accurate measurement of the flue gas SO2 concentration is
essential to obtaining meaningful results in an S02 removal test
program. The system at RAFB utilized a DuPont Photometric ana-
lyzer with sintered stainless steel probes for these measurements.
Regular calibration usincr certified SO2~nitroqen mixtures coupled
with the analyzer's inherent reliability normally produced results
within the expected range of uncertainty 4-3%. However, even the
most sophisticated and reliable sampling systems are susceptible
to occasional problems that result in inaccurate measurements. A
rapid field wet test was developed to verify the measurements"
obtained from the DuPont analyzer. This method which is des-
cribed in detail in Appendix B has an expected uncertainty of
+ 6%.
(1) The Uncertainty Analysis Principle used in the program are
listed in Experimental Methods for Engineers bv B T
McGraw Hill, 1966, p. 37 et seq.
47
-------
ft.
00
TABLE 4. 2 MATERIAL BALANCE RESULTS
Calcium Sulfur Solids
#m/hr . #m/hr. #/hr.
Test % %
Period 1976 In Out Variation In Out Variation In Out
5/19 to 5/21 3.34 2.66 -16.2 3.21 2.68 -16.6 411 401
5/26 to 5/28 2.48 2.59 + 4.5 2.91 2.53 -13.2 358 407
%
Variation
-2.4
+12.0
Notes: Results are presented as hourly averages for the test period including accumulations in system
inventory.
-------
Slurry Solids Concentration
Accurate measurements of slurry solids concentrations were
necessary to determine the change in the inventory of solids in
the scrubbing system as well as the quantity of solids leavina
the system.
Both hydrometer (specific gravity) and moisture balance measure-
ments were taken to determine the solids content of slurry samples.
Correlations using the results of these measurements were developed
to provide a rapid method for monitoring the slurry solids concen-
trations. The following correlation best represents slurry
solids concentration - specific gravity data from 2% to approxi-
mately 60% solids.
Wt.% Solids = 38.KS.G.)- 0.973)
(S.G.) (4.1)
The following linear relationship was developed for 5% to 25%
solids.
Wt.% Solids = 140.6 (S.G. - 1.0196) (4.2)
The slurry solids and specific gravity data used to develop
the correlations in Figure 4-2 are listed in Table 4.3. An
examination of the results in Figure 4^2'indicate that there are
substantial variations between predicted and observed values of
slurry solids concentrations. These variations are due mainly to
differences in the chemical composition of the solids since there
are substantial difference in specific gravity between different
solid species.
In order to obtain accurate data for the material balance,
test samples were routinely analyzed on a moisture balance.
which has an expected uncertainty of + 3%.
Sludge Flow Rate
The sludge flow rate was measured by periodically taking
samples from the sludge line at the pond. The sludge flow rate
was determined by weighing a sample collected in a known time
period. The average of the observed rates and the total pumping
time, which was accumulated on a totalizing timer on the control
panel, were used to determine the total quantity of sludge leaving
the system. These rates had an expected uncertainty of approxi-
mately + 2%.
49
-------
1400
INLET
-0 OUTLET
1200
1000
§: aoo
C/3
LU
H-
UJ
600
400
200
I
I
200 400 600 800
DuPONT ANALYZER READING, PPM
1000
FIGURE 4-1: Verification of inlet and outlet SC>2
concentration data.
50
-------
1.80
1.70
1.60
O DISSOLVER
X FIRST STAGE
A CYCLONES
D POND
OMILL
1.50
< 1.40
C3
O
at
1.20
LINEAR MODEL
NONLINEAR MODEL/
1.10
1.00
spa
I
10
20 30
Wt. % SOLIDS
40
50
60
FIGURE 4-2: The relationship between slurry
solids concentration and
specific gravity.
51
-------
TABLE 4.3 SLURRY SOLIDS CONCENTRATION AND SPECIFIC GRAVITY
Hydrometer Moisture Balance
Location Specific Gravity % Solids
Dissolver 1.195 24.8
Dissolver 1.144 20.0
Dissolver 1.160 23.1
Dissolver 1.132 17.7
Dissolver 1.118 17.9
Dissolver 1.136 18.3
Dissolver 1.151 20.4
Dissolver 1.156 21.1
1st stage seal 1.238 - 27.3
1st stage seal 1.181 22.1
1st stage seal 1.168 19.7
1st stage seal 1.164 16.6
1st stage seal 1.161 16.3
1st stage seal 1.142 19.3
1st stage seal 1.161 19.2
Dewatered slurry 1.800 58.0
Dewatered slurry 1.160 23.3
Dewatered slurry 1.700 55.0
Dewatered slurry 1.190 23.0
Dewatered slurry 1.370 40.0
Mill 1.032 6.5
Mill 1.043 7.9
Mill 1.047 8.5
52
-------
Gas Flow Rate
The determination of gas flow rates was essential for suc-
cessful completion of the material balance tests. The flue
system at RAFB was designed to accommodate existing site con-
ditions at a minimum cost. As usual, an ideal location for gas
flow measurements was lacking. In order to minimize the uncer-
tainty in measuring gas flow rates two independent methods were
utilized. A multiport pitot tube was installed in the flue
system downstream from the flue gas inlets in the longest straight
run available. In addition fan differential pressures and fan
motor current readings were used in conjunction with their respec-
tive performance curves^ to provide an independent measurement of
gas flow rates.
The gas flow data obtained from these two methods of measure-
ment are listed in Table 4.4 and plotted in Figure 4-3.
An inspection of Figure 4-3 indicates that the fan flow data
verify the measurements made with the pitot tube. These flow
measurements had an expected uncertainty of + 5%.
Lime Feed Rate and Composition
The amount of reagent added to the system and its chemical
composition are essential elements in completing an accurate
material balance. Lime was fed to the system via a weigh belt
feeder. For these tests the feed rate was set manually to mini-
mize variations. Samples were taken from the discharge of the
feeder at regular intervals to check the feed rate and to obtain
samples for subsequent chemical analysis. Table C.2 in Appendix C
contains the chemical analyses of the lime samples taken. The
lime feeder calibration curve is also included in Appendix C,
Figure C. The expected uncertainty in the lime feed was approxi-
mately + 2%.
53
-------
TABLE 4.4 COMPARISON OF GAS FLOW MEASUREMENT TECHNIQUES
Gas Flow Rate, SCFM X 10
-3
Run
#
14
5
24
10
2 15
7
30
22
1
9
35
33
Scrubber
Inlet
Temp.
260
227
270
240
290
315
249
327
213
281
245
267
Fan
Current
Draw
(amps)
95
99
90
91
78
67
66
59
70
78
87
85
Pitot
Tube
AP
(IWC)
1.02
1.14
0.92
0.92
0.56
0.44
--
__
Multiport
Pitot Tube
Flow Data
58
63
55
56.5
42.5
36.8
30.4
26.8
44.1
45.1
50.9
51.6
Fan
Flow
Data
56
60.2
53.9
54.6
44.8
35.8
33.8
28.0
38.4
43.8
50.4
49.5
Note:
This data has been corrected by standard conditions, 14.7 psia and 32°F.
-------
70
X
LL
o
Cfi,
LLJ
CD
I
Q.
ill
en
<
O
60
50
40
30
20
10
IDEAL CORRELATION LINE
10
20
30
40
50
60
GAS FLOW RATE FROM FAN PERFORMANCE DATA
(SCFMX10-3)
FIGURE 4-3: A comparison of gas flow rates
derived from fan performance data
and pitottube measurements.
55
-------
SECTION 5
S02 REMOVAL TESTS
PRELIMINARY S02 REMOVAL TESTS
Initial SO, removal tests confirmed the high performance
capability of tfie R-C/Bahco scrubber for S02 removal. The
removal efficiencies observed, 87 to 99%, were higher than the
84% required to comply with the State regulation of 1.0 Ib.
S02/MM Btu heat input initially in effect. Since then, the
requirement has been changed to 2.2 Ib. S09/MM BTU requiring 69%
SO2 removal. Table 5.1 presents the S0_ removal data collected
during startup and early operation. Figure 5-1 illustrates the
effect of lime/SO2 stoichiometry on S02 emission rates. Emission
rates below 0.10 Ibs. SO2/MM BTU were readily attained at lime/S09
stoichiometries of 1.0 to 1.1. In these tests, inlet levels of
300 - 500 ppm were reduced to less than 10 ppm in the outlet gas
stream. Virtually complete lime utilization was attained at
SO- removal efficiencies up to approximately 90%. Lime utilization
between 90 and 95% was achieved at S02 removal efficiencies in
the 97 to 99% range. For these tests, EPA Method 6 was used to
determine S02 concentrations in the scrubber outlet gas. A
DuPont SO2 analyzer was used to measure inlet concentrations.
LIME SCREENING TESTS
The lime screening tests, as outlined in Section 3, were
conducted to identify variables which have a significant effect
on SO- absorption. In a series of 21 tests conducted in December 1976
and February 1977, eight variables which were susceptible to in-
dependent control were screened. Table 5.2 shows average values
for the three levels selected for each of the eight variables
studied. Some difficulties were encountered in attaining the
precise levels planned for certain variables. A comparison be-
tween these values and values listed in Table 3.2 indicates that
the low levels of 1st and 2nd stage pressure drops and slurry
pumping rates were different from the target levels. The low
levels of pressure drop were difficult to achieve because they
were too close to the lower limits of the system in certain
cases. The pumping rates varied from the desired levels because
variations in the system volume changed conditions on the suction
side of the 2nd stage slurry pump. The levels achieved, however,
were satisfactory for the purposes of the screening study since
56
-------
TABLE 5.1
S02 REMOVAL EFFICIENCY DATA
1976
Date
3/30
4/8
5/19
5/26
5/26
5/26
5/27
5/27
Coal
Sulfur
Content
3.24
3.24
3.25
2.64
2.64
2.64
2.01
2.01
Coal
Firing Rate
(MM BTU/hr.)
132,2
115.2
47.9
54.0
52.8
43.3
44.8
44.3
Inlet SO,
Concentration
(ppm)
1,392
1,200
454
555
489
401
327
323
Outlet SO,
Concentration
(ppm)
156
45
24
5
8
8
5
5
SO, Removal
Efficiency
(%)
87.6
95.7
94.4
99.0
98.2
97.9
98.3
98.2
SO, Emission
Rate
(Ibs./MM BTU)
0.621
0.21
0.29
0.045
0.084
0.095
0.061
0.061
Lime
Utilization
(%)
100.0
94.0
98.8
90.3
90.5
91.2
94.2
95.4
Note: Removal efficiencies were corrected for increased outlet gas volume due to water evaporation in the scrubber.
-------
ui
00
fe
O
CO
to
m
I
O
CO
CO
5
LU
6!
CO
2.5
2.0
1.5
1.0
0.5
RICKENBACKER EPA LIMIT
GUARANTEE EMISSION RATE
_L
I
90% LIME USAGE
I
0.7 0.8 0.9 1.0 1.1 1.2 1.3
LIME STOICHIOMETRY, MOLES LIME/MOLE SO2
FIGURE 5-1: The relationship between SOz
emission rates and lime
stoichiornetry.
1.4
1.5
-------
TABLE 5.2 LIME SCREENING TEST VARIABLES LEVELS - HIGH, CENTER, AND LOW POINT MEAN VALUES
AND STANDARD DEVISIGNS
VARIABLE
x,v Inlet Gas Flow Rate , SCFM
x2 1st Stage AP, in. W.C.
x_ 2nd Stage AP, in. W.C.
x. 2nd Stage Slurry Pump
4 Rate , GPM
x5 Mill Pump Rate , GPM
x,- Stoichiometry
D
x? Scrubber Volume, gal.
XQ Slurry Cone., % solids
HIGH LEVEL
MEAN
50,680
11.2
11.9
2,703
2.024
0.992
16,800
14.06
STD. DEV.
1548
2,85
0.88
318
233
0.06
1255
2.24
CENTER POINT
MEAN
42,000
8.2
8.1
2,342
1,476
0.872
15,540
10.12
STD. DEV.
620
1.3
0.74
377
150
0.16
1155
1.00
LOW LEVEL
MEAN STD. DEV.
35,420
8.3
7.25
1,903
1,258
0.61
13,011
6.21
828
2.8
2.0
383
148
0.13
1715
1.08
-------
the differences between the high and low levels were sufficient
to determine significant effects. The intermediate levels
provided information to determine the inherent variability of
the test data. The data obtained during these test is summarized
in Appendix D.
The following system responses were monitored during the
lime screening tests:
SO- removal efficiency
<£
First stage venturi liquid pickup
Second stage venturi liquid pickup
Slurry pH
Slurry alkalinity
Significant Relationships
The results of the statistical analysis of the screening
tests are presented in Table 5.3 Independent variables which
were found to have a significant relationship with a system re-
sponse have a check (y') in the appropriate row in the table.
The following relationships between independent variables and
system responses were determined.
SO- Removal Efficiency
S0_ removal efficiency is the most important response
variable studied in the screening tests. Lime/SO- stoichiometry,
as indicated in Table 5.3, is the only variable of significance
at the 95% confidence level. Table 5.4 presents S02 removal
efficiencies determined in each of the screening tests along
with lime/SO- stoichiometry, lime utilization and liquid pickup
in each of tfie R-C/Bahco scrubber venturi stages. Over a range
of SO- removal efficiencies from 36% to almost 99%, lime utilizations
of over 90% were achieved under all process conditions investigated.
First Stage Liquid Pickup
First stage liquid pickup is significantly influenced by
first stage venturi pressure drop and by second stage slurry
pumping rate, i.e., the total slurry flow rate to the scrubber.
Liquid pickup is normally controlled only by the venturi pressure
drop in the R-C/Bahco scrubber. However, the first stage liquid
pickup at high pressure drops exceeded the slurry feed rate when
the rate was set at low values. Under these conditions the
liquid pickup was limited by the slurry feed rate.
60
-------
TABLE 5.3 SCREENING TEST RESULTS
inlet Gas
Flow Rate
Independent Variables
2nd Stage
1st Stacre 2nd Stage Slurry Mill Pump
AP iP Flow Rate Rate
Stoichiometrv
Slurry
System Cone.
Volume (% Solids)
Removal
Efficiency No effect No effect No effect No effect No effect
1st Stage
Liquid
Pickup
2nd Stage
Liquid
Pickup
1st Stage
Drop
Collector
pH
Total
Alkalinity
Scrubber
Slurry
No effect
/ No effect >/ No effect
No effect No effect
No effect No effect
No effect No effect
No effect No effect
No effect
No effect
No effect No effect No effect No effect No effect
V
No effect No effect
No effect No effect
No effect No effect
No effect No effect
No effect
Note: A confidence level of 95% was selected for the analysis of the lime screening tests.
-------
TABLE 5.4 LIME SCREENING TEST DATA SUMMARY
Stat. Stoichiometry
Test Run Lb. Moles Lime/
No. No. Lb. Moles
1st Stage
Liquid
Pick-up, GPM
2nd Stage
Liquid
Pick-up, GPM
%S02
Removal
Efficiency
% Lime
Utilization
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
4
14
16
5R
8
13
12
9
14R
7R
6
15
18
3
2R
19
1
10
11
20
21
0.98
0.92
0.66
0.65
0.36
0.80
0.99
0.97
0.54
0.60
0.93
1.06
1.08
1.00
0.60
0.67
0.94
0.99
0.98
0.60
0.86
1400
750
1050
1110
1170
600
1050
720
1080
750
1080
780
1110
450
1020
360
1200
1050
1080
820
900
840
660
300
90
750
1050.
600
640
1080
600
360
450
1260
900
210
150
600
570
600
300
300
95.4
87.4
65.8
64.8
36.0
78.8
94.6
92.7
54.5
59.8
92.5
98.2
98.7
96.8
59.6
63.4
89.4
95.6
93.3
57.8
82.3
96.9
95.3
100.0
99.1
100.0
97.9
95.2
95.9
100.0
100.0
99.7
92.5
91.4
96.5
100.0
95.1
95.5
96.5
95.6
95.7
96.3
-------
Second Stage Liquid Pickup
Second stage liquid pickup is solely a function of second
stage pressure drop. The actual quantity of slurry picked up in
the second stage is lower than that for the first stage at
similar conditions. In the present series of tests, the minimum
slurry pumping rates to the second stage always exceeded second
stage liquid pickup; therefore, the pumping rate was not found to
be limiting.
Results obtained from these types of screening tests are
often very useful in analyzing a system when no significant
relation-ship is determined. The liquid pickup in both the first
and second stage is not significantly influenced by gas flow
rate, i.e. no adjustments of process variables need be made as
the gas flow rate changes to maintain adequate levels of liquid
pickup in the Venturis.
Slurry pH
First stage drop collector pH, i.e., the pH of the spent
slurry leaving the scrubber, was affected by lime/SC>2 stoichio-
metry and by the second stage venturi pressure drop. The in-
fluence of stoichiometry on pH is readily apparent. At high
stoichiometry, more lime is added to the system for a given
amount of SCu present, resulting in more available alkali and a
higher pH. At low stoichiometry, the opposite holds true and a
lower pH results.
The effect of the second stage pressure drop on first stage
drop collector pH is related to operation near the lower limits
of the systems operating range. At high slurry feed rates and
low second stage liquid pickup levels, which occurred when low
second stage pressure drops prevailed, slurry containing makeup
lime with a high pH tended to overflow from the second stage
venturi pan into the first stage drop collector. This overflow
caused an increase in pH in the first stage drop collector
slurry.
Slurry Alkalinity
Total alkalinity in the scrubber slurry, i.e. the amount of
unreacted lime or carbonate present as a fraction of total
solids, is a function of lime/SO^ stoichiometry and slurry
solids concentration. The influence of stoichiometry on alka-
linity is analogous to the effect of stoichiometry on pH noted
above, namely, higher alkalinity means higher pH. The effect of
slurry solids concentration on dissolver slurry alkalinity is
merely a dilution effect. If the same amount of lime is added to
a larger or smaller solids inventory in the scrubber as determined
from the solid concentration and the total volume of slurry in
the system, the proportion of lime increases or decreases accor-
dingly.
63
-------
LIME VERIFICATION TESTS
Since the screening tests indicated that only lime/SO_
stoichiometry controlled SO2 removal efficiency, a series of
tests was conducted to verify this finding. The effects of
stoichiometry, gas flow rate, and mill pump rate on SO- removal
were studied in more detail. In addition, the effect of low
slurry solids concentration, namely 2 wt.%, on SO., removal was
investigated. The above variables were selected for these tests
because of their possible contribution to SO,, absorption based
on conventional mass transfer theories1 which relate gas absorption
accompanied by chemical reaction to both gas phase and slurry
phase parameters.
Verification Test Results
As Figure 5-2 shows, a nearly linear relationship exists
between lime/SO- st°ichiometry and S0~ removal efficiency. Lime
utilization approached 100% in the verification tests in the
range of stoichiometry from 0.3 to 0.9 moles lime/mole S02 and
dropped gradually to 90-95% as SO2 removal approached 1001.
These results confirmed the screening tests results and indicate
that essentially any S02 removal efficiency desired can be
achieved by controlling by the lime/SO- stoichiometry.
Over the range of conditions studied, gas flow rate, mill
pump rate and slurry solids concentration had no effect on
either S02 removal or lime utilization as indicated in Table 5.5
and Figures 5-2 and 5-3.
The only negative effect that was observed when operating
at 2% solids related to the capacity of the sludge disposal
system. The thickener was designed to handle a 10% solids feed
at 60 to 100 gpm and produce a 40% solids underflow. In order
to produce a 40% solids underflow at the load conditions which
existed during the verification tests at 2% slurry solids concen-
tration a 200 gpm feed rate would have been required. In order
to accommodate this feed rate, an increase in thickener volume
of 100% would have been required. Since neither of these increased
requirements was possible, a dilute underflow in the range of 10
to 15% solids was produced.
During the lime tests reagent purity was checked on a re-^
gular basis as a part of the data package needed for this portionf
of the test program. Although Air Force specifications required
a minimum of 83 wt.% CaO, our laboratory analyses showed that
the actual levels varied from 74 to 95%^. During the December
TH D~7~V. Danckwerts, Gas Liquid Reactions, McGraw-Hill, 1970.
(2) See Appendix E for reagent and coal specifications and analyses,
64
-------
TABLE 5.5
LIME VERIFICATION TEST RESULTS
STOICHIOMETRY
SLURRY SOLIDS
RUN
NO.
23
24
25
29
30
o\
en 31
32
33
34
36
37
MOLES LIME/
MOLE SO0
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
«
572
417
565
674
698
710
901
985
746
944
930
GAS FLOW RATE
SCFM
62,
64,
65,
47,
53,
65,
65,
72,
68,
70,
74,
900
300
,150
900
200
400
600
100
700
900
400
MILL PUMP
RATE, GPM
1,
It
2,
1,
2,
1,
2,
1,
2,
2,
2,
150
750
300
600
400
550
200
700
150
200
150
CONCENTRATION % S0?
WT. % REMOVAL
17.
16.
18.
3.
2.
2.
9.
10.
10.
11.
9.
7
1
0
2
4
4
2
3
6
5
8
56.
41.
55.
65.
68.
69.
88.
96.
73.
92.
92.
7
3
9
5
4
0
2
6
2
5
1
% LIME
UTILIZATION
99.08
98.92
-
97.22
98.06
97.22
97.90
98.04
98.19
97.97
98.13
-------
o>
O
UJ
ct
C\J
O
U)
100
80
60
40
20
RAFB CODE REQUIREMENT
LIME
UTILIZATION '/
/>
100%
,
/
O SCREENING TESTS
A VERIFICATION TESTS
I
I
0.2 0.4 0.6 0.8 1.0
LIME STOICHIOMETRY, MOLES. LIME/SO2
1.2
FIGURE 5-2: SC>2 removal efficiency as a function of lime
stoichiometry.
-------
3?
Q
90
LU 'V80
2
70
60
MILL PUMP TEST RESULTS
__ ------- 0_ __._
X
GAS FLOW RATE RESULTS
I | t I
1000 1250 1500 1750 2000 2250
MILL PUMP RATE, GPM
I I I
35 42'.5 50
INLET GAS FLOW RATE, SCFM X1Q~3
FIGURE 5-3:SO2 verification test results.
67
-------
1976 tests, when low CaO content lime was utilized in the scrubber,
the reagent feed rates selected resulted in lower lime/S09 stoichi-
ometries than desired. This became evident only after analyzing
the lime - samples after the tests were performed. These variations
in reagent purity account for some of the variations observed
between target stoichiometries and the actual levels achieved in
the lime tests.
Conclusions
Lime/SO2 stoichiometry is the controlling factor in deter-
mining S02 removal efficiency. Virtually any desired S02 removal
efficiency can be achieved in the R-C/Bahco scrubber, when using
lime, simply by adjusting the lime/SO2 stoichiometry.
Lime utilization approaching 100% is achieved at stoichi-
ometric ratios up to about 0.9. At stoichiometric ratios up to
1.1, producing up to 99% SO- removal, lime utilization is above
90%.
LIMESTONE PROCESS VARIABLE TESTS
In May 1977., a series of limestone process variable tests,
modeled after the lime screening test design, were conducted. A
program of twenty-one runs, in which eight operating variables
were investigated was completed.
Limestone Test Results
The results of these process variable screening tests listed
in Table 5.6 indicate that limestone/S02 stoichiometry and second
stage slurry pumping rates control S02 removal.
The following mathematical model was developed to predict
S02 removal efficiency when limestone is used as the scrubbing
reagent:
%S02 removal = (St) °'52 X (L) °'55 (5.I)3
where:
St = stoichiometry, moles of CaC03 per mole S02 in the
inlet gas and
L = second stage slurry flow rate,
GPM.
Figure 5-4 shows predicted performance using Equation (5.1) and
observed SO0 removal data. Figure 5-5 provides for a more
direct comparison between the observed S02 removal efficiency
and the value predicted from Equation (5.1). If there were no
inherent errors in the measurements used to determine S02
removal efficiency and Equation (5.1) predicted S02 removal
T3lA General Electric multiple regression analysis program out-
lined in the G.E. Mark III Foreground User's Guide, Dec. 1973,
was used to develop this equation.
68
-------
TABLE 5.6 LIMESTONE SCREENING TEST DATA SUMMARY
Actual Operating Conditions
Run
No.
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
Limestone/SO_
Stoichiometry
0.80
1.53
1.55
1.42
1.08
1.41
1.19
1.30
1.14
0.86
0.94
1.19
0.63
0.96
0.72
1.19
0.94
1.39
0-59
0.94
1.01
2nd Stage
Slurry Rate
GPM
2280
2250
1910
2660
1890
2780
1960
2780
2050
2400
2280
2470
1875
2680
2000
2720
1750
2660
1930
2250
2280
SO, Removal
Efficiency
DuPont Analyzer
73.46
' 92.46
90.02
88.73
67.74
93.22
85.18
88.73
75.07
75.49
81.55
74.82
48.38
77.39
51.55
78.74
63.41
80.96
42.26
80.33
82.60
% Limestone
Utilization
91.79
60.34
58.18
62.32
62.41
66.21
71.74
68.10
66.06
87.92
86.52
66.81
77.38
80.76
71.93
66.02
67.16
58.44
71.20
85.14
81.42
69
-------
0 .2
1.2 1.4 1.6
LIMESTONE/SOa STOICHIOMETRY
(LB-MOLE CACO3/LB MOLE SO2)
FIGURE 5-4: SOz removal efficiency as a function
of limestone/862 stoichiometry and
slurry pumping rate.
70
-------
o
UJ
CC
C\J
O
w
Q
UJ
a
ui
DC
Q.
100
90
80
70
60
50
40
30
//
'/'''
/fa'
, / / ' / LIMESTONE MODEL
// /' £S
% S02 Removal = St° 52L° 55
where
St = moles limestone/mole SO2
L = GPM slurry
I
30 40
80
50 60 70
OBSERVED SO2 REMOVAL %
90
100
FIGURE: 5-5: A comparison of predicted and observed
removal efficiencies for limestone.
-------
Idfa? rn.L?effeCt^' a11 °f the data P^nts would fall on the
m^nr^f.? J°n "^ SinCS there are inherent errors in the
nrad^?r?nS aPProximately 15%, 4 the model, i.e. Equation (5.1)
pected 2 rem°Val efficiency within the range of accuracy ex-
_ S02 removal (efficiency and limestone utilization were found
to- improve with increased second stage slurry pumping rates.
Figure 5-4 shows the positive effect on S00 removal and limestone
utilization of increasing the slurry rate from 2000 to 2600 gpm.
Figure 5-4 also indicates that limestone utilization is
about 75% to 90% at lower SO, removal levels but decreases signifi-
cantly above 80% S02 removal: The scatter experienced in the
runs at 75-80% SO, removal is within the uncertainty limits of
the data. Limitations on the scope of this test program precluded
running verification tests to investigate this further.
Limestone Tests and Stoichiometry
The limestone variable screening test series was a duplication
of the tests performed with lime. Minor adjustments in some
variables were made to avoid system limit problems and to set
stoichiometric ratios at levels suitable for limestone.
Reagent utilization with lime was very often nearly 100%.
This resulted in a very small in-process lime inventory and
large changes in effective Stoichiometry could be made by
merely adjusting the lime feed rate. Limestone utilization,
however, ranged from 60 to 90% and resulted in a relatively
large reagent inventory.
Since the system was being tested in a transient condition,
as described in Section 3 Test Procedures, and the effective
Stoichiometry is more a function of the reagent inventory than
the reagent feed rate, it was virtually impossible to obtain
the desired levels of limestone/S02 Stoichiometry. However,
stoichiometric ratios from 0.59 to 1.55 were investigated
during these tests. The analysis of this data incorporated a
two step process. In the first step, a linear regression
analysis was performed to approximate the effect of limestone/S02
Stoichiometry on S02 removal efficiency. The remaining variations
in SO., removal efficiency were analyzed in the same manner as
the lime tests to determine the significance, if any, of the
other controlled variables relative to S02 removal.
Conclusions
Limestone/SO,, Stoichiometry and second stage slurry pumping
rate are the significant variables controlling S02 removal
efficiency.
See footnote 2, Section 4.
72
-------
A considerable excess of limestone is needed to absorb
S0~, especially at high SO- removal levels.
Limestone can be used to meet the requirements for SO2
removal at RAFB and other similar coal-fired installations.
LIME VS. LIMESTONE
Both lime and limestone scrubbing reagents have been
demonstrated to be very effective in controlling the SO*
emissions from the boilers at RAFB. Lime is capable of removal
efficiencies in excess of 98% with reagent utilizations approaching
100%. Limestone can remove as much as 93% of the inlet S02.
However, limestone utilization drops below 75% at S02 removal
efficiencies above approximately 80%.
Reagent Economics
At RAFB, limestone is more economical to use than lime
despite the fact that more than twice as much limestone is
needed to attain the same S02 emission rate. The chemical
equations illustrating lime and limestone stoichiometry and
weight utilization are:
Lime absorption of SO^
CaO + H2° = Ca(OH)2 (5.2)
1 g-mole CaO = 56g.
Ca(OH)2 + S02 = CaS03 + H20 (5.3)
1 g-mole S02 = 64g.
Limestone absorption of S02
CaC03 + SO2 = CaSO3 (5.4)
1 g-mole CaC03 = 100 g.
1 g-mole SO- = 64 g.
Therefore, to absorb one gram-mole of SO- (64 grams), 56
g. of lime or 100 g. of limestone are required if a stoichiometric
equivalent of either reagent is used. Since lime was completely
utilized compared to an average of 75% for limestone, more
representative numbers are 62 grams5 of lime used versus 133
TB1These values include allowances for typical impurities found
in these reagents.
73
-------
grams of limestone. The price of limestone delivered to the
RAFB during 1977 was $12.72/ton compared to $40.35 for lime.
This large price differential gives limestone the economic
advantage, for every $100 spent for limestone $141.20 must
be spent for an equivalent amount of lime.
In addition limestone is not hygroscopic and need not
be slaked, thus eliminating the need for a complex slaking
device. Finally, limestone is far less likely to cause
injuries to operating or maintenance personnel since it does
not exhibit the potentially damaging caustic properties
inherent in lime.
Limestone and Oxidation
A large increase in the ratio of sulfate to sulfite in
the scrubber slurry was observed when the shift from lime to
limestone was made during the test program. Table 5.7 shows
an average calcium sulfate (CaSO. 2H20) and calcium sulfite
(CaSCU ^H-O) content of 36% and 57%, respectively, when
lime was being used. The limestone slurry is' more fully
oxidized and contained 75% sulfate and less than 1% sulfite.
The comparison of average lime and limestone slurry analyses
during similar boiler load periods listed in Table 5.7
indicates that this oxidation trend is probably attributable
to the effect of the reagent used on the system's average pH
as discussed below since all other operating conditions were
essentially the same.
Dissolver Slurry pH
As illustrated in Figure 5-6, dissolver slurry pH
measurements taken during the limestone tests were signi-
ficantly different from those taken during the lime tests.
Although the limestone/S02 stoichiometry was varied over a
range of 0.59 to 1.53, the accompanying slurry pH variation
was only 4.9 to 6.2. In comparison, the lime slurry pH was
much more sensitive to stoichiometric changes. The dissolver
slurry pH varied from 4.3 to 9.6 at stoichiometric levels
from 0.36 to 1.08.
In the lime test series, the relationship between the
dissolver slurry pH and stoichiometry was used as a means to
estimate stoichiometry during the test program.
74
-------
TABLE 5.7 LIME AND LIMESTONE SLURRY ANALYSES
Slurry Solids
CaS04 2H20
CaS03 %H20
CaC03
MgC03 -
Acid Insoluables
TOTAL
Lime Slurry
May 1976
wt %
33.4
54.5
3.7
4.6
96.2
Limestone Slurry
May 1977
wt %
77.5
1.0
17.3
0.8
3.4
100.0
75
-------
a\
9.0
8.0
I
D.
DC
111
>
o
in
w
Q
7.0
6.0
5.0
OLIME TESTS
OLIMESTONE TESTS
O w
o
o 9>°.0
O
o
o
o
o
o o
4.0
I
0.2 0.4
0.6 0.8 1.0
STOICHIOMETRY
1.2
1.4 1 6
FIGURE 5-6: The effect of lime and limestone stoichiornetry on
dissolver pH.
-------
CONCLUSIONS
o Lime is a more efficient reagent than limestone for S02
removal in the R-C/Bahco scrubber.
o Limestone is a more economical reagent than lime because of
the large reagent price differential.
o Limestone scrubber solids contain substantially higher
sulfate levels than those from lime scrubbing. This is due to
lower system pH levels normally associated with limestone.
Note: See Section 7 page 117 regarding the relation between the
use of limestone and scrubber solids sulfate levels.
77
-------
SECTION 6
PARTICULATE REMOVAL TESTS
INITIAL PARTICULATE REMOVAL TESTS
Initial particulate removal tests on the R-C/Bahco scrubber
performed in March, April and May of 1976 revealed the presence
of unexpected amounts of.particulate in the stack gas. The
average particulate emission rate for these tests, listed in
Table 6.1,-was 0.-23 Ibs./MM Btu.l Overall particulate removal
averaged 93 to 94% for these tests.2 The Ohio emission standards
require an overall removal efficiency of about 96% at a particulate
inlet loading of 1.5 grains per SCFD to achieve an emission
rate of 0.16 lbs./MM Btu.
The R-C/Bahco system at RAFB was designed with extra fan
capacity to carry out this test program. This extra capacity
was utilized to help reduce particle emissions levels. Venturi
pressure drops were increased to nearly double the design value
of 7 in. w.c. to accomplish this end. Data from these preliminary
tests are plotted in Figure 6 -1 to show the effect of the
combined pressure drop of the two Venturis in the R-C/Bahco
scrubber on particulate emissions. Below approximately 18 in.
H~0 total pressure drop, particulate emissions increased rapidly.
Since outlet particulate loadings were higher than anticipa-
ted, tests were run and the results analyzed to determine the
cause. Particulate size distribution data from several sources,
listed in Table 6.2, were collected and analyzed.
The particle size data shown in Figure 6-2 was derived
from Pratt-Daniel data supplied for the original installation
of the generators. This particle size distribution is typical
for stoker fired coal burning equipment operating without any
particulate controls. This information was used to estimate
the particle size distribution and loading for the R-C/Bahco
scrubber at RAFB.
THAdditional particulate emission data is listed in Appendix H«
(2) Including particulate removed by the mechanical collectors.
78
-------
Date
3/18/76
3/30/76
3/30/76
4/8/76
4/29/76
4/30/76
5/20/76
Particle
Loading,
gr/SCF
Inlet Outlet
0.178
0.075
0.059
0.036
0.072
0.038
0.024
Gas Flow
Rate,
SCFM*
42,800
37,500
37,500
50,500
45,500
47,200
54,200
Boiler Firing
Rate,
MM Btu/hr.
128.2
132.6
132.6
115.2**
94.8
94.8
57.7
TABLE 6.1 PARTICULATE EMISSION RATES
Particulate
Total Pressure Emissions Rate,
Drop,in H90 LBS/MM Btu
15 0.51
19 0.18
19 0.14
27 0.14
15 0.30
18 0.16
23 0.19
* Flue gas flow rates were based on fan motor current readings and fan pressure differentials
** The total firing rate was 132.3 MM Btu/hr.; however, 13% of the flue oas was bvoassed.
TABLE 6.2 INLET PARTICLE SIZE DISTRIBUTIONS
Flue Gas Stream
Location Source Remarks
Boiler Outlet Pratt-Daniel' Boilers Manufacturer's
Data
Scrubber Inlet Design Model Calculated For Two
Mechanical Collectors
in Series
Scrubber Inlet R-C Particulate
Tests May 1976 Samples were collected
in Andersen Impactors
-------
i
*
LU
o
CO
CO
0.60
Q.5O
0.40
0.30
O
CT 0.20
0.10
I
I
GUARANTEE EMISSION RATE
I
I
I
10 14 18 22 26- 30
TOTAL SCRUBBER PRESSURE DROP - IN. W.C.
FIGURE 6-1: The effect of scrubber pressure
drop on particulate emission rates.
80
-------
60.0
40.0
20.0
10.0
CO an
2 8.0
O
DC
O 6.0
111
a «
UJ
tr
2.0
1.0
0.8
0.6
0.4
CH PRATT-DANIEL DATA
O TWO MECHANICAL COLLECTORS IN SERIES
A SCRUBBER INLET SAMPLES, MAY 1976
0.2l
JL
±
20 40 60 80
CUMULATIVE WEIGHT, %
90 95
FIGURE 6-2: A comparison of various particle
size distributions for RAFB.
81
-------
The amount of soot present in the flue gas at RAFB is sub-
stantially higher than normal for stoker fired generators of
this type. The USAP has undertaken an extensive program to
upgrade the heat plant at RAFB. Data obtained during this test
program contributed substantially to information used by the
USAF to plan this upgrading program.
The following items have been undertaken or completed:
o Installation of a new 60 Btu/hr. generator to replace two
old units.
o Replacement of hot water distribution piping.
o Installation of flue gas oxygen monitoring equipment.
o Repair of firing air distribution equipment and fire box
pressure controls in the generators.
o Rebuilding mechanical collectors and I.D. fans on the
generators.
o Replacement of burned out ledge plates, which regulate
combustion air flow around the grates, and repair of the
traveling grates.
FRACTIONAL EFFICIENCY TESTS
Tests were run at RAFB over a wide range of operating conditions
to determine fractional particulate removal efficiencies. The
effect of gas flow rates, and other process variables were
determined for a number of particulate size fractions. Andersen
Impactors were used during these tests. Table 6.3 summarizes
the results of these tests. Major system variables and the
particle diameters at which 50% and 90% collection efficiencies
were obtained are also listed.
The particle diameter at which 90% collection efficiency was
obtained varied from 0.67 microns to 1.24 microns. Increasing
the total scrubber pressure drop decreased the particle diameter
at which 90% collection was observed, i.e. collection efficiency
increased. Figure 6-3 presents the effect of the combined
pressure drop of the first and second stages of the scrubber on
the particle size at which 90% collection was achieved. Particulate
collection efficiency in this range of sizes appears to be
unaffected by the gas flow rate.
For 50% particulate collection efficiency, the diameter
ranged from 0.42 to 0.72 microns and was related to total
scrubber pressure drop and gas rate. Figure 6.4 shows that
increasing the pressure drop and increasing the gas rate acted
to increase particulate collection as evidenced by the decrease
in the diameter observed for 50% collection efficiency. Pressure
drops beyond about 16 in. W.C. had no effect on improving
particulate collection efficiency.
82
-------
TABLE 6.3 FRACTIONAL EFFICIENCY TEST RESULTS
Test
No.
1
2R
4
10
11
oo 12R
OJ
15
18R
2
7
12
18
Note:
Inlet Gas
Flow
SCFMxlO"3
40.5
35.3
53.1
41.5
42.8
53.7
30.3
33.2
32.9
34.2
55.3
33.7 - 41.8
1st Stage
AP, in.W.C.
7.0
4.5
11.0
9.5
9.5
5.3
12.7
4.5
12.0
9.5
8.0
13.0
Gas flow rates were based
2nd Stage
AP, in W.C.
9.0
4.8
13.0
8.0
8.5
5.3
6.6
13.0
9.0
10.7
8.0
12.0
on booster
Flue Gas
Temp, °F
440
428
' 480
473
478
353
492
435
478
485
468
448
fan motor
1st Stage 2nd Stage
Liquid Pickup Liquid Pickup
GPM GPM
1200
165
1400
1050
1100
255
780
220
1350
400
1050
1050
current
600
90
840
540
600
105
450
810
810
480
600
1140
readings and
50% Part.
Diameter
Microns
0.66
0.67
0.43
0.58
0.72
0.70
0.72
0.67
0.55
0.59
0.44
0.66
90% Part
Diameter
Microns
1.00
0.98
0.72
0.92
0.97
1.24
1.06
0.96
0.78
0.82
0.67
0.94
differential pressures,
-------
00
1.2
1.1
O
g 1.0
0.9
0.8
0.7
O
10
O
GAS FLOW
O = 50,000 SCFM
O = 42,000 SCFM
£ =35,000 SCFM
12
14
16
18
20
22
TOTAL SCRUBBER PRESSURE DROP IN. W.C.
24
FIGURE 6-3: The effect of total scrubber pressure drop on the
cut off diameter for 90% collection efficiency.
-------
0.9
00
Ul
0.8
0.7
I
y 0.6
8
o.
Q
0.5
0.4
= 50,000 SCFM
-42,000 SCFM
=35,000 SCFM
0.3
J_
10
12 14 16, 18 20
TOTAL SCRUBBER PRESSURE DROP IN. W.C.
22
24
FIGURE 6-4: The effect of total scrubber pressure
drop and gas rate on the cut off
diameter for 50% collection efficiency.
-------
Flue gas from each generator passes through two mechanical
collectors before entering the scrubber. The first mechanical
collector is mounted on each hot water generator upstream from
its I.D. Fan. The other is located upstream from the booster
fan in the main flue.
The second particle size distribution illustrated in
Figure 6-2 was derived from the Pratt-Daniel data by calculating
the particulate collected from the original distribution for
each of the mechanical collectors.3 Average flue gas rates
resulting in a pressure drop of 2 in. w.c. for each collector
and a fly ash density of 2.0 gm/cm3 were selected for this
calculation.
Based on a 1.5 grains per SCFD loading from the generators,
a grain loading of 0.1 grains per SCFD entering the scrubber
was estimated from the above analysis.
The particulate tests conducted in May, 1976 , to gain
insight into the particulate emissions problem, revealed the
presence of considerably more fine particulate than anticipated.
The data from these tests is plotted as the third distribution.
Figure 6.2 shows that 70 to 80 wt. % of the particulate
entering the scrubber is less than 1 micron as opposed to an
expected 8 to 10 wt.%. Samples of this particulate appear very
sooty on visual inspection. In addition, the material has a
low specific gravity (^ 1.3) and a high loss on ignition (^ 35%).
A typical grain loading for the flue gas at RAFB is 1.0
grain per SCFD. The first mechanical collector should have
reduced the grain loading to 0.14 grains per SCFD and the
second to 0.07 grains per SCFD before the particulate matter
entered the scrubber. The scrubber must remove half of the
remaining particulate in order to meet code requirements.
Based on size distribution tests and chemical analyses, it is
estimated that 10-20% or 0.1-0.2 grains per SCFD of the total
particulate matter is soot.
A soot level of 0.1 grains per SCFD, if uncollected, corresponds
to an emission rate of about 0.5 Ibs. particulate/MM Btu. The
test results indicate that soot levels well over three times
the allowable emission rate of 0.16 Ibs./MM Btu were present.
Actual removal efficiencies at loadings of approximately 0.15
to 0.25 grains per SCFD were in the 40 to 80% range, resulting
in an overall particulate removal efficiency of 93-94%. This
data indicates the R-C/ Bahco scrubber performed very well on
material for which it was not designed to collect.
T3~]See" Appendix H for particulate removal efficiency data for the
mechanical collectors.
(4) The individual test data is also located in Appendix H.
86
-------
As indicated in Figure 6-5, collection efficiency for
particles above 3 microns is essentially complete. In addition,
Figure 6-5 shows the particle size range at which 50, 90 and 98%
particulate collection was observed during this portion of the
test program.
PARTICULATE COLLECTION AT SYSTEM LIMITS
While analyzing the test results it became apparent that a
few tests were conducted outside the scrubber operating limits
described in Table 4.1 above for gas flow rate and venturi pressure
drops.
Two phenomena which resulted in decreased particulate
collection were observed. The results of these tests, runs 2R,
12R and 7, are presented in Figure 6-6 and Figure 6-7. Tests 2R
and 12R exhibited a phenomenon called slurry droplet "entrainment"
and test 7 illustrates what is termed gas "bypassing". Pertinent
gas flow and pressure drop data for these tests is listed in
Table 6.3.
Entrainment
The inlet and outlet particle size distribution curves in
Figure 6-6 reveal that particles as small as about 2 to 3 microns
are removed effectively by the scrubber. In run 2R, however, for
particles above 3 microns, the apparent collection efficiency
dropped off drastically. Collection decreased to the point where
very little collection occurred in the 9 to 12 micron range,
instead of the normal outlet level indicating essentially 100%
removal. Run 2R was conducted at a second stage venturi pressure
drop of 4.8 in. w.c. and a gas rate of 35,300 SCFM. The most
likely explanation for this behavior is the following: at these
operating conditions fine slurry droplets are generated in the
second stage venturi which are not removed in the second stage
mist eliminator. This phenomenon was observed at both high and
low gas flow rates, but occurred only when operating at very low
pressure venturi drops, i.e., 4 to 5 in. W.C.
There are further indications which point to slurry droplet
carryover at these operating conditions. The amount of carryover
observed in Test 2R is high enough to raise the emission of
particles in the 9-12 micron range almost as high as they were in
the inlet gas. Pressure drops in Test 12R were similar to those
in Run 2R (4.8 in. w.c. vs. 5.3 in. w.c.), but the scrubber gas
rate was higher, i.e. 53,700 SCFM. In Run 12R, some carryover
occurred but not nearly as much as experienced at the lower gas
rate. This difference may be due to the ability of the second
stage mist eliminator to collect droplets of slurry in the 10
micron range more effectively at the higher gas rate experienced
in Run 12R.
87
-------
oo
00
go
80-
fc"
o
u.
Hi 70
§
O 60
O
50
40
0.2
0.4 0.6 0.8 1.0 2.0
PARTICLE DIAMETER, MICRONS
FIGURE 6-5: Particulate collection efficiency as
related to particle size at RAFB.
4.0
6.0
-------
oo
12
10
d
N
CO
LL
O
LU
g 6
H-
oc
o
I
o
OUTLET WITH
REENTRAINMENT
RUN 12R
\
INLET
OUTLET WITH
REENTRAINMENT
RUN2R
EXPECTED OUTLET
DISTRIBUTION
\
8
10
12
14
PARTICLE DIAMETER D, MICRONS
FIGURE 6-6: Examples of reentrainment from the
R-C/Bahco scrubber at RAFB.
-------
Bypassing
The second phenomenon observed during Run 7 is called
bypassing, illustrated in Figure 6-7. Particulate collection
is uniformly poor over the entire range of particle sizes
observed. It is as if part of the gas stream was bypassing
around the Venturis. This condition is characterized by pulsating
flow through the system and occurs when relatively high pressure
drop in either stage is coupled with low slurry flow to the
scrubber. When the system is operating in this manner, the
current drawn by the fan motor varies over a range of 5 amps at
a frequency near one cycle per second. In addition, liquid
pickup in the Venturis is less than expected for the observed
combination of gas flow and pressure drop. It appears that the
flue gas picks up more slurry than is being pumped to the
scrubber thereby dropping the slurry level and back pressure,
allowing a surge of flue gas through the venturi. As the
slurry begins to catch up, the pressure drop increases and the
flue gas flow decreases. However, since the average slurry
rate is too low, the process is repeated until conditions are
changed to eliminate the imbalance between slurry flow and
venturi liquid pickup.
Table 6.4 presents the liquid pickup rates measured on the
first and second venturi stages of the R-C/Bahco scrubber as a
function of pressure drop, gas rate and slurry circulation
rate. These results were obtained over a pressure drop/stage
range up to about 15 in. H^O and gas flow rates from 33,000
SCFM to 55,000 SCFM. Figure 6-8 illustrates the normal amounts
of liquid pickup in each venturi and the results from Run 7R
where bypassing occurred. For the first stage at 9.5 in. H-O
pressure drop, a slurry pickup rate of only 400 GPM was obtained
in Run 7R compared to the expected rate of 1000 - 1400 GPM. In
the second stage at 11.2 in. H2O pressure drop, a 500 GPM
pickup rate was observed compared to an expected 900-1,200 GPM.
It is important to note that this condition only occurs
when the slurry flow to the scrubber is intentionally reduced
to less than 50% of the normal flow rate of 2600 GPM and the
venturi pressure drops are increased to more than 50% above
normal operating levels of 6-8 in. w.c.
PARTICULATE PERFORMANCE MODEL
The R-C/Bahco scrubber treats flue gas using two Venturis
in series. The most commonly used collection mechanism for
describing particulate removal in the size range observed at
RAFB is inertial impaction. This is the prevailing mechanism
for collection of particles above 0.5 microns in diameter.
15)W7 Strauss, Industrial Gas Cleaning, Pg. 215, Pergamon Press
1966.
90
-------
LU
N
_
O
LU
g
i-
cc
10
8
INLET-RUN 7
O
EXPECTED OUTLET DISTRIBUTION
OUTLET-RUN 7
4 6 8 10
PARTICLE DIAMETER D, MICRONS
12
FIGURE 6-7: An example of bypassing from the
R-C Bahco scrubber at RAFB.
14
-------
TABLE 6.4 VENTURI LIQUID PICKUP RATES
Inlet Second Stage
Gas Flow Total Pressure Drop,in.w.c. Slurry Pumping Liquid Pickup,
Test NO. SCFM
1
2R
4
7
10
11
12R
15
18R
2
12
13
40,500
35,300
53,100
34,200
41,500
42,800
53,700
30,300
33,200
32,900
55,300
33,700-41
First Stage
7.0
4.5
11.0
9.5
9.5
9.5
5.3
12.7
4.5
12.0
3.0
,300 13.0
Second Stage Rate, GPM
9.0
4.3
13.0
10.7
3.0
3.5
5.3
6.6
13.0
9.0
3.0
12.0
2730
3000
3100
1750
2230
2700
3000
2400
3000
3000
2800
2400
1st Staere
1200
165
1400
400
1050
1100
255
780
220
1350
1050
1050
2nd Staae
600
90
340
480
540
600
105
450
810
810
600
1140
92
-------
1600
NORMAL RANGE
1400
1200
a.
C3
a: 1000
x.
o
Q.
Q
g 800
600
400
200
2ND STAGE
2ND STAGE
RUN 7
LIQUID PICKUP
I
I
I
6 8 10
VENTURI PRESSURE DROP IN. W.C.
12
14
FIGURE 6-8: The effect of venturi pressure
drop on liquid pickup.
93
-------
Inertial impaction occurs when the inertia of a particle causes
it to continue to move toward and collide with the collecting
medium, slurry droplets, as the gas stream changes direction in
flowing around the collecting medium. According to inertial
impaction theory, collection efficiency can be related to
scrubber variables and particle size by using the following
relationship:
P = 1 - n - exp. (a(L/G) ₯} (6.1)
where P = penetration or the fraction of particulate not collected
r\ = the fraction of particulate collected
a = an empirical constant
L/G = stage liquid slurry/gas ratio, GPM/1000 CFM
₯ = inertial impaction parameter defined below:
y = c1 (p - p ) d v
*P *V P o
18 y D (6.2)
In equation (6.2), C1 = Cunningham correction factor, dimensionless
p = Particle density, gm/cm
p = Gas Density, gm/cm
d = Particle diameter, cm
P
v = Gas velocity, cm/sec
o
y = Gas- viscosity, Pas.
D = Droplet diameter, cm
For the purposes of this analysis several simplifications
in the inertial impaction model, i.e. Equation (6.1), were 7
made. First, the collector droplet size, i-l, was assumed to be
inversely proportional to gas velocity at average temperature
conditions in the venturi6:
1 492 _
D = v~ = G~ - (6.3)
O
where G = Gas flow rate, SCFM
ThTs is based on a simplification of the equation developed by
Nukiyama and Takasawa, Trans. Soc. Mech. Eng. (Japan), 4, (14)
86, (1938).
94
-------
T = Average temperature, R
The average gas temperature was estimated using a root-mean-
square absolute temperature:
T = (T^ . , ^ + 460) (T , + 460) (6.4)
av v Gas inlet slurry
This particular mean was chosen to emphasize the effects of gas
velocity before gas cooling occurred in the scrubber. The
slurry was used to represent the outlet gas temperature and was
in the temperature range of 125°F for all runs. Since all the
other factors in Equation (6.1) are constant for a given particle
size and temperature, the inertial impaction parameter can be
approximated by:
ty = a, d G T (6.5)
1 p av
By combining Equations (6.1) and (6.5), the particulate penetration
for the first scrubber stage can be determined.
p
1st stage = exp. {a, (L,G, ) d G T } (6.6)
j. J. -L p av
or more simply,
Plst stage = exp {a, L, d T } (6.7)
-L .L p clV
The penetration expression for the second or upper stage
is analogous to the first stage but simpler in form since the
inlet gas temperature in the second stage is essentially the
same as the outlet or slurry temperature. The average temperature
in Equation (6.5) can be incorporated into the constant term:
(6.8)
Thus, the second stage penetration can be expressed as,
P = exP {a L d} (6.9)
2nd stage 2 2 p
For the R-C/Bahco scrubber with two stages in series, P , the
overall particulate penetration is the product of the pinetration
for each stage,
95
-------
Po = Plst P2ndd = (exP {al Ll d Tav» (exp {a2L2d} (6.10)
Analysis of the Test Results
A regression analysis of the fractional efficiency test
results was performed using the penetration model developed in
Equation (6.10). The penetration model coefficients a., and a7
for the first and second stages are given in Table 6.5. Average
particle size and collection efficiency test results were used
for the regression analysis.
The results predicted from the Equation (6.10) and the
values determined in the test program are compared in Figure
6.9. Figure 6.10 compares the observed and predicted collection
efficiency for the particle size ranges selected for analysis.
An analysis of the results in Table 6.5 indicates that
collection of particles above and below one micron occur in
different stages of the scrubber. This conclusion is supported
in part by the relatively small changes which occurred in
correlation coefficients when single stage models were used.
It appears that particles larger than 1 micron are collected
in the first stage of the scrubber. For example, Table 6.6
shows that in the 2.0-5.0 micron range the predicted collection
efficiency changes from 99.8% to 98.4% when the second stage is
eliminated from the model. As a further indication of the
minimal significance of the second stage on particulate collection
above 1 micron, one should note the effect of eliminating the
second stage on the first stage model coefficient and the
correlation coefficient. The first stage coefficient changes
somewhat from -0.232 x 10~5 to -0.172 x 10 but more significantly
the correlation coefficient decreases by only 1% from 0.97 to
0.96.
A similar situation occurs for the second stage for fine
particulate collection. For 0.3 - 0.5 micron particles, the
second stage is the primary collector. Typical values for
collection efficiency are 50.9% for a one stage model and 43.8%
for a two stage model. Thus, each scrubber stage functions to
collect a different portion of the total particulate distribution.
The comparison presented in Figure 6-9 and 6-10 indicates
that there is some variability between the observed and predicted
particulate penetration.
Two factors which contributed to scatter in the test
results and complicated particulate testing were:
96
-------
TABLE 6.5 PENETRATION MODEL COEFFICIENTS (1)
Particle Size
Range,
No. of Stages
In
First Stage ,
Model Coefficient (3)
al
Second Stage
Model Coefficient Correlation
a Coefficient
micro
0.3 -
0.3 -
0.5 -
Oc
* D
1.0 -
1A
. u
2.0 -
2n
. u
hs
0
0
1
2
5
t
.5
.5
.0
.0
.0
(2)
(0
(0
(0
/ n
\ u
(1
/ 1
V -L
(3
/ *3
.4)
.4)
.7)
7 \
/ J
.4)
)
-2)
o\
4)
Model
2
r
2
.
2
i
-L
2
0.
0.
-0
n
u
-0
f\
u
-6
376 X 10
519 X 10~6
.210 X 10~5
ORT V 1 A~~
. /Di A J.U
.232 X 10~5
-5
170 Y TA
, i. 1 £, A -LU
-0.269
-0.221
-0.229
A 1 Q R
U . J.OD
-0.478
0.710
£.
-2
X 10
-2
X 10
X 10~2
Xi n~ *
-LU
X 10~3
x io'3(2)
0.
0.
0.
.
0.
.
0.
.
91
91
83
Q T
O J
94
Q A
7ft
97
O £
y o
(1) The regression analyses in which these coefficients were determined are located
in Appendix H.
(2) Average particle size used to determine penetration model coefficients shown.
in parentheses.
(3) These positive correlation coefficients are strictly empirical since they
must posses negative values to *>ave physical significance
-------
100
40
10.0
4.0
ill
Q.
Q
LU
Hi
§ '<
0.4
0.1
PARTICLE SIZE RANGE
O 0.3-0.5/i
a 0.5-1 .o/u,
A 1.0-2.0ju,
O 2.0-5.0 M.
0.1
0.4 1.0 4.0 10.0
PREDICTED PENETRATION %
40 SO
FIGURE 6-9: A comparison of observed and
predicted participate penetration.
98
-------
100
80
yj
o
Q
b
o
40
20
OBSERVED EFFICIENCIES
PREDICTED EFFICIENCIES
0.2
0.4 0.6 0.8 1.0 2.0
PARTICLE SIZE, MICRONS
FIGURE 6-10: A comparison of observed and
predicted particulate collection
efficiencies for the R-C/Bahco
scrubber at RAFB.
4.0
-------
1. The variability of boiler operation which resulted in
fluctuating flue gas rates and compositions to the scrubber
during the tests.
2. Changes in the particulate characteristics when fine soot
was generated along with fly ash.
However, the results indicate that the simplified inertial
impaction model used in this analysis does adequately represent
the observed particulate collection.
TABLE 6.6 A COMPARISON OF ONE AND TWO STAGE MODELS
Particle Size
Range, Microns
One Stage Model* Two Stage Model*
Collection Efficiency % Collection Efficiency
0.3 - 0.5 (0.4)
0.5 - 1.0 (0.7)
1.0 - 2.0 (1.4)
2.0 - 5.0 (3.2)
50.7
64.5
93.0
98.4
43.8
63.5
92.6
99.8
*These values were calculated using the following values for
system variables:
L = 1100 GPM
L_ = 800 GPM
= 688 °R
= 350F)
Since large sized particles are collected efficiently in the
first stage, very few large particles remain to be collected in the
second stage. For certain applications, where high combustion
efficiency in the boiler can be maintained, it may not be necessary
to use two stages for particulate collection since the generation
of fine particulate would be minimal.In addition to collecting
large particles, the first stage may also help condition fine
particulate in the flue gas to enhance collection in the second
stage. Gas cooling and humidification processes occurring in the
first scrubber stage may condition the fine particles via nucleation
and growth mechanisms to make them more susceptable to capture in
the second stage.
100
-------
CONCLUSIONS
The particulate removal efficiency of the R-C/Bahco
scrubber is comparable to that of low energy venturi
scrubbers for particles larger than 1 micron and appears
to be better for particles smaller than 1 micron.
In an R-C/Bahco scrubber the second stage is the
primary collector for fine particles.
Slurry carryover and gas bypassing limit particulate
collection in an R-C/Bahco scrubber when operating
beyond the systems gas handling and venturi pressure
drop limits.
Particulate emissions from stoker fired coal burning
equipment can be reduced to acceptable levels if ex-
cessive soot formation does not occur.
101
-------
SECTION 7
SCRUBBER SLUDGE CHARACTERIZATION TESTS
A series of scrubber sludge characterization tests were
carried out in the R-C laboratories to accomplish the following
objectives:
o Determine scrubber sludge dewatering characteristics
o Evaluate transportability of dewatered sludge
o Determine physical/structural properties of dewatered
sludge
c Measure sludge leachate for environmental acceptability
The laboratory tests outlined below were performed on
samples collected at RAFB during the lime and limestone testing
described in Section 5 of this report.
SLURRY DEWATERING
o Settling
o Centrifugation
o Vacuum Filtration
TRANSPORTABILITY
0 Truckability
o Angle of Repose
o Slump
o Slide
102
-------
PHYSICAL/STRUCTURAL
o California Bearing Ratio
o Unconfined Compressive Strength
LEACHATE
o Trace Metal Analysis
o Total Dissolved Solids, Sulfate and Chloride
o Chemical Oxygen Demand
SLURRY DEWATERING
Settling Tests
Six settling tests were run on lime and limestone slurry
samples to determine rates and final sludge solids. A commonly
used flocculant, Betz 1100, was tested at 2 and 5 ppm to study
its effectiveness. Slurries containing 16 to 17 wt. % solids
were used for settling tests. These tests were performed in
2,000 ml graduated cylinders.
Limestone slurries which contained substantial amounts of
gypsum settled much more rapidly than lime slurries as indicated
in Figure 7-1. The flocculant had no effect on lime slurry
settling rates but improved settling substantially on the gypsum
limestone slurries. As Table 7.1 shows, the settled sludge
concentration was not affected by the addition of flocculant for
either type of slurry. However, limestone slurries with their
high gypsum content produced a settled layer with a higher solids
content, 58 wt.% vs. 43-45 wt.% for lime.
Centrifuge Tests
Laboratory tests indicated that centrifugation produces a
denser cake from limestone slurries which contained substantial
amounts of gypsum than from lime slurries. As Figure 7-2 shows,
this effect is more pronounced at higher slurry feed concentrations.
With a 37-38 wt.% solids feed, limestone slurry dewatered to
65-67 wt.% solids while the lime slurry reached only 56-58 wt.%.
Centrifugation produced a 50-55% solids concentration for both
slurries when the feed concentration was reduced to 20-25 wt.%.
Figure 7-2 also reveals that the slurry solids reached nearly
maximum compaction within the first five minutes. Continued
operation at 800 g's for up to twenty minutes increased solids
content by only 2-3 wt%. Lime and limestone centrifuge test
results are summarized in Table 7.2.
103
-------
2000
0 PPM FLOCCULANT
0 PPM FLOCCULANT
2 PPM
5 PPM
lil
IU
O 1000
cc
LU
2 750
500 -
250 -
LIMESTONE SLURRY
LIME SLURRY
12
18
24
30
36
42
48
SETTLING TIME - HOURS
FIGURE 7-1: A comparison of lime and limestone
slurry settling rates.
-------
TABLE 7.1 SETTLING TEST RESULTS
Test No.
L2
LS, LS.
LS.
Scrubber
Reagent
Lime Lime Lime
Lime- Lime- Lime-
stone stone stone
Floe Type
Added
None
Betz Betz
1100 1100
Betz Betz
None 1100 1100
Floe Cone.
ppm
Feed
Wt.% Solids
16.15 16.15 16.15 16.66 16.66 16.66
Settled
Sludge
Wt. % Solids 43.31 45.32 44.03 58.35 58.86 57.50
Settling Rate*
Ibs/ft2 day 20.1 22.7 22.3 164.3 350.0 578.1
* at an underflow concentration of 35% solids
105
-------
70
65
D RUNS U-La
O RUNS LSi - LS4
O RUNSLSs-LSs
37 WT. % LIMESTONE SLURRY FEED
O
G\
CO
Q
O
CO
LU
o
60
55
50
38 WT. % LIME SLURRY FEED
23 WT.% LIMESTONE SLURRY FEED
10
CENTRIFUGE OPERATION
TIME-MINUTES
15
20
FIGURE 7-2: The effect of operating time and slurry feed
concentration on centrifuge cake density.
-------
TABLE 7.2 CENTRIFUGE TEST RESULTS
Feed Centrifuge Cake
Run No. Wt.% Solids Time (min) Wt.% Solids
L-l 26.27 5 50.63
L-2 26.27 10 50.16
L-3 26.27 15 52.04
L-4 26.27 20 50.44
L-5 38.35 5 55.66
L-6 38.35 10 56.95
L-7 38.35 15 57.28
L-8 38.35 20 58.68
LS-1 22.92 5 51.04
LS-2 22.92 10 53.19
LS-3 22.92 15 53.56
LS-4 22.92 20 54.31
LS-5 37.40 5 65.18
LS-6 37.40 10 66.44
LS-7 37.40 15 66.82
LS-8 37.40 20 67.41
107
-------
Filter Leaf Tests
Filter leaf tests show that limestone slurry filtration
rates are significantly lower than lime slurry rates.
Figure 7-3 and Table 7.3 present filtration rates for the five
composite filter leaf tests performed. Filtration rates measured
for lime and limestone slurries ranged from 43 to 150 Ibs of
solids per hour for each square foot of filter cloth. The tests
were conducted on a 0.1 ft* filter with an 853 Eimco polypropylene
filter cloth (59 x 38 thread count).
Figure 7-3 also shows that the filtration rate increases
with increasing slurry solids concentration and decreases with
increasing form time. As shown in Table 7.3, filtration produced
a more concentrated filter cake from limestone slurries (74 wt.%)
which contained substantial amounts of gypsum than from lime
slurries (58%). Form filter time had no effect on the filter
cake solids content for either type of slurry in our tests using
a vacuum of 26.0 in. Hg.
Lime vs. Limestone Dewatering
Photomicrographs, Figures 7-4 and 7-5, taken of the slurry
samples listed in Table 7.4 show that the lime slurry crystals
which were predominately calcium sulfite tend to be small and
needlelike, whereas the limestone slurry crystals which were
predominately gypsum (calcium sulfate) are larger and more block-
like. These large crystals in the limestone slurry promote more
rapid dewatering in most instances and result in higher solids
concentrations in dewatered products.
108
-------
500
200
CO
CD
LU
CC
100
<
cr
CC
£
50
TEST NO. 1 -32 WT.% LIME SLURRY
0 TEST NO. 2-24 WT. % LIME SLURRY
A TEST NO. 3-41 WT. % LIME SLURRY
O TEST NO. 4-28 WT.% LIMESTONE SLURRY
D TEST NO. 5-37 WT. % LIMESTONE SLURRY
20
0.2
0.5 1.0
FORM TIME (MINUTES)
2.0
FIGURE 7-3: Filtration rate as a function of form
time and slurry concentration.
109
-------
TABLE 7.3 FILTER LEAF TEST RESULTS
Form Filter Rate
Filter Cake
Test #
1
2
3
4
5
(L)
(L)
(L)
(LS)
(LS)
Initial
Slurry
Cone.
% Solids
32.32
24.58
41.47
28.13
37.40
Ibs/hr-ft %
At
1 min
100.5
86.9
150.0
55.8
75.2
Form Time
1^ min.
__
69.9
124.4
49.1
63.9
2 min.
__
58.
102.
43.
53.
7
4
1
8
1
58
58
57
72
76
At
min.
.4
.0
.8
.6
.8
Solids
Form
1%
^ ^
58.
59.
74.
74.
Time
min.
2
0
7
0
2 min.
_
58.6
57.3
74.2
74.3
Component wt.%
Acid Insolubles
CaS04 . 2H20
CaSO3 . 1/2H2O
Ca(OH)2
CaCO3
Loss on Ignition
Total (Excluding LOI)
TABLE 7.4 SLURRY SOLIDS COMPOSITIONS
Lime Slurry
6.1
40.4
54.8
0.0
0.5
(2.5)
101.8
Limestone Slurry
6.7
62.3
16.1
0.0
14.2
( 2.2)
99.5
-------
Figure 7-4 A Photomicrograph of Lime Sludge Showing
Calcium Sulfite Crystals.
Scale: Approximately 20 microns per inch | 20,u
111
-------
Figure 7-5 A Photomicrograph of Limestone Sludge
Showing Gypsum Crystals.
Scale: Approximately 20 microns per inch | 20/j
112
-------
Nearly all of the acid insolubles in the slurry solids repre-
sent collected particulate matter rather than insolubles from the
lime or limestone.1 The relatively high 2-2.5% loss on ignition
shows that substantial carbonaceous material, or soot is present.
The presence of soot and fly ash appeared to have minimal influence
on lime slurry dewatering and cake solids. The effect, however, on
limestone slurry dewatering especially in filtration tests was
significant. Dewatering limestone sludges had a high solids con-
tent but the dewatering rates were lower than those measured for
lime slurries.
TRANSPORTABILITY
Several tests were conducted to determine the ease of trans-
porting and discharging dewatered sludge from trucks or railroad
cars. The results are presented in Table 7.5.
Sludge Transport
Slump tests indicate that, above 70 wt.% solids, either
lime or limestone slurry is readily transported.- Both
slump and angle of repose tests show that dewatered lime
slurry, below about 70 wt.% solids, is too fluid to hold its
shape. Because of its larger particle sizes, limestone
slurry is expected to begin to exhibit fluidity below about
60 wt.% solids. Slump tests were performed according to
ASTM Method D-2435.
Although laboratory tests provide a good indication of
sludge transportability, it would be prudent to carry out
full scale tests on a specific sludge to determine the
minimum solids content suitable for transport. If dewatered
sludge is too fluid to transport, blending with fly ash or
other materials would improve transportability and facilitate
unloading. If blending was not possible, relatively fluid
materials could be transported in containers with provisions
to eliminate leakage.
Sludge Unloading
The slide and truckability tests measure the minimum
angle at which sludge will move on an inclined plane.
These values are measured to indicate the degree of ease or
difficulty likely to be encountered when unloading sludge.
(1) The acid insolubles listed for the reagents in Appendix E
are diluted by approximately 3:1 for lime and 1.5:1 for
limestone in the sludge produced in the scrubber.
113
-------
TABLE 7.5 TRANSPORTABILITY TESTS RESULTS
SAMPLE NO.
Scrubber Reagent Lime Lime Limestone
Solids Content, Wt.% 74.9 66.3 83.1
Slump, Inches 0 1-1/8 0
Angle of Truckability,
Deg. 90+ 90+ 90+
Angle of Slide, Deg. 36.1 (1) 37.2
Angle of Repose, Deg. 34.8 (1) 36.0
(1) Sample was too fluid to test.
Sludge is normally transported to disposal or landfill
sites in large trucks or railroad cars. A certain amount of
compaction normally occurs during transport. The slide tests
show that uncompacted, dewatered sludge can be discharged
readily from a truck. A minimum angle of 35-40° is required to
initiate sludge movement. However, the truckability tests
indicated that compacted sludge samples which contain 65-85
wt.% solids may not discharge readily from a truck by gravity
alone. Thus, it appears that difficulty may be experienced in
discharging compacted loads of sludge even at solids concentration
as high as 85 wt.%. Blending the sludge with fly ash is recommended
to enhance the discharge of sludge from transport devices. If
adhesion of sludge to the transport vehicle is encountered, a
layer of fly ash could be placed under the sludge to facilitate
unloading.
PHYSICAL/STRUCTURAL PROPERTIES
Two tests were used to determine the strength of dewatered
sludge, California Bearing Ratio (CBR) and unconfined compressive
strength. The CBR, which is an important parameter for designing
structural landfills, is a measure of the load bearing capacity
of a confined material. Unconfined compressive strength tests
measure the ability of a material to withstand a compressive
force. CBR tests were conducted in accordance with ASTM
Method D1883-73 and unconfined compressive strength tests
according to ASTM D2166.
114
-------
TABLE 7.6 CBR TEST SUMMARY
RUN NO.
1 23 4 _ 5
Scrubber Reagent Lime Lime Lime Limestone Limestone
Samples % Solids 80.83 80.83 83.05 83.08 83.08
Pressure at 0.1
Inch Penetration
psi 100 133 200 575 383
CBR, % Standard
Load 10.0 13.3 20.0 57.5 38.3
The CBR tests showed that the gypsum solids from limestone
slurries exhibit more desirable physical properties than those
from lime slurries which were predominantely calcium sulfite.
Table 7.6 illustrates that limestone solids at similar moisture
contents, i.e., 17 to 20%, have a far greater capacity bearing
ratio, 38% to 57%, than the lime solids, 10%-20%. However, both
the lime and limestone specimens exhibited sufficiently high
compressive strengths at the 80+% solids content to support most
vehicles.
Unconfined compressive strength tests were run on the two
limestone solids samples reported in Table 7.6. The specimens
showed a tendency to break apart upon removal from the miter box
so more extensive testing was not undertaken. The limestone
samples exhibited unconfined compressive strengths of 186 and
307 psi, respectively before showing vertical cracks. These
strengths were approximately one-half of those obtained in the
CBR tests.
LEACHATE TESTS
Leachate tests were performed on samples of lime and lime-
stone slurry solids to evaluate potential environmental problems
associated with the contamination of rain water or runoff per-
colating through sludge at disposal sites. Leachates were
prepared by mixing sludge solids with distilled water. These
samples were held at room temperature for two hours under well-
stirred conditions to obtain a suitable leachate sample. This
procedure, which is used by the State of Indiana, is described
in Appendix B. Leachates were analyzed for the chemical com-
ponents listed in Table 7.7 below.
115
-------
TABLE 7.7 SLUDGE LEACHATE ANALYSES
Lime Limestone
Analysis Leachate Leachate
TDS (mg/1) 2,960 2,760
S04 (mg/1) 1810.6 1613.1
COD (mg/1) 8.4 6.8
Cl (mg/1) 72.52 48.04
Pb (ppb) <100 <100
Cd (ppb) <10 <10
Cr (ppb) 50 50
Hg (ppb) <25 <25
Leachate compositions from lime and limestone sludges are,
within the limits of experimental error, the same. Total dis-
solved solids (TDS) in the range of 2500-3000 mg/1 and the
sulfate levels of 1600-1800 mg/1 indicate that the leachates
were saturated with respect to CaSO.. Chemical oxygen demand
(COD) levels of 6.8 and 8.4 mg/1 are typical for these types of
sludge-^. Both sulfites in the sludge and organic matter in the
fly ash contribute to COD levels.
While the chloride level in the lime leachate is somewhat
higher than the limestone leachate, the remainder of the trace
elements listed in Table 7.7 are present at essentially the same
concentrations in each leachate. Fly ash is thought to be the
main source of the trace metals present. Among the three sources
of chloride i.e., lime, makeup water and coal the latter is the
major contributor.
The leachate analyses indicate no unusual or unexpected
results. The constituents found in these leachates are similar
in type and concentration to those reported in other studies.
Total dissolved solids, sulfate, chloride and chemical
oxygen demand were determined by wet chemical methods. Trace
metals were determined using a Jarrell-Ash 850 atomic absorption
spectrophotometer.
IT)P. O- Leo and J. Rossoff, "Control of Waste and Water
Pollution from Power Plant Flue Gas Cleaning Systems,"
EPA-600/7-76-018, October 1976, p. 27.
116
-------
CONCLUSIONS
o Limestone slurries containing high percentages of gypsum
yield more concentrated sludge than lime slurries which
were predominately calcium sulfite but dewatering rates
are similar.
o Sludge containing more than 70 wt.% solids is readily
transportable but may exhibit poor unloading character-
istics. Use of fly ash is recommended to improve the
transportability and unloading of dewatered sludges.
o Limestone slurry solids because of their high gypsum
content exhibit greater compressive strength than those
from lime slurries.
o Leachates from RAFB lime and limestone sludges contain
typical constituents at concentrations similar to those
reported for other FGD sludge leachates.
The above conclusions were based on results obtained
during this test program. There have been many other test
programs involving comparisons between lime and limestone as
scrubbing reagents and the resulting sludges.
The high levels of gypsum (calcium sulfate) observed
during this program in the limestone sludges have not been
observed in other programs. A comprehensive investigation of
oxidation in the R-C/Bahco system would be necessary to deter-
mine why these high levels of oxidation occurred. This type of
investigation was beyond the scope of this test program.
The effects of temperature, pH, S02 concentration and
reagent S02 stoichiometry on oxidation with lime and limestone
were scrutxnized. Based on the available data only the system
pH varied significantly when limestone was substituted for lime
(see Figure 5-6).
The strong effect of the lower pH levels experienced with
limestone on increasing oxidation coupled with the very high
oxygen content of the flue gas at RAFB (14% or more) are very
likely the factors which produced the high levels of gypsum
observed during this program.
117
-------
SECTION 8
OPERATING EXPERIENCE
This section of the report describes the operation of the
R-C/Bahco system at RAFB and circumstances which prevented its
operation.
EVALUATION OF DOWNTIME
An analysis of the downtime indicates that the system operated
with minimal requirements for maintenance under maximum load condi-
tions during the months of December, January and February. It is
also obvious that there were substantial amounts of downtime.
Figure 8-1 summarizes the overall operation of the system and
outages which lasted for more than 24 hrs.
The majority of downtime can be attributed to difficulties
with auxiliary equipment, including the booster fan, thickener,
second stage slurry pump and lime slaker. Some of the downtime
which resulted from the problems with auxiliary equipment was
attributable to time involved in obtaining replacement parts.
During the test period, spare parts, as a general rule, were
not on hand. Brief outages were required to modify spray mani-
folds and some of the control panel wiring. Heat plant outages
for annual maintenance and for tying in new heating equipment are
included in the overall analysis because some maintenance or
installation work was performed during these periods.
In general, routine maintenance and minor repairs resulted
in very short outages and did not contribute significantly to
the overall downtime. In addition, some time was lost due to
interruptions in the supply of water and electric power to the
system.
The downtime associated with auxiliary equipment, including
delays resulting from the lack of spare parts, repair time and
total downtime, is summarized in Table 8.1.
The following table summarizes the downtime contributed by
other sources:
118
-------
i
i
i
k\\\\\\\\\\\\\\\\VIL\\\\\\\N U K\\\\\\\\\\
t
Scrubbei
Startup
t
Replace Slurry
Pump Lining
i
Scrubber
Inspection
i
Replace Slurry
Pump Lining
I
\7//\ i
* ' T
Sludge
Thickener
Rake
March
April
1976
May
June
I
I i
kv^x\\\xi KXI K^V^^V\\N^^N\\V^J: kvxN
Replace
Torque
Limiler
i
Correct
Fan Vibration
i i
Repair
Sludge &
Slurry Lines
July
August
1976
September
October
ill
t ,
Install Improved
Booster Fan Bearing
i i
November la-ia December
i y /D
vVvVOsN
January
\\\\\N K\\\\\\\\\\VC
Replace
Blow down
Valves
i
1977 February
i i
KN l^^^s^^^scc\sNs^^^^sl l^^\s^^^l K^NNS^
t x *
Replace \ Inspection & Repair & Modification
Water Booster Grit Removal of Fan Wheel
t
Replace
Motor
Slaker
K^-.O Scrubber
«^J Operability
r^a Boiler
*'" Shutdown
FIGURE 8-1 Downtime related to Auxiliary Equipment
-------
TABLE 8.1 DOWNTIME RELATED TO AUXILIARY EQUIPMENT
Estimated
Procurement Time % of Repair Time % of Downtime % of % of
Tima Hrs. Total Time Hrs. Total Time hrs. Total Time Downtime
Booster Fan
Thickener
Slurry Pump
i j
o Water Booster Pumps
Lime Slaker
514
471
252
190
122
(4
(4
(2
(1
(1
.7)
.3)
.3)
.7)
.1)
2252
8
18
16
11
(20
(0
(0
(0
(0
.4)
.1)
.2)
.2)
.1)
2766
479
270
206
133
(25.
(4.
(2.
(1.
(1.
1)
4)
5)
9)
2)
57.2
9.9
5.6
4.3
2.8
1549
(14.1)
2305
(21.0)
3854 (35.1)
(79.8)
Note: Total time referred to in this table was 11,024 hrs., i.e., the total number
of hours from startup in March 1976 to the end of the test program in June 1977.
-------
TABLE 8.2 DOWNTIME FROM OTHER SOURCES
Item
Downtime
Hours
% Of
Total Time
% Of
Downtime
Heat Plant
Outages 388
Modifications 139
Maintenance 116
Loss of Utilities 99
Miscellaneous -
inadvertent, unkown
frozen lines etc.
234
(3.5)
(1.3)
(1.1)
(0.9)
Total
976
(2.1)
(8.9)
8.0
2.9
2.4
2.1
4.8
20.2
The total downtime was 4830 hours or 44.0% of the total of
11,024 hours in the period.
The modifications of the booster fan wheel in April - May 1977
and the improved thrust bearing installed in October - November of
1976 should eliminate the extensive fan outages experienced during
the test program.
Thickener related downtime was caused by a fabrication error
which in itself was easy to remedy. The extensive downtime resul-
ted from the thickener tank drying out while the necessary shaft
extension was being designed and fabricated. The problem can be
avoided, if future work on the thickener is required, by keeping
the tank full of water.
Downtime associated with the second stage slurry pump was
caused by the spare parts situation rather than the time required
to work on the pump. Replacement of wet end parts, including
rubber liners, shaft sleeves or impellers, requires approximately
one or two shifts (8 to 16 hrs). Packing was routinely added or
replaced in less than two hours and new drive belts in 2 to 3 hrs.
The mill pump is identical to the second stage slurry pump. The
only maintenance required for this pump included replacing one
set of drive belts and one set of packing.
121
-------
Downtime associated with, the water booster pump and lime
slaker was extended significantly because replacement bearings
or other parts, which would normally be available, were not on
hand. Among the other causes of downtime, heat plant outages
were longer than normal because of the new equipment being
added to the system. Of the remaining categories, only the
downtime for modifications is not likely to recur. Some of the
losses of utilities resulted from work in the heat plant.
Familiarity with the system should enable operators to reduce
downtime in the miscellaneous category.
With major equipment problems cleared up and an adequate
stock of spare parts on hand, the system should be able to
operate at well over 95% availability.
Each month of operation, from March 1976 to May 1977,
is described below, including overall operating or test conditions,
operating time for the period and a summary of all interruptions. 1
MONTHLY OPERATING SUMMARIES
March 1976
The system was started up on February 27, 1976. The flue
gas flow rate was set at approximately 40,000 SCFM and the
first and second stage pressure drops were set at approximately
7 in. W.C. After approximately 2 hrs. of operation the booster
fan started vibrating. Subsequent inspection revealed that the
motor side fan bearing had failed causing damage to the fan
shaft. The fan was repaired and the system started up again on
March 11, 1976. The system operated without incident for the
remainder of the month. Performance of the system during this
period is described in Section 4.
Operation 504 hours
Interruptions 24° hours
1) Fan bearing failure caused by binding of the oil lubrication
rings. 240 hours
April 1976
The system operated at conditions similar to those in
March. System variables were changed to obtain qualitative
data on the response of the system to adjustments in operating
parameters. While these^variables were being investigated it
became apparent that the second stage venturi pressure drop had
become insensitive to adjustments and scrubber particulate
performance seemed to deteriorate. On April 14 the system was
shutdown and the interior of the scrubber was inspected. The
IT)TTomplete log of the operation can be found in Appendix J.
122
-------
results of this inspection are described in detail later in
this section of the report. Grit from the lime slaker, which
had filled the slurry pan in the second stage venturi and
other areas, was removed from the system before it was restarted.
When the second stage slurry pump was started, the rubber
lining collapsed. Delays in obtaining a replacement liner for
the pump resulted in an outage which lasted until April 29.
Operation 453 hours
Interruptions 267 hours
1) The scrubber inspection and cleanout was caused by indequate
grit removal in the lime slaker. Operating procedures were
modified to improve grit removal. 15 hours
2) The second slurry pump lining collapse was caused by mismatch
between the slurry pump suction line and the slurry pump inlet.
An adaptor plate was installed with the new liner to eliminate
this problem. 252 hours
May 1976
In May, a series of material balances were conducted to
quantitatively determine the overall performance of the system.
Complete details of this work can be found in Section 4. For
these tests the system was operated at a gas flow of approximately
49,000 SCFM, and a total pressure drop of 22 in. w.c. The
inlet gas temperature averaged 320°F.
Early in the month, two plastic external spray manifolds
on the upper part of the scrubber were removed. These were
replaced by stainless steel manifolds built into the shell of
the scrubber.
A few days later, dirt contaminated the motor side fan
bearing resulting in a failure. Rebabbiting of the bearing was
necessary since a spare was not on hand.
Operation 445 hours
Interruptions 299 hours
1) Replacement of water spray manifolds 57 hours
2) Failure of the booster fan motor side bearing was caused by
dirt entering the bearing. After the repair was completed, in-
strument air purges were added to the bearings to prevent dirt
from entering them. 226 hours__
3) Unknown 16 hours
123
-------
Operation 459 hours
Interruptions 285 hours
1) Resealing the wood thickener tank. 159 hours
2) Procuring and replacing the lime slaker torque
limiter. 43 hours
3} Miscellaneous maintenance, training and loss of
utilities 9 hours
4) High fan vibration. 58 hours
5) Replacing sludge pump diaphragm. 16 hours
August 1976
During the month of August the fan vibration which began in
July was analyzed. This analysis revealed that the concrete
outboard fan bearing support had become separated from the slab
upon which it rested. External braces for the bearing support
were designed, fabricated and installed to rectify this problem.
When the system was started up near the end of August, the motor
on the lime slaker grit conveyor failed. These problems resulted
in a shutdown for essentially the entire month of August.
Operation 16 hours
Interruptions 728 hours
1) Diagnosis of the problem and repair of the fan
support. 701 hours
2) Replacement of the grit conveyor motor. The motor
was obtained from a local distributor. 19 .hours
3) Miscellaneous. 8 hours
September 1976
In September the system operated at conditions similar to
those for July. Performance tests for S02 removal efficiency
were conducted. The results of these tests are presented in
Appendix J. The level controller in the lime dissolving tank
malfunctioned causing some downtime during the month. During the
last week in September the heat plant was shutdown to tie in
piping for a new hot water generator. The fan alignment was
checked and adjusted and the fan was rebalanced during this
outage.
_ 532 hours
Operation_ 18g hours
Interruptions
1) Heat plant outage and realigning and balancing
the booster fan. 142 hours
2) Lime dissolver level control malfunction. 40 hours
124
-------
3) Maintenance. 6 hours
October 1976
The outage for tying in the new generator extended three
days into October. When the system was started up the load on
the heat plant required an increase in gas flow to the scrubber.
In the middle of the month one of the sludge pumps stopped
operating when the hose to the pond was full of high density
(50+% solids) sludge. The spare hose had already been plugged by
a similar mishap and had not been cleaned. The hose was removed
from its casing and unplugged. Shortly after the system was
restarted a slurry hose which runs from the first stage liquid
seal to the lime dissolver developed a leak. A stainless steel
coil had been placed inside this hose to prevent it from collapsing
This coil had a sharp end which worked its way through the hose
causing the leak. The damaged section of the hose was replaced
with a stainless steel pipe and the coil was modified to prevent
the problem from recurring. During the last week in October, the
fan began vibrating and the system was shutdown.
Operation 451 hours
Interruptions 293 hours
1) Heat plant outage. 78 hours
2) Unplugging the sludge line to the pond. 18 hours
3) Fabrication and installation of a stainless steel
line to repair the damaged slurry line. 64 hours
4) Loss of utilities - water and electric
power. 47 hours
5) Miscellaneous. 8 hours
6) Fan vibration. 78 hours
November 1976
The entire month of November as well as several days in
December were spent replacing the booster fan bearing on the
motor side and repairing the fan shaft. Problems in obtaining
replacement parts contributed substantially to the length of this
outage.
Operation 0 hour
Interruptions 720 hours
125
-------
December 19:7S
The system functioned smoothly throughout the month. The
inlet gas flow rate was 50,000 scfm and the scrubber inlet tem-
perature reached 46Q°F, typical high-load winter operating con-
ditions. The first phase of the process variable screening study
using lime as a reagent was undertaken. Both SO- and particulate
removal efficiencies were evaluated while eight primary operating
parameters were varied. These tests are described in detail in
Sections 5 and 6 of this report.
Approximately three days in December were lost while com-
pleting the repair to the fan bearing. Aside from minor outages
due to the losses of water and plant air and some frozen instrument
air lines availability of the system was nearly 100%.
Operation 623 hours
Interruptions 121 hours
1) Completion of fan bearing repairs. 82 hours
2) Loss of water and electric power. 24 hours
3) Frozen air lines. 9 hours
4) Maintenance and miscellaneous. 6 hours
January 1977
The system ran very smoothly in January. Despite
record breaking sub-zero weather, availability was approximately
94%. The scrubber operated at its rated capacity, approximately
50,000 scfm, as it had in December. S02 test work planned for
January was cancelled due to the extreme weather conditions.
Frozen air lines which supply instrument air to operate the
scrubber blowdown valves caused some downtime. In addition,
some time was lost repairing frozen water lines.
Operation 69,9 hours
Interruptions 45 hours
1) Frozen air lines and cleanout of accumulated grit
due to the inability to operate the blowdown valves 30 hours
2) Repair of frozen water lines. 9 hours
3) Maintenance and miscellaneous. 6 hours
126
-------
February 1977
The system functioned very well for the third consecutive
month at operating and ambient conditions similar to those in
December and January. In February the system operated at slightly
above the design gas rate. Testing was resumed and the variable
screening tests started in December were completed. The blowdown
valves again were the most significant source of downtime.
During this month, however, time was spent replacing the air
operated valves with more reliable manual valves.
Operation 588 hours
Interruptions 84 hours
1) Blowdown valve blockages and replacement of
these valves. 70 hours
2) Repair of frozen water lines. 4 hours
3) Maintenance and miscellaneous. TO hours
March 1977
In March the lime reagent testing was completed. Prior to
starting this work, the system was inspected and the grit which
had accumulated over previous months from the inoperative blowdown
valves was removed. Early in March a defective bearing in the
water booster pump was replaced and some wiring changes were made
in the control panel.
Operation 465 hours
Interruption 279 hours
1) Procuring and replacing the water booster pump
bearing. 206 hours
2) Rewiring control panel. 39 hours
3) Inspecting the system and removing accumulated
grit. 30 hours
4) Miscellaneous. 4 hours
April 19-77
Work on the particulate removal efficiency tests begun in
December was completed in April. The results of these tests are
described in Section 6 of this report. In April limestone was
substituted for lime as the SO2 scrubbing reagent. Heat plant
loads were significantly lower than in prior months. The system
performed smoothly until the middle of the month when booster
127
-------
fan vibration levels rose. An inspection revealed cracks in the
rim of the fan wheel. This assembly was removed and subsequently
modified by removing the rims and reducing the wheel diameter
from 81 inches to 76 inches. The resulting outage began in mid
April and lasted until early May.
Operation 218 hours
Interruptions 439 hours
1) Repair and modification of the fan wheel. 430 hours
2) Maintenance and miscellaneous. 9 hours
May 1977
In May a series of limestone reagent SO,, performance tests
were completed. This work was the final phase of the field test
program for this contract at RAFB. The results of these tests
are described in Section 5 of this report.
The fan repair was completed in the first part of May. Some
additional downtime resulted from repairs to a fan bearing resistance
temperature detector which was disturbed during the fan repairs.
In addition, a motor on the lime slaker paddle drive was replaced.
Operation 435 hours
Interruptions 309 hours
1) Complete fan repairs. 201 hours
2) Repair fan bearing RTD. 30 hours
3) Procure and replace lime slaker motor. 71 hours
4) Maintenance and miscellaneous. 7 hours
SCRUBBER INSPECTIONS
Scrubber inspections were an integral part of the program to
monitor scrubber performance. A thorough internal inspection was
made in April of 1976, approximately one month after start-up,
and a follow-up inspection was made two months later in June.
Subsequently, inspections were made when opportunities were_
available during outages up to the end of the test program in
June of 1977. These inspections helped to determine the effectiveness
of the water makeup system in keeping key areas in the scrubber
clean and provided an opportunity to see if any other problems
were developing. The inspections revealed that solids had accumula-
ted in the following areas as shown in Figure 8-2 during the test
program.
128
-------
STACK
MANHOLE
PLATFORM
2ND STAGE
<==
MIST ELIMINATOR
- MAN DOOR
1ST STAGE
DROP COLLECTOR
SCHEDULE
0 GAS FLOW
-* SLURRY FLOW
INLET Q SPRAY MANIFOLDS
1ST STAGE
RECYCLE PUMP
Figure 8-2 Bahco Scrubber Module
129
-------
1. Scrubber inlet wet/dry zone
2. Bottom of the 1st stage drop collector
3. 2nd stage venturi slurry pan
4. Top of the 2nd stage venturi spin assembly
5. Top of the 2nd stage mist eliminator
6. Straightening vanes in the stack and the stack wall
7. Bottom of the flue gas inlet manifold
April 1976 Inspection
The inspection in April 1976 revealed the following:
Scrubber Inlet Wet/Dry Zone
The solids in the wet/dry zone in the inlet area were a
mixture of dried slurry and fly ash which had accumulated on the
spray manifold in this location. Since there was no reduction in
the area of the flue gas inlet ports and no appreciable decrease
in flow area between the scrubber wall and the first stage
venturi. this material was left in place.
Bottom of the First Stage Drop Collector
A coarse sandy or gritty material covered the bottom of the
drop collector. The buildup was approximately 18 inches deep
opposite the slurry outlet and sloped down towards the outlet. ~~
This material was left in place since slurry flow was not impeded.
Second Stage Venturi Slurry Pan
The pan in the second stage just below the venturi had a
substantial accumulation of coarse sandy material similar to that
in the first stage drop collector. This material had accumulated
to the point that the second stage venturi performance was
adversely affected. Since the pan was nearly full to the rim^
(about 24 inches deep) over more than half of its area, allowing
flue gas to flow through the venturi with little slurry pickup
this material was removed.
Top of the Second Stage Venturi Spin Assembly
The buildup on top of the second stage spin assembly was a
soft muddy material several inches thick. This material which appeared
to be typical scrubber solids which had settled out in a stagnant
area in the centrifugal mist eliminator was left in place.
130
-------
Top of the Second Stage Mist Eliminator
The solids which were observed in the top of the second
stage mist eliminator had essentially filled in the corner at the
top of the mist eliminator just above the spray nozzles. This
material was left in place.
Straightening Vanes in the Stack and the Stack Wall
The material which accumulated in the stack area appeared to
be a mixture of fly ash and slurry which had deposited and
recystallized. The buildup on the stack wall was approximately
% inch thick. The buildup on the straightening vanes was approxi-
mately % inch thick. These accumulations were left in place.
The accumulation in the second stage venturi slurry pan location
(3) was the only accumulation which posed any problem to the
operation of the scrubber. The pan was emptied and solids were
dislodged from the wet/dry zone and removed from the system.
This operation took approximately 15 hours. Subsequent investi-
gation into the cause of the accumulation in the pan revealed
that the lime slaker grit removal circuit was not operating
effectively and virtually all of the gritty material in the
unslaked lime was entering the scrubber. The grit removal cir-
cuit was readjusted to remove this material. Figure 8-3 shows
the scrubber internals in the second stag.e. A light coating of
solids had accumulated on the venturi spray manifold and shell.
This coating which was observed during subsequent inspections
remained essentially unchanged during the entire fourteen month
test period.
June 1976 Inspection
A follow-up inspection during an annual heat plant outage in
June 1976 revealed the following:
Scrubber Inlet Wet/Dry Zone
The inlet wet/dry zone had an accumulation which was very
similar to that observed in April. Again, since there was no
reduction in the gas inlet port area or in the flow area to the
first stage venturi this material was left in place.
Bottom of the First Stage Drop Collector
There was a slight increase in the amount of material
located in the first stage drop collector compared to the amount
observed in April. Again, this material was left in place.
131
-------
Venturi
Legend:
Gas Flow
Figure 8-3 R-C/Bahco Scrubber Internals Showing the Venturi,
Pan and Spray Manifold in the Second Stage
132
-------
Second Stage Venturi Slurry Pan
This area had accumulated an insignificant amount of
gritty material which partially covered the bottom of the pan.
This material was left in place. The revised lime slaker operating
procedures had eliminated the qrit accumulation problem and the
resulting rapid buildup which had occurred in April.
Top of the Second Stage Venturi Spin Assembly
The muddy accumulation at this location was thicker than it
was in April. It appeared to be the same type of material
observed earlier, again :.it was left in place.
Top of the Second Stage Mist Eliminator
The solids which accumulated in this area did not seem to be
any heavier than the deposit which had been observed in April.
This indicated that a stable accumulation had occurred. Again,
the material was left in place.
Straightening Vanes in the Stack and the Stack Wall
This area continued to accumulate material. The straightening
vanes had approximately 1-1/2" of buildup and the stack wall
approximately 3/4". This material was left in place.
Bottom of Flue Gas Inlet Manifold
Between April and May some flooding occurred in the first
stage causing slurry to enter the flue gas inlet manifold. This
flooding resulted in a deposit of dried solids approximately 4"
deep. This material was left in place.
September 1976 Inspection
The heat plant was shut down again in September 1976 to tie
in new hot water distribution lines. The scrubber was inspected
again at this time.
Additional accumulations of solids had occurred in two
locations. The straightening vanes, location 6, had accumulated
further solids which caused an undesirable pressure drop of
approximately 2 in. W.C. The material, which was 2 to 4 in.
thick, was removed at this time. The bottom of the flue gas
inlet manifold, location 7, had accumulated more solids from
additional overflow from the first stage venturi. The solids
were 8" to 10" deep at this time. Since the continued flooding
of the first stage and the resulting buildup were unacceptable
occurrences, they had to be eliminated. Subsequent investigation
revealed that operation of the first stage at pressure drops
above 12" to 13" w.c., coupled with a second stage slurry pumping
rate more than 50% higher than the design rate of 2600 gpm,
caused flooding when the gas flow was reduced below design'
133
-------
rates or the booster fan was shut down. In order to eliminate
this problem, the speed of the second stage slurry pump was
decreased to reduce the slurry pumping rate to design levels. In
addition, an interlock was added to stop the second stage slurry
pump when the booster fan shut down.
March 1977 Inspection
A somewhat unusual winter caused a number of freezing
problems in instrument air supply lines. These freezeups rendered
the scrubber blowdown valves inoperative. This resulted in a
substantial accumulation of gritty material in the system,
partially blocking the scrubber's slurry outlets. In March of
1977, in preparation for a continuation of the SO- and particulate
tests, the scrubber was inspected. The bottom of the first stage
drop collector and the second stage venturi pan had accumulated
quantities of gritty material which were substantial enough to
require removal. In addition, the scrubber inlet and straightening
vanes in the stack were cleaned.
The Scrubber Inlet Wet/Dry Zone
This area had a buildup which appeared to be similar to that
observed in June of 1976. Again there was no reduction in gas
flow area. This material was dislodged and removed.
Bottom of the First Stage Drop Collector
The material which deposited in this area as a result of the
inoperative blowdown valves was a mixture of muddy and gritty
material as much as 30 inches deep. This material settled out
when the slurry outlet from this part of the scrubber was obstructed
by grit accumulated above the inoperative blow down valves. This
material was removed from the scrubber.
Second Stage Venturi Pan
The retention of gritty material in the system caused_by the
inoperative blowdown valve, resulted in an accumulation which
half filled the 24" deep pan. This accumulation could have
adversely affected the operation of the second stage venturi at
.low pressure drops; therefore, it was removed from the scrubber.
Straightening Vanes in the Stack and the Stack Wall
Again, a crystalline deposit occurred in this area. These
deposits were 1 to 2 in. thick on the straightening vanes and 1/2
in. thick in the stack. At this time, these deposits were
removed.
The cleanout of the four areas described above took approx-
imately 15 hours.
134
-------
June 1977 Inspection
In June of 1977, after completion of the test program, the
system was inspected during an annual heat plant outage. The
material which had accumulated in the bottom of the flue gas
inlet manifold during the first six months of operation was
removed. As expected, the inspection revealed the following:
Scrubber Inlet Wet/Dry Zone
The usual accumulation observed in the past was present.
This material was not removed.
Bottom of the First Stage Drop Collector
Some muddy deposits which were thickest opposite from the
slurry outlet had accumulated. These deposits which were left
in place tapered off near the outlet leaving the slurry flow
passage 100% open.
Second Stage Venturi Slurry Pan
A small amount of gritty material which partially covered
the bottom of the pan was observed. This material was leftr-in
place.
Top of the Second Stage Venturi Spin Assembly
A layer of soft deposits which were approximately 6" thick
was observed on top of the venturi. This material was left in
place.
Top of the Second Stage Mist Eliminator
As it appeared in earlier inspections, the upper corner was
filled with solids. This material was left in place.
Straightening Vanes in the Stack and the Stack Wall
As observed earlier, a buildup had accumulated on the vanes
which was approximately 1/2" thick. In addition, approximately
1/4" of accumulation had occurred in the stack. This material
was left in place.
Conclusions
Three areas which can result in deterioration of scrubber
performance include infiltration of grit into the system through
the lime slaker, inadequate operation of the scrubber blowdown
valves, and the slow accumulation of solids in the straightening
vanes and stack.
The infiltration of grit can be kept to a minimum by paying
close attention to the operation of the lime slaker grit removal
circuit. The blowdown valves need to be operated two to four
135
-------
times a shift, depending on scrubber load, to prevent an accumulation
of solids in the slurry outlets. The straightening vanes and
stack wall accumulate some material which must be checked on a
semiannual basis. Removal of this material is required when the
accumulation approaches 3" - 4" in thickness in the vanes. The
rate of accumulation can be minimized by operating the second
stage venturi to minimize the possibility of slurry droplet
carryover (i.e., by operating at pressure drops below 12" w.c.).
Figure 8-4 illustrates the buildup which accumulated on the vanes
in the stack area during approximately six months of operation.
It is important to note that the reduction in gas flow area even
at this point is relatively insignificant and did not adversely
affect the operation of the scrubber.
The foregoing indicates that there are no significant
problems related to the accumulation of solids in this system.
The scrubber has the ability to tolerate substantial accumula-
tions of solids resulting from external operating problems
before scrubbing performance is adversely affected. In addition
any deterioration in performance, if it occurs, is gradual and
can be rectified at a convenient time.
136
-------
Figure 8-4 Buildup on the Vanes in the Stack Area
After Six Months of Operation
137
-------
SECTION 9
OPERATING AND MAINTENANCE COSTS
In order to arrive at operating and maintenance costs for
the R-C/Bahco system, data was collected during the period from
March 11, 1976 to May 31, 1977. This data included information
on the consumption of power, water, and chemicals as well as
operating and maintenance costs.
OPERATING COSTS
During the test period, the scrubbing system treated flue
gas from the combustion of 27,216 tons of coal; sulfur content
ranged from 2 to 3.5% and averaged 2.5%. The total operating
cost, summarized in Table 9.1, for utilities, reagent, supplies
and operating labor was $5.07/ton of coal burned. These costs
were projected to be $5.92/ton of coal burned and a cost ceiling
of $7.56/ton, based on current reagent and power cost, was
guaranteed in the contract. Maintenance, labor and materials
added $0.21 to the cost of $5. 07/ton of coal burned. An opera-
ting cost of $4.06/ton can be achieved if limestone is used or
the scrubbing reagent and fan settings and makeup water consump-
tion are optimized.
TABLE 9.1 OPERATING COST SUMMARY
Power
Booster
Auxiliary
Equipment
Water
Potable water
Well water
Chemicals
Lime
Limestone
Units
3,065,000 KW
602,500 KW
4,448,120 Gal
1,768,700 Gal
721 Tons
130 Tons
Unit Cost
$0.024/KWH
$0.024/KWH
$0.36/M gal
$40.35/ton
$12.72/ton
Total Cost
$ 73,560
14,460
1,601
29,092
1,654
ofifting labor 1,860 man-hrs . $7 52/man-hr 13,987
Supervision 25% of operating labor (estimate^ 3,497
Total
$137,851
Cost per ton of coal = $5.07
138
-------
Data in Table 9.1 on operating labor requirements and rates,
lime and limestone deliveries, coal consumption and water and
power costs were obtained from the USAF.
Power Consumption
As part of the test program, devices to measure power con-
sumption of the booster fan as well as other equipment in the
system were installed. These are listed in Table 9.2. Pumps,
controls, lighting, heat tracing, control house heating and
cooling, etc. were included.
TABLE 9.2 EQUIPMENT POWER REQUIREMENTS
Booster Fan
Slurry circulating pumps
Makeup water booster pump
Thickener
Lime Slaker
Freeze Protection 7
Other misc. equipment f
Lighting -'
No. Used
(1)
(2)
(1)
(1)
(1)
Total KW
or (H.P.)
518 (700)
74.0 (100)
22.4 (30)
1.5 (2)
2.2 (3)
7.5 (10)
Power consumed by the auxiliary equipment, as listed in
Table 9.1, is typical for this type of system. The booster fan
power consumption, however, is substantially higher than normal.
Several factors contributed to the high fan power consumption,
including the extra capacity required for flexibility in running
this test program as well as the need to run higher than normal
pressure drops to cope with high levels of soot.
In order to insure that the capacity of the system would not
be limited by fan capacity, a fan with extra pressure and flow
capacity requiring a 700 H.P. motor was installed. A 500 H.P.
model would normally have been selected for this application.
Under normal circumstances the scrubber would have been operated
at at total pressure drop of approximately 15 in. W.C. Routine
operation at total pressure drops over 20 in. W.C. to attempt to
maximize the removal of soot resulted in a significant increase
in power consumption and a power cost penalty of approximately
$14,000. Near the end of the test program the fan was modified
to reduce its pressure capacity from 32 to 25 in W.C. This modi-
fication will alleviate some throttling losses but the major power
cost penalty will be eliminated only when improvements in the
heat plant result in better combustion.
139
-------
Water Consumption
The consumption of potable water did not contribute significantly
to the overall cost of operating the system. This cost, however,
could be reduced by more efficient water management. The reuse
of booster fan bearing cooling water for lime slaking or on
slurry pump seals and optimizing lime slaker water consumption
could reduce potable water consumption to approximately half of
the present rate of 159 gpm.
Chemical Consumption
The average stoichiometry for the test period based on the
total coal consumption was approximately 0.50 moles of reagent
per mole of S02. During the test period 90% of the reagent
used was lime, the balance was limestone.
The average stoichiometry of 0.5 was adequate for 2.5%
sulfur coal to meet the existing emissions standards of 2.2#S02
per MM Btu of coal burned.
If limestone were substituted for lime, this SO- emission
standard could be achieved at a stoichiometry of 0.55. Sub-
stituting limestone for lime during the test program would have
resulted in a reagent cost reduction of $14,000.
Labor Costs
Operating manpower needs were met by a scrubber technician,
who operated the system on day shift during the week while per-
forming routine maintenance tasks, and heat plant personnel who
operated the system, in addition to running the heat plant, on
off shifts and weekends.
The scrubber technician handled routine operation and
maintenance items including lubrication, instrumentation and
electrical equipment in addition to routine mechanical maintenance
such as adjusting pump packings and belt tensions. Routine
operation of the scrubber includes monitoring the system's
auxiliary equipment and periodically removing gritty material
from the lime slaker grit removal circuit and the scrubber
blowdowns.
Since the USAF at Rickenbacker does not keep a separate
account of supervision requirements, the cost of supervision
attributable to scrubber operation was estimated to be 25% of
the operating labor cost.
140
-------
Optimum Operating Costs
Table 9.3, summarizes a projected cost for operating the
R-C/Bahco system at RAFB at optimum conditions, for a period
comparable to the test period.
TABLE 9.3 OPTIMUM OPERATING COSTS
Power
Fan $ 59,500
Auxiliary Equipment 14,460
Water
Potable 800
Well
Chemicals
Lime 32,375
Limestone (if used instead of lime) (18,375)
Labor
Operating labor 13,987
Supervision 3 ,497
Total for lime $142,990
Cost per ton of coal $4. 58 ($4.06 for limestone)
MAINTENANCE COSTS
The maintenance costs presented in this section include
items related to normal maintenance. Problems such as fan
repairs which were handled on a warranty basis, and are not
likely to recur, have not been included.
The following labor and materials costs, for routine main-
tenance listed in Table 9.4, were incurred during the test
period:
141
-------
TABLE 9.4 MAINTENANCE LABOR AND MATERIAL COSTS
Maintenance Labor
Repair lime feeder $300.
Repair sludge control valve 100.
Replace slurry pump shaft,
bushing and sleeve 800.
Repair water booster pump 200.
Scrubber maintenance 600.
Rebuild delumper 1000.
Service Dupont Analyzer 400.
Clear lime unloading line 100.
Sub-total $3,500
Maintenance Material
Replace slurry pump belts 50.
Replace sludge pump'diaphragm ' 20.
Replace fan bearing oil 10.
Repack slurry pumps 120.
Replace lh in. pinch valve liner 75.
Replace slurry pump bushing & sleeve 500
Rebuild delumper 200.
Replace slaker torque limiter 10.
Replace pH probe 250.
Touchup paint 25.
Recorder ink & paper 600.
Sub-total 1,860.
Total Maintenance Cost (Labor
and Material) $5,360
The maintenance labor listed above does not include_the
time spent by the scrubber technician in performing routine ad-
justments, oil changes, etc. However, this time was included
in the operating cost summary under operating labor.
As indicated by the modest maintenance requirements,
the system is relatively simple to maintain and should prove to
be relatively trouble-free.
142
-------
RECOMMENDED OPERATING CONDITIONS
Reagent Selection
Lime was originally chosen as the S02 scrubbing reagent
for RAFB. This choice was made to insure that S02 emissions
would be well within the allowable limits in effect at that
time, i.e. 1.0 pounds of S02 per million Btu gross heat input.
Dependable SO2 emissions of 0.83 pounds S02 per million Btu at
a stoichiometry of 0.87 were achieved using lime.
Since that time new air pollution requirements have been
proposed to allow emissions of 2.2 pounds per million Btu.
Either lime or limestone can be used effectively to meet this
proposed standard. However, lime delivered at the site costs
$40.35 per ton while limestone costs $12.72 per ton. Economics
favor limestone even though nearly double the weight of lime-
stone is required to compensate for the higher molecular weight
and lower utilization (85%) expected for limestone. The amount
of lime with 90% available CaO required to treat S02 produced
by burning one ton of 3.5% sulfur coal costs $1.88 while lime-
stone with 95% available alkalinity can produce the same result
for $1.05.
Limestone is easier to feed into the system since it is
not hygroscopic and requires no slaking. In addition, lime-
stone is less hazardous since it does not exhibit the caustic
properties of lime dust and slurries which can cause damage to
sensitive tissue. The fact that the use of limestone results
in approximately 25% more weight of sludge due to higher gypsum
levels and the presence of unreacted limestone is offset by the
fact that limestone sludges settle more rapidly and produce
higher final solids content than lime sludge. These factors
will essentially balance the excess weight resulting in an
equivalent volume required for storage, i.e., pond life at RAFB
will not be decreased by switching to limestone. Again, it is
important to note that the high gypsum content of the limestone
sludge results from the particular circumstances at RAFB which
produce high levels of oxidation.
Slurry Pumping Notes
The second stage slurry pumping rate should be set for a
flow of 2200 to 2400 gpm to maximize limestone utilization and
SO2 removal efficiency.
Flue Gas Notes
Flue gas flow rates should be maintained between 35,000
and 55,000 SCFM as heat plant load variations permit, to avoid
unstable operating conditions which result in high particulate
emissions.
143
-------
Venturi Pressure Drops
First and second stage pressure drops should be maintained
between 7 and 10 in. W.C. to achieve good liquid pickup while
avoiding excessively high or low total pressure drops. Pressure
drops close to 7 in. W.C. should be selected when operating with
the new generator to minimize power consumption. When older
generators are being operated, pressure drops near 10 in. W.C.
may be needed to reduce particulate emissions to acceptable
levels if combustion problems persist.
Table 9.5 summarizes the recommended stoichiometries, gas
rates, pressure drops and slurry pumping rates necessary to
achieve optimum performance in the R-C/Bahco system at RAFB with
the present particulate and SO- emissions requirements. The
recommendations made above witn regard to minimizing operating
costs should be implemented in conjunction with the recommended
operating conditions listed below.
TABLE 9-5 SUMMARY OF RECOMMENDED OPERATING CONDITIONS
Reagent Limestone Lime
Stoichiometry 0.75 0.7
Slurry rate to
scrubber 2200-2400 gpm 2200-2400 gpm
Flue gas rates 35-55,000 scfm 35-55,000 scfm
First stage AP 7-10 in. W.C. 7-10 in. W.C.
Note: The stoichiometries listed in this table are for 3.5%
sulfur, 11,500 Btu/lb coal. Adjustments to accommodate
other coal types used at RAFB are described in the system
operating manual.
144
-------
APPENDIX A
CONVERSION FACTORS: BRITISH TO SI UNITS
To Convert From
LENGTH
inch(in)
foot(ft)
AREA
inch2(in2)
foot2(ft2)
VOLUME
3 3
inch (in )
foot3(ft3)
foot3(ft3)
gallon(gal)
gallon(gal)
MASS
ounce(oz)
pound(Ib)
pound(Ib)
grain(gr)
Ton(T)
PRESSURE
inches W.C.(in w.c.)
pounds/inch2(psi)
TEMPERATURE
degree Fahrenheit(°F)
degree Rankin(°R)
ENERGY
British Thermal Unit
(Btu)
British Thermal Unit
(Btu)
To
meter(m)
meter(m)
2 2
meter (m )
meter2(m2)
meter3(m )
meter3(m3)
liter (1)
meter (m3)
liter(1)
kilogram(kg)
gram(g)
kilogram(kg)
gram(g)
kilogram(kg)
kilopascal (kPa)
kilopascal (kPa)
degree centrigrade(°C)
degree Kelvin(°K)
joule(J)
kilojoule(kJ)
Multiply by
2.540x10
0.3048
-2
6.452x10 4
9.290xlO~2
1.639x10
2.832xlO~2
28.32
3.785x10
3.785
-3
_ o
-2
2.835x10
453.6
0.4536
6.480x10
907.2
0.2488
6.895
tc = (t--32)
0.5555 r
1055.
1.055
145
-------
POWER
British Thermal
Unit/hour
(Btu/hr)
British Thermal
Unit/hour
(Btu/hr)
British Thermal
Unit/horsepower(hp)
DENSITY
pound/foot3(lb/ft3)
pounds/gallon(Ib/gal)
VISCOSITY
pound foot p
second/foot
(Ib. ft/sec ft2)
MISCELLANEOUS
cubic
feet/minute(CFM)
gallons/1000 ft3
(gal/M
gallon/minute
(gal/min)
grains/standard
cubic foot
(gr/SCF)
feet/second
(ft/sec)
pounds/1,000,000 BTU
(Ib/MM Btu)
British Thermal
Units/pound
(Btu/lb)
watt(w)
kilowatt(kw)
kilowatt(kw)
kilogram/meter
(kg/m3)
kilogram/meter3
(kg/m3)
pascal-second
(Pas)
meter /hour
(m3/hr)
liter/meter3
(1/m3)
liter/minute
(1/min)
grams/normal meter
(g/nm3)
meter/second
(m/sec)
grams/kilojoule
(g/kJ)
kilojoule/kilogram
(kJ/kg)
0.2931
2.931x10
0.7457
16.02
119.8
47.89
-4
1.699
0.1337
3.785
2.288
0.3048
429. 9
2.326
146
-------
APPENDIX B
ANALYTICAL AND TESTING METHODS
B1 Tnermogravimetrlc Analysis
B-2 Analytical Procedure for S02 Wet Tes-ts
B3 Procedure for Preparing Leachate Samples
B.-4 Analytical Methods-
147
-------
APPENDIX B-l
THERMOGRAVIMETRIC ANALYSIS OF SOLIDS FROM BAHCO SCRUBBING PROCESS
General Procedure:
All analyses performed on lime-based scrubbing
solids from the Banco S02 Gas Removal Process at
Rickenbacker Air Force Base utilized the specific
technique of thermogravimetric analysis (TGA).
This technique involved heating a prepared sample of
solid phase material at a specific rate over a pre-
determined temperature range and observing the weight
change which results from solid state reaction occurring
at some characteristic temperature.
After in-laboratory treatment, which includes
drying at 35°C for 24 hours,breaking up of the dried
solids and riffling as many times as needed to obtain a
representative sample of about 2.5 gms.; the prepared
solid phase sample is then subjected to analysis on the
thermobalance over a preprogrammed temperature rangef
(ambient to 980°C) at a specific heating rate (80°C
per minute). The resulting thermogram will exhibit,
in a general case, associated weight losses of 2 waters
of hydration from CaSO4.2H20(130-200°C), 1/2 water of
hydration from CaSO3-l/2H20(400-450°C), dehydration of
Ca(OH)2 (575-625°C), loss on ignition from combustibles
(700-750°C) and finally evolution of C02 from CaC03 (800-
900°C). Measuring these losses and back calculating each
148
-------
-2-
for the particular constituent results in a total analysis
of the solids/with the exception of any inert material,
which would require separate testing.
In addition to the various calcium compounds which
were determined by TGA, it also became necessary to de-
termine the concentration of MgCO- present in the solid
phase during the limestone phase of the test program.
Since the limestone used in this phase of the study
was dolomitic, the presence of MgCC^ was noticed during
initial thermogravimetric testing. To effectively separate
the weight losses of C02 from MgCC^, and the weight
loss from the ignition of combustibles, which overlap at
650°C, it was necessary to reanalyze these samples in a
nitrogen environment where combustion would not take
place. The resulting weight loss in percent was then
back calculated in the same manner as previously stated.
To insure the data obtained from TGA analyses pro-
duced a high degree of accuracy, an alternate wet chemi-
cal method was employed as a check. This wet test pro-
cedure uses the same prepared sample which is reacted
in an absorption train assembly using concentrated hydro-
chloric acid to digest the sample. As the sample digests
in the acid medium f CO2 and S02 are evolved and forced
through the train. The evolved C02 and S02 gases are
passed through a small gas washing bottle filled with a
3% hydrogen peroxide solution, which traps any SO2
forming H2SO4. The C02 gas also passes through the
149
-------
-3-
peroxide trap, into a series of acid and moisture traps
to a preweighed Miller bulb containing 20 mesh Ascarite,
which will absorb the C02.
After the digestion has been completed, the reaction
flask solution is tested for insolubles, calcium contentt
by EDTA titration (also.magnesium, content if applicable)
and total sulfate by gravimetric means. The solution in
the peroxide trap is titrated for S02 using a BaCl2 titrant
and Thorin as an indicator. The Miller bulb is weighed
to determine the weight of C02 absorption. From data
obtained from these tests we are able to perform
a complete analysis of all the constituents previously
mentioned. Data obtained by this procedure match TGA
results quite closely in all samples tested.
In addition, to the use of wet chemical methods to
verify TGA data, the use of laboratory prepared samples
using reagent grade chemicals similar to those to be
determined were also tested by thermal -means. Various
ratios of CaSO4 2H20 to CaSO 3 1/2H20 with amounts
of CaCO3 and MgC03 were analyzed by TGA. The data ob-
tained from these TGA analyses also yielded results which
correlated closely to calculated percentages in the
sample formulations.
150
-------
-4-
Uae of a th^rmogravimetric, Balance, for lime, or limestone.
Based solids analysis- is: a rapid, reliaBle method for the.
determination of CaSO4'2H20, CaSD3-1/2H2O, Ca(OHl2, MgC03 and
CaC03. ^^ "QS"e or tnis" ins-trument witn. occasional wet chemical
methods prod-aces data which is highly accurate and in subs-tantially
less time than compara&le wet chemical analyses.
151
-------
APPENDIX B-2
ANALYTICAL PROCEDURE FOR SO, WET TESTS
This method for determining the SO, content of gas streams
is only approximate and should be used only as a semi-
quantitative check on SO2 concentrations.
No temperature or pressure corrections have been
incorporated, and the method should not be used below 100
ppm.
Apparatus: Reagents:
1) 250 ml impinger with an open 1) 3% Hydrogen Peroxide
glass dip tube. 2) 0.1N NaOH or 0.01 N
NaOH
2) A dry test meter. 3) Methyl/Orange-Xylene
3) A source of vacuum. Cyanol indicator
4) 25 ml pipette.
5) Vacuum tubing.
6) Hose clamp.
Procedure:
Inlet Samples (i.e., 500 + ppm SO,) pipette 25 ml of 0.1 N
NaOH into the 250 ml impinger, add 50 ml of 3% hydrogen
peroxide. Add approximately 25 ml of deionized water. Add
several drops of Methyl/Orange-Xylene Cyanol indicator.
Draw the gas- sample through the impinger at 0.1 to 0.2
ft.3/min. Record the gas meter reading when the indicator
turns from green to purple.
Outlet Samples (100 to 600 ppm SO,) substitute 0.01 normal
NaOH for the 0.1 normal NaOH in tne above procedure. Follow
the same procedure as above.
The following equation can be used to calculate the S02
concentration:
10,000 X (NaOH Normality)
SO, ppm = '
Meter Volume ft.-3
Note: Add the indicator within 15 minutes of running_the
test. If the indicator is added at an earlier time, it may
be destroyed by the hydrogen peroxide in the impinger.
152
-------
APPENDIX B-3
GENERAL PROCEDURE FOR PREPARING A LEACHATE OF A SOLID
PROMULGATED BY THE INDIANA STATE BOARD OF' HEALTH, SOLID
WASTE MANAGEMENT SECTION 9-13-74.
1. Weigh some convenient amount of sample (10-20 gm) into
a tared dish, dry at 103° C for one hour; cool, reweigh
to determine the amount of moisture in the sample.
2. Place the dried sample in a flask, add distilled water
or rainwater (500-1000 ml), and place on a magnetic
stirrer for two (2) hours (or some other period that
may be specified by the engineer who submitted the
sample).
3. Filter the sample, then dry and weigh the residue to
determine the percentage of insoluble material.
4. Retain the leachate (filtrate) in a capped bottle, and
make the determinations for all parameters using this
solution.
5. Calculate all results on a dry basis.
6. Record all steps, times, weights, etc. throughout the
entire process.
7. This type of sample should have a high priority in the
order of analysis.
153
-------
APPENDIX B-4
ANALYTICAL METHODS'
Listed Below- are various- physical and analytical methods
which. were employed in testing samples from the BAHCO GAS Gleaning
Project at Rickenbacker Air Force Base:
Particle Size CSub-Sieve) - BAHCO micro-particle classifier as per
ASTM procedures
Particle Size (.Sieve) - U.S. Standard Sieves as per ASTM
procedures
Specific Gravity Determination Use of calibrated cement pycro-
meter and ASTM procedure
Bulk Density - Use of ASTM procedure for compacted bulk density
Thermo-gravimetric Analysis - Used in analyzing solids from
scrubbing process; limestone and lime samples for CaS04-2H20,
CaS03'%H2O/ Ca(OH)2/ MgCOs, CaC03 and loss on ignition (see
detailed description supra.)
Total Sulfate Analysis - Standard gravimetric procedure for total
sulfate measurements. Ref. Scott's Standard Methods of Chemical
Analysis
Total Calcium Content - Research.-Cottrell analytical procedure
using EDTA titrant and Hydroxy Napthol-Blue indicator
TQT:aT'-Matrnfe3;i:tim' Content - ResearcH-Cottrell analytical procedure
using' EDTA titrant and Erichrome BlacK. "T" indicator
' AlRallni'ty"De,teTraiaiatit>n - Used ResearcIL-Cottrell analytical
inetRods' for' determining alkalinity- present in lime and slurry
samples^
.,-^ np^rmination - Conducted as: per Agentometric procedure
outlined in Standard Methods of Water and Wastewater Analysis
154
-------
Coal Analysis Methods as per those, specified by U..SU Bureau of
Mines publication PB-209---O36,. * Joistrumen tat Ion used for the
various: tests are as: follows::
Percent Sttlfur in coal Lec'o Sulfur Analyzer
B.T.U. Values - Parr Calorimeter
Percent Carbon, Hydrogen, Nitrogen - Perfcin Elmer 24Q-Analyzer
Trace Elemental Analysis - Methods used to determine concentrations
of Eg, Cd, PB and Cr were derived from Varian Te.ch.tron publication
85-100224-00 and Jarrell Ash. reference material dealing with flame-
less Atomic Absorption Spectroscopy. Methods from these sources
were employed in conjunction with a Model 850 Jarrell Ash. Atomic
Absorption Spectrophotometer
155
-------
APPENDIX C
MATERIAL BALANCE TEST DATA
C-l Coal
C-2 Lime. Analyses:
C~3 Lime/SO2 Stoicfiiometry-, Lime Utilization and
SO2 Emission Rates
C-^4 Slurry- Cnemical Analyses
C-5 A Procedure for Determining Gas Flow Rate
and the Lime Feeder Calibration Curve
156
-------
TABLE C-l MATERIAL BALANCE COAL ANALYSES
Ui
1976
DATE
3-21
3-30
4-29
5-1
5-19
5-20
5-26
5-27
CARBON
67.17
67.29
65.95
63.52
65.34
62.91
62.21
61.04
MOISTURE
7.50
7.97
8.90
11.80
6.10
8.80
8.10
9.58
* As-Fired
SULFUR
ORGANIC
1.83
1.74
1.73
1.68
1.78
1.70
1.31
0.95
PYRITIC
1.36
1.19
1.02
1.23
0.70
1.35
0.69
0.66
SULFATE
0.26
0.31
0.39
0.38
0.77
0.66
0.64
0.40
ASH BTU/Lh*
6.88
6.64
7.13
6.84
8.65
8.67
10.86
11.70
12,265
12,164
11,957
11,598
11,748
11,359
11,167
10,898
TABLE C-2 MATERIAL BALANCE LIME ANALYSES
1976
DATE
3-30
5-19
5-20
5-21
5-26
5-26
TIME
1500
1045
1030
1100
0900
1100
ALKALINITY
AS CaQ
88.26
89.82
82.77
84.34
74.14
74.70
DATE
5-26
5-26
5-27
5-27
5-28
5-28
TIME
1300
1500
0900
1300
0900
1100
% ALKALINITY
AS CaO
76.38
76.94
73.33
81.87
79.63
73.86
-------
TABLE C-3
00
MATERIAL BALANCE TESTS RESULTS
LIME STOICHIOMETRY UTILIZATION AND SO2 EMISSIONS
1976
Date
3-30
5-19
5-26
5-26
5-26
5-27
5-27
Time
1045
0900
0900
1300
1500
1300
1500
Outlet
156
24
5
8
8
5
5
SO2 Removal
Efficiency %
87.59
94.39
98.98
98.23
97.85
98.25
98.23
Stoichi-
ometry
0.876
0.955
1.096
1.086
1.072
1.043
1.030
SO2 Emissions,
lbs/106 BTU
0.615
0.289
0.045
0.078
0.095
0.060
0.061
Lime
Utilization
100.0
99.4
89.8
90.5
91.2
94.4
94.8
Lime Feed
Rate, Ibs./hr.
440
611
146
164
174
182
194
-------
TABLE C-4
MATERIAL BALANCE CHEMICAL ANALYSES (WT.%)
1976
Date
3-18
3-18
3-29
3-29
3-30
3-30
3-31
3-31
4-8
5-19
5-19
5-19
5-19
5-20
5-20
5-20
5-21
5-21
5-21
5-25
5-25
5-25
5-25
5-26
5-26
5-26
5-26
Time
0930
0945
1630
1630
1600
1600
1500
1500
1530
0945
0945
1700
1700
0845
0915
1015
0930
1000
1315
1300
1300
1500
1500
0900
0900
1100
1100
Location*
D
S
S
D
S
D
S
D
D
D
S
D
S
S
Pond
D
S
D
S
D
S
D
S
D
S
Pond
S
% Acid
Insolubles
4.87
5.03
2.44
3.32
3.15
%CaSOQ-
2H;,0
3.48
3.38
3.34
8.69
7.
2.
1.
91
20
80
2.17
1.12
50
82
1
1
4.11
8.83
2.52
3.17
3.24
3.42
3.47
4.37
3.76
3.50
4.23
10.63
9.87
78.13
70.60
56.39
49.70
59.16
60.35
27.95
50.89
50.17
35.12
34.88
23.89
37.03
29.63
46.35
40.86
51.37
47.31
46.11
45.63
45.87
34.88
34.88
70.48
_ _ ~ ^ _
80.83
80.18
15.06
24.02
37.71
46.24
35.85
35.70
63.09
40.86
46.89
56.64
58.07
70.26
55.20
64.52
47.32
50.18
40.50
42.30
40.15
40.15
38.00
53.06
51.62
24.37
(OH)
32.02
55.92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.14
0.12
0
0
0
0
0
0
0
0
0.80
,02
,14
,12
.84
.14
0.80
1.25
0.84
10.46
10.23
9.63
9.55
7.28
6.48
5.12
1.
1.
1.
2.
1,
% Ignition
Loss
0.73
0.80
0.32
0.42
0.75
0.74
0.25
0.38
1.50
6.14
* D - Dissolving Tank
S - First Stage Seal Tank
-------
TABLE C-4
MATERIAL BALANCE CHEMICAL ANALYSES (WT.%) (CONT.)
1976
Date
5-26
5-26
5-26
5-26
5-26
5-26
5-27
5-27
5-27
5-27
5-27
5-27
5-28
5-28
5-28
5-28
5-28
Time
1100
1300
1300
1300
1500
1500
0900
0900
1300
1300
1500
1500
0900
0900
0900
1100
1100
Location*
D
Pond
D
S
D
S
D
Pond
D
S
D
S
Pond
D
S
D
S
% Acid
Insolubles
2.29
3.36
4.56
4.70
4.31
5.25
4.89
3.90
5.27
5.74
5.67
5.67
2.78
5.68
5.75
4.20
4.35
%CaSO4 -
2HnO
31.78
70.96
29.63
29.15
27.48
27.71
22.46
49.93
23.68
23.68
25.09
25.09
61.40
28.43
24.13
26.28
24.33
%CaSO3
0.5H,0
55.92
24.37
57.35
57.35
63.80
60.22
65.24
40.86
64.52
63.80
62.37
62.37
29.39
61.65
62.37
61.87
63.97
%Ca(OH).
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
%CaCO,
6.60
2.96
6.37
6.48
6.37
6.03
89
73
64
87
89
07
62
5.91
5.75
5.12
5.23
% Ignition
Loss
* D
S
Dissolving Tank
First Stage Seal Tank
-------
APPENDIX C-5
GAS FLOW ESTIMATION FROM FAN AND MOTOR PERFORMANCE CURVES
AND LIME FEEDER CALIBRATION
The following figures, A for fan motor current vs. horse-
power output, B, fan horsepower vs. flow, were used extensively
during the test program to determine the scrubber gas flow rate.
Fan motor current, pressure differential and gas temperature-
measurements were routinely taken. Gas flow was determined as
follows: First determine the motor horsepower from Figure A,
based on the observed fan motor current draw. Then determine
the gas flow rate from Figure B, based on the motor horsepower,
gas temperature and pressure differential observed.
Figure C illustrates the calibration of the lime feeder.
161
-------
100 200 300 400 500 600 700 800 900
HORSEPOWER OUTPUT
FIGURE A: Fan motor current vs. horsepower.
162
-------
900
u>
cc
111
O
Si
CO
DC
O
I
800
700
600
500
400
GAS TEMP. 275°F
25 IN. W.C.
GAS TEMP. 475°F
19-25 IN. W.C.
NOTE:
GAS DENSITY
@ 275°F - 0.0540 LBS/FT3
@475°F = 0.0426 LBS/FT3
300
I
20
30 40 50 60 70
GAS FLOW TO SCRUBBER, SCFMX10 3
FIGURE B: Fan horsepower and gas flow rates.
80
-------
i
CO
_i
LU
<
oc
Q
LU
LU
U.
LU
5
TOO -
FEED RATE = 27.12 x METER READING
_L
10
20 30 40
METER READING (%)
50
FIGURE C : Lime Feeder Calibration Curve.
164
-------
APPENDIX D
LIME TEST DATA
D-l Lime/S02 Stoichiometry, Lime Utilization and SC>2
Emission Rates
D-2 Lime Test pH Data
D-3 Slurry Chemical Analyses
D-4 Scrubber Operating Data
D-5 Inlet and Outlet SO2 Concentration Data
D-6 Numerical Results of the Significant Effect Analysis
165
-------
TABLE D-l LIME STATISTICAL TESTS - STOICHIOMETRY AND S02 EMISSIONS
AVERAGE
TEST
NO.
1
2R *
3
4
5R
6
7R
8
9
10
11
' 12
13
14R
15
16
17
18
19
20
21
so2
INLET
1110
1500
1000
992
1140
1100
1350
1275
910
1075
1150
842
940
1200
1500
1095
1000
1155
348
370
1075
ppm
OUTLET
104
550
28
40
365
81
494
742
58
41
70
40
174
546
25
340
115
14
348
370
172
COAL FIRING
RATE lbs/hr
14,098
13,213
14,994
13,991
13,361
13,214
13,368
13,229
12,712
13,229
14,627
15,522
10,789
14,983
16,639
10,448
15,878
20,313
12,539
8,954
10,430
COAL HEATING
VALUE, Btu/lb
11,012
11,143
11,233
11,861
11,143
10,877
10,687
11,143
11,861
11,585
11,585
10,958
10,958
10,664
11,656
10,664
10,819
11,261
10,819
11,107
11,145
AVERAGE
LIME/SO2
STOICHTOMETRY
0.936
0.596
1.003
0.985
0.654
. 0.928
0.598
0.360
0.967
0.991
0.976
0.994
0.805
0.545
1.061
0.658
0.917
1.080
0.667
0.604
0.85.5
AVERAGE LIME
UTILIZATION
95.4
100.0
96.7
96.8
98.7
99.7
100.0
100.0
95.8
96.5
95.9
95.1
98.0
100.0
92.5
100.0
97.5
91.2
95.1
95.6
96.2
AVERAGE
SO- EMISSIONS
ff /-
lbs/10DBtu
0.342
1.566
0.072
0.143
2.075
0.237
1.497
3.019
0.239
0.135
0.217
0.150
0.903
1.424
0.058
1.944
0.422
0.028
1.052
1.803
0.737
* The "R" designation indicates a repeat test run.
-------
TABLE D-2 AVERAGE pH DATA FOR LIME STATISTICAL TESTS
TEST NO.
1
2R*
3
4
5R
6
7R
8
9
10
11
12
13
14R
15
16
17
18
19
20
21
DISSOLVER pH
7.4
6.1
8..6
6.4
9.4
6.3
7.5
7.1
8.9
5.5
5.0
7.6
9.6
8.2
9.6
7.4
8.2
DROP COLLECTOR pH
FIRST STAGE SECOND STAGE
6.2
5.3
6.5
5.8
5.4
7.5
5.4
7.8
5.8
5.7
8.0
5.0
4.3
7,
5,
,0
.0
7.1
6.2
5.6
4.5
5.6
5.6
5.0
7.6
5.6
5.2
8.4
5.1
8.0
5.5
5.4
8.2
4.8
4.4
8.0
5.0
8.0
7.2
5.9
6.0
LEVEL TANK pH
FIRST STAGE SECOND STAGE
8.0
5.8
8.5
8.1
6.2
9.3
5.9
7.0
7.0
5.1
7.9
8.6
4.9
8.3
7.9
7.1
9.3
5.3
5.2
7.7
8.5
8.7
9.3
5.9
8.1
The "R" designation indicates a repeat test run.
-------
TABLE D-3 LIME TESTS, CHEMICAL ANALYSES (HT.%) AND UTILIZATION
1976
Date
12-13
12-13
12-13
12-13
12-14
12-14
12-14
12-14
12-15
12-15
12-15
12-15
12-15
12rl5
12-15
12-15
12-15
12-15
12-15
12-15
12-16
12-16
12-16
12-16
12-17
12-17
12-17
12-17
Time
1930
1930
2200
2200
1800
1800
1945
1945
1200
1200
1300
1300
1715
1715
2030
2030
2130
2130
2230
2230
1630
1630
1815
1815
1500
1500
1645
1645
*
Location
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
Test
15
15
15
15
7
7
7
7
9
9
9
9
4
4
4
4
5
5
5
5
1
1
1
1
2
2
2
2
% Acid
Inaolubles
4.49
3.93
3.67
4.52
4.71
4.83
4.45
4.87
4.25
4.32
4.14
4.33
3.90
4.24
4.62
4.35
4.26
4.31
4.37
4.32
4.86
4.50
4.74
4.82
4.79
4.32
4.69
4.37
%CaSC>4
2H2O
29.29
36.79
28.43
26.90
35.60
39.18
32.97
33.21
28.05
27.91
28.62
27.67
23.94
23.99
25.18
26.04
28.72
31.54
29.15
31.63
21.50
22.12
26.19
27.48
36.55
43.96
39.90
46.59
% CaSO3 -
0.5H2O
60.65
53.91
57.78
61.65
55.92
54.48
57.35
60.22
66.10
66.82
66.67
65.67
71.55
70.54
67.68
67.53
64.23
61.51
61.22
60.79
73.55
67.39
66.96
65.95
56.35
51.87
55.92
47.03
% Ca(OH) j
1.58
1.91
1.58
1.19
0.82
0.62
1.03
0.62
2.06
1.73
0.95
0.95
1.03
0.41
0
0.41
1.03
1.03
1.03
1.03
1.77
1.77
1.44
1.23
1.11
1.07
1.27
1.48
% CaCO^
2.55
2.25
3.41
3.66
2.73
2.50
3.30
1.41
0.80
0.45
1.29
1.64
1.41
2.30
1.48
1.25
0.91
0.73
0.91
0.91
1.14
1.09
1.07
1.18
1.09
0.64
0.84
0.50
% Ignition
Loss
4.73
5.67
5.40
5.53
4.83
4.66
4.66
5.01
5.33
6.12
5.83
5.10
5.32
4.92
4.36
5.26
3.96
5.00
3.84
4.16
5.30
4.80
4.43
4.44
3.54
3.61
4.04
3.68
% Lime
Utilization
93.18
92.90
91.71
92.33
94.35
95.12
93.13
96.71
94.96
96.06
96.37
95.82
96.12
96.01
97.84
97.40
96.65
96.89
96.55
96.61
95.16
94.92
95.70
95.41
96.16
96.93
96.30
96.21
D - Dissolver tank
S - First stage drop collector
-------
TABLE D-3 LIME TESTS, CHEMICAL AHALYSES (WT.%) AND UTILIZATION (CONT.)
a\
1976
Date
12-17
12-17
12-17
12-17
12-18
12-18
12-18
12-18
12-18
12-18
12-18
12-18
12-19
12-19
12-19
12-19
12-19
12-19
12-19
12-19
12-20
12-20
12-20
12-20
12-20
12-20
12-20
12-20
Tirae
2000
2000
2045
2045
1500
1500
1700
1700
2030
2030
2215
2215
1800
1800
2000
2000
2300
2300
2350
2350
1715
1715
1845
1845
2200
2200
2345
2345
Location*
D
S
D
S
D
S
D
S
D
3
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
Test
3
3
3
3
10
10
10
10
11
11
11
11
12
12
12
12
13
13
13
13
14
14
14
14
18
18
18
18
% Acid
Insolubles
386
4.03
3.76
4.09
4.35
4.48
4.74
4.79
4.62
4.65
4.35
4.36
2.79
3.18
2.90
3.14
4.45
5.14
4.95
5.18
5.94
3.33
4.87
2.65
4.45
3.92
4.28
3.23
%CaSO4"
2H2Q
47.59
45.39
41.81
41.57
19.11
18.49
19.21
19.11
17.97
19.02
23.03
24.90
17.20
16.15
16.19
13.33
13.14
20.31
20.31
22.60 '
90.69
69.91
89.60
94.37
63.27
77.70
72.01
75.26
% CaSC>3
0.5H0O
** ' A
45.02
48.32
51.04
51.62
76.71
74.84
74.84
74.99
77.57
74.99
71.40
68.82
74.56
78.14
77.20
80.87
81.87
72.69
72.69
70.11
1.58
22.08
3.30
1.43
27.39
16.20
19.07
17.78
% Ignition % Lime
% Ca(OH)0
1.48
1.11
1.11
1.28
1.78
0.99
1.52
1.44
1.44
1.44
1.85
2.06
1.03
0.66
0.66
0.66
0
0
0.41
0.41
0
2.30
: 0
0
2.55
3.54
3.99
2.63
% CaCCh
0.50
0.50
0.39
0.45
0.98
0.84
0.61
0.56
0.56
1.02
0.56
0.45
2.23
2.20
2.66
3.05
1.82
1.43
1.02
0.34
0
0.23
0
0
1.71
0.91
1.14
1.16
Loss
3.43
3.84
4.25
3.93
5.99
5.72
6.26
5.67
6.03
5.39
5.71
5.64
6.47
6.37
6.19
6.68
6.10
5.37
5.39
5.01
1.80
2.20
1.50
1.41
2.72
1.83
1.14
1.76
Utilization
96.16
96.96
97.13
96.72
96.30
96.93
96.29
96.51
96.57
95.89
95.74
95.45
94.93
95.77
95.12
94.70
97.50
97.95
97.74
98.69
100.00
94.54
100.00
100.00
91.84
91.02
89.66
92.43
* D - Dissolver tank
S - First stage drop collector
-------
TABLE D-3 LIME TESTS, CHEMICAL ANALYSES (WT.%) AND UTILIZATION (CONT.)
1977
Date
2-10
2-10
2-10
2-10
2-13
2-13
2-13
2-13
2-13
2-13
M 2-13
Lj 2-13
o 2-14
2-14
2-14
2-14
2-15
2-15
2-15
2-15
2-15
2-15
2-15
2-15
2-16
2-16
2-16
2-16
2-17
2-17
Time
2000
2000
2100
2100
1330
1330
1430
1430
1845
1845
1945
1945
1430
1430
1515
1515
1300
1300
1400
1400
1930
1930
2030
2030
1800
1800
1900
1900
1200
1200
Location*
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
Test
20
20
20
20
14R
14R
14R
14R
16
16
16
16
21
21
21
21
19
19
19
19
17
17
17
17
6
6
6
6
5R
5R
% Acid
Insolubles
4.42
4.35
4.63
4.96
4.27
8.96
6.59
10.85
6.74
9.82
6.93
8.18
5.39
5.87
5.10
5.16
4.40
4.92
4.26
4.76
3.71
4.53
3.84
4.35
4.51
4.28
3.99
4.34
4.74
4.54
%CaSO,,
4
2H;0
60.49
63.07
56.58
56.24
89.60
85.10
89.60
89.16
82.67
80.95
79.42
79.18
46.35
56.15
49.41
51.13
30.92
43.15
31.06
37.13
22.98
20.11
21.79
22.65
23.18
22.50
20.98
21.54
6.69
7.50
% CaSO,
O.SH-jO
"ii"
33.84
29.82
37.71
34.98
3.58
6.45
3.58
2.58
8.89
9.03
12.62
12.47
4Q.03
35.70
42.87
39.86
64.09
48.61
63.66
55.97
70.97
71.40
68.39
68.39
78.09
77.36
77.43
77.00
87.75
84.59
% Ca(OH)2
0.82
1.89
0.62
0.70
0
0
: 0
0
0
0
0
0
0.98
1.69
1.15
2.47
0.98
1.03
0.41
0.90
0.62
1.02
0.72
0
0
0
0
0
0
% CaCO3
1.34
1.66
1.55
1.16
0
0
0
0
0
0
0
0
0
4.09
0.45
0.22
1.11
2.84
0.34
0.34
0.23
0.80
0.73
0.57
0.19
0.28
0.57
0
0
3.98
% Ignition
Loss
2.61
2.36
2.83
2.42
1.19
3.15
2.82
3.06
2.67
2.43
3.03
3.79
4.62
2.60
2.99
2.47
4.16
2.97
5.23
3.11
5.16
3.57
4.11
3.45
5.36
5.14
5.29
5.05
3.68
3.52
% Lime
Utilization
96.17
93.42
96.30
96.60
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
91.76
95.77
97.16
93.83
90.22
97.50
98.64
97.93
97.62
96.89
97.72
99.74
99.62
99.22
100.00
100.00
94.62
-------
TABLE D-3 LIME TESTS, CHEMICAL ANALYSES (WT.%) AND UTILIZATION (CONT.)
1977
Date
2-17
2-17
2-17
2-17
2-17
2-17
2-17
2-17
2-17
2-17
2-18
2-18
2-18
2-18
Time
1300
1300
1815
1815
1915
1915
2230
2230
2330
2330
1145
1145
1230
1230
Location
D
S
D
S
D
S
D
S
D
S
D
S
D
S
Teat
5R
5R
2R
2R
2R
2R
8
8
8
8
7R
7R
7R
7R
% Acid
Insolubles
4.38
4.76
4.84
5.07
4.77
4.51
5.54
4.82
5.69
4.63
4.44
5.25
4.61
4.52
%CaSO. .
2H20
5.88
7.07
9.32
9.56
8.94
9.22
14.96
15.60
14.10
16.72
16.72
18.64
18.01
20.21
% CaCO.
O.5H2Q'
87.61
88.90
84.88
84.88
85.31
83.73
76.56
76.13
79.58
78.14
78.14
74.27
75.27
74.70
of. % Ca(OH)2 %CaCO3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Ignition
Loss
3.52
3.82
3.40
3.79
3.45
3.13
2.97
2.42
3.18
3.20
3.43
2.84
2.92
3.48
% Lime
Utilization
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
-------
TABLE D-4 SCRUBBER OPERATING DATA LIME TEST
K)
AVERAGE
SCRUBBER GAS
TEST
NO.
1
2R
3
4
5R
6
7R
8
9
10
11
12
13
14R
15
16
17
18
19
20
21
FLOW
a _
SECOND STAGE
PRESSURE DROP(in.H90)
SLURRY PUMP SLURRY
DATE (acfmxlO* 120°F) RATE (gpm)
12-16-76
2-17-76
12-17-76
12-15-76
2-17-76
2-16-76
2-18-76
2-17-76
12-15-76
12-18-76
12-18-76
12-19-76
12-19-76
2-13-76
12-13-76
2-13-76
2-15-76
12-20-76
2-15-76
2-10-76
2-14-76
56.2
46.2
47.6
65.4
68.1
46.3
47.7
66.1
68.6
55.7
57.9
70.2
67.6
45,9
49.3
70.2
64.5
50.4
45.2
53.4
54.9
2780
2350
2130
3100
1675
2500
1700
3000
2320
2230
2700
2800
2120
2500
2400
2230
1425
2400
1475
2025
1975
% SOLIDS
10.8
14.2
17.0
16.9
13.0
11.8
12.9
14.2
14.3
10.5
8.5
7.5
5.2
4.9
7.8
6.1
66.7
6.1
6.5
10.4
8.1
TOTAL
16.0
14.0
16.8
24.0
17.6
21.5
18.4
21.8
19.1
17.5
18.0
16.4
18.8
25.5
19.3
19.7
17.0
25.0
9.2
16.0
13.8
FIRST
STAGE
7.0
8.5
5.0
11.0
12.8
14.5
7.6
8.8
9.6
9.5
9.5
8.0
7.0
12.5
12.7
10.5
6.0
13.0
4.2
8.0
6.8
SECOND
STAGE
9.0
5.5
11.8
13.0
4.8
7.0
10.8
13.0
9.5
8.0
8.5
8.0
11.8
13.0
6.6
9.2
11.0
12.0
5.0
8.0
7.0
_ . 3
AVERAGE LIQUID PICKUP (gem) AVERAGEL/G (gal./lO acf
FIRST STAGE
1200
1005
525
1400
1110
1065
735
1170
1290
1050
1080
1050
600
1080
780
1050
750
1110
390
820
900
SECOND STAGE FIRST STAGE
600
180
900
840
90
330
585
750
960
570
600
600
1050
1065
450
300
630
1260
150
300
330
21.4
21.8
11.0
21.4
16.3
23.0
15.4
17.7
18.8
18.9
18.7
15.0
8.9
23.5
15.8
15.0
11.6
22.0
8.6
15.4
16.4
SECOND STAGE
10.7
3.9
18.9
12.8
1.3
7.1
12.3
11.3
14.0
10.2
10.4
8.5
15.5
23.2
9.1
4.3
9.8
25.0
3.3
5,6
6.0
-------
TABLE D-5 LIME TESTS, S02 CONCENTRATION DATA
Test
No.
1
2R
3
4
5R
6
7R
8
9
10
11
12
13
14R
15
16
17
18
19
20
21
DuPont/
Wet Test
D
D
W
D
D
D
W
D
D
W
D
W
D
D
D
D
D
D
W
D
D
W
D
W
D
D
W
D
W
D
W
Concentration (ppin)
Inlet Outlet Corrected Outlet
1110
1500
1000
992
1140
1100
1350
1275
1000
1155
1045
950
934
1075
104
28
40
365
75
910
1075
1150
842
940
1200
__-.
1500
1095
58
41
70
40
174
_
546
25
«».._
90
115
20
348
370
104
550
28
40
365
81
494
742
58
41
70
40
174
546
25
340
97
115
14
348
370
172
Overall
Efficiency (%)
89.35
59.60
96.79
95.36
64.51
92.50
59.77
35.98
92.67
95.60
93.30
94.56
78.83
54.50
98.17
65.84
89.45
87.49
98.67
63.40
57.78
82.27
173
-------
TABLE D-6 NUMERICAL RESULTS OF THE SIGNIFICANT EFFECT ANALYSIS
Independent Variables
2nd Stage Slurry
System Inlet Gas 1st Stage 2nd Stage Slurry Mill Pump System Concent.
Responses Flow Rate A p A p Flow Rate Rate Stoichiometry Volume (% Solids
SO2
Removal
Efficiency -.0149 .0084 -.0423 -.0486 .0559 .3429 .0446 -.0501
1st Stage
Liquid
Pickup 152.5 190.0 17.5 430,0 55.0 25.0 62.5 115.0
2nd Stage
Liquid
Pickup -17.5 -155.0 JL5JLJ 115.0 -12.5 192.5 55.0 -152.5
1st Stage
Drop
Collector
pH .20 -.275 -.75 -.30 .075 1.825 -.325 .10
Total
Alkalinity
in Scrubber
Slurry .0095 -.01885 -.0023 -.0076 .0093 -037 .00043 -.02j
Note: Independent variables which have an effect at the 95% confidence level on system responses are under
Minimum
Significant
Effect (95%
Conf. Level)
0.0996
172.4
233.6
0.6
0.02167
lined
in this table.
-------
TABLE D-6 NUMERICAL RESULTS OF THE SIGNIFICANT EFFECT ANALYSIS (CONT.)
System Inlet Gas
Responses Flow Rate
M Total
-J Alkalinity
01 in Drop
Collector
Slurry -.0057
% Sulfate
in Scrubber
Slurry -12.09
% Sulfate
in Drop
Collector
Slurry -14.14
2nd Stage
1st Stage 2nd Stage Slurry
A p A p Flow Rate
-.0126 -.0059 -.0099
7.57 9.58 15.9
5.35 10.09 14.33
Independent Variables
Slurry
Mill Pump System Concent.
Rate Stoichiometry Volume (% Solids)
.0105 .0316 .0056 .018
14.85 -1.59 -5.63 -23.71
-11.77 -2.58 -7.66 -25.71
Minimum
Significant
Effect (95%
Conf . Level)
0.0256
28.78
28.7
-------
APPENDIX E
REAGENT AND COAL DATA
Er-1 Coal Properties
EL2 Lime Specifications
E~3 Lime Analyses
E-4 Limestone Specifications
E-5 Limestone Analyses
176
-------
APPENDIX E-l COAL PROPERTIES
COAL (Data are reported on an as-fired basis)
Supplier: Peabody Coal Co., Columbus, Ohio
Type: Mixture of Sunny Hill and Broken Arrow
mines, New Lexington, Ohio
Analysis;
Sunny Hill Coal
6 to 12%,
9.8 av. wt.% moisutre
11.2 av. wt.% ash
2.7 av. wt.% sulfur
10,400 to 12,200, Btu/lb
11,080 av.
Broken Arrow Coal
7 to 12%,
9.2 av. wt.% moisture
6.2 av. wt.% ash
3.5 av. wt.% sulfur
11,900 to 12,800, Btu/lb
12,200 av.
APPENDIX E-2 LIME SPECIFICATIONS
Source: Black River Mine
Butler, Kentucky
Specifications: Quicklime
>_ 83% CaO
72°F temperature rise in three minutes and slaking
reaction complete in ten minutes or less when added
to water at a four to one ratio weight.
Particle size not to exceed 3/4 inch.
Should be freshly burned and substantially free of
carbonate solids and silicious residue. Amount of
such materials (insolubles) not to exceed 5%.
177
-------
APPENDIX E-3 LIME ANALYSES
Total Alkalinity
Date
12-13-76
12-14-76
12-15-76
12-16-76
12-17-76
12-18-76
12-19-76
12-20-76
2-10-77
2-13-77
(as % CaO)
84.90
80.86*
85.90
74.70*
75.26*
75.71*
75.26*
75.26*
95.09
95.09
Total Alkalinity
Date
2-13-77
2-14-77
2-15-77
2-16-77
2-17-77
2-18-77
3-14-77
3-15-77
3-16-77
3-17-77
(as % CaO)
96.21
96.21
96.21
96.21
96.21
96.21
87.92
88.48
88.48
88.48
*These samples contained less CaO than required by the reagent
specification listed in Appendix E-2.
178
-------
APPENDIX E-4 LIMESTONE SPECIFICATIONS
Source: Armco Piqua Quarry
Piqua, Ohio
Specifications: YA-stonedust (pulverized limestone)
13% MgCO^ quarry specifications
Particle size distribution =
100% through 40 mesh (375u)
99.9% through 60 mesh (250y)
99.7% through 80 mesh (177y)
99% through 100 mesh (149y)
85% through 200 mesh (74 u)
70% through 30 mesh (47 u)
APPENDIX E-5 LIMESTONE ANALYSES
Date
5-11-77
5-12-77
5-16-77
5-17-77
5-19-77
5-20-77
5-23-77
5-24-77
5-25-77
% CaCO.
89.5
89.6
90.4
86.6
87.9
87.2
86.6
86.6
86.4
8.6
8.4
8.9
10.4
9.2
10.0
10.7
9.9
11.8
179
-------
APPENDIX F
LIME. VERIFICATION TEST DATA
F~l Stoichioroetry and SD2 Emissions
F-2 pE Data
F-3 Chemical Analyses- and Utilization
F4 Scrubber Operating Data
F-5 SO2 Concentration Data
180
-------
TABLE F-l LIME VERIFICATION TESTS - STOICHIOMETRY AND SO0 EMISSIONS
00
AVERAGE
TEST
NO.
22
23
24
25
26
27
28
29
30
31
32
33
34
36
37
so2
INLET
800
910
720
660
740
735
725
945
950
855
550
650
725
700
665
ppm
OUTLET
309
355
386
265
207
229
271
289
265
239
61
21
176
48
48
COAL FIRING
RATE* Ibs/hr
9916
7822
7755
7658
9526
9702
9225
12,810
9694
9668
6813
7492
7955
7976
10,823
COAL HEATING
VALUE, Btu/lb*
10,619
11,082
11,082
11,082
11,082
11,082
11,082
11,139
11,139
11,139
10,707
10,707
10,707
10,707
10,707
AVERAGE
LIME SO2
STOICHIOMETRY
0.576
0.572
0.417
0.565
0.710
0.662
0.592
0.674
0.698
0.710
. 0.901
0.985
0.746
0.944
0.938
AVERAGE LIME
UTILIZATION
99.4
99.1
98.9
99.0
99.2
99.0
99.2
97.2
98.1
97.2
97.9
98.0
98.2
98.0
98.1
AVERAGE
SO2 EMISSIONS
lbs/106Btu
1.768
2.337
2.620
1.844
1.123
1.260
1.616
0.880
1.185
1.317
0.498
0.171
1.288
0.362
0.280
* As Fired
-------
TABLE F-2 AVERAGE pH DATA FOR LIME VERIFICATION TESTS
DROP COLLECTOR pH
TEST NO. DISSOLVER pH
CO
ro
22
23
24
25
26
27
28
29
30
31
32
33
34
36
37
5.4
6.2
5.8
5.7
6.6
7.1
8.0
7.9
8.3
8.4
8.5
8.6
8.3
8.4
8.6
FIRST STAGE
4.8
-
4.9
5.4
5.4
5.0
5.2
4.9
4.6
5.0
6.0
7.0
5.1
5.9
6.3
SECOND STAGE
5.4
4.6
4.7
5.4
5.2
4.9
5.1
5.0
4.8
4.7
5.1
5.4
5.3
5.7
5.2
LEVEL TANK pH
FIRST STAGE fiRPOWn fiTAflF
_
6.2
5.5
5.4
6.0
7.1
7.4
7.6
7.7
8.5
9.0
8.6
8.4
8.0
8.6
4.6
6.4
5.6
6.4
7.6
8.0
8.4
8.3
8.6
8.8
8.8
8.3
8.4
8.5
-------
TABLE F-3LIME VERIFICATION TESTS, CHEMICAL ANALYSES (WT.%) AND UTILIZATION
1977.
Date
3-14
3-14
3-14
3-14
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
£ 3-15
S 3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-15
3-16
Time
1800
1800
1900
1900
0930
0930
1030
1030
1245
1245
1400
1400
1445
1445
1600
1600
1645
1645
1800
1800
1820
1820
1920
1920
2000
2000
2100
2100
1545
Location*
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
0
S
D
Test
22
22
22
22
23
23
23
23
24
24
24
24
25
25
25
25
26
26
26
26
27
27
27
27
28
28
28
28
29
% Acid
Insolubles
4.01
4.83
4.06
3.74
3.41
3.87
3.53
4.39
3.43
3.95
3.25
4.42
4.64
6.59
4.13
4.75
4.29
4.11
4.27
4.73
3.77
3.49
4.30
3.90
6.50
3.37
6.63
3.71
6.15
%CaSOd.
*t
42.81
42.96
40.76
42.91
47.50
47.69
45.44
47.16
46.68
47.64
47.45
48.41
47.74
48.48
49.07
50.08
47.98
49.55
47.74
50.46
49.17
48.98
46.59
50.27
46.59
48.31
46.35
47.83
88.97
%CaSO3-
0.5H20
50.76
49.75
54.05
51.04
48.18
46.60
47.74
45.45
46.74
46.60
45.59
45.45
45.74
45.59
45.59
44.88
46.31'
43.87
46.17
44.88
45.74
44.88
47.17
44.88
47.17
45.16
47.75
45.16
3.73
% Ca(OH)2
. 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
. 0
0
0
0
0
0
0
0
0
0
0
% CaCO,
Ignition
Loss
0
0.43
0.55
0.63
0.54
0.64
0.61
0.59
0.68
0.77
0.77
0.55
0.68
0.56
0.56
0.77
0.86
0.52
0.86
0.57
0.77
0.50
0.57
0.68
0.57
0.56
0.45
0.39
1.14
3.80
2.61
3.15
2.13
2.86
2.26
2.31
2.26
2.29
2.40
2.06
2.26
2.08
2.13
2.10
2.19
2.15
2.01
2.13
2.13
2.22
2.13
2.17
2.29
2.17
2.19
2.36
2.31
1.70
% Lime
Utilization
100.00
99.33
99.17
99.03
99.18
99.01
99.05
99.07
98.94
98.81
98.79
99.14
98.94
99.13
99.13
98.81
98.67
99.18
98.67
99.12
98.81
99.22
99.11
98.59
99.11
99.12
99.30
99.38
97.96
* D - Dissolver tank
S - First stage drop collector
-------
TABLE P-3 LIME VERIFICATION TESTS, CHEMICAL ANALYSES (WT.%) AND UTILIZATION (CONT.)
*»
1977
Date
3-16
3-16
3-16
3-16
3-16
3-16
3-16
3-16
3-16
3-16
3-16
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
3-17
Time
1545
1645
1645
1730
1730
1830
1830
1900
1900
1950
1950
0945
0945
1100
1100
1130
1130
1245
1245
1500
1500
1600
1600
1700
1700
1800
1800
1845
1845
2000
2000
Location
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
Teat
29
29
29
30
30
30
30
31
31
31
31
32
32
32
32
33
33
33
33
37
37
37
37
36
36
36
36
34
34
34
34
% Acid
Insolubles
5.72
6.94
10.31
3.92
4.17
3.93
5.55
7.55
5.01
5.20
4.26
4.49
4.08
6.06
3.53
4.04
3.06
4.31
2.99
3.28
2.69
3.15
3.06
2.84
3.34
3.21
3.21
3.91
4.01
3.84
3.66
% CaSO4
2H20
66.18
87.83
85.05
79.99
84.91
77.89
71.63
75.36
63.31
69.33
61.88
47.78
35.36
49.12
37.65
54.48
46.02
33.54
42.00
43.29
36.55
35.31
38.18
41.72
32.25
34.07
41.81
31.63
30.53
33.54
35.69
% CaCO3
O.5H2O
22.51
5.54
3.58
16.78
7.46
18.21
20.07
17.35
28.53
24.23
31.11
44.45
61.37
44.02 .
57.78
42.87
46.89
61.37
52.19
50.18
60.94
60.79
'55.06
53.77
63.80
60.36
53.48
61.65
64.81
60.94
60.08
% Ca (OH)2 % CaCO3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4.25
1.14
0
1.59
1.68
1.34
0
1.41
2.16
32
84
36
60
36
1.25
07
36
66
11
18
1.71
16
02
61
14
75
02
25
36
25
1.09
Ignition
Loss
1.95
1.49
5.57
1.38
2.32
2.04
3.15
1.85
3.13
2.31
2.95
3.38
4.05
3.11
3.95
3.31
3.04
4.39
3.95
3.59
4.04
4.78
3.80
3.57
3.91
3.89
4.34
4.18
4.71
3.96
3.89
% Lime
Utilization
92.94
97.98
100.00
97.40
97.04
97.79
100.00
97.60
96.46
97.81
97.03
97.86
97.71
97.88
98.16
98.38
97.89
97.58
98.32
98.19
97.56
98.31
98.45
97.62
98.36
97.44
98.47
98.15
98.04
98.16
98.41
-------
TABLE F-4 SCRUBBER OPERATING DATA LIME VERIFICATION TEST
AVERAGE
SCRUBBER GAS +
SECOND STAGE
TEST 1Q77 FLOW SLURRY PUMP SLURRY
NO. DATE (acfnucl03 $120°F) RATE (gpm) %SOLIDS
H
00
Ul
22
23
24
25
26
27
28
29
30
31
32
33
34
36
37
3-14
3-15
3-15
3-15
3-15
3-15
3-15
3-16
3-16
3-16
3-17
3-17
3-17
3-17
3-17
66.4
62.9
64.3
65.1
63.1
65.2
67.2
47.9
53.2
65.4
65.6
72.1
68.7
70.9
74.4
2400
2200
2350
2100
2650
2650
2250
2600
2550
>3000
2400
2400
2400
2400
2500
14.5
17.7
16.1
18.0
19.1
20.8
19.8
3.2
2.4
2.4
9.2
10.3
10.6
11.5
9.8
PRESSURE DROP
TOTAL
19.5
21.0
23.5
22.3
22.4
21.4
22.8
19.2
17.9
18.0
21.0
20.7
22.5
21.5
19.9
FIRST1
STAGE
8.5
10.5
11.5
10.5
10.5
10.2
11.0
11.8
9.0
8.5
8.0
8.2
13.0
9.0
9.2
(in.H,o)
SECOND
STAGE
11.0
10.5
12.0
11.8
11.9
11.2
11.8
7.4
8.9
9.5
13.0
12.5
9.5
12.5
10.7
AVERAGE LIQUID PICKUP (qpm) AVERAGE L/G
FIRST STAGE
990
960
1035
1005
1020
990
1020
1020
1020
990
1020
1020
1170
780
1005
SECONDTfAGE
180
120
165
150
165
150
180
150
225
180
210
180
390
600
225
FIRST STAGE
14.9
15.3
16.1
15.4
16.2
15.2
15.2
21.3
19.2
15.1
15.5
14.2
17.0
11.0
13.5
(gal ./ i
-------
TABLE p-5 LIME VERIFICATION TESTS, SO2 CONCENTRATION DATA
SO0 Concentration (ppro)
% Removal Efficiency*
Test No.
22
23
24
25
M
oo 26
en
27
28
29
30
31
DuPont/
Wet Test
D
W
D
W
D
W
D
W
D
W
D
W
D
W
D
W
D
W
D
W
Inlet
800
910
720
660
740
735
725
945
950
855'
Midpoint
492
425
338
350
328
382
426
Outlet
340
390
460
425
396
292
408
228
328
252
322
298
272
332
242
305
332
275
379
Corrected Outlet
309
355
386
396
265
207
229
271
272
289
242
265
239
First Stage
39.92
29.31
51.38
46.71
60.82
54.47
44.79
Second Stage Overall
57.20
56.65
41.26
39.74
55.92
70.22
65.54
58.73
22.29 58.58
65.48
26.22 71.09
68.42
60.43
69.03
-------
TABLE P-5 (CONT.) LIME VERIFICATION TESTS, SO2 CONCENTRATION DATA
Test No.
SO0 Concentration (ppm)
% Removal Efficiency *
DuPont/
Wet Test Inlet Midpoint Outlet Corrected Outlet First Stagfe Second Stage
Overall
32
33
34
36
37
D
W
D
W
D
W
D
W
D
W
550
650
725
700
665
1X1 J1.
304
-.
308
261
480
___
378
64
22
185
242
50
55
61
21
176
48
48
~ _ W
41.05
___
49.56
__ _
60.33
24.81
___
37.72
88.17
96.56
73.25
92.48
92.09
-------
APPENDIX G
LIMESTONE TEST DATA
G~l Stoichiometry and S02 Emissions
G-2 pH Data
G3 Chemical Analyses and Utilization
G-4 Scrubber Operating Data
G-5 SOo Concentration Data
^
G-6 Regression Analysis. Results
188
-------
TABLE G-l LIMESTONE STATISTICAL TESTS - STOICHIOMETRY AND SO-) EMISSIONS
oo
10
TEST
NO.
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
AVERAGE
SO2 ppm
INLET OUTLET
438
425
350
290
300
360
410
320
295
390
410
375
254
483
425
350
375
6QO
525
375
335
109
30
33
31
92
23
57
35
70
90
71
90
125
102
194
71
131
107
284
69
55
AVERAGE
COAL FIRING
RATE Ibs/hr
4312
4928
4531
3955
3338
4158
4992
5252
3984
5006
3949
4765
3958
3960
3631
5408
4165
4158
4092
3393
3966
AVERAGE . AVERAGE
COAL HEATING LIME/SO2
VALUE, Btu/lb *STOICHIOMETRY
10,992
11,003
11,168
10,751
11,003
10,646
10,646
10,646
10,646
10,992
10,992
11,894
10,469
11,481
11,481
11,107
11,107
11,107
11,107
10,992
10,992
0.800
1.532
1,547
1.424
085
408
187
298
136
0.859
0.943
1.186
0.625
0.958
0.717
1.193
0.944
1.385
0.593
0.943
1.014
AVERAGE
LIMESTONE
UTILIZATION
91.8
3
2
3
60
58
62
62.4
66.2
71.8
68.1
66.1
87.9
86.6
63.4
77.4
80.8
71.9
66.0
67.2
58.4
66.4
85.1
81.4
AVERAGE
S02 EMISSIONS
lbs/106Btu
1.018
0.181
0.253
0.345
1.232
0.186
0.437
0.342
0.958
0.760
0.674
0.781
1.484
0.753
1.828
0.617
1.513
0.685
2.228
0.781
0.525
*As Fired
-------
TABLE G-2 AVERAGE pH DATA FOR LIMESTONE STATISTICAL TESTS
vo
o
DISSOLVER DROP COLLECTOR pH LEVEL TANK pH
TEST NO.
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
5.8
5.3
5.6
5.8
5.5
5.5
5.1
5.1
5.7
6.2
5.6
7.7
5.9
5.5
6.1
6.2
6.3
6.7
6.2
5.6
5.6
FIRST STAGE
5.2
5.2
5.2
5.6
5.4
5.5
4.9
5.0
5.3
5.8
5.5
6.3
5.3
5.4
5.5
5.8
6.0
6.2
5.8
5.5
5.5
SECOND STAGE
5.3
5.4
5.5
5.9
5.4
5.7
5.1
5.5
5.2
5.6
5.4
6.3
5.2
3.8
5.7
5.7
5.8
6.3
5.7
5.6
5.6
FIRST STAGK
5.7
5.9
6.1
6.4
6.9
6.2
5.6
5.6
6.4
6.0
6.0
6.6
5.4
4.4
6.0
6.0
6.6
6.1
6.1
6.2
6.2
SFirrnsm .STAGE
5.7
5.8
6.0
5.9
6.3
5.6
5.2
5.8
6.0
6.1
6.0
9.1
6.1
6.2
6.2
6.5
6.1
6.7
6.6
5.7
5.9
-------
TABLE G-3 LIMESTONE TESTS CHEMICAL ANALYSES (WT.%) AND UTILIZATION
1977
Date
5-11
5-11
5-11
5-11
5-12
5-12
5-12
5-12
5-12
5-12
5-12
5-12
5-16
5-16
5-16
5-16
5-17
5-17
5-17
5-17
5-17
5-17
5-17
5-17
5-17
5-19
5-19
Time
1720
1720
1815
1815
1130
1130
1230
1230
1630
1630
1730
1730
1600
1600
1700
1700
1030
1030
1410
1410
1700
1700
1800
1900
1900
1015
1015
Location*
P
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
D
S
S
D
S
D
S
Test
50
50
50
50
51
51
51
51
52
52
52
52
49
49
49
49
54
54
55
55
56
56
56
53
53
38
38
% Acid
Insolubles
3.21
6.94
5.39
6.26
4.86
7.04
3.36
5.37
2.78
6.27
2.60
5.40
2.63
4.63
3.49
4.19
5.47
2.58
1.37
2.67
3.01
2.77
2.06
2.46
3.30
2.14
%CaSO. .
2H20
76.93
74.16
83.53
83.00
85.34
78.89
82.67
84.20
78.60
80.99
70.00
74.30
51.61
74.64
69.29
74.11
72.82
75.36
62.45
63.31
76.45
61.40
79.08
69.00
74.07
93.75
92.03
%CaS03
Q.5H00
J £
2.87
1.43
1.43
1.43
1.72
3.58
2.44
2.58
1.86
3.30
1.72
2.29
1.43
0.72
1.43
1.43
0.00
0.00
0.00
1.43
0.72
0.00
0.00
2.87
0.75
0.00
0.00
%MgCOq
1 * ' J
1.25
0.92
0.77
0.77
0.59
2.39
2.53
0.67
1.15
0.57
0.73
0.69
0.24
0.61
0.90
0.80
0.48
0.77
1.25
1.34
0.79
1.49
2.13
0.67
0.71
0.36
0.36
%CaCO3
16.40
18.76
8.62
8.35
9.92
8. BO
10.87
10.87
16.49
10.50
21.88
19.67
28.43
19.33
22.40
19.90
21.04
19.60
25.24
24.45
18.88
25.31
14.49
22.63
20.17
4.09
4.71
% Limestone
Utilization
72.41
69.01
83.89
84.19
82.75
80.68
78.25
81.37
72.52
81.62
64.88
68.70
52.00
68.67
63.81
67.94
66.21
68.11
57.60
59.28
69.43
56.86
72.98
64.38
67.66
92.34
91.24
* D - Dissolver tank
S - First stage drop collector
-------
TABLE G-3 LIMESTONE TESTS CHEMICAL ANALYSES (WT.%) AND UTILIZATION (CONT.)
1977
Date
5-19
5-19
5-19
5-19
5-19
5-19
5-19
5-19
., 5-19
vo 5-20
ro 5-20
5-23
5-23
5-23
5-23
5-24
5-24
5-24
5-25
5-25
5-25
5-25
5-25
5-25
5-25
5-25
Time
1230
1230
1515
1515
1615
1700
1700
1900
1900
1045
1045
1500
1500
1715
1715
1600
1600
1700
0930
0930
1150
1150
1705
1705
1915
1915
Location*
D
S
D
S
S
D
S
D
S
D
S
D
S
D
S
D
S
S
D
S
D
S
D
S
D
S
Test
47
47
48
48
48
57
57
58
58
41
41
42
42
39
39
40
40
40
45
45
44
44
46
46
43
43
% Acid
Insolubles
2.52
-1.83
3.79
2.77
3.62
4.18
3.70
3.15
3.40
4.89
3.65
4.02
4.16
3.06
3.67
2.94
2.41
6.54
4.88
6.37
3.95
3.29
3.57
5.16
5.24
4.72
%CaSO4 -
2H 0
89.60
91.08
88.54
72.87
89.60
85.53
89.11
84.43
87.49
68.43
65.85
68.09
68.14
67.85
63.03
63.07
56.15
69.29
74.64
75.83
79.18
79.32
72.58
72,63
71.77
73.01
%CaSO3 .
0.51^0
0.57
0.00
0.00
0.00
0.43
0.00
0.00
0.00
0.00
2.29
0.43
0.00
0.00
0.43
0.72
0.00
0.43
0.00
0.43
0.00
0.00
2.15
0.00
0.00
1.43
0.00
%MgCO3
0.44
0.56
0.44
0.31
0.29
0.46
0.42
0.47
0.42
0.65
1.13
0.94
1.03
1.32
2.78
1.44
1.88
0.42
1.21
0.96
0.67
0.73
0.52
0.67
0.75
0.84
%CaCO0
7.12
6.18
7.62
6.07
7.89
8.12
8.55
11.83
9.92
22.70
23.63
22.72
22.63
22.70
23.02
24.45
27.25
22.77
19.33
19.22
17.76
17.51
21.04
20.92
21.06
20.58
% Limestone
Utilization
87.30
88.55
86.34
86.81
86.42
85.16
85.13
79.85
83.00
63.91
60.73
62.41
62.41
62.11
58.56
58.36
52.79
63.39
67.80
68.41
71.27
72.22
66.08
66.04
66.12
66.30
-------
TABLE G-4 SCRUBBER OPERATING DATA LIMESTONE TEST
AVERAGE
SCRUBBER GAS SECOND STAGE PRESSURE DROP (in.HoO)
TEST 1977 FLOW SLURRY PUMP
NO. DATE (acfmaclO3 120°F) RATE (gpm)
38
39
40
41
42
43
44
45
46.
47
48
49
50
51
52
53
54
55
56
57
58
5-19
5-23
5-24
5-20
5-23
5-25
5-25
5-25
5-25
5-19
5-1
5-1
5-1
5-1
5-1
5-17
5-17
5-1
5-1
5-19
5-19
48.8
36.0
42.8
52.2
54.2
39.5
44.9
60.2
64.0
51.2
45.4
54.2
54.2
37.0
43.3
57.5
58.9
32.6
39.3
46.5
45.9
2280
2550
1910
2660
1890
2780
1960
2780
2050
2400
2280
2470
1875
2680
2000
2720
1750
2660
1930
2250
2280
;LURRY
.OLIDS
6.5
13.1
15.7
9.0
12.9
8.1
7.8
8.8
9.2
4.4
5.0
3.6
4.7
2.5
2.8
2.0
1.3
2.9
1.5
7.2
4.6
TOTAL
17.0
16.3
15.2
24.7
15.5
20.0
16.8
22.2
11.2
18.3
17.9
19.8
18.8
24.8
13.3
25.3
16.3
18.4
10.0
18.0
18.0
FIRST
STAGE
7.5
8.8
5.2
12.0
9.5
12.0
7.8
7.2
6.0
8.8
9.0
9.0
6.0
12.0
7.5
15.3
7.5
7.0
5.0
9.0
9.0
SECOND AVERAGE LIQUID PICKUP (gpm) AVERAGE L/G (gal./10Jac
STAGE FIRST STAGE
9.5
7.5
10.0
12.7
6.0
8.0
9.0
15.0
5.2
9.5
8.9
10.8
12.8
12.8
5.8
10.0
8.8
11.4
5.0
9.0
9.0
1080
1050
435
1755
1320
1515
1020
1050
630
1440
1440
1500
585
1560
1050
1710
945
945
495
1380
1395
SECOND STAGE
690
435
750
915
120
540
510
1260
120
615
585
600
750
1065
150
525
240
780
90
525
480
FIRST STAGE
22.2
29.2
10.2
33.6
24.4
38.3
22.7
17.4
9.8
28.1
31.7
27.7
10.8
42.1
24.2
29.7
16.1
29.0
12.6
29.7
30.4
SECOND STAGE
14.2
12.1
17.5
17.5
2.2
13.7
11.4
20.9
1.9
12.0
12.9
11.1
13.8
28.7
3.5
9.1
4.1
24.0
2.3
11.3
10.5
Saturated gas flow inside scrubber =
(Inlet scfm) x [~H- (T inlet-120°F) CP x 580 R
L HI X MWT^ 492°R
Hl=970 Btu/lb water
MWt= 181b/lb mole
CPQ= 7.26 Btu/lb mole- F
-------
TABLE G-5 LIMESTONE TESTS, AVERAGE S02 CONCENTRATION DATA
SO2 Concentration (ppra)
% Removal Efficiency *
VO
Test No.
38
39
40
41
42
43
44
45
46
47
DuPont/
Wet Test
D
W
D
W
D
W
D
W
D
W
D
W
D
W
D
W
D
W
D
W
Inlet Midpoint
438
354 169
425
411 76
350
332 158
290
270
300
233
360
294
410
358 119
320
298 114
295
208 105
390
- 330 120
Outlet
122
81
34
43
38
57
35
20
105
60
26
15
65
64
40
31
80
60
101
42
Corrected Outlet First Stage Second Stage Overall
109
81
30
43
33
57
31
20
92
60
23
15
57
64
35
31
70
60
90
42
73.46
49.28 52.07 75.69
92.46
80.24 43.42 88.82
90.02
49.64 63.92 81.83
88.73
92.20
67.74
72.91
93.22
94,58
85.18
64.58 46.22 80.95
88.43
59.55 72.81 89.00
75.07
46.96 42.86 69.69
75.49
61.37 65.00 86.48
-------
TABLE G-5 (CONT.) LIMESTONE TESTS, AVERAGE SO CONCENTRATION DATA
^ Concentration (ppm)
Test No.
48
49
50
51
52
53
54
55
56
57
58
DuPont/
Wet Test
D
W
D
W
D
W
D
W
D
W
D
W
D
N
D
W
D
W
D
W
D
W
Inlet
410
320
375
330
254
270
483
449
425
395
350
309
375
312
600
563
525
438
375
278
335
258
Midpoint
139
140
166
163
190
166
202
249
112
84
Outlet
80
70
100
48
120
53
112
41
212
129
80
46
147
106
120
80
318
204
78
48
62
46
Corrected Outlet
71
70
90
48
125
53
102
41
194
129
71
46
131
106
107
80
> 284
204
69
48
55
46
% Removal Efficiency *
First Stage Second Stage
53.71
55.49
35.50
61.13
48.94
44.27
61.70
39.32
57.20
65.48
49.64
65.71
68.07
74.85
32.11
36.14
60.40
18.07
57.14
45.24
Overall
81.55
76.69
74.82
84.74
48.38
79.41
77.39
90.22
51.55
65.33
78.74
84.40
63.41
64.42
80.96
84.83
42.26
50.29
80.33
81.66
82.60
81.10
* DuPont SO2 Removal based on corrected outlet value
-------
APPENDIX G-6
Regression Analysis of the Limestone Test Data
Equation (5.1) represents the best two-coefficient model
obtained from the analysis of the lime test data. The choice
of limestone/SO2 stoichiometry and slurry circulation rate for
the regression analysis was based on the results of the statis-
tical screening tests described in Section 5.
A General Electric regression analysis program was used to
obtain the exponents in equation (5.1). A copy of the computer
printout indicating the coefficients, correlations and observed
and predicted values of the natural logs of the removal efficiency
is attached.
The analysis was run on a model in the following form:
In (S02 removal efficiency %) = Zl X In (slurry circulating
rate, gpm) + Z2 X In (limestone/S02 stoichiometry).
196
-------
PTP-PENDSNT
VARIABLE
21
22
REGRESSION
COEFFICIENT
0.554521
0.523039
CHECK _ STANDARD T SIS
NUMBER _ ERROR VALUE LEVEL
-I.03S-16 _0.295509E-02 137.650 100.00?,
-6.36S-19 _3.323090E-01 6.355 100.00*
CODE ?2
ANALYSIS OF VARIA;j£S FOR TXE NO-INTERCSPT MODEL-
SOURCE
REGRESSION
ERROR
TOTAL
DF
2
19
21
SS
390.50
0.20222
390.71
MS-
195.25
0.10643S-OI-
13344.943 = F-RATIO* A 100.00% VALUE.
0.9997 = MULTIPLE CORRELATION COEFFICIENT.
0.9995 =' IM^E" OF DETERMINATION.
0.9994 = ."ADJUSTED" IMDEX OF DETSRMIMATIOM.
CCOMPUTED FJJOM MOMEMTS ABOUT THE OR1SIM3
0.103L7 = 3TAWDARD ERROR OF ESTIMATE
2.295% OF MSAM OF Y3
CODS --
CASE
:-JO.
1
2
3
4
5
6
7
3
9
10
11
12
13
14
15
16
17
13
19
20
OBSERVED
VALUE
4.4856
4. 1496
4.3662
4.2157
4.4356
3.3791
4.3151
4.3134
4.3439
4.4443
4.5330
3.9425
4.3940
4.5000
4.. 5263
3.7433
4.296.7
4.. 3240
4.4012
4.3361
PREDICTSD-
_VALUE
_ 4.5564
4. LOSS
4.4J54
4.2233
_ 4.5347
3-2290
4.».3912
4.2925
4. .2558
21 ~ 4.4140
STANDARD ERROR OF Td:
A..5772
4.J1430
_ 4.5415
4.4417
4.5720
3.9213
4.1703
4.2371
4.2552
4..247S
4.2927
3TIMATE =»
197
RESIDUAL
0.70323E-01
0.41171E-01
-0.11021
-0.80745S-02
0.491 I1E-01
0.49933S-01
0.761022-01
0.25934E-01
0.69433E-02
0.1501 1
0.42235E-01
-0.10047
-0.14752
0.53349E-01
0.45259E-01
-0.17792
0.12594
0.86930E-OI
0.14606
0.13333
0.12123
0. 10317
% DEVIATION
-1 .55
1 .00
-2.46
-0.19
-1 .08
-1 .27
-1.73
0.60
-0.16
3.50
-0.92
-2.49
-3.25
1 .31
-0-99
-4.54
3.02
2.05
3.43
3.26
, 2.33-
-------
APPENDIX HI
PARTICULATE TEST DATA SUMMARY
H--1 Particulate Collection Efficiency and Particulate
Emission Rate Data
H.-2 Mechanical Collector Efficiency Calculations
H-3 Scrubber Inlet Particle Size Distribution
H.-4 Particulate Penetration Model Regression
Analyses
H-5 Fractional Collection Efficiencies
198
-------
TABLE H-ls PARTICULATE COLLECTION EFFICIENCY AND PARTICULATE EMISSION RATE
vo
Test No.
1
2H
4
10
11
12R
15
18R
2
7
12
18
Date
12-16-76
4-5-77
12-15-76
12-18-76
12-18-76
4-7-77
12-13-76
4-6-77
12-17-76
4-12-77
12-19-76
12-20-76
Particulate Wt.
Collected (gins)
Inlet Outlet
0.7024
0.5717
0.9050
0.7563
0.5239
0.2236
0.4742
0.6003
0.3930
0.1678
0.9368
0.4003
0.0894
0.1682
0.1509
0.2996
0.2755
0.0847
0.2845
0.1832
0.2111
0.0423
0.1504
0.1312
Concentration
(Grains/SCF Dry) Collection
Inlet Outlet Efficiency,%
0.2403
0.2736
0.2738
0.2624
0.1700
0.1139
0.1828
0.2463
0.1654
0.0785
0.2543
0.1509
5-11-77* -
5-11-77* -
5-11-77* -
0.0570
0.1169
0.0539
0.1035
0.0054
0.0539
0.1024
0.1139
0.1073
0.0334
0.0439
0.0573
0.0144
0.01RR
0.0211
76.27
57.29
80.30
60.55
49.76
52.68
43.99
53.74
35.16
57.53
R2.73
62.05
-
-
-
Total
Pressure
Drop (in. Ha 0)
16.0
20.5
24.0
17.5
18.0
10.6
19.3
17.5
21.5
21.8
18.0
25.0
19.0
19.0
19.0
Boiler
Firing Rate
MM ntu/hr
155.2
133.1
165.9
153.3
169.5
-
193.9
114.6
-
77.7
-
-
-
-
-
Particillate
Emissions
(lbs/106Btu)
0.171
0.292
0.177
0.314
0.244
0.210
0.219
0.311
-
0.189
-
-
0.146
0.188
0.219
Orsat Analyses
Inlet % Outl
CO; O^ CO,
8.2 11.0 7.4
_
7.1 12.3 7.0
8.6 11.0 8.2
7.8 11.8 7.2
_
9.0 10.6 8.6
_
7.6 12.0 8.8
_
6.2 13.0 6.2
8.8 9.8 8.8
-
- - -
_
Inlet
et % Exces
Oj Air
12.2 106
-
12.4 137
12.0 107
12.2 125
-
10.8 100
-
12.1
-
13.2
10.4
-
-
_ _
*These tests were not part of the fractional efficiency test program. EPA method 5 was used for these tests.
-------
APPENDIX H-2
MECHANICAL COLLECTOR EFFICIENCY CALCULATIONS FOR EACH MECHANICAL COLLECTOR
Ap = 2.0 in. w.C. FLY ASH S.G. = 2.0 gm/cm3
to
o
o
First
Mechanical
Collector
Dp , Microns
1
1-3
3-5
5-7
7-9
9-11
11-19
20 +
Collector
Efficiency
4
10
44
63
83
95
96
98
Wt. of
Particulates
0.5
4.0
5.5
6.0
6.0
5.0
17.0
66.0
100.01
Wt.
Collected
-0-
0.40
2.42
4.08
4.98
4.75
16.32
64.68
86.63#
Wt.
Uncollected
0.50
3.60
3.08
1.92
2.02
0.25
0.68
1.32
13.37#
Cumulative
Wt.% of
Uncollected
Particulates
3.74
30.67
53.71
68.07
83.18
85.05
90.14
. 100.00
Second
Mechanical
Collector
Dp, Microns
1
1-3
3-5
5-7
7-9
9-11
11-19
20+
Collector
Efficiency
4
10
44
68
83
95
96
98
Wt. of
Particulates
0.5
3.6
3.08
1.92
2.02
0.25
0.68
1.32
Wt.
Collected
-0-
0.36
1.36
1.31
1.68
0.24
0.65
1.29
Wt.
Uncollected
0.50
3.24
1.72
0.61
0.34
0.01
0.03
0.03
Cumulative
Wt.% of
Uncollected
Particulates
7.72
57.72
84.27
93.68
98.93
99.08
99.54
100.00
13.37#
6.89#
6.48#
Total Collected - 93.52# for 93.52% Removal Efficiency
-------
TABLE H-3
SCRUBBER INLET PARTICLE SIZE DISTRIBUTION
Start S.G. Temp. Dia.,u Cum. Wt.%
1335 hrs. 1.3 220°F 10.0 99.99
6.3 99.12
4.2 96.12
2.9 88.66
1.8 ' 77.06
0.94 65.60
0.53 56.07
0.38 47.51
<0.38 39.50
5/24/76 1400 hrs. 1.3 232°F 15.6 99.99
9.66 99.62
6.44 97.75
4.48 95.13
2.82 90.64
1.45 81.65
0.90 66.67
0.61 49.44
<0.61 44.57
5/19/76 1400 hrs. 1.3 222°F 11.49
7.09
4.80 99.98
3.35 98.70
2.10 95.63
1.08 87.17
0.65 75.39
0.44 62.56
<0.44 43.20
5/19/76 1645 hrs. 1.3 222°F 11.49 100.00
7.09 99.53
4.80 96.73
3.35 92.28
2-10 88.19
1.08 81.29
0.65 72.51
0.44 59.30
<0.44 45.96
201
-------
APPENDIX
This section of Appendix H: contains the results: of a series
of regression analyses which. were performed to determine the
coefficients in the one and two stage participate penetration models,
As stated in Section S, a
Qn ^^ tmpfatim
selected. Coefficients for both, models for particles in the
following size ranges were developed:
0.3 to 0.5 micron, 0.5 to 1.0 micron,
1.0 to 2.0 microns and 2.0 to 5.0 microns .
The following diameters were used to represent an average
particle in each range:
0.4 micron, 0.7 micron, 1.4 microns and 3.2 microns.
Input data for these regression analyses included liquid pickup
and flue gas temperature data collected during the particulate
tests and the fractional particulate collection efficiency data
selected from the results of the Andersen Impactor tests run for
this part of the program.
A total of eight regression analyses were performed. They
are presented in the following order.
Particle Size No, of Data File
Micron Stages _
0.3-0.5 2 Fijal 1A
1 Final 1A
o 2 Final 2A
0 I o 1 Frnal 2A
10-20 2 Final 3A
10-2-0 1 Final 3A
J. \J £* W __ ^ 14
9 n-q n 2 Final 4
2:0-5:0
The data files listed above are presented first. Each line in
202
-------
the data file Hats the, observed collection efficiency as. a
percentage, first stage, liquid pickup in Gi?M> second stage, liquid
pickup in GPM and the temperature of gas- entering the first stage
in °R. The notation used in the regression analyses is as follows.
Y2 is the natural log of the penetration
Zl is the first stage venturi correlation coefficient
(as in the body of the report)
Z2 is the second stage venturi correlation coefficient
(as in the body of the report).
Final 1A
48.23
66.66
15.24
21.01
33.67
29.90
Final 2 A
72.85
86.68
40.83
26.74
22.05
70.06
19.60
53.90
85.40
54.40
Final 3A
95.89-
97.56
93.13
71.24
86.06
86.60
94.40
98.40
85.30
1200.
1400
1050
1100
780
400
1200
400
1050
1100
78'0
220
1350
400
1050
1050
12QQ
1400
1050
1100
78 Q
220
1350
1050
1050
Data Files
600 .
840
540
600
450
480
600
840
540
600
450
810
810
480
600
1140
600
84 Q
540
600
450
810
810
600
1140
900
940
933
9.38.'
952
945
900
940
933
238.
952
895
938
945
928
908
9-33
938
89-5
938
928
908
203
-------
Final 4
9-9.9-9 1200 6"QO, 9J50.
98.78 1400 840 940
95.04 1050 540 933
98.28 1100 600 938
97.66 780 450 952
99.99 1350 810 938
95.70 1050 1140 908
204
-------
REGRESSION ANALYSIS FOR 0.3 td 0.5 microns Particles
TWO STAGE MODEL DATA FILE: FINAL 1A
INDEPENDENT
VARIABLE
21
22
REGRESSION
COEFFICIENT
0.37534SE-06
-0.268697E-02
CHECK _ STANDARD T SIS
NUMBER _ ERROR VALUE LEVEL
2.04E-13 _0.206379E-05 0.132 13.57S
1.30S-15 _0-264399E-02 -1.016 63.30?
CODE ?i2
ANALYSIS OF VARIANCE FOR THE NO-INTERCEPT MODEL-
SOURCE
REGRESSION
ERROR
TOTAL
OF
2
4
6
SS
1.6337
0.32933
2.0131-
MS
0.34434
O.S2346E-01-
10.254 =_ F-RATIO* A 97.342 VALUE.
0.9148 a MULTIPLE CORRELATION COEFFICIENT.
0.8363 = INDEX OF DETERMINATION.
0.7552 » 'ADJUSTED" INDEX OF DETERMINATION.
CCOMPUTED_FROM MOMENTS ABOUT THE ORIGIN!
0.23696 =_ STANDARD ERROR OF ESTIMATE
_-53.333S OF MEAN OF Y2
CODE
CASE
IJO.
1
2
3
4
5
OBSERVED
VALUE
-0.65843
-1.0935
-0.16536
-0.23587
-0.41057
PREDICTED
VAL'JE_
-0.51397
-0.7^674
-0.^6376
-0.52237
-0.32614
-0.35523
-0^47119
STANDARD ERROR OF THE .ESTIMATE =
RESIDUAL
-0.14446
-0.35173
0.29340
0.28650
-0.14423E-01
0.I 1590
0.28696
DEVIATION
23.11-
47. It
-64.34
-54.35
3.64
-24.60
205
-------
r ANALYSIS' FOR 0.5 TO 1.0 MICHON PARTICLES-
TITO STAGE MODEL
INDEPENDENT
VARIABLE
21
22
REGRESSION
COEFFICIENT
0.518732E-06
-0.229168E-G2
DATA FILE: FINAL .1A
CHECK_ STANDARD T
NUMBER _ ERROR VALUE
-1.08E-12 _0.978248S-06 0.530
-1.19E-1S JJ.941933E-03 -2.433
SIG
LEVEL
3S.97S
95-90?
CODE ?2
ANALYSIS OF VARIANCE FOR THE MO-INTERCEPT MODEL-
SOURCE
REGRESSION
ERROR
TOTAL
DF
2
8
10
ss
8.7278
3.8934
12.621-
MS
4.3639-
0.48667
8.967 =_ F-RATIO., A 99.09% VALUE.
0.3316 <= MULTIPLE CORRELATION COEFFICIENT.
0.6915 = . INDEX OF DETERMINATION.
0.6144 = "ADJUSTED" INDEX OF DETERMINATI ON.-
CCOMPUTED EROM MOMENTS ABOUT THE ORIGIN]
0.69762 = .STANDARD ERROR OF ESTIMATE
-74.903% OF MEAN OF Y2
CODE ?1
CASE
NO.
1
2
3
4
5
6
7
8
9
10
OBSERVED
VALUE
-1.3039
-2.0161
0.52481
0.31119
0.24913
-1.2061
0.21813
.0.77444
-1.9243
-0.73534
PREDICTED
VALUE.
-0.6JS634
-1.2398
-0.5S.45S
-0.6^663
-0.5J.052
-1.^416
-0.93.626
-0.6
-------
REGRESSI.OH ANALYSIS FOR 1.0. TO 2.0 MICRON PARTICLES.
INDEPENDENT
VARIABLE
21
22
REGRESSION
COEFFICIENT
-0.2104SOE-05
-0.477507E-03
CHECK _- STANDARD T SIS
NUMBER _ ERROR VALUE LEVEL
3.24E-13 _0.395516S-06 -2.350 94.393
2.98E-15 JJ.962452E-03 -0.496 36.50*
CODE --?2
ANALYSIS OF VARIANCE FOR THE MO-INTERCEPT MODEL
SOURCE
REGRESSION
ERROR
TOTAL
DF
2
7
9-
SS
70.314
9.4732
79.792-
MS-
35.157
1.3540
25.965 = F-RATIO* A 99.94% VALUE.
0.9387 a MULTIPLE CORRELATION COEFFICI£?JT.'
0.8812 2. IMDEX OF DETERMINATION.
0.8473 = "ADJUSIED" IMDEX OF DETSRMINATION.-
CCOMPUTED JROM MOMENTS ABOUT THE ORIGIN]
1.1636 =_STANDARD ERROR OF ESTIMATE
-Jll.797% OF MEAN OF Y2
CODE --?l
CASE
NO.
I
2
3
4
5
6
7
3
9
.OBSERVED
VALUE
-4.5013
-3.7136
-2.6733
-1.2463
-1.9706
-2.
-2,
-4.
-1 ,
0101
3827
1356
9175
PREDICTED-
VALUE,
RESIDUAL
-2.2569
-3.62J33
-2.6JI63
-2.30.22
-2.01.61
-1.Q.I06
-3.4^33
-2.fi.30S
-3 . 0 12 1
-1.5344
-0.92S04E-0!
-0.31444E-01
1.5559
0.45476E-01
-0.99955
0.60560
-1.4548
1 .0996
X DEVIATION
51.72
2.56
1.19
-55.52
-2.26
98.91
-17.36
54.27
-36.45
STANDARD ERROR OF THE_ESTIMATS =
1.1636
207
-------
REGRESSION ANALYSIS FOP. 2.0 TO 5.0 MICRON PARTICLES
TWO STAGE MODEL DATA FILE: FINAL 4
INDEPENDENT
VARIABLE
REGRESSION
COEFFICIENT
-0.232419E-05
0.709920E-03
CHECK_ STANDARD T 513
NUMBER_ ERROR VALUE LEVEL
2.13E-11 _0.720005E-06 -3.228 97.67*
1.98E-I4 J3.31996SE-03 0.366 57.33*
CODE ?2
ANALYSIS OF VARIANCE FOR THE MO-INTERCEPT MODEL-
SOURC" DF _ S<5 MS
REGRESSION
ERROR
TOTAL
2
5
_ 7
154.52
9.3926
154.41-
77.261
1.9735
39.050 j* F-RATIOj A 99.91* VALUE.
0.9694 = MULTIPLE CORRELATION COEFFICIENT.-
0.9398 = INDEX OF DETERMINATION.
0.9153 = -ADJUSTED' INDEX OF DETERMINATION.
CCOMPUTED_FROM MOMENTS ABOUT THE ORIGINS
1.4066 = .STANDARD ERROR OF ESTIMATE
_-30.585% OF MEAN OF Y2
CODE ?1
CASE
NO.
1
2
3
4
5
6
7
OBSERVED
VALUE
-6.9035
-4.4063
-3.0041
-4.0633
-3.7554
-6.9035
-3.1469
PREDICTED-
VALUE_
-5.1129
-5.2.1 30
-4.5.426
-4.6.972
-3.^070
-5.5975
-3.UD13
STANDARD ERROR OF T2E ESTIMATE
RESIDUAL
-1.7956
1.4063
1 .5336
0.63397
-0.44845
-1.31 10
-0.45113E-01
1.4066-
DEVIATION
35.12
-24.19
-33.37
-13.50
13.56
23.42
1.45
208
-------
REGRESSION ANALYSIS FOR 0... 3 TO 0.5 MICRON PARTICLES
SINGLE- STAGE MODEL DATA PILE: FINAL 1A.
INDEPENDEN1
.VARIABLE
Z2
REGRESSION
COEFFICIENT
-0.221377E-02
CHECK_ STANDARD T SI3
NUMBER_ ERROR VALUE LEVEL
-4.51E-17 JJ.439410E-03 -5.033 99.60*
CODE ?2
ANALYSIS OF VARIANCE' FOR THE MO-INTERCEPT MODEL
SOURCE
REGRESSION
ERROR
TOTAL
DF
1
5
6
55
1.6859
0.33211
2.0131-
MS
1.6859-
0.66423E-01
25.382 * F-RATlOj A 99.60S VALUE.
0.9140 = MULTIgL'E CORRELATION COEFFICIENT.-
0.3354 = INDEX OF DETERMINATION.
0.3025 = "ADJUSTED- IMDEX OF DETERMIMATION.
[COMPUTED J^ROM MOMENTS ABOUT THE ORIGIM3
0.25773 =_ STANDARD ERROR OF ESTIMATE
_-52.384% OF MEAM OF Y2
CODE ?1
CASE
NO.
1
2
3
4
5
OBSERVED
VALUE
0.65843
-1.0985
0.16536
0.23587
0.41057
PREDICTED
VALUE _
-0.53131
-0.74J333
-0.41317
-0.5,3131
-0^39843
6 -0.35523 -Q.42L504
STANDARD ERROR OF THE ESTIMATE =
RESIDUAL
-0.
-0.
0.
0,
12712
35470
31281
29543
-0.12092E-01
0.69760E-01
0.25773-
DEVIATION
23.93
47.69
-65.42-
-55.60
3.03
-16.41
209
-------
REGRESSION ANALYSIS FOR 0.5 TO 1. a MICRON, PARTICLES
SINGLE STAGE MODEL DATA FILE: PI-7AL ?A
INDEPENDENT REGRESSION
VARIABLE COEFFICIENT
22 -0.1S5015E-02
CHECX_ STANDARD T
NUMBER_ ERROR VALUE
"2.95E-17 _D«422405E-03 -4.330
SIC3
LEVEL
99.32S
CODE ?2
ANALYSIS OF VARIANCE FOR THE NO-INTERCEPT MODEL
SOURCE DF
REGRESSION 1
ERROR 9
TOTAL 1Q
SS
S.5910
4.0302
12.621-
MS
8.3910
0.44730
19.135 = F-RATIO* A 99.32% VALUE.
0.3250 = MULTIPLE CORRELATION COEFFICIENT.-
0.6807 _= INDEX OF DETERMINATION.
0.6452 = "ADJUSTED- INDEX OF DETERMINATION.
CCOMPUTED .FROM MOMENTS ABOUT THE ORIGIN}
0.66913 = STANDARD ERROR OF ESTIMATE-
-71 .350Z OF MEAN OF Y2
CODE ?1
CAS;
NO.
1
2
3
4
5
6
7
3
9
10
OBSERVED
VALUE
-1.3039
-2.0161
0.52481
0.31 I 19
0.24913
-1.2061
0.21318
0.77444
-1.9243
0.73534
PREDICTSD-
VALUEL
-0.71706
-1- . OSJ 9
-0.69933
-0.77.706
-0.53530
-1 ..Q.490
-1.0J190
-0.62J65
-0.77106
-1 .47.64
RES IDUAL
-0.52636
-0.92322
0.17455
0.46583
0.33367
-0.15706
0.33036
-0.15279
-1.1473
0.69108
DEVIATION
67.30
85.32
-24.96
-59.95
-57.25
14.97
-79.20-
24.53
147.64
-46.81
STANDARD ERROR OF THE ESTIMATE
0.66913-
210
-------
REGRESSION ANALYSIS FOR 1.0 TO 2.0 MICRON. PARTICLES
SINGLE STAGE MODEL DATA FILE: FINAL 3A
INDEPENDENT REGRESSION
VARIABLE COEFFICIENT
21 -0.251383E-05
CHECK_ STANDARD T SIG
NUMBER _ ERROR VALUE LEVEL
I.26S-12 _ 0.3327S3E-06 -7.554 99.99%
CODE ~?2
ANALYSIS OF VARIANCE FOR THE NO-INTERCEPT MODEL
SOURCE
REGRESSION
ERROR
TOTAL
DF
I
3
9
SS
69.981
9.31 15
79.792-
MS
69.981
1 .2264
57.061 »_ F-RATIQ* A 99.99% VALUE.
0.9365 = MULTI£LE CORRELATION COEFFICIENT.
0.877TD. = INDEX OF DETERMINATION.
0.8617 = "ADJUSTED" INDEX OF DETERMINATION.-
CCOMPUTED £ROM MOMENTS ABOUT THE ORIGIN}
1.1074 =_STANDARD ERROR OF ESTIMATE
i39.779% OF MEAN OF Y2
CODE --?!
CASE
NO.
1
2
3
4
5
6
7
8
9
.OBSERVED
VALUE
-4.5013
-3.7136
-2.6783
-1.2463
-1.9706
-2.0101
-2.3827
-4.1356
-1.9175
PREDICTED-
VALUE
-3.0644
-3.6^37
-2.7101
-2.8£77
-2. Q486
-O.S6&24
-2.2.227
-2.6^32
STANDARD ERROR OF THS_ESTIMATE
RESIDUAL
-1.4369
0.59853E-01
0.51772E-01
1.6214
0.77975S-01
-1.4499
0.63676
-1.4129
0.77571
1.1074
* DEVIATION
46.39
1.64
-1.90-
-56.54
-3.31
258.80-
-13.09
51.39-
-28.30
211
-------
REGRESSION ANALYSIS FOR 2.0 TO 5.0 MICRON PARTICLES
SINGLE STAGE MODEL DATA FILE: FINAL 4
INDEPENDENT REGRESSION
VARIABLE COEFFICIENT
21 -0.172437E-05
CHECK_ STANDARD T S1G
NUM3ER_ ERROR VALUE LEVEL
-2.95E-12 JD.191930E-06 -3.934 99.9915
CODE --?2
ANALYSIS OF VARIANCE FOR THE MO-INTERCEPT MODEL
SOURCE DF _ SS MS-
REGRESSION I _ 153.04 153.04
ERROR 6 _ 11.376 1.3959-
TOTAL 7 164.41-
80.719 = F-RATIO, A 99.99t VALUE.-
0.9648 = MULTIPLE CORRELATION COEFFICIENT.-
0.9308. = INDE:"? OF DETERMINATION.
0.9193 = -ADJUSTED" INDEX OF DETERMINATION.
CCOMPUTED F30M MOMENTS ABOUT THE ORIGIN]
1.3769 =_ STANDARD ERROR OF ESTIMATE-
-29.940* OF MEAN OF Y2
CODE ?1
CASE
NO.
1
2
3
4
5
A
Vrf
7
.OBSERVED
VALUE
-6.9085
-4.4063
-3.0041
-4.0633
-3.7554
-6.9035
-3.1469
PREDICTED-
VALUS_
-4.J046
-5.7.236
-4.2.204
-4. j.'">63
-3.2L23
-5.5J32
-4.2227
RESIDUAL
-2.1033
1.3219
1.2764
0.4.1301
-0.54344
-1.3903
1.0753
% DEVIATION
43.79
-23.07
-29.32
-9.63
16.92
25.20
-25.43-
STANDARD ERROR OF THS' ESTIMATE
1.3769-
212
-------
TABLE IT-5. FRACTIONAL COLLECTION EFFICIENCIES
Particle Size Range, Microns
Test No.
1
2R
4 .
10
11
12R
15
18R
2
7
12
13
0.3-0.5
48.23
25.66
66.66
15.24
21.01
21.93
33.67
29.90
0.5-1.0
72.85
27.08
36.68
40.83
26.74
59.46
22.05
70.06
19.60
53.90
85.40
54.40
1.0-2.0
98.39
91.83
97.56
93.13
71.74
86.06
86.60
94.40
98.40
88.30
2.0-5.0
99.
98.
95.
98.
97.
99.
95.
9
78
04
28
66
90
70
Note: This data was used to develop particulate performance
models
213
-------
Date
4/14/76
5/3/76
5/5/76
5/7/76
6/4/76
6/7/76
6/8/76
6/9/76
6/10/76
7/7/76
7/13/76
7/14/76
7/16/76
7/15/76
7/19/76
7/20/76
7/22/76
7/26/76
7/26/76
7/28/76
APPENDIX J
R-C/BAHCO SCRUBBER OPERATING LOG
STARTUP MARCH 11, 1976 11:00 p.m.
Duration
(of outage, hrs*)
15
252
57
16
226
1
2
16
5
11
1
6'48
10
1
3
1
3
1
16
4
41
759
Comments
Inspection of scrubber
Second stage slurry
pump liner collapsed
Modify scrubber spray
manifolds
Unknown
Repair booster fan
bearing
Low water level in
dissolver tank
Unknown
Repack 2nd stage slurry
pump
Replace belt on 2nd
stage slurry pump
Loss of fan bearing
cooling water
Loss of plant air
Power outage
Heat plant shutdown, in-
stallation of thickener
mechanism and resealing
wood thickener tank
Control panel mainten-
ance
Unknown
Fan vibration
Loss of fan bearing
cooling water
Lime feeder not oper-
ating
Power outage
Training of operators
Training of operators
Replace sludge pump
diaphragm
Training
Replace torque limiter
on lime, slaker
Repair and modification
of fan support
-Outages which dcurred in one month- and carried over to the
next month are underlined.
214
-------
APPENDIX J (.Contl
Date
8/30/76
8/31/76
9/1/76
9/2/76
9/3/76
9/7/76
9/24/76
10/5/76
10/8/76
10/12/76
10/13/76
10/14/76
10/14/76
10/16/76
10/19/76
10/28/76
10/28/76
10/28/76
10/28/76
12/6/76
12/10/76
12/11/76
Duration
(of outage> hrs*)
19
16
16
16
1
Comments
220
0.5
0.5
21
24
1
18
64
1
6
0.5
0.5
380
1
3
23
Replace lime, slakes
conveyor motor
Lime dissolver level
controls malfunction ,
system shutdown over-
night
See 8/31/76
See 8/31/76
Clean blockage in
mechanical collector
hopper
Repack 2nd stage slurry
pump and clean block-
age in mechanical
collector hopper
Heat plant outage +
realign and balance
booster fan
Inadvertent
Control panel main-
tenance
Loss of fan bearing
cooling water
Loss of fan bearing
cooling water
Loss of fan bearing
cooling water
Remove and clean
plugged sludge line to
the pond
Repair hole in slurry
line
Loss of power to con-
trol panel
Control panel mainten-
ance
Unknown
Overload fan motor
Repair of the fan thrust
bearing
Power outage
Maintenance of 2nd
stage slurry pump
packing
Loss of water in heat
plant
215
-------
APPENDIX J CCont.)
Date
12/14/76
12/21/76
12/22/76
12/24/76
12/30/76
1/4/77
1/5/77
1/20/77
1/21/77
1/25/77
1/26/77
1/28/77
2/1/77
2/2/77
2/4/77
2/6/77
2/9/77
2/10/77
2/11/77
2/14/77
2/15/77
2/17/77
2/24/77
2/27/77
3/1/77
3/10/77
3/12/77
Duration
(of outage, hrs*)
2
3
4
1
2
2
2
3
3
3
11
4
7
55
1
1
1
1
1
2
3
5
206
0.5
39
Comments
Changed oil in fan
bearing
Frozen air lines to
blowdowns
Frozen air lines to
blowdowns
Overloaded fan motor
Frozen air lines to
falowdowns
Frozen air lines to
blowdowns
Repack 2nd stage
slurry pump
Repack 2nd stage slurry
pump
Frozen air line to by-
pass damper
Frozen air line to
blowdowns
Remove accumulated
grit from 2nd stage
Frozen air line to by-
pass damper
Replace blowdown valves
Replace blowdown valves
Unknown
Frozen air line to by-
pass damper
Remove grit accumula-
tion from slurry line
Repair frozen valve
Maintenance on booster
pump controls
Clear 1st stage blow-
down valve
Repair frozen valve
Repair frozen valve
Repack 2nd stage slurry
pump
Overloaded fan motor
Repair bearings in water
booster pump
Changed oil in fan
bearing
Modify control panel
wiring
216
-------
APPENDIX J (Cent.)
Date
3/14/77
3/14/77
3/16/77
3/18/77
3/24/77
4/1/77
4/4/77
4/5/77
4/8/77
4/12/77
4/13/77
5/9/77
5/10/77
5/11/77
5/12/77
5/12/77
5/13/77
5/16/77
5/19/77
5/25/77
5/27/77
Duration
(of outage'> hrs.*)
0.5
0.5
2
0.5
29
1
2
5
0.5
0.5
430
19
10
1
0.5
1
71
0.5
0.5
2
Comments
Inadvertent
Changed oil in fan
bearing
Preparation for EPA tests
Inspection of scrubber
interior
Removal of grit accumula-
tion from scrubber
Check low water level
shutdown
Replace belt on 2nd stage
slurry pump
Overloaded fan motor
Control panel maintenance
Install EPA test equip-
ment
Repair and modification
of fan
Fan bearing resistance
temp, detector (RTD) in-
operative
RTD inoperative
Repair RTD
Control panel mainten-
ance
Unknown
Replace lime slaker motor
Inadvertent
Change oil in fan bearing
Repack 1st stage slurry
pump
Inadvertent
217
-------
APPENDIX K
SULFUR DIOXIDE PERFORMANCE TEST RESULTS
K-l S02 Performance Test Results
K-2 Gas Flow and Coal Firing Rate Calculations
K-3 SO-7 Emission Rate Calculations
218
-------
APPENDIX K-l
SO2 PERFORMANCE TEST RESULTS
On. September 12, 1976, Research.-Cottrell personnel performed
a series, of three C3) tests to determine the rate of sulfur dioxide
emissions from a Bahco flue gas desulfurizat-ion facility at
Rickenbacker Air Force Base, Ohio. These tests, conducted according
to Method 6 stipulated by the U.S. Environmental Protection Agency,
showed the Bahco facility to be operating in compliance with, the
guarantee stipulated by Research-Cottrell, Inc. to the U.S. Air
Force with regard to S02 emissions in the stack gases.
The results of these tests are presented in Table 'J-l. Note
that the average S02 emission rate obtained for the three tests was
0.8347 Ibs. per million Btu, 16% below the maximum acceptable
rate of 1.0 Ib. per million Btu.
At the time that these, tests were performed, the R-C/Bahco
scrubbing system was being operated by automatic controls which.
provide for, among other things, automatic lime, feed control.
This lime feed control is designed to regulate the lime additions
to the. system to meet emission requirements and at the. same tine
to minimize lime consumption.
219
-------
TABLE K-l S02 PERFORMANCE RESULTS
Gas Volume Collected, ft3 15.3363 15.4750 15.5230
SO4 = Level, gm/1 0.2027 0.1572 0.1293
S02 Emission Rate, Ibs/hr 52.35 42.34 34.72
SO2 Emission Rate, Lbs/million
Btu 1.0243 0.8130 0.6667
Average S02 Emission Rate = 0.8347 Ib/million Btu
220
-------
TABLE K-2
GAS FLOW AND COAL FIRING RATE CALCULATIONS
Annubar reading = 42,000 scfm
Temperature at annubar = 278 F
Pressure at annubar = 25" H-0 -?
Corrected gas flow rate = 42,000 scfm x 37° + 46° R
278 + 460°F
14.7 + 25 (14.7/407)psia = 49
14.8 psia
(The Annubar was supplied and calibrated for a gas temperature
of 370°F and 14.8 psia)
Between 10:45 AM and 5:55 PM, total coal burned = 32,736 Ibs.
= 4568 Ibs/hr.
Sulfur content of coal = 2.62%, heating value = 11,400 Btu/lb
as received. (These figures are based on coal^analysis)
Assume 93% of sulfur in the coal is burned to S02 and emitted
from boiler. (This assumption is based on overall heat plant
efficiency data collected in Oct. 1974)
S0? rate = (4568 Ibs/hr)(0.0262 Ib.S/lb-coal) (0.93) (64#SO-/
32#S) = 222.61 Ibs/hr. ^
Heat production = (11,400 Btu/lb) (4568 Ib/hr) =
52.08 x 106 Btu/hr
Moisture in gas = 5.0%
Corrected (dry) gas flow = (49,798) (0.95) = 47,308 scfm
221
-------
TABLE E-3 SO-? EMISSION RATE CALCULATIONS
7est
Volume collected = 3.0769 + 7.7594 = 15.3363 ft3
3arium chloride titration: 0. 2027gmS04=/l
3n, T-ato _ 47,308scfm x 0.7027cm/l 64cm SO?/om mole
^ ~ ~ 15.3363scf 96 ~gm S04=s/g^i mole *
60 min/hr. ^ 53.35£/hr.
454gm/lb.
302 rats per million Btu = " 1.0243*/MM3tu
Test 2
Voluae collected = 7.7531 + 7.7213 = 15.4750 ft3
Barium chloride titration: 0.1572 gm SO4=/1
3C, r-,^e _ 47,308 x 0.1572 x 64_ 60_ = 42.34|/hr. = 0 . 3130 ^
a<-2 r^-e -- 15.4750 96 454
'Jest :
Volume.- collected = 7.7612 + 7.7613 = 15.5230 ft3
Barium chloride titration: 0.1293 gin SO4~/1
30, ,ate = 47.308 X 0.1293 x 64 x 60_ = 0. 66S7*/MM3tu
302 -^^e 15.475b 96 454
222
-------
TECHNICAL REPORT DATA
fPlease read Inurucrions on the reverse before completing}
I. REPORT NC. 2.
EPA-600/7-78-115
-. TITLE AMD SUBTITLE
EPA Evaluation of Bahco Industrial Boiler Scrubber
System at Rickenbacker AFB
7. AUTHOR(S)
E.L.Biedell, R.J.Ferb, G.W.Malamud, C.D.Ruff,
and N.J.Stevens
9. PERFORMING ORGANIZATION NAME AND ADDHESS
Research-Cottrell, Inc.
P.O. Box 750
Bound Brook, New Jersey 08805
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
5. REPORT DATE
June 1978
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
IAG-D5-0718 *
13. TYPE OF REPORT AND PERIOD COVERED
Final; 3/76-6/77
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES IERL-RTP project officer is John E. Williams , Mail Drop 61,
919/541-2483. (*) IAG with U.S. Air Force: USAF contract F33617-75-90100 with
Research-Cottrell.
16. ABSTRACT i
The report gives results of an 18-month evaluation of the R-C/Bahco com-
bined flue gas desulfurization and particulate removal system on a stoker-fired
industrial boiler at Rickenbacker AFB, Ohio. Particulate emissions were reduced
to as low as 0.15 Ib/million Bfcu. SO2 emissions were reduced to as low as 0.1 lb/
million Btu with lime and 0.2 Ib/million Btu with limestone while burning 2-4%
sulfur midwestern coal at firing rates from 20 to 200 million Btu/hr. Operating
costs, including maintenance, were #5.28 per ton of coal burned for lime and $4.27
per ton for limestone, exclusive of capital costs. There were no problems with
scale formation or plugging with either reagent. The system met all emission and
operating cost guarantees.
17.
a.
Air Pollution
Flue Gases
Desulfurization
Sulfur Oxides
Dust Control
Boilers
Unlimited
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Industrial Engineering
Stokers
Coal
Scrubbers
Limestone
Calcium Oxides
VtENT
b.lOENTlFIERS/OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Bahco Process
Particulate
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COSATI Field/Group
13 B 13H
2 IB
07A,07D 21D
07B
08G
13A
21. NO. OF PAGES
22-3
22. PRICE
223
------- |