cvEPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-90-016
August 1990
Air
Small Industrial-
Commercial-
Institutional
Steam Generating
Units --
Background
Information for
Promulgated Standards
-------
EPA-450/3-90-016
Small Industrial-Commercial-lnstitutional
Steam Generating Units --
Background Information for
Promulgated Standards
Emission Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
August 1990
-------
This report has been reviewed by the Emission Standards Division of the Office
of Air Quality Planning and Standards, EPA, and approved for publication.
Mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use. Copies of this report are available
through the Library Services Office (MD-35), U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina 27711; or, for a fee, from the
National Technical Information Services, 5285 Port Royal Road, Springfield,
Virginia 22161.
11
-------
ENVIRONMENTAL PROTECTION AGENCY
Background Information Document
for New Small^Jtndustrial-Commercial-Institutional Steam Generating Units
Prepared by:
/
Jack R. Farmer (Date)
Director, Emission Standards Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
1. The standards of performance limit emissions of sulfur dioxide from new
coal- and oil-fired industrial-commercial-institutional steam generating
units of 29 MW (100 million Btu/hour) or less, but greater than or equal
to 2.9 MW (10 million Btu/hour). Particulate matter emissions from new
wood-, coal-, and oil-fired industrial-commercial-institutional steam
generating units in this same size category are also limited. Section
111 of the Clean Air Act (42 U.S.C. 7411), as amended, directs the
Administrator to establish standards of performance for any category of
new stationary source of air pollution that "...causes or contributes
significantly to air pollution which may reasonably be anticipated to
endanger public health or welfare."
2. Copies of this document have been sent to the following Federal
Departments: Office of Management and Budget, Commerce, Interior, and
Energy; the National Science Foundation; and the Council on
Environmental Quality. Copies have also been sent to members of the
State and Territorial Air Pollution Program Administrators; the
Association of Local Air Pollution Control Officials; EPA Regional
Administrators; and other interested parties.
3. For additional information contact:
Mr. Rick Copland
Standards Development Branch (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919) 541-5265
4. Copies of this document may be obtained from:
U.S. EPA Library (MD-35)
Research Triangle Park, NC 27711
Telephone: (919) 541-2777
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: (703) 487-4600
-------
List of Abbreviated Terms
$/Mg
$/ton
ABMA
ASTM
BACT
BDT
Btu
Btu/hr
Btu/lb
CAA
OEMS
CIBO
C02
CTG
DMC
EGF
EPA
ESP
FBC
FR
ID
kJ/kg
kPa
Ib/million Btu
L/G ratio
Mg
MW
NAAQS
ng/J
NOX
NRDC
NSPS
PM
PM10
POM
PSD
RCRA
S02
THROX
WEPCO
dollars per megagram
dollars per ton
American Boiler Manufacturers Association
American Society for Testing and Materials
best available control technology
best demonstrated technology
British thermal units
Btu per hour
Btu per pound
Clean Air Act
continuous emission monitoring system
Council of Industrial Boiler Owners
carbon dioxide
Control Technique Guidelines
double mechanical collector
electrostatic gravel bed filter
Environmental Protection Agency
electrostatic precipitator
fluidized bed combustion
Federal Register
induced draft
kilojoules per kilogram
kilopascals
pounds per million Btu
1iquid to gas ratio
megagram
megawatt
National Ambient Air Quality Standards
nanograms per Joule
nitrogen oxides
Natural Resources Defense Council
new source performance standards
particulate matter
PM with mean diameter less than 10 microns
polycyclic organic matter
prevention of significant deterioration
Resource Conservation and Recovery Act
sulfur dioxide
Thermal Heat Recovery Oxidation
Wisconsin Electric Power Company
-------
TABLE OF CONTENTS
Section Page
1.0 Summary 1-1
1.1 Summary of Changes Since Proposal of the Standard 1-1
1.2 Summary of Impacts of Promulgated Action 1-2
1.2.1 Alternatives to Promulgated Action 1-2
1.2.2 Environmental Impacts of Promulgated Action .... 1-2
1.2.3 Energy and Economic Impacts of Promulgated Action . . 1-3
1.2.4 Other Considerations 1-4
2.0 Summary of Public Comments 2-1
2.1 Selection of Pollutants 2-11
2.2 Selection of Affected Facility 2-12
2.3 Standard for Sulfur Dioxide 2-17
2.3.1 Emission Limit 2-17
2.3.2 Emission Credits 2-20
2.3.3 Percent Reduction Standard 2-22
2.3.4 Emerging Technology Standard 2-32
2.4 Standard for Particulate Matter 2-36
2.4.1 Emission Limit - Coal 2-36
2.4.2 Emission Limit - Wood 2-45
2.5 Standard for Nitrogen Oxides 2-53
2.6 Performance and Reliability of Demonstrated Control
Technology 2-57
2.6.1 Sulfur Dioxide 2-57
2.6.2 Particulate Matter 2-64
2.6.3 Nitrogen Oxides 2-71
-------
TABLE OF CONTENTS (CONTINUED)
Section Page
2.7 Cost and Economic Impacts 2-72
2.7.1 Auxiliary Steam Generating Units 2-72
2.7.2 Fuel Cost 2-75
2.7.3 Low Capacity Units 2-76
2.7.4 Alaskan Coal 2-80
2.7.5 Lignite Industry 2-82
2.7.6 Anthracite Coal Industry 2-83
2.8 Non-Cost National Impacts 2-90
2.9 Environmental Impacts 2-97
2.9.1 Sulfur Dioxide 2-97
2.9.2 Particulate Matter 2-100
2.9.3 Other 2-103
2.10 Selection of Format of Standard 2-105
2.11 Test Methods and Monitoring 2-106
2.11.1 Sulfur Dioxide 2-106
2.11.2 Particulate Matter 2-119
2.11.3 Nitrogen Oxides 2-124
2.12 Reporting and Recordkeeping 2-124
2.12.1 Sulfur Dioxide 2-124
2.12.2 Particulate Matter 2-126
2.13 Wording of Regulation 2-126
2.14 Miscellaneous 2-131
VI
-------
1.0 SUMMARY
On June 9, 1989 the Environmental Protection Agency (EPA) proposed
standards of performance limiting emissions of sulfur dioxide (S02),
particulate matter (PM) and nitrogen oxides (NOX) from small industrial-
commercial -institutional steam generating units with a maximum design capacity
of 29 megawatts (MW) [100 million Btu per hour (Btu/hr)] or less, but greater
than or equal to 2.9 MW (10 million Btu/hr) heat input (54 FR 24792;
Subpart DC) under authority of Section 111 of the Clean Air Act. Public
comments were requested on the proposal in the Federal Register. There were
48 commenters, composed mainly of industries, trade associations, and State
regulatory agencies. Also commenting were a U.S. Government agency and
several nonaffiliated commenters. The comments that were submitted (see
Docket A-86-02), along with responses to these comments, are summarized in
this document.
1.1 SUMMARY OF CHANGES SINCE PROPOSAL OF THE STANDARD
There are five changes to the originally proposed standards. First, .
the proposed NOX limit of 430 nanograms of pollutant per joule (ng/J)
[1.0 pound of pollutant per million Btu (1.0 Ib/million Btu)] has been
eliminated; no NOX standards are being promulgated. Secondly, the PM standard
has been changed from 43 ng/J (0.10 Ib/million Btu) to 130 ng/J
(0.30 Ib/million Btu) for wood-fired units with heat input capacities greater
than 8.7 MW (30 million Btu/hr) and with annual capacity factors of less than
30 percent. Third, distillate oil-fired units have been exempted from the
requirement for continuous opacity monitoring. Fourth, oil-fired units
operating at capacity factors of less than 10 percent are no longer required
to perform an initial 24-hour performance test. The last change is to allow
the use of fuel supplier certification to verify the use of low sulfur coal
and very low sulfur residual oil in units with heat input capacities of 2.9 to
8.7 MW (10 to 30 million Btu/hr). In the proposed regulation, only distillate
oil-fired units were allowed to use supplier certification in lieu of fuel
sampling and analysis. Owners and operators using supplier certification in
1-1
-------
lieu of sampling and analysis must maintain records of the appropriate
certification.
1.2 SUMMARY OF IMPACTS OF PROMULGATED ACTION
1.2.1 Alternatives to Promulgated Action
The regulatory alternatives for these new source performance
standards (NSPS) are discussed in the following background documents for the
proposed standards: "Overview of the Regulatory Baseline, Technical Basis,
and Alternative Control Levels for Particulate Matter (PM) Emission Standards
for Small Steam Generating Units" (EPA-450/3-89-11), "Overview of the
Regulatory Baseline, Technical Basis, and Alternative Control Levels for
Sulfur Dioxide (S02) Emission Standards for Small Steam Generating Units"
(EPA-450/3-89-12), and "Overview of the Regulatory Baseline, Technical Basis,
and Alternative Control levels for Nitrogen Oxides (NOX) Emission Standards
for Small Steam Generating Units" (EPA-450/3-89-13). The regulatory
alternatives discussed in these background documents reflect the different
levels of emission control from which the promulgated standards have been
selected. These alternatives remain the same as those discussed in the
preamble to the proposed standards.
1.2.2 Environmental Impacts of Promulgated Action
A discussion of the environmental impacts of the proposed
standards is found in the following background documents: "Projected Impacts
of Alternative New Source Performance Standards for Small Industrial-
Commercial -Institutional Fossil Fuel-Fired Boilers" (EPA-450/3-89-17), and
"Projected Impacts of Alternative Particulate Matter New Source Performance
Standards for Industrial-Commercial-Institutional Nonfossil Fuel-Fired Steam
Generating Units" (EPA-450/3-89-18).
Since the changes to the proposed standards are not expected to
significantly impact air emissions, these analyses of environmental impacts of
the proposed standards now become the final Environmental Impact Statement for
the promulgated standards.
1-2
-------
The changes to the proposed standards are not expected to significantly
impact air emissions. The proposed standards for NOX had no environmental
benefit, therefore the decision to delete the NOX standards from the final
regulation will result in no environmental impact. The PM emissions projected
for small wood-fired steam-generating units in the proposal notice were based
on units operating at a 55 percent annual capacity factor. Since the less
stringent emission limit of 130 ng/J (0.30 Ib/million Btu) will only apply to
wood-fired units operating at annual capacity factors of 30 percent or less,
PM emissions are not expected to exceed the level projected in the proposal
notice. The elimination of the opacity monitoring requirement for distillate
oil-fired units is not expected to result in increased PM emissions because
distillate oil is a cleaner fuel than residual oil, where the opacity
monitoring requirement is being retained.
1.2.3 Energy and Economic Impacts of Promulgated Action
The energy and economic impacts associated with this NSPS are
discussed in the background documents "Projected Impacts of Alternative New
Source Performance Standards for Small Industrial-Commercial-Institutional
Fossil Fuel-Fired Boilers" (EPA-450/3-89-17) and "Projected Impacts of
Alternative Particulate Matter New Source Performance Standards for
Industrial-Commercial-Institutional Nonfossil Fuel-Fired Steam Generating
Units" (EPA-450/3-89-18). For the most part, these impacts remain unchanged.
However, the changes to the proposed regulation that could affect the energy
and economic impacts include the increase in the PM emission limit for units
operating below 30 percent annual capacity factor, the exemption from opacity
monitoring for oil-fired units, and the provisions for supplier certification
in lieu of fuel sampling and analysis for residual oil- and coal-fired units
with heat input capacities between 2.9 and 8.7 MW (10 and 30 million Btu/hr).
These changes are projected to apply to only a very small number of steam
generating units, so the energy and economic impacts anticipated from the
final standards would be only slightly less than those at proposal.
1-3
-------
1.2.4 Other Considerations
1-2.4.1 Irreversible and Irretrievable Commitment of Resources.
Other than the fuels required for power generation and the materials required
for the construction of the control systems, there is no apparent irreversible
or irretrievable commitment of resources associated with this regulation.
1-2.4.2 Environmental and Energy Impacts of Delayed Standards. The
results of delay in the standards are that new small steam generating units
would be built that may not meet the emission limits established by these
standards. This would delay improvement of ambient air quality and other
environmental benefits associated with this NSPS.
1-2.4.3 Urban and Community Impacts. Neither plant closures nor
significant adverse impacts on small businesses are forecast. No significant
adverse impacts on urban areas or local communities are anticipated as the
result of the promulgation of these standards.
1-4
-------
2.0 SUMMARY OF PUBLIC COMMENTS
A total of 48 letters commenting on the proposed standards were
received. Comments were provided by industry representatives and governmental
entities. These comments have been recorded and placed in the docket for this
rulemaking (Docket A-86-02, Category IV). Table 2-1 presents a listing of all
persons submitting written comments, their affiliation and address, and the
recorded Docket Item Number assigned to each comment letter.
In addition, 5 industry representatives presented oral comments on the
proposed standards at a public hearing held on August 8, 1989. A verbatim
transcript of the comments at the public hearing has been prepared and placed
in Docket A-86-02, Category IV. Table 2-2 presents a listing of all persons
presenting comments at the public hearing, their affiliation and address, and
the recorded Docket Item Number assigned to the public hearing transcript.
The comments summarized in this chapter have been organized into the
following categories:
2.1 Selection of Pollutants
2.2 Selection of Affected Facility
2.3 Standard for Sulfur Dioxide
2.4 Standard for Particulate Matter
2.5 Standard for Nitrogen Oxides
2.6 Performance and Reliability of Demonstrated Control Technology
2.7 Cost and Economic Impacts
2.8 Non-Cost National Impacts
2.9 Environmental Impacts
2.10 Selection of Format of Standard
2.11 Test Methods and Monitoring
2.12 Reporting and Recordkeeping
2.13 Wording of Regulation
2.14 Miscellaneous
2-1
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. Bryce E. Harthoorn IV-D-01
Supervisor, Environmental Services
John Deere Waterloo Works
Post Office Box 270
Waterloo, Iowa 50704-0270
Mr. Harvel M. Rogers Jr., P.E. IV-D-02
Air Pollution Control Officer
Physical and Environmental Services
914 East Broadway
Jefferson Country Air Pollution
Control District
Louisville, Kentucky 40204
Mr. Robert H. Collom, Jr. IV-D-03
Branch Chief, Air Protection Branch
Georgia Department of Natural Resources
205 Butler Street S.E., Floyd Towers East
Atlanta, Georgia 30334
Mr. Richard E. Grusnick IV-D-04
Chief, Air Division
Alabama Department of Environmental
Management
1751 Congressman W. L. Dickinson Drive
Montgomery, Alabama 36130
Mr. J. Leonard Ledbetter IV-D-05
Commissioner
Georgia Department of Natural Resources
.205 Butler Street, S.E., Suite 1252
Atlanta, Georgia 30334
Mr. Ben A. Brodovicz, Chief IV-D-06
Commonwealth of Pennsylvania
Department of Environmental Resources
Bureau of Air Quality Control
Post Office Box 2063
Harrisburg, Pennsylvania 17120
Mr. David Ellsworth IV-D-07
Project Manager - Ladysmith Plant
NORENCO Corporation
45 South Seventh Street
Plaza VII - Suite 3140
Minneapolis, Minnesota 55402-1621
2-2
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. Leonard D. Verrelli IV-D-08
Chief, Air Quality Management Programs
State of Alaska
Department of Environmental Conservation
Division of Environmental Quality
Post Office Box 0
Juneau, Alaska 99811-1800
Mr. R. Dean Cooper IV-D-09
Legal Department
Dow Chemical U.S.A.
Willard H. Dow Center
2030 Building
Midland, Michigan 48674
Mr. David J. Smukowski IV-D-10
Manager, Environmental Controls
Boeing Support Services
Post Office Box 3707
Seattle, Washington 98124-2207
Mr. Robert E. Reilly IV-D-11
President
Enertrac Corporation
182 Turnpike Road, Suite 102
Westborough, Massachusetts 01581
Mr. Tracey S. Narel IV-D-12
Senior Policy Analyst
The Commonwealth of Massachusetts
Executive Office of Energy Resources
100 Cambridge Street, Room 1500
Boston, Massachusetts 02202
Mr. Jeffrey C. Smith IV-D-13
Executive Director
Industrial Gas Cleaning Institute, Inc.
Suite 570, 1707 L Street, N.W.
Washington, D. C. 20036
2-3
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. Brett W. Thacher IV-D-14
EFB Incorporated
Participate Control Systems
16 Upton Drive
Wilmington, Massachusetts 01887
Mr. Charles D. Bennett IV-D-15
Environmental Coordinator
Ashland Petroleum Company
Post Office Box 391
Ashland, Kentucky 41114
Mr. William H. Prokop IV-D-16
Consulting Director
Engineering Services
National Renderers Association, Incorporated
O'Hare Lake Office Plaza
2250 East Devon Avenue
Des Plaines, Illinois 60018
Mr. David C. Rinebolt IV-D-17
Director of Research
National Wood Energy Association
Suite 610
1730 North Lynn Street
Arlington, Virginia 22209-2009
Mr. David C. Branand IV-D-18
Counsel and Director
Environmental Affairs
National Coal Association
1130 Seventeenth Street, N.W.
Washington, D.C. 20036-4677
Mr. David F. Hobson IV-D-19
Executive Director
International District Heating and
Cooling Association
1101 Connecticut Avenue
Suite 700
Washington, D.C. 20036
2-4
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. F. William Brownell IV-D-20
Counsel for the Utility
Air Regulatory Group
Hunton & Williams
2000 Pennsylvania Avenue, N.W.
Post Office Box 19230
Washington, D.C. 20036
Mr. Richard L. White IV-D-21
Director of Environmental Services
Texas Utility Services
2001 Bryan Tower
Dallas, Texas 75201
Mr. F. P. Partee IV-D-22
Principal Staff Engineer
Ford Motor Company
Suite 608
15201 Century Drive
Dearborn, Michigan 48120
Mr. Kyd D. Brenner IV-D-23
Director of Public Affairs
Corn Refiners Association, Inc.
1100 Connecticut Avenue, N.W.
Washington, D.C. 20036
Mr. Russell Mosher IV-D-24
Executive Director
American Boiler Manufacturers
Association
Suite 160
950 North Glebe Road
Arlington, Virginia 22203
Mr. D. M. Anderson IV-D-25
General Manager
Environmental Affairs
Bethlehem Steel Corporation
Bethlehem, Pennsylvania 18016
Mr. William B. Marx IV-D-26
Council of Industrial Boiler Owners
6035 Burke Centre Parkway
Suite 360
Burke, Virginia 22015
2-5
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. Robert E. Pocock IV-D-27
President
Alternate Inputs, Inc.
Post Office Box 43345
Cleveland, Ohio 44143
Ms. Geraldine V. Cox, Ph.D. IV-D-28
Vice President-Technical Director
Chemical Manufacturers Association
2501 M Street, N.W.
Washington, D.C. 20037
Mr. Walter R. Quanstrom IV-D-29
Vice President
Amoco Corporation
200 East Randolph Drive
Chicago, Illinois 60601
Mr. U. V. Henderson, Jr. IV-D-30
Director, Environmental Affairs
Research and Environmental
Affairs Department
Texaco, Inc.
Post Office Box 509
Beacon, New York 12508
Mr. Wallace N. Davis IV-D-31
Executive Director
Commonwealth of Virginia
Department of Air Pollution Control
Room 801, Ninth Street Office Building
Post Office Box 10089
Richmond, Virginia 23240
Mr. M. E. Miller, Jr., P.E. IV-D-32
Manager, Environmental Management
R. J. Reynolds Tobacco Company
Winston-Salem, North Carolina 27102
Dr. John E. Pinkerton IV-D-33
Program Director
National Council of the Paper Industry
for Air and Stream Improvement, Incorporated
260 Madison Ave.
New York, New York 10016
2-6
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. Joel D. Patterson IV-D-34
Manager, Environmental Services
Entergy Services Incorporated
Post Office Box 61000
New Orleans, Louisiana 70161
Mr. Richard P. Lewis, P.E. IV-D-35
Principal Plant Engineer
University of Minnesota
Physical Plant Operations
200 Shops Building
319 15th Avenue S.E.
Minneapolis, Minnesota 55455
Mr. J. R. Smith IV-D-36
Manager, Air Resources Division
Environmental Department
Houston Lighting and Power
Post Office Box 1700
Houston, Texas 77251
Mr. John W. Drake IV-D-37
Chief, Air Quality Service
Oklahoma State
Department of Health
Post Office Box 53551
1000 N.E. Tenth Street
Oklahoma City, Oklahoma 73152
Mr. Michael E. McKay, P.E. IV-D-38
Princeton University
Department of Engineering
The MacMillan Building
Post Office Box 2158
Princeton, New Jersey 08543-2158
Mr. Uttam Trivedi IV-D-39
Manager, Environmental Programs
Environmental and Energy Affairs
Kimberly-Clark Corporation
1400 Hoi comb Bridge Road
Roswell, Georgia 30076
2-7
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS OF PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. J. C. Edwards IV-D-40
Manager, Environmental Affairs
Eastman Kodak Company
Eastman Chemicals Division
Post Office Box 511
Kingsport, Tennessee 37662
Mr. John W. Dwyer IV-D-41
President
The North Dakota Lignite Council
Post Office Box 2277
1016 East Owens Avenue, Suite 200
Association Office Center
Bismarck, North Dakota 58502
Mr. Robert D. Bessette IV-D-42
President
IES
Post Office Box 21887
Lexington, Kentucky 40522
Mr. Peter L. Rozelle IV-D-43
405 Academic Activities Building
Pennsylvania State University
University Park, Pennsylvania 16802
Mr. Arthur A. Davis IV-D-44
Secretary
Commonwealth of Pennsylvania
Department of Environmental Resources
Post Office Box 2063
Harrisburg, Pennsylvania 17120
Mr. Peter N. Brush IV-D-45
Acting Assistant Secretary
Environment, Safety and Health
Department of Energy
Washington, D.C. 20585
Mr. William B. Marx IV-D-46
President
Council of Industrial Boiler Owners
6035 Burke Centre Parkway,
Suite 360
Burke, Virginia 22015
2-8
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS Of PERFORMANCE
FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Commenter and Affiliation Docket Item Number
Mr. E. Allen Womack, Jr. IV-D-47
Vice President, R & D Division
Babcock & Wilcox
1562 Beeson Street
Alliance, Ohio 44601
Mr. M. K. Hopkins IV-D-48
Ford Motor Company
Environmental Control Engineer
15201 Century Drive
Suite 608
Dearborn, Michigan 48120
2-9
-------
TABLE 2-2. LIST OF PUBLIC HEARING SPEAKERS ON PROPOSED STANDARDS
OF PERFORMANCE FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Speakers and Affiliation Docket Item Number
Mr. William Marx F-l.l
Council of Industrial Boiler Owners
6035 Burke Centre Parkway
Suite 360
Burke, Virginia 22015
Mr. Robert Pocock F-1.2
President
Alternate Inputs, Incorporated
Post Office Box 43345
Cleveland, Ohio 44143
Mr. Peter L. Rozelle F-1.3
405 Academic Activities Building
Pennsylvania State University
University Park, Pennsylvania
Mr. Dan Blaschak F-1.4
Blaschak Coal Company
Saint Nicholas
Mahanoy City, Pennsylvania 17948
Mr. Robert Bessette F-1.5
President
IES
Post Office Box 21887
Lexington, Kentucky 40522
2-10
-------
2.1 SELECTION OF POLLUTANTS
1. Comment: One commenter (IV-D-04) stated that emissions of trace
quantities of regulated pollutants such as arsenic, benzene, mercury, and
radionuclides emitted during the combustion of oil or coal should be addressed
in the proposed regulation. He pointed out that the EPA's Prevention of
Significant Deterioration (PSD) regulations require that quantities, effects,
and control options for these pollutants must be evaluated during individual
permitting actions at the State and local level. These pollutants should not
be ignored in the NSPS. The commenter added that the EPA has a better ability
to acquire and analyze information on trace toxic air contaminants than State
or local programs. Further, these data could be used in developing more
meaningful EPA standards, which would lead to more consistent best available
control technology (BACT) determination at the State and local level.
Response: The PSD regulations apply to any pollutant regulated under
the Clean Air Act (CAA). The proposed standards are being developed under
Section 111 of the CAA, which applies to criteria pollutants. The control of
hazardous air pollutants, such as those listed above, is handled under
Section 112 of the CAA. When evaluating standards under Section 111,
hazardous air pollutants are considered in the evaluation of other
environmental impacts of the standards. A control technology resulting in
lower hazardous emissions would be selected over a control that would increase
emissions of hazardous air pollutants. Emissions of a number of trace toxic
pollutants would be controlled by the technologies used to comply with the
small steam generating unit NSPS for PM. Fabric filters, for example, are
demonstrated to be very efficient in controlling emissions of trace metals and
other trace constituents including arsenic, beryllium, cadmium, copper,
chromium, manganese, nickel, and radionuclides.
2-11
-------
2.2 SELECTION OF AFFECTED FACILITY
1. Comment: Four commenters (IV-D-24, IV-D-26, IV-D-28, IV-D-40) stated
that the applicability of the proposed standards in Section 60.40c(a) should
be changed to exclude steam generating units in the 2.9 to 8.7 MW (10 to
30 million Btu/hr) range. Five commenters (IV-D-24, IV-D-26, IV-D-28,
IV-D-40, IV-D-45) stated that these small units are usually monitored by
operators only periodically, unlike larger units, which usually are monitored
constantly by full-time, highly skilled operators. The commenters stated that
EPA has acknowledged that the testing and monitoring requirements of the
standards would be very burdensome for steam generating units with heat input
capacities of less than 2.9 MW (10 million Btu/hr), and suggested that EPA
exempt units in the 2.9 to 8.7 MW (10 to 30 million Btu/hr) range for the same
reasons. One commenter (IV-D-26) stated that the smaller units are usually
designed differently from larger units. This commenter pointed out that many
of these smaller units are operated by automated control systems and,
therefore, are left unattended. This commenter stated that the EPA
acknowledged in the preamble to the proposed regulation that commercial -
institutional units with heat input capacities less than 2.9 MW (10 million
Btu/hr) are often left unattended. The commenter maintained that unattended
operation is not unique to the commercial-institutional units; it also occurs
in small industrial units.
Three commenters (IV-D-24, IV-D-26, IV-D-40) further argued that the
emissions impacts of exempting units in the 2.9 to 8.7 MW (10 to 30 million
Btu/hr) range would be insignificant because the number of new coal-fired and
residual oil-fired units of this size is relatively insignificant. According
to American Boiler Manufacturers Association (ABMA) sales data estimates cited
by one commenter (IV-D-26), there will be 20 new coal-fired units in this size
range over the next 5 years. Furthermore, two commenters (IV-D-24, IV-D-26)
stated that natural gas and distillate oil are the predominant fuel types for
new units smaller than 8.7 MW (30 million Btu/hr). The commenter (IV-D-26)
concluded by stating that even if emission limits can be met by 2.9 to 8.7 MW
(10 to 30 million Btu/hr) facilities, the sampling, analysis, monitoring, and
2-12
-------
recordkeeping requirements will impose unreasonable burdens for these
facilities.
Response; As discussed in the preamble to the proposed regulation, all
of the above-mentioned factors were examined in selecting the 2.9 MW
(10 million Btu/hr) heat input size cutoff for affected facilities in this
source category. The reporting and recordkeeping burden, as well as the
differences in unit end use application (i.e., whether commercial-
institutional or industrial), unit type and design, and unit fuel-use patterns
were considered.
The only standards that apply to small steam generating units in the 2.9
to 8.7 MW (10 to 30 million Btu/hr) size range are the SOg standards, which
can be achieved by firing low sulfur fuels. The absence of full-time, skilled
operators was a factor in the decision not to require compliance with PM
standards in the 2.9 to 8.7 MW (10 to 30 million Btu/hr) range. Thus, no
emission control equipment is required to meet the standards, and no employees
are needed to operate and maintain control equipment. As a result, the
absence of full-time operators at these units would not present a problem in
complying with the standard.
As discussed in the preamble to the proposed regulation, it was found
that units less than 2.9 MW (10 million Btu/hr) are different from units
2.9 MW (10 million Btu/hr) or greater and thus warrant separate consideration.
Units below 2.9 MW (10 million Btu/hr) are used almost exclusively at smaller
commercial-institutional facilities, such as elementary and secondary schools,
shopping centers, and office buildings. Units above 2.9 MW (10 million
Btu/hr) tend to be used at industrial- and larger commercial-institutional
facilities, such as major hospitals, large colleges and universities, and
large commercial laundries. In addition, firetube and cast-iron designs
predominate in units below 2.9 MW (10 million Btu/hr), while watertube and
firetube designs are found in units above 2.9 MW (10 million Btu/hr). These
different unit designs have different emission characteristics.
In addition to differences in end use and unit type, the 2.9 MW
(10 million Btu/hr) heat input size is also a reasonable divider with respect
to fuel-use patterns. Natural gas and distillate oil tend to be the
2-13
-------
predominant fuels for units below 2.9 MW (10 million Btu/hr) heat input
capacity. Higher polluting fuels, such as residual oil and coal, are used in
units above 2.9 MW (10 million Btu/hr) heat input capacity. Because of the
different applications, types of units, and fuels fired, the potential
emission reductions from a typical unit below 2.9 MW (10 million Btu/hr) are
much less than those from a typical unit above this size. As a class, the
steam generating units less than 2.9 MW (10 million Btu/hr) represent a
relatively large percent of the total number of units but only a small percent
of the total S02 emissions for the entire source category. Since the emission
reduction potential is small, the reporting and recordkeeping burden that the
standards would impose on units in this size range is not justified.
Based on this information, it was concluded at proposal that standards
would be reasonable for units in the 2.9 to 8.7 MW (10 to 30 million Btu/hr)
size range, but unreasonable for units below 2.9 MW (10 million Btu/hr) heat
input capacity. No new information was presented during the public comment
period to support a change to the level of the proposed cutoff. Therefore, a
size cutoff of 2.9 MW (10 million Btu/hr) heat input capacity has been
retained for the final standards.
Although the cutoff of 2.9 MW (10 million Btu/hr) was determined to be
reasonable for the definition of affected facilities (i.e., units below this
level are not subject to any provisions of the standards), the lower size
cutoff was considered separately during the development of standards for each
pollutant. As stated above, since the S02 standards are based on the use of
low sulfur fuels, the absence of full-time operators is not a problem in the
2.9 to 8.7 MW (10 to 30 million Btu/hr) heat input range. Further, a full-
time operator is not needed to conduct fuel sampling and analysis because the
proposed regulation has been amended to allow coal and residual oil-fired
units with heat input capacities of 2.9 to 8.7 MW (10 to 30 million Btu/hr) to
use supplier certification of fuel sulfur content in lieu of fuel sampling and
analysis. Since costs and other impacts are reasonable, the cutoff for the
S02 standards remains at 2.9 MW (10 million Btu/hr) heat input capacity. When
evaluating the PM standards, the lack of full-time, skilled operators was a
factor in the decision to raise the lower size cutoff for PM to 8.7 MW
(30 million Btu/hr).
2-14
-------
2. Comment: One commenter (IV-D-01) stated that the proposed regulation
should address such nontraditional fuels as densified refuse-derived fuel,
which is not regulated now but may be used, either alone or in combination
with other fuels, as an energy source in the future.
Response; Densified refuse-derived fuel is not addressed in the NSPS
for small steam generating units because it is proposed for regulation under
the NSPS for municipal waste combustors (54 FR 52251, December 20, 1989).
However, if densified refuse-derived fuel is combusted in a steam generating
unit which also combusts coal, oil, or wood, then the steam generating unit is
subject to the S02 and/or PM standards under the small steam generating units
NSPS when coal, oil and/or wood is combusted.
3. Comment: Three commenters (IV-D-05, IV-D-15, IV-D-29) questioned
whether specific units would be covered by the proposed standards. One
commenter (IV-D-29) stated that the current definition of process heater
overlooks a-whole class of process heaters, specifically, fired preheaters and
reboilers used in the distillation process. The current definition of process
heaters includes devices "primarily used to heat a material to initiate or
promote a chemical reaction. . . ." However, distillation is technically not
a chemical reaction. He recommended the following definition:
"Process heater" means a device that is primarily used to
heat a material to promote separation, like distillation,
or to initiate or promote a chemical reaction in which
the material participates as a reactant or catalyst.
One commenter (IV-D-05) suggested a clarification of the definition of
"steam generating units". "Steam generating unit" is currently defined in the
proposed regulations as "a device which combusts any fuel to produce steam or
to heat water or any other heat transfer medium." According to the commenter,
the phrase "any other heat transfer medium" could also include air used in
wood and mineral process dryers, which do not qualify for the process heater
exception.
2-15
-------
One commenter (IV-D-15) stated that the definition of steam generating
unit as currently stated in the proposed regulation includes process
circulating hot oil heaters. The commenter pointed out that these types of
heaters were not mentioned in the background documents and, therefore, should
not be covered in the proposed NSPS. The commenter recommended that the
definition of steam generating unit be modified to reflect this exemption.
Response; The PM and S02 emission characteristics of circulating hot
oil heaters, preheaters, process heaters used to heat a material to promote
separation, like distillation, and process dryers using air as the heat
transfer medium are no different from the PM and S02 emission characteristics
from any other steam generating unit burning the same fuels under the same
conditions. Similarly, the emission control techniques, the performance of
these techniques in reducing emissions, and the costs of these control
techniques are essentially the same. For these reasons, the analysis of steam
generating units presented in the preamble is representative of the energy,
environmental, and economic impacts for hot oil heaters, preheaters, process
heaters used to heat a material to promote separation, like distillation, and
process dryers using air as the heat transfer medium. Therefore, these units
are considered small steam generating units for the purpose of this regulation
and are subject to the final standards.
4. Comment: One commenter (IV-D-28) stated that some of the terms and
definitions used in the proposed standards are vague and need to be clarified
in order to define the affected facility more clearly. He offered the
following language to clarify those terms of concern:
"Steam generating unit" means a device that combusts any
fuel or by-product/waste to produce steam or to heat water
or any other heat transfer medium. This term includes any
steam generating unit that combusts fuel and is part of a
cogeneration system or a combined cycle system. The term
"steam generating unit" does not include process heaters
as defined in this subpart.
2-16
-------
"By-product/waste" means any liquid or gaseous substance
produced at chemical manufacturing plants or petroleum
refineries (except natural gas, distillate oil, or residual oil)
combusted in a steam generating unit for heat
recovery or for disposal.
"Chemical manufacturing plants" means industrial plants
that are classified by the Department of Commerce under
Standard Industrial Classification Code 28.
Response: Inclusion of "by-product/waste" in the definition of a steam
generating unit is relevant only when special provisions are adopted for units
firing such fuel. However, since there are no special provisions in the rule
for units combusting "by-product/waste," the term itself need not be included
nor defined in the final standards. Nor is a definition needed for "chemical
manufacturing plants," a term used in the "by-product/waste" definition.
2.3 STANDARD FOR SULFUR DIOXIDE
2.3.1 Emission Limit
1. Comment: Three commenters (IV-D-22, IV-D-24, IV-D-26) recommended that
the S02 emission limit be increased from 520 to 690 ng/J (1.2 to
1.6 Ib/million Btu) heat input based on the availability of low sulfur coal.
One of the commenters (IV-D-22) stated that a 520 ng/J (1.2 Ib/million Btu)
limit would require burning coal in the 0.5 to 0.75 percent sulfur range,
which is usually a higher priced metallurgic coal (0.6 percent sulfur content)
with limited availability rather than a steam coal (0.9 percent sulfur
content). At present, the price is $28 per ton ($28/ton) for steam coal and
$33 to $35/ton for metallurgic coal; however, as demand increases the prices
will rise and usage levels will become an issue. According to the commenter,
users with small demands would not be able to compete with major utilities
requiring high tonnages of coal. The commenter concluded that coal in the
range of up to 1 percent sulfur would be able to meet a 690 ng/J
(1.6 Ib/million Btu) S02 emission limit and would be more readily available in
a less competitive market.
2-17
-------
Another commenter (IV-D-26) pointed out that significant costs would be
incurred to transport coal with 0.75 percent or less sulfur content, costs
that would be avoided if burning the more widely available 1 percent sulfur
coal were allowed. The commenter also stated that a 690 ng/J (1.6 Ib/million
Btu) emission limit would provide fuel flexibility in allowing units to
combust a higher sulfur coal and use less costly S02 control systems to
achieve 40 to 50 percent reduction. This commenter pointed out that the use
of high sulfur coal would encourage the development of new cost-effective S02
control technologies and would benefit the local communities that supply this
coal. The commenter concluded that fuel flexibility is important to the small
fuel user facing fuel supply and price uncertainties.
Another commenter (IV-D-24) stated that the 520 ng/J (1.2 Ib/million
Btu) S02 emission limit was unrealistic. Because only an annual average of
10 units below 29 MW (100 million Btu/hr) heat input capacity fire coal, the
commenter recommended that a fuel sulfur limit of 1 percent, equal to 690 ng/J
(1.6 Ib/million Btu), would be more realistic.
Response: As discussed in the preamble to the proposed standards, the
costs and availability of coals capable of meeting a 520 ng/J (1.2 Ib/million
Btu) S02 emission limit were examined. Coal capable of achieving a 520 ng/J
(1.2 Ib/million Btu) limit is widely available at a reasonable cost. This
emission limit was chosen to represent the best demonstrated technology for
small coal-fired steam generating units. The selection of the 520 ng/J
(1.2 Ib/million Btu) emission limit was not based on the use of metallurgical
grade coal. It was based on the use of steam coal. The NSPS, therefore,
should have little impact on either the supply or cost of metallurgical coal.
Low sulfur steam coal has become increasingly available throughout the
United States. This is largely due to the demand for low sulfur coal
generated by the 1971 NSPS for utility steam generating units and the 1987
NSPS for large industrial units with an annual capacity of less than
30 percent. Both of these NSPS established an emission limit of 520 ng/J
(1.2 Ib/million Btu) for coal-fired units. In response to this demand, coal
suppliers have increased the availability of this coal across the
United States. As a result, 520 ng/J (1.2 Ib/million Btu) coal has become
2-18
-------
more widely available to all coal users, including small steam generating unit
owners and operators.
A new analysis of regulatory alternatives, which was prepared to examine
the potential impact of a 690 ng/J (1.6 Ib/million Btu) standard, shows that
the incremental cost effectiveness of a 693 ng/J (1.6 To/million Btu) standard
over the 1,560 ng/J (3.6 To/million Btu) baseline is $830 per megagram
($830/Mg) ($750/ton) of S02 removed. The incremental cost-effectiveness value
of a 520 ng/J (1.2 To/million Btu) standard over a 690 ng/J (1.6 Ib/million
Btu) standard is $750/Mg ($680/ton). The coal costs included in this analysis
included the cost of transportation. Each of these increases is considered to
be reasonable. Therefore, no changes have been made to the S02 standard.
2. Comment: One commenter (IV-D-42) stated that it is not accurate to use
national average numbers from State implementation plans (SIP's) to represent
the diverse fuel user/equipment population of small steam generating units.
The commenter recommended that a statistical sampling of small steam
generating unit users would more accurately represent this population than
available SIP data, which may not be representative of the population as a
whole.
The commenter stated that a baseline emission level of 1,550 ng/J
(3.6 Ib/million Btu), which equates to about 2.15 percent sulfur, is too high.
The commenter estimated that the average coal used by small steam generating
unit owners or operators probably has an emission level of 1,130 ng/J
(2.6 Ib/million Btu), which equals a 1.5 percent sulfur content. The
commenter maintained that if this lower, more accurate, baseline were used in
establishing the proposed regulation, then the cost/benefit level for a
520 ng/J (1.2 Ib/million Btu) emission level over a 690 ng/J (1.6 Ib/million
Btu) emission level would be about $3,300/Mg ($3,000/ton) nationwide. In
light of this information, the commenter suggested that a more appropriate
emission level would be 690 ng/J (1.6 ID/million Btu), equaling a 1 percent
sulfur coal.
2-19
-------
Response: This 1,550 ng/J (3.6 Ib/million Btu) baseline, which was used
in the "model boiler" cost analysis, was chosen to represent a typical SIP
emission limit. As discussed in the preamble to the proposed regulation, SIP
emission limits range from 1,400 to 1,510 ng/J (3.3 to 3.5 To/million Btu).
The 1,550 ng/J (3.6 la/million Btu) level was selected as baseline because it
is close to these limits and available economic models contain substantial
information about the price and supply of coals at this sulfur range. Other
baselines could have been selected. However, if a more stringent baseline
were used, the annualized costs for S02 control at the baseline would also
increase to account for the additional costs associated with more stringent
emission limitations. Therefore, no direct calculation of new cost
effectiveness values could be made until the costs at the new baseline were
evaluated.
An additional cost analysis has been performed using a new baseline
emission level of 1,130 ng/J (2.6 Ib/million Btu). The incremental cost
effectiveness of a 690 ng/J (1.6 To/million Btu) standard over a 1,130 ng/J
(2.6 Ib/million Btu) standard is $l,200/Mg ($l,080/ton) of S02 removed. This
is considered reasonable. The incremental cost effectiveness of a 520 ng/J
(1.2 Ib/million Btu) standard over a 690 ng/J (1.6 Ib/million Btu) standard is
$750/Mg ($680/ton). This is also considered reasonable and, therefore, the
standard remains at 520 ng/J (1.2 Ib/million Btu).
2.3.2 Emission Credits
1. Comment; Two commenters (IV-D-27, IV-D-36) expressed concern about
mixed fuel firing. One commenter (IV-D-27) noted that EPA does not allow
emission credits for dual-firing natural gas in utility steam generating
units. However, if natural gas were burned in industrial steam generating
units along with coal and credit were given for burning natural gas in the
total count of Btu's going into the unit, then the micronized coal process
could be used to meet the 90 percent reduction standard. The commenter
recommended that natural gas be counted toward the total Btu's fired.
2-20
-------
Another commenter (IV-D-36) stated that EPA has not addressed the option
of dual-firing natural gas and residual oil as a means of achieving compliance
for a 30-day S02 averaging period. This commenter stated that since it
maintains air quality and meets the intent of both the proposed regulation and
the proposed S02 standard, this option should be recognized as valid by EPA.
Response; The merits of emission credits were thoroughly examined and
rejected during development of the NSPS for large industrial-commercial-
institutional steam generating units (Subpart Db). The rationale for not
providing emission credits for large units likewise applies to small
industrial-commercial-institutional units. Emission credits effectively
negate any environmental benefits, in terms of reduced S02 emissions,
associated with the combustion of nonsulfur-bearing fuels in mixed fuel-fired
steam generating units. Emission credits would permit S02 emissions from a
mixed fuel-fired steam generating unit to increase to the same level they
would if the steam generating unit fired only oil or coal. A mixed fuel-
fired steam generating unit firing a 50/50 mixture of coal and wood waste, for
example, would be permitted to emit twice the S02 emissions with an emission
credit as it would without an emission credit. The same would be true when
cofiring natural gas.
The merits of emission credits for mixed fuel-fired steam generating
units in particular were thoroughly examined and discussed in several
Subpart Db supporting documents: "Summary of Regulatory Analysis for New
Source Performance Standards: Industrial-Commercial-Institutional Steam
Generating Units of Greater than 100 Million Btu/hr Heat Input"
(EPA-450/3-86-005), "An Analysis of Costs and Cost Effectiveness of S02
Control for Mixed-Fuel-Fired Steam Generating Units" (EPA-450/3-86-001), and
"Impacts of New Fuel Prices on S02 Emission Credits for Cogeneration Systems
and Mixed-Fuel-Fired Steam Generating Units". To assess the merits of
emission credits, the costs, SC>2 emissions, and cost effectiveness of SC>2
control were analyzed in these documents both with and without an emission
credit. These analyses show that granting an emission credit for mixed
fuel-fired steam generating units results in very small reductions in costs
while allowing significant increases in S02 emissions. The costs without a
2-21
-------
credit are considered reasonable in view of the significant additional
emission reductions achieved by not providing emission credits. Consequently,
the final standards do not include provisions for emission credits for mixed
fuel-fired steam generating units.
2.3.3 Percent Reduction Standard
1. Comment; Two commenters (IV-D-08, IV-D-28) requested that the
90 percent SOg reduction requirement be eliminated and replaced with an
emission limit of 520 ng/J (1.2 Ib/million Btu) heat input. One commenter
(IV-D-08) objected to applying the 90 percent S02 reduction requirement to all
coal regardless of sulfur content. This commenter stated that the EPA's
conclusion that no units will be built in the size range between 22 and 29 MW
(75 and 100 million Btu/hr) heat input capacity and operating at greater than
55 percent capacity factor is flawed. This commenter stated that the S02
standard of 520 ng/J (1.2 Ib/million Btu) heat input for coal-fired plants
should apply to all steam generating units in this source category, regardless
of size. This commenter further recommended that the full 90 percent $62
removal be required only when the 520 ng/J (1.2 Ib/million Btu) limit could
not be met by using low sulfur coals or by pretreating the coals.
Another commenter (IV-D-28) stated that the 90 percent S02 reduction
requirement should be removed and that coal-fired steam generating units in
the 8.7 to 29 MW (30 to 100 million Btu/hr) range should be required only to
meet the 520 ng/J (1.2 ID/million Btu) S02 limit. The commenter stated that
the percent reduction requirement would place an unjustified cost and
performance burden on units in this range that either already meet or are
close to meeting the 520 ng/J (1.2 Ib/million Btu) S02 limit.
Response: Section 111 of the CAA requires standards to reflect
application of the best demonstrated technology considering costs, nonair
quality health and environmental impacts, and energy requirements.
Section 111 also requires that for fossil fuel-fired steam generating units a
percent reduction standard be established. Read together, this means that the
2-22
-------
Administrator is compelled to include a percent reduction standard unless the
impacts associated with the requirements would be unreasonable. As discussed
in the background document, "Model Boiler Cost Analysis for Controlling Sulfur
Dioxide ($02) Emissions from Small Steam Generating Units" (EPA-450/3-89-14),
a small coal-fired steam generating unit of 22 MW (75 million Btu/hr) size and
operating at a 55 percent capacity factor has an incremental
cost-effectiveness value of about $3,600/Mg ($3,300/ton) relative to an
emission limit standard of 520 ng/J (1.2 Ib/million Btu). Capital and
annualized costs are projected to increase by approximately 20 percent over
the regulatory baseline for the percent reductions standard. However, these
values increase significantly for units less than 22 MW (75 million Btu/hr)
heat input capacity and for any unit less than 29 MW (100 million Btu/hr)
operating at an annual capacity factor for coal of less than 55 percent.
Imposing these high costs for these units was considered to be unreasonable
when compared to the increase in emission reductions achievable by the percent
reduction requirement on these units. Therefore, in keeping with the
requirements of the CAA, the final standards will not require percent
reduction for any units operating at less than a 55 percent annual capacity
factor for coal or any unit with a heat input capacity of 22 MW (75 million
Btu/hr) or less.
Finally, no conclusion was made that coal-fired steam generating units
greater than 22 MW (75 million Btu/hr) heat input and greater than 55 percent
capacity factor would not be built. Rather, this was a projection of sales
over the next five years based on sales trends over the past several years.
The sales projections for coal-fired units had no influence on the conclusion
of the reasonableness of the percent reduction requirement. (The assumption
was used in generating national impacts of the standards.) The model steam
generating unit analysis examined the potential impacts of the percent
reduction requirement on a coal-fired unit greater than 22 MW (75 million
Btu/hr) and greater than 55 percent capacity factor. Therefore, should a unit
be built, requiring 90 percent reduction of emissions would be reasonable.
2-23
-------
2. Comment: One commenter (IV-D-32) stated that the proposed S02 emission
limits cannot be supported by the cost and benefits analysis information
presented in the preamble to the proposed regulation. According to the
commenter, a very small number of coal-fired steam generating units are likely
to be built in the affected size category, so that emission reduction benefits
compared to the costs would be very minimal. The commenter estimated that,
when compared to the use of low sulfur coal, the increased costs of S02
removal to meet the proposed percent reduction requirement would be almost
$4,400/Mg ($4,000/ton). The commenter concluded that the decision of whether
to use S02 scrubbing should be the choice of the steam generating unit owner
or operator.
Response: As discussed in the preamble to the proposed standards, an
estimated national emission reduction of 11,000 tons/year would result from a
standard based on a 520 ng/J (1.2 Ib/million Btu) emission level for units
greater than 2.9 MW (10 million Btu/hr) heat input with a 90 percent SOg
reduction for units greater than 22 MW (75 million Btu/hr) heat input and
operating at greater than 55 percent capacity factor. National annualized
emission control cost increases over baseline for this standard are expected
to be $5.9 million, and national incremental cost effectiveness is expected to
be $770/Mg ($700/ton). For individual units, cost-effectiveness values range
from $550 to $3,600/Mg ($500 to $3,300/ton).
When assessing the reasonableness of the percent reduction requirement,
the number of steam generating units likely to be built in the affected size
category is not as important as the balance between the overall costs of the
standard and the overall benefits achieved. As shown above, the cost/benefit
analysis indicated that the upper range of cost-effectiveness values would be
more in the range of $3,300/Mg ($3,000/ton) rather than the $4,400/Mg
($4,000/ton) stated by the commenter. In considering all of the associated
costs of the standard, including national annualized cost, national capital
cost, and national cost effectiveness, as well as individual cost
effectiveness and other non-cost impacts, these standards are considered
reasonable.
2-24
-------
Finally, the choice of whether or not to use scrubbing to achieve the
standard remains at the discretion of the unit owner or operator. Scrubbing
is only one choice. The owner or operator can elect to use another control
device capable of a 90 percent reduction, or can avoid the percent reduction
requirement altogether by firing a fuel other than coal (such as oil or gas),
operating the unit at less than 55 percent capacity factor, or building units
of less than 22 MW (75 million Btu/hr) heat input capacity.
3. Comment: One commenter (IV-D-22) stated that based on previous studies
of flue gas desulfurization (FGD) systems, scrubbing costs were estimated to
at least equal, and probably exceed, the delivered cost of coal.
Response: Calculated from Table 4 in the background document entitled
"Model Boiler Cost Analysis for Controlling Sulfur Dioxide (S02) Emissions
from Small Steam Generating Units" (EPA-450/3-89-14), the cost of scrubbing
(i.e., the annualized cost) was determined to be less than 13 percent of the
delivered cost of low sulfur coal for a 22 MW (75 million Btu/hr) steam
generating unit operating at 55 percent capacity factor. The relationship
between increased cost due to scrubbing and the delivered cost of coal is not
as important as other factors in determining the reasonableness of the
standard. As stated in the previous comment, the impacts of the percent
reduction standard were determined to be reasonable.
4. Comment: One commenter (IV-D-26) stated that the 90 percent reduction
requirement requires the installation of complex emissions control systems
that cause unnecessary losses of steam supply reliability without significant
environmental gain. This commenter asserted that, in the case of small steam
generating units, the technology required to meet the percent reduction
requirement has not been adequately demonstrated and shown to be "reasonably
reliable" and "reasonably efficient . . . without becoming exorbitantly
costly". One commenter (IV-D-24) stated that the complexity of scrubber
systems presents an excessive burden to small steam generating facilities.
2-25
-------
Response: As discussed in the preamble to the proposed regulation,
available data show consistently high (96 percent) S02 removal efficiencies
for sodium scrubbing FGD systems. Available data for lime FGD systems
indicate that a 92 percent $03 removal efficiency could be achieved, and the
reliability of the lime wet scrubbing exceeded 91 percent. Likewise, lime
spray drying systems are capable of achieving 90 percent reduction in S02
emissions. Available data for fluidized bed combustion (FBC) units indicated
that an S02 removal efficiency of 91 to 99 percent can be achieved.
Therefore, the 90 percent reduction used in evaluating these controls is
reasonable.
To assure steam supply in case of a scrubber malfunction, a small steam
generating unit could be designed and constructed with alternative fuel firing
capability (such as natural gas or very low sulfur oil). If the unit
experiences a scrubber malfunction, the unit would not need to be taken
offline. Rather, the unit could switch over to firing the alternative backup
fuel until the malfunction is corrected. While the backup fuel is fired, use
of the scrubber is not required. Therefore, using a scrubber system to
achieve the percent reduction requirement would not cause a loss of steam
supply reliability. The additional costs for firing an alternative fuel
during a period of scrubber malfunction were included in the cost analyses.
The additional costs represent only a small percentage of the total steam
generating unit costs and were considered reasonable.
5. Comment: One commenter (IV-D-31) stated that requiring a 90 percent
reduction and achieving an emission limit of 520 ng/J (1.2 Ib/million Btu) on
units larger than 22 MW (75 million Btu/hr) penalizes low sulfur coal users by
encouraging use of coal with sulfur contents as high as 7.4 percent. He
questioned why an emission limit only, as proposed for oil-fired units, was
not proposed for coal-fired units. He further stated that the percent
reduction requirement should be eliminated and replaced with a finite emission
limit. He contended that the ultimate reduction of emissions to the
atmosphere is the most important factor, not the method used to achieve the
reduction.
2-26
-------
Response; Section 111 of the CAA requires standards to reflect
application of the best demonstrated technology taking into consideration
costs, nonair quality health and environmental impacts, and energy
requirements. Section 111 also requires that a percent reduction standard be
required unless the impacts associated with the requirements would be
unreasonable. The percent reduction standard applied to large coal-fired
units achieves greater emission reductions than would be achieved by an
emission limit only, and the costs, nonair quality health and environmental
impacts, and energy requirements of the final standards are considered
reasonable for large coal-fired units as described in the response to
Comment 2 of this section. The impacts on oil-fired units, on the other hand,
are very different. The 90 percent reduction requirement for oil would
increase capital costs by about 80 percent, annualized costs by about
40 percent, and the incremental cost effectiveness would be in excess of
$ll,000/Mg ($10,000/ton). Based on these costs, the application of a percent
reduction requirement for oil-fired units was judged to be unreasonable and
inappropriate under Section 111 of the CAA. However, the alternative option
of firing very low sulfur oil would increase capital costs for a typical unit
by less than 1 percent, annualized costs by about 20 percent, and the
incremental cost effectiveness is about $l,500/Mg ($l,400/ton). This is
considered reasonable and is the basis for the SC<2 standard for oil-fired
steam generating units.
6. Comment: One commenter (IV-D-25) stated that it was not clear whether
the three types of by-product fuels used in the steel industry (coke oven gas,
coal tar, and blast furnace gas) are included in the definition of coal in the
proposed standards. The commenter pointed out that prohibiting SC>2 emissions
from these types of by-product fuels would ensure that existing units firing
these fuels will not be replaced with new, more energy efficient units.
According to the commenter, it is uncertain whether 90 percent SC>2 reduction
is achievable for these by-product fuel-fired units with capacity greater than
22 MW (75 million Btu/hr). The commenter stated that due to the relatively
2-27
-------
low level of S0£ emissions from combusting these fuels, they should be
exempted from any percent reduction requirement and either be allowed to be
combusted with no emission limits or be regulated by a Ib/million Btu limit.
Response; The definition of "coal" in the regulation applies only to
conventional solid coal fuels, coal refuse, petroleum coke, or to coal-
derived synthetic fuels, such a solvent-refined coal, gasified coal, and
coal-oil or coal-water mixtures. The definition for "coal" does not include
coke oven gases, coke oven coal tars, or blast furnace gases. Similarly, coke
oven gases and blast furnace gases are not covered under the definition of
"natural gas," nor are coke oven coal tars included under the definitions of
"distillate oil," or "residual oil". There are, therefore, no emission limits
applicable to coke oven gases, coke oven coal tars, or blast furnace gases
under these standards.
7. Comment: One commenter (IV-D-25) provided an example of an 24 MW
(80 million Btu/hr) steam generating unit firing 50 percent oil and 50 percent
blast furnace gas. The commenter stated that the unit's allowable emission
rate would be 12 MW (40 million Btu/hr) (heat input from oil) times 215 ng/J
(0.50 Ib/million Btu) (emission limit for oil), or 20 Ib/hr. The commenter
further stated that such a unit would actually emit 20 Ib/hr from the oil and
2.8 Ib/hr from the gas (22.8 Ib/hr total), resulting in an apparent violation
even though the unit is firing compliance oil. The commenter stated that, in
order to remain in compliance with the new regulation, this unit would need to
fire exclusively oil even though such operation would result in greater S02
emissions (i.e., 24 MW [80 million Btu/hr] times 215 ng/J [0.50 Ib/million
Btu], or 40 Ib/hr).
Response: In the example described above, the commenter was correct
when he excluded the heat input from the blast furnace gas in determining the
allowable SOg emission rate. There are no emission credits for mixed fuel-
fired units. On the other hand, there are also no emission penalties for such
units. The commenter was incorrect when he included the S02 emissions from
2-28
-------
the blast furnace gas in determining whether or not the unit was in
compliance. As discussed in the response to the previous comment, there are
no emission limits applicable to blast furnace gas under these standards. In
the situation described by the commenter, the owner or operator could
demonstrate compliance with the standards either by obtaining fuel receipts
that state that the oil meets the ASTM specifications for distillate oil, or
by fuel sampling to determine the sulfur content of the oil.
8. Comment: Four commenters (IV-D-24, IV-D-28, IV-D-37, IV-D-40) stated
that the cost-effectiveness value that was used as the basis of the 90 percent
reduction requirement, $3,700/Mg ($3,300/ton) of S02 removal, is not
justifiable and not consistent with established precedents. The commenters
stated that the cost-effectiveness levels of the 90 percent reduction
requirement exceed the $3,300/Mg ($3,000/ton) guideline and violate the legal
requirement that an adequately demonstrated system not be "exorbitantly
costly". One commenter (IV-D-37) stated that the percent reduction
requirements seem unnecessarily severe based on the cost-effectiveness
figures, and suggested that the percent reduction requirement be eliminated.
Response: The cost-effectiveness value of $3,600/Mg ($3,300/ton) is
considered reasonable (i.e., "justifiable"). There is no legal or other
requirement that limits the acceptable cost-effectiveness value to $3,300/Mg
($3,000/ton). Consideration of this value, however, was not the sole basis
for selecting the 90 percent reduction requirement. Selection of this percent
reduction requirement was based on consideration of all the impacts associated
with the requirement, including S02 emission reductions, annualized and
capital costs, non-air quality environmental impacts, and energy requirements.
Taking all of these factors into consideration, it was concluded that the
potential impacts of the percent reduction requirement are reasonable.
9. Comment: Two commenters (IV-D-24, IV-D-26) predicted that coal-fired
units in the heat input capacity range greater than 22 MW (75 million Btu/hr)
2-29
-------
and greater than 55 percent capacity factor may disappear due to the
imposition of this percent reduction requirement.
Response: No information is available to support the claim that the
percent reduction standard would preclude the construction of steam generating
units above 22 MW (75 million Btu/hr) operating at annual capacity factors
greater than 55 percent. The market price among fuels and the reliability of
fuel supplies has a greater impact on fuel selection than the potential costs
of the percent reduction requirement. The percent reduction requirement does
increase the annualized cost of firing coal by about 20 percent; however, this
increase, in and of itself, is not sufficient to cause a shift of new coal-
fired units in the future. Over the past decade, fluctuations in coal, oil,
and natural gas fuel prices have frequently and often dramatically exceeded
20 percent. Therefore, if the absolute level of coal use for this source
category declines in the future, it would most likely be due to abundant
supplies and low prices for oil and natural gas, not the imposition of the
percent reduction requirement.
10. Comment: Three commenters (IV-D-24, IV-D-28, IV-D-40) stated that the
proposed conditional exemption of steam generating units operating at
55 percent capacity factor or less is not useful because it would not be
prudent for an industrial steam generating unit owner or operator to accept
such a capacity restriction on a new unit. To do so would restrict plant
expansion and would limit the use of the new unit should other units develop
problems.
Response: The standards do not require that an owner or operator accept
a capacity factor restriction on a new unit. It is left to the discretion of
an owner or operator whether that owner or operator wishes to operate a new
unit at less than a 55 percent annual capacity factor. The intention of the
55 percent capacity factor cutoff is to ensure that the percent reduction
requirement is applied only where the impacts of the percent reduction
requirement are reasonable.
2-30
-------
Also, the owner or operator of an affected facility would not
necessarily have to accept capacity factor restrictions for the life of the
affected steam generating unit. When the unit is initially installed, the
owner or operator can apply for a permit condition that restricts operation of
the unit to below a 55 percent capacity factor for coal. If the plant is
expanded and a scrubber is installed to comply with the percent reduction
requirement, the owner or operator can apply for a modification to the permit
to remove the capacity factor restriction. Thus, the cost of complying with a
percent reduction requirement is not incurred until the additional capacity is
needed. At that higher capacity level, the impacts of the percent reduction
requirements are reasonable.
Finally, it is possible that the situation may occur where other units
develop operational problems and must shutdown, requiring an affected unit to
increase its operating capacity to cover the steam demand. This situation can
also be solved without an undue burden on a facility. The percent reduction
requirement applies only when the annual capacity factor for coal exceeds
55 percent. Backup fuel supplies, such as natural gas or very low sulfur oil,
can be fired to prevent the annual capacity factor for coal exceeding
55 percent. In this way the owner or operator can avoid the need to meet the
percent reduction requirement.
11. Comment; One commenter (IV-D-40) recommended that the EPA establish a
more reasonable percent reduction requirement for units with an annual heat
input capacity of less than 29 MW (100 million Btu/hr), that could be achieved
by precleaning the coal at the mine.
Response: The background document entitled "Overview of the Regulatory
Baseline, Technical Basis, and Alternative Control Levels for Sulfur Dioxide
(S02) Emission Standards for Small Steam Generating Units," (EPA-450/3-89-12)
includes a discussion of the available control technologies that were used in
developing regulatory alternatives for reducing S02 emissions from coal-fired
steam generating units. In the section on low sulfur coal, consideration was
given to coal treatment and physical coal cleaning for reducing the naturally
2-31
-------
occurring sulfur content of coal. However, as pointed out in this background
document, coal cleaning is just one method used for producing a coal capable
of achieving the 520 ng/J (1.2 To/million Btu) emission limit. By examining
the reasonableness of the standards based on a 520 ng/J (1.2 Ib/million Btu)
coal, the reasonableness of the standard based on coal cleaning was also
examined.
The percent reduction requirement applies only to small coal-fired steam
generating units with a heat input capacity greater than 22 MW (75 million
Btu/hr) and an annual capacity factor greater than 55 percent. Units with a
heat input capacity of 22 MW (75 million Btu/hr) or less, and units with an
annual capacity factor less than 55 percent, are not subject to the percent
reduction requirement and can use cleaned coal to meet the 520 ng/J
(1.2 Ib/million Btu) emission limit. As discussed in Comment 2 of this
section, the percent reduction requirement achieves greater emission
reductions than would be achieved by an emission limit only, and the costs,
nonair quality health and environmental impacts, and energy requirements of
the percent reduction requirement are considered reasonable. Furthermore, the
costs and emission reductions associated with a 90 percent reduction
requirement and a 520 ng/J (1.2 ID/million Btu) emission limit (whether
achieved by coal cleaning or through use of naturally low sulfur coal) were
examined in "Model Boiler Cost Analysis for Controlling Sulfur Dioxide ($62)
Emissions from Small Steam Generating Units" (EPA-450/3-89-14). Based on the
analyses therein and the overall evaluation of regulatory options, the
requirement for a 90 percent S02 reduction was found to be reasonable for the
larger, more polluting units (i.e., those with a heat input capacity greater
than 22 MW [75 million Btu/hr] and annual capacity factor greater than
55 percent) in the small steam generating unit source category.
2.3.4 Emerging Technology Standard
1. Comment: Three commenters (IV-D-26, IV-D-27, IV-D-45) objected to the
proposed S02 emission limit of 260 ng/J (0.60 Ib/million Btu) for new
technologies. One commenter (IV-D-26) stated that the proposed limit for new
2-32
-------
technologies would be biased against the use of high sulfur coal and act as a
deterrent for developing new control technologies to use with medium and high
sulfur domestic coals. This commenter also stated that, as a matter of
policy, emerging technologies should always be subject to a less stringent
performance standard that encourages the development of effective new
technologies. This commenter recommended that the proposed 50 percent
reduction and 260 ng/J (0.60 Ib/million Btu) S02 emission limit be modified to
50 percent reduction and 690 ng/J (1.6 Ib/million Btu) S02 emission limit.
Another commenter (IV-D-27) agreed with the proposed 50 percent
reduction, but stated that the proposed S02 emission limit of 260 ng/J
(0.60 Ib/million) Btu for new technologies is too low. This commenter pointed
out that owners/operators will not want to use an emerging technology if they
have to use low sulfur coal. One commenter (IV-D-45) suggested that until
sufficient data are available, emerging technologies should be explicitly
exempted from the S02 standard. The commenter continued that, if an S02
standard for specific emerging technologies were to be set, then the standard
should be the same 520 ng/J (1.2 Ib/million Btu) standard as applies to
conventional coal-fired technologies. The proposed $62 standard for emerging
technologies is inappropriately biased against medium and high sulfur coals.
Response: The purpose of the emerging technologies provisions of the
standard are to encourage the development of new emission control technologies
for steam generating units that may be more effective or more economical than
conventional technologies, while ensuring adequate protection of air quality.
In the absence of special provisions for these technologies, all S02 control
technologies applied to steam generating units larger than 22 MW (75 million
Btu/hr) that operate at a capacity factor greater than 55 percent would be
required to meet the 90 percent reduction requirement. A 90 percent reduction
requirement poses a potential obstacle to the development and adoption of new
emission control technologies that have an unproven record in actual
applications. Therefore, by providing that bona fide emerging technologies
can comply with a lower percent reduction requirement of 50 percent, this
standard allows the continued development of emission control technologies
2-33
-------
that show promise of achieving levels of performance comparable to existing
technologies.
The emerging technology provisions of the standard couple the less
stringent percent reduction requirement with a more stringent emission
limitation of 260 ng/J (0.60 Ib/million Btu) heat input. This emission limit
protects air quality that might otherwise be adversely affected by the
50 percent reduction requirement by effectively requiring steam generating
units that use emerging technologies to fire low sulfur coal. Requiring use
of low sulfur coal during the development phase of a rrew technology is not
unreasonable in light of the need to protect air quality. Also, to the extent
that reductions greater than 50 percent can be achieved, coals with higher
sulfur contents can be used.
It should be reemphasized that the emerging technology provisions of the
standards apply only to the largest coal-fired units in this source category
(i.e., those with heat input capacities greater than 22 MW [75 million Btu/hr]
and operating at a capacity factor greater than 55 percent). Below that size
or capacity factor, coal-fired units are subject to an SOg emission limit of
520 ng/J (1.2 Ib/million Btu) and no percent reduction requirement.
2. Comment: One commenter (IV-D-27) stated that the economics of a new
sulfur removal method that achieves a 65 percent removal of S02 in a very
small 20 million Ib/hr steam generating unit by firing micronized coal and
micronized limestone would eliminate the need for the SOg percent reduction
requirement. The commenter viewed the percent reduction requirement as a
barrier to the development of this new technology. The commenter cited the
cost effectiveness of the new technology, indicating that in the first year of
use it pays for the total cost of using coal. He further stated, however,
that the percent reduction requirement in the proposed standard prevents those
steam generating units between 22 to 29 MW (75 and 100 million Btu/hr) heat
input and greater than 55 percent capacity factor from using this new coal
micronization method. The commenter recommended dropping the 90 percent
removal requirement to 50 percent to encourage development of this as well as
other sulfur removal methods.
2-34
-------
Response: The NSPS in no way would discourage the development of
micronized coal technology. The technology as described by the commenter
would qualify as an emerging technology under Section 60.42c(b)(2) of the
rule. Therefore, units in the range from 22 to 29 MW (75 to 100 million
Btu/hr) heat input capacity that use this technology would be subject to the
standards for an emerging technology, which is a 50 percent reduction in
potential S02 emissions and an S02 emission limit of 260 ng/J (0.60 Ib/million
Btu) heat input. If this technology is installed on a unit not subject to the
percent reduction requirement (i.e., any small steam generating unit with a
heat input capacity of 22 MW [75 million Btu/hr] or less, or operated at an
annual capacity factor for coal of 55 percent or less), the emerging
technology provisions do not apply. In these cases, the control device must
only be capable of achieving the 520 ng/J (1.2 ID/million Btu) emission rate.
3. Comment: One commenter (IV-D-45) stated that the disallowance of credit
for precombustion cleaning of fuel toward the S02 percent reduction
requirement for emerging technologies is inappropriate.
Response: Although the allowance of full credit for fuel pretreatment
toward the 90 percent reduction requirement is appropriate and is allowed in
the final standards, giving credit for fuel pretreatment toward the 50 percent
reduction requirement for emerging technologies is inappropriate. The primary
objective of the 50 percent reduction requirement is to stimulate and
encourage the development and use of "add-on" control systems or "in-situ"
combustion types of emerging S02 control technologies that show promise of
ultimately achieving emission reductions comparable to the 90 percent
reduction that can currently be achieved by conventional technologies. Any
emerging technology that is unable to reduce its emissions by 50 percent on
its own merits, without the use of fuel pretreatment credits, is unlikely to
have the potential for achieving S02 removal levels of 90 percent in the
future.
2-35
-------
4. Comment: One commenter (IV-D-45) stated that, although approval is
required for a facility to operate as an emerging technology, it is important
that an owner or operator is certain that the facility's technology so
qualifies before substantial financial commitments are made. The commenter
suggested that the rules impose a time limit of 30 days on EPA approval of
emerging technology applications, similar to the time limit imposed under
40 CFR 60.5 on construction/modification determinations.
Response: It is the EPA's practice to respond to such matters in a
timely manner. Requiring a facility to demonstrate that it is indeed
incorporating an emerging technology and not merely incorporating some minor
change to a conventional technology is not considered unreasonable, given the
significant benefits associated with qualifying as an emerging technology.
While a 30-day time period may be sufficient time for review in some cases,
other situations may require a somewhat longer period to complete a review.
2.4 STANDARD FOR PARTICULATE MATTER
2.4.1 Emission Limit - Coal
1. Comment: One commenter (IV-D-45) recommended a PM standard of 22 ng/J
(0.05 Ib/million Btu) for coal-firing units greater than 22 MW (75 million
Btu/hr) and 55 percent capacity factor, a standard of 130 ng/J
(0.30 ID/million Btu) for all other coal-firing units greater than 14.6 MW
(50 million Btu/hr), and no standard for units under 14.6 MW (50 million
Btu/hr). The commenter stated that dividing the general source category into
smaller "subcategories" is in accord with both Section lll(b)(2) of the Clean
Air Act which permits the Administrator to "distinguish among classes, types,
and sizes within categories," and the statute's directive to "consider the
cost of achieving such emission reduction."
Response: As stated in the preamble, differences in design, fuel-use
patterns, capacity, end use, emission characteristics, and potential for
2-36
-------
emission reduction were considered in determining size categories to be
addressed in the proposed standards. The size categories chosen for
regulatory purposes (i.e., 8.7 to 29 MW, 2.9 to 8.7 MW, and less than 2.9 MW
[30 to 100 million Btu/hr, 10 to 30 million Btu/hr, and less than 10 million
Btu/hr] heat input) reflect the units' differences in end use application.
Units in the 8.7 to 29 MW (30 to 100 million Btu/hr) category are used
primarily in industrial facilities; units between 2.9 and 8.7 MW (10 and
30 million Btu/hr) are used primarily at large institutional or commercial
facilities such as major hospitals, large hotels, colleges and universities,
or large commercial laundries. The smallest units, those with less than
2.9 MW (10 million Btu/hr) heat input, primarily serve schools, shopping
centers, or other small facilities.
A 14.6 MW (50 million Btu/hr) unit is considered to be representative of
the 8.7 to 29 MW (30 to 100 million Btu/hr) range; the impacts for a unit
below 14.6 MW (50 million Btu/hr) will be somewhat higher than those for units
above 14.6 MW (50 million Btu/hr). However, these differences in impacts were
considered and are discussed in the background documents, and it was
determined that the impacts are reasonable. Consequently, the PM standards
apply to all boilers with greater than 8.7 MW (30 million Btu/hr) heat input
capacity. The commenter provided no data to support the emission limits and
size cutoffs recommended.
2. Comment; One coal industry commenter (IV-D-18) stated that since the
National Ambient Air Quality Standards (NAAQS) for particulate matter smaller
than 10 microns in diameter (PMjo) are more stringent than the proposed NSPS
for PM, EPA should not consider health risks for PMjQ in selecting a PM
standard for steam generating units. The commenter stated that in NAAQS
attainment areas there would be no health effects of PMio; in nonattainment
areas, the more stringent NAAQS for PM would apply.
The commenter further stated that he agreed with the EPA's rejection of
PM standard Alternative III because of unreasonable cost effectiveness and the
significant burdens that it placed on industry; however, the commenter
asserted that the same considerations should lead to the rejection of
2-37
-------
Alternative II, as well. The commenter stated that selection of
Alternative II over Alternative I represents a cost-effectiveness increase of
267 percent for an increased emission reduction of 360 Mg (400 tons).
Response: Ambient concentrations of PMjn, are regulated by NAAQS, which
are intended to achieve and maintain ambient air quality in order to protect
public health and welfare. The purpose of NSPS is to minimize emissions
through application of best demonstrated technology, and to prevent new air
pollution problems or exacerbation of existing problems.
In developing an NSPS, alternatives are evaluated to determine which
alternative represents best demonstrated technology (BDT), considering costs
and other impacts. The cost effectiveness of each alterative is an important
consideration, but is not the single or primary criterion for concluding which
alternative represents best demonstrated technology. As stated in the
preamble, Alternatives II and III are based on the use of fabric filters or
electrostatic precipitators (ESP's), both of which are demonstrated control
technologies for PM at 22 ng/J (0.05 Ib/million Btu), while Alternative I is
based on the use of double mechanical collectors, which are demonstrated to
achieve only 130 ng/J (0.30 ID/million Btu).
Although the costs of achieving a 22 ng/J (0.05 Ib/million Btu) standard
are relatively high, PM and PMjo emission reductions are significantly greater
than those achieved by a 130 ng/J (0.30 Ib/million Btu) standard. In
addition, trace metals and other toxic compounds from coal combustion are more
effectively controlled by Alternatives II and III. It was further determined
that Alternative III, although based on best demonstrated technology, placed
an undue burden on units with less than 8.7 MW (30 million Btu/hr) heat input
capacity. Alternative II, however, which requires an emission limit of
22 ng/J (0.05 Ib/million Btu) for coal-fired units with heat input capacity of
8.7 MW (30 million Btu/hr) or greater, is considered reasonable and is,
therefore, the basis of the standards.
3. Comment; Four commenters (IV-D-22, IV-D-24, IV-D-26, IV-D-28) stated
that requiring the use of a fabric filter or ESP for all coal-fired steam
2-38
-------
generating units above 8.7 MW (30 million Btu/hr) heat input capacity to meet
the PM emission limit of 22 ng/J (0.05 ID/million Btu) is not justified. One
commenter (IV-D-24) stated that the cost of a fabric filter is very high
relative to the cost of the unit. The commenter recommended a PM standard of
130 ng/J (0.30 Ib/million Btu), which would allow for the use of dual
mechanical collectors or sidestream separators.
The commenters pointed out that in the preamble for the proposed rule,
the SIP PM limits for small units range from 142 to 198 ng/J (0.33 to
0.46 Ib/million Btu). The commenters stated that EPA should consider the
different baseline emission fates of various types of steam generating units
in the rulemaking process in order to apply a reasonable standard to all types
of units. The commenters continued that it is important to consider the type
of steam generating unit as well as the control device in evaluating control
technology and cost effectiveness, and that a more reasonable and cost
efficient standard would vary according to the type of unit involved. The
commenters recommended the following PM limits for different sizes and types
of units: (1) stoker-fired units in the 8.7 to 22 MW (30 to 75 million
Btu/hr) range should be limited to 130 ng/J (0.30 ID/million Btu), based on
the use of mechanical collectors; (2) stoker-fired units in the 22 to 29 MW
(75 to 100 million Btu/hr) range should be limited to 86 ng/J (0.20 Ib/million
Btu), based on the use of sidestream separators with mechanical collectors;
and (3) pulverized coal and fluidized bed units in the 8.7 to 29 MW (30 to
100 million Btu/hr) range should be limited to 22 ng/J (0.05 Ib/million Btu),
based on the use of fabric filters or ESP's.
t
Two commenters (IV-D-08, IV-D-45) objected to applying the PM standard
to coal-fired steam generating units smaller than 14.6 MW (50 million Btu/hr)
heat input capacity. One commenter (IV-D-08) stated that the PM standard
would require use of an electrostatic precipitator or a fabric filter, and
that these technologies are not appropriate for units smaller than 14.6 MW
(50 million Btu/hr) heat input capacity.
Two commenters (IV-D-22, IV-D-26) presented data showing that the
difference in tons of PM removed under the EPA's proposed standard and under
the commenter's proposed alternative is 91 Mg/yr (83 tons/year). The
commenters presented further data comparing the cost effectiveness of a
2-39
-------
130 ng/J (0.30 Ib/million Btu) standard, an 86 ng/J (0.20 Ib/million Btu)
standard, and a 22 ng/J (0.05 Ib/million Btu) standard for three different
size ranges of steam generating unit. The commenters stated that these data
indicate that the 130 ng/J (0.30 Ib/tnillion Btu) standard is reasonably cost-
effective for all size ranges, whereas the costs of the 86 and 22 ng/J
(0.20 and 0.05 Ib/million Btu) standards for all units far exceed the EPA's
own cost guidelines of $3,300/Mg ($3,000/ton). One commenter (IV-D-22) also
pointed out that, for the NSPS for utility steam generating units, the cost-
effectiveness values ranged from $44 to $77/Mg ($40 to $70/ton) for PM.
Response: Fabric filters and ESP's have been demonstrated to achieve PM
emission levels of 22 ng/J (0.05 Ib/million Btu) for coal-fired units with
greater than 8.7 MW (30 million Btu/hr) heat input. The background document
entitled "Overview of the Regulatory Baseline, Technical Basis, and
Alternative Control Levels for Particulate Matter (PM) Emission Standards for
Small Steam Generating Units" (EPA-450/3-89-11) presents data for coal-fired
units from 14 to 60 MW (48 to 208 million Btu/hr) heat input capacity equipped
with fabric filters and ESP's. Using fabric filters, PM emission levels were
less than 22 ng/J (0.05 Ib/million Btu) for all units tested. Using ESP's, PM
emission levels were also less than 22 ng/J (0.05 Ib/million Btu) for all
units tested. The cost-effectiveness value for the PM standard ranges from
$2,800 to $7,300/Mg ($2,500 to $6,600/ton). The cost of PM control relative
to the cost of the unit was also calculated and ranges from about 8 to
12 percent. Although the cost-effectiveness and the cost of the fabric filter
in comparison to the cost of the entire unit are important factors in
determining the "reasonableness" of an NSPS, these are not the only factors
that are considered. The CAA states that the selection of a standard must be
based on the "best demonstrated technology," taking into account cost, nonair
quality health and environmental impacts and energy requirements. All these
aspects were considered in determining that these requirements are reasonable
in light of the significant reductions achieved for PM and PMjQ emissions, as
well as for emissions of trace metals and other toxics.
In determining the baseline emission level for the PM standard, data
from different types and sizes of steam generating units, including stoker
2-40
-------
fired and fluidized bed units, were considered. Mechanical collectors, the
demonstrated control technology for baseline emissions, are a well established
technology for removing PM from a gas stream, and have been widely used to
control PM emissions from all types of coal-fired steam generating units.
Available data indicate that coal-fired units equipped with single mechanical
collectors can achieve emission levels of 260 ng/J (0.60 Ib/million Btu) for
spreader stokers and 245 ng/J (0.45 Ib/million Btu) for underfeed stokers.
Underfeed stokers are predominant in to 2.9 to 8.7 MW (10 to 30 million
Btu/hr) range while spreader stokers are most prevalent above this size range.
Therefore, these types of units and controls are considered representative of
small coal-fired steam generating units, and the baseline emission levels for
coal-fired units used in the analyses have been set accordingly.
For the purpose of regulation, affected facility size categories were
determined through analysis of steam generating unit population data. These
data indicate that units in the size range of 29 MW (100 million Btu/hr) heat
input and less fall into two distinct segments: commercial-institutional and
industrial units. Industrial units serve manufacturing and other production
facilities; they are primarily in the size range from 8.7 to 29 MW (30 to
100 millon Btu/hr) heat input. Commercial-institutional units serve smaller
facilities and are usually below 8.7 MW (30 million Btu/hr) heat input. Units
less than 8.7 MW (30 million Btu/hr) heat input capacity are not subject to
the PM standards because the costs are unreasonable and the availability of
operators capable of handling the control systems is questionable in this
range. The impacts of regulating units above 8.7 MW (30 million Btu/hr) heat
input capacity is reasonable considering the emission reductions achievable by
the standard.
In setting emission limits for an NSPS, the alternative that achieves
the greatest emission reductions using demonstrated control technology, and
for which the cost, energy, and other impacts are reasonable, is selected.
The absolute amount of emissions reduced from a source category is not
directly considered when evaluating a standard. If the impacts of achieving
the reduction (however large or small) are reasonable, the standard will be
set.
2-41
-------
It is not an objective of the standard to equalize cost effectiveness
across industries; rather the objective is to determine the most effective
control technologies to accomplish the greatest emission reduction,
considering cost and environmental impacts. Therefore, a cost-effectiveness
value for one source category does not set the limit for all future standards.
The standard for utility steam generating units cited by the commenter, for
instance, was based on the performance of the most effective control
technologies applicable to utility sources. The costs of these technologies
were determined to be reasonable in the utility rulemaking. The relative cost
effectiveness of different control alternatives was not a controlling factor
in setting that standard. The $44 to $77/Mg ($40 to $70/ton) cost
effectiveness value associated with that standard was the result of the
selection of the most effective control technology, and was not intended to be
a benchmark for subsequent standards, or to imply that any control with a cost
effectiveness of greater than $77/Mg ($70/ton) would be unreasonable.
4. Comment: Two commenters (IV-D-24, IV-D-26) stated that the EPA's
argument that mechanical collectors require a high level of maintenance is not
accurate. The commenters argued that mechanical collectors are very simple
devices with no moving parts, but that baghouses and ESP's are much more
complicated and require much more attention. For these reasons, the
commenters asserted that mechanical collectors are more appropriate for small
steam generating units than baghouses or ESP's because the maintenance and
operating expertise level required is lower with a mechanical collector.
Response: As stated in "Overview of the Regulatory Baseline, Technical
Basis, and Alternative Control levels for Particulate Matter (PM) Emission
Standards for Small Steam Generating Units" (EPA-450/3-89-11), although
mechanical collectors have no moving parts (i.e., internal working
mechanisms), they are subject to erosion from abrasive fly ash. Also, air
leakage tends to occur over time. Thus, regular maintenance is necessary to
ensure good performance at design levels.
2-42
-------
It is true that ESP's and fabric filters require maintenance.
Facilities that operate steam generating units larger than 8.7 MW (30 million
Btu/hr), which are typically industrial facilities, have the resources and the
operators to give these systems the maintenance and operating assistance they
require. The costs for maintenance of these technologies were included in the
cost analyses and were considered reasonable. Therefore, ESP's and fabric
filters are considered best demonstrated technology for PM control,
considering costs and other impacts.
5. Comment: One commenter (IV-D-32) stated that the EPA based the 22 ng/J
(0.05 Ib/million Btu) PM standard on the assumption that affected facilities
will be located predominantly in urban areas and emissions will be released
through short stacks. The commenter stated that the permitting process for
steam generating units in the affected size range generally requires "good
engineering practice" heights to ensure sufficient dispersion. Furthermore,
the commenter stated that it is typical for smaller commercial units to be
located in urban areas and to be constructed with short stacks, but such units
have been exempted from the PM requirements. Therefore, the commenter stated
that the EPA's justification for requiring fabric filters on small industrial
coal-fired units is based on faulty logic. The commenter stated that the cost
of these control devices will be nearly twice the normal cut-off level for
each ton removed. The commenter recommended a PM limit of 86 ng/J
(0.20 Ib/million Btu).
Response: Although many affected facilities will be located in urban
areas and may have short stacks, this information was offered as an indication
of the significance of this source of PM, not as the basis for setting the PM
standard. The PM standard is based on emission reductions achievable using
the best demonstrated control technology. Available data on steam generating
units ranging in size from 14 to 60 MW (48 to 208 million Btu/hr) indicate
that fabric filters are capable of reducing PM emissions to less than 22 ng/J
(0.05 Ib/million Btu).
2-43
-------
Smaller commercial-institutional units (i.e., those less than 8.7 MW
[30 million Btu/hr] heat input) are exempt from the PM standard because units
in this size range are unlikely to have full-time, professional operators
capable of operating and maintaining emissions control equipment, and because
of very high cost-effectiveness levels. The impacts of compliance, therefore,
would be unreasonable.
The costs associated with the use of fabric filters to control PM
emissions on steam generating units over 8.7 MW (30 million Btu/hr) heat input
capacity are estimated at $2,800 to $7,300/Mg ($2,500 to $6,600/ton). This
cost effectiveness value reflects the use of best demonstrated technology and,
considering the substantial reductions of PM and PMjo emissions, as well as of
trace metals, is considered reasonable. Selection of NSPS regulatory
alternatives is not intended to equalize cost effectiveness across source
categories; rather, it is intended to reflect the use of best demonstrated
technology, considering costs and other impacts.
6. Comment: One commenter (IV-D-24) recommended a PM standard of 130 ng/J
(0.30 Ib/million Btu) based on the use of double mechanical collectors (DMC's)
or sidestream separators, and stated that this approach would allow units to
meet the 20 percent opacity limit and would establish the cost of compliance
at a reasonable level.
Response: The data assessed in the background documents indicate that
DMC's are capable of achieving PM emission levels of 130 ng/J (0.30 Ib/million
Btu). Fabric filters and ESP's, on the other hand, are capable of achieving
PM emission levels of 22 ng/J (0.05 Ib/million Btu). The costs associated
with the PM standards are described in the proposal preamble; for coal-fired
units with greater than 8.7 MW (30 million Btu/hr) heat input, the cost
effectiveness value is $2,800 to $7,300/Mg ($2,500 to $6,600/ton). Although
this value is higher than the cost effectiveness value for a standard based on
DMC's, it is not considered unreasonable in light of the increased reduction
of PM and PMjo emissions, as well as the reduction of trace metals and other
toxics afforded by the standard.
2-44
-------
2.4.2 Emission Limit - Wood
1. Comment: Two commenters (IV-D-07, IV-D-33) objected to the PM limit of
43 ng/J (0.10 Ib/million Btu) for wood-fired steam generating units with
greater than 8.7 MW (30 million Btu/hr) heat input capacity. They felt that
the emission limit is too low.
One commenter (IV-D-07) said that his plant was equipped with a wet
scrubber, and achieved particulate emissions of 0.12 Ib/million Btu. To
achieve the 43 ng/J (0.10 Ib/million Btu) limit would require modifications to
or replacement of the wet scrubber and would increase induced draft (ID) fan
horsepower by 60 percent and overall plant horsepower requirements by
25 percent. The additional annual fan electric power operating cost was
estimated at $65,000, or 4 percent of the plant's operating budget. The
commenter, therefore, claimed that achieving a 43 ng/J (0.10 Ib/million Btu)
PM limit with wet scrubber technology is a difficult and costly process. He
recommended replacing the PM limit of 43 ng/J (0.10 Ib/million Btu) with a
limit of 86 ng/J (0.20 Ib/million Btu) for small units based on good
combustion and high efficiency cyclones.
One commenter (IV-D-33) stated that EPA appears to have based the PM
limit on the fact that control technology to achieve the limit is available to
larger steam generating units, and has not fully addressed the costs of these
controls for small units.
Response: It should be noted that the standards apply only to new,
modified, or reconstructed units. Thus, there would be no need to modify or
replace the wet scrubbers on an existing unit with a more efficient system
unless the unit has been modified or reconstructed as defined under
Sections 60.14 or 60.15.
Wet scrubbers have been used widely for control of emissions from small
wood-fired steam generating units. The available data on the effectiveness of
wet scrubbers in reducing PM emissions indicate that the proposed standard of
43 ng/J (0.10 Ib/million Btu) is achievable in wood-fired steam generating
units with heat input capacities between 8.7 and 29 MW (30 and 100 million
Btu/hr). The background document entitled "Overview of the Regulatory
2-45
-------
Baseline, Technical Basis, and Alternative Control Levels for Participate
Matter (PM) Emission Standards for Small Steam Generating Units"
(EPA-450/3-89-11) also presents data showing that wood-fired units equipped
with an ESP or electrostatic gravel bed filter (EGF) can achieve PM emission
levels of less than 43 ng/J (0.10 To/million Btu). Therefore, these control
technologies are also considered to be demonstrated for reducing PM emissions
from wood fired steam generating units to 43 ng/J (0.10 Ib/million Btu) heat
input or less.
The PM control data analyzed for wood-fired steam generating units
include measurements of PM emissions from units ranging in size from 50 to
181 MW (170 to 615 million Btu/hr) equipped with ESP's, and from 16 to 67 MW
(55 to 230 million Btu/hr) equipped with wet scrubbers. Although many of the
units from which these data are drawn are larger than those included in this
source category, the characteristics of the emissions from these larger units
are similar to those of smaller units in terms of the concentration of
particles in the gas stream, the size distribution of those particles, the
resistivity of those particles, and their chemical composition. The primary
difference in emissions between the larger and smaller steam generating units
is the volume of the gas stream to be treated. The gas volume affects the
size and capacity of the control device, rather than the effectiveness of the
control technology in reducing the level of particulate matter emitted.
Consequently, these technologies are considered appropriate for units in the
size range from 8.7 to 29 MW (30 to 100 million Btu/hr). There are no
available data suggesting that these technologies are not transferable to
units in this size range.
The cost analysis performed for the various control options includes
estimates of capital, operation and maintenance, and annualized costs.
Although the cost-effectiveness values for achieving the standard are
relatively high, the standard results in greater overall PM emission
reductions and much greater reductions in PMjo emissions than the other
alternatives that were considered. Consequently, the impacts of compliance
for small units were found to be reasonable for units with annual capacity
factors above 30 percent.
2-46
-------
However, for wood-fired steam generating units with heat input
capacities between 8.7 and 29 MW (30 and 100 million Btu/hr) and operating at
an annual capacity factor below 30 percent for wood, the PM standard has been
increased to 130 ng/J (0.30 Ib/million Btu). This decision is based on
consideration of public comments and analysis of impacts on units operating at
low capacity factors. (See Section 2.4.2, Comment 3).
2. Comment: One commenter (IV-D-08) stated that small wood- and wood
waste-fired steam generating units should be subject to the same PM standard
as coal-fired units. The commenter stated that the impact of wood- and wood
waste-fired units on ambient air quality is very significant, and that the
EPA's reasons for applying a more lenient PM standard to these units were not
well presented. The commenter recommended a PM standard of 22 ng/J
(0.05 Ib/million Btu) for wood- or wood waste-fired steam generating units of
14.6 MW (50 million Btu/hr) heat input capacity and greater with no PM
standards for units smaller than 14.6 MW (50 million Btu/hr) heat input
capacity.
Response: Under Section 111 of the CAA, NSPS are based on application
of best demonstrated technology; available data indicate that an emission
level of 43 ng/J (0.10 Ib/million Btu) represents the performance of best
demonstrated technology for the control of PM emissions from small wood-fired
units. As discussed in the proposal preamble and in the background document
"Overview of the Regulatory Baseline, Technical Basis, and Alternative Control
Levels for Particulate Matter (PM) Emission Standards for Small Steam
Generating Units" (EPA-450/3-89-11), the control technologies analyzed for PM
emissions from small wood-fired steam generating units include DMC's, wet
scrubbers, and ESP's.
Considering data from units ranging in size from 7 to 43 MW (25 to
150 million Btu/hr), DMC's are demonstrated technology for reducing PM
emissions from small wood-fired steam generating units to 130 ng/J
(0.30 Ib/million Btu) or less. Wet scrubbers were tested on units ranging
from 16 to 67 MW (55 to 230 million Btu/hr) and are considered demonstrated
2-47
-------
control technology for controlling PM from small wood-fired units to 86 ng/J
(0.20 Ib/million Btu) or less with low pressure drops, and 43 ng/J
(0.10 Ib/million Btu) or less with medium pressure drops. For ESP's, test
data from units ranging in size from 49 to 178 MW (170 to 615 million Btu/hr)
were considered. The emissions test data indicate that ESP's are demonstrated
technology for controlling PM from small wood-fired units to 43 ng/J
(0.10 ID/million Btu) or less. Considering these three technologies, their
respective emission control efficiencies, and their associated cost, energy,
and other impacts, ESP's and wet scrubbers at medium pressure drops were
selected as best demonstrated control technologies for small wood-fired units.
Accordingly, the PM standard for these units is set at 43 ng/J
(0.10 Ib/million Btu). There are no available data that support a PM standard
for wood-fired units of 22 ng/J (0.05 Ib/million Btu).
As described in the proposal preamble, different size cutoffs for the PM
standard were considered. The impacts of a 43 ng/J (0.10 Ib/million Btu)
standard for units with heat input capacity greater than 8.7 MW (30 million
Btu/hr) are considered reasonable; however, the impacts of such a standard for
units below 8.7 MW (30 million Btu/hr) are considered unreasonable. This
occurs because steam generating units smaller than 8.7 MW (30 million Btu/hr)
heat input capacity are primarily commercial-institutional units, which
include those units located at offices and apartments, shopping centers,
hospitals, laundries, hotels, elementary and secondary schools, colleges and
universities, and other non-industrial facilities. Relatively high cost-
effectiveness values are associated with control of PM at these units, and
commercial-institutional facilities are not likely to have full-time skilled
operators capable of operating wet scrubbers or ESP's. Industrial units,
however, include those units that provide steam for manufacturing and other
production facilities. The cost effectiveness of PM control is more
attractive and these facilities are likely to hire full-time skilled operators
capable of operating the control equipment. Because industrial wood-fired
units are predominately found in the size range above 8.7 MW (30 million
Btu/hr), the PM standard for wood applies to units with heat input capacities
greater than 8.7 MW (30 million Btu/hr).
2-48
-------
3. Comment: One commenter (IV-D-33) stated that the EPA's cost-
effectiveness calculations for achieving the PM limit in wood-fired steam
generating units should not have been based on an average annual capacity
factor of 55 percent. The commenter noted that these calculations were based
on figures from a study attributed to the commenter's own organization, but
that he was unaware of any such study. Further, the commenter stated that an
annual capacity factor of 55 percent is unrealistically high for small wood-
fired units. The commenter presented figures from the background document
entitled "Model Boiler Cost Analysis for Controlling Particulate Matter (PM)
Emissions from Small Steam Generating Units" (EPA-450/3-89-15) to demonstrate
that annual capacity factor has a major effect on cost-effectiveness
calculations; lower annual capacity factors produce higher dollars/ton cost-
effectiveness estimates.
Response: The annual average capacity factor of 55 percent was not
taken from the report noted by the commenter. A capacity factor of 55 percent
has been determined to be typical of wood-fired steam generating units based
on an analysis of the data presented in the background document "Projected
Impacts of Alternative Particulate Matter New Source Performance Standards for
Industrial-Commercial-Institutional Nonfossil Fuel-Fired Steam Generating
Units" (EPA-450/3-89-18). As shown in that document, virtually all new wood-
fired steam generating units in the affected size range will be high capacity
factor units due to economic consideration. Other fuels would be more
economical below this capacity factor. Impacts of the 43 ng/J
(0.10 Ib/million Btu) standard were carefully analyzed for units operating at
a capacity factor of 55 percent and were determined to be reasonable for units
greater than 8.7 MW (30 million Btu/hr) in size.
It is acknowledged, however, that some lower capacity units may also be
constructed, for reasons other than cost. Consequently, the impacts of the PM
standard on units operating at low capacities (i.e., below 30 percent annual
capacity factor) were also carefully analyzed. Compared to small wood-fired
units with higher capacity factors, the cost effectiveness of the proposed PM
standard for low capacity factor wood-fired units below 29 MW (100 million
Btu/hr) in size can be relatively high. For example, the estimated cost
2-49
-------
effectiveness for small wood-fired units between 29 and 8.7 MW (100 and
30 million Btu/hr) operating at a capacity factor of 26 percent ranges from
about $11,000 to $18,000/Mg ($10,000 to $16,000/ton), depending on unit size.
In order to reduce the costs to these seasonal and standby-type units, which
are sometimes operated at extremely low annual capacity factors, a less
stringent PM standard of 130 ng/J (0.30 Ib/million Btu) is being included in
the final regulation for wood-fired units operating with annual capacity
factors of 30 percent or less based on the use of double mechanical
collectors. Such a standard will significantly reduce capital and operating
costs and cost effectiveness.
4. Comment: Two commenters (IV-D-33, IV-D-45) stated that the EPA's cost-
effectiveness estimate for achieving the PM emission limit is considerably
higher, on a dollars/ton basis, than has been considered reasonable for
previous NSPS rulemakings. Both commenters further stated that the EPA has
acknowledged that cost-effectiveness figures for wood-fired steam generating
units to achieve the PM standard are significantly greater than those for
other source categories. One commenter (IV-D-33) noted that the EPA has
estimated incremental cost-effectiveness values for small coal-fired units
that are much lower than the cost-effectiveness values for small wood-fired
units. Both commenters stated that there is clearly an inequitable cost
burden placed on new wood-fired steam generating units.
One commenter (IV-D-45) stated that the wide range of relatively high
cost-effectiveness values indicates the need to further distinguish among
sizes of wood-fired units, and to establish standards for subcategories based
on size and capacity factor. The commenter suggested a PM standard of
130 ng/J (0.30 Ib/million Btu) for wood-fired units; with this standard,
compliance could be achieved using DMC's.
Response: The estimated cost effectiveness of the PM standard for wood-
fired units, based on use of wet scrubbers or ESP's capable of reducing PM
emissions to 43 ng/J (0.10 ID/million Btu) or less, ranges from about $6,000
to 9,200/Mg ($5,400 to $8,300/ton), depending on the size of the steam
2-50
-------
generating unit. These values represent the cost effectiveness for small
wood-fired units with an annual capacity factor of 55 percent, which, as
mentioned above, is typical for industrial size wood-fired units. These
values are generally consistent with the cost-effectiveness range estimated
for small coal-fired units. Therefore, although it is not necessary to
equalize cost-effectiveness values across all classes of sources within the
source category, the cost effectiveness of the PM standard for the bulk of the
population of new wood-fired units is entirely consistent with the cost
effectiveness of the PM standard for small coal-fired units. Additionally, as
discussed in the previous response, the impacts of the PM standard on wood-
fired units operating at lower annual capacity factors (i.e. below 30 percent)
were examined and the standard for these units has been raised to 130 ng/J
(0.30 Ib/million Btu) to reduce the cost burden. This change results in
reductions in both the capital cost of the equipment and the annual operating
and maintenance costs. Consequently, the final PM standard does not pose an
inequitable burden on new wood-fired units.
The steam generating unit size categories used in establishing the PM
standard reflect differences in the end use, design, fuel-use, and emissions
characteristics of small steam generating units. The data presented in the
background documents indicate that small steam generating units fall into the
three size categories of 0 to 2.9 MW, 2.9 to 8.7 MW and 8.7 to 29 MW (0 to
10 million Btu/hr, 10 to 30 million Btu/hr and 30 to 100 million Btu/hr) heat
input capacity. The largest size units 8.7 to 29 MW (30 to 100 million
Btu/hr) generally serve industrial facilities such as manufacturers. These
units are primarily watertube units. Units in the middle category (2.9 to
8.7 MW [10 to 30 million Btu/hr]) serve large commercial and institutional
facilities where both watertube and firetube units are used. The smallest
units are generally in small commercial or institutional facilities (such as
churches and public schools) where firetube and cast-iron units predominate.
Available data on small steam generating units indicate that these size
categories are the most appropriate for setting PM standards because emission
reduction potential varies according to these size categories and because
there are differences in the typical uses of the units and the availability of
operating personnel among these size categories.
2-51
-------
The impacts of PM control on each of these three size categories of
small steam generating units were carefully examined. It was determined that
PM controls were both effective and did not impose unreasonable costs or other
impacts for units larger than 8.7 MW (30 million Btu/hr) heat input. For the
two smaller size categories, however, the costs of control are not considered
reasonable and trained operators capable of operating the PM control systems
are not typically available. Therefore, no PM standard has been established
for units smaller than 8.7 MW (30 million Btu/hr) heat input.
5. Comment: One commenter (IV-D-17) stated that the reasons cited in the
preamble to the proposed standard for exempting small wood-fired steam
generating units (less than 8.7 MW [30 million Btu/hr]) from the PM emission
limits also apply to units above this size range. The commenter pointed out
that wood-fired steam generating units have a much higher capital cost than
gas- and oil-fired steam generating units in this size range, 50 percent
higher in some instances. The commenter stated that the proposed standard
would increase capital costs for the larger units by 19 percent over the
regulatory baseline. He emphasized that such an increase would be very
detrimental to the wood energy industry and would effectively prevent the
construction of wood-fired steam generating units in the affected size range.
Response: The reasons cited in the proposal preamble for exempting
wood-fired steam generating units less than 8.7 MW (30 million Btu/hr) in size
from the PM emission limit are based on the determination that the burden on
these smaller units would be unreasonable, due to relatively higher cost-
effectiveness values and the fact that such units are unlikely to have skilled
operators on site to operate sophisticated PM control equipment.
For units in the size range above 8.7 MW (30 million Btu/hr) heat input,
control costs are estimated to be a 19 percent increase over baseline capital
costs, as indicated by the commenter. These increased capital costs were
included in the cost analyses of regulatory alternatives.
It should be noted, however, that annualized costs play a much larger
role in selecting what type of steam generating unit to install than capital
2-52
-------
costs. Annualized costs, in turn, are "driven" or determined primarily by
fuel costs. For example, at today's fuel prices, oil- or gas-fired units are
likely to be more attractive to install than wood-fired units. Likewise,
future changes in relative fuel prices between coal, oil, gas, and wood are
likely to exert far more influence and have much more impact on fuel choice
for a new unit than this NSPS.
It is estimated that, assuming "full cost pass-through," the cost of
services associated with most establishments using wood-fired steam generating
units would increase by less than 0.5 percent. These impacts are considered
reasonable for units above 8.7 MW (30 million Btu/hr). Therefore, the PM
standard is not expected to be detrimental to the construction of new wood-
fired units.
2.5 STANDARD FOR NITROGEN OXIDES
1. Comment: Seven commenters (IV-D-05, IV-D-09, IV-D-20, IV-D-26, IV-D-28,
IV-D-29, IV-D-45) expressed concern about the proposed NOX standard. Five
commenters (IV-D-09, IV-D-20, IV-D-26, IV-D-28, IV-D-29) stated that the EPA
has misinterpreted the district court's ruling when deciding to set NOX
standards after determining that the standards would be unreasonable. These
commenters stated that EPA has discretion and is ordered to make a decision
only on whether to promulgate NOX regulations for small steam generating
units. One commenter (IV-D-26) noted that the court specifically stated that
it did not have the authority to grant the plaintiffs' request that EPA be
ordered to promulgate regulations covering three specific pollutants and two
different kinds of pollutants, stating that the CAA does not compel EPA to
issue NSPS in any particular form.
Two commenters (IV-D-20, IV-D-28) stated that as the agency responsible
for implementing the regulatory program, EPA has the discretion to determine
if such a standard is necessary. The commenters maintained that if EPA
determines that promulgating such a standard is neither reasonable nor
practical, that determination would be given deference by the court. One
commenter (IV-D-28) further stated that the EPA's own data show that the
2-53
-------
proposed NOX standard imposes excessive costs for compliance on small units.
The commenters concluded by stating that EPA should not interpret the statute
and the court decision to mean that a standard is required even if it is
unreasonable.
Another commenter (IV-D-29) contended that any NOX standard would
require additional costs in permitting, testing, and compliance reporting. He
concluded that the administrative burden of this standard is unnecessary. One
commenter (IV-D-09) stated that he believes the court would be likely to
conclude that setting a standard in the face of overwhelming data to the
contrary would be "arbitrary and capricious."
Another commenter (IV-D-05) stated that including the NOX standard in
the proposed standards for small natural gas-fired steam generating units is
not required by the court order. This commenter pointed out that the proposed
standard is based upon the court order to issue NOX standards for small steam
generating units. This commenter maintained that this court order would still
be met if the NOX standards apply only to those units subject to the PM and
S02 standards.
This commenter stated that the proposed NOX standards should be met only
by those steam generating units subject to the requirements of the proposed PM
and S02 standards. The commenter explained that requiring natural gas-fired-
only units under 29 MW (100 million Btu/hr) and wood-fired-only units under
8.7 MW (30 million Btu/hr) to comply with the NOX standards would also require
their owners or operators to undergo the NSPS review process. These units
would be considered "reportable" sources that require routine inspections and
reporting.
According to this commenter, because these small natural gas-fired units
represent a very large number of the potentially regulated steam generating
units, State agencies would find it impossible to support this increased
workload with any level of confidence. Requiring routine reporting and
recordkeeping for these units would result in fewer resources being available
for important work.
The seventh commenter (IV-D-45) stated that, although they have no
specific comment on the proposed NSPS for NOX, if reevaluation of the standard
2-54
-------
following consideration of public comments results in a more stringent
standard, the commenter urged reproposal of that portion of the NSPS.
Response: Pursuant to the order of the Court in Sierra Club v. Reillv
(D.D.C. No. 84-0325), control technologies for reducing NOX emissions from
small steam generating units were examined thoroughly in developing the
proposed regulations. Low excess air, flue gas recirculation, staged
combustion, thermal de-NOx, and selective catalytic reduction were examined,
and the best information on the effectiveness, availability, and costs of
these technologies evaluated. Of these technologies, thermal de-NOx was found
to be inapplicable to small steam generating units for technical reasons.
Selective catalytic reduction is an expensive technology which is considered
unreasonable costly for small steam generating units. Although the data on
the performance of the remaining three technologies in reducing NOX emissions
from small steam generating units is limited, the cost effectiveness of these
technologies was calculated for representative units and conservatively
estimated to range from $3,300 to $33,000/Mg ($3,000 to $30,000/ton). These
costs were considered to be unreasonably high for national NOX standards for
this source category.
In discussing the NOX standard in the preamble to the proposed
regulations, it was pointed out that the specific language of the Court order
seemed to require EPA to propose a standard for nitrogen oxides. The
memorandum opinion, however, recognized the EPA's discretion in determining
which pollutants emitted by the source category ought to be regulated.
Although a NOX standard based on the available control technologies was
determined to be unreasonable for this source category, a standard was
proposed to avoid the possibility that EPA would be in technical violation of
the Court order. This standard of 430 ng/J (1.0 Ib/million Btu) heat input is
attainable by units firing coal, oil, natural gas, or mixtures of these fuels
with any other fuels, without NOX control. Public comments were invited on
the appropriateness of applying NOX control techniques to this source
category.
2-55
-------
No additional information has been presented that would alter the
original conclusion that the cost of NOX control for this source category is
unreasonably high and that the proposed NOX standard results in no
environmental benefits. Therefore, the proposed NOX standard is being
withdrawn, and small steam generating units will not be subject to an emission
limit for NOX under this regulation.
2. Comment: Two commenters (IV-D-02, IV-D-04) stated that the proposed EPA
standard for NOX is ineffective and does not satisfy the requirements of the
Clean Air Act (CAA). One commenter (IV-D-02) stated that the proposed NOX
standards could be met too easily and do not meet the intent of the CAA. He
further stated that many small natural gas-fired steam generating units can be
designed to meet a 22 ng/J (0.05 To/million Btu) emission limit or less and
that an NSPS standard of 430 ng/J (1.0 Ib/million Btu) is not stringent
enough. The commenter argued that a reasonable NOX standard and a
verification procedure should be adopted, especially considering the role that
NOX plays in ozone formation and in acid rain. This commenter concluded by
recommending that all natural gas-fired steam generating units with heat input
capacities below 29 MW (100 million Btu/hr) should be required to meet a NOX
standard of 22 ng/J (0.05 ID/million Btu) and the larger units in this
category (12 MW [40 million Btu/hr] and greater) should be required to meet a
NOX standard of 17 ng/J (0.04 Ib/million Btu).
The other commenter (IV-D-04) stated that EPA should not propose such a
lenient NOX standard given the expectations for best available control
technology (BACT) under the Prevention of Significant Deterioration (PSD)
requirement. This commenter observed that the proposed 430 ng/J
(1.0 Ib/million Btu) NOX limit would not even require the installation of
conventional burners, much less the use of "preventative" control measures
such as low NOX burners. The commenter argued that because EPA is not
requiring NOX controls for small steam generating units, permitting
authorities must redevelop a NOX standard on a case-by-case basis during the
New Source Review process. The commenter concluded by stating that it is
inappropriate for EPA to avoid responsibility for setting a meaningful NSPS
2-56
-------
for NOX emissions from small steam generating units while expecting the State
and local agencies to set a similar standard when reviewing individual
permits.
Response: The preamble to the proposed regulations recognized that
there are technologies that can be used to reduce NOX emissions from small
steam generating units. Standards based on these technologies were not
proposed, however, because the impacts of installing and operating these
controls were determined to be unreasonable when compared to the small
emission reductions that would be achieved. These impacts include control
costs which range from $3,300 to $33,000/Mg ($3,000 to $30,000/ton), as well
as the additional costs of employing a full-time, skilled operator to manage
the control system. The comments received in response to the proposed
standards provide no additional information on the feasibility, performance,
costs, or other impacts of these technologies that would show the impacts of
these technologies to be reasonable or that would warrant further analysis.
Therefore, the proposed NOX standard is being withdrawn.
The commenters are correct that some new sources may be required to
install NOX controls under other regulatory programs, such as New Source
Review. The information on technologies, performance levels, costs, and other
impacts developed in this rulemaking and contained in the background documents
can be used by local and State authorities in reviewing and permitting new
sources under these programs.
2.6 PERFORMANCE AND RELIABILITY OF DEMONSTRATED CONTROL TECHNOLOGY
2.6.1 Sulfur Dioxide
1. Comment: One commenter (IV-D-13) stated that the 90 percent S02
reduction requirement for small steam generating units ignores some of the
newer methods of accomplishing SOg removal, such as a conditioning tower and
dry injection of hydrated lime or other alkali directly into the baghouse.
The commenter pointed out that this newer method would reduce system
2-57
-------
maintenance, overall system cost, and space requirements. According to the
commenter, this method can easily remove 50 to 70 percent of the S02, and by
combining this method with the use of medium to low sulfur coal, the 520 ng/J
(1.2 Ib/million Btu) S02 limit could easily be met.
Response: As discussed in the preamble to the proposed standards,
90 percent emission reduction has been determined to be a demonstrated control
level for small coal-fired steam generating units. Available data indicate
that sodium scrubbing FGD systems, dual alkali FGD systems, lime/limestone FGD
systems, lime spray drying systems, and FBC are all capable of consistently
achieving 90 percent reduction in S02 emissions on a 30-day rolling average
basis from coal-fired units. Based on the demonstrated performance of these
control technologies and the reasonableness of costs, nonair quality health
and environmental impacts, and energy requirements of operating these
technologies, these technologies are selected as the best demonstrated control
technologies for controlling S02 emissions from small coal-fired units. Any
control method, however, that is not specifically recognized as a best
demonstrated control technology that is capable of achieving the 90 percent
S02 reduction requirement would be considered an acceptable control method for
meeting the standard.
If a control technology that cannot meet the 90 percent reduction
requirement is in the development stages and has the potential for achieving
the 90 percent reduction level in the future, then this technology would
qualify as an emerging technology and would be subject to only a 50 percent
emission reduction requirement.
2. Comment: Four commenters (IV-D-22, IV-D-24, IV-D-26, IV-D-28) objected
to scrubbers being used as the basis for the percent reduction requirements.
One commenter (IV-D-22) stated that wet or dry scrubbers would be required to
achieve the percent reduction required by the proposed standard. The
commenter pointed out that scrubbers would be both inefficient and
economically infeasible to install on his steam generating units because most
2-58
-------
of his units are used for building heating and do not operate continuously.
The commenter claimed that his units, which range from 10 to 100 percent
capacity load depending on weather conditions, would not be able to achieve
the 90 percent S02 reduction requirement. According to the commenter, varying
the unit loads causes pressure fluctuations, and control systems do not work
efficiently with fluctuating pressures. The commenter also pointed out that
fabric filters used with dry scrubbers cannot maintain a temperature above dew
point at low operating loads.
Three commenters (IV-D-24, IV-D-26, IV-D-28) claimed that no operating
test data are available to show that scrubbers can meet the 90 percent removal
requirement on a 30-day rolling average basis that includes startup, shutdown,
and malfunction. These commenters also claimed that no test data are
available to show how these scrubbers operate during severe load swings or at
maximum capacity for extended amounts of time.
Another commenter (IV-D-40) stated that, by using data from units larger
than 29 MW (100 million Btu/hr), EPA apparently recognizes that industry has
very limited, if any, installation, operational experience, demonstrated
effectiveness, or cost information on FGD systems for small steam generating
units. For these reasons, the commenters stated that this technology should
be considered inappropriate for application to these steam generating units.
Response: In the proposed rule, "annual capacity factor" is defined as
the ratio between the actual heat input to a steam generating unit from the
fuels combusted during a calendar year and the potential heat input to the
steam generating unit had it been operated for 8,760 hours during a calendar
year at the maximum design heat input capacity. Therefore, if a unit has an
annual capacity factor of 55 percent, it means that during one calendar year
the unit operated at an average of 55 percent of its designed heat input
capacity. If a steam generating unit is not operating continuously, such as
the units mentioned by the commenter which operate only during cold weather,
then the unit is probably not operating at 55 percent annual capacity. The
90 percent reduction requirement applies only to units greater than 22 MW
(75 million Btu/hr) that operate at greater than 55 percent annual capacity
factor.
2-59
-------
Concerning the appropriateness of applying FGD systems to small steam
generating units, there are sufficient data on the performance of these
control technologies applied to industrial steam generating units as discussed
previously. These controls would perform as well when applied to small steam
generating units as to industrial steam generating units. The gas stream
characteristics (such as concentration and chemical composition) of industrial
and small units are similar, and the operating efficiencies of FGD systems do
not vary appreciably with unit size. Therefore, use of data from larger units
is appropriate for assessing control technology performance on smaller units.
Steam generating units with average loads ranging from 5 to 100 percent
were included in the data to assess FGD and FBC system performance. In
addition, steam generating unit loads were varied during tests of individual
units to simulate load swings that might be experienced in some applications.
Based on this data, S02 removal efficiency was found to be insensitive to
changes in steam generating unit load over the ranges observed.
The primary concern for FGD systems operating on steam generating units
which experience load swings is a sudden increase in the S02 loading. This
can result from an increase in either the flue gas flowrate or the flue gas
S02 concentration. As discussed in the background document, "Overview of the
Regulatory Baseline, Technical Basis, and Alternative Control Levels for
Sulfur Dioxide (S02) Emission Standards for Small Steam Generating Units"
(EPA-450/3-89-12), changes in flue gas flowrate are matched by corresponding
changes in the scrubbing solution flow rate according to a set liquid-to-gas
(L/G) ratio. In a well-designed and operated system, a safety margin is
maintained in the L/G ratio to account for delays in control loop response;
thus, an increase in flue gas flowrate would be adequately handled. Also, FGD
systems which experience highly variable S02 loadings typically operate at
high alkaline reagent concentrations. This provides a buffering capacity
against large swings in solution pH caused by dramatic changes in S02
concentration. As a result, sufficient excess alkaline reagent is present to
ensure adequate S02 removal performance during load swings.
2-60
-------
3. Comment: Three commenters (IV-D-22, IV-D-26, IV-D-42) stated that small
steam generating units require coarse coal (1 to 1/4 inch in diameter) to
obtain optimum combustion and performance. According to the commenters, the
crushing process used to reduce sulfur content to the size necessary to meet a
520 ng/J (1.2 ID/million Btu) S02 standard would produce a fine-grade coal
that is not acceptable for use in small industrial steam generating units.
The commenters concluded that increasing the S02 emission limit from 520 to
690 ng/J (1.2 to 1.6 ID/million Btu) would alleviate this problem.
Response: The "coarse" coal used in small steam generating units is
typically stoker-grade coal. The commenters assume that all coal would need
to be processed to meet the 520 ng/J (1.2 ID/million Btu) emission limit, thus
reducing the size of the coal and creating combustion problems for small
units. However, a review of the availability of low sulfur stoker coal to
supply the small industrial-commercial-institutional steam generating unit
market indicates that sufficient low sulfur stoker coal supplies are available
to meet market demands following implementation of this regulation. This low
sulfur stoker coal would be in the coarse size ranges used by small steam
generating units. Although an individual unit owner/operator may need to
change coal suppliers, there is ample low sulfur stoker coal available on a
regional basis to meet the needs of small units. Owners/operators of small
units can switch to use of low sulfur stoker coal with no or, at most,
negligible impacts on national or regional coal markets because coal-fired
units below 29 MW (100 million Btu/hr) in size account for such a small
fraction of overall coal demand. Thus, suppliers of low sulfur stoker coal
suitable for use in small coal-fired units are expected to have no difficulty
meeting the needs of owner/operators in the future.
4. Comment: Two commenters (IV-D-40, IV-D-45) stated that for units with
FGD systems, the EPA's proposed SC>2 standard requires the use of either a
continuous emission monitoring system (CEMS) or daily fuel sampling, in the
absence of demonstrated technological or cost data. One commenter (IV-D-40)
continued that installation of CEMS on small units is not cost-effective due
2-61
-------
to the high cost and complexity of monitoring equipment. This commenter
recommended that the $63 emissions compliance methods should be reconsidered
and revised.
Response: As discussed in the preamble to the proposed standards, the
available information on the application of either CEMS or daily fuel sampling
to units with FGD systems was analyzed. The CEMS and daily fuel sampling were
found to be reasonable methods of monitoring considering that monitoring is
necessary for enforcement personnel to determine that the percent reduction
requirement is being achieved by the control device. Furthermore, the
annualized costs above baseline for emissions monitoring using either CEMS or
daily fuel sampling were considered reasonable in both cases.
5. Comment: One commenter (IV-D-45) stated that the S02 percent reduction
requirement should not be applied to small steam generating units because the
data on reliability of FGD in these units are inadequate. The commenter
stated that further investigation is required before a percent reduction
requirement can be imposed.
Response: The reliability of FGD systems was discussed in the
background document, "Overview of the Regulatory Baseline, Technical Basis,
and Alternative Control Levels for Sulfur Dioxide (S02) Emission Standards for
Small Steam Generating Units," (EPA-450/3-89-12). The reliability of sodium
scrubbing FGD systems on coal-fired units is well-documented in available
emissions test data. The performance test data include 30 days of certified
CEMS test data and show removal efficiencies averaging 96.2 percent for the
test period. Although the sodium scrubber in this 30-day test was applied to
a unit rated above 29 MW (100 million Btu/hr) heat input, the performance data
from this scrubber are applicable to small steam generating units because the
gas stream characteristics (e.g., concentration, chemical composition) are
similar and FGD design and operating characteristics do not vary significantly
with size in this general size range.
2-62
-------
For lime spray drying systems, reliability levels ranging from 70 to
97 percent were reported for various test sites. Some decrease in reliability
was reported with increasing SOg removal efficiency. However, an examination
of the reasons for decreased reliability indicated that the FGD system
failures were not generally the result of increased system stress, but were
due to other factors unrelated to S02 removal efficiency and could often have
been prevented with improved operating and maintenance procedures or
maintaining a spare parts inventory. With proper operation and maintenance,
high FGD reliabilities can be achieved and maintained on steam generating
units operating at high S02 removal levels. For compliance during periods of
FGD malfunction, the cost of firing alternative low sulfur fuels, such as
natural gas, in the steam generating unit was included in the cost
calculations. As an alternative approach, installation of backup FGD modules
to ensure system reliability, may be used on small steam generating units.
6. Comment: One commenter (IV-D-45) stated that EPA should explain the
rationale for the relaxed S02 removal requirement for steam generating units
using fluidized bed combustion (FBC) and firing coal refuse alone. The
commenter argued that if the purpose of this partial exemption is to encourage
coal refuse recovery, then the same consideration should be given to all coal
technologies using coal refuse or coal refuse-derived fuel. If, on the other
hand, the FBC technology cannot meet the 90 percent reduction requirement
while firing coal refuse, then a technical basis for the 80 percent
requirement should be provided. The commenter noted that none of the test
data supporting FBC as demonstrated technology for 90 S02 percent removal were
derived from units firing coal refuse.
Response: The purpose of the more relaxed $62 percent removal
requirement for FBC units firing coal refuse is to encourage the use of coal
refuse. Many abandoned coal refuse piles remaining from coal mining
operations, particularly in areas of the United States such as northeastern
Pennsylvania, contribute to contamination of surface and ground waters through
acid water run-off, present a fire hazard in terms of spontaneous combustion,
2-63
-------
and are an unsightly blight on the environment in terms of aesthetics. Thus,
there are a number of benefits to be gained by eliminating these abandoned
coal refuse piles.
Fluidized bed combustion units are the only type of steam generating
unit which is capable of firing coal refuse as a fuel. The Btu content of
coal refuse is too low to serve as a viable fuel in conventional stoker or
pulverized coal-fired steam generating units. As a result, all of the steam
generating units currently in operation or in the various stages of planning
that fire coal refuse are FBC units.
The FBC units firing coal refuse are capable of achieving a 90 percent
reduction in S02 emissions. The vendors of these units, however, indicate
that achieving this level of S02 control in FBC units firing coal refuse
requires such large quantities of lime that such units could no longer compete
economically with FBC units firing coal or conventional stoker coal or
pulverized coal-firing steam generating units. Only by reducing the SC>2
percent reduction requirement to 80 percent could FBC units firing coal refuse
remain economically competitive.
As mentioned above, the benefits to be gained by eliminating abandoned
coal refuse piles are substantial and these benefits outweigh the loss in air
pollution benefits associated with reducing the $62 percent reduction
requirement for FBC units firing coal refuse from 90 percent to 80 percent.
Therefore, to encourage the use of coal refuse, the standards provide this
more relaxed $62 percent reduction requirement for FBC units firing coal
refuse. The question of whether to provide a relaxed percent reduction
requirement for other types of steam generating units firing coal refuse has
already been addressed.
2.6.2 Particulate Matter
1. Comment: One commenter (IV-D-07) objected to the selection of wet
scrubbers and ESP's as BDT for wood-fired steam generating units with greater
than 8.7 MW (30 million Btu/hr) heat input. He stated that the use of these
technologies does not ensure that the proposed emission limits would be met.
2-64
-------
He also stated that attempts at his small wood-fired cogeneration plant to
develop a PM compliance plan indicated that his wet scrubber would require
modification or replacement to meet the 43 ng/J (0.10 Ib/million Btu) PM
limit. These modifications would result in an increased overall plant
horsepower requirement of 25 percent.
The commenter noted that although wet scrubbers have a relatively
inexpensive capital cost, operating costs can be high. According to the
commenter, small steam generating units in the 8.7 MW to 29 MW (30 to
100 million Btu/hr) range affected by these proposed regulations have fewer
pounds of steam over which to amortize costs; however, pollution control
capital costs would be similar to those for large industrial units. The
commenter explained that the electrified gravel bed filter is compact, only
slightly more capital intensive than a wet scrubber, has low operating costs,
and can obtain removal efficiencies comparable to the ESP.
The commenter also stated that, if not operated properly, wet scrubbers
can lead to emission excursions due to gas-side fouling of collection and wet-
dry interface surfaces. High gas-side pressure drops also lead to the use of
undersized induced draft fans. This in turn increases the chances for poor
combustion due to inadequate excess air and operator inclination to reduce
scrubber flow rates or to turn off scrubber pumps to achieve production goals.
The commenter recommended that the wet scrubber be eliminated from the list of
BDT and replaced with the electrified gravel bed filter (EGF), also known as
the electrified filter bed or electrostatic gravel bed filter.
Response: As described in the background document, "Overview of the
Regulatory Baseline, Technical Basis, and Alternative Control Levels for
Particulate Matter (PM) Emission Standards for Small Steam Generating Units"
(EPA-450/3-89-11), available data indicate that wet scrubbers operating at
medium pressure drops (3.8 to 6.0 kilopascals (kPa) [15 to 26 inches water]}
and preceded by a mechanical collector can consistently achieve PM emissions
of 43 ng/J (0.10 ID/million Btu) or less. The operating costs involved were
considered and found to be reasonable. However, an owner/operator is not
constrained to use a wet scrubber for PM emissions control; any technology,
including the EGF, that achieves compliance with the standard is acceptable.
2-65
-------
2. Comment: Two commenters (IV-D-07, IV-D-14) stated that the electrified
filter bed (also referred to as an EGF, as discussed above) should be included
in the EPA's description of available control technologies for PM emissions
from small steam generating units. One commenter (IV-D-14) noted three
advantages of the EGF over a wet scrubber or an ESP: (1) the EGF does not
generate effluent water or wet sludge; (2) the EGF does not require high
pressure drops, thereby offering greater operating efficiency; and (3) the EGF
is not sensitive to high levels of oxygen that are often present in small
wood-fired steam generating units. The commenter included as attachments a
list of 89 plants using EGF's; two emissions test reports for on-line EGF's
showing that the EGF can reduce emissions; a Staff Activity Report from the
Michigan Department of Natural Resources citing EGF's as BACT for PM emission
control in certain types of wood- and oil-fired chip dryers; and a journal
article presenting three case studies of EGF's used to control PM emissions
from bark-and wood-fired steam generating units.
Response: Owners/operators of affected facilities are not constrained
to use ESP's or wet scrubbers to achieve compliance; EGF's can also be used.
3. Comment: One commenter (IV-D-13) noted that hopper evacuation or
sidestream separator systems are attractive for steam generating units smaller
than 14.6 MW (50 million Btu/hr), and that they achieve emission rates of
86 ng/J (0.20 Ib/million Btu) or less for coal-fired units and 0.15 Ib/million
Btu for chain grate stokers firing coal.
Response: Sidestream separators were considered as a control option for
PM emissions, but test data indicate that this technology cannot achieve the
PM emission levels of fabric filters or ESP's. Test data were taken from
coal-fired units ranging in size from 8.7 to 29 MW (30 to 100 million Btu/hr)
equipped with sidesteam separators; PM emissions ranged from 52 to 71 ng/J
(0.120 to 0.165 Ib/million Btu). Therefore, sidestream separators are
demonstrated to be capable of reducing PM emissions to 86 ng/J
(0.20 Ib/million Btu) or less. Fabric filters and ESP's, on the other hand,
2-66
-------
are demonstrated at 22 ng/J (0.05 Ib/million Btu) heat input. Since the costs
and other impacts associated with a 22 ng/J (0.05 Ib/million Btu) standard
based on the use of fabric filters and ESP's were found to be reasonable, the
PM standard for small coal-fired units is set at 22 ng/J (0.05 Ib/million
Btu).
4. Comment; One commenter (IV-D-13) stated that steam generating units
firing hardwood cannot easily meet the proposed PM standard of 43 ng/J
(0.10 Ib/million Btu) using the demonstrated control technology of wet
scrubbers. The commenter stated that much higher pressure drops or other
special auxiliaries would be needed in order to meet the PM standard.
Response: As described in the background document "Overview of the
Regulatory Baseline, Technical Basis, and Alternative Control Levels for
Particulate Matter (PM) Emission Standards for Small Steam Generating Units"
(EPA-450/3-89-11), test data from 16 units in the affected facility size range
show that wood-fired units equipped with wet scrubbers preceded by a
mechanical collector are capable of achieving PM emission limits of 43 ng/J
(0.10 Ib/million Btu) or less at medium pressure drops 3.8 to 6.0 kPa (15 to
26 inches water). The ESP's and EGF's are also demonstrated at 43 ng/J
(0.10 Ib/million Btu) and can be used to achieve the PM standard. In some
special cases, higher pressure drops might be required in order to achieve the
proposed standard. However, the test data demonstrated that, even with higher
pressure drops, the proposed standard is achievable at a reasonable cost.
5. Comment: One commenter (IV-D-33) stated that EPA has chosen stringent
control technology for PM, based on a concern over PMjo emissions from less
efficient devices, but presents no particle size data to indicate that
different control devices are more or less effective on PMjo- The commenter
recommended that EPA examine particle size data to determine if there are, in
fact, significant differences among alterative control technologies.
2-67
-------
Response: The PM standard is not, as the commenter impli.es, based
primarily on effective control of PMiQ. The control of PMjo was, however,
considered when evaluating additional impacts of the proposed standards. As
stated in the preamble, the PM limits are based on the best demonstrated
control technology considering costs and other impacts. Fabric filters and
ESP's are the most effective PM control technologies available for coal-fired
steam generating units, and are capable of achieving emission levels of
22 ng/J (0.05 To/million Btu) or less. It was also noted in the preamble
that, in addition to providing effective control of total PM emissions, these
technologies are very effective in controlling emissions of PMio- As stated
in the preamble to the proposed regulations, data from AP-42 indicate that the
PMjo removal efficiency of mechanical collection systems on coal-fired units
is less than 40 percent. This is compared to a greater than 99 percent PMjo
removal efficiency for fabric filters and ESP's.
6. Comment: One commenter (IV-D-17) stated that the predictions in the
background document of performance characteristics for various control devices
on small steam generating units are not valid because they are based on
arbitrary capacity figures, data from large units, and data from units that
are probably running on green chips. The commenter further stated that PM
emissions vary greatly depending on the feedstock and the size of the wood-
fired steam generating unit. The commenter also stated that the PM standard
would eliminate emissions of PM which will mostly be larger than 10 microns
and not respirable by humans. The commenter recommended the adoption of
Control Option A, an emission limit of 130 ng/J (0.30 Ib/million Btu).
Response: The data on which the PM standard is based are summarized in
the background document "Overview of the Regulatory Baseline, Technical Basis,
and Alternative Control Levels for Particulate Matter (PM) Emission Standards
for Small Steam Generating Units" (EPA-450/3-89-11). These data were
generated by tests performed over a wide range of actual operating conditions.
The tests included units of varying sizes within the range from 8.7 to 29 MW
(30 to 100 million Btu/hr). The capacity factors used in the tests were not
2-68
-------
arbitrary; they reflect actual operating conditions representative of wood-
fired units in the affected size range. In order to consider emissions
variation with size and design, the regulatory baseline included two levels
depending on the unit size: one baseline of 194 ng/J (0.45 Ib/million Btu) for
units less than 8.7 MW (30 million Btu/hr), and one of 260 ng/J
(0.60 Ib/million Btu) for units greater than or equal to 8.7 MW (30 million
Btu/hr).
The PM emissions that the standard will eliminate include respirable
PM (i.e., PMio)- As explained in the proposal preamble, ESP's and wet
scrubbers are considerably more effective than DMC's (Control Option A) for
PMjo control and have been selected as BDT for typical wood-fired units in the
affected size range. The final standard does, however, include an emission
limit of 130 ng/J (0.30 Ib/million Btu) based on the use of DMC's, for wood-
fired units subject to a Federally enforceable requirement limiting the annual
capacity factor to 30 percent or less. This change has been made only because
of the unreasonable cost effectiveness which results from the application of
the proposed standard to low capacity factor units.
7. Comment: One commenter (IV-D-13) objected to the conclusion reached by
EPA that the sidestream separator is an "inferior" control option for small
coal-fired steam generating units because of excessive costs. The commenter
disagreed with the cost analysis presented in the preamble to the proposed
standards. The commenter claimed that in that analysis, the cost of a
baghouse handling about 15 to 20 percent of the gas volume was assumed to be
the same as the cost of a baghouse handling 100 percent of the gas volume.
According to the commenter, the actual cost of a baghouse associated with a
sidestream separator is only about 30 to 50 percent of the cost of a baghouse
handling 100 percent of the gas volume.
Response: In considering PM control options, the costs for both fabric
filters and sidestream separators were analyzed. Analysis of the sidestream
separator's costs and design did not include a fabric filter sized for
100 percent capacity; the costs are based on actual sidestream separator
2-69
-------
design. Sidestream separators were not deemed inferior due to excessive
costs; rather, fabric filters are considered superior because they give
greater reductions (demonstrated to achieve 22 ng/J [0.05 1 fa/mill ion Btu])
than sidestream separators (demonstrated to achieve 86 ng/J [0.20 Ib/million
Btu]), and the cost effectiveness of emission control for fabric filters is
more attractive than that for sidestream separators.
8. Comment: One commenter (IV-D-45) stated that the proposed PM standard
for units firing coal is based on fabric filters and ESP's. However, the
commenter continued, smaller units with overfeed and underfeed stoker designs
typically emit high-carbon content flyash particles that create an increased
fire hazard in fabric filters. The commenter stated that EPA had not
considered this point when transferring the application of this technology to
units smaller than those tested. The commenter stated that the high carbon
content of flyash particles from small overfeed and underfeed stokers is also
known to be detrimental to ESP devices. The commenter stated that this
performance effect was not considered by EPA.
Response: Emissions test data were gathered on five spreader stoker
coal-fired units and two FBC coal-fired units equipped with fabric filters.
Emissions test data were also gathered on six spreader stoker coal-fired units
equipped with cold-side and hot-side ESP's. During the collection and
evaluation of these test data, none of the owners/operators of these units
indicated any undue concern about these PM control systems presenting an
unacceptable fire hazard, nor have vendors of such systems indicated any
similar concern. In addition, the BACT/LAER Clearinghouse lists fourteen
States that require fabric filters for permits for coal-fired units between
14.6 and 29 MW (50 and 100 million Btu/hr) heat input, and fifteen States with
such requirements for units below 14.6 MW (50 million Btu/hr) heat input.
Therefore, the proposed standard has been demonstrated to be achievable, and
the costs of control (including fire precautions for fabric filters or use of
an adequately sized ESP) are considered reasonable.
2-70
-------
2.6.3 Nitrogen Oxides
1. Comment: One commenter (IV-D-09) stated that EPA was incorrect in
concluding that all units in the regulated community would be able to meet the
NOX limit of 430 ng/J (1.0 Ib/million Btu). He further stated that EPA had
not evaluated the impact of the proposed NOX standard on the units that could
not meet the limit and, therefore, EPA had not followed the proper decision
scheme in setting the standard. The commenter concluded that the NOX standard
should not be promulgated.
The commenter said that his company operated a number of Thermal Heat
Recovery Oxidation (THROX) units designed to destroy "hard-to-destroy" wastes
and to generate steam. According to the commenter, some of these units, which
would be subject to the provisions of Section 60.49c based on the use of
natural gas as a supplemental fuel, would not be able to meet the proposed
standard. The units that cannot meet the limit either burn high nitrogen-
containing waste or combust waste at unusually high temperatures. In order to
meet the limits, the high-temperature THROX units would have to operate at
reduced temperatures and, therefore, at reduced efficiencies. The units
burning high nitrogen-containing wastes would have to have installed the
technology that EPA has determined is not cost effective or operate the THROX
units below their design capacity for burning the waste.
Response: The NOX standards were proposed to comply with the Court
order handed down in Sierra Club v. Reillv (D.D.C. No. 84-0325). Compliance
with the proposed 430 ng/J (1.0 Ib/million Btu) heat input NOX emission
limits, however, would result in insignificant emission reductions.
Compliance with more stringent NOX standards would require the application of
unreasonably expensive technological controls. Therefore, these proposed
standards are being withdrawn and no NOX standards are being adopted for this
source category.
2-71
-------
2.7 COST AND ECONOMIC IMPACTS
2.7.1 Auxiliary Steam Generating Units
1. Comment; One commenter (IV-D-34) was concerned about the effect that
the proposed standards would have on auxiliary steam generating units located
at electric utility plants. The commenter stated that, without an exemption,
auxiliary units firing medium to high sulfur fuel oil would not be able to
comply with the proposed S02 standards without either installing expensive
emissions control devices or switching to lower sulfur fuels, which would
require the installation of a separate fuel handling system.
The commenter recommended that an exemption be given for affected
facilities that have an annual capacity factor for oil of 30 percent or less,
similar to the exemption granted in the Subpart Db standard for S02-
Alternatively, the commenter recommended that a specific exemption be granted
from the proposed standards for auxiliary units capable of burning oil and
located at electric utility steam generating units. The commenter stated that
no health or environmental benefits are gained by requiring auxiliary units to
comply with stricter emission standards than main units located at the same
plant. The commenter pointed out that incremental emissions from the small
units are only a very small fraction of the total emissions from the site.
Response: In the situation described by the commenter, the
owner/operator has the option of either installing a separate fuel handling
system to provide very low sulfur oil to the new unit, or firing very low
sulfur oil in all units supplied by the existing fuel handling system. The
first option is no different than the situation at an entirely new plant with
a new steam generating unit. In both situations, a new fuel handling system
must be installed to provide the new unit with an appropriate fuel supply, and
the costs of installing this new system are considered reasonable. The second
option would avoid the need to install a new fuel handling system, and would
achieve additional reductions in S02 emissions as well. Therefore, basing the
S02 standard on use of very low sulfur fuel is considered reasonable for small
2-72
-------
oil-fired units, including auxiliary and other low capacity factor units that
might otherwise be fired with medium to high sulfur fuel.
It should be noted that the exemption for auxiliary oil-fired units in
Subpart Db pertains to the percent reduction requirement only. These units
are still required to meet the emission limit. In the S02 standard for small
steam generating units, all oil-fired units are exempt from the percent
reduction requirements.
2. Comment: One commenter (IV-D-21) stated that the proposed $03 standard
for distillate oil-fired steam generating units would impose new or expanded
testing, monitoring, reporting, and recordkeeping requirements for auxiliary
units located at electric utility plants, which would result in a substantial
cost to customers without any additional air quality benefits.
Response: Distillate oil-fired units are required only to maintain
records of fuels fired (a practice which most unit owners or operators already
follow), and to submit quarterly reports showing that distillate oil was
fired. When receiving a shipment of distillate oil, the unit owner or
operator also receives information concerning the shipment (such as percent
sulfur content, etc.). These records are kept on file for accounting
purposes. The standard only requires that the owner or operator retains these
records and submits quarterly reports to show that the shipment is certified
to comply with ASTM specifications, which are minimal requirements.
3. Comment: Two commenters (IV-D-20, IV-D-21) stated that they objected to
the EPA's proposed opacity monitoring requirements for auxiliary oil-fired
units. Although they agree with the EPA's decision not to set a PM limit for
units firing very low sulfur oil, one commenter (IV-D-20) questioned whether
the cost of a continuous opacity monitor is justified for utility auxiliary
steam generating units combusting very low sulfur oil and operating at low
capacity factors. The commenter added that opacity is a problem only if a
unit is operated and maintained poorly. Since poor operation and maintenance
2-73
-------
waste expensive fuel, there is an incentive even without continuous opacity
monitoring to maintain and operate the units properly. The commenter
concluded by requesting that EPA reconsider whether continuous opacity
monitors are needed for low capacity factor units burning very low sulfur oil.
Response: As stated in the proposal preamble, the intent of the opacity
limit on small oil-fired steam generating units is to allow identification of,
and appropriate enforcement action to be taken against, units operating with
incomplete combustion. However, distillate fuel oil is a very clean-burning
fuel compared to residual fuel oil. Steam generating units firing distillate
oil require little maintenance, so there is little chance for incomplete
combustion to occur in such a unit.
Opacity from a distillate oil-fired unit will be very low unless, as
pointed out by the commenters, the unit is improperly operated. In view of
this, violations of the opacity standard are unlikely to occur and the costs
of requiring all small distillate oil-fired steam generating units in this
source category to install a CEMS have been determined to be unreasonable.
Therefore, the standards are being amended to exempt from the opacity
CEMS requirement distillate oil-fired steam generating units with .heat input
capacities between 8.7 and 29 MW (30 and 100 million Btu/hr), including
auxiliary units located at electric utilities. However, the opacity standard
for all oil-fired units is being retained so that enforcement personnel who
become aware that a problem with opacity exists at a certain oil-fired unit
will have the authority to intervene and see that the problem is corrected.
Unlike distillate oil-fired units, residual oil-fired steam generating
units are required to have continuous opacity monitors. Residual oil is a
dirtier fuel requiring more maintenance of the burner to ensure proper
combustion. There is a greater chance for incomplete combustion in a residual
oil-fired steam generating unit. Therefore, the opacity limit of 20 percent
is retained for all oil-fired units for the reasons discussed in the proposal
preamble, but the requirement of an opacity CEMS is only required for residual
oil-fired steam generating units.
2-74
-------
4. Comment: One commeriter (IV-D-20) examined the cost impact of NOX
control and monitoring on auxiliary steam generating units and found that it
exceeds 26,500/Mg ($24,000/ton). The commenter provided a test report to
support these calculations. The commenter stated that these costs were
unreasonable.
Response; As stated previously, the proposed NOX standards are being
withdrawn and no NOX standards are being adopted for this source category.
2.7.2 Fuel Cost
1. Comment: One commenter (IV-D-16) disagreed with the current fuel prices
quoted for medium sulfur residual oil and distillate oil in the background
document entitled "Model Boiler Cost Analysis for Controlling Sulfur Dioxide
($02) Emissions from Small Steam Generating Units" (EPA-450/3-89-14). The
commenter estimated an average cost differential between residual and
distillate oil of 29 cents/gallon, meaning an estimated increase in annual
fuel cost of $500,000 rather than $231,000 as indicated in the background
document. The commenter stated that this estimated fuel cost increase would
have a significant impact on the operating costs of the rendering plant
industry. The commenter pointed out that, because rendering plants must
compete on the commodity market, these increased costs cannot be passed on to
customers, meaning that this adverse economic effect may force some rendering
plants to close. To avoid such adverse industry impacts, the commenter
recommended setting the S02 emission limit for small steam generating units at
690 ng/J (1.6 Ib/million Btu).
Response: The fuel costs incorporated in the economic analysis reflect
regional-average price projections that are amortized over a 15-year life
cycle. Current fuel prices may differ from the prices analyzed for a number
of reasons (e.g. current versus projected prices, or site-specific factors).
Any change in impacts associated with these factors are, however, considered
to be insignificant to the rendering industry. Only the fraction of the
2-75
-------
industry installing new small steam generating units will experience higher
fuel costs, and these costs can be mitigated to some degree by the flexibility
the affected facility has to choose between fuels. The industry currently
uses dual-fired steam generating units to allow fuel flexibility. New units
that are subject to this standard will have the flexibility to switch fuels in
response to market prices, choosing between very low sulfur residual oil,
distillate oil, and natural gas.
2.7.3 Low Capacity Units
1. Comment; Three commenters (IV-D-01, IV-D-10, IV-D-24) questioned
whether the S02 limits should be required when oil is used intermittently as a
backup fuel to natural gas.
One commenter (IV-D-10) stated that due to the fluctuating availability
of various fuels, the proposed standard should provide allowances for fuel
flexibility. According to the commenter, natural gas is the preferred fuel
type in his company because it is generally the most cost-effective and poses
minimal environmental impacts; however, natural gas cannot be the sole fuel
source for a steam generating unit because major users are supplied natural
gas on an "interruptible basis," meaning that large manufacturing facilities
are often cut off by natural gas suppliers during periods of heavy natural gas
use by small users. In these situations, large facilities are forced to
switch over to an alternate fuel source to maintain operations.
The commenter pointed out that the older units tend to use residual oil
because of the cost, availability, and environmentally safe storage in
underground tanks (the viscous residual oil poses less threat to groundwater
contamination than does distillate oil). New units will tend to use residual
oil as the emergency backup fuel because it is more efficient to install
residual oil capability when residual oil is already in use in other units at
a facility than to have the old units firing residual oil and the new units
firing distillate oil. Use of one fuel as the backup fuel also provides for
more uniform operations company-wide. The commenter maintained that the
amount of residual oil fired as a backup fuel in these natural gas-fired units
2-76
-------
is minimal and the installation of control devices would be costly. In
conclusion, the commenter recommended that the proposed standards include an
exemption to the emission requirements for residual oil when firing it as an
alternate fuel to natural gas during supplier or transporter-imposed
interruptions.
One commenter (IV-D-24) stated that in the northeastern part of the
country, natural gas is not always available in sufficient quantities to meet
the fuel demand. Distribution systems frequently require interruptible
service to ensure adequate supplies of gas for residential demand during
severe weather. The commenter stated that when gas is unavailable,
commercial-industrial-institutional steam generating units require backup oil.
The commenter asserted that it is unrealistic to require backup oil to comply
with S02 requirements.
Response: The cost of the SOg standard was carefully examined for units
with a range of capacity factors, including units where oil is used primarily
as a backup fuel. An analysis of the cost of the S02 standard for oil-fired
units shows that the cost effectiveness of S02 control is not significantly
influenced by the capacity factor of the unit. For a standard based on use of
very low sulfur oil, there are no significant capital costs associated with
compliance that must be "spread over" the hours that the unit operates. The
incremental cost/ton of pollutant removed is almost entirely a function of the
cost differential between very low sulfur and medium to high sulfur oils.
This cost differential remains essentially constant regardless of the amount
of oil used.
For example, the incremental cost effectiveness of the standard for a
14.6 MW (50 million Btu/hr) unit operating at a capacity factor for oil of
55 percent is about $l,500/Mg ($l,400/ton). The incremental cost
effectiveness of the standard for the same size unit operating at a capacity
factor for oil of 26 percent is also about $l,500/Mg ($l,400/ton). In view of
this, the cost effectiveness of applying the standard to units operating at
low capacity factors would be equally reasonable for both high and low
capacity oil-fired units. Therefore, if a small steam generating unit
owner/operator chooses to fire oil as a backup fuel, the owner/operator is
2-77
-------
required to comply with the S02 standard of 215 ng/J (0.50 Ib/million Btu)
during those periods.
In the situation described by the commenter, the owner/operator has the
option of either using medium to high sulfur residual oil as the backup fuel
for the "old" units while using very low sulfur oil as the backup fuel for the
new unit, or using very low sulfur oil as the backup fuel supply for all
units. The second option would avoid the need to install a new fuel storage
and handling system since the existing system could support all units. At
most, some additional fuel storage capacity may be needed. This option would
also achieve additional reductions in S02 emissions.
If the owner/operator chooses to continue to use medium to high sulfur
residual oil as the backup fuel for the "old" units, installation of a new
fuel storage and supply system may be required to support the new unit.
However, this is no different than the situation at an entirely new plant with
a new steam generating unit. In both situations, a new fuel storage and
supply system must be installed to provide the new unit with an appropriate
backup fuel supply, and the costs of installing this new system are considered
reasonable in both cases.
The duration and frequency of natural gas supply interruptions each year
are not significant. In view of this, it is reasonable to require compliance
with the SOg requirement when stored backup fuel is used during these
infrequent periods of natural gas unavailability. The additional cost of
compliance for units firing distillate oil as a backup fuel is minimal because
only the maintenance and reporting of fuel supply records is required. If the
unit fires residual oil as the backup fuel, then the S02 limit must be
achieved and demonstrated either by CEMS or shipment fuel sampling. Residual
oil that is fired as a backup fuel in this situation would be used
infrequently and, therefore, shipment sampling of backup residual oil would be
infrequent, resulting in minimal additional cost.
2. Comment: One commenter (IV-D-31) stated that intermittently operated
coal-fired steam generating units should not be required to install fabric
filters. He said that it is unlikely that resources would be available at
2-78
-------
these facilities to maintain such equipment in an acceptable fashion. He
recommended that an emission limit of 130 ng/J (0.30 To/million Btu) would be
more realistic for these facilities.
Response: Fabric filters are not the only technology capable of
achieving the 22 ng/J (0.05 Ib/million Btu) emission limit at small coal-
fired steam generating units. If the owner or operator of the unit believes
fabric filters may require too much maintenance, ESP's may also be used to
achieve this emission level. Fabric filters and ESP's are the two most
effective PM emission control technologies available and the technologies are
widely available for all sizes and types of steam generating units in this
source category. The cost impacts of a 22 ng/J (0.05 Ib/million Btu) emission
limit on small coal-fired steam generating units operating at various capacity
factors were evaluated. Maintenance costs were included in the analysis.
Although the cost effectiveness of emission control is relatively high (i.e.,
in the range of $2,800 to 7,300/Mg [$2,500 to $6,600/ton]), these impacts are
considered reasonable for units above 8.7 MW (30 million Btu/hr) heat input
capacity in light of the fact that many small coal-fired units are located in
urban areas with short stacks, are significant emitters of PMjo> and emit
trace metals and other compounds considered toxic to humans.
3. Comment: One commenter (IV-D-47) requested a special exemption for
small steam generating units of 14.6 MW (50 million Btu/hr) or less that are
used exclusively for combustion research. The commenter stated that the units
would be used on a temporary or intermittent basis only, and that the
resulting cost-effectiveness of emission control would be unreasonable.
Response: Extensive analyses of the impacts of the proposed standards
on small steam generating units in the 2.9 to 14.6 MW (10 to 50 million
Btu/hr) heat input size categories, and across a range of operating capacity
factors, were performed prior to proposal. The size cutoffs in the
promulgated standards are supported by these analyses.
2-79
-------
The promulgated standards are least stringent toward the smallest units
regulated. Small steam generating units between 2.9 and 8.7 MW (10 and
30 million Btu/hr) firing coal are only subject to an SQz emission limit of
520 ng/J (1.2 Ib/million Btu), while oil-fired units in this size range are
subject an S02 emission limit of 215 ng/J (0.5 ID/million Btu). These units
may easily achieve these emission limits using low-sulfur coal, distillate
oil, very low sulfur residual oil, or natural gas. Additionally, the
promulgated standards will allow the use of fuel supplier certification in
lieu of daily fuel sampling and analysis for distillate oil-fired units and
for coal and residual oil-fired units between 2.9 and 8.7 MW (10 and
30 million Btu/hr) heat input.
For small steam generating units between 8.7 and 14.6 MW (30 and
50 million Btu/hr) (the top of the range the commenter spoke about), the
additional requirements under the standard are minimal. A 20 percent opacity
limit and continuous opacity monitoring are required for coal and residual
oil-fired units; a 22 ng/J (0.05 Ib/million Btu) PM emission limit is required
for coal-fired units; and a 43 ng/J (0.10 Ib/million Btu) PM emission limit is
required for wood-fired units. Additionally, in the promulgated standard,
wood-fired units operating at less than 30 percent annual capacity factor are
subject to a less-stringent PM standard of 130 ng/J (0.30 Ib/million Btu).
Overall, these "smaller" units are subject to very little regulation.
The analyses indicate that the emissions, potential emission reductions, costs
and impacts of the promulgated standards are reasonable, whether a small steam
generating unit is being used for research purposes only or not.
Consequently, because research units have not been revealed by these analyses
to differ in cost, emission reductions, or impacts from other steam generating
units in this size range which operate infrequently, no exemption is being
included in the final standard.
2.7.4 Alaskan Coal
1. Comment: One commenter (IV-D-08) stated that when analyzing the costs
of the mandatory S02 scrubbing technology, EPA did not consider the additional
2-80
-------
energy cost incurred by units operating in Alaska where it would be necessary
to protect a wet scrubbing system from freezing for 7 months of the year. The
commenter also stated that additional waste disposal costs associated with the
discharge of the slurry or lime and lime sulfate sludge would be significant,
especially the costs of managing a disposal site to contain these wastes in an
Arctic environment.
Response: The CAA requires scrubbing where the impacts are considered
reasonable. Scrubbers can be designed to operate in different types of
climatic conditions. In freezing climates, insulation could be built into the
scrubber when it is installed, or a dry scrubbing technology could be used
that does not freeze under extremely cold conditions. The additional waste
disposal costs associated with the discharge of the slurry or lime and lime
sulfate sludge would not be a significant burden, even in an Artie
environment.
Steam generating units generally do not operate as independent entities,
but are most often part of an industrial, commercial, or institutional
establishment that produces other types of wastes requiring disposal. In
addition, coal-fired steam generating units generate ash, which also requires
disposal. These other operations impose an additional energy cost on the
facility owner/operator in order to protect them from freezing. In this
sense, the scrubbing system is no different than these other operations in
that each of them may need to be designed and operated to withstand the
extremely cold conditions of Alaskan winters.
In most cases, disposal of wastes generated by FGD systems presents no
more of a problem than disposal of other wastes generated at the unit site, or
steam generating unit ash. In certain locations where landfill capacity may
be limited or restricted, disposal of all wastes generated at the unit site is
likely to present as many problems -- and in some cases more problems, given
the nature of these wastes --as disposal of wastes from FGD systems. For the
unit, such constraints may necessitate substantial changes to the operation
process in order to minimize the wastes generated or to alter their
characteristics. For the small steam generating unit and $03 control system
operating in an Arctic environment, this may necessitate selection of a dry
2-81
-------
system over a wet system or selection of an alternative fuel, such as natural
gas, with little or no waste disposal requirements.
2.7.5 Lignite Industry
1. Comment: One commenter (IV-D-41) stated that the proposed standards are
more stringent than those required by Section 111 of the CAA. The commenter
further stated that in North Dakota, the proposed standards would prevent any
new, modified, or reconstructed small steam generating unit from using
North Dakota lignite because the economics of the standard would force unit
owners and operators to switch to an alternate fuel. The commenter pointed
out that North Dakota is one of the few States to have attained all of the
NAAQS. The commenter concluded that in a rural State such as North Dakota,
which is struggling economically, the proposed standards would provide very
limited environmental benefits at an extremely high cost for small units.
Response: As stated in Section lll(a)(l)(C) of the CAA, "a standard of
performance shall reflect the degree of emission limitation and the percentage
reduction achievable through application of the best technological system of
continuous emission reduction which (taking into consideration the cost of
achieving such emission reduction, any nonair quality health and environmental
impact and energy requirements) the Administrator determines has been
adequately demonstrated." The costs and impacts of the proposed standards for
coal-fired steam generating units (including lignite-fired units) have been
considered and are deemed reasonable. In the preamble to the proposed
standards, it was estimated that national S02 emission reductions of
9,500 Mg/year (11,000 tons/year) would result from the S02 standard for coal-
fired units. Based on these potential emission reductions, the cost of the
standards is not unreasonable.
Although it is true that the standards will increase the cost of unit
operation because of the additional costs of pollution control, this increased
cost is likely to be much less significant in fuel selection decisions than
considerations such as fuel price, fuel availability, and fuel supply
2-82
-------
reliability. Changes in relative fuel prices among coal, oil, and natural gas
will have a much greater effect than this standard on the decision of a unit
owner or operator to select lignite (a low cost, readily available, and
reliable fuel source in North Dakota) versus conventional coal, oil, or
natural gas.
Finally, the NAAQS and NSPS requirements are developed under separate
sections of the CAA and for separate purposes. The purpose of Section 111
(NSPS) standards is to avoid future air quality problems through application
of best demonstrated control technologies to new sources, regardless of
whether they are located in areas where NAAQS are attained.
2.7.6 Anthracite Coal Industry
1. Comment: One commenter (IV-D-43) asserted that anthracite-fired steam
generating units are sufficiently different from units firing other types of
coal to justify separate treatment within the context of the proposed
regulations. The commenter stated that the proposed regulations were based on
emissions data from units firing coal other than anthracite, which generate
substantially more flyash. According to the commenter, these nonanthracite-
fired units burn fuel in suspension above the grate; whereas, anthracite-
fired units use cross-feed stokers and burn fuel on the grate. The commenter
stated that anthracite stokers produce less flyash because they have lower
burning rates/square foot of grate area and use lower air velocities.
The commenter pointed out that anthracite is double-screened before it is
burned and that this procedure further reduces the potential for producing
flyash in anthracite-fired units. The commenter acknowledged that few data
are available to document flyash generation in cross-feed stoker units;
however, the commenter stated that he would collect and submit some data to
the EPA to support his assertions after the data become available when the
upcoming heating season begins (October/November 1989).
The commenter recommended that Control Option A, cyclone-type
separation, be used as the basis for regulations governing anthracite
particulate emissions from small steam generating units. The commenter
2-83
-------
pointed out that a precedent exists for considering emissions from anthracite
units separately from other coal-fired units in the 1977 CAA Amendments
governing utility generating stations.
Response: The proposed PM standard for coal-fired units was based on an
evaluation of PM emissions and control technologies for all types of coal,
including anthracite coal. The commenter is correct in saying that
anthracite-fired units generally have inherently lower PM emissions than other
coal-fired units because of the factors described by the commenter. However,
the CAA requires that NSPS be based on the best demonstrated control
technology, considering cost, energy, and other impacts. For coal-fired
units, including anthracite-fired units, the best demonstrated technology for
the control of PM emissions was determined to be fabric filters. Thus, the
proposed PM standard for coal-fired units was based on the performance of
fabric filters. As stated in the preamble to the proposed regulations, fabric
filters are capable of reducing PM emissions from such units to 22 ng/J
(0.05 Ib/million Btu) heat input or less.
The use of double mechanical collectors would only reduce PM emissions
from anthracite coal-fired units to 130 ng/J (0.30 Ib/million Btu) or less,
which is not different than the performance of double mechanical collectors on
other types of coal-fired units. As a result, the incremental reduction in PM
emissions achieved by installing a fabric filter is about the same, whether
the steam generating unit fires anthracite or other types of coal. Thus, in
considering the incremental costs and the incremental benefits of fabric
filters over double mechanical collectors for units firing anthracite or other
types of coal, there is no substantial difference and, therefore, no basis for
a more lenient standard for small units firing anthracite coal.
Consequently, the final standards include the same PM emission limit for
all coal-fired units with heat input capacities greater than 8.7 MW
(30 million Btu/hr), regardless of the type of coal combusted.
2. Comment: One commenter (IV-D-43) stated that in the proposed
regulation, flyash was defined as having toxic characteristics due to trace
2-84
-------
elements present in the material. The commenter pointed out that trace
elements constitute a portion of the earth's crust in its natural state so
that if a material is toxic due to the mere presence of trace elements, then
nature itself could be defined as toxic. The commenter explained that
anthracite ash is similar in composition to burned mineral products such as
brick or expanded aggregate materials. To support his assertions, the
commenter attached to his comments a trace element composition breakdown of
anthracite ash.
Response: The toxic constituents of anthracite are no different from
the toxic constituents of other coals. These harmful elements are present in
the inhaleable size range of the particulate matter in the flyash from coal-
fired steam generating units and are released into the atmosphere. The
problem is not that these trace elements are present in the coal; rather, the
problem is that they are released into the atmosphere and can enter the body
through inhalation.
3. Comment; One commenter (IV-D-43) argued that the decision scheme used
in developing the proposed regulations did not consider the major impact the
proposed regulations would have on the anthracite industry. The commenter
stated that anthracite competes with natural gas and fuel oil on the fuel
market for facilities of the affected size range. The commenter pointed out
that although anthracite is considerably less expensive than gas or oil in its
marketing area, the equipment to fire anthracite is more expensive. The
commenter stated that if an anthracite firing system is installed, it
typically pays for itself through fuel cost savings within a span of five
years. He further stated that an increase in system capital costs would
result in a longer payback time, thereby reducing the incentive to purchase an
anthracite-fired unit.
The commenter stated that control costs for PM Control Option D seem to
have been generated by mathematical manipulation of numbers derived from
larger units. The commenter also stated that preliminary price estimates
based on manufacturers' quotations rather than derived cost estimates suggest
2-85
-------
that Control Option D could increase the capital cost of an anthracite steam
generating unit in the 8.7 MW (30 million Btu/hr) range by over 20 percent,
which is not "negligible" as the EPA states in the preamble. The commenter
concluded that these costs would result in adverse economic impacts affecting
the use of anthracite-fired units, and that these units would be replaced by
units firing other, more expensive fuels, which in turn would increase
national dependency on imported energy.
Response: The potential impact of the standard on small coal-fired
units firing anthracite is not substantially different than small coal-fired
units firing other types of coal. Anthracite is not unique. The capital cost
impacts of the PM standard for anthracite coal-fired units are comparable to
the impacts for other types of coal-fired units. For the smallest bituminous
coal-fired units subject to the new PM standard (i.e., 8.7 MW [30 million
Btu/hr]), the regulatory impacts analysis showed that the total capital costs
of the steam generating unit would increase approximately 12 percent compared
to baseline levels. For the smallest anthracite coal-fired units, the total
capital cost is expected to increase by approximately 15 percent. The
estimate of the increased capital cost for anthracite coal-fired units is
somewhat lower than the commenter's estimate but, more significantly, is not
substantially different from that for other small coal-fired steam generating
units.
The analysis also included estimates of the increased annualized cost of
a new unit, the potential impact of the regulations on fuel-use patterns, an
evaluation of non-air quality environmental impacts, and consideration of
other possible impacts. With respect to annualized cost, the analysis showed
that the PM standards are reasonable for coal-fired units greater than 8.7 MW
(30 million Btu/hr) even considering the effect of increased PM control costs.
This analysis employed actual fabric filter costs for units as small as 8.7 MW
(30 million Btu/hr). The analysis was not based on mathematical manipulation
of numbers derived for larger units, as suggested by the commenter.
With respect to fuel-use patterns, the analysis showed that the cost of
complying with the PM standard would have very little, if any, impact on the
use of coal relative to natural gas and oil. This is considered true for both
2-86
-------
anthracite and bituminous coal. In part, PM control cost would have little
impact because of the many factors that enter into fuel choice decisions
(e.g., fuel availability, reliability, past practices, fuel storage space)
apart from control system costs. Of the many factors involved, fuel price
generally affects fuel choice decisions more than regulatory requirements.
Because anthracite coal is readily available in the Northeast and has a lower
cost than other fuels in the region, the competitive position of anthracite
coal versus other fuels in its market area is not expected to substantially
change as a result of the PM standard.
4. Comment: Two commenters (IV-D-44, IV-F-1.4) were concerned about the
environmental impact that the proposed S02 regulation would have on the
anthracite industry. One commenter (IV-F-1.4) stated that the methods of
mining anthracite coal are very different from those used to mine bituminous
coal. Anthracite mining is conducted by reclamation -- digging out old
abandoned mines where deep mining and strip mining were conducted earlier this
century. After the remaining anthracite coal is mined by reclamation, the old
abandoned mines are cleaned up and the earth is replaced to its former state.
This commenter stated that the environmental impact that anthracite mining has
through cleaning up these old mines would be eliminated if the anthracite
industry is no longer able to compete on the commercial market.
One of the commenters (IV-D-44) stated that anthracite-fired steam
generating units should be exempted from the 90 percent S02 emission reduction
requirement because an exemption would permit anthracite to compete with other
sources of coal that are cheaper to mine, thereby stimulating the anthracite
industry in Pennsylvania. This commenter stated that anthracite coal is
primarily obtained from mining previously abandoned mine areas, many of which
pose safety hazards as a result of mine subsidence and water quality problems
due to acidic mine drainage. According to this commenter, uncontrolled mining
in the past has left hundreds of thousands of acres of unreclaimed strip mines
and abandoned deep mines. This commenter pointed out that increased
anthracite mining activity will lead to reclamation of previously abandoned
mine sites and will alleviate mine subsidence, improve water quality, increase
2-87
-------
aesthetic value, and return the site to productive use.
This commenter asserted that an exemption from the 90 percent reduction
requirement would have little impact on S02 emissions, due to the natural low
sulfur content of anthracite coal and the relatively small size of units
affected. The commenter further stated that failure to grant an exemption for
anthracite would place anthracite at a competitive disadvantage because it is
almost entirely obtained from mining previously worked areas, which makes
anthracite more expensive to mine than many other sources of coal.
The commenter further asserted that anthracite is capable of meeting a
520 ng/J (1.2 Ib/million Btu) S02 limit without flue gas treatment. Meeting
this emission limit is possible because the majority of the sulfur present in
raw mined anthracite is pyrite, which is removed readily during the mechanical
cleaning processes. The commenter stated that no measurable improvement in
air quality would be achieved by requiring anthracite units to achieve
emission reductions below this level (i.e., the 90 percent emission reduction
requirement).
Response: Steam generating units firing either anthracite coal or
bituminous coal are subject to the standards. In both cases, only those coal-
fired units greater than 22 MW (75 million Btu/hr) heat input capacity that
operate with an annual capacity factor above 55 percent are subject to the
S02 percent reduction requirement. Units below this size are subject only to
the 520 ng/J (1.2 Ib/million Btu) for S02, a limit that the commenters stated
anthracite-fired units could achieve without flue gas treatment. If units
below this size firing anthracite can meet the 520 ng/J (1.2 Ib/million Btu)
limit for $62 without controls, the cost burden for anthracite units to comply
with the standard would not be great. As previously described, the S02
percent reduction requirement has been determined to be achievable and
reasonable for units with heat input capacities greater than 22 MW (75 million
Btu/hr) and an annual capacity factor greater than 55 percent. The same add-
on control or coal cleaning technologies can be used for both anthracite and
bituminous coal to comply with the percent reduction requirement.
The exemption from the S02 percent reduction requirement granted for
anthracite in Subpart Da was provided to encourage reclamation of anthracite
2-88
-------
mines, resulting in environmental benefits such as improvement of mine
drainage acid-water conditions, elimination of old mining scars on the
topography, and eradication of dangerous fires in deep mines and culm banks.
The exemption from the percent reduction requirement provided under Subpart Da
for anthracite created a market for this fuel in the utility sector, and the
environmental benefits associated with this large-scale utility reclamation
were judged to outweigh any ambient air quality impacts of burning anthracite
without a post-combustion S02 control system.
The small projected overall coal demand of the industrial-commercial-
institutional steam generating unit sector combined with the predominant use
of locally available coals would generally result in anthracite being used as
a local fuel only, even if an exemption from the percent reduction requirement
were granted for anthracite. The small quantities of coal demanded by the
industrial-commercial-institutional steam generating unit sector in
northeastern Pennsylvania and other areas of localized anthracite deposits
would not result in the large-scale utility-type reclamation of abandoned
mines that might have resulted from the Subpart Da exemption. For that
reason, no special provisions for anthracite were included in the final
standards under Subpart Db applicable to industrial-commercial-institutional
steam generating units greater than 29 MW (100 million Btu/hr) heat input
capacity, and none are being included in the final standards under Subpart DC
for units below that size for the same reason.
A different situation exists, however, with the firing of anthracite
mining waste and other coal mining and washing wastes (collectively referred
to as coal refuse). These wastes raise concerns similar to those addressed in
the Subpart Da exemption for anthracite. These waste piles are not only
unsightly, but they are responsible for acid-water drainage problems and can
also lead to fires from spontaneous combustion. Therefore, it is important to
encourage the use of these wastes as fuels in steam generating units
(specifically FBC steam generating units) to eliminate a potential
environmental hazard. Consequently, a less stringent requirement of
80 percent reduction combined with an emission limit of 520 ng/J
(1.2 Ib/million Btu) has been provided for FBC steam generating units which
fire coal refuse. This action balances the need to minimize air pollution
2-89
-------
from combustion of these wastes against the environmental benefits resulting
from eliminating coal refuse piles.
2.8 NON-COST NATIONAL IMPACTS
1. Comment: Two commenters (IV-D-07, IV-D-17) stated that the proposed
standards and associated capital costs would prevent the building of new wood-
fired steam generating units. One commenter (IV-D-17) pointed out that most
wood-fired steam generating units burn what are essentially waste products
that would have to be landfilled if not burned. The commenter stated that
landfill ing clean wood waste is not a good practice because, as the wood
breaks down, methane is produced and settling of the landfill occurs. The
commenter also noted that there is a current national shortage of landfill
sites, and that several States have designated combustion as the preferred
disposal method for wood waste. The commenter stated that the virtual
elimination of new wood energy systems in the regulated size range, due to the
cost of compliance, will cause more environmental problems than it solves.
Another commenter (IV-D-07) stated that small wood-fired steam
generating units should not be required to meet the same emission limits as
large industrial units.
Response: It is true that wood-fired steam generating units do burn a
substantial amount of wood waste products and that landfill ing of clean wood
waste is not a preferred practice considering the national shortage of
landfill sites and other environmental, health, and safety problems with
landfills. Combustion using state-of-the-art control equipment and good
operating practices is becoming the preferred method for disposing of wood
waste, as discussed in the EPA report entitled "The Solid Waste Dilemma: An
Agenda for Action". However, with the possible increase in the amount of wood
use, there is a need for standards that achieve adequate emissions control
from small wood-fired steam generating units combusting wood waste in order to
protect air quality.
2-90
-------
As shown in the background document entitled "Projected Impacts of
Alternative Particulate Matter New Source Performance Standards for
Industrial-Commercial-Institutional Nonfossil Fuel-Fired Steam Generating
Units" (EPA-450/3-89-18), the impact of the 43 ng/J (0.10 Ib/million Btu)
standard applied to wood-fired units is expected to increase the national
annualized cost of wood combustion by about 19 percent over baseline costs,
which, while somewhat high, is considered reasonable given the substantial air
pollution control benefits achieved. Moreover, analysis of these compliance
costs indicate they are not sufficiently high to eliminate new wood-fired
steam generating units in favor of new coal, oil, or gas fired units. Thus,
the standard will not encourage a shift to landfill ing.
The standard of 43 ng/J (0.10 Ib/million Btu) heat input capacity
applied to small wood-fired units greater than 8.7 MW (30 million Btu/hr) was
evaluated on its own merits and found to be reasonable without regard to the
emission limit set under Subpart Db for industrial-commercial-institutional_
steam generating units above 29 MW (100 million Btu/hr) in size. Consistent
with the purposes of Section 111 of the CAA, this standard reflects the use of
the most effective PM control technologies which, considering cost, energy,
and environmental impacts have been adequately demonstrated (i.e., ESP's or
wet scrubbers). This standard would not only result in greater overall PM
emission reduction, but would also result in more efficient control of PMjQ
emissions and toxic emissions, such as polycyclic organic matter.
2. Comment: Two commenters (IV-D-16, IV-D-42) expressed concern that the
proposed regulation would cause fuel prices to increase. One commenter
(IV-D-16) pointed out that the price of natural gas is tied closely to the
price of fuel oil used in steam generating units. The 215 ng/J
(0.5 Ib/million Btu) limit for S02 from oil-fired units would eliminate the
use of medium sulfur residual oil, which would result in price increases for
both natural gas and distillate oil.
One commenter (IV-D-42) stated that by increasing the application costs
of coal and oil through regulation, the market value of all alternate fuels
will rise, especially the cost of natural gas. The commenter stated that this
2-91
-------
will raise the cost of natural gas and other alternate fuels to all people.
The commenter stated that the EPA needs to conduct an economic impact study
assuming that the cost of all alternate fuels will rise to the application
value of the lowest cost regulated fuel.
Response: Although competition from residual oil may influence the
price of natural gas and distillate oil, the prices of natural gas and
distillate oil are also "driven" by other factors, such as marketplace
competition between natural gas and distillate oil, international trade
regulations, and foreign price controls. In addition, new small steam
generating units will combust only a very small percentage of the total fuel
oil combusted in the United States. Compared to the amount of fuel oil
consumed in existing small, industrial, and utility steam generating units,
the amount of fuel consumed by new small steam generating units is so small
that the standard would have only a minimal, if any, impact on fuel market
prices.
3. Comment: Three commenters (IV-D-18, IV-D-22, IV-D-45) were concerned
that the proposed standards would affect the coal market. One commenter
(IV-D-22) stated that the additional restrictions imposed by the proposed
standard on coal-fired steam generating units would eliminate the future use
of coal at his facilities, as well as other facilities, thereby affecting the
national energy policy.
A second commenter (IV-D-45) stated that a more thorough analysis of the
impact upon potential coal use by facilities subject to the proposed percent
reduction requirement is necessary. The commenter continued that the failure
to analyze the possibility of units switching from coal to natural gas is
particularly worrisome, given the historical predominance of coal in the
industrial steam generating market.
A third commenter (IV-D-18) was concerned that the percent reduction
requirement would discourage the use of coal in the 22 to 29 MW (75 to
100 million Btu/hr) range for both new and existing steam generating units.
The commenter believed that, based on the EPA's interpretation in the
2-92
-------
Wisconsin Electric Power Company (WEPCO) case, applying the percent reduction
requirement to existing units would increase the total cost of the percent
reduction requirement. The commenter further stated that it would create a
potential for a large displacement of coal in the existing market as well as
in the future expanded market. The commenter pointed out that an impact on
the coal market would affect U.S. competitiveness in the world market.
Response; The standards would not eliminate the future use of coal in
small steam generating units. For facilities in this size range, the choice
of firing a specific fuel is based on a combination of factors including fuel
cost, fuel availability, reliability of supply, and other operating factors.
Evaluating whether an owner or operator would switch fuels based on cost alone
ignores the other factors that impact fuel selection. Since other
considerations impact fuel selection, the relatively small increase in costs
due to the standard is not likely to eliminate the use of coal when all other
factors are also considered.
The impacts of a percent reduction requirement were evaluated for a
range of steam generating unit size and capacity factors and found to be
reasonable for units in the 22 to 29 MW (75 to 100 million Btu/hr) range with
a capacity factor greater than 55 percent. As discussed in the preamble to
the proposed rule, the incremental cost of applying the percent reduction
standard to units smaller than 22 MW (75 million Btu/hr) heat input capacity
or to units with less than 55 percent capacity factor was considered
unreasonable. Therefore, percent reduction is not required on units in this
range. However, the percent reduction requirement was considered reasonable
when applied to units greater than 22 MW (75 million Btu/hr) and 55 percent
capacity factor considering the amount of emission reductions achieved.
As discussed in the preamble to the proposed standard, the number of
existing coal-fired units in the size range covered by the percent reduction
requirement is relatively small, and few if any additional new coal-fired
units in this range are expected to be built even in the absence of this NSPS.
No coal-fired units were projected in the baseline (i.e., without the NSPS)
because, based on cost considerations alone, no unit of this capacity would
fire coal. However, due to other considerations, as discussed above, new
2-93
-------
units in this size range may be built in the future. Even if these relatively
few new small coal-fired units were to switch from coal to natural gas or oil,
the magnitude of coal use represented by these few units is not sufficient to
cause any significant displacement of coal nor would it have a substantial
impact on the U.S. or world markets.
The percent reduction requirement is expected to have little, if any,
impact on the use of coal in existing steam generating units. The
Administrator's determination of October 14, 1989, in the WEPCO case should
affect only a very small number, if any, existing coal-fired steam generating
units. Because of the unusual set of facts present in the WEPCO case, few
existing coal-fired steam generating units are likely to fall within the scope
of the decision. Furthermore, owners or operators of existing coal-fired
units are not expected to switch to a fuel other than coal simply because
there is a remote possibility of being subject to the percent reduction
requirement as a result of future modifications to the unit.
4. Comment: Two commenters (IV-D-27, IV-D-42) expressed concern about the
economics and availability of trained technical personnel. One commenter
(IV-D-27) argued that most owners/operators of small steam generating units
have fewer technical and maintenance employees than owners or operators of
utility plants. This commenter also stated that the operators and maintenance
employees working at small steam generating units are familiar with the
current equipment and methods of operation; however, if changes were made in
unit operation (such as switching over to natural gas or to fluidized bed
combustion), this commenter predicted that problems would arise with teaching
and supervising the new operating procedures.
This commenter also noted that the majority of plants with small steam
generating units are not located in large cities so that hiring technically
trained employees would be difficult and the plants would be forced to hire
technical consultants, a costly practice. As a result of these ensuing
problems, the commenter believed that the proposed standards would cause the
owners/operators of small steam generating units to switch to natural gas,
which is not in the best interest of the country.
2-94
-------
The other commenter (IV-D-42) stated that, as a result of this proposed
regulation, small steam generating unit owners or operators will be forced to
hire at least one highly technical person to operate the pollution control
equipment and instrumentation. This commenter pointed out that small units
would not be able to afford this added cost of operation. This commenter
further stated that the added cost of operation was not included in the
background documents to the proposed regulation.
Response: First, it should be noted that the standards apply only to
new, modified, or reconstructed units, not existing units. Thus, any impact
the standards may have will only be on operations for new, modified, or
reconstructed units. The vast majority of owners/operators of existing small
steam generating units will not be affected by the standards.
As discussed in the preamble to the proposed regulations, units above
about 8.7 MW (30 million Btu/hr) heat input capacity are located predominately
at industrial facilities. Industrial facilities normally employ full-time
professional operators. Operators for new units above 8.7 MW (30 million
Btu/hr) may require additional training to deal with the PM standards, but
operators are normally given training in the operation of a new unit when one
is installed. This additional training on how to comply with the PM standards
could easily be incorporated into the startup training for a new unit. The
cost of such additional training would be minimal.
The population of units below 8.7 MW (30 million Btu/hr) is made up
largely of commercial-institutional units. Very few commercial or
institutional units are found in the size range above 8.7 MW (30 million
Btu/hr). Operators at commercial-institutional facilities, such as churches,
schools, offices and apartments, shopping centers, and laundries, are more
likely to work part-time or have other duties in addition to operating the
steam generating unit. As a result, these operators have little or no
experience or training in the operation of sophisticated PM emissions control
systems such as fabric filters and ESP's, and the cost of a full-time operator
trained in the operation of these controls would be relatively high for new
units below 8.7 MW (30 million Btu/hr) in size. In light of these
considerations, together with the higher cost effectiveness of applying these
2-95
-------
PM emissions control systems to units in this size range, no PM standards
apply to units below 8.7 MW (30 million Btu/hr) in size.
For owners/operators of units between 2.9 and 8.7 MW (10 and 30 million
Btu/hr) heat input capacity, the only standard that applies is for S02. The
S02 standard is based on use of low sulfur fuels, such as low sulfur coal,
wood, very low sulfur oil, or natural gas. No additional training should be
necessary for operators of these units, since there is no additional or new
equipment that needs to be operated. In this size range, the predominant
fuels used at existing facilities and expected to be used at new units are
natural gas and distillate oil. Thus, operators of these new units will in
most cases be operating the same type of unit for which they are already
trained and experienced in operating.
Units below about 2.9 MW (10 million Btu/hr) in size are found almost
exclusively at smaller commercial and institutional facilities, such as public
schools, churches, or laundries. Because these facilities usually employ
part-time or volunteer operators rather than full-time, professional
personnel, complying with the standards would be much more difficult for
owners/operators of these units than for owners/operators of larger ones. In
addition, units in this size range tend to be of a different type than larger
units (i.e., cast iron rather than firetube and watertube) and burn clean
fuels, such as natural gas or distillate oil. As a result, the emissions
potential for these very small units is much less. Because of the extra
burden required of the unit owner/operator and the small emission reductions
involved, units below 2.9 MW (10 million Btu/hr) have been exempted from the
standards.
It is to the advantage of the facility that adequately trained personnel
are available to supervise the operation and maintenance of a steam generating
unit. Adequately trained personnel improve the efficiency of unit operation
and also ensure that the unit is operating in a safe manner. Although units
above 8.7 MW (30 million Btu/hr) may have need to further train operators as a
result of the NSPS, the impact of the standards on these activities is not
considered unreasonable considering the emission reduction achievable.
2-96
-------
5. Comment: One commenter (IV-D-42) stated that the proposed regulation
would significantly raise the costs for small businesses as well as for
education in the U.S. and would force some firms to go out of business. The
commenter stated that this would increase our dependence on foreign
manufactured goods for many items that are necessary for our national
independence, including our national defense needs. The commenter also
maintained that the proposed regulations would stifle our innovative spirit to
try new ventures that could earn us revenues in the world market.
Response; The proposed regulation would not significantly increase the
operating costs for small businesses or schools. Most small businesses and
schools that use fossil-fuel fired steam generating units are below the size
cut-off of the standard. Those who are larger than the lower size cut-off
predominately burn a mix of distillate oil and/or natural gas. The impact of
the standard, therefore, is limited to new units that would have chosen medium
or high sulfur residual oil, or coal in the absence of the standard. The
small steam generating units impacted have the flexibility, in most cases, to
choose between very low sulfur residual oil, distillate oil, low sulfur coal
and natural gas.
2.9 ENVIRONMENTAL IMPACTS
2.9.1 Sulfur Dioxide
1. Comment: One commenter (IV-D-17) stated that wood energy systems
produce virtually no S02 and that every wood plant eliminates the need for
fossil fuel-fired capacity, which generally releases S02- The commenter
stated that, consequently, regulatory standards should not discourage the
building of new wood-fired steam generating units.
Response: It is not true that every wood-fired unit eliminates the need
for a fossil fuel-fired unit that generates S02 emissions. In fact, most
small steam generating units fire natural gas which releases virtually no S02
2-97
-------
emissions, and releases no PM emissions. Thus, it could be argued that many
small wood-fired units are likely to be eliminating small gas-fired steam
generating units, and thus replacing a source that does not emit PM with a
source that does emit PM.
The standards will not, in and of themselves, discourage the
construction of new wood-fired units. As discussed previously, the market
price among fuels and the reliability of fuel supplies has a greater impact on
fuel selection than the potential costs of these standards.
Under Section 111 of the CAA, emission limits for new sources are based
on the use of the best demonstrated technology, considering cost, energy and
other environmental impacts. Costs and impacts of the proposed standards for
wood-fired steam generating units have been considered and are deemed
reasonable. Although the cost-effectiveness value for the PM standard is
relatively high, there are significant emission reductions afforded by the
standard.
2. Comment: Three commenters (IV-D-22, IV-D-24, IV-D-26) stated that the
emission reductions achievable under the proposed S02 standard are
insignificant and the associated industry burden is unreasonable. One
commenter (IV-D-26) presented calculations demonstrating that the proposed
standard would reduce $62 emissions by 445 Mg (494 tons) in excess of the
reduction that a standard of 690 ng/J (1.6 Ib/million Btu) would achieve. The
commenter also presented figures showing that the proposed NSPS addresses S02
emissions that constitute only 0.07 percent of annual industrial coal-fired
steam generating unit emissions or about 0.00005 percent (5 x 10" ) of annual
national S02 emissions. The commenter concluded that, given what he viewed as
the unreasonable and exorbitantly costly burdens it would impose, the EPA's
proposed standard is not justified by this insignificant additional S02
removal.
Another commenter (IV-D-22) stated that the environmental benefits of
the proposed standard are negligible. The commenter presented figures showing
that the annual emissions of all new steam generating units affected by this
2-98
-------
regulation over the next 5 years would be only 0.155 percent of the annual
S02 emissions of one major utility plant.
Response: As stated in the proposal preamble, the emissions from small
steam generating units contribute significantly to air pollution, and
therefore new small steam generating units need to be regulated with NSPS. An
NSPS is intended to set standards that reflect use of the best demonstrated
control technology, considering cost and other factors. As discussed
elsewhere, all the cost and other potential impacts associated with the
proposed standards are considered reasonable. The incremental cost
effectiveness of a 520 ng/J (1.2 Ib/million Btu) standard over a 690 ng/J
(1.6 Ib/million Btu) standard, for example, is $750/Mg ($680/ton) of SC>2
removed. Thus, the proposed standard is considered reasonable in light of the
emission reductions that it will achieve.
New source performance standards are issued for stationary sources which
cause or contribute significantly to air pollution which may reasonably be
anticipated to endanger public health and welfare. The CAA of 1977 required
that EPA publish a priority list of all major stationary sources that were not
then regulated by NSPS, and to promulgate regulations for those sources on a
prescribed schedule. Within this legal framework, the relative amounts of
emission reductions among the various source categories is not important. The
important consideration is to develop NSPS for all major source categories, as
prescribed by mandate.
On August 21, 1979, a priority list for development of additional NSPS
was published in accordance with Sections lll(b)(l)(A) and lll(f)(l) of the
CAA (44 FR 49222). This list identified 59 major stationary source categories
that were judged to contribute significantly to air pollution that could
reasonably be expected to endanger public health or welfare. Fossil fuel-
fired steam generating units ranked eleventh on this priority list of sources
for which NSPS would be established in the future. The fossil-fuel category
later was amended to include nonfossil fuel-fired units and commercial and
institutional units.
Fossil fuel- and wood-fired steam generating units are significant
sources of emissions of PM (including PMjo), S02, and NOX. While emissions
2-99
-------
from individual units may not always appear to be substantial by themselves,
the construction of numerous new small steam generating units as a result of
industrial, commercial, and institutional sector growth is expected to result
in an increase in emissions of these pollutants that warrant control.
2.9.2 Particulate Matter
1. Comment: Two commenters (IV-D-07, IV-D-45) stated that using wet
scrubbers results in additional waste disposal problems, specifically water
pollution from scrubber blowdown and sludge waste from collected PM. One
commenter (IV-D-07) also argued that wet ESP's have the same problem with
water and waste disposal as wet scrubbers.
Response: As stated in the proposal preamble, projected water pollution
and solid waste impacts from PM control have been considered. Analysis
included impacts of both of these waste streams, and it was determined that
the volumes projected to be generated can be handled by conventional waste
management technologies. The wastes produced by PM control processes are not
defined as hazardous by the Resource Conservation and Recovery Act (RCRA), and
can be disposed of using traditional treatment and disposal techniques at a
reasonable cost.
2. Comment: One commenter (IV-D-45) pointed out that the preamble to the
proposed regulations acknowledges that polycyclic organic matter (POM), a
carcinogen, will be captured by wet scrubbers and ESP's, but elsewhere states
that waste generated by proposed PM controls are nonhazardous. The commenter
asserts that further review and resolution of this issue are needed.
Response; Polycyclic organic matter is an air pollutant that is
generated by incomplete combustion of fuel. When inhaled into the lungs, it
is a potential carcinogen. As part of a wet scrubber's or ESP's waste, POM is
not an air pollutant; rather, it is a solid or water waste that is not defined
2-100
-------
as hazardous waste by RCRA. Consequently, liquid and solid wastes containing
POM can be disposed of using traditional treatment and disposal techniques.
3. Comment: One commenter (IV-D-33) stated that he was concerned that the
EPA has not included any analysis of ambient air quality impacts of
alternative emissions standards. The commenter stated that such analyses have
been included in most previous NSPS proposals. The commenter further stated
that dispersion modeling for PMjo and trace metals would be helpful in
examining the impact of alternative PM standards.
Response: Ambient concentrations of air pollutants are regulated by
National Ambient Air Quality Standards (NAAQS), which are intended to achieve
and maintain specific ambient air quality concentrations in order to protect
public health and welfare. The purpose of NSPS is to minimize emissions
through application of best demonstrated technology, considering cost, so as
to prevent creation of new air pollution problems or exacerbation of existing
problems. In setting NSPS, ambient air quality analysis may be conducted if
such analysis might aid in selecting the basis for best demonstrated
technology. Since this was not judged to be the case for small steam
generating units, air quality modeling was not done.
4. Comment: One commenter (IV-D-17) pointed out that the EPA data
concerning wood states that wood-fired steam generating units with or without
flyash reinjection produce PM with less than 10 percent PMjQ- Since EPA has
reported that PM larger than PMjo is not respirable, the commenter saw no
justification for regulating PM from wood-fired units. Also, the commenter
stated that most wood-fired facilities are not located in urban areas and
that, consequently, provides even less justification for regulating these
facilities.
The commenter further stated that the background document fails to cite
any basic references for wood combustion and recommended that the EPA perform
2-101
-------
a series of analytical studies of wood-fired steam generating units before
attempting to regulate them.
Response: The most current data on PMjo emissions (AP-42, 1985)
indicate that for wood-fired steam generating units controlled with mechanical
collectors alone, PMjn accounts for 30 to 90 percent of total suspended
particulates (91 percent with fly ash reinjection, 32 percent without).
The emissions and control technologies for wood-fired steam generating
units were studied and documented in substantial detail during the development
of the NSPS for industrial-commercial-institutional steam generating units
(Subpart Db) that was published on November 25, 1986. The same technologies
found effective in large units are equally applicable to small steam
generating units as well. The gas stream characteristics of small wood-fired
units are very similar to those of larger units because unit design and
operation is essentially the same for all wood-fired units within the
industrial steam generating unit population. Furthermore, the basic design
and operation of the PM emission control systems for wood-fired units do not
vary substantially according to the size of the industrial unit to which they
are applied. Therefore, there was no need to perform additional or redundant
technical analysis. Control technology options for wood-fired units are
described in the background document "Overview of the Regulatory Baseline,
Technical Basis, and Alternative Control Levels for Particulate Matter (PM)
Emission Standards for Small Steam Generating Units" (EPA-450/3-89-11), and
also in the proposal preamble to Subpart Db (49 FR 25123).
5. Comment: Three commenters (IV-D-24, IV-D-26, IV-D-28), in response to
the EPA's request for comments on the lower PMjQ collection efficiency of
mechanical collectors as an air toxics issue, stated that this is an
insignificant concern and is not justification for requiring baghouses or
ESP's on small steam generating units. The commenters presented figures,
based on information from an EPA document, indicating that a 14.6 MW
(50 million Btu/hr) unit operating at 55 percent capacity with a multiclone
mechanical collector will emit approximately 0.1 ton (200 Ibs) per year more
2-102
-------
PM than a similar unit equipped with an ESP. The same comparison at
26 percent capacity indicates that only 0.05 ton (100 Ib) higher emissions
will occur with a multiclone. The commenters concluded that the additional
PM emission reduction gained by the use of an ESP or baghouse is not
justification for imposing a 22 ng/J (0.05 Ib/million Btu) standard on small
steam generating units.
Response: Emission limits that are set by an NSPS are based on the
emission levels achievable using best demonstrated control technology,
considering costs. Best demonstrated technology is not determined by
considerations of potential toxics emissions alone, but by consideration of
all factors influencing the overall benefits and the overall costs associated
with the use of best demonstrated technology. Fabric filters and ESP's are
capable of achieving PM emission levels of 22 ng/J (0.05 Ib/million Btu) or
less, while mechanical collectors are only capable of achieving 130 ng/J
(0.30 Ib/million Btu) or less. Therefore, since the emission reductions are
significantly greater and the associated costs are not unreasonable, the
promulgated PM emission limit is 22 ng/J (0.05 Ib/million Btu) for coal-fired
steam generating units.
2.9.3 Other
1. Comment: Two commenters (IV-D-28, IV-D-40) stated that EPA should
provide incentives for the environmentally beneficial and cost-effective use
of chemical by-products/wastes as fuels in combination with coal or oil. The
commenters stated that the EPA's Control Technique Guidelines (CT6) for
control of VOC from industrial wastewater encourages this practice and,
therefore, EPA should not propose a regulation that discourages it by making
it cost prohibitive.
Response: The promulgated standards do not regulate the combustion of
by-product or waste fuels. As a result, the proposed standards neither
encourage nor discourage such practices.
2-103
-------
2. Comment: One commenter (IV-D-17) stated that, while all combustion
technologies produce carbon dioxide (C02), biomass combustion is the only one
that effectively recycles C02 through regrowth of the resource. He further
stated that the increase in capital costs associated with the proposed
standards would effectively prohibit construction of new wood-fired steam
generating plants in the size range affected by the regulations. Further, the
increased costs would also cause the loss of the environmental benefits of a
renewable energy source that does not contribute to S02 emissions, lessens
pressure on landfills, and helps limit atmospheric loading of C02 and methane.
Response: The commenter may be correct in pointing out that C02
produced by fuel combustion is, in some sense, "recycled" through the
absorptive capacity of trees and use of C02 by trees and other vegetation in
producing oxygen. However, the commenter apparently assumes that firing wood
in steam generating units somehow influences the renewal of our forest
resources by "causing" new trees to be planted. This relationship is not
clear. Most wood is not grown for use as a fuel, but is grown and harvested
for production of wood products, such as paper, lumber, and wood furniture.
Most wood combusted in steam generating units is wood scrap or other types of
wood wastes. Consequently, most wood used as a fuel is what is left over or
otherwise unusable after trees are harvested for production of wood products.
Similarly, most replanting of trees is not undertaken to maintain wood
as fuel, but rather is undertaken to maintain the supply of wood for use in
various wood products. Thus, the number of new wood-fired steam generating
units constructed in the future will have little or no effect on the rate of
tree replanting.
The increase in capital costs associated with the proposed standard is
estimated to amount to approximately 19 percent over baseline capital costs.
Assuming "full cost pass-through", the cost of services from most units will
increase by less than 0.5 percent. The increase in costs associated with the
proposed standards on the selection of wood as a fuel in steam generating
units will have less effect on fuel choice and investment decisions than
changes in prices of different fuels. Therefore, it is not anticipated that
2-104
-------
costs of PM controls will discourage the construction of new wood-fired steam
generating units.
2.10 SELECTION OF FORMAT OF STANDARD
1. Comment: One commenter (IV-D-30) questioned the consistency between the
preamble and regulation as it pertains to the proposed S02 standard. He
stated that the preamble discusses regulations and compliance demonstrations
in terms of fuel oil sulfur contents, whereas the actual proposed regulation
is written in terms of an S02 emission rate in Ib S02/million Btu. He stated
that this apparent inconsistency could result in problems implementing and
enforcing the proposed NSPS. He recommended that EPA resolve the
inconsistency.
Response: The proposed S02 standard for oil is an emission limit of
215 ng/J (0.5 Ib S02/million Btu). As stated in the preamble and in the
proposed regulation, compliance with this S02 emission limit can be
demonstrated through fuel sampling for sulfur content rather than stack
sampling or operation of a CEMS to measure S02 emissions in terms of
Ib S02/million Btu. Both the preamble and the proposed regulation describe
the test methods for all three of these options for determining compliance
with the standards. The preamble used the terms "percent sulfur" and
"ID/million Btu" interchangeably for oil because 0.5 percent sulfur is
approximately equal to 215 ng/J (0.5 Ib S02/million Btu). Therefore, a fuel
that conforms with either specification would be able to comply with the
standard.
2. Comment: Two commenters (IV-D-24, IV-D-45) recommended an alternative
format for the standards. One commenter (IV-D-24) stated that several
jurisdictions (e.g., New York City, Boston, Washington, D.C., and New Jersey)
require a specific fuel sulfur content for legal sales of residual fuel. The
commenter recommended that a legal sulfur content limit be established for
2-105
-------
residual oil covered by the American Society for Testing and Materials (ASTM)
grades 4 through 6.
The commenter presented market data for oil-fired water-tube units
indicating that at least two-thirds of these units fire distillate oil. The
commenter also recommended that a fuel sulfur limit be established for these
units.
Response: Section 111 of the CAA authorizes regulation of new
stationary sources. A standard such as that suggested by the commenter would
affect existing units, which the NSPS clearly is not authorized to regulate.
Nor can an NSPS hold suppliers accountable for the use of fuel by affected
facility owners or operators after a sale. An emitting facility is the only
entity for which an NSPS can be enforced. However, the CAA does authorize the
regulation of the emission potential of fuels as they are combusted by
affected facilities.
2.11 TEST METHODS AND MONITORING
2.11.1 Sulfur Dioxide
1. Comment: Two commenters (IV-D-32, IV-D-40) stated that EPA should allow
compliance with fuel sampling requirements to be determined on a 30-day
average basis, relying on composite samples collected daily. The commenters
stated that this approach would provide sufficient compliance monitoring while
significantly reducing analytical costs compared to the 30-day rolling average
approach.
Response; A rolling average basis was selected as the format for the
fuel sampling requirements because it is not significantly more costly than a
30-day composite average format, and yet facilitates enforcement of the S0£
standard. Both a rolling average format and a composite average format would
require fuel samples to be taken on a daily basis. The only cost advantage to
a composite average is that analysis of the samples collected would only have
2-106
-------
to be performed every 30 days, instead of daily. The estimated cost
difference between daily versus monthly analysis of composite samples is
$28,000/year. In light of the enforcement advantages discussed below, this
cost difference is not considered significant.
The purpose of a rolling average is to provide compliance data on a
daily basis. A 30-day composite average, on the other hand, would not provide
a daily record that could be used as the basis for enforcing the standard.
Instead, a composite average would allow enforcement only on a monthly basis.
Thus, a 30-day composite average would reduce the enforceability of the
standards by lengthening the period before which a compliance problem could be
identified. In balancing the modest increase in analytical cost against the
loss of enforcement capability, it is reasonable to require fuel sampling and
analysis for compliance on a rolling average basis, rather than a 30-day
composite basis.
2. Comment: One commenter (IV-D-16) asked if it would be acceptable to EPA
for the supplier of distillate oil to certify whether the distillate oil
complies with ASTM specifications. The commenter also asked whether it would
be acceptable to EPA for the supplier of the residual oil to provide an
analysis of the oil to the steam generating unit owner in order to meet the
S02 monitoring requirements.
Six commenters (IV-D-22, IV-D-24, IV-D-26, IV-D-28, IV-D-40, IV-D-45)
stated that requiring daily fuel sampling at small coal-fired steam generating
units would impose an unreasonable cost burden without any significant
improvement in air quality. These commenters stated that the fuel sampling
procedures, as specified in EPA Reference Method 19, are overly complicated
and costly. They pointed out that the facilities with small units cannot take
samples as frequently and as accurately as required without using an automatic
sampling system, which would be expensive. The commenters stated that the
equipment is complex, and facilities with small steam generating units usually
lack trained personnel to run it. One commenter (IV-D-45) also pointed out
that the reasons for selecting a cutoff of 8.7 MW (30 million Btu/hr) for the
PM standard concerning emission control equipment likewise apply to the S02
2-107
-------
standard concerning compliance procedures. In both instances, employees may
work part-time and lack the necessary skill for operating sophisticated
equipment. The commenters stated that no justification was offered for
requiring this rigorous sampling method.
The commenters pointed out that coal companies certify the sulfur, Btu,
and ash content of the coal at the time of purchase for each shipment to a
facility. The commenters said that coal or oil purchased at a specific sulfur
content will meet the defined $03 emission limit, and that duplicate testing
and paperwork by the fuel purchaser is unnecessary. The commenters stated
that the certification of the sulfur content of fuel when purchased and a
composite sample and analysis would be adequate to prove compliance on a
monthly basis. The commenters recommended, however, that the fuel supplier
should not be held liable for the owner's or operator's compliance. The owner
or operator of the unit should take and analyze spot samples to ensure
compliance with the 30-day average emission limit.
Response: In the case of distillate oil, the fuel supplier's
certification that the oil shipment meets ASTM specifications for distillate
oil is sufficient to demonstrate compliance with the S02 standards. A
supplier certification is considered sufficient for distillate oil because
ASTM specifications include a requirement that the fuel sulfur content be
0.5 weight percent or less to qualify as distillate oil. Thus, any fuel
certified as distillate oil has, by definition, a sufficiently low sulfur
content capable of meeting an S02 emission limit of 22 ng/J (0.5 Ib/million
Btu) on an as-delivered basis.
Unlike distillate oil, the sulfur content of residual oil ranges from
below 0.5 weight percent sulfur to well above this sulfur content level. In
addition, fuel supplier certifications are seldom the results of sampling and
analysis of the fuel shipment actually delivered to a steam generating unit.
In almost all cases, these certifications are the result of sampling and
analysis of very large shipments of fuel oil before they leave a petroleum
refinery. The route a fuel oil shipment may follow between a petroleum
refinery and a local fuel oil supplier can be long and tortuous. Trying to
confirm that the results of fuel sampling and analysis at a distant petroleum
2-108
-------
refinery are indeed representative of the fuel shipment received from a local
supplier is very difficult. Consequently, sampling and analysis of the local
fuel shipment is necessary to ensure compliance with the standards.
Fuel sampling and analysis for residual oil is a relatively simple
procedure, calling for a steam generating unit operator to collect a sample of
the fuel oil immediately after the fuel tank is filled (i.e., after each
shipment is delivered) and to submit it to a laboratory for analysis of the
sulfur content. Residual oil-fired steam generating units with heat input
capacities larger than 8.7 MW (30 million Btu/hr) are typically located at
industrial facilities and, as a consequence, have full-time professional
operators. These operators are trained and capable of sampling a residual oil
shipment and preparing the sample for submission to a laboratory. Further,
because the analytical equipment for determining the sulfur content of the oil
is not unreasonably costly, the analysis may be performed on-site.
Oil-fired steam generating units smaller than 8.7 MW (30 million Btu/hr)
heat input capacity, however, are typically located at institutional or
commercial facilities (e.g., shopping centers, schools, hospitals) and
consequently may not have a full-time operator. For this reason, fuel
sampling and analysis would be a greater burden on the owners and operators of
units with heat input capacities smaller than 8.7 MW (30 million Btu/hr).
Therefore, although somewhat less reliable, certifications of residual oil
sulfur content from fuel suppliers can be used to determine compliance for
residual oil-fired units smaller than 8.7 MW (30 million Btu/hr) heat input
capacity. This alternative may result in some increase in $03 emissions from
these units, but this consequence is considered reasonable in light of the
increased burden that would otherwise be placed on these smaller units.
For coal-fired steam generating units, the circumstances are much the
same. The coal sampling and analysis procedures described in Method 19
require the attention of a full-time, trained operator to collect the samples
with the required frequency and to prepare the sample for analysis properly.
Industrial facilities typically have full-time operators for their steam
generating units who can be trained in this procedure. Consequently,
Method 19 is an appropriate alternative to continuous emission monitoring at
facilities larger than 8.7 MW (30 million Btu/hr) heat input capacity.
2-109
-------
For smaller steam generating units, which typically do not employ a
full-time operator, coal sampling and analysis would be a greater burden.
Consequently, these standards allow operators of steam generating units with
heat input capacities smaller than 8.7 MW (30 million Btu/hr) to rely on a
coal supplier's certification as to the sulfur content of the coal purchased
and fired in the steam generating unit. Again, this alternative may result in
some increase in S02 emissions from these units, but this consequence is
considered reasonable in light of the increased burden that would otherwise be
placed on these smaller units.
3. Comment: One commenter (IV-D-36) stated that units with federally
enforceable annual capacity factor limits of ten percent or less that fire
very low sulfur oil should not be required to perform a 24-hour, full load
performance test. The commenter pointed out that, under the proposed
standards, similar units firing very low sulfur oil, yet operating at higher
capacity factors, are not subject to performance tests. The commenter stated
that fuel sampling and analysis procedures are adequate to demonstrate
compliance with S02 emission limits.
The commenter further stated that the proposed requirement to include at
least one period of 24-hour operation at full load during the 30-day
performance test entails significant additional cost if full load operation is
not normally required for an entire 24-hour period. The commenter stated that
this requirement will afford no additional air quality benefit, and is in
contradiction with the EPA requirement to conduct performance tests under
conditions "representative of future operating conditions". The commenter
recommended deleting the requirement for a period of 24-hour full load
operation during the 30-day performance test.
Response: The requirement contained in Section 60.44c(d) of the
proposed regulation for a 24-hour performance test for very low capacity
factor units firing only very low sulfur oil was intended to be less
burdensome than the 30-day initial S02 performance testing requirement
applicable to other oil-fired units. Under the proposed regulation, all
2-110
-------
oil-fired units except very low capacity factor units were required either to
conduct a 30-day initial performance test pursuant to Section 60.44c(c) or
maintain records to demonstrate use of a very low sulfur oil pursuant to
Section 60.44c(j). Rather than a full 30-day initial performance test, the
proposed standards specified only a 24-hour initial performance test for very
low capacity oil-fired units.
Based on further consideration of the S02 performance testing and
compliance requirements for very low capacity factor oil-fired units, it
appears that the 24-hour test, rather than being less burdensome, is actually
more burdensome than the initial performance test required for other units.
Therefore, the requirement for a 24-hour initial compliance test has been
deleted in the final standards. In lieu of an initial performance test,
owners or operators of distillate oil-fired units are allowed to certify that
the oil they combust meets ASTM specifications for distillate oil, provided
they maintain records of fuel supplier certifications that they have purchased
oil meeting ASTM specifications for distillate oil. Similarly, supplier
certification of fuel sulfur content is allowed as an alternative to an
initial performance test for residual oil-fired units between 2.9 and 8.7 MW
(10 and 30 millon Btu/hr) heat input capacity. For residual oil-fired units
between 8.7 and 29 MW (30 and 100 million Btu/hr) heat input capacity, the
initial performance test consists of sampling and analysis of the initial tank
of fuel oil to demonstrate compliance with the standard. Thereafter, the
owner or operator must sample and analyze the sulfur content of each new
shipment of fuel oil and calculate compliance with the standard on a 30-day
rolling average basis.
The requirement for a period of 24-hour full load operation during the
30-day initial performance test was included in the proposed standards to
demonstrate the maximum design heat input capacity for units seeking a more
lenient standard based on low annual capacity factor operation. Because the
standard for oil-fired units is independent of annual capacity factor, there
is no need for this demonstration of maximum heat input capacity. Only
coal-fired units between 22 and 29 MW (75 and 100 million Btu/hr) heat input
capacity seeking an exemption from the percent reduction requirement by
operating at an annual capacity factor of 55 percent or less must perform the
2-111
-------
24-hour demonstration of maximum heat input capacity. This demonstration is
required under Section 60.44c(i). Therefore, the requirement for a period of
24-hour full load operation during the initial performance test has been
deleted from the final rule.
4. Comment: One commenter (IV-D-22) stated that the proposed S02 testing
and monitoring requirements for coal-fired steam generating units are
excessive. The commenter stated that CEMS are an inefficient and overly
expensive method of measuring SOg emissions. According to the commenter, the
initial capital cost for an SOg monitoring system with software and
installation ranges from $139,000 to $170,000, with annualized costs of
$60,000.
Response: As discussed in the preamble to the proposed regulation, the
costs of the standards, including the cost of a CEMS, were evaluated and
determined to be reasonable. For example, the capital cost of a new coal-
fired unit 22 MW (75 million Btu/hr) in size would increase by less than
1 percent by requiring a CEMS. Annualized cost would increase by only about
.1.5 percent. In view of the enforcement benefits of CEMs on coal-fired units,
these additional costs are considered reasonable.
It should be noted that CEMS are not the only S02 monitoring option
allowed under the standards for coal-fired units. Owners or operators can
elect to install a CEMS, use Method 6B, or conduct daily fuel sampling and
analysis. Therefore, use of a CEMS is only one of three options available for
complying with the monitoring requirements of the NSPS. For coal-fired units
with heat input capacities of 8.7 MW (30 million Btu/hr) or less, fuel
supplier certification of each shipment's sulfur content is allowed. Thus,
for very small coal-fired units, even a fourth option is available.
5. Comment: One commenter (IV-D-22) recommended that determining the
percent sulfur content of the coal would be a more direct method of
2-112
-------
calculating S02 emissions because CEMS give SOg emission readings in Ib/hr
rather than in Ib/million Btu as-specified by the standard.
Response: It is correct that CEMS output is not expressed in Ib/million
Btu. However, a standard procedure is available to convert the emissions data
to the form expressed in the S02 standard. Also, daily coal sampling may be
used instead of CEMS to determine sulfur content and heat content, from which
emissions can be calculated.
6. Comment: Five commenters (IV-D-24, IV-D-26, IV-D-28, IV-D-40, IV-D-45)
stated that requiring a CEMS for small coal-fired steam generating units would
impose an unreasonable cost burden without any significant air quality
improvement. The commenters stated that CEMS equipment is complex, and
facilities with small steam generating units usually lack trained personnel to
run it.
Response: In evaluating possible S02 monitoring choices for small coal-
fired units, CEMS for units subject to the percent reduction requirement was
determined to be a reasonable choice based on the ability of this monitoring
technique to determine compliance with the standard. The use of CEMS will
ensure that emission reductions are constantly being achieved and provide for
the greatest enforcement capability. The successful operation of CEMS has
been demonstrated at many facilities without problems. As a result, it has
been determined that this method is neither too complex nor cost prohibitive
in light of the increased enforcement capability it provides.
As discussed above, however, steam generating units smaller than 8.7 MW
(30 million Btu/hr) heat input capacity are typically located at non-
industrial facilities and often do not have a full-time operator. The proper
maintenance and use of a CEMS requires a trained operator capable of providing
attention at any time during the operation of the CEMS. Consequently, this
monitoring approach is not appropriate for many steam generating units smaller
than 8.7 MW (30 million Btu/hr) heat input capacity. The final standard will
2-113
-------
allow operators of coal-fired steam generating units smaller than 8.7 MW
(30 million Btu/hr) heat input capacity to-demonstrate compliance with the
standards through fuel supplier certification.
7. Comment: One commenter (IV-D-36) stated that the requirement to perform
Appendix F testing on an S02 CEMS is too expensive, especially for utility
auxiliary units. The commenter recommended that only the cylinder gas audit
portion of the Appendix F testing be required.
Response; The requirements to perform daily CEMS drift checks and
quarterly accuracy audits as required under Appendix F were evaluated along
with other costs of monitoring and found to be reasonable and necessary.
Appendix F testing is important to ensure that the CEMS is operating properly
and is indicating accurately that compliance with the S02 standard is being
achieved.
8. Comment: One commenter (IV-D-03) expressed concern with the wording of
Method 19. He stated that, as currently written, Method 19 does not specify
clearly that the procedures are to be used for compliance determination or for
monitoring, except for fuel treatment cases or for control device inlet
sampling. The commenter recommended that Method 19 be revised and that
paragraphs similar to 3.2.2 and 3.2.3 of Method 19 be added to paragraph 2.0.
The commenter also recommended that the reference to ASTM D270-65 should be
updated to include new ASTM designations.
Response: Method 19 has been updated recently to specify procedures to
be followed when using fuel sampling methods for determining compliance
(see 52 FR 47853). These revisions should sufficiently clarify the compliance
procedures.
When establishing a standard, current ASTM methods are referenced that
are specified by dates because the wording of these specifications is known to
be in accordance with the standard at the time of promulgation. The newest
2-114
-------
ASTM specification is not automatically adopted for use in the regulation
because a revised ASTM specification could become less stringent or change in
a way that no longer applies to the standard. To ensure that the reference
specification is changed to remain abreast of current practices while
retaining its application in the standard, the ASTM specifications will be
reviewed and updated, as applicable.
9. Comment: One commenter (IV-D-06) acknowledged that the proposed
Section 60.46c(b) would allow some small steam generating units to sample the
sulfur content of fuel in lieu of operating a CEMS for SOg. The commenter
recommended, however, that to avoid sampling systems that would not provide
S02 data on a time basis consistent with outlet $63 data, relative accuracy
testing of the coal sampling and analysis system should be required at the
inlet to the SOg control device.
This commenter also stated that the proposed Section 60.46c(e)(3)
specifies setting the span value of the outlet S02 CEMS at 50 percent of the
maximum estimated hourly potential S0£ emissions of the fuel combusted. He
pointed out that this setting could result in the inability to determine
exceedances of the S02 emission limit when a relatively low sulfur fuel is
used. The commenter recommended that the span value of the outlet S02 CEMS
should be at least 125 percent of the applicable emission standard.
Response: The proposal of the fuel sampling and analysis procedures in
Section 60.46c(b) as an alternative performance monitoring method is based on
the Administrator's judgement that these procedures are sufficiently accurate
to be used as a basis for determining compliance with the SC>2 standard for
small steam generating units. Quality assurance and control plans are not
deemed necessary at this time due to the inherent precision and accuracy built
into the sampling and analysis procedures specified in Method 19.
The span values listed in Section 60.46c(e)(3) of the proposed
regulation were relevant only to those instances where a percent reduction in
S02 emissions is required in addition to compliance with an emission limit.
Section 60.46c of the final regulation has been amended to require a span
2-115
-------
value of 125 percent where only an emission limit (no percent reduction) is
required.
10. Comment: One commenter (IV-D-09) stated that the EPA should allow
alternative methods of fuel analysis for oil-fired steam generating units with
only one fuel tank. The commenter pointed out that according to the current
wording of the standard, an owner/operator with only one fuel tank would have
to shut the unit down while filling and analyzing the oil in the fuel tank.
The commenter stated that it was unreasonable either to shut the unit down or
to build a new tank and suggested the following alternatives: 1) sample the
tank immediately after refilling, or 2) sample and analyze the incoming
shipment prior to unloading.
Response: The language of the proposed regulation, requiring oil
samples to be collected "from the fuel tank after each new shipment of oil is
received and before any amount of oil is combusted" (Section 60.46c(b)(2))
would accommodate the commenter's first alternative suggestion that the tank
be sampled "immediately after refilling." As use of this new fuel begins, the
owner or operator can send the fuel sample off to be analyzed. If the owner
or operator subsequently learns that this fuel analysis shows a sulfur content
too high to comply with the standard, then the owner or operator must ensure
that the sulfur content of the next fuel shipment is low enough to meet the
standard using a 30-day rolling average for compliance. The final regulation
has been clarified to reflect this procedure in Section 60.46c.
11. Comment: One commenter (IV-D-36) stated that Section 60.46c(b) of the
proposed rule seems to require fuel sampling and analysis of both as-stored
and as-fired fuel. This requirement is unnecessary and imposes additional
costs for no additional air quality benefits. The commenter suggested
allowing either as-stored or as-fired fuel sampling and analysis.
2-116
-------
Response: Section 60.46c(b) of the proposed rule was intended to
require fuel sampling and analysis of either as-stored or as-fired fuel, but
not both. The type of sampling and analysis selected for oil-fired units will
be the choice of the owner or operator. To clarify this provision, the
wording of Section 60.46c has been changed to show clearly that the fuel
sampling provisions are alternative rather than supplemental requirements.
12. Comment: One commenter (IV-D-06) stated that the proposed
Section 60.48c(e)(5) apparently allows CEMS operators to decide which data
should or should not be included when calculating emission rates. The
commenter recommended that specific criteria be given in the proposed standard
to ensure that all valid data are used and to determine whether or not some
data should be included in the emission averages.
The commenter suggested that data that do not meet these specific
criteria, but that the CEMS operator believes to be invalid, should be
included in the calculations along with his/her justification for considering
the data invalid. The ultimate decision of whether to include these data in
calculating the emission rate should be made by the reviewing agency.
Response: Section 60.44c(e) of the proposed rule required that all
valid emissions data must be used in calculating the emission rate and the
percent S02 reduction. Also, Section 60.48c(e)(5) of the proposed rule
required owners or operators to identify all data excluded and the reasons for
excluding such data. If the reasons for excluding these data are not
considered sufficient, the enforcement agency has the authority to require
that such data be included in the emission rate and percent reduction
calculations.
Every piece of data gathered in determining compliance is required to be
submitted as part of the reporting and recordkeeping requirements. The
75 percent requirement does not mean that only 75 percent of the collected
data must be submitted. Rather, 100 percent of all valid data collected must
be submitted. The 75 percent requirement means that all data must be gathered
during at least 75 percent of the total hours that the unit is operating.
2-117
-------
13. Comment: One commenter (IV-D-06) stated that when fuel sampling and
analysis data or Method 6B data are used in lieu of inlet S02 CEMS data, daily
averages are calculated and subsequently averaged to yield the required 30-day
averages. The commenter pointed out that if this level of accuracy is
acceptable for calculating the fuel sampling and Method 6B data, then CEMS
data should also be averaged on a daily rather than hourly basis to yield the
required 30-day averages.
The commenter argued that the current method used for calculating 30-day
averages is not useful in determining either the impact of pollutant levels on
the atmosphere or the operating condition of pollution control equipment and
should be discontinued. The commenter stated that some 30-day averages may
include as few as 30 hours of source operation, whereas others may include as
many as 720 hours. Therefore, the same reported S02 emission rates for any
two time periods may not indicate the actual or relative mass emissions during
these periods. He added that 30-day averages calculated using "valid" hours
from days that also contain many "invalid" hours would not be representative
of the entire day, since actual emissions during the "invalid" time may be
drastically different from the actual emissions during the "valid" time.
The commenter recommended that in order to obtain information useful in
determining both the impact of pollutants on the atmosphere and the operating
condition of pollution control equipment, actual 30-day averages should be
computed using data from the previous 30 calendar days regardless of the
operation of the source. The commenter also noted that days when the source
is not operating and not emitting the pollutant of interest should be
considered invalid. He added that criteria should be established to determine
when there are sufficient valid data and valid days to consider a 30-day
average valid. Appropriate requirements to provide valid alternative
averages, other than 30-day averages, should also be included.
Response: Hourly averaging procedures are considered the preferred
method for determining compliance with the SOg standards. Because hourly
values are readily available from an operating CEMS, hourly averaging is
considered appropriate. For Reference Method 6B or fuel sampling and
2-118
-------
analysis, however, hourly sampling would be an unreasonable burden.
Therefore, only daily values are required for these monitoring methods.
The primary purpose of establishing an averaging time for compliance
purposes is to minimize the effect of variability in fuel sulfur content and
short-term performance of control devices on compliance with the standards.
The longer the period selected for averaging SC>2 emissions data, the lower the
variability exhibited by the data and the more realistically it reflects the
long-term performance of the system. However, in terms of enforcing
compliance with the standards, this averaging period must also be short enough
to permit timely enforcement of a standard once a source begins operation. An
averaging period of 30 days is considered long enough to yield data
representative of long-term performance, while also being short enough to
allow timely enforcement of the standards. In addition, use of a 30-day
rolling average (as opposed to a discrete average) allows enforcement of the
standard on a daily basis. To use the previous 30 calendar days could result
in an insufficient number of daily averages to comprise a statistically valid
30-day average if the unit has operated for only a few days during the 30-day
period.
2-11.2 Particulate Matter
1. Comment: Two commenters (IV-D-24, IV-D-33) objected to the
requirement that opacity monitors be installed on all coal-, oil-, or wood-
firing steam generating units with heat input capacities over 8.7 MW
(30 million Btu/hr). One commenter (IV-D-33) stated that operation of an
opacity monitoring system for consistently high quality data may be beyond
present capabilities at many smaller plants, especially where there are no
environmental engineers on site. The commenter stated that the EPA's
estimated annual operating cost of $8,000 seems unrealistically low, and that
the EPA should reconsider less costly alternatives for small units. The
commenter suggested periodic Method 9 observations as an alternative for small
steam generating units. The other commenter (IV-D-24) stated that CEMS for
opacity is an unnecessary burden for small steam generating units.
2-119
-------
Response: Opacity monitoring costs are based on a survey of several
instrumentation vendors, which indicated that the capital cost for an opacity
monitoring system (in June 1985 dollars) is about $59,000. Annual operating
and maintenance costs are about $8,000/year, resulting in total annualized
costs for opacity monitors of about $16,000/year. This annual cost of
operating a CEMS is considered reasonable given that continuous monitoring is
a necessary and reasonable method for enforcement personnel to ensure that
compliance with the opacity standard is being achieved on a continuous basis.
As discussed in the proposal preamble, units larger than 8.7 MW (30 million
Btu/hr) heat input capacity usually employ full-time, skilled operators so
that their employees would be able to operate and maintain the control
equipment necessary for meeting the opacity standard. Because units with heat
input capacity of 8.7 MW (30 million Btu/hr) or less are not as likely to
employ full-time, skilled operators, these units are not subject to any
opacity monitoring requirements.
2. Comment: One commenter (IV-D-24) stated that the cost of testing for PM
is approximately $5,000/test, and recommended that testing for PM be required
only at the time of startup for initial compliance purposes.
Response: Both the proposed and final NSPS require that coal- and wood-
fired units with greater than 8.7 MW (30 million Btu/hr) heat input perform a
compliance test for PM emissions at initial startup. (Units smaller than
8.7 MW [30 million Btu/hr] heat input are not covered by the NSPS for PM.)
The standard applies, however, to affected facilities at all times subsequent
to startup. Therefore, additional performance testing can be required at any
time at the discretion of the Administrator in enforcing the standard.
3. Comment: Six commenters (IV-D-19, IV-D-21, IV-D-35, IV-D-36, IV-D-38,
IV-D-39) expressed concern about applying CEMS for opacity to oil-fired steam
generating units. One commenter (IV-D-19) did not see the need or
justification for opacity monitoring on oil-fired stream generating units that
are used for district heating and cooling. This commenter's objections to the
2-120
-------
opacity monitoring requirements are based on the following points: (1) a
properly operating oil-fired unit will have minimal opacity emissions and,
although the EPA states that opacity monitoring is needed to determine periods
of incomplete combustion, the EPA identifies no environmental benefit
resulting from determination of periods of incomplete combustion; (2) the
proposed $03 standards essentially allow only distillate fuel oil to be used.
Consequently, emission limits will not be exceeded by oil-fired units;
(3) there is a limited number of oil-fired steam generating units used for
district heating and cooling. Steam generating units of the size covered by
Subpart DC are generally used as backup or emergency units and operate at
annual capacities 20 to 35 percent. They are often "last on and first off";
(4) the cost of installing, operating, and maintaining opacity monitors would
be disproportionate to the environmental benefit derived; and (5) steam
generating units of this size are often designed to share a stack with other
units that may not require opacity monitors. This commenter recommended that
the word "oil" be removed from Section 60.47c.
One commenter (IV-D-36) stated that a CEMS for opacity should not be
required for oil-fired units with annual capacity factors of 10 percent or
less. This commenter stated that the capital operating and maintenance costs
are not justified on low capacity units with low annual emissions.
Another commenter (IV-D-38) stated that the requirement for applying
CEMS for opacity on steam generating units in the 2.9 to 29 MW (10 to
100 million Btu/hr) range and that burn distillate oil or natural gas would
cause an added expense for his facility of $30,000 to $60,000. This commenter
stated that the requirement would not result in subsequent improvement of air
quality and, consequently, serves no purpose.
Two commenters (IV-D-21, IV-D-39) objected to the continuous opacity
monitoring requirement for distillate oil-fired units. One commenter
(IV-D-39) stated that a distillate oil-fired unit would have to be very poorly
operated in order to violate the 20 percent opacity limit. The commenter
continued that no cost-conscious operator would allow such inefficient
operations to continue for extended periods of time. He stated that the
continuous opacity monitor would cost $15,000 to $30,000/steam generating unit
stack. The other commenter (IV-D-21) stated that the expected capital cost of
2-121
-------
an opacity continuous monitoring device is at least $110,000, and the annual
operating and maintenance costs would be about $50,000. Both commenters
stated that the expense is not defensible for distillate oil-fired units on a
cost/ton basis, since it will not achieve any reduction in PM emissions.
One commenter (IV-D-35) stated that the proposed rule was an excessive
regulation and an unnecessary expense for small steam generating units. The
commenter explained that the proposed standard would require him to install an
opacity monitor on his 80,000 Ib/hr natural gas and distillate oil-fired steam
generating unit; this unit is intended to add emergency capacity to a coal-
fired heating plant. The commenter requested that the proposed standard be
modified to exclude opacity monitoring for gas- or distillate oil-fired units
with heat input capacities less than 29 MW (100 million Btu/hr).
Response: The intent of an opacity limit on small oil-fired steam
generating units is to identify units operating with incomplete combustion and
to allow appropriate remedial and enforcement actions to be taken to improve
the operation of those units. Because distillate oil is a very clean burning
fuel, with a relatively low ash content, steam generating units firing
distillate oil typically require little maintenance to ensure proper
combustion. Because violations of the opacity standard are not expected to
occur at distillate oil-fired units, these units are being exempted from the
opacity monitoring requirements in the final standards.
The combustion of residual oil in a small steam generating unit is more
likely to result in opacity problems that the combustion of distillate oil.
The higher ash content of residual oil and the carbonaceous material that
forms when residual oil does not combust completely results in residual oil-
fired units having much higher maintenance requirements than distillate oil-
fired units to achieve low opacity. In addition, because residual oil has a
higher ignition temperature and lower viscosity than distillate oil, it is
much more difficult to achieve and maintain proper combustion in a residual
oil-fired unit. Therefore, because residual oil-fired steam generating units
are more likely to lead to opacity exceedances if not carefully maintained,
residual oil-fired units are not being exempted from the opacity requirements.
The opacity monitor will identify units operating with incomplete combustion
2-122
-------
and indicate to the source owner or operator the need for some maintenance
activity.
4. Comment: One commenter (IV-D-30) questioned whether the benefit of
opacity monitors justifies the high costs of the monitors. He stated that
continuous monitors are costly from both initial cost and maintenance
standpoints. He said that operators with the necessary skills and training to
maintain these instruments are frequently unavailable in remote locations. He
noted that there is minimal additional benefit derived from continuous
monitoring versus periodic compliance testing. He recommended deleting the
requirements for continuous monitoring.
This commenter also pointed out a technical concern regarding continuous
opacity monitors used on units with wet scrubbers. He stated that opacity
monitoring of the wet plume from these sources is not technically feasible
unless the units install a wet flue gas reheat system. The commenter
recommended that a specific exemption be granted from continuous opacity
measurements for units using wet scrubbers.
Response: It is reasonable to require opacity CEMS for wood-, residual
oil-, and coal-fired steam generating units because it is the only way to
provide for continuous enforcement capabilities in determining compliance with
the standards. The additional costs of training for operation of this
instrumentation are not unreasonable in light of the increased enforcement
capabilities obtained.
Section 60.13(i)(l) of the General Provisions provides for alternative
monitoring requirements when installation of a CEMS would not provide accurate
measurements due to water or other interferences. This alternative monitoring
provision would apply to the situation of determining opacity if a wet plume
interferes.
2-123
-------
2.11.3 Nitrogen Oxides
1. Comment: One commenter (IV-D-37) stated that since the EPA has
consistently insisted that States, as part of the SIP regulations, have
specific requirements for monitoring, recordkeeping, and performance tests,
the EPA should do the same when promulgating their own regulations.
Response: The proposed NOX standards are being deleted from the final
standards for reasons discussed in Section 2.5. Because no NOX standards are
being promulgated, no monitoring, recordkeeping, or performance tests will be
necessary for NOX.
2.12 REPORTING AND RECORDKEEPING
2.12.1 Sulfur Dioxide
1. Comment: Two commenters (IV-D-21, IV-D-24) objected to the proposed S02
reporting and recordkeeping requirements. One commenter (IV-D-21) stated that
the proposed standard should exempt distillate oil-fired steam generating
units from $03 performance testing and monitoring requirements. The commenter
found no benefit in submitting quarterly reports certifying fuel quality
because these records are retained at the plant. Another commenter (IV-D-24)
stated that reporting of fuel sulfur limits should be required only with each
fuel delivery.
Response: Small steam generating units that combust distillate oil are
not subject to performance testing and monitoring requirements. As stated in
Sections 60.44c(j) and 60.48c(f) of the proposed standard, these units are
required only to maintain and report records of fuel supplies used. Quarterly
reporting of fuel supplies is a reasonable way for enforcement personnel to
determine that a distillate oil-fired unit is in compliance. Quarterly
reporting would not be a burdensome requirement for a facility firing
distillate oil because the report would be only a certification summarizing
2-124
-------
readily available records. This quarterly reporting would result in greater
enforcement capability because the enforcement agency could monitor compliance
at all facilities, instead of only the few that are inspected each year.
As the second commenter suggests, the standard requires reporting of
fuel sulfur limits only with each fuel delivery. In most cases, however,
units require more than one fuel shipment every three months. Therefore, if
required to report quarterly, most units would have multiple shipments to
report.
2. Comment: One commenter (IV-D-21) recommended that the language
requiring compliance with a single ASTM fuel specification (ASTM-D-396-78) be
changed and the definition of distillate oil be modified to include all
distillate oils with a sulfur content less than or equal to 0.5 percent by
weight because sulfur content is the criterion which affects S02 and PM
formation. The commenter believed that compliance with ASTM-D-396-78 is
unnecessary as long as the distillate oil contains less than 0.5 percent
sulfur. .
Response: For purposes of this regulation, distillate oil is defined by
the ASTM fuel specification because it is a widely recognized and universally
accepted definition. Defining distillate oil by the ASTM specification is
reasonable because it saves the steam generating unit operator the costs of
fuel sampling to determine compliance. The standard, however, does not
require compliance with the ASTM fuel specification as a condition for burning
oil. Unit operators burning oil that does not have the ASTM certification are
subject to the standard for residual oil (0.5 percent sulfur) and can, as the
commenter suggested, demonstrate compliance by fuel sampling. For these
reasons, the definition of distillate oil is not being changed. To adopt the
suggested revision would increase the cost of complying with the NSPS by
requiring all oil to be sampled for sulfur for no apparent beneficial reasons.
2-125
-------
2.12.2 Particulate Matter
1. Comment: One commenter (IV-D-06) stated that quarterly rather than
semi-annual reporting of no excess opacity readings should be required in
Section 60.48c(c), since the reports would contain other information besides
emissions levels that should be monitored quarterly. The commenter
recommended quarterly reporting of all opacity readings so that the reviewing
agency could detect possible problems with the CEMS within a reasonable time.
Response: As a means of minimizing the reporting burden associated with
compliance, semi-annual rather than quarterly reporting is permitted in cases
where there are no monitored exceedances of an opacity standard. This
reporting frequency is considered adequate and there is no reason to change
this policy for the small steam generating unit NSPS.
2.13 WORDING OF REGULATION
1. Comment: One commenter (IV-D-03) stated that proposed
Section 60.45c(b)(7) should be changed to read "Method 9 and the procedures of
Section 60.lid shall be used to determine opacity," to be consistent with
other similar provisions. (Commenter's suggested addition is underlined.)
Response: Section 60.lid is a part of the General Provisions to the
NSPS rules. The General Provisions (including Section 60.lid) apply to all
NSPS unless a specific exception is stated in a particular NSPS. Therefore,
there is no need to state that 60.lid applies to small steam generating units.
2. Comment: One commenter (IV-D-03) cited the first paragraph of proposed
Section 60.46c(b) and questioned whether the first occurrence of the word
"outlet" is in error and if the word "inlet" should be substituted.
2-126
-------
Response: The first occurrence of the word "outlet" in the first
paragraph of proposed Section 60.46c(b) is correct as written. The wording in
the regulation has been revised to make the meaning clear.
3. Comment: One commenter (IV-D-09) was concerned that a reading of
proposed Section 60.46c(b)(2) would imply that even if distillate oil were
burned, the sulfur content of the feed tank would have to be tested after each
new shipment. He recommended that the following wording be used (new wording
underlined): "For affected facilities combusting only oil, other than
distillate oil. oil samples shall be collected. ..."
Response: The commenter is correct in that units firing distillate oil
are not required to collect oil samples after each new shipment of oil is
received. For distillate oil-fired units, maintenance of fuel supplier
certifications is sufficient for demonstrating compliance. Section 60.46c has
been modified in the final rule to clarify these requirements.
4. Comment: One commenter (IV-D-20) stated that the S02 compliance and
monitoring provisions should be clarified in several respects in order to be
consistent with the EPA's intent underlying these proposals.
First, in Section 60.46c(b) of the proposed rule, the commenter stated
that the "and" in the phrase "pursuant to paragraph (b)(l) and (b)(2) of this
section" should be "or." That is, for oil-fired units, (b)(l) and (b)(2) are
alternative monitoring approaches; the "and" suggests that both approaches
must be used. The commenter also stated that (b)(2) should clarify the
procedure as it applies to low capacity factor units that must use 24-hour
averaging.
Second, the commenter found Section 60.44c (the compliance provisions)
confusing as applied to units firing very low sulfur oil. The commenter
pointed out that under paragraph (j), units firing very low sulfur oil can
avoid the performance testing specified under (c)(4); yet, paragraph (j) is
unclear on how compliance is to be determined for such units. The commenter
2-127
-------
argued that based upon the preamble and other rules, it appears that the EPA
intended for units burning distillate oil simply to certify that distillate
oil is being used. [See Section 60.48c(f)(2) of the proposed rule.] For
units burning residual oil, it appears that fuel sampling and analysis
pursuant to the procedures spelled out in Section 60.46c(b)(l) or (b)(2) would
be appropriate. Accordingly, the commenter suggested that paragraph (j) be
amended to address these issues.
Third, the commenter recommended that proposed Section 60.44c(d) also
needs clarification. In particular, although the section appears to require
24-hour averaging for low annual capacity factor units (10 percent or less),
its failure to reference Section 60.46c(b) leaves some questions on how to
conduct compliance tests when using any of the available methods (CEMS,
Method 6B, or fuel sampling and analysis).
Response: The commenter is correct that for purposes of emission
monitoring for S02, the proposed regulation requires either the Method 19
approach in paragraph (b)(l) or the fuel sampling approach in paragraph (b)(2)
and not both approaches. To clarify this provision, the final rule will
incorporate the commenter's suggestion to replace the "and" with "or."
Because the final rule will not require 24-hour averaging for low capacity
units (see Section 2.11.1, Comment 3), a clarification of this procedure is
not necessary in paragraph (b)(2).
With regard to the second comment, the commenter is correct in
interpreting the compliance provisions for oil-fired units. Owners or
operators of distillate oil-fired units are required only to retain records
certifying that distillate oil is used. Only owners or operators of residual
oil-fired units are required to collect and analyze fuel samples. To avoid
any confusion, the final rule has been modified to clarify the intended
meaning.
With regard to the third comment, Section 60.44c(b) of the proposed rule
clearly stated that compliance is determined on a 30-day rolling average no
matter which monitoring method is selected. As discussed above, 24-hour
averaging is not required in the final rule.
2-128
-------
5. Comment: One commenter (IV-D-24) disagreed with definitions in the
proposed regulation that describe fuel oil in terms of the amount of nitrogen
it contains. The commenter suggested that the reference to fuel nitrogen
content be deleted from the definition of distillate oil. In the definition
for residual oil, the commenter recommended deleting the fuel oil nitrogen
requirement and defining residual oil on the basis of ASTM grades 4, 5, or 6
and fuel sulfur content.
Response: Fuel nitrogen content is not used in the definitions of
either distillate or residual oil. Distillate oil is defined on the basis of
ASTM grades 1 or 2 and residual oil on the basis of ASTM grades 4, 5, or 6.
6. Comment: One commenter (IV-D-45) stated that the definition of "coal
refuse" in the proposed 40 CFR 60.41c should be clarified to include coal
refuse-derived fuels such as beneficiated culm.
Response: No additional clarification is necessary. The definition of
"coal refuse" in the proposed and final standards includes anything that is a
by-product of coal mining or coal cleaning operations with an ash content
greater than 50 percent, by weight, and a heating value less than
13,900 kilojoules per kilogram (kJ/kg) (6,000 Btu/lb) on a dry basis.
7. Comment: One commenter (IV-D-06) stated that the proposed
Section 60.48c(d) apparently allows S02 continuous emissions monitoring system
reporting for time periods other than calendar quarters. He recommended that
to allow for consistent comparison of data, calendar quarters should be
specified.
Response: The proposed SOg standard requires only that quarterly
reports be submitted; the standard does not specify that these reports have to
be submitted on specific calendar quarters. However, a State that has been
2-129
-------
delegated the authority to implement and enforce NSPS has the authority to
require specially tailored reporting requirements as long as they meet the
minimum requirements of the NSPS. If a State wishes to adopt reporting
requirements specifying that reports must be submitted on specific calendar
quarters, then the State may do-so.
8. Comment: One commenter (IV-D-15) stated that Section 60.44c(a) of the
proposed rule, which reads that "the sulfur dioxide standards under
Section 60.42c apply at all times, including periods of startup, shutdown, or
malfunction," does not adequately address those affected facilities that burn
a combination of fuel oil and fuel gas and that are subject to the S(>2
standards of Subpart J. Section 60.40c(c) of the proposed rule stated that
these affected facilities are subject to the S02 standard of Subpart J. This
commenter recommended that Section 60.44c(a) be amended to exempt affected
facilities that are also Subpart J facilities. The commenter added that this
exemption would be consistent with the Subpart DC Section 60.45c(a) exemptions
from the particulate matter (PM) emission and opacity standards.
Response: Section 60.40c(c) of the proposed rule stated: "Affected
facilities that also meet the applicability requirements under Subpart J ...
are subject to ... the sulfur dioxide standards under Subpart J ..." (emphasis
added). This paragraph was intended to clarify that affected facilities
subject to Subpart DC are not exempt from Subpart J, but must also meet the
S0£ standards under Subpart J. If the facility fires only fuel gas, there are
no emission limits under Subpart DC, but the facility must comply with
Subpart J. If the facility fires only fuel oil, it is subject to Subpart DC
and not Subpart J. If the facility fires both fuel gas and fuel oil, it is
subject to both Subparts DC and J. This is also true for Section 60.40c(b) of
the proposed rule as it refers to Subpart E. Affected facilities may be
subject to more than one subpart. Rather than list all of the subparts that
Subpart DC facilities may also be subject to, Sections 60.40c(b) and (c) have
been deleted from the final rule.
2-130
-------
2.14 MISCELLANEOUS
1. Comment: One commenter (IV-D-25) was concerned that when evaluating the
availability of low sulfur fuels or natural gas to meet the proposed NSPS, the
impact of replacing a unit supplied by a fuel distribution system was not
considered thoroughly. The commenter pointed out that the cost of providing
natural gas or very low sulfur oil to one new source may be small; however,
the cost of providing this same type of fuel to all the other units on a fuel
distribution system could be great. The commenter stated that, due to space
and access restrictions, it is often not practical to install new fuel
supplies near a replaced steam generating unit.
Response: If a new steam generating unit is added to or replaced in a
matrix of units, the new unit does not have to be connected to the same fuel
distribution system as the existing units. In fact, if the existing units are
combusting fuel that does not comply with the NSPS, the new unit would have to
install a separate fuel storage handling system. For either a new unit at a
new plant or for new and replacement units at existing plants, the additional
cost of a separate fuel storage and handling system to house a different type
of fuel was included in the cost estimations, and those impacts are considered
reasonable.
If, on the other hand, very low sulfur fuel oil is used for all units,
new and existing, a separate fuel storage and supply system would not be
needed. In this case, additional reductions in S0£ emissions would be
achieved because S02 emissions from existing units would also be reduced. In
light of these additional S02 emission reductions, the cost impacts of firing
very low sulfur fuel oil in all units (both new and existing) are considered
reasonable.
2. Comment: One commenter (IV-D-27) cited as incorrect the statement in
the preamble to the proposed standards reading "(C)oal markets supplying coals
with low sulfur contents (520 ng/J [1.2 Ib/million Btu]). . . ." It should
2-131
-------
instead read 0.6 to 0.9 percent sulfur coal, depending on the Btu value of the
coal."
Response: The intent of the preamble statement was that the coals
supplied by these coal markets would be coals with a potential to emit
520 ng/J (1.2 Ib S02/million Btu) or less when combusted. Although the
commenter is correct that this may represent a range of fuel sulfur content,
the relevant unit of measure from the point of view of emissions potential is
mass/unit of heat input.
3. Comment: Two commenters (IV-D-11, IV-D-12) stated that based on the
California standard (the generally accepted industry standard), 10,000 Btu is
required to produce 1 KW. In the proposed standards, however, 29 MW equals
100,000,000 Btu. According to the commenter, 10,000,000 Btu produces 1,000 KW
and 100,000,000 Btu produces 10,000 KW (not 2.9 and 29, respectively, as the
proposed standards state).
Response: The applicability of the proposed (and final) standard is
based on the heat input, expressed as either MW or Btu/hr, and does not
consider electric output. One hundred million Btu/hr of heat input is
equivalent to 29 MW of heat input using standard conversions. Similarly, if
30 percent efficiency is assumed, 29 MW of heat input would produce
approximately 10 MWe of electrical output, or 10,000 KWe. Using these
standard conversion factors, the NSPS is consistent with the California
standard cited by the commenter.
4. Comment: One commenter (IV-D-45) stated that energy efficiency should
be encouraged by setting standards based on energy output rather than input.
This would encourage energy efficiency in accordance with the EPA draft policy
on pollution prevention and national energy goals. (The commenter offered
assistance in ensuring that future standards recognize energy efficiency.)
2-132
-------
Response: Emission limits for small steam generating units are based on
energy input because, under Section 111 of the CAA, NSPS must be based on the
use of best demonstrated technology to reduce air pollution emissions. Best
technology for steam generating units is not a function of energy efficiency.
These technologies operate equally well whether the steam generating unit
operates at high or low efficiency. Efficiency does, however, affect
emissions. Under the final rule, as combustion efficiency is increased,
emissions would decrease because less fuel would be fired. Output-based
standards, however, would allow total mass emissions to increase to the level
that would have occurred had the increase in efficiency not taken place. As a
result, output-based standards eliminate the environmental benefit, in terms
of reduced air pollution, that otherwise would occur from high efficiency
operation. Consequently, output-based standards would result in increased
emissions.
A number of factors influence the decision to design and operate high
efficiency steam generating units, including unit life expectancy, maintenance
requirements, and the cost of fuel. The cost savings associated with relaxed
emission control for achieving increased efficiency is relatively
insignificant compared with these other considerations and is not likely to
play a substantial role in the decision.
Moreover, standards based on heat output could result in increased
control and, therefore, increased costs, for some steam generating units. If
a credit for increased efficiency were incorporated into the current NSPS,
then it would be necessary to "reference" the standard to some standard
efficiency. A unit operating above this selected standard efficiency would be
permitted to achieve a less stringent standard; however, a unit operating
below the standard efficiency would have to meet increased emission control
requirements. Because there are a number of good operational reasons why a
small steam generating unit operator may opt for a lower efficiency unit, an
NSPS based on efficiency could well be regarded as inequitable for some units.
In summary, a standard based on heat output clearly would result in
increased emissions. It is not clear, however, that this would result in a
significant cost savings or any significant inducement for owners to build
more efficient units. Other reasons, such as the cost of fuel, provide
2-133
-------
of
0" "- output ^ ' "-"' '""
to offset this increase in emisslo s Th? I' *" "° 'PParent """"^
=nd environmenta, impacts ch nq s I t^T '""'^n, cost, energy,
not granted ' f0>1nat °f the pro'!osed «*"
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-90-016
2.
3. RECIPIENTS ACCESSION NO.
4. TITLE AND SUBTITLE
Small Industrial-Commercial-Institutional Steam
Generating Units -- Background Information for
Promulgated Standards
5. REPORT DATE
August 1990
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Standards Division
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park. North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-4378
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air and Radiation
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document summarizes EPA's response to public comments
received on proposed new source performance standards for
sulfur dioxide and particulate matter emissions from new
small industrial-commercial-institutional steam generating
units (54 FR 24792, June 9, 1989).
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Pollution Control
Standards of Performance
Steam Generating Units
18. DISTRIBUTION STATEMENT
Release unlimited
Small Boilers
Industrial Boilers
Air Pollution Control
Unclassified
EPA Form 2220-1 (R«v. 4-77) PREVIOUS EDITION is OBSOLETE
-------
U.S.Envlfonrr^nta» Protection Ai«»cy
Region 5, Library (PL-12J)
77 West Jackson'Boulevard, 12th Floor
Chicago, IL 60604-3590
------- |