vxEPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-82-020a
July 1983
Air
Electric Arc
Furnaces and
Argon-Oxygen
Decarburization
Vessels in
Steel Industry —
Background
Information for
Proposed Revisions
to Standards
Draft
EIS
-------
-------
EPA-450/3-82-020a
Electric Arc Furnaces and
Argon-Oxygen Decarburization Vessels
in Steel Industry -
Background Information for
Proposed Revisions to Standards
Emission Standards and Engineering Division
U.S ENVIRONMENTAL PROTECTION AGENCY
Office of Air Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
July 1983
-------
This report has been reviewed by the Emission Standards and Engineering 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, N.C. 27711, or from National Technical
Information Services, 5285 Port Royal Road, Springfield, Virginia 22161
-------
ENVIRONMENTAL PROTECTION AGENCY
Background Information
and Draft
Environmental Impact Statement
for Electric Arc Furnaces and
Argon-Oxygen Decarburizatioh Vessels in the Steel Industry
Prepared by:
Jack R. Farmer '
Director, Emission Standards and Engineering Division
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
(Date)
The proposed standards of performance would limit emissions of
part icul ate matter from new, modified, and reconstructed electric arc
furnaces and argon-oxygen decarburization vessels in the steel
industry. 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. The comment period for review of this document is 60 days. Mr. ^
Crowder may be contacted regarding the date of the comment period.
'3. For additional information contact:
Mr. Jim Crowder
Industrial Studies Branch (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
telephone: (919) 541-5601.
4. Copies of this document may be obtained from:
U.S. EPA Library (MD-35)
Research Triangle Park, N.C. 27711
National Technical Information Service
5285 Port Royal Road
Springfield, Va. 22161
-------
-------
Section
List of
List of
1
1.1
1.2
1.3
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
3
3.1
3.2
3.3
3.4
4
4.1
4.2
4.3
TABLE OF CONTENTS
Title
Figures
Tables .
SUMMARY . '.
Regulatory Alternatives • •
Environmental Impact >
Economic Impact .
INTRODUCTION • •
Background and Authority for Standards
Selection of Categories of Stationary Sources . . .
Procedure for Development of Standards of
Performance
Consideration of Costs •
Consideration of Environmental Impacts
Impact on Existing Sources
Revision of Standards of Performance .......
ELECTRIC ARC FURNACES AND ARGON-OXYGEN
DECARBURIZATION VESSELS IN THE STEEL INDUSTRY:
PROCESSES AND POLLUTANT EMISSIONS . . . .
General •
Process Facilities and Their Emissions
Emissions
References for Chapter 3
EMISSION CAPTURE AND CONTROL TECHNIQUES
Introduction
Capture of EAF Process and Fugitive Emissions . . .
Capture of AOD Vessel Process and Fugitive
Emissions
Page
viii
X
1-1
1-1
1-2
1-2
2-1
2-1
2-5
2-6
2-9
2-10
2-11
2-11
3-1
3-1
3-15
3-35
3-44
4-1
4-1
4-2
4-19
-------
TABLE OF CONTENTS (continued)
Section Title Page
4.4 Fugitive Emissions Capture System Combinations . . 4-21
4.5 Exhaust Gas Cleaning Devices • . 4-24
4.6 Emission Source Test Data 4-29
4.7 References for Chapter 4 . 4-43
5 MODIFICATION AND RECONSTRUCTION ' . 5-1
5.1 Summary of Modification and Reconstruction
Provisions 5-1
5.2 Applicability to Electric Arc Furnaces and Argon-
Oxygen Decarburization Vessels in the Steel
Industry 5-3
6 MODEL PLANTS AND REGULATORY ALTERNATIVES ..... 6-1
6.1 Introduction 6-1
6.2 Model Plants 6-1
6.3 Regulatory Alternatives 6-15
6.4 References for Chapter 6 ; . 6-19
7 ENVIRONMENTAL IMPACTS 7-1
7.1 Air Pollution Impact 7-1
7.2 Water Pollution Impact 7-21
7.3 Solid Waste Disposal Impact 7-21
7.4 Energy Impact . 7-25
7.5 Other Environmental Impacts 7-27
7.6 Other Environmental Concerns 7-27
o
7.7 References for Chapter 7 „ . . . 7-28
8 COSTS ; . 8-1
8.1 Cost Analysis of Regulatory Alternatives 8-1
8.2 Other Cost Considerations 8-20
VI
-------
TABLE OF CONTENTS (continued)
Section
8 3
g
9 1
9 2
9.3
9 4
APPENDIX A-
APPENDIX B:
APPENDIX C:
C.I
C.2
C.3
C 4
r 5
APPENDIX D:
n i
'n 9
D.3
Title
References for Chapter 8
ECONOMIC IMPACTS . . .
Economic Impacts of Regulatory Alternatives . . .
EVOLUTION OF THE BACKGROUND INFORMATION
DOCUMENT
INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS . . ,
EMISSION TEST DATA FOR FABRIC FILTERS ON
ELECTRIC ARC FURNACES AND ARGON-OXYGEN
DECARBURIZATION VESSELS IN THE STEEL INDUSTRY . .
Emission Test Data for Fabric Filters on Electric
Arc Furnaces in the Carbon Steel Industry . . .
Emission Test Data for Fabric Filters on Electric
Arc Furnaces and Argon-Oxygen Decarburization
Vessels in the Specialty Steel Industry ....
Visible Emission Data for Dust-Handling Systems
for Both Carbon and Specialty Steel Plants . . .
English/Metric Conversions
EMISSION MEASUREMENT AND CONTINUOUS MONITORING . .
Emission Measurement Methods
Monitoring Systems and Devices
Performance Test Methods . . . . „ .
Page
8-22
9-1
9-1
9-3
9-59
9-119
A-l
B-l
C-l
C-2
C-16
C-22
C-22
C-47
D-l
D-2
D-18
D-26
vn
-------
No.
3-1
3-2
3-3
3-4
3-5
3-6
3-7
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
4-9
7-1
9-1
9-2
9-3
9-4
LIST OF FIGURES
Title
Flow Diagram of an Iron and Steel Plant
Schematic Flowchart for Integrated and Nonintegrated
Steel making
Steel Production by Furnace Type
Electric Arc Steel Furnace . . .
Material Balance of Electric Arc Furnace Based on 1,000 kg
of Steel Produced • .
Argon-Oxygen Decarburization Vessel
Carbon-Chromium Equilibrium Curves
Direct-Shell Evacuation Control (Two Views) . . . .
Side Draft Hood (Two Views)
Partial Furnace Enclosure
Total Furnace Enclosure at Lone Star Steel Company
Canopy Hood Capture System
Diverter Stack With Canopy Hood
Close-Fitting Hood With Canopy Hood
Summary of Particulate Matter Source Data for Carbon Steel
EAF Fabric Filters (Reference Method 5)
Summary of Particulate Matter Source Data for Specialty
Steel Shop Fabric Filters (Reference Method 5) ...
Typical Furnace Building and Fabric Filter Layout (With
Roof Monitor Open Across Furnace Building)
Supply Schedules for Constructed and Unconstructed Plants
NSPS Effects on Market Price and Quantity
NSPS Effects on the Average Total Cost of New Plants,
Equilibrium Price and Equilibrium Quantity . .
Current Steel Market Disequilibrium and Long-Run
Equilibrium . . .
Page
3-2
3-3
3-14
3-17
3-26
3-29
3-32
4-4
4-6
4-8
4-10
4-13
4-20
4-22
4-31
4-37
7-10
9-93
9-94
9-100
9-109
-------
LIST OF FIGURES (continued)
No. Title
9-5 NSPS Effects on U.S. Steel Market Long-Run Equilibrium . . .
C-l Summary of Particulate Matter Source Test Data From Fabric
Filters on EAF's at Carbon Steel Shops .
C-2 Summary of Particulate Matter Source Test Data for
Specialty Steel Shop Fabric Filters .
D-V Particulate Sampling Train at Plant Q Outlet ........
D-2 Particulate Sampling Train at Plants A and Q Inlets . > .• .
D-3 Fabric Filter Outlet at Plant P
D-4 Sampling Location at the Fabric Filter Outlet at Plant P . .
Page
9-111
C-3
C-17
D-5
D-6
D-8
D-9
IX
-------
LIST OF TABLES
No.
1-1
»
1-2
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
3-12
4-1
4-2
4-3
4-4
4-5
Title
Environmental and Economic Impacts of Various Alternatives
Compared to Alternative 1 (Baseline) in the Fifth Year
(1987)
Matrix of Environmental and Economic Impacts for
Regulatory Alternatives
Electric Arc Furnaces in the United States — 1981 . . . . .
Location of Steel Plants With EAF's and AOD Vessels—
1981
Industry Percentages of EAF's and AOD Vessels by
Capacity— 1981
Argon-Oxygen Decarburization Vessels in the United
States—I 981
Steel Industry Data and Projections
Raw Materials and Products-- EAF and AOD Operations ....
Particulate Matter Emission Factors (Uncontrolled) ....
Trace Constituent Emission Factors (Uncontrolled) ....
Chemical Analysis of Electric Arc Furnace Dust by Phase of
Furnace Operation
Exhaust Gas Particulate Matter Composition
Size Distribution of Particulate Matter Emissions From
Steel making EAF and AOD Facilities
Summary of State Air Pollution Regulations ...
Fugitive Emissions Capture Technology Combinations
(Capture and Specialty Steel EAF)
Fugitive Emissions Capture Technology Combinations
(Specialty Steel AOD)
Summary of Carbon Steel Plant Data .....
Summary of Visible Emission Data from Fabric Filters on
EAF's at Carbon Steel Shops
Summary of Opacity Data From Shop Roof Monitors on Carbon
Steel EAF Shops
Page
1-3
1-4
3-6
3-9
3-10
3-11
3-13
3-27
3-37
3-38
3-39
3-40
3-41
3-43
4-23
4-25
4-30
4-33
4-34
-------
LIST OF TABLES (continued)
No.
4-6
4-7
4-8
4-9
4-10
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-$
6-9
6-10
6-11
7-1
7-2
7-3
Title
Summary of Specialty Steel Plant Data
'Summary of Visible Emission Data From Fabric Filters on
EAF's and AOD's at Specialty Steel Shops
Summary of Visible Emission Data From Shop Roof Monitors
on Specialty Steel Shops
Summary of Visible Emission Data From Dust-Handling
Systems at EAF and ADD Vessel Steel Mill Facilities . .
Summary of Trace Constituent Concentrations Analysis . . .
Model Plants . . . .
Model Furnace Parameters—Carbon Steel . . . .
Model Furnace Parameters—Specialty Steel ........
Model Plants With Capture Configuration Options
Air Flow Rates Per Unit of Furnace/Vessel Capacity . . . .
Model Furnace Parameters—Carbon Steel Fabric Filter
Information (Metric Units) .
Model Furnace Parameters—Carbon Steel Fabric Filter
Information (English Units)
Model Furnace Parameters—Specialty Steel Fabric Filter
Information (Metric Units) ...
Model Furnace Parameters—Specialty Steel Fabric Filter
Information (English Units)
Regulatory Alternatives—Carbon and Specialty Steel EAF .
Regulatory Alternatives—Specialty Steel AGO Vessel . . .
Annual Particulate Matter Emissions for Each Model Furnace
and Regulatory Alternative ....
Annual Particulate Emission Reduction Below Uncontrolled
and Baseline Levels for Each Model Furnace/Vessel Size .
Projected Construction of Model EAF/AOD Plants Industry
Wide, 1983-1987
Page
4-36
4-38
4-39
4-41
4-42
6-2
6-4
6-5
6-7
6-8
6-9
6-10
6-11
6-13
6-17
6-18
7-2
7-3
7-5
-------
LIST OF TABLES (continued)
No. Title
7-4
7-5
7-6
7-7
7-8
Total Particulate Matter Emissions for New Shops Added
in 1983-1987 and 5-Year Cumulative
General Data and Options of ISCST Model Used in Atmospheric
Disperson Modeling Analysis
Modeling Data for Fabric Filter Sources
Modeling Data for Roof Monitor Sources .
Highest Second-Highest 24-Hour Particulate Matter
Concentration Impacts From Fabric Filter Sources . . . .
Page
7-6
7-7
7-11
7-12
7-16
Highest Second-Highest 24-Hour Particulate Matter
Concentration Impacts From Roof Monitor Sources
Maximum Annual Arithmetic Mean Particulate Matter
Concentration Impacts From Fabric Filter Sources .
Maximum Annual Arithmetic Mean Particulate Matter
Concentration Impacts From Roof Monitor Sources .
7-18
7-19
7-20
7-9
7-10
7-11
7-12 Solid Waste (Fabric Filter Catch) Generation for Each
Model Plant 7-22
7-13 Summary of Potential Industry Wide Solid Waste Generation
From Fabric Filters—Carbon Steel Industry 7-23
7-14 Summary of Potential Industry Wide Solid Waste Generation
From Fabric Filters—Specialty Steel Industry ...... 7-24
7-15 Industry Wide Electrical Energy Requirements for 1983-1987
and 5-Year Cumulative 7-26
8-1 Capital and Annualized Costs for Uncontrolled Model
Plants 8-2
8-2 Component Capital Cost Factors for Fabric Filters as a
Function of Equipment Cost—New Facilities 8-4
8-3 Basis for Estimating Annualized Cost—New Facilities . . . 8-5
8-4 Capital and Annualized Costs of Pollution Control
Equipment—EAF/AOD Process and Fugitive Emissions . . . 8-6
8-5 Capital Costs for Fabric Filter Control Method—New
Carbon Steel Facilities ' 8-8
xn
-------
LIST OF TABLES (continued)
No.
8-6
8-7
8-8
8-9
8-10
8-11
9-1
9-2
9-3
9-4
9-5
9-6
9-7
9-8
9-9
9-10
9-11
9-12
9-13
Title
Capital Costs for Fabric Filter Control Method— New
Specialty Steel Facilities
Annual ized Costs for Fabric Filter Control Method— New
Carbon Steel Facilities .
Annual ized Costs for Fabric Filter Control Method— New
Specialty Steel Facilities
Projected Nationwide Control System Expenditures Under
Various Regulatory Alternatives (by 1987)
Average Cost Effectiveness of EAF/AOD Process and Fugitive
Emissions Control Over Baseline .
Cost Estimates for OSHA Compliance— EAF/AOD Vessel
Facilities • •
Important Inputs to the Blast Furnaces and Steel Mills
Industry— SIC 3312
Raw Steel Production by Process Type
Pig Iron and Scrap Inputs to Raw Steel Production ....
U.S. Real Gross National Product and Apparent Consumption
of Steel Mill Products
Important Purchasers of Output From the Blast Furnaces and
Steel Mills Industry— SIC 3312
Plant Integration •
After-Tax Profit to Stockholders' Equity ....
U.S. Companies Operating Electric Arc Furnaces: Financial
Data if Available (1979)
Financial Ratios for U.S. Companies Operating Electric
Arc Furnaces (1979)
Financial Ratios for Selected Industries (1979)
Steel Mill Products and Total Industrial Output Indexes .
Real Value of Output for SIC 3312
Steel Price Index and GNP Price Deflator
Page
8-10
8-13
8-15
8-17
8-18
8-21
9-9
9-13
9-15
9-17
9-18
9-22
9-31
9-33
9-36
9-37
9-39
9-40
9-42
XTU
-------
No.
9-14
9-15
9-16
9-17
9-18
9-19
9-20
9-21
9-22
9-23
9-24
9-25
9-26
9-27
9-28
9-29
9-30
9-31
LIST OF TABLES (continued)
Title
Indexes of Real New Investment
Index of Output per Employee-Hour
Steel Mill Products Trade
Projected Domestic Raw Steel Production
Summary of Electric Arc Furnace Capacity Projection
Projected Construction of Electric Furnace Plants,
1982-1987
Average Total Cost Impacts Assuming 6.2 Percent Weighted
Average Cost of Capital .
Total Annualized Cost of Compliance Above Baseline
as a Percent of Annual Plant Sales
Compliance Capital Cost Above Baseline as a Percent of
Baseline Plant Capital Cost
Model Plant Cost Data
Model Parameter Values
Project Net Present Values Assuming 6.2 Percent Weighted
Average Cost of Capital . . .
Net Present Value Impacts Assuming 6.2 Percent Weighted
Average Cost of Capital
Project New Present Values Assuming 10.0 Percent Weighted
Average Cost of Capital
Net Present Value Impacts Assuming 10.0 Percent Weighted
Average Cost of Capital
Average Total Cost Assuming 6.2 Percent Weighted
Average Cost of Capital
Average Total Cost Impacts Assuming 6.2 Percent Weighted
Average Cost of Capital . .
Average Total Cost Assuming 10.0 Percent Weighted
Average Cost of Capital
9-43
9-45
9-46
9-47
9-49
9-51
9-54
9-55
9-57
9-68
9-69
9-72
9-74
9-75
9-76
9-78
9-80
9-81
XIV
-------
LIST OF TABLES (continued)
No.
9-32
9-33
9-34
9-35
9-36
9-37
9-38
9-39
9-40
9-41
9-42
9-43
A-l
B-l
C-l
C-2
C-3
C-4
C-5
Title
Average Total Cost Impacts Assuming 10.0 Percent Weighted
Average Cost of Capital
Domestic Steel Production Impacts for 1987
Domestic Steel Industry Employment Impacts for 1987 .
Steel Imports Impacts for 1987 . . . .
Net Present Value and Average Total Cost Data for Model
Plants 4 and 7
Summary of Economic Impacts From Baseline to Regulatory
Alternative 2
Summary of Economic Impacts From Baseline to Regulatory
Alternative 3 ......
Capital Requirements of Regulatory Alternatives
Industry Debt Ratios . .
Total Cost of Regulatory Alternative 2 in 1987 ......
Total Cost of Regulatory Alternative 3 in 1987
Total Cost of Regulatory Alternatives in the Very Long Run
Evolution of the Background Information Document . . . . ,
Cross-indexed Reference System to Highlight Environmental
Impact Portions of the Document
Summary of Plants Tested and Type of Tests Performed .
Summary of Visible Emission Data From Fabric Filters on
EAF's at Carbon Steel Shops
Summary of Opacity Data From Shop Roof Monitors at Carbon
Steel EAF Shops
Summary of Particulate Matter.Results—Plant A (Fabric
Filter Inlet) . . . .
Summary of Particulate Matter Results—Plant A (Fabric
Filter Outlet) .
9-82
9-85
9-86
9-88
9-90
9-102
9-104
9-106
9-107
9-115
9-116
9-118
A-2
B-2
C-23
C-24
C-25
C-26
C-27
XV
-------
LIST OF TABLES (continued)
No. Title
C-6 Summary of Particulate Matter Results—Plant B (Fabric
Filter Outlet—Canopy Hood))
C-7 Summary of Particulate Matter Results—Plant B (Fabric
Filter Outlet—Side Draft Hood)
C-8 Summary of Particulate Matter Results—Plant C (Fabric
Filter Outlet)
C-9 Summary of Particulate Matter Results—Plant D (Fabric
Filter Outlet)
C-10 Summary of Particulate Matter Results—Plant E (Fabric
Filter Outlet)
C-ll Summary of Particulate Matter Results—Plant F (Fabric
Filter Outlet)
C-12 Summary of Particulate Matter Results—Plant G (Fabric
Filter Outlet
Page
028
029
030
O31
C-32
C-33
C-34
C-13 Summary of Particulate Matter Results—Plant H (DEC Fabric
Filter Outlet c"35
C-14 Summary of Visible Emission Data From Fabric Filters at
Specialty Steel Shops
C-36
C-15 Summary of Visible Emission Data From Shop Roof Monitors on
Specialty Steel Shops ; c"37
C-16 Summary of Particulate Matter Results—Plant P (South
Fabric Filter Inlet) .... C-38
C-17 Summary of Particulate Matter Results—Plant P (North
Fabric Filter Inlet) C-39
C-18 Summary of Particulate Matter Results—Plant P (Fabric .
Filter Outlet) .....' .••'•' C"40
C-19 Summary of Particulate Matter Results—Plant Q (Fabric
Filter Inlet) c~41
C-20 Summary of Particulate Matter Results—Plant Q (Fabric
Filter Outlet) C-42
C-21 Summary of Particulate Matter Results—Plant Q (Fabric :
Filter Outlet) C-43
xvi
-------
LIST OF TABLES (continued)
No. Title
C-22 Summary of Particulate Matter Results—Plant R (Fabric
Filter Inlet) . .• C-44
C-23 Summary of Particulate Matter Results—Plant R (Fabric
Filter Outlet) . , C-45
C-24 Summary of Visible Emission Data From Dust-Handling Systems
at EAF and AOD Vessel Steel Mill Facilities . C-46
xvi i
-------
-------
1. SUMMARY
1.1 REGULATORY ALTERNATIVES
Standards of performance for new stationary sources are developed
under Section 111 of the Clean Air Act (42 U.S.C. 7411), as amended.
Section 111 requires the establishment of standards of performance for
any new stationary source which ". : . causes, or contributes significantly
to air pollution which causes or contributes to the endangerment of
public health or welfare." The Act requires standards of performance
for such sources to ". . . 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 standards
apply only to stationary sources, the construction, modification, or
reconstruction of which starts after regulations are proposed in the
Federal Register.
Three regulatory alternatives were selected for study. The first
alternative would require no additional Federal regulatory action. The
existing new source performance standards (NSPS) for electric arc furnaces
(EAF's) would be applicable to new, modified, or reconstructed EAF
sources and only State and local regulations would be applicable to new,
modified, or reconstructed argon-oxygen decarburization (ADD) vessel
sources. This alternative is considered to be the baseline condition
from which the impacts of the other alternatives are calculated. The
second and third alternatives would require Federal regulatory action
and would place a more stringent limitation on the allowable levels of
particulate matter and visible emissions than that allowed by the baseline
1-1
-------
condition. The same emission control systems are used for the second
and third altneratives as for the first, but improvements in the emission
capture systems would be required.
1.2 ENVIRONMENTAL IMPACT
The beneficial and adverse environmental impacts associated with
each regulatory alternative over the baseline levels for the carbon and
specialty steel shops are summarized in Table 1-1. As shown, a moderate-
to-large beneficial air impact would result from adoption of Regulatory
Alternative 2 or 3, and solid wastes would increase negligibly. The
percent increase in solid wastes is the same for both the carbon and
specialty steel shops. A modest increase in the energy requirements is
also indicated under Regulatory Alternative 3. There would be no water
or noise impacts. A matrix of the environmental and economic impacts
for the regulatory alternatives is presented in Table 1-2.
1.3 ECONOMIC IMPACT
Capital and annualized costs were estimated for the regulatory
alternatives. The capital and the annualized costs in the fifth year
for the capture and control systems in the carbon and specialty steel
shops under Regulatory Alternatives 2 and 3 show an increase from the
baseline levels. These increases in the capital and annualized costs
are the result of higher costs for the control and capture systems, which
offset the decrease in the costs related to the use of periodic observation
of emission capture systems operating parameters and of the opacity of
visible emissions discharged from the pollution control device installed
on the EAF or ADD vessel in lieu of continuous flow monitoring and of
continuous opacity monitoring.
Cost impacts on product prices are generally quite small for the
regulatory alternatives. Compared to Regulatory Alternative 1, Regulatory
Alternative 2 would result in an increase of about 0.11 percent in the
average total cost of producing carbon' and specialty steel billets. The
impacts of Regulatory Alternative 3 on average total cost of producing
carbon and specialty steel billets are al'l under $1.90 per megagram (Mg).
Regulatory Alternative 3 would result in a worst-case product price increase
1-2
-------
a
UJ
-
C£
i—
u_
U_ I-H
o u_
*j
*^
JZ.
^e.
O)
to
re
0)
S-
u
O)
-P
to
re
3
."^
r—
O
C
O
u
13
O)
c
o
tfl
•g
u.
\ +J
s- c
0) 0)
CL. u
/*->
$-
to
C
0
4«)
\^rf
S-
^>^
*N^
O)
s:
i -P
S- C
(U a)
rt (j
S-
^>^
'X^
to
C
0
^
J_
^1.
N»^
O5
Lf>
*
r—
CO
o
^
t-O
•
Lf)
CM
/^*s
0
cn
to
«*t
'VM^
0
c*o
LO
*»
i—
*JD
cn
x—^
o
cn
**
^
o •,•
co
Lf)
-"
co
p^»
d>
^
4^
r™*
re
•i—
u
CM
o
o
o
CO
CM
(0
o
X
o
o
CM
o
CM
CM
O
O
CM
CTV
o o
<— CM
r- CM
O O
O CD
r- CM
CM CO
to
re
o
•a
oo
cn
.cr
u
re
c
-a
O)
-P
o
a.
0)
a>
re
tn
•P
to
o
o
re
1-3
-------
TABLE 1-2. MATRIX OF ENVIRONMENTAL AND ECONOMIC
IMPACTS FOR REGULATORY ALTERNATIVES
Solid ;
Regulatory Air Water waste Energy Noise
alternative impact impact impact impact impact
1
2
3
Key: +'
0
1
2
3
4
*
**
***
0 0 0 00
+3** 0 -1** 0 0
+4** o -1** -1** 0
Beneficial impact
Adverse impact
No impact
Negligible impact
Small impact
Moderate impact
Large impact
Short-term impact
Long-term impact
Irreversible impact
Economic
impact
0
_•]**
_2**
1-4
-------
of approximately 0.32 percent, including monitoring costs. There would be
no significant effect on the growth of the carbon and specialty steel shops.
1-5
-------
-------
: 2, INTRODUCTION
2.1 BACKGROUND AND AUTHORITY FOR STANDARDS
Before standards of performance are proposed as a Federal regulation,
air pollution control methods available to the affected industry and the
associated costs of installing and maintaining the control equipment are
examined in detail. Various levels of control based on different technolo-
gies and degrees of efficiency are expressed as regulatory alternatives.
Each of these alternatives is studied by EPA as a prospective basis for
a standard. The alternatives are investigated in terms of their impacts
on the economics and well-being of the industry, the impacts on the
national economy, and the impacts on the environment. This document
summarizes the information obtained through these studies so that
interested persons will be able to see the information considered by EPA
in the development of the proposed'standards.
Standards of performance for new stationary sources are established
under Section 111 of the Clean Air Act (42 U.S.C. 7411) as amended,
hereafter referred to as the Act. Section 111 directs the Administrator
to establish standards of performance for any category of new stationary
source of air pollution which ". . . causes, or contributes significantly
to, air pollution which may reasonably be anticipated to endanger public
health or welfare."
The Act requires that standards of performance for stationary
sources 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 standards apply only to stationary
2-1
-------
sources, the construction or modification of which commences after the
standards are proposed in the Federal Register.
The 1977 amendments to the Act altered or added numerous provisions
which apply to the process of establishing standards of performance.
For example:
1. EPA is required to list the categories of major stationary
sources which have not already been listed and regulated under standards
of performance. Regulations must be promulgated for these new categories
on the following schedule: :
a. 25 percent of the listed categories by August 7, 1980
b. 75 percent of the listed categories by August 7, 1981
c. 1.00 percent of the listed categories by August 7, 1982
A governor of a State may apply to the Administrator to add a category
which is not on the list or may apply to the Administrator to have
certain standards of performance revised.
2. EPA is required to review the standards of performance every
4 years and, if appropriate, revise them.
3. EPA is authorized to promulgate standards based on design,
equipment, work practice, or operational procedures when standards based
on emission levels are not feasible.
4. The term "standards of performance" is redefined, and a new
term, "technological system of continuous emission reduction," is defined.
The new definitions clarify that the control system must be continuous
and may include a low- or nonpolluting process or operation.
5. The time between the proposal and promulgation of standards
under Section 111 of the Act is extended to 6 months.
Standards of performance, by themselves, do not guarantee protection
of health or welfare because they are not designed to achieve any specific
air quality levels. Rather, they are designed to reflect the degree of
emission limitation achievable through application of the best adequately
demonstrated technological system of continuous emission reduction,
taking into consideration the cost of achieving such emission reduction,
any nonair quality health and environmental impact and, energy requirements.
Congress had several reasons for including these requirements.
First, standards having a degree of uniformity are needed to avoid
2-2
-------
situations where some States may attract industries by relaxing standards
relative to other States. Second, stringent standards enhance the
potential for long-term growth. Third, stringent standards may help
achieve long-term cost savings by avoiding the need for more expensive
retrofitting when pollution ceilings may be reduced in the future.
Fourth, certain types of standards for coal-burning sources can adversely
affect the coal market by driving up the price of low-sulfur coal or by
effectively excluding certain coals from the reserve base-due to their
high untreated pollution potentials. Congress does not intend that new
source performance standards contribute to these problems. Fifth, the
standard-setting process should create incentives for improving technology.
Promulgation of standards of performance does not prevent State or
local agencies from adopting more stringent emission limitations for the
same sources. States are free under Section 116 of the Act to establish
even more stringent emission limits than those established under Section 111
or than those necessary to attain or maintain the National Ambient Air
Quality Standards (NAAQS) under Section 110. Thus, new sources may in
some cases be subject to State limitations that are more stringent than
standards of performance, under Section 111, and prospective owners and
operators of new sources should be aware of this possibility in planning
for such facilities.
A similar situation may arise when a major emitting facility is to
be constructed in a geographic area that falls under the prevention of
significant deterioration of air quality provisions of Part C of the .,.
Act. These provisions require, among other things, that major emitting
facilities to be constructed in such areas are to be subject to best
available control technology. The term "best available control technology"
(BACT), as defined in the Act, means
. . . an emission limitation based on the maximum degree of
reduction of each pollutant subject to regulation under this
Act emitted from or which results from any major emitting
facility, which the permitting authority, on a case-by-case
basis, taking into account energy, environmental, and economic
impacts and other costs, determines is achievable for such
facility through application of production processes and avail-
able methods, systems, and techniques, including fuel cleaning
or treatment or innovative fuel combustion techniques for
control of each such pollutant. In no event shall application
2-3
-------
of "best available control technology" result in emissions of
any pollutants which will exceed the emissions allowed by any
applicable standard established pursuant to Sections 111 or 112
of this Act. (Section 169(3))
Although standards of performance are normally structured in terms
of numerical emission limits where feasible, alternative approaches are
sometimes necessary. In some cases, physical measurement of emissions
from a new source may be impractical or exorbitantly expensive.
Section lll(h) provides that the Administrator may promulgate a design
or equipment standard in those cases where it is not feasible to prescribe
or enforce a standard of performance. For example, emissions of hydro-
carbons from storage vessels for petroleum liquids are greatest during
tank filling. The nature of the emissions (i.e., high concentrations
for short periods during filling and low concentrations for longer
periods during storage) and the configuration of storage tanks make
direct emission measurement impractical. Therefore, a more practical
approach to standards of performance for storage vessels has been equipment
specification.
In addition, under Section lll(j) the Administrator may, with the
consent of the Governor of the State in which a source is to be located,
grant a waiver of compliance to permit the source to use an innovative
technological system or systems of continuous emission reduction. In
order to grant the waiver, the Administrator must find that: (1) the
proposed system has not been adequately demonstrated; (2) the:proposed
system will operate effectively and, there is a substantial likelihood
that the system will achieve greater emission reductions than the otherwise
applicable standards require or at least an equivalent reduction at
lower economic, energy, or nonair quality environmental cost; (3) the
proposed system will not cause or contribute to an unreasonable risk to
public health, welfare, or safety; and (4) the waiver when combined with
other, similar waivers, will not exceed the number necessary to achieve
conditions (2) and (3) above. A waiver may have conditions attached to
ensure the source will not prevent attainment of any National Ambient
Air Quality Standard (NAAQS). Any such condition will be treated as a
performance standard. Finally, waivers have definite end dates and may
be terminated earlier if the conditions are not met or if the system
fails to perform as expected. In such a case, the source may be given
2-4
-------
up to 3 years to meet the standards and a mandatory compliance schedule
will be imposed.
2.2 SELECTION OF CATEGORIES OF STATIONARY SOURCES
Section 111 of the Act directs the Administrator to list categories
of stationary sources. The Administrator ". . . shall include a category
of sources in such list if in his judgment it causes, or contributes
significantly to, air pollution which may reasonably be anticipated to
endanger public health or welfare." Proposal and promulgation of standards
of performance are to follow.
Since passage of the Clean Air Amendments of 1970, considerable
attention has been given to the development of an approach for assigning
priorities to various source categories. The approach specifies areas
of interest by considering the broad strategy of the Agency for imple-
menting the Clean Air Act. Often, these areas are pollutants that are
emitted by stationary sources rather than the stationary sources themselves.
Source categories that emit these pollutants were evaluated and ranked
considering such factors as: (1) the level of emission control (if any)
already required by State regulations; (2) estimated levels of control
that might be required from standards of performance for the source
category; (3) projections of growth and replacement of existing facilities
for the source category; and (4) the estimated incremental amount of air
pollution that could be prevented in a preselected future year by standards
of performance for the source category. Sources for which new source
performance standards were promulgated or under development during 1977,
or earlier, were selected using these criteria.
The Act amendments of August 1977 establish specific criteria to be
used in determining priorities for all source categories not yet listed
by EPA. These are: (1) the quantity of air pollutant emissions which
each such category will emit, or will be designed to emit; (2) the
extent to which each such pollutant may reasonably be anticipated to
endanger public health or welfare;.,and (3) the mobility and competitive
nature of each such category of sources and the consequent need for
nationally applicable new source standards of performance.
2-5
-------
The Administrator is to promulgate standards for these categories
according to the schedule referred to earlier.
In some cases, it may not be immediately feasible to develop standards
for a source category with a high priority. This might happen if a
program of research is needed to develop control techniques or if techniques
for sampling arid measuring emissions require refinement. In the developing
of standards, differences in the time required to complete the necessary
investigation for different source categories must also be considered.
For example, substantially more time may be necessary if numerous pollutants
must be investigated from a single source category. Further, even late
in the development process the schedule for completion of a standard may
change. For example, inability to obtain emission data from well-controlled
sources in time to pursue the development process in a systematic fashion
may force a change in scheduling. Nevertheless, priority ranking is,
and will continue to be, used to establish the order in which projects
are initiated and resources assigned.
After the source category has been chosen, the types of facilities
within the source category to which the standard will apply must be
determined. A source category may have several facilities that cause
air pollution, and emissions from these facilities may vary according to
magnitude and control cost. Economic studies of the source category and
of applicable control technology may show that air pollution control is
better served by applying standards to the more severe pollution sources.
For this reason, and because there is no adequately demonstrated system
for controlling emissions from certain facilities, standards often do
not apply to all facilities at a source. For the same reasons, the
standards may not apply to all air pollutants emitted. Thus, although a
source category may be selected to be covered by standards of performance,
not all pollutants or facilities within that source category may be
covered by the standards.
2.3 PROCEDURE FOR DEVELOPMENT OF STANDARDS OF PERFORMANCE
Standards of performance must: (1) realistically reflect best
demonstrated control practice; (2) adequately consider the cost, the
nonair quality health and environmental impacts, and the energy require-
2-6
-------
ments of such control; (3) be applicable to existing sources that are
modified or reconstructed as well as to new installations; and (4) meet
these conditions for all variations of operating conditions being considered
anywhere in the country.
the objective of a program for development of standards is to
identify the best technological system of continuous emission reduction
that has been adequately demonstrated. The standard-setting process
involves three principal phases of activity: (1) information gathering;
(2) analysis of the information; and (3) development of the standard of
performance.
During the information gathering phase, industries are questioned
through telephone surveys, letters of inquiry, and plant visits by EPA
representatives. Information is also gathered from other sources,
including a literature search. Based on the information acquired about
the industry, EPA selects certain plants at which emission tests are
conducted to provide reliable data that characterize the pollutant
emissions from well-controlled existing facilities.
In the second phase of a project, the information about the industry
and the pollutants emitted is used in analytical studies. Hypothetical
"model plants" are defined to provide a common basis for analysis. The
model, plant definitions, national pollutant emission data, and existing
State regulations governing emissions from the source category are then
used in establishing "regulatory alternatives." These regulatory
alternatives are essentially different levels of emission control.
EPA conducts studies to determine the impact of each regulatory.
alternative on the economics of the industry and on the national economy,
on the environment, and on energy consumption. From several alternatives,
EPA selects the single most plausible regulatory alternative as the
basis for standards of performance for the source category under study.
In the third phase of a project, the selected regulatory alternative
is translated into performance standards, which, in turn, are written in
the form of a Federal regulation. The Federal regulation, when applied
to newly constructed plants, will limit emissions to the levels indicated
in the selected regulatory alternative.
2-7
-------
As early as is practical in each standard-setting project, EPA
representatives discuss the possibilities of a standard and the form it
might take with members of the National Air Pollution Control Techniques
Advisory Committee. Industry representatives and other interested
parties also participate in these meetings.
The information acquired in the project is summarized in the background
information document (BID). The BID,, the standard, and a preamble
explaining the standard are widely circulated to the industry being
considered for control, environmental groups, other government agencies,
and offices within EPA. Through this extensive review process, the
points of view of expert reviewers are taken into consideration as
changes are made to the documentation.
A "proposal package" is assembled and sent through the offices of
EPA assistant administrators for concurrence before the proposed standard
is officially endorsed by the EPA Administrator. After being approved
by the EPA Administrator, the preamble and the proposed regulation are
published in the Federal Register.
As part of the Federal Register announcement of the proposed regulation,
the public is invited to1 participate in the standard-setting process.
EPA invites written comments on the proposal and also holds a public
hearing to discuss the proposed standard with interested parties. All
public comments are-summarized and incorporated into a second1volume of
the BID. All information reviewed and generated in studies in support
of the standard of performance is available to the public in a "docket"
on file in Washington, D.C.
Comments from the public are evaluated, and the standard of performance
may be altered in response to the comments.
The significant comments and EPA's position on the issues raised
are included in the "preamble" of a promulgation package, which also
contains the draft of the final regulation. The regulation is then
subjected to another round of review and refinement until it is approved
by the EPA Administrator. After the Administrator signs the regulation,
it is published as a "final rule" in the Federal Register.
2-8
-------
2.4 CONSIDERATION OF COSTS
Section 317 of the Act requires an economic impact assessment with
respect to any standard of performance established under Section 111 of
the Act. The assessment is required to contain an analysis of: (1) the
costs of compliance with the regulation, including the extent to which
the cost of compliance varies depending on the effective date of the
regulation and the development of less expensive or more efficient
methods of compliance; (2) the potential inflationary and recessionary
effects of the regulation; (3) the effects the regulation might have on
small business with respect to competition; (4) the effects of the
regulation on consumer costs; and (5) the effects of the regulation on
energy use. Section 317 requires that the economic impact assessment be
as extensive as practicable.
The economic impact of a proposed standard upon an industry is
usually addressed both in absolute terms and by comparison with the
control costs that would be incurred as a result of compliance with
typical, existing State control regulations. An incremental approach is
taken because both new and existing plants would be required to comply
with State regulations in the absence of a Federal standard of perfor-
mance. This approach requires a detailed analysis of the economic
impact of the cost differential that would exist between a proposed
standard of performance and the typical State standard.
Air pollutant emissions may cause water pollution problems, and
captured potential air pollutants may pose a solid waste disposal problem.
The total environmental impact of an emission source must, therefore, be
analyzed and the costs determined whenever possible.
A thorough study of the profitability and price-setting mechanisms
of the industry is essential to the analysis so that an accurate estimate
of potential adverse economic impacts can be made for proposed standards.
It is also essential to know the capital requirements for pollution
control systems already placed on plants so that the additional capital
requirements necessitated by these Federal standards can be placed in
proper perspective. Finally, it is necessary to assess the availability
2-9
-------
of capital to provide the additional control equipment needed to meet
the standards of performance.
2.5 CONSIDERATION OF ENVIRONMENTAL IMPACTS
Section 102(2)(C) of the National Environmental Policy Act (NEPA)
of 1969 requires Federal agencies to prepare detailed environmental
impact statements on proposals for legislation and other major Federal
actions significantly affecting the quality of the human environment.
The objective of NEPA is to build into the decision-making process of
Federal agencies a careful consideration of all environmental aspects of
proposed actions.
In a number of legal challenges to standards of performances for
various industries, the United States Court of Appeals for the District
of Columbia Circuit has held that environmental impact statements need
not be prepared by the Agency for proposed actions under Section 111 of
the Clean Air Act. Essentially, the Court of Appeals has determined
that the best system of emission reduction requires the Administrator to
take into account counterproductive environmental effects of proposed
standards, as well as economic costs to the industry. On this basis,
therefore, the Courts established a narrow exemption from NEPA for EPA
determinations under Section 111.
In addition to these judicial determinations, the Energy Supply and
Environmental Coordination Act (ESECA) of 1974 (PL-93-319) specifically
exempted proposed actions under the Clean Air Act from NEPA requirements.
According to Section 7(c)(l), "No action taken under the Clean Air Act
shall be deemed a major Federal action significantly affecting the
quality of the human environment within the meaning of the National
Environmental Policy Act of 1969." (15 U.S.C. 793(c)(l))
Nevertheless, the Agency has concluded that the preparation of
environmental impact statements could have beneficial effects ;on certain
regulatory actions. Consequently, although not legally required to do
so by Section 102(2)(C) of NEPA, EPA has adopted a policy requiring that
environmental impact statements be prepared for various regulatory
actions, including standards of performance developed under Section 111
2-10
-------
of the Act. This voluntary preparation of environmental impact statements,
however, in no way legally subjects the Agency to NEPA requirements.
To implement this policy, a separate section is included in this
document which is devoted solely to an analysis of the potential
environmental impacts associated with the proposed standards. Both
adverse and beneficial impacts in such areas as air and water pollution,
increased solid waste disposal, and increased energy consumption are
discussed.
2.6 IMPACT ON EXISTING SOURCES
Section 111 of the Act defines a new source as ". . . any stationary
source, the construction or modification of which is comme.nced ..."
after the proposed standards are published. An existing source is
redefined as a new source if "modified" or "reconstructed" as defined
in amendments to the General Provisions (40 CFR Part 60, Subpart A),
which were promulgated in the Federal Register on December 16, 1975
(40 FR 58416).
Promulgation of standards of performance requires States to establish
standards of performance for existing sources in the same industry under
Section lll(d) of the Act if the standard for new sources limits emissions
of a designated pollutant (i.e., a pollutant for which air quality
criteria have not been issued under Section 108 or which has not been
listed as a hazardous pollutant under Section 112). If a State does not
act, EPA must establish such standards. General procedures for control
of existing sources under Section lll(d) were promulgated on
November 17, 1975, as Subpart B of 40 CFR Part 60 (40 FR 53340).
2.7 REVISION OF STANDARDS OF PERFORMANCE
Congress was aware that the level of air pollution control achievable
by any industry may improve with technological advances. Accordingly,
Section 111 of the Act provides that the Administrator ". . . shall, at
least every 4 years, review and, if appropriate, revise ..." the
standards. Revisions are made to ensure that the standards continue to
reflect the best systems that become available in the future. Such
revisions will not be retroactive but will apply to stationary sources
constructed or modified after the proposal of the revised standards.
2-11
-------
-------
3. ELECTRIC ARC FURNACES AND ARGON-OXYGEN DECARBURIZATION VESSELS
IN THE STEEL INDUSTRY: PROCESSES AND POLLUTANT EMISSIONS
3.1 GENERAL
This chapter presents a description of the electric arc furnace
(EAF) and argon-oxygen decarburization (AOD) vessel processes and their
emissions. The source category is covered under the Standard Industrial
Classification (SIC) Code 3312, Blast Furnaces and Steel Mills. Also
discussed in this chapter is the selection of baseline emissions for
EAF's. and AOD vessels which will be .used in later chapters to determine
incremental environmental, economic, and ehergy impacts of the regulatory
alternatives. .
3.1.1 General Industry Description
Major sources of air pollution in the steel industry are the basic
oxygen process furnaces (BOPF's), EAF's, AOD vessels, open-hearth furnaces,
blast furnaces, and coke plants (Figure 3-1). All processes emit large
quantities of air pollutants, primarily particulate matter, if not
properly controlled. This document will deal primarily with the EAF and
AOD vessel segments of the industry.
The EAF's are typically utilized in semi-integrated and nonintegrated
steel mills and specialty steel shops (Figure 3-2). The semi-integrated
steel mills are a new class of integrated plants that utilize direct
reduced iron (DRI) in addition to iron and steel scrap as a source of
ferrous material (the manufacture and use of DRI will be briefly discussed
later in the chapter). The nonintegrated steel mills operate melting
units, such as an EAF, casting units, and fabrication mills and produce a
limited range of products for a regional market. Mini-mills, a term that
applies to nonintegrated steel mills that produce less than 544,200 Mg_^
3-1
-------
C
(O
O)
(1)
T3
S= •
ea
O
c
(C
1
en
o
i
CO
cu
=5
CD
V)
3-2
-------
STEELMAKING
FURNACE
_j BASIC OXYGEN
OPEN HEARTH
L ELECTRIC ARC
FINAL HOT ROLLING,
COLD ROLLING,
FINISHING,
ANNEALING, ETC.
INGOT
BREAKDOWN,
PRIMARY
ROLLING
INGOT
CASTING
CONTINUOUS CASTING
INTEGRATED:
NONINTEGRATED:
SEMI-INTEGRATED:
POSSIBLE MAJOR ROUTES
COKING—BLAST FURNACE—BASIC OXYGEN—INGOT CASTING—FINISHING
SCRAP + DIRECT REDUCTION—ELECTRIC ARC + SECONDARY REFINING—
CONTINUOUS CASTING—FINISHING
DIRECT REDUCTION + SCRAP—ELECTRIC ARC + SECONDARY REFINING—
CONTINUOUS CASTING—FINISHING
Figure 3-2. Schematic flowchart for integrated
and nonintegrated steelmaking.2
3-3
-------
(600,000 tons) of steel per year, typically use EAF's as metal melters.2
Specialty steel companies produce stainless and alloy steels from steel
making units and usually do not deal with raw iron ore. A specialty
steel plant typically has electric arc furnaces, some secondary steel
refining equipment such as an AOD vessel or vacuum degassing unit,.and
forming and rolling facilities.
There are two basic types of steel-producing EAF facilities: those
that produce the common grades of steel (carbon steel) and those that
produce stainless and alloy steels (specialty steel). Electric arc
furnaces are used extensively as metal melters for both types of steel
facilities with oxygen blowing used to assist in scrap metal meltdown.
In the production of carbon steels, the EAF is also used as the refining
vessel where oxygen blowing is performed for the final steel chemical
adjustment. In addition, various additives needed to produce specific
grades of carbon steel are added directly to the furnace or added to the
ladle during a tap.
The EAF is used primarily as a metal melter in the specialty steel
shop and the molten steel from the EAF is charged to an ADD vessel or
other secondary refining vessel for refining. This practice allows
shorter heat times and better quality control over the final product.
Fewer additions are made and less oxygen blowing for molten steel chemistry
adjustments is performed in the EAF when the AOD vessel is used for
steel refining. This innovation in specialty steel production has
gained widespread industry acceptance over the last 6 to 8 years.
Several factors tend to favor the trend toward the increased
utilization of EAF's in steel production. They include the higher blast
furnace energy costs, shorter starting periods for EAF's, large supplies
of available steel scrap, the ability to utilize DRI, the growing use of
specialty steels by industry, the growing number of mini-steel plants
(which normally use EAF's exclusively), the adoption of ultrarapid steel
melting technology, and the increased use of water-cooled panels and
roofs to reduce refractory costs.2-4
The utilization of AOD vessels has increased in the last few years
for several reasons. These reasons are: an interest in the reduction of
EAF operating costs (less refining is required in the EAF when an AOD
3-4
-------
vessel Is used), the ease of molten bath temperature adjustment, the
simplicity of alloy refinement, the high quality of AOD-produced steel
(similar in quality to vacuum-degassed steel), the improved operating
control (over that of an EAF) in refining molten steel, the significant
savings in the use of alloys, and the increase in the melt capacity of
the shop when the EAF is used only as a melting device.2,5,6
In 1980, EAF's in the United States produced 28.2 xlO6 megagrams
(Mg) (31.2 xlO6 tons) of steel.7 Of this total, 71.6 percent was carbon
steel, 23.0 percent was alloy steel, and 5.4 percent was stainless
steel. This accounts for 23.6 percent of the carbon steel, 46.3 percent
of the alloy steel, and TOO percent of the stainless steel produced from
all furnace types in the United States.7 In 1980, 96 percent of the
steel produced by EAF's was made from recycled iron and steel scrap, and
the remainder of the steel was produced with blast furnace hot metal and
DRI.7,8 EAF's and ADD vessels are well suited for producing alloy
steels where only small batches are needed. In 1980, AOD vessels in the
United States refined approximately 2,131,900 Mg (2,350,000 tons) of
specialty steels, with much of this tonnage being stainless steel.
In 1981, there were 327 EAF's in the United States that were operated
by 87 companies in 130 locations (Table 3-1). These plants are in 36 States,
with 63 percent of the plants located in 8 States (Table 3-2). Of the
327 EAF's approximately 10 percent or 32 furnaces are subject to the
existing new source performance standard. The EAF capacities range from
4.5 Mg (5 tons) to 363.9 Mg (400 tons).9-11 The percentages of EAF's
within a range of capacities are listed in Table 3-3. Many of the smaller
furnaces are in shops that produce small quantities of specialty steels.
There were 27 AOD vessels in operation in the United States in
1981. These vessels were operated by 19 companies with 23 locations in
9 States (Table 3-4). AOD vessel capacities range from 3.6 Mg (4 tons)
to 158.8 Mg (175 tons).13 The percentages of AOD vessels within a range
of capacities are listed in Table 3-3, along with the EAF size ranges.
3.1.2 Industry Growth and Projections
The demand for steel has increased from nonresidential and nonauto-
motive segments of industry, such as nonresidential construction and the
production of durable goods (notably machinery, industrial equipment,
3-5
-------
M
1
r—
CO
r—
1
•
CO
LU
CO
o
UJ
>— i .
2;
=5
UJ
z
t— i
i3
_i
TE.
UJ
UJ
I—
CO
1— <
co
UJ
o
°LS
. 3 10
C
o
(D
U
o
^J
c
a!
jijj,
'lotj^i
lv IB SI
fll CL****
a.re
o o
t- vt
o i ei
i. u
• 3 RJ
owe
|
*J
ra
•
vt • :£*"•
(9 21 * •
wo. >, OP-
i/> 4-> - OP-
'S C •*- 3 LU •
•p- re u u uj w
re »— a) TO H- J->
v> +•* *a o. t_ to _e
UJ v* •— i/J >» en
>«• S vj — J •»-
O£ O O O OJ
^^ . ^; . ox
en JQ . ja • i—
S 2° 2 Q S3
t-j" ° ig
s ' ' s
o co CM r-.
cn o CNJ CM
CM CM CM CM
^
C
o
S,
W 35 i—
cu e ^- ^ 1-4
" C U» -J -4->
— J w y> CM CM »—
t
>
O
at
*-/ (0 ^y; "^ o Q O QJ m
fc. ^- 1_ O +J .r» p- 4-> O.
. O «C O . »—
uj js jz 4J *J vi jns cu o r"» re m •*•
H^c1cVcc o oil'*- s-*oai
es .£ £ § § £§ u ex §" m *c ^ "C
|COCO EX g| gUJE => 03
2 So
•*• co «d- co ro r*» CM
^- un m «— CM .CM
-a
m
r— P- CM CM IT) r— CM
a.
o±
M S
p-« • VT
>* C ~J •
- CD C — 1 CB
VI O C O *~* (- at a a •>-
O 4J O UJ O a. O O
i0"^ i31 1" ' s
•-co *n— —o r- p«*
us eo MUD cn o r* o
o trt *•" PO CM r^ cn
CM ^ CM «<*• •— r- CM r—
c
o
>
JH: • • u re
££ ^-t . g"
U) .-3: CXr- 0 « -
^ x -Q.CD o^
re o»2 5*^* w>° _i oo ^^ S
*>cc reu h-reit>tniA uj
+J 33 CC 4J't- )-H »— r— fc.1-
3 oo ns rt) re "o H- -M *J rere 2:
... _j . . ca
P— r>
• » o
H- J- O >i VI *
S LU C -J «r- «3 OJ
t— *m Lum o -P- ^£Jr
2 S ui 2
cn ID 03 co
m •-- o
CNJ CO CO CM
»-• do
d CJ CJ
-J O) O)
LU e c
LU -f— «r-
p- C, U
>- CL,' O - ^r o r-. _ «d|
«a- c\j LT> co o «*
re
p- CM CM CM CO CM •
. CL.
cj m . ; o
al.3 °
O - oc ^ —i
C t- h- t- LU
i- :* aj co o 3:
3 O > , LL
< CO CO >- 5C H-
r1. O O
jif "o "o "c "e
S J= J= ^ "=
gOO l-IP-4
_j
LL.
O PO ro
J3
^_*_ _
a
u
=f
V—
J
• UJ =*
Sx *
d1- S~:
UJ - p— P—
H" C OHH
v> re LU
— I J= LU O
ce r— LU u
a.;g a js.
3= 3=
o o
CMCn^f CM f>
r* cn <* T ro
re
r- CM CM m CM
C ' 4J
f— a. c • m re
a. re m *— a.
mm ,^* . c c
ca co •*-* •*•> o o
. . t
3-6
-------
/— •, :
•o
CD
3
c
4->
C
o
u
•^s
•
CO
LU
_l
03
*
<- +>
g'si
oi a.*— '
a. m
0 0
u— w
O 1 -r- --\
£££
OJ Q.^-1
a. 03
o o
U- \f>
O f 1)
s- u
• 3 CO
o **- c
3E
c
o
*J
.Plant/loca
en
'Z£^
«3 U O
1- CtJ 3E
OJ 0.*.-
o. ra
0 U
H- u>
O 1 QJ
£- CJ
. 3 fO
0*- =
C
O
^
u
c
c
n
a.
- ^.
*
UJ
O
r— OJ CO
CO P» -—
f— csj en
^^ r?* r~.
ut r- (0
r~ u^
•o 'O. a.
— E e
re re re
en H- H-
•
r^. r*- CM *S* r*>- r^
csj csj r^ ^ csi r-
CM Csl Csi CO f— CO
(SI CO CSJ CSJ Csi r—
C)
° >
Z3 * *"**
5 Cxi •
Z ' O O -J «
Q£ . (J • tj « UJ (/>
f— 1 — ^.^ , 'p uj;? '^O*
ij i t/itf LLJ° vi- o!"5^
2— 3C » COC 3IO •— • O 5
0> 0(W 0 f=4-> 0000
3:0) 3:0. z*»- rDC oo^r—
h— c H-e ot- 001 t-iou.
rf — ' rf1 — 2?^ Vj*™" t^l^*
*
co in <-o <^> . •**• r~
S ^ R S ' J2S2
»— ro
re
Csl Csi if) CSJ CSJ r—
ol
o
°
1/1
LU O UJ
LU V) CJ UJ
t— i/> O J^ . 1—
t/^j«:-«-fc.re 31 "^-ri
1-.JZ.C I-'i— UJW LU
S"c"c P1?-*"^" e^s zoT
^ KQ 01 3 3 -O H-iC
^^U^a^JtlSl .4J m3
03410 •+->-*•> s- s_. 2; +J
r-^— -r--*- itsro -*J z:*J
coa_o:2e3B uJre >-s-
UJ _l (1) _J O
z . . . ce i/» u.
s? S %
o «*• ^ S .—JO
Jid ^ •— »J3 t/JO Z 3: UJin
O (J G£ Ul CL ^ -H— CD
• c£ ^~J ^LLJ ^>~
o o S :n
r
P
o
CJ
C£
£
1
<«
z
o
^
o
_J
UJ
to
1
0
=3
~3
PO
U3
en
r-
'
~^
UJ
>— «
Q£
0
§ X
i— i e
t-
"3 ^
X
-p — CO
0 U3 r— «3-
-CSI ^- *—
e
o
ul
>
Q
41
5 03 LU
— O- *
3. O) CD
- C C
2-S > >
t- s. s- s-
jj UJ »-•»-*
* *
CO «ej-
s s
r- CS|
o
CJ
UJ
. £
<4~ LO
H- .
flT UJ ^*
1— — 1
Emery vil
KENTUCKY i
Ashland,
en
O
CSI
CSI
J—
~J
a
CQ
t
S^:
a* "^
^ •— 1 CO
§j*:
c:
>-* re
i — j s^
MH
S
__l
u.
h-
Z
S
§
U
2
*"
<*•
r^
Csl
>
0
D_ r~-
O +^ -C
Lu aj
trt 1o aT
_j t.
o
z re u
O 0) LU
5
CSt
-.
oT
OJ
3Z.
0
S'
CSI
UJ
QZ.
h-
o
1
=a
i *
1/5 Ol"
;_ £;
1 1
3 O
r-
m
CSJ
. Si
— 4)
— 3= >—
""* § »
* I~H ,2
U —1 •*-
0 -J >
0
>= c
O ^
g
ro
.S
§ X t
II §
z S
^ >—
— «*• 03
rC «3- CO
r-*. tr> a
fO
Csl i—
8 +J C
~ ^ ^T 'tl ^
UJ 1? V^ i— r— Cl.
UJ ,33 -
if.pSIi
uJ O
0 CO —
' S S o
ro csj csi
• F- x
1_ ej> t—
O •*-*
O » CO «•
0 -P
•— . w c
a> x e
CO 3s O O
fO C CD _j .
O) O - UJ r—
£j "oj o* "a! Q. t— *— *
§ 0_C3 00 UJ.C
t-H O)
fe . . ds
0 <
^ -J
co *"•• cr>
cd co cri
C\i C\J t—
Q- CZ
K 0
o -_n
•a "3
^ ffi m III
o" t^^" u 2*"!
03 1 gj I_M C
.,- t— «3 - d) -^>
O ^ O. Ni C i~"
o re «c nj ^~
.*-> ro
t/i oc QJ ry* o O
(T3 UJ -C UJ O
Z 2:
•-H )-H
**;
^r
m
r-
C ?
a_
c: c*
II
IT) CSI
**£
i—
-------
i=£~
•a
0)
t ^
4-*
O
(J
4
|M~
cr>
UJ
CO
ta u os
*" roO
o'S
w in
O ! e*
i. u
» «3 re
.gv-c:
t/ location
c
a.
f*
III
g-8
**- U)
* 3 (O
2*" C
Plant/location
|^
i o.S
a-s
jp u- o
a re
3 u- c
£ 0
Plant/location
CO CM
r— CM
*— CO
eu
cc
o
«
LU
LU
S«c
S«i
o • 3 is
cj re ce *
CO O. O C C
4 LU re re
si HI
a *J LU as z
LU re z:
LU LU ,
H- H-
r-»5 co
m-«
o >, po
°^*2 re re
_j »— ^ ^j ^u
UJ J= !S
t— r— - CJ CJ
cn re e
^t. 0 J= J=
"*C"C 3 "=
_j a> re o o
CQ CJ CJ t/> VI
UJ
ce
r*. cn
CO CO
CO
CM CO
8
UJ
ce
i 3
NORTHWEST STEEL ROLL!
HILLS, INC.
Kent, Wash.
NORTHWESTERN STEEL &
Sterling, 111.
«n cn r-
r^ o CM r^
r— CM Csl
X
CJ
d1-
"s
UJ U
CO L.
S£
LU
H-
O CO »—
O CO f*^
CM CO f»»
re
CM«— *-
al
ce
o
o
_j
ROANOKE ELECTRIC STEE
Roanoke, Va.
f"- r-
CO CO
re
CO CM
r-
«r-
"§ 0
H- •»-
|°
C
0 •— 0
CJ O) •*->
(U C
-c ••-» re
UJ CO O
E
J—
CM CM
r-. r*.
CM CM
8
UJ
t—
_J
en
§
ro ^-
CO V
re
CM ro
El
4-1 £
«'
CSJ^-
w . ^
3 If — 1
^> *r~ ' '
J0 to
VJ
g
m^co
re »"
CM CM r—
U
r— (U
•£==
'o 'o
co «s* loesi^-io^-r^^^i—
co co CM*— cor«*cncococ3
f— , *— ^— C\l
. (/) -^
Q_ . .^ t-
ce o. t- 5
o ce . • .0 2e ^
^ *^" £_ C ^^
UJ t — i • CJ «>
LU re uj u) -u o> *
H-O- LU^r—crea- o :
Lo h— t, »— a* o_ cr re cn x
o e ftS£ V c- a0" f 3 ^"
§1 ^ss II « s" IS 1 ="
SJS S^^ S3 § S ^^ «• i
cd 3 •" i i 1 5 S & ==3>,
z E1 Si££ ^^ zo «5 |2.
55 LU- ^" ^^H. uj'ac
33 uj-a oto uao.0 i-r^g
ii KS a - ^o§^^s>
5. >, r— LU cu •••* s •*<£ ^^ re re
S£ 53 S5 o .
00 0. : CL,
CO
CJ
CM
C3
_J
uj re
UJ O-
-fc
?r o o
{,- «
-2 ^ o
CO LU
s
r--
CM
CM
!
to
_J
Z *^
LU re
i—
CO »
re
LU C
1— O
H- U_
<5'
co co co in CM rj- 10
us to o
r— CO T
LU
1/3
ce
LU' t-
I5
CU
—j cn •
LU s- re
U.I O O.
t— u_
"'SB
I'll
Ci 01 S.
z; J- 3
S,
cn
jr .
U J=
cJ oj »
ce a) c
O 4-> O
CJ to m
:>< u u
u j ja re
|T"
ro r*«
tO CM
re
CL
O)
, |
re
«r
in
CM
±
dt
re
re s-
!5
to
to
%t
in *—
•0- CO
re
y
STRUCTURAL METALS, II
Sequin/ Tex.
cn
5
re
o
o
-J .
UJ -3
to
UJ O
ce «£
as JZ
t-t CU
-------
TABLE 3-2. LOCATION OF STEEL PLANTS WITH
EAF'S AND AOD VESSELS—198110-13
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut .
Delaware
Florida
Georgi a
Hawai i
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maryland
Michigan
Minnesota
Mississippi
Missouri
Nebraska
North Carolina
New Jersey
New York
Ohio ' '
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Plants
EAF
3
1
2
6
1
1
1
3
2
1
12
5
1
3
2
2
6
1
1
1
1
1
3
5
7
1
2
28
1
3
4
13
1
2
3
1
130
AOD
vessel
1
1
3
?a
1
2
2
10
1
23
aAnother plant in Michigan has not operated its
AOD vessel since it was installed in 1977.
3-9
-------
TABLE 3-3. INDUSTRY'PERCENTAGES OF EAF'S AND AOD VESSELS
BY CAPACITY—198110-13 ''
EAF's
Capacity
range (Mg)
Under 50
50-99.9
100-149.9
150-199.9
200-299.9
Greater than 300
Percentage
51
26
10
11
1
1
AOD vessels!
Capacity
range (Mg)
Under 25
25-49.9 '•
50-99.9
Greater than 100
Percentage
41
33
22
4
3-10
-------
CO
f__
00
•""
oo
LLJ
t
re
u
t— r~*
o tv
LO
> to
O (U
z >
c
o
*J
re
u
0
T3
C
re
c
re
a.
^" 4^
to'u'o)
S 2.S
u
to
'o "3
to
O flJ
c:
o
n
u
o
•a
c
re
n
£
•a- «s- f>i
•3- • r- CM
O
CJ
i n
LU
EASTERN STAINLESS SI
Baltimore, Md.
ELECTRALLOY CORP.
Oil City, Pa.
o r~
en o
csj en
CM ' —
?v n
o °z
cj o
"i t
AL TECH SPECIALTY STEE
Watervliet, N.Y.
ALLEGHENY LUDLUM STEEI
Brackenridge, Pa.
LQ
ro
i
1
s
— J
LU
i i.i
INGERSOL JOHNSON STI
New Castle, Ind.
ro
sd
m
ARMCO, INC.
• Baltimore Works
Baltimore, Md.
_
CO
to
. s. re
o o o.
CJ3!
-J C C
UJ O O
1— O) Ol
LO C C
O to to
LosIS
LU
CO
co
"
to
^- re
S°;
t- i.
3 3
CO CO
10,
CM
^
ol
ce
o
u
_i
LU
LU
t—
LO
Z
HH
-J
O tn J=
Z5 J^ u
5|3E
O "
to i~ i-
LU re re
i33
j — i
CM (
CM
O
U
*: re
o o.
CJ
t
5£
o re
Z u- o-
<£ ee
s- O
o •>
o re 1—
CJ tV O
eo u
CO CM
ro r—
_ _
t
0
o
_J LO
LU -J
LU LU
I — LU
EARL M. JORGENSEN S
Seattle, Wash.
JOSYLN STAINLESS ST
Fort Wayne, Ind.
O LO «*•
V) fO LO
al
a:
r t
Stellite Div.
Kokomo, Ind.
CARPENTER TECHNOLOGY 1
• Steel Div.
Bridgeport, Conn.
'r- (*.
CD O
en en
JO.
0— f—
•^
to
0. i-
a. an o
en OS
CD * CJ
U — IOO
_4 T- UJ i— -i-
LU 3E LU r- JZ
UJ 1— < O
H- - IO
o o *re e
31 4J »-H t- O
=3 01 co c c:
—j H" cVj o o
U UJ
•Z. ce.
LO r— P^
r- r— en
.
tp_
• Reading Plant
Reading, Pa.
CRUCIBLE, INC.
• Stainless Steel D
Midland, Pa.
to i—
ro CO
•
OJ .
s: -z.
r— LO
re 3
.r- U
u re
tV i-
a. >i
LO LO
S. '
o
en
r«
ol
O to i —
cj JB; 1-1
s-
_J 0 -
LU 3: o
LU O)
t— J= re
LO *-> U
3 -r-
• O Si
LO LO O
re
un
•er
J--
CYCLOPS CORP.
• Universal Cyclops
Specialty Steel
Bridgeville, Pa.
ro
ro
' ^~
WASHINGTON STEEL CO
Fitch Works
Houston, Pa.
r-- ,
en
,„
>
• Empire-Detroit Di
Mansfield, Ohio
O)
c
•r-
C.
C
o
c
3
t-
0)
c:
. v>
tv
jz
c:
o>
3
r--
+j en
re -—
5_c
01 c
z: o
. re
CM r—
CM •—
^^ re
4_>
. — to
oj c:
to ••-
tv dj
> u
^Facility has second
Has not operated sin
3-11
-------
and railroad equipment).14. The decline in residential construction has
been'accompanied by a decline in steel orders from the appliance industry,
Shipments to the automotive industry declined in 1980 because of the
demand for smaller, lighter cars, which require less steel to;produce.2
In addition, the use in the automotive industry of lighter-than-steel
substitutes, such as plastics and aluminum, has resulted in a;further
decline in demand of•steel for cars.2 • ;
A summary of the steel industry production, apparent domestic
consumption, shipment weight and value, export-import weights: and values,
and employment for the period 1976 to 1988 is presented in Table 3-5.12,14
Industry predictions for 1981 call for 94.3 xlO6 Mg (104 xlO6!tons) in
apparent domestic consumption.15 Weaknesses in the automotive, service
center, and agricultural markets in 1980 will be partially offset by an
increased demand for oil- and gas-drilling equipment and rail; transportation
equipment, as well as a steady market for cans and closures.14,15 The
projected apparent domestic consumption of steel for 1985 is 108 xlO6 Mg
(119 xlO6 tons).14 In order to meet the demand that industry projects
for 1990, a 12 percent increase in total steel production capacity
(19.1 xlO6 Mg [21 xlO6 tons]) will be needed to provide for 85 percent
of the U.S. consumption.15 This figure assumes that imported steel will
capture 15 percent of the market.15 This production capacity increase
would mean a 2 percent annual growth rate between 1980 and 1990 for all
furnace types.
Growth in the use of electric arc furnaces in steel making has been
steady for the past 18 years except for the recession period ,of 19.74 to 1975
and the depressed economy in 1980-1981. Figure 3-3 shows the trends for
the past 18 years for the total production of steel and for the production
of steel from the three major types of furnaces: open hearth, basic
oxygen process, and electric arc. The total raw steel production in
1980 was 101.5 xlO6 Mg (111.8 xlO6 tons).7 The basic oxygen process
furnace accounted for 60.4 percent of the total steel produced (up from
8 percent in 1963) while electric arc furnace production accounted for
27.9 percent (up from 10 percent in 1963). These gains were at the
expense of open hearth furnace production, which accounted for 11.7 percent
in 1980 (down from 82 percent in 1963).7 :
3-12
-------
Q
LO
03
01
CJ '"^
<£ <£ CO Cn
-v.^ Or-
CO LO «/ co
<• 10 LO
i— i— co
*~* LO
en
CM
"^ en
m -a-
O
LU
O
Q-
a
co
r- CM
CM o
cn >— r~
--CM •-
s.^ cn
o
C£
to
CO r— CM
03 cn o
h» CM
co en
CO O
r-^ CM
r—
ro
O
r-- en •—
r— r— CO
•^ LO
o
r—
LU
LT>
I
co
CO
ID
r~
en
u> co
I— CSJ
r- en
co co
CM f". i—
•cf r-- LD
. . Lft
CM CM CM
CM
O
cn
CO
•3-
^< en o
to (6
o o
C.
o
c
o
u
c
0)
O-
5-
-a
£
a.
•01=
Q.
!£
•o
c
S 2
•—«
ra >
>
•r-'i- O
-p *J C
u u
01 01 I!
•Tr-j
O O «C
3-13
-------
106 Mg
160
106 tons
o
H—«
O
o
a.
_j
LU
LU
CO
150 .
140 .
130
120 -
110 -
100
90
30 -
70 -
60 .
50 -
40 -
30 -
20 -
10 -
170
.160
-150
r!40
-130
120
. 10
1963 1965 1967 1969 1971 1973 1975 1977 1979 1981
;>
"1963-1968 open hearth data included Bessemer furnace production.
Figure.3-3. Steel production by furnace type.7
3-14
-------
The use of electric arc furnaces is expected to continue to increase.
The growth in EAF capacity in 1981 was very large. During 1981, 4.4 xlO6 Mg
(4.9 xlO6 tons) of capacity came on line.12 Estimates of future EAF
growth vary. One source Indicates that EAF's will produce 40.8 xlO6 to
42.6 xlO6 Mg (45 xlO6 to 47 xlO6 tons) of steel per year, or 30 percent
of the steel made in this country, by 1985.16 Another forecast calls
for an annual growth rate of 5 percent, reaching a level of 44.4 xlO6 Mg
(50 xlO6 tons) per year by 1990.17,18 The most optimistic prediction
calls for an even more prolific future with EAF's producing half the
nation's raw steel by 1990.16
For this document, the projected EAF production share of the total
domestic raw steel production in 1987 (119.7 xlO6 Mg [132 xlO6 tons])
will be 31.5 percent. It is projected that the total EAF capacity in 1987
w-m be 44.4 xlO6 Mg (48.9 xlO6 tons) and that with an 85 percent utiliza-
tion rate EAF's,will produce 37.7 xlO6 Mg (41.6 xlO6 tons) of raw steel
in 1987.1S These estimates are conservative compared to other growth
projections and may prove to be low when the economy improves as the
recession ends. In addition, if import restriction are placed into
effect, the domestic raw steel production should improve.
Growth in the use of AOO vessels is expected to be small in the
stainless steel industry since most of the existing steel mills that
produce stainless steal already have ADD vessels. Any new AOD vessels
will be part of a new steel plant rather than retrofits into an existing
shop. The vendor that supplies AOD vessels estimates that one or two
vessels per year will be installed in the specialty steel industry.19
This number would be higher if the use of AOD vessels was increased in
the shops that produce alloy steels and no stainless steels.20
3.2 PROCESS FACILITIES AND THEIR EMISSIONS
3.2.1 Electric Arc Furnace Components
The electric arc furnace process for steel making was developed in
Europe during the 1800's. The first successful commercial installation
was placed in operation in France in 1899 by Paul Heroult. The first
EAF installation in the United States was at the Halcomb Steel Company,
Syracuse, New York, in 1906.21
3-15
-------
The direct electric arc furnace used today is a refractory-lined,
cylindrical vessel made of heavy, welded steel plates and having a
bowl-shaped hearth and a dome-shaped roof (Figure 3-4). Many of the new
furnaces use water-cooled wall panels to reduce refractory costs and
cool the surrounding working environment. Three graphite or carbon
electrodes are mounted on a superstructure above the furnace land can be
lowered and raised through holes in the furnace roof. The etectrodes
convey the energy for melting the scrap charge. Water-cooled glands are
provided at the holes to cool the electrodes and minimize the gap between
the electrodes and roof openings to reduce fugitive emissions', noise
levels, electrode oxidation,-and heat losses. The furnace is! usually
mounted on curved rocker trunnions. Hydraulic cylinders or electro-
mechanical means are used for tilting the furnace.
The main items of electrical equipment are a circuit breaker, a
i
step-down transformer, and, for small transformers, a tapped reactor to
give arc stability and to dampen current surges. The transformers that
regulate the electricity to an EAF are designed for connection to a high
voltage supply. The transformer is provided with equipment to give a
range of secondary voltages ("taps") to suit the melting, superheating,
and refining conditions in the furnace. The transformer taps are changed
in preprogrammed steps td provide the necessary voltage for the melting
and refining of the scrap. ;
The electrodes are raised or lowered by electromechanical or
electrohydraulic devices. At a given transformer voltage, lowering the
electrodes shortens the arc and increases the current and power input
and raising the electrodes has the reverse effect. Electrode movement
is accomplished by automatic control in normal operation.
A recent development in the use of EAF's has been the ultra-high
power (UHP) furnace. These furnaces utilize much larger transformers
than traditional EAF's. A traditional 100-ton EAF might use;20-megavolt-
ampere (MVA) transformers whereas a 100-ton UHP furnace would use 40- to
50-MVA transformers.16 The typical figure used for determining the trans-
former size for a UHP furnace is 0.55 MVA per Mg (0.5 MVA per ton) of heat
size, compared to approximately 0.22 MVA per Mg (0.2 MVA per ton) of heat
size for a conventional EAF.16 The new UHP furnaces use larger electrodes
3-16
-------
SCRAP
CHAR6E
RETRACTABLE
' ELECTRODES
FURNACE
SHELL
TAP
HOLE
REFRACTORY
LINING
Figure 3-4. Electric arc steel furnace.
3-17'
-------
than those used by normal powered EAF's that allows more power input to
the charge (and thus a faster melting rate) and increases the production
rate (a 100-ton normal power EAF has a heat time of about 3 hours while
a lOO-.ton UHP EAF has a heat time of about 1 to 2 hours). Oxyfuel
burners and oxygen lances may also be used to increase the melt rate in
UHP furnaces.16
The faster melting had a detrimental effect on refractory linings
until the use of water-cooled side wall panels and roofs was;revived.
The use of water-cooled side panels and roofs allows for faster melting
without the associated detrimental effects on the furnace refractory and
lining. One company utilizes two 317.5-Mg (350-ton) UHP furnaces that
are melting at a rate of 95.3 Mg/h (105 tons/h) each.16 This speed of
melting approaches the speed of other melting processes, which increases
*
the competitiveness of the EAF in the steel market. '.
The furnace refractories used in producing steel are mainly basic.
Acidic refractories in EAF's are typically used only in steeil foundries
and forging shops. The bottoms of basic-lined arc furnaces used in the
steel industry consist of a burned magnesite brick subheartlr with a
working surface approximately 30.5 centimeters (cm) (12 inches [in.])
thick of magnesia rammed material. Basic arc furnace roofs jare generally
constructed of high-alumina brick with high-alumina rammed or castable
materials for the center section around the electrodes.21,221
With the electrodes' raised, the furnace roof can be swung aside to
permit the charge materials to be dropped into the furnace (top charged).
Additional alloying agents, as required, are added through the slag door
of the furnace or through a separate hole in the furnace roof. Top
charging of materials is the most economical method because a furnace
can be completely charged within a few minutes. '
Openings are provided on both sides of the furnaces (the tap hole,
and the slag door). The temperature of the molten bath is checked and
samples for laboratory analysis are taken through the slag door. For
furnace operations that require oxygen blowing, the oxygen liance can be
inserted through the partially open slag door or through a separate
i
opening in the side of the EAF. Oxygen lances supply oxygen that is
used as a fuel gas to increase the rate of scrap melting and speed up
3-18
-------
the oxidizing of carbon and silicon and the impurities in the molten
steel.16,21
3.2.1.1 EAF Operation. The production of steel in an EAF is a
batch -process where "heats" or cycles range from 1 to 5 hours, depending
upon the size and quality of the charge, the power input to the furnace,
and the desired quality of the steel produced. Each heat consists of.
charging and backcharging, meltdown and refining, and tapping. Cold
steel scrap and sometimes DRI are charged to begin a cycle, and alloy
materials and fluxing agents are added for refining. Direct reduced
iron is produced from iron ores that are reduced in the presence of
excessive quantities of a reducing agent (natural gas, noncoking bituminous
coal, anthracite, lignite, etc.) to produce low carbon iron which is
used as melting stock along with scrap iron and steel.2 The DRI is used
as a scrap supplement and as a diluent for residuals in the scrap. Many
of the new electric shops are designed to allow for continuous DRI
charging through a slot in the roof or side wall. The use of DRI is
currently limited in the United States because of the high cost and the
availability of the primary reducing agent, natural gas, and because of
the relatively low cost and adequate supply of scrap. Currently there
are several demonstration plants in the United States that produce DRI
with coal as the reducing agent.23 The coal-based reduction.process may
provide a more economical means of producing DRI.
There are circumstances that may delay the operation of an electric
arc furnace. "Sheds" (the cutting off of the power to the electrodes)
occur frequently and might not be considered unusual operating delays.
Other delays that might be considered typical, are:
1. Lack of room on the ingot floor (teeming aisle delay);
2. Malfunction of an ADD vessel or continuous casting unit;
3. Delay or malfunction of the crane;
4. Failure of the electrode lifting mechanism;
5. Improper arcing of electrodes; ••
6. Loss of plant power; and
7. An inoperative furnace roof which cannot be opened or closed.24
3.2.1.1.1 Charging. Iron and steel scrap are loaded into a drop-
bottom (clam-shell type) charge bucket with an electromagnet that is
3-19
-------
suspended from an overhead crane. The charge bucket.is filled to a
specified weight and weighed on a.scale that has a digital display that
is observed by the crane operator. When the roof of the furnace has
been opened, charging is normally performed by carefully dropping the
charge into the open arc furnace from the charge bucket. Some smaller
furnaces are charged with scrap directly from the suspended electromagnet
and do not utilize a charge bucket. :
All steel plants except one charge cold scrap to the electric
furnaces. One melt shop routinely charges blast furnace metal to the
EAF's, and the molten metal is 36 to 40 percent of the total 'charged
material. Molten steel can be refurnaced (i.e., the molten steel is
tapped into a ladle and then the steel is poured back into the furnace).
Refurnacing may be done to salvage an off-grade or cold heat batch or to
keep the steel hot due to a production delay or breakdown elsewhere.21
A large variety of scrap is charged to EAF's. According to the
Institute of Scrap Iron and Steel, all grades of scrap are tq be almost
free of dirt, nonferrous metals, and foreign material of any kind.
Carbon steel shops typically use No. 1 and No. 2 grades of scrap, while
specialty shops typically use No. 1 scrap, stainless scrap, and alloys
such as ferromanganese, ferrochrome, high carbon chrome, nickel, molybdenum
oxide, aluminum, manganese-silicon, and others.
Scrap size and bulk density vary from light scrap, such as machine
turnings, to heavy scrap, such as ingot butts. The charge bucket is
usually loaded with light scrap at the bottom for two reasons: to
provide a cushioning for the impact of the charge on the bottom of the
EAF and because the light scrap melts quicker than denser scrap which
aids in forming a molten metal pool in the bottom of the furnace.22
Alloying materials that are not easily oxidized (such as copper, nickel,
and molybdenum) and lime are charged before, or along with, the scrap
metal charge. The lime is a fluxing agent to reduce the sulfur and
phosphorus content in the molten steel. Depending on the desired carbon
content of the steel and the finished product specifications,, iron ore
and coke may be charged prior to meltdown.
Charging the open furnace produces emissions that are difficult to
control. The intensity level of emissions during charging varies depending
3-20
-------
on the cleanliness and the makeup of the scrap. Most charging emissions
result from (1) vaporization of oil, grease, or dirt introduced with any
turnings, borings, or chips; (2) oxidation of organic matter that may
adhere, to the scrap; and (3) the vaporization of water from wet or icy
scrap.22,24 Charging emissions are made up of particulate matter,
carbon monoxide, hydrocarbon vapors, and soot.24 the carbon monoxide is
quickly oxidized to. carbon dioxide in ambient air. Backcharging produces
a large eruption of reddish-brown fumes with a strong upward thermal
driving force. The emissions during backcharging are higher than during
the initial charge because of the intense reaction that occurs due to
the heat of the molten steel bath in the furnace.
During the charging process, the scrap must be introduced into the
furnace so that there is no damage to the refractory. If scrap pieces
remain above the furnace ring, the pieces must be repositioned so that
the roof can swing back into"place for meltdown. This repositioning can
be done by hand or by compressing the scrap with the charge bucket or
other large mass of metal suspended from the crane. An oxygen lance is
sometimes used to cut any pieces blocking the roof. After the roof is
rotated into place, it is lowered onto the furnace in preparation for
meltdown. Repositioning of the scrap delays the closing of the roof,
allowing more emissions to escape from the furnace.
3.2.1.1.2 Meltdown and refining. After the roof is in place, the
electrodes are mechanically lowered to within 2.5 cm (1 in.) of the
scrap, and the power is turned on.21 When the current is applied to the
electrodes, the electrodes are slowly lowered by automatic controls
until they touch the scrap. During the first 3 to 5 minutes, an inter-
mediate voltage is applied to the charge to allow the electrodes 'to bore
into the scrap, which, in effect, shields the sides and roof of the
furnace from the heat of the arc.21 Melting is accomplished by the heat
supplied by direct radiation from the arcs formed between the electrodes
of the furnace and the metallic'ch'arge, by direct radiation from the
furnace lining, and by the resistance of the metal between the arc
paths. The arcs melt scrap directly beneath and around the electrodes,
"boring" through the scrap charge and forming a pool of molten metal on
the furnace hearth.21,25 The molten steel pool, in turn, enhances
3-21
-------
meltdown by the radiation of heat from below into the cold scrap. After
the initial period, the maximum voltage is applied in order tp melt the
charge as fast as possible. Before the scrap is entirely melted, a bank
of refractory material (such as dolomite) is built in front of the
slagging door to prevent the molten steel from spilling out the door.
When the initial scrap charge is almost entirely molten, a backcharge
i
of scrap may be added to the furnace (in some shops there mayi be more
than one backcharge). Following the backcharge, the roof is replaced,
and electrodes are lowered and energized to melt the scrap. Near the
end of the meltdown, oxygen lancing may be performed.
Oxygen lancing in arc furnaces is used mainly for adjusting of the
chemistry of the steel, for speeding up the melting process, and for
superheating the bath. Oxygen lancing results in increased b&th and gas
temperatures, gas evolution, and generation of particulates. Oxygen is
now used almost universally (instead of iron ore or mill scale) for
"boiling" a heat of steel to flush out gases, mainly hydrogen! and nitrogen.
Oxygen may increase the steel temperature without the arcs because the
carbon boil reaction is exothermic. Oxygen lancing can be carried out
with moderate rates of oxygen addition, thereby avoiding excessive
generation of high temperatures, gas evolution, and particulates.
However, extended periods of oxygen lancing can increase refractory wear
and oxidation of the bath but at the same time increase the production
rate. ;
During the meltdown, phosphorus, silicon, manganeses carbon, and
other elements in the scrap metal are oxidized. Slag formation begins
and is carefully monitored during the meltdown stages to control the
chemical concentration and product quality. Basic EAF's use leither
single.or double slagging operations depending upon the desired quality
of the end product. The single slagging process uses an oxidizing slag
that is formed by the addition of lime and coke breeze (or other source
of carbon) during the initial scrap metal charge. Other flux additions,
such as fluorspar, silica, and ferrosilicon, may be made through the
slag door. The carbon reacts with the calcium in the slag to form
calcium carbide, which makes the slag basic.22 The oxidizing treatment
3-22
-------
under a basic slag removes most of the phosphorus and carbon from the
melt, thus lowering the concentrations to the desired level.24
The double slagging process develops as oxidizing slag first,
followed by a reducing slag. The initially formed oxidizing slag is
raked off, with the power to the electrodes cut off, and is followed by
additions of burnt lime, powdered coke, fluorspar, silica, sand, and
ferrosilicon.21 The purposes of the reducing slag are: (1) to return
the reducible oxides, such as those of manganese, chromium, vanadium,
tungsten, iron, etc., from the slag to the metal; (2) to eliminate the
sulfur as calcium sulfide; and (3) to finish the steel to the specified
composition.21,24 Prior to the metal tap, the reducing slag is poured
off into a slagging pot or onto the ground.
During the meltdown operations, the emissions consist of (1) metallic
and mineral oxide particulates .generated from the vaporization of iron
and the transformation of mineral additives; (2) some carbon monoxide
from combustion losses of the graphite electrodes, carbonaceous additives,
and the carbon content of the steel; and (3) hydrocarbons from the
vaporization and combustion of oil and impurities remaining on the scrap
charge.24 Fluoride and trace constituents, such as nickel, hexavalent
chromium, lead, cadmium, and arsenic, are emitted from EAF's.26,27 The
carbon monoxide is combusted where the exhaust gases are exposed to
ambient air, i.e., the electrode ports and the off-gas duct.
During the melting process, emissions escape through the electrode
holes, the slag door, the roof ring, and sometimes the tap spout.
Furnace evacuation with direct-shell evacuation control (DEC) can control
most of these emissions by maintaining a slightly negative pressure
within the furnace. Emission capture and control techniques will be
discussed in Chapter 4.
• 3.2.1.1.3 Tapping. After the proper temperature has been reached
and the steel composition has been adjusted, the molten steel is tapped
from the furnace into a ladle. To tap a heat, the power is shut off and
the electrodes are raised sufficiently to clear the bath. The furnace
is tilted (sometimes as much as 45 degrees), and the molten steel is
tapped into a ladle. The ladle is placed close to the tapping spout to
capture the batch of steel without excessive splashing and to reduce the
3-23
-------
exposure of the molten steel to the air and thus minimize excessive
oxidation and cooling of the steel. Additions of ferromanganese, ferro-
silicon, and aluminum are sometimes made to the ladle to adjust the
oxygen-content of the steel.21 Depending upon the final product require-
ments, various alloying agents can also be added to the ladle; These
alloying agents include aluminum, titanium, zirconium, vanadium, and
boron.21,22 Specific amounts of alloys are added manually to the molten
steel stream during the tap. For certain steel alloys, chrome is added
just prior to the tap to avoid oxidation of the chromium durihg meltdown.22
During tapping, fumes consisting of iron oxides are generated; in addition
to oxide fumes resulting from alloys that were added to the ladle.
After the molten steel is tapped into the ladle, the ladle is transferred
to either an ingot teeming area, a continuous caster, or a refining
vessel (in a specialty steel shop).
The bottom-tapped EAF, recently introduced in West Germapy, is a
technological change in the use of EAF's. Only one bottom-tapped furnace
is in operation but the cost-saving aspects may be an impetus: for worldwide
distribution of these furnaces. This bottom-tapped furnace was
built by Mannesmann-Demag Metal!gewinnung of West Germany to enable the
installation of water-cooled panels further down the furnace ;sidewall.16
The company claims that money is saved through reduced capital investment,
operating costs, refractory consumption, repair and maintenance, electrode
usage, and pollution control expenses. In this furnace, wheq the metal
has been melted and is ready to be tapped, a closing flap is swung out
of the way, and the sinter plug is pushed out by the weight of the
molten steel. If the plug does not come out, an oxygen lance is used to
remove the plug. The molten steel is tapped, then the slag is tapped
into a slag pot. Some slag is left in the furnace to protect the refractory
from the thermal shock of starting a new melt.and the impact1of the
subsequent charge. Since the furnace does not tilt, electrodes last
longer because they are not subjected to stress, and a DEC system is
operational during the tap. The greater coverage achieved by larger,
water-cooled side panels and a water-cooled roof greatly reduces the
refractory replacement requirement.16
3-24
-------
3.2.1.2 Material Balances. The steel yield from basic refractory
electric arc furnaces is relatively high. Approximately 88 to 92 percent
by weight of a typical total material charge (i.e., scrap, alloys and
fluxes-) is returned as product and 8 to 12 percent is poured off as slag
or escapes as particulate matter.28-36 The amount lost as dust varies
from 0.6 to 2.9 percent of the.metal charged.37 Figure 3-5 provides a
general diagram of the material balance for steel production from an
EAF. The amount of particulate matter collected varies from 7.5 to 20 kg/Mg
(15 to 40 Ib/ton) of steel produced.36 The remainder of the lost metal
is poured off with the slag. Table.3-6 provides a summary of the typical
material balance for EAF's and ADD vessels,,,,.___^_
The rate
-------
at
o
=9
CO
-a
•a
o
a.
O)
O)
CO
<+-
o
en
o
o
o
•o
O
•P
LO
O)
3
O)
tw
3-26
-------
(M
CO
1
CO
M
CO
z.
o
t— (
1
\
•p
(Mi
(0
U
0)
a
CO
c
0
a
c
4-
,_
a
a
+-
i/-
a
3
J_
LLJ
rnace
> 3
> U_
^ /-N
co •—
d o
^**
?
- CU
2; JC
•> O "-P
r— C S- - T- O
(O O •>- S- -CO
4j ja o •<— c -
cu i. >o cos: £
E U 3 oE U. -
C C r- "
Q) »i~ •> •• CU C
"o •<- <— CO O -r- •'-
2! — J ro i— SI 21 31
C
1 - O
CU r— 5-
E CU •!-
O CU
S- -P -
J= W r—
O CU
C CU
«^" i— JQ tn to cn LO P"*-
co CU i- O i— O O
O-PU3 o'do'cD
S^^* I/} ^B ^^^ %l*^ *"^^ **1**^
3 o
to o -C ^>
to i— Q. •
O O)
i- -P O 3 CD E C_3 -C
U CO •!- O 1- •!- T- •!- -P
__^
•r- U
(0 -P
S- CU
,x^, « - /— % CM CM /— > LO ID
rom Q. tnoO'— oo
cn 01 to OOOOOCD
tr^ "Q m CJ O O O CD O O
3 C
JD. - u
m -i- j= -P
CU w co CD,
^_^ i*"N ^*N
OO O CM
00 i— O
CD OO
^*^
C
$ la cn-p
i — -p ITS 10
0 CU -—3
2 E s-
C 3 /— »
•r- 4J •
J3 O U
3 3 4->
-P
sT to
i — ro «t3
CU jQ CU C D5-P
CU OJ CL 3 (T3 tO
.p Q£ T- O I — 3
CO ^—^ O. S- CO O
• 3
t1 '
tj
T3
flX ^
(A ^
+* •— *=
(j cu Q-
3 CU 1
•O -P >>
O CO CO
Q_
^^
o
^3.
o
^
CD
1 —
i
jf-
o
^
-p
U
3'
T3
O
i.
O.
•O
4-
U
3
73
o
Q.
t
^
fO
J_
CU
ns
E
3
4-
^
fll
B
^^^
CU
U
(O
-Q
ro
• r-
CU .
E
C
r
-------
The kilovoIt-ampere (kVA) rating of the furnace transformer generally
increases with the size of the furnace. If water-cooled panels are
used, the voltage applied to the EAF electrodes can be increased without
detrimental effects on the furnace refractories. Refractory wear increases
at a rate proportional to the increased production rate when water-cooled
panels are not used.16
3.2.2 Argon-Oxygen Decarburization Vessels and Their Operatipn
The argon-oxygen decarburization process was first develbped by the
Linde Division of the Union Carbide Corporation at their Niagjara Falls
facility.5 Joslyn Stainless Steel in Fort Wayne, Indiana, assisted in
the process development and started a full-scale AOD vessel operation in
1969. Computer techniques assist in optimizing the AOD vessel! operation.
The AOD process is used to produce tool-and-die steels, high-speed and
forging steels, as well as stainless steel and other alloy steels.5
The AOD vessel is used in conjunction with an EAF in specialty
steel shops. The AOD vessel is used to refine the steel that was melted
in the EAF, thus allowing the EAF to be used solely as a scrap melter,
resulting in quicker heats. The AOD vessel refines the steel quicker
and more economically than an EAF. The AOD vessel is a significant
source of air pollution emissions and is frequently ducted to the same
control device as the EAF.
Argon-oxygen decarburization vessels are closed-bottom, refractory-
lined, pear-shaped converter vessels with submerged tuyeres i,n the lower
portion of the vessel (Figure 3-6). The AOD vessel is constructed of
welded steel plate and mounted such that it may pivot for charging,
slagging, and tapping. Argon, oxygen, and/or nitrogen gases ^are blown
through the tuyeres into the molten steel to adjust the bathjchemistry.
The,gases are introduced at various ratios of argon/oxygen/nitrogen at
various stages of the heat to control the metallurgical reaction, control
the bath temperature, and cool and maintain the air passage in the
shrouds and tuyeres. Computer controlled orifice meters or vortex
precision meters are used to control the gas flow rates and mixtures
chosen by the AOD vessel operator.
3.2.2.1 Vessel Operation. The molten steel from an EAF is transferred.
by ladle to the AOD vessel, which rotates forward to accept -j:he molten
3-28
-------
VESSEL TOP
Figure 3-6. Argon-oxygen decarburization vessel.5
3-29
-------
charge. When the charging operation is complete, typically in 1 to
8 minutes depending on the size of the AOD vessel, it is rotated back to
an upright position so that refining can begin.32-34 Before refining
begins-, additives such as lime or alloys are added to the molten steel
with a crane-held charge pan or through a charge chute.
To begin a heat, while the vessel is in a horizontal position,
small volumes of air are blown through the tuyeres in the sides of the
vessel for cooling purposes. As the AOD vessel is turned into an upright
position, argon is blown through the tuyeres as they become submerged.
The argon-oxygen ratio is approximately 1:3 at the beginning ;of a heat,
and the argon concentration is raised throughout the heat until the
final ratio is 3:1. In many AOD operations, nitrogen gas is |used in
addition to argon, since nitrogen is less expensive than argon. However,
for some specialty steels, the nitrogen concentration in the isteel must
be kept low, and nitrogen is not used in these heats.5 The process gas
flow rates depend upon heat size and design of the vessel. The gas
consumptions vary from shop to shop with argon fluctuating fr|om 9.9 to
19.6 cubic meters/megagram (ms/Mg) (390 to 770 cubic feet per ton [ft3/ton])
of steel produced, nitrogen from. 2.8 to 13.7 mVMg (112 to 540 ftVton),
and oxygen around 18.8 ms/Mg (700 ftVton).32,34,42 Refining is accom-
plished by blowing argon, oxygen, and/or nitrogen gases through the
molten steel bath. The control of the gas mixture and flow is important
to avoid the oxidation of alloys necessary for specialty steel production.
Refining in an AOD vessel generates a dense cloud of emissions with a
strong thermal driving force. As the heat progresses, alloys and fluxing
agents are added to the molten steel in quantities that are determined
by the chemical analyses performed on samples of the bath. The fluxing
agents are typically lime and fluorspar, and the alloys include aluminum,
chrome, nickel, manganese, boron, silicon, vanadium, and .titanium.
Limited amounts of scrap generated at the steel mill (home scrap) may
also be periodically added to the vessel as additives to help reach the
desired chemical makeup of the final product.
The carbon-chromium equilibrium relationship is very important in
controlling the quantity of chromium in the final product. [The amount
of chromium in the melt is in an equilibrium relationship with the
3-30
-------
carbon. The carbon level is decreased with oxygen blowing; however, any
excess chromium may also be oxidized and lost. The amount of chromium
that the melt can retain decreases as the carbon content of the melt
decreases. Before the AOD vessel was used, the carbon level was reduced
by oxygen lancing in the EAF, and expensive low-carbon ferrochrome was
added at the end of the heat.5 The carbon-chromium relationship (Figure 3-7)
determines the extent to which decarburization (carbon removal) will
occur before the chromium is oxidized.5 Raising the temperature of the
melt will lower the limits of decarburization without reducing the
•chromium content. However, the increased temperature has a detrimental
effect on furnace refractories. If an inert gas is injected into the
AOD vessel along with oxygen, the partial pressure of the carbon monoxide
gas is reduced and~the carbon monoxide is diluted. These two conditions
increase the rate of the decarburization process while greatly increasing
the amount of chromium that is retained in the molten steel.5
The oxygen flow rate is programmed to maintain the bath temperature
through an exothermic reaction. Oxygen is also used to reduce the
carbon content of the bath.6 Just prior to the tap, pure argon is blown
through the molten bath to assure uniform temperature distribution and
to reduce the bath temperature (for improved steel quality) by decreasing
the dissolved gas and oxide content.5,6 Furthermore, the argon is blown
through the molten steel to perform effective mixing of the slag and
steel in order to reduce metallic oxides from the slag and to decrease
the dissolved gas, oxide and sulfur levels of the molten steel.
3.2.2.2 Material Balances. The steel yield from an AOD vessel is
very high. About 91 percent by weight of a typical charge of molten
steel and fluxes to an AOD vessel is returned as product (specialty
steel).28,32,34 The metallic yield, i.e., the steel tapped as a percent
of the metal charge, is approximately 97 percent. Table 3-6 presents a
summary of the material balances for AOD vessels along with EAF's. In
specialty shops, the EAF molten steel is typically of a higher grade _
than that produced in common steel shops; however, to reach the desired
concentrations of alloys in the final product, several alloy additions
are made throughout the heat. The typical quantity of steel charged
3-31
-------
C
O)
o
-------
that is lost as dust is 8.0 kg/Mg (16.0 Ib/ton), and the remainder of
the lost metal is poured off with the slag.26,27
3.2.2.3 Energy Considerations. The main source of energy for
refining the molten metal in the ADD vessel is the exothermic oxidizing
reaction that is promoted by the injection of oxygen gas. Energy is
required only to operate the compressors that inject the gases and to
operate the tilting mechanism for the vessel. Furthermore, some electricity
is used to operate cranes, fans, pumps, and the emission control system.
3.2.3 New Technology
Technological changes and improvements are continually being made
to improve the speed, cost effectiveness, and flexibility of EAF steelmaking.
Several of the changes that are being introduced at the time of preparation
of this document are listed below.' These changes will need to be addressed
when the next review of the NSPS occurs. No emissions data are available
for any of the described changes because of their relative newness in
this county and their limited usage at this time.
One of these changes is the use of DRI in the production of steel.
This relatively new technology is not being used to a great extent at
present because of low scrap prices and the high cost of producing DRI;
however, it could see more use in the future. The fugitive emissions
from charging would be greatly reduced because the DRI would be continuously
fed into the furnace. The quantity of dust generated will have to be
determined because no plants using DRI were observed during this study.
Recyling of the EAF and AOD dust is presently practiced by many
stainless steel producers for the recovery of nickel, chromium, and
iron. In addition, several EAF plants have their dust recycled for the
zinc content or pelletize the dust and feed it back to the furnaces.
Numerous studies are underway to explore all possible means of recycling
the dust to reduce the loss of iron oxides and valuable alloys and to
reduce the impact of the Resource Conservation and Recovery Act (RCRA)
requirements. Recycling the dust will help offset some of the costs for
pollution control equipment.
Energy recovery is also being attempted on a trial basis at several
steel mills. The energy is recovered from the water that circulates
3-33
-------
through the water cooled roof and well panels.43 The approximate amount
of the energy in the water cycled through the furnace wall panels and
furnace roof approaches 16 percent of the total energy input..44 In
addition to this, some of the heat from AOD vessel off gases :can be
salvaged with a heat exchanger. The energy could be used for heating
purpose within a melt shop or in buildings near the melt shop. The
recycled heat will improve the overall economy of EAF's and AOD vessels.
New ways to utilize EAF's are currently being investigated. One of
these techniques, the bottom tapped EAF, has already been discussed.
This type furnace would require a smaller capital outlay and it would be
easier to control the air pollution emissions since the tapping emissions
of the furnace could be captured with the primary emission capture
systems. Another change to the furnace that is receiving experimental
use is the use of greater power to the furnace, raising the transformer
rating to nearly 1 MVA per megagram of furnace capacity.16 this would
reduce the heat times below those used in existing ultra-high powered
(UHP) furnaces. ;
New methods or combinations of methods for refining the;steel from
EAF's, in addition to the use of AOD vessels, are currently being introduced.
These secondary refining processes, collectively known as ladle refining,
are being used to reduce the amount of refining that is performed in the
EAF, allowing the EAF to be used strictly as a metal melting^device.
Argon bubbling is one technique where a gentle stream of arg6n is injected
through a porous plug at the bottom of the ladle. This technique improves
the quality of the steel and controls the composition of low-alloy and
medium alloy grades of steel.44 Another technique is vacuum-oxygen
decarburization (VOD) that is used to reduce the carbon content in the
steel without oxidizing the chromium. A consumable oxygen lance is
inserted into the molten steel through the ladle cover. While the ladle
is under reduced pressure, oxygen is blown into the melt. After the
desired amount of oxygen is blown, the vacuum is continued while argon
is bubbled through the melt so that the oxygen remaining in the steel
can react with the remaining carbon. This technique takes 2\ to 2.5 hours
to refine the molten steel.
3-34
-------
Vacuum arc degassing (VAD) is another ladle refining technique.
The ladle cover is equipped with three electrodes that heat the molten
steel to 1680°C (3050°F). After the molten steel reaches the desired
temperature, the heat is degassed, desulfurized (after the lime and spar
fluxes are added to the molten steel), and alloy corrections are made.45
Holding the steel under low pressure with argon stirring serves to
'promote the oxidation of unwanted carbon and to purge the hydrogen gas.
TMs ladle refining technique also requires 2 to 2.5 hours. Desulfurization
can also be performed by the lance injection method, in which desulfurizing
reagents are pneumatically injected deep into the bath.45 This method ;....
is performed with the ladle under reduced pressure and while argon gas :
is bubbled through the molten steel. The advantage to this technique of.
desulfurization is the short treatment time of 5 to 10 minutes.
3.3 EMISSIONS
3.3.1 General
The quantity and type of emissions from an electric furnace depend .
upon many factors: furnace size, type and composition of scrap, quality;
of scrap, type of furnace, process melting rate, number of backcharges,
refining procedure, and tapping duration and temperature.24 The majority
of the emissions from EAF's are particulates, both ferrous and nonferrous
oxides. Furnace emissions are the highest during meltdown and refining
operations, but charging and tapping emissions can also be significant,
especially if ladle additions are made during the tap and dirty scrap is
charged. The charging and tapping emissions represent approximately
5 percent each of the total emissions during a heat.24 Increases in
electrical power to the furnace and the use of oxygen landing will cause
emissions to increase during meltdown and refining.
; Electric arc furnace emissions are classified as process or fugitive.
Emissions generated at the furnace during periods when the furnace roof
; is closed (e.g., during melting and Defining) and the primary emission
capture device (e.g., DEC system, side draft hood) is'Operative are
considered to be process emissions. Those emissions generated during
periods when the furnace roof is open (elgf, charging) or when the
3-35
-------
primary emission capture device cannot operate (e.g., charging and
tapping) are considered to be fugitive emissions.
i
The emissions from an ADD vessel are primarily particulates of both
ferrous and nonferrous oxides. The quantity and type of emissions from
an ADD vessel depend upon several factors: the quality of tlie molten
steel charge, the quality of the final product desired, and the types
and quantity of alloys added. Almost all the emissions occur during the
blowing (refining.) stage, with the greatest emissions occurring when the
concentration of oxygen in the gas stream is the highest at the beginning
of the heat. When the AOD vessel is in a tilted position fovj temperature
checks and sample-taking, there are almost no emissions because no gases
are blown through the molten steel. The charging and tapping emissions
are minimal because the charge is made to an empty vessel, arid the tap
occurs after the carbon content has been greatly reduced. Table 3-7
presents a summary of the emission factors for EAF's and AOD ivessels.
Table 3-8 presents a summary of the trace constituent emission factors... •
The chemical composition of the typical EAF fume during ^various
stages of a melt is presented in Table 3-9. Iron oxide is th,e main '.:;••
component of the EAF fume, with a large amount of calcium oxide emitted >
during refining and a large amount of manganese oxide emitted during ;
charging. The exhaust gas particulate composition for both EAF's and f
AOD vessels is presented in Table 3-10. The distribution of :the particu-
late matter in EAF and AOD vessel fumes indicates that the particles are
quite small. A particle size distribution is presented in Table 3-11.
A majority of the particulates are in the inhalable size range (less
than 15 micrometers [urn]).
3.3.2 Baseline Emissions :
The baseline emission level is the level of emission control reached
by the furnace shop in the absence of additional standards. The baseline
is used to evaluate the incremental environmental, economic, |and energy
impacts associated with the regulatory alternatives selected^for analysis.
The baseline emission level for EAF's is the existing new source performance
standard (NSPS). ;
3-36
-------
TABLE 3-7. PARTICULATE MATTER EMISSION FACTORS
(UNCONTROLLED)1,24,26,27,36
Melt and, refine
Charge and tap
EAF's
kg/Mg
7.5-20
0.8-1.0
Ib/ton
15-40
1.6-2.0
AOD
kg/Mg
8.0
--.
vessels3
Ib/ton
16.0
?Draft test data, includes charge and tap emissions.
With alloys added to ladle.
3-37
-------
CO
CtL
O
0
LUM
LU -
M S
1— LU
CO — 1
Z —1
O 0
o D; -
LU Z
o o
0£ Z
1- =3
^
CO
1
co
UJ
CO
-C •(->
UJ -P ^>
CO CU -Q
> CTr—
O
Qi ^
•r*
o
^C ™^3
CU
•a r—
C i—
03 O OJ
UJ [ ^ ^*,
- C OJ
Lt_ 0 -^
r— 3
*f UJ i™™
Z CO Li_
r— CO LfJ
CO i— CM
• • •
O O O
i — CO CO
CO CO.. r—
O
0 • 0
o
r>. en co
CO CO CO
o
0 • 0
o
co en en
^ r— • r—
0
0 • 0
•o
1 1 1
1 1 '
1 1 1
1 1 1
PS
r;
•r— r— •
E CU
o -a ^.
i- ro ,U
fm CU *f™
O — J Z
3-38
-------
a
03
C
03
O
5-
C
C
O
•p
c
a.
g
L
•P
01
™
O
CO
O
a.
C>4
O .
to
CO
o
S-
o
0
c
2:
CO
o
fHH
1™"" ^^ **J"
CO ;f— r—
o d CD
in o r--
f^ 0 CO
co co in
vo r»> co
«?t* co co
d <=> <-^
G*i ^"^ €2
CO CO r—
co co co
psi VO CSi
r>- r- <*•
en o CM
0)
c
ft**
o
c
O) CO
C i—
O5 -i-
C N ' C
+J "O O5
"a! x x
200
LO
in
o
LT)
in
p*^
co
m
o
o
r-.
•
CO
m
O
o
CO
co
OvJ
^^i
p^»
CM
csj
CM
in
.co
.
i-
f—
C
o
•I—
'-P
o
=5
•a
02
•
CO
O
O4
a)
LL.
O
-a
.p
&..
(U
>
o
o
•o
c.
fO
c
o
i.
•r"
'«
o
+*
(ft
(0
n
CD
C
0)
*J
•o
I/)
(O
^J
c
cu
-------
TABLE 3-10. EXHAUST GAS PARTICULATE
MATTER COMPOSITION22,26,27,32,46
(Percent)
Constituent
Fe203
CaO
A1203
Si02
MgO
Mn203
ZnO
NiO
Cr203
CuO
MnO
W03
Mo03
Cu20
Cl
V205
Ti02
PbO
Nb203
FeO
C
P
S
Na20
LOIC
Other
Process
EAFa
19-53
3-14
1-13
0.9-9
2-15
0.6
0-16.3
0-3
0-14
0.1
0.6-12
—
—
—
1.2
—
—
0-4
—
4-10
•
—
—
1.5
4.3-6.8
4.8
AODb
--
7.4
1.6
8.9
3.2
--
3.4
3.1
11.4
15.6
0.2
0.9
0.4
0.4
0.1
0.8
1.2
0.1
34.4
1.7
0.1
0.7
;:
3.9
Carbon steel.
.Specialty steel.
'Loss on ignition.
3-40
-------
TABLE 3-11. SIZE DISTRIBUTION OF PARTICULATE MATTER
EMISSIONS FROM STEELMAKING EAF AN§ ADD FACILITIES22,26,27
Particle size
range, \M
<0.5
0.5-1.0
1.0-2.5
2.5-5.0
5-10
10-20
20-40
>40
Size
EAF
—
57-72
— '
—
8-38
3-8
2-15
0-18
distribution, percent
EAF + AOO
49
8
10
7
6
20b
—
--
by weight
ADD
15
22
29
10
5
19b
— •"
aParticles smaller than 15 pm are considered to be
.respirable or in the respirable categories.
Larger than 10 pm.
3-41
-------
The existing NSPS is defined below: ;
1. Particulate emission limit, 12 mg/dscm (0.0052 gr/dscf)*;
2. Opacity limit from control device, <3 percent (based on a
6-minute average);
3. Opacity limit from shop, zero percent, except:
a. During charging, the opacity could be greater than zero
but less than 20 percent (based on a 6-minute average);
b. During tapping, the opacity could be greater than zero but
less than 40 percent (based on a 6-minute average); and
4. Opacity limit from dust handling system, <10 percent Abased on
a 6-minute average). j
Process weight regulations are commonly used to determine regulations
for sources of particulate matter emissions such as AOD vessel's. The
emission level for AOD vessels, which are not regulated by the existing
NSPS for EAF's, is the emission rate determined by one of the process
weight equations listed below: |
1. E = 4.10 P°-67 for P ^30 ;
E = 55 po.11-40 for P >30 !
2. E = 3.59 P0-62 for P ^30
E = 17.31 P0-16 for P >30
In equations 1 and 2, -
P = process weight rate in tons of raw material input per hour
E = allowable particulate emission in pounds per hour
Table 3-12 presents a summary of the air pollution regulations for
the States that have steel mills with AOD vessels. In addition to the
process weight regulations, most States have a visible emission regulation.
Typically, sources are required to maintain less than 20 percent opacity
except for one 3- to 6-minute period every hour when average opacity
readings can be as high as 40 or 60 percent. Fugitive emission levels
are also controlled by most States. Typically, no visible emissions
from fugitive sources are allowed at the property line.
*mg/dscfm = milligrams per dry standard cubic meter.
gr/dscf = grains per dry standard cubic foot.
3-42
-------
z
o
HH
1 _LT|
3
O
CD
LU
C£
2:
0
1— *
r~"
— 1
0
Q_
"^
U_
o
>—
C£
£
rs
to
C\j
1
oo
LU
— 1
ca
t c
•i- 0
"cL'-M
CLIO
fH
41
O 4J
X 0
41 •—
O 10
ro
VII O
cn ro
CM *
«a- o
r— CM
•I- .O U
1/1
C
o
in
•o 01 in
CM CO •<-
Ol 01
w in 41
<= c S
O O •!-
•r- »r- V)
+J 4J -t-
10 <0 >
3 '3
cr cr o
LU LU Z
in
o
c
t— 4
§
Ol
^ :
tf)
"g 0
IO 03
cn
ro 1—
~
I
8
^
i
ro
»
o
i
i
LO
ro
ift r~
Of—
C
o
a.
S
X
41
o" s
CM 10
v co
0,
CM
*
r—
Ol 01
4)
»0
c
o
in
I
r-
-
XI .Dl U
—
0
'^3
a.
01
u
X
41
^ "
*^* fj
*y in
"41^: S
in cn o
c «*•
g V« 01
crs: o
UJ O-Z
ro
C
>
^
in
c
c
01
a.
i— i
O VD
o cn
i f~*
o
o «
T§
ro o
*" 0
C-3 ^t"
^R
U)
O1O
c
o
c.
01
u
X
41
8
•r-
•^
cn
0
in
•a
CD
o
o
VII
s:
a.
o
cn
c
'.£
1
o
CM
1
IO
cn
S
c
1
c
'£ •
C 41
i*- O
VII in
ro
S
o
CM
XI U
01
XI
10
c
. in
ro
SI
4J
QJ
Table"1
Controlled to ex
(B
•r-
£-
>
-i->
U)
. *^
g;
O
.£
ro
Si «
41 01
O • i- *J
C 3 10
•r- O +J
in -*J .c: LO
u ^-,
it- -a 3 « s- c c
o 01 -a *> o •>-•'-
in *J - u> S
•a . '" rr> O 1-
o n -x. o^c v Q- 01 H-
2.C OS.C^All'o^
ct Jr? m jm *^~ d UJ f
c 'O s- •!- o o o> s-
O 5! C 4-1 O S m M~ 01
UO10 O Olr*- 4J+-1
*— cn 10 c it— cr c:
40 XlfSt ••- r— O in 3
ratnu • in4Jja<*-T3i— *J
r- rO -f-O X in O ro -^ if- •<-
3 D)S OS OI3SLUt-i*-e
LJ LL. O "O O O) 41 •'"
•r— c a 10 o o *-" o **"
4J 0 -r- r-- II S. !- r- 0 CM II
S_ .O.O.IOOO C
lOtnaoce « • iu o
Q.TJ O || || II VO 0 -f-
C «C II 41 41 in
|| 41 !- LL. 3! LU II II t- UJ
a. ai LU a • srs^'i
at d CD 4^ "» a. a. 3= uj
JC -r- -r-JJi r— S
^
U
jS.
>fc^. t-
r— O
M- in
c
41 O
4-1 4->
10
i_ C
•r-
•2 4J O O
ooinra LOLOOO • •
VII A "- VII A VII A C C
S 4J 41 41
O-O.41X: Q-CuO-Cu.£E
4-1 41 33
ro 3 • 0 0- cr
i — in ^f 41 01
04 (0 3 in C ^ 1
*o t- u u) • Ort to r-« "i " Co
• . •!— Ol 41 •»"• in »-* tO »^ 41 C
a- a. i- o i- (no o o o •»- 4J
i— o. a. o 'e °~ °~ °~ °~ -5 •"
LO • II II - m • i— • E oi
. r~ on 4) • -a- . in .
ro r— LU a, c >CMCM«TLO-O^
U || a> *J 4J H || || (| ro ro
llJ ' ' f *-cn
3-43
-------
3.4 REFERENCES FOR CHAPTER 3 ' •
1. Background Information for Standards of Performance: Elpctric Arc
Furnaces in the Steel Industry. Vol I: Proposed Standards. U.S.
Environmental Protection Agency. Research Triangle Parki, N.C.
Publication No. EPA-450/2-74-017a. October 1974.
2. Technology and Steel Industry Competitiveness. Congress: of the
United States. Office of Technology Assessment. Washington, D.C.
June 1980. pp. 186-188, 194-201. j
3. New EF Technology Stretches Steel making Capacity With Low Capital
Investment. World Steel Industry Data Handbook/United States.
1/90-95. November 1975. :
4. Brown, J.W. 1980's Decade of Opportunity for Electric Steelmaking.
Presented at Mini Mill Conference. Milan, Italy. 1980.: 16 p.
5. Aucott, R. B«, D. W. Gray, and C. G. Holland. The Theory and
Practice of the Argon/Oxygen Decarburizing Process. Journal of the
West of Scotland Iron and Steel Institute. 79(5):98-127|. 1971-1972.
6. Kurzinski, E. and W. S. Buzzard. Process Gas Control—Argon-Oxygen
Decarburization. Iron and Steel Engineer. 55(6):31-34. > June 1978.
7. Annual Statistical Report—American Iron and Steel Institute—1980.
The American Iron and Steel Institute. Washington, D.C.: 1980.
pp. 55 and 72. '•
8. Telecon. Terry, W., Midwest Research Institute, with Lepinski, J.,
Midrex Corporation. March 2, 1982.' Information on production of
direct reduced iron.
9. "Living Electrically: The Power Behind the Switch-on to;EF Technology.
World Steel Industry Data Handbook/United States. Jh79-89. 1978.
i
10. Nicole, A. G. Electric Arc Furnace Shops in the U.S. and Canada.
Iron and Steel Engineer. 55(11):69-71. November 1978.|
11. Deily, R., Electric Arc Furnaces in Steel Plants, United States,
1982. Iron and Steelmaker. 9(13):25-32. July 1982.
12. Developments in the Iron and Steel Industry. U.S. and Canada—1979.
Iron and Steel Engineer 58(2):D1 to D22. February 1982.
13. AOD: The New Common Denominator in Stainless Steel Making. World
Steel Industry Data Handbook/United States. !_: 109-111.' 1978.
14. 1981 U.S. Industrial Outlook for 200 Industries with Projections
for 1985. U.S. Department of Commerce, Industry and Trade
Administration. Washington, D.C. January 1981. pp. 199-205.
3-44
-------
15 Steel at the Crossroads: The American Steel Industry in the 1980's.
American Iron and Steel Institute. Washington, D.C. January 1980.
pp. 33-41.
16 Technology Leads the Way as Electric Furnace Steel making Heads for
• ' New Heights in the U.S. 33 Metal Producing. 18(7):41-48. July 1980.
17 Telecon. Terry, W. , Midwest Research Institute With Brown, J. W. ,
Union Carbide Corporation. September 14, 1981. Information on
growth projections for EAF's.
18 Letter from Brown, J. W. , Union Carbide Corporation, to Terry* W.
•V. , Midwest Research Institute. February 13, 1981. .Information, on
the growth projections for electric arc furnaces,
19 Telecon. Terry, W. , Midwest Research Institute with Sarlitto, R. ,
Union Carbide Corporation. September 4, 1981. Information on the
use of AOD vessels in the steel industry.
20. Telecon. Terry, W. , Midwest Research Institute with Sarlitto, R. ,
Union Carbide Corporation. September 14, 1981. Information on the
use of AOD vessels in the steel industry.
21. The Making, Shaping and Treating of Steel. United States Steel ,
Pittsburgh, Pa. December 1970. pp. 403, 551, 553, 574.
22.
Sahagian, H. , P. F. Fennelly, and M. Rei. Inspection Manual _f or
Enforcement of New Source Performance Standards— Steel Producing
Electric Arc Furnaces. U.S. Environmental Protection Agency.
Washington, D.C. Publication No. EPA 340/1-77-007. May 1977.
pp. 8-23.
23 Hogan, W. T. Does Direct Reduction Have a Future? Iron and Steel
Engineer. 59(2)57-58. February 1981.
24
Fennelly, P. F. and P. D. Spawn. Air Pollutant Control Techniques
for Electric Arc Furnaces in the Iron and Steel Foundry Industry.
US. Environmental Protection Agency. Research Triangle Park, N.C.
Publication No. EPA 450/2-78-024. pp. 2-1 through 2-9, 2-16.
25. Technical Guidance for Control of Industrial Process Fugitive
' Particulate Emissions. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Publication No, EPA 450/3-77-010.
: March 1977. pp. 84-100.
26. Emission Test Report. AL Tech Specialty Steel Corporation. U.S.
Environmental Protection Agency, Research Triangle Park, N.C.
Publication No. EMB report 80-ELC-7. March 1981.
27. CarTech test report (not finalized).
3-45
-------
28. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R. EPA/ISB. May 18, 1981. Report of source test trip to
Carpenter Technology Corporation, Reading, Pennsylvania.
i
29. Memo and attachments from Banker, L., Midwest Research Institute,
to Iversen, R., EPA/ISB. March 30, 1981. Report of trip to U.S.
Steel, Baytown, Texas. ,
30. Memo and attachments from Banker, L., Midwest Research Institute,
to Iversen, R. E., EPA/ISB. March 26, 1981. Report of trip to
Chaparral Steel, Midlothian, Texas. i
i
31. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R. E., EPA/ISB. May 8, 1981. Report of source itest trip
to AL Tech Specialty Steel, Watervliet, N.Y.
32. Letter and attachments from Andolina, A, V., AL Tech Specialty
Steel Corporation, to Iversen, R. E., EPA/ISB. August 20, 1980.
Submittal of requested information. ;
33. Letter and attachments from Hauris, F. C., Allegheny Ludlum Steel
Corporation, to Iversen, R. E. •, EPA/ISB. September 24, 1980.
Submittal of requested information.
34. Letter and attachment from Clouse, R. L., Armco, Incorporated, to
•Iversen, R. E., EPA/ISB. Submittal of requested information.
35. Letter and attachments from Heintz, J. R., Sharon Steel Corporation,
to Iversen, R. E., EPA/ISB. February 2, 1981. Submittal of requested
information. ;
36. Goodfellow, H. Solving Fume Control and Ventilation Problems for
an Electric Melt Shop. Air Pollution Control Association. Montreal,
Hatch and Associates. June 1980. 22 p.
i
37. Pollution Effects of Abnormal Operations in Iron and Steel Making—
Vol. V. Electric Arc Furnace Manual of Practice. U.S. Environmental
Protection Agency. Research Triangle Park, N.C. Publication
No. EPA 600/2-78-118e. pp. 3-10. !
i
38. Letter and attachments from Dauksch, W. E., Nucor Corporation, to
Iversen, R. E., EPA/ISB. December 26, 1980. Submittal of requested
information.
i p '
39. Letter and attachments from Martin, R. W., Northwestern Steel and
Wire Company, to Iversen, R. E., EPA/ISB. March 16, 1981. Submittal
of requested information. j
40. Letter and attachments from Sussman, V. H., Ford Motor Company, to
Iversen, R. E., EPA/ISB. October 24, 1980. Submittal of requested
information. . :
3-46
-------
41.
42.
43.
44.
45.
46.
Letter and attachments from Hilton, A., Florida'Steel Corporation,
to Iversen, R. E., EPA/ISB. Submittal of requested information.
Letter and attachments from Petrel!a, P., Babcock and Wilcox Tubular
Products Division, to Iversen, R. E., EPA/ISB. November 24, 1980.
Submittal of requested information.
Whittaker, D. A., and A. R. Palmer. Ladle Metallurgy at the Welland
Plant of Atlas Steel. Iron and Steelmaker. 8(6):31-35. June 1981.
Junker, A. Electric Steelmaking-The Bottom Tapping Combined Process
Furnace (CPF), Part 1-Technical Innovations. Iron and Steel Engineer.
59(12):25-28. December 1981.
Gray, R.D. Vacuum Refining and Secondary Steelmaking. Iron and
Steel Engineer. 58(112):44-47. December 1981.
Brough, J. R., and W. A. Carter. Air Pollution Control of an
Electric Furnace Steel Making Shop. Journal of Air Pollution
Control Association. 22:167-171. March 1972.
3-47
-------
-------
4. EMISSION CAPTURE AND CONTROL TECHNIQUES
4.1 INTRODUCTION
This chapter presents the capture and control techniques for EAF
and AOD units. The requirements for emission capture and control equipment
vary with each plant's individual design and operating practices. The
engineering factors that must be addressed when choosing an emission
capture system include the size of the EAF or AOD vessel and the operational
practices of the individual furnace or vessel, such as the oxygen blow
rate, the type and amount of alloys added, the number of backcharges
added to the EAF, the melt rate of the EAF, and the grades of steel
produced. The size, layout, and number of openings in the melt shop
building have an impact on the choice of which emission capture arrangement
and air flow rate will meet the required emission limit at the most
favorable cost. The Federal, State, or local emission regulations for
each plant will also influence the choice of emission capture equipment.
Control of emissions from EAF's and AOD vessels requires two separate
steps: (1) the evacuation and containment (capture) of the emissions
and (2) the removal of various pollutants—primarily particulate matter--
from the evacuated gas stream (control). Emissions must be captured
during the melting and refining processes (process emissions) and the
charging and tapping processes (fugitive emissions).
The air pollution capture systems to be discussed in the following
sections are compatible with processes used to make the many different
grades of steel. Fabric filters are the most widely used control devices
to treat the exhaust gases from EAF's. There is one ESP (installed in
1958) in operation at an EAF plantain Cleveland, Ohio. Only one scrubber
has been installed on an EAF, and no ESP's have been installed since
1974. Only fabric filters are known to be in use on AOD vessels.
4-1
-------
New developments and improvements in the steel industry [have resulted
in the use of higher air flows per megagram of steel produced to effectively
evacuate the process and fugitive emissions. These include the use of
UHP EAF's, the use of ADD vessels in specialty steel shops* and shortened
heat times in both carbon and specialty steel shops to increase the
production rate. These changes have resulted in increased use of large
single or segmented canopy hoods and closed roof monitors over the
furnace, local tapping hoods, and scavenger systems to capture emissions
that bypass the canopy hood. These fugitive emissions capture systems
are the most significant improvements over the capture systems that were
in use during the development of the existing standards of performance.
An alternative to the canopy hood/scavenger duct capture systbm or
closed roof shop is the total furnace enclosure (TFE). Several TFE's
have been installed in carbon shops in the past 5 years. Thesje various
capture systems are discussed in the following sections.
4.2 CAPTURE OF EAF PROCESS AND FUGITIVE EMISSIONS
Several capture systems are used by the industry to meet1 the
requirements of State and local regulatory agencies and the existing ""
standards of performance for EAF's. These systems include: •]
1. Direct-shell evacuation control systems;
2. Side draft hoods; ' ;
3. Partial furnace enclosures; !
4. Total furnace enclosures; !
5. Canopy hoods;
6. Tapping hoods; ', . •
7. Scavenger duct systems; ', . • •
8. Shop roof configurations; and
9. Building evacuation. ' ; '• •
Each system is described below, along with design and operational factors
that affect its performance. :
4.2.1 Direct-Shell Evacuation Control System
The DEC system, also known as the fourth-hole evacuation^ system,
requires a hole in the furnace roof in addition to the three holes
required for the electrodes. A water-cooled or refractory-lined duct
4-2
-------
attaches to the furnace roof and, when the furnace roof is in place,
joins a duct that is connected with the emission control device (Figure 4-1).
At the connecting point of the two ducts, there is a small gap that
allows dilution air to enter the duct. The dilution air cools the
exhaust gases and causes the combustion of the carbon monoxide and
unburned hydrocarbons. The gap also allows room for the furnace roof to
be elevated and rotated to the side for furnace charging and for the
furnace to be tilted for tapping molten steel or for slagging. During
the times when the furnace is tilted or the furnace roof is rotated
aside for charging, the DEC system is ineffective, and the fugitive
emissions drift toward the building roof or canopy hood.
When the furnace roof is in place, the DEC system provides good
emission control with a minimum of energy since the air volume withdrawn
is the lowest of the process emissions capture devices. During melting
and refining operations, a slight negative pressure is maintained within
the furnace to withdraw effectively the emissions through the DEC system.
The DEC withdraws between 90 and 100 percent of the melting and
refining (process) emissions from the furnace before they escape the
furnace and are diluted with ventilation air. A typical particulate
matter emission capture efficiency with a properly operated DEC system
is estimated to be 99 percent of the process emissions.1
The DEC system of fume extraction has been widely used in the steel
industry for many years to capture EAF emissions. It can be used on
EAF's that produce any grade of steel, including common carbon grades
and alloy steel grades. In the past, when EAF's performed both the
melting and refining operations, the DEC system could not be used in
specialty steel shops when a second or reducing slag operation was
performed. The reducing slag was used to remove impurities from the
molten steel, and the introduction of outside air into the furnace (due
to the negative pressure created by the DEC system) oxidized the slag
and rendered it ineffective. With the wide acceptance of ADD vessels
and other secondary refining operations (i.e., duplexing, or the use of
a vessel other than the EAF in which to carry out refining), the use of
a reducing slag has been diminished. Duplexing allows the use of the
DEC fume extraction system in most EAF shops.
4-3
-------
A. PLAN
ELECTRODES (3). ' _„
FURNACE ROOF
TAP —
SPOUT
. DUCT TO
AIR GAP /CONTROL DEVICE
"""---.^
^\
ELECTRIC ARC FURNACE
r^—
REFRACTORY LINED
OR WATER-COOLED
DUCT
B. ELEVATION
Figure 4-1. Direct-shell evacuation control (two vieWs).
4-4 . •••• •
-------
the direct evacuation system can be retrofitted to an existing
furnace. However, careful design is needed to avoid problems such as:
excessive weight on the furnace roof of small furnaces, excessive deterio-
ration of shell refractories and roofs, inadequate water cooling, and
inadequate clearance for the DEC when rotating the furnace roof for
charging.1,2 The DEC system, however, is very popular in new installations,
and no problems are known to exist when the DEC system is built as a
part of the new furnace.
4.2.2 Side Draft Hoods
The side draft hood is another fume extraction system that is used
on EAF's to capture melting and refining (process) emissions (Figure 4-2).
The side draft hood is mounted on the EAF roof, with one side open to
avoid restricting the movement of the electrodes. This system requires
a tight fit of the furnace roof so that all the emissions that leave the
furnace escape only around the electrode annuli. The side draft hood,
like the DEC system, operates only when the furnace roof is in place and
when the furnace is in an upright position.
Side draft hoods are not used as widely as DEC systems and, because
of higher operating costs, are typically used only on small furnaces.2
The side draft hood requires a larger exhaust volume than a DEC system.1
The exhaust volume serves to introduce dilution air to cool the exhaust
emissions and ensure combustion of the carbon monoxide and unburned
hydrocarbons.
The side draft hood has an estimated particulate emission capture
efficiency of between 90 and 100 percent of the melting and refining
emissions. The typical particulate capture efficiency is estimated to
be 99 percent.1
Retrofitting an existing EAF with a side draft hood generally
presents few problems. The side draft hood allows easy access to the
electrodes and annuli to perform needed maintenance. It is believed
that the use of this system on new furnaces will be limited to small
furnaces. '*
4.2.3. Partial Furnace Enclosures
The partial furnace enclosures (PFE's) have walls on three sides
of the furnace area that act as a chimney directing the fugitive emissions
4-5
-------
ooo
A. PLAN
ELECTRODES (3)
SIDE DRAFT,
-HOOD
FURNACE
ROOF
TAP
SPOUT
ELECTRIC ARC FURNACE
SMALL GAP TO
FACILITATE:
ROOF MOVEMEN"
B. ELEVATION
Figure 4-2. Side draft hood (two views)
4-6
-------
to the canopy hood. The PFE systems are used in association with a DEC
system for process emissions and sometimes with local hoods for the
capture of slagging and tapping emissions (Figure 4-3). The walls of
the PFE help reduce the impact of cross drafts in deflecting the upward
flow of emissions. Significant secondary benefits are that the furnace
noise and heat radiation are reduced outside the partial furnace enclosure
walls.' The enclosure walls are designed to' allow adequate room for the
furnace roof to swing open for charging, for crane passage above the
enclosure, and for furnace, maintenance.
Partial furnace enclosures are in use at several steel plants that
have EAF's ranging in size from 154 Mg (170 tons) to 204 Mg (225 tons).
One company uses PFE's on two 204-Mg (225-ton) EAF's. This system uses
a DEC system to capture the process emissions and local hoods above the
tapping ladle and slag pot to capture the fugitive emissions from the
tapping and slagging operations, respectively. These hoods are stationary,
and the molten steel or slag is poured through an opening in the hood
(see Figure 4-3).
Two other examples of PFE's are located at new facilities that were
completed during 1981. The PFE's were installed on two new 154-Mg
(170-ton) UHP EAF's and on two new 168-Mg (185-ton) UHP EAF's.3,4 Both
plants incorporated the furnaces and enclosures into an existing shop
when old or damaged equipment was replaced.
Partial furnace enclosures are easier to install and less expensive
than total furnace enclosures (Section 4.2.4). Because the furnace is
only partially enclosed, the crane operator can see the furnace during
charging. Any periodic explosions due to wet or icy scrap will not
cause damage to a partial enclosure since the force will be vented out
the top and front of the partial furnace enclosure. In contrast to the
total furnace enclosure, however, a PFE does not function to capture
emissions, but serves to direct them to another capture device. Crane
passage above the furnace will still disrupt the emission plume. One
plant has partially overcome the problem of the emission plume deflec-
tion by installing additional enclosure walls on the crane so that when
the crane is in position for a charge or a tap, the enclosure walls
extend from the floor to the roof.5 Retrofitting a PFE into an old shop
4-7
-------
»-.' CANOPY HOOD
CHARGING
ENCLOSURE
OPS-HATING FLOOR
SLAG POT
Figure 4-3. Partial furnace enclosure.
4-8
-------
with a hew EAF is easier than retrofitting some of the other emission
capture systems.
The amount by which PFE's increase canopy hood capture efficiency
has not been documented. However, based on their design considerations
(i.e, decreased cross-draft problems), their use should increase overall
fugitive emission capture by reducing the quantity of emissions that
escape capture by the canopy hood. Overall fugitive capture efficiency
should be in excess of the 80 percent (estimated) capture efficiency for
canopy hoods (see Section 4.2.5).
4.2.4 Total Furnace Enclosure
The total furnace enclosure completely surrounds the furnace with a
metal shell that acts to contain all the charging, melting and refining,
slagging, and tapping emissions,as well as to reduce the furnace noise
and heat radiation outside the enclosure (Figure 4-4). The enclosure is
typically designed to capture all the process and fugitive emissions
because the emissions are confined to a small area. Total furnace
enclosures operate with a greatly reduced air flow compared with building
evacuation or canopy hood systems. The volume of air that must be
removed from the total furnace enclosure is estimated to be only 30 to
40 percent of that required for an efficient canopy hood system.1 A
duct at the top of the enclosure removes charging and me!ting/refining
emissions, and a local hood under the enclosure collects emissions from
molten steel and slag tapping.
The crane brings in the charge bucket through the front charge
doors on the.furnace enclosure. A mechanical roof door opens to provide
a slot for the crane cables holding the charge bucket. During charging,
the front charge doors are closed, and an air curtain blows any emissions
directly into the large duct at the top of the enclosure.3 Some emissions
can escape capture through this roof door; therefore, a scavenger duct
or a single canopy hood can be installed over the TFE for the purpose of
capturing these fugitive emissions. After the charge, the front charge
doors are reopened, the charge bucket is-removed, the furnace roof is
moved into place to begin melting, and the front charge doors are closed.
Tapping emissions are collected by diverting (using dampers) part of the
4-9
-------
/\_/\
*k*
A
4
Concrete
Floor
~:
Roof R°°f Venf :
,
Roof Trusses : j*"'
t :
5' ^- Top Charge Door
f - • ;
//"T „ \
/ '-Mam Exhaust Duct ^\ .'-Front Charge Doors
•nfin
^jJ-i-iji, ;
f*** "*% ^s- Alloy Addition Chute
Ji *r
! Furnace -T -.^s' ,
( ' O i '
: ;
/ "'•>--._-,—*' |^-"-"-"-n ^—Topping Exhaust Duct
'—Rear Enclosure Door J ft j
0' i Ladle |
cTrrri
5 log Ptt .". '.!• ts.
y I -^ : . ,
SIDE VIEW
Front Cnarge Doors —3 _..,_..
_ _. _ y7 r?- To Control Device
Top Charge Door -7 // sf~if
/ / / iA^ '
Air Curtain ran--''
Alloy Addition
Chute IT""?
C
t
/
Jl
y
«
I '?
Furnace
A
-^ -"
e Ladle
4====
'
i
*
i
. , i •
•""z:
•
— 1 Main Exhaust Duet
\, "^Damper
•n t
I
i
I !
1
i
I ;
t
i Concrete :Floor
— . — i __, i
^•Tapping Exhaust Duct
i
Shop Floor
FRONT VIEW
Figure 4-4. Total furnace enclosure at Lone Star Steel Cpmpany.
4-10
-------
flow from the main exhaust duct to a duct adjacent to the tapping ladle.
The tapping ladle is positioned under the furnace by a transfer car.
The first TFE on a domestic furnace was installed in 1976.1 Total
furnace enclosures were installed on each of two 54.4-Mg (60-ton) furnaces.
One TFE unit was installed in 1980 on a new steel plant EAF, and three
additional TFE's were retrofitted in 1981 onto existing steel plant
EAF's.3,6 An existing DEC system was maintained on the retrofitted
furnaces to capture the primary emissions, and the TFE was used to solve
a fugitive emission problem. A retrofit TFE was installed on a medium
size (165-Mg, 182-ton) furnace with a 22-ft shell in Italy in
September 1980.7 This enclosure is on a furnace considerably larger
than any of the furnaces with TFE's in the United States and should
provide valuable operational data for the use of TFE's on medium-to-large
EAF's.
Proper design of a TFE must include analysis of the size and
transformer capacity of the EAF (which have a significant impact on the
TFE size and the air flow necessary to evacuate the enclosure), the
quality of the scrap used, the oxygen blow rate, the amount of space
that is needed inside the enclosure, the number of doors in the sides of
the enclosure, and the amount of space above the furnace needed for
crane clearance.5,8 In addition, the need for explosion venting panels
to vent the pressure from explosions due to wet or oily scrap must be
evaluated.8
Industry has shown some reluctance to the use of TFE's. Some crane
operators believe that the TFE will hamper operations such as charging
and alloy addition since the crane operator will have difficulty getting
the bucket into position and will not be able to see the bucket when the
charge doors are closed.3 Another problem that industry considers to be
important is that the use of a TFE may cause too many process delays,
especially on larger furnaces.5 Problems with the charge doors leaking
have arisen because the seals are degraded by the high heat, and the
doors will not open or close when they are bumped and damaged by the
charge bucket.5 Vendor representatives indicate that these problems may
be solved after operating exprience has been gained and recommended
4-11
-------
maintenance is performed.5,8 Operator reluctance to TFE's is being
reduced by the experience gained at domestic and foreign steel plants.6,7
Emission capture efficiencies of TFE systems are estimated to range
from 90 to TOO percent.1 The capture efficiency may be enhanced through
the use of a canopy hood or a scavenger duct system. :
4.2.5 Canopy Hoods | '
The canopy hood system involves one or more canopy hoods suspended
from, or built into, the melt shop roof directly above each furnace
(Figure 4-5). The hood has to be high enough to provide clearance for ,
crane movement and space for the upward movement of the electrodes when
moving the furnace roof. Canopy hoods are widely used either alone, to
capture both process and fugitive emissions, or in combination with
other capture technologies to capture fugitive emissions only. This
system is one of the oldest and most well-known capture technologies in
use. It is likely that canopy hoods will continue to be use;d, typically
in combinations with other capture technologies discussed in other
sections of this chapter. ;
The thermal currents from the hot furnace help the fume's from the
charging, melting and refining, and tapping operations rise !to the
canopy. The fumes are sometimes deflected away from the canopy by
impingement on the crane and charge bucket when the furnace !is being
charged and also by impingement on the crane during tapping ;when the
crane holds the ladle. The emissions can also be disrupted >by cross-drafts
within the building due to open doors, open ends and side walls of the
building, passage of shop vehicles,, temperature gradients within the
shop, and other hoods that ventilate nearby processes. For small furnaces,
the canopy is not generally as effective because there is less thermal
uplift generated by the furnace.1 High-pressure weather systems and low
humidity tend to aid efficient upward flow of the exhaust plume. However,
during periods of low-pressure weather systems, high humidity, and/or
strong winds, the thermal columns above the furnace may be insufficient
to carry all fumes directly to the hoods.1
Partition walls and air curtains have been successfully employed to
reduce the effects of cross-drafts and improve the flow of emissions to
the canopy hoods. The partition walls and air curtains can'be used to
4-12
-------
OS
LU
OS
CO
if
o.
o
OJ
•t-3
CO
2
. Q.
«3
O
-a
o
o
Si
o
c
(C
Lf)
I
qj
CD
4-13
-------
screen the steelmaking furnace and help contain and direct the fumes to
the canopy. Unfortunately, the air curtain often cannot completely
overcome the force of cross-drafts. The emissions that are not completely
captured bypass the hood and accumulate in the upper part of the melt
shop roof or escape to the outside air. A scavenger duct system can be
installed in the exhaust ductwork immediately above the canopy to capture
the emissions that elude the canopy and accumulate under the shop roof
(see Section 4.2.7). . :
The canopy hoods can be divided into sections that can be closed
off by dampers. At many steel plants, the canopy hood is divided into
two sections, one section for charging and melting emissions iand the
other section for tapping emissions. 'At some plants the canopy is
divided into three sections that are arranged to capture slagging emissions,
charging and melting emissions, and tapping emissions. Sections of the
canopy hoods can be closed off with dampers, if a particular operation
i
is not in progress, to maximize the draft directly above the point of
greatest emissions. Tapping hoods can be located at the same level as
the charging canopy hood or directly above the tap ladle (see Section 4.2.6).
Retrofitting an existing furnace with a canopy hood sometimes
requires extensive structural modifications. The. trusses and roof beams
must often be relocated, reconstructed, and strengthened to accommodate
the canopy and exhaust ductwork. In some shops, there may not be enough
i
clearance between the crane and the roof, or the roof configuration
itself may not be adaptable to a canopy installation. Also, ispace
outside the shop must be available for a baghouse capable of handling
the required exhaust volume. ,
The capture efficiency of a single canopy hood is typically 75 to
85 percent (average 80 percent), depending on the amount of cross-drafts
present in the shop. The capture efficiency rises to 85 to 95 percent
(average 90 percent) with a segmented (sectioned) or a large canopy
hood. These percent capture efficiencies are estimates based on observa-
tions made at several facilities, engineering judgment, and review of
available technical information.11-14 Industry representatives believe that
there may be trade-offs since the segmented canopy hood has a higher face
4-14
-------
velocity to capture the fumes while the single canopy hood has greater
storage volume.
4.2.6 Tapping Hoods
Tapping hoods have received increased use in recent steel plant
emission capture designs. They are used to supplement the emission
capture of TFE's, PFE's, canopy hoods, and even DEC and side draft hood
systems. (The air flow utilized by a DEC system can be diverted from
the DEC during the tap when the DEC is ineffectual.)
The tapping hood can have several different adaptations, but it
usually involves a movable or stationary hood that is located immediately
above the tapping ladle when the tapping operation is in progress.
Movable tapping hoods are swung aside when tapping is not in progress.
A hood located right above the ladle is considerably more efficient than
a canopy hood at the roof level.14 In the past, the industry has used
the crane to hold the ladle during the tap. The crane deflects the
fumes that are generated during the tapping operation, making it very
difficult for the canopy hood to effectively capture the emissions.
Currently, many shops set the ladle on a pedestal or ladle car prior to
tapping, and this eliminates deflection by the crane.
One company has developed a tapping pit enclosure that is simple in
design and is efficient. The overhead crane places the ladle in a
tapping pit, then a powered, removable cover seals the tapping pit. The
molten steel from the EAF is tapped out a short spout to a chute that
extends through the side wall of the ladle pit.1 The exhaust gases are
drawn out through a duct at the top of the tapping pit enclosure. The
tapping pit enclosure captures 90 to 100 percent of the fumes.1
Another design utilizes a retractable hood that can either be
manually or automatically operated when the furnace starts to tilt for a
tap. The ladle is put in place with the overhead-crane, and then the
crane is withdrawn before the retractable hood is projected outward so
that no damage is done to the hood with the crane hooks. When the
furnace starts to tilt, the flow from the DEC is dampered off and redirected
to the tapping hood. The capture efficiency of this local tapping hood
has not been documented but is expected to greatly exceed that of a
canopy hood.
4-15
-------
There are other possible designs for tapping hoods, but the primary
emphasis is on placing a hood much closer to the source of emissions,
thereby eliminating the problems of fume deflection by the crane and
emission plume disruption by cross drafts. Because of space limitations,
retrofitting a tapping hood can be more difficult than installing such a
hood in a new shop.
4.2.7 Scavenger Duct System ;
A scavenger duct system consists of small auxiliary duetts that are
located above the main canopy hood near the shop roof. A relatively low
flow rate (approximately 10 percent of the total flow rate for a canopy
hood) is maintained through these ducts to capture fugitive emissions
that are not captured by the canopy hood system. The ducts have openings
through which the fugitive emissions are withdrawn to be cleaned by the
emission control device.
The factors that should be considered when designing a scavenger
duct system include the estimated capture efficiency of the canopy
hoods, the regulatory requirements for fugitive emissions containment,
the engineering necessary to install the scavenger system, the shop
configuration, and worker health and safety.
The capture efficiency of scavenger ducts is enhanced through the
use of cross-draft partitions and can only be used with a closed roof
configuration, semi-closed roof (only closed above the furnace area,
open elsewhere), or louvered roof. . :
4.2.8 Shop Roof Configurations
There are three basic shop roof configurations: completely open,
open except over the furnace, and closed over the entire melt shop. A
variation of the closed roof shop involves a louvered roof monitor that
is mechanically controlled to allow for closing the louvers during
periods of fugitive emissions and opening the louvers when the melt shop
is clear.
Completely open roof monitors allow for natural ventilation within
the shop but do not provide adequate residence time for effective emissions
capture by a canopy hood and preclude the use of scavenger systems.
Those emissions not captured on contact with the capture system (i.e.,
canopy hood) are emitted to the atmosphere.
4-16
-------
A shop roof system that is closed above the furnace but open over
'other areas provides the advantages of both the open and closed systems.
The open monitor sections provide for natural ventilation, while the
closed roof section, particularly when combined with cross-draft partitions
or a scavenger duct system, aids in effective emissions capture
(Section 4.4).
A shop roof system that is closed over the entire shop promotes
effective capture of emissions because the emissions are contained
within the shop and can be captured by the canopy hood or a scavenger
duct system. However, closed roof shops do not allow for natural ventilation
and industry claims this may contribute to a worker heat stress problem.10
The National Institute for Occupational Safety and Health (NIOSH)
has concluded that radiant heat is. the primary contributing factor to
heat stress. High air ventilation rates may lower the temperature in an
electric arc furnace shop to some extent; however, they do not reduce
the effects of radiant heat. Other means available to protect workers
from heat stress are:15,16
1. Decrease the number of workers and the time they are exposured
to the heat;
2. Provide an air conditioned rest area to decrease the time-weighted
average temperature;
3. Use portable fans- to blow air into the work area;
4. Blow air ducted from outside the shop (and possibly cooled)
into the work area; and
5. Utilize radiation shields and protective clothing.
There are several recently constructed closed roof shops operating
in warm areas of the country, which may indicate that adequate protective
measures are available to compensate for the loss of natural ventilation.
However, sufficient operating experience has not yet been gained to
determine if heat stress at these plants will be a problem.
An optional shop roof configuration for use where heat stress may
' ;.
-------
shops as it Is for single furnace shops because of overlapping periods
of high emissions. Sectioning of the shop with cross-draft partitions
could alleviate some of the problems encountered with a multifurnace
shop. Two EAF shops are known to use the louvered monitor system.
4.2.9' Building; Evacuation |
A building evacuation system involves a closed shop roof with
ductwork at the peak of the roof to collect all emissions from the shop
operations. This system of .capture is very similar to a canopy hood
capture system with a closed roof. Sheet metal partitions can be permanently
attached to the shop roof to prevent the emissions from drifting to
other parts of the melt shop.
Building evacuation requires a greater air flow than a canopy hood
system but will capture-all emissions generated in a furnace 'shop. Some
!
estimates are that building evacuation requires 25 percent more air flow
than a well-operated canopy hood. The capture efficiency for the building
evacuation system is very good, ranging from 95 to 100 percent removal
of the particulate matter; the typical maximum pa'rticulate removal
efficiency is 99 percent.1 The emissions capture rate, howevW, is
sometimes slower than that of a canopy hood. '
There are several plants utilizing this type of emission capture
system. One of these plants has two 22.7-Mg (25-ton) EAF's, a 45.5-Mg
(50-ton) EAF, a 68.2-Mg (75-ton) EAF, and a 22.7-Mg (25-ton) ;AOD vessel.
Another plant has one 45.5-Mg (50-ton), one 68.2-Mg (75-ton), and three
90.7-Mg (100-ton) EAF's. |
The factors that influence the selection of a building evacuation
system over other systems for controlling emissions are:1,9 ;
1. Insufficient space and structural limitations that preclude the
use of a canopy hood; ;
. 2. Desire to control fumes from all EAF operations including
charging, melting, refining, slagging, tapping, and other operations in
the shop; ;
3, Roof design well suited to serve as the hood;
4. Desire to exhaust the entire shop's internal atmosphere to reduce
pollutant concentrations for worker hygiene; and
4-18
-------
5. Only minor interference with existing shop operation would
occur when a building evacuation system is installed.
4.3 CAPTURE OF AOD VESSEL PROCESS AND FUGITIVE EMISSIONS
The emissions from the AOD vessel primarily occur when the vessel
is in an upright position for blowing or .Stirring with a combination of
argon, oxygen, and nitrogen gases. The emissions during these times are
very heavy, but the upward thermal lift quickly brings the emissions to
the capture device. The emissions during molten steel charging are
minimal since the molten steel is hotter than the refractory and the
gas-injecting tuyeres are not in operation. The gas injection is normally
stopped before the vessel turns down for alloy and flux additions or for
sampling and temperature checks. However, sometimes the gas injection
operation is not stopped until the vessel has started to turn down,
resulting in emissions that are largely deflected by the overhead crane
if it is in position for adding fluxes or alloys. Slagging and tapping
emissions are small since the gas injection is stopped to prevent damage
to the tuyeres while the vessel is in a horizontal position.
4.3.1 Diverter Stack With Canopy Hood
The diverter stack with canopy hood capture configuration is a
common arrangement currently used for collection of emissions from AOD
.vessels (Figure 4-6). The diverter^stack is typically located 1,5 to
3 m (5 to 10 ft) above the mouth of the vessel. The diverter stack can
either be fixed in position or movable so that it can swing out of the
way during charging and tapping. The diverter stacks that are in a
.fixed position are typically at a greater distance from the AOD vessel
than movable stacks.
The diverter stack is tapered, with the narrow end at the top, and
acts to accelerate the AOD vessel off-gases toward the canopy hood. The
stack greatly reduces the dispersing effect of melt shop cross-drafts on
the upward flow of the emission plume.
The canopy hood is typically built into the shop roof so that
adequate clearance is provided for the overhead cranes. The emission
capture by canopy hoods is usually very good. However, the emissions
can be deflected by the overhead crane or by shop cross-drafts if the
4-19
-------
UJ
se.
CO
u_
on
LU
I— S«£
as o
i— i on
o
XJ
o
•o
(C
CJ
O
(O
O)
3
U>
4-20
-------
distance from'.the vessel mouth to the diverter stack or from the diverter
stack to the canopy hood is great. Emissions that miss the canopy hood
can.be captured by a closed roof and scavenger duct system, if utilized;
otherwise, these emissions will drift out of the shop through openings
in the roof and walls.
4.3.2 Close-Fitting Hood With Canopy Hood
Another emission capture system for AQD vessels involves a close-
fitting hood for process emissions and a canopy hood for fugitive emissions
(Figure 4-7). The close-fitting hood, which may be moved away for
charging and tapping, is typically situated 0.3 to 0.6 m (1 to 2 ft)
above the top of the vessel when the vessel is in an upright position.
The refining emissions are captured by the close-fitting hood, and any
fugitive emissions are captured by the canopy hood. This system overcomes
the problem of deflection of fumes during the steel refining operation
by the overhead crane or by cross-drafts in the shop. Most new AOD
installations are expected to use the close-fitting hood system because
it achieves efficient fume capture with generally lower air flow volumes
and better emissions capture than a diverter stack-canopy hood system.17
The close-fitting hood requires more maintenance than the diverter stack
system because of the very high temperature of the exhaust gases.
4.4 FUGITIVE EMISSIONS CAPTURE SYSTEM COMBINATIONS
Individual fugitive emissions capture technologies'have been discussed
in the previous sections. These technologies are seldom used alone and
are typically used in combination with one or more technologies. This
section presents those combinations of capture technologies that would
be suitable for use industry wide. The fugitive emissions capture
efficiency achievable by each combination is also presented. All of the
fugitive emission capture system combinations have not been observed in
operation. However, it is the engineering judgment of EPA that there is
no technological reason that these capture combinations could not be
used. - « ,
Table 4-1 presents six combinations of fugitive emissions capture
technologies suitable for use on EAF's. The ranges of capture efficiencies
(emission reductions) are based on observations of the operation of
4-21
-------
QS
LU
OS
03
2
-a
o
o
.
to
o
•M
»^-
•o
o
o
id
en
i
07
4-22
-------
TABLE 4-1. FUGITIVE EMISSIONS CAPTURE TECHNOLOGY COMBINATIONS
(CARBON AND SPECIALTY STEEL EAF)
Fugutive
emission
capture
system a
combination Capture equipment
Estimated
fugitive
emission .
reduction,
percent
4
Single canopy hood, open roof monitor.
Segmented canopy hood, closed roof (over
furnace)/open roof monitor elsewhere.
Single canopy hood, local tapping hood, local
slagging hood, closed roof (over furnace)/
open roof monitor elsewhere.
Segmented canopy hood, scavenger duct,
cross-draft partitions, closed roof (over
furnace)/open roof monitor elsewhere.
Single canopy hood, total furnace enclosure,
closed roof (over furnace)/open roof
monitor elsewhere.
Segmented canopy hood, scavenger duct,
cross-draft partitions, closed roof.
75-85
85-95
85-95
90-95
90-95
95-100
j*DEC system used for process emissions capture on all alternatives.
Estimate based on observation of technologies and engineering judgment.
4-23
-------
individual technologies at various plants and on engineering studies and
judgment as to the effectiveness of these technologies when used in
combination.14,18 Limited data are available to substantiate these
capture efficiency estimates due to the difficulties involved in performing
tests'on the systems. Some plume photographic scaling techniques have
been performed to help determine the hood efficiencies and help in the
design of fume collection equipment.18,19 Visible emission tests have
been performed that demonstrate 100 percent fugitive emission reduction
for some of the systems since there were no visible emissions observed
coming from melt shop roof.20,21 Table 4-2 presents four combinations
of fugitive emissions capture technologies suitable for use on AOD
vessels.
4.5 EXHAUST GAS CLEANING DEVICES
Following the capture and evacuation of the melt shop emissions, a
control device is used to clean the dust-laden gas stream before the
gases are vented to the atmosphere. The control device used :in most
U.S. steel mills with EAF's arid AOD vessels is the fabric filter. Only
i
one electrostatic precipitator (ESP) installation was found in use at an
!
older EAF shop (the ESP's were installed in 1958). Less than 2 percent
of the existing EAF units and no AOD units are known to use Wet scrubbers.
The predominate use of fabric filters is expected to continue; ESP's and
scrubbers are expected to have only limited application. No'process
parameters or conditions are known to exist that preclude the use of a
fabric filter on either an EAF or an AOD or that require the use of an
ESP or scrubber. •
i
Fabric filters have many advantages that make them suitable for
control of EAF and AOD vessel emissions. Fabric filters useiless energy
than either scrubbers or ESP's for equivalent outlet particulate concen-
trations, are efficient collectors of very fine emissions,- are tolerant
of fluctuations in the inlet particle size distribution (which affects
ESP's), and collect emissions in a dry form. The dust from the baghouse,
which is easier to handle and recycle than the wastewater and sludge
from scrubbers, can be wetted in a pug mill or pelletized before it is
recyled or landfilled. ;
4-24
-------
TABLE 4-2. FUGITIVE EMISSIONS CAPTURE TECHNOLOGY COMBINATIONS
(SPECIALTY STEEL AOD)
Fugitive
emission
capture
system
combination
Capture equipment*
Estimated
fugitive
emission ,
reduction,
percent
Single canopy hood, open roof monitor.
Single canopy hood, closed roof (over
vessel)/open roof monitor.
Single canopy hood, scavenger duct,
cross-draft partitions, closed roof (over
vessel)/open roof monitor.
Single canopy hood, scavenger duct,
cross-draft partitions, closed roof.
75-85
85-95
90-95
95-100
Close-fitting hoods are used for process emissions capture on all
combinations.
Estimate based on observation of technologies and engineering judgment.
4-25
-------
A fabric filter system (baghouse) consists of a number of filtering
elements (bags) along with a bag cleaning system contained in a main
shell structure with dust hoppers. Particulate-laden gases are passed
through the bags so that the particles are retained oh the fabric, thus
cleaning the gas. Typically, a baghouse is divided into several compart-
ments or sections. In larger installations, an extra section is often
provided to allow one compartment to be out of service for cleaning at
any given time without affecting the overall efficiency of the fabric
filter. • ;
The basic mechanisms available for cleaning particulate-laden gases
are inertia! impactioni diffusion, direct interception, and sieving.
!
The first three mechanisms prevail only briefly during the first few
i
minutes of cleaning with new or recently cleaned bags, while'the sieving
action of the dust layer accumulating on the fabric surface soon predomi-
nates. The sieving mechanism leads to high efficiency collection of
particulates unless defects such as pinhole leaks in the bags or cracks
in the filter cake appear.
In fabric filtration, both the collection efficiency and the pressure
drop across the bag surface increase as the dust layer on the bag builds
up. . Since the system cannot continue to operate efficiently with an
increasing pressure drop, the bags are cleaned periodically by reverse
air flow, pulse-jet, or a shaker mechanism. Reverse air flow is typically
used on fabric filter units controlling EAF and AOD vessel emissions.
The air is forced through the bags being cleaned, causing them to collapse
and the dust cake to fall into the hoppers below. Pulse-jet^cleaning is
used at a few EAF and AOD vessel installations. A sharp pulse of compressed
air released into the bag causes a shock wave and reverses the air flow
in the bag. The pulse of air deforms the bag and dislodges the dust
cake into the hoppers below. The shaker mechanism physically shakes the
bags to be cleaned, causing the dust cake to fall into the hoppers
below. i
The design of a fabric filter requires that certain information be
known about the gas stream to be cleaned. The information usually
needed is mass emission rate, volumetric flow rate, and particle charac-
teristics. However, once initial experience has been gained iwith the
4-26
-------
performance of fabric filters in a particular industry (as it has with
EAF's and AOD vessels in the steel industry), control device vendors can
predict the range within which the key design, parameters (e.g., the
air-to-^cloth [A/C] ratio, .pressure drop, cleaning mechanism and frequency,
and bag construction [e.g., material, weave]) must be maintained to
achieve desired outlet emission concentrations. Baghouse design in the
steel industry no longer requires specification of site-specific gas
characteristics except the gas volume to be cleaned. Historical performance
and the experience gained in the industray dictate that, in most cases,
the A/C ratio will be around 3:1, the pressure drop across the bags will
be between 7.6 and 12.7 cm (3 and 5 in.) water column (w.c.), the cleaning
mechanism will be reverse air, and the bags will be constructed of a
Dacron® polyester blend. After installation of the unit, the cleaning
frequency is adjusted to optimize operation of the fabric filter.
Pressure drop sensors or timers are used to initiate sequential compartment
cleaning automatically on a preset time schedule.
The two types of fabric filters used in the industry are the positive-
pressure type and the .negative-pressure type. Pressurized fabric filter
systems are those in which the effluent gases are forced through the
fabric filter by a fan placed between the emission collection system and
the fabric filter. The compartments in the positive-pressure fabric
filter do not need to be airtight since only the dirty air side of the
collector needs to be sealed.1 Bag inspections and maintenance are
easier to perform than on negative-pressure fabric filters, and the
compartments can be entered while the positive-pressure fabric filter is
in operation if the temperature is low enough for worker safety. Dirty
air entering the fabric filter is filtered through the cloth and then
vented to the atmosphere through louvers, stub stacks, or a ridge vent
(monitor) on the top of the positive-pressure fabric filter.
The alternative to the pressurized system is a negative-pressure or
suction-type fabric filter. The fan is placed on the clean air side of
the fabric filter, and the effluent gas is drawn through the fabric.
With this type of system, the bags must be kept airtight, and thus each
compartment must be taken off-line for bag maintenance and replacement.
Negative-pressure filters usually require less fan maintenance and
4-27
-------
less operating horsepower than the pressure type.1 However, 'negative-
pressure fabric filters may require more ductwork due to the!need for a
stack, may be more expensive because they must be built to withstand the
suction created by the fans, and must be well sealed to prevent the
introduction of dilution air. :
Suction-type fabric filters are typically vented to the;atmosphere
through a common stack, which can easily be sampled for particulate
emissions. Pressurized fabric filters usually do not have a;stack;
!
thus, testing these fabric filter systems for compliance with State and
Federal regulations is difficult. Since use of the pressurized fabric
filter system predominates on EAF and ADD units in the steel|industry, a
potential enforcement problem exists. A common method for testing these
systems for particulate emissions has been the use of ambient high
volume (hi-vol) samplers placed in each of the compartments. These
hi-vol samplers are placed on the clean-air side of the bags'so as to
obtain representative samples of the exhaust gas. Because of the question-
able accuracy of the results obtained, EPA does not recommend the use of
hi-vol sampling procedures on future pressurized fabric filter emission
tests (this recommendation does not affect previous agreements between
governmental agencies and the steel companies). Two of the causes of
the inaccuracies are the high bias of the gas volume sampled, which
results in a low bias to the results, and the practice of turning the
samplers upside down, which can result in particulate matter;being lost
i
when the sampler is turned off and handled. EPA prefers a Reference
Method 5 sampling train with the probe inserted into the compartment of
i
a pressurized fabric filter between the bags and the monovent, or in the
monovent (ridgevent). Emission testing is discussed further in Appendix D.
Problems have also been recognized when performing continuous
opacity monitoring on pressurized fabric filters. Some of these problems
include an excessive path length in some applications, stratification of
the gas flow, and the inherent difficulties involved when converting the
opacities obtained to an equivalent stack diameter or to EPA1Reference
Method 9. These problems are also discussed in Appendix D. '
4-28
-------
4.6 EMISSION SOURCE TEST DATA
This section presents data obtained by EPA on the capture and control
of particulate matter emissions from EAF and ADD- fabric filter systems.
The data are from field tests performed by EPA contractors or from perform-
ance tests submitted by industry or State agencies. Data are also presented
on visible emissions obtained for fabric filter systems, shop roof monitors,
and dust-handling equipment.
These data indicate that properly designed and operated fabric filters
can achieve emission concentrations of 6 mg/dscm (0.0026 gr/dscf)* or less
when operated on EAF, AOD, or combined EAF/AOD units. These same fabric
filter units can achieve exhaust gas opacities of less than 3 percent when
operated on EAF> AOD, or combined EAF/AOD units. Emissions from the shop
roof monitors can be maintained below 6 percent opacity. The dust-handling
equipment (i.e., that equipment from the baghouse dust hopper to and
including the transfer equipment to the truck that hauls the dust) can be
designed tp exhibit less than 10 percent opacity.
4.6.1 Carbon Steel Shops
Source test data were obtained from 15 carbon steel shop fabric filter
systems. Table 4-3 presents a summary of data from the plants tested. The
tested plants represent the range of facilities expected to be regulated by
any new NSPS with regard to EAF size, ducting configuration, capture
equipment, scrap type, and level of furnace power.
Figure 4-8 presents the results of measurements of particulate
concentrations at eight plants: A, B, C, D, E, F, G, and H. Each data
point represents a separate sample collected by Reference Method 5. The
maximum value for an average of three test runs was 6 mg/dscm (0.0026 gr/
dscf) at Plants B and G. Data for Plants B, C, D, E, F, and G were obtained
during NSPS compliance tests. Data from Plants A and H were obtained during
an EPA source test and from a State compliance test, respectively. Data from
the plants noted on Figure 4-8 have been determined to be suitable for
setting and supporting an emission, limit for EAF fabric filters. A detailed
summary of these test results is presented in Appendix C.
*mg/dscm = milligrams per dry standard cubic meter.
gr/dscf = grains per dry standard cubic foot.
4-29
-------
"
C£
^•^
^
5>
«
CO
"^t"
UJ
_J
CO
H-
u
a.
cu ro '
a. s-
>> u
H-1 in
s
r*>
^^
c
0)
E
0.
•r-
3
CT
0)
CU
5-
3
•P
Q.
ro
O
s-
in a)
CU -P
o o •—
ro 4-5 TJ'i-
c cu *t—
S- T3 4J
3
f-
cu
•p
c
ro
r—
a.
c
3
o >>
C 4-3
C •!-
ID a
en
u
1C
a; o
o
*
— ZC
LU O
CQ to
CM CM
^^/*™^
Lf) LO
M *
O O
LO «=J-
s_ss^
o r*-
CO CM
tD CM
*« *
M
"Tf
H-
z
o
M
LU
U_
1—
r— •
f—\
m
LO
v»^
t
en
en
^j-
r>-
CM
M
to
O
c
3
c o
re c
CO 36
i— c
O =3
rv*
o
C£ CK
O O
CJ O
LU LU
0 0
*» M
3= rc
O CJ
^w 1—
/""^ *^™N
o o
CM CO
^^v_^
3
a.
:n
=3
oo en
00 P**-
O r—
r™ r—
CO
CM
M
r— en
i— CM
Q LU
c c
3 3
o o
c c
c c
f^ Q*"^
O CJ
nz o
O LU
CO Q
»» W
z z
O CJ
CO r—
,^^
*^""> CO
O CM
^^-^^
*?J" en
LO CO
>* 0
r™
O r—
CO CO
U_ CD
C
O
4-^ C
&• i^TifC
•r- C
a ZD
rv* Q*^
o o
o o
LU LU
0 0
* A
^^ "T^
O O
CM CSJ
S^*t ^**l
o o
CD CM
•*^*^*>
r**- en
O CO
en o
r—
CO
CO
*v
CM CO
ro ro
a>
^|*^ H~l
C
^» o
4*^ C
g_ \i/
•p- C
Q =D
P»^
O
•^v
C^j
•>o:
z o
to
—'CD
z to
1—
o a
UJ to
a
** J^~
z o
o to
CM r-
/^N<^>
O O
LO CO
co i—
•a
ex
z
=3
LO en
r-~ r*»
P-— i~*
CO i—
LO
oo
A
•* o
00 CM
i~Hj s^
5>fc >»
+?-p^
S- S-
•r— *^
o o
Qi
O
n** CC
O O
- «
LU LU
O O
M M
Z Z
o o
to to
CO CM
>»-N/— S
O LO
O CM
i— CM
v«"«-^
f**» r~"
o ^t-
en o
CM
r-. oo
CO CO
* #i
to CM
CO r-
— 1 S
^>. ^j
•P -P
i. i.
•p- «r-
Q CD
££
O
<7jg|
O
*s
CD
to
«l
a. cc
O —1
u
O LU
LU CD
Q
M
** T*
z o
o to
r— CM
/•"^
O
LO
ft S***,
LO O
i— O
^^, <^s
a.
z
Z3
uo
CO
#«
co r^*«
«:»• O
o en
p—
O CM
•* •<*•
A M
en r-
co •=*•
z: o
• W D_
o- ro
r- 0 , i-
O |J= U
J- ' in O) in
-P ; C
C C -r- >,
O | •!- U r-
0 S-.r-
i"O O 0
c • i oo <*-
o in • in c >i
•r- S- C ' 3 •!- i-
•p o o ,£_>
O C •!— r— ""O
ro o -P j re . •!-
> E >- ; 3 O) O
co 4-^ ro .1/5 • >
<4- u a. . 3 a) re
r- O 3 | •—'
r- O T3 -P in O
CU S- M- ; •!- r- 4^
sz. s- re cu
in T3 CO S- "O CO TJ
co cn"O c -P e
•P s- c i : ro in co
CJ CU CU W. i +J
co > > in -P , »cuc:oa>t-inro
. a. o re x: tnl a. co s-
o re u r- -r- o
c c 4J cu 3 4-J m
re s- -a o 'r-
O 3 OJ ro r— p- ti-
lt- 4,i i. if_; .r- O
cu c: -a o in u
i — i — co o re re co
C 4-> O5T3 • ' >,
•r- O CU ••- C OL 4-5
v) 4-5 in in O) re
a. s- p- >>
II || || || 0 U 0) C
^x, m a) re
, Z LU Z Z /•"-. 4-5
OU-OCDCJCinflJ
t— to to u ro in
re cu <*- 3
C r— O
t4 O 'i—
3 in i—
C M-J CO •!-
• O • "O 3
•a -r- s- cu re
CO 4-> CO > S- , — ,
4-5 re > T- O) m
O 3 04-5 CO
re t3 o r— +j re
in ?> *fr-" o o M—* cO *p™ ^"*
incooo-coS-r-in
cu . o jz o re
r— OOt. O)t-in3r—
c c co c T- o" re
3 T- TJ C -r- TJ S-
TQ gj «r— Q) (y f^ $„ ^j
S-p— in Q.O)in re CU4-5
co-i-ocxreoJ-x:u
3 3 P— (V3 r— r-: U O) 3
a. .c 4->
1! II II II It: 4- in
p— o en
« LU ai z z as:- c -
EenoH-toci ojT- m
s- •-< a.-i4 s-
o ce. >, re re
Z .01-^ E-J3
re ^s u T
•
•a
cu
4^
crt
O
4>J
(/)
ITJ
3s
OJ
4->
r~*
i. -r-
CO M—
3
O U
CL-r-
S_
r- r^
o^ re
•r- M-
,c^
o
re LU
£-. PT^
r— CO
3 JC
4-5
II
a_ r—
z c
3D 0
3 CO
4-30
-------
dscf
0.005 -
0.004 -
C.003 -
0.002
0.001
q/dscra
2
•fl
0 i
1
1
I
1
1
1
1
8 • 1
. 1
1
t
1
1
8 '
' i L
- 6 >L-H 'If
II . !
11
' P, y.
n
4 | ' a P
1 ' H P ' !
i! ii : tt
ft ii s. \
i! h ? t'
CO iff 1 I i -
1 ' 1 ^I-S-^
-2 !! § -
y i
KEY
0 TEST RUN
1 1 AVERAGE
J O
I
PLANT A B
CAPTURE EQUIPMENT BE CH.
REFERENCE 22 24
8 C 0 E F G H
SD TFE, DEC, DEC, SDH, DEC, DEC,
CH CH CH CH CH CH
25 27 28 29 30 31 32
Figure 4-8. Summary of Particulate Matter Source Data for
Carbon Steel EAF Fabric Filters (Reference Method 5).
4-31
-------
Other data are available from tests at six carbon steel plants
where the hi-vol sampling technique was used. These data will not be
presented here because the results obtained are biased, and the EPA does
not recommend the use of this technique. Further discussion ion the
i
weaknesses of this technique is presented in Appendix D.
Opacity data were obtained from the fabric filter exhaust gas
stream at Plants B, C, F, G, H, I, J, and K with Reference Method 9 and
continuous opacity monitors. Plants B, C, and F have negative-pressure
fabric filters with stacks, while Plants G, H, I, J, K, L, M,: and 0 have
positive-pressure fabric filters with monovent exhausts. Plant N has a
positive-pressure fabric filter with a monovent exhaust and a negative-
pressure fabric filter with a stack. The opacity data for Plants B, C,
F, G, H, I, J, and K are presented in Table 4-4. The maximum opacity
observed by EPA Reference Method 9 was 0 percent. These opacity data
were obtained concurrently with the particulate data presented in
Figure 4-8. The fabric filters at Plants B, C, F, G, H, I, .J, K, and
i
the negative-pressure fabric filter at Plant N represent best demonstrated
technology, (BDT); however, the test data at Plant N are not ;yet available.
Plants L, M, 0, and N were visited primarily to discuss the worker heat
stress problem'at steel facilities located in the Southern United States.
The fabric filters at Plants L, M, and 0 and the positive-pressure
fabric filter at Plant N are not considered BD'T because of plant age,
level of emission limit required, degree of maintenance, and;design
considerations.43 Although some visible emission data were obtained,
they are not being reported here nor used in the data base.
Table 4-5 presents visible emission data obtained from the shop
roof monitors at Plants C, G, H, I, J, K, and N. Emissions from the
shop roof monitors taken during periods of normal EAF operation were
below 6 percent opacity. The roof monitors at Plants C, G, H, I, and K
are closed over the entire melt shop while the roof monitors:at Plant J
and N are closed over the furnace and open elsewhere.
Visible emission data have also been obtained from the Shop roof
monitors at Plants B, L, M, and 0. The,overall emissions capture technology
at these plants do not represent BDT•, and the data obtained are not used
4-32 .
-------
TABLE 4-4. SUMMARY OF VISIBLE EMISSION DATA FROM FABRIC
FILTERS ON EAF'S AT CARBON STEEL SHOPS15,25,27,29-54,44
Plant
Ba,b
cb,d
cb,d
Fb
Ga,e
He
Ie
Je'f
Ke
Length of
observations,
minutes
960
36
560
18
936
120
60
' 577
300
Maximum
6-minute
average,
percent
2.5C
. 0
0
0
2.8C
o
0
of /
0
From continuous monitor; not
.Reference Method 9.
Single stack exhaust.
°Not 6-minute average; highest
average from continuous
.monitor during a Method 5 test.
Two separate tests at Plant C.
^Monovent exhaust.
Reference Method 22.
4-33
-------
TABLE 4-5. SUMMARY OF OPACITY DATA FROM SHOP ROOF MONITORS
ON CARBON STEEL EAF SHOPS15,20,27,31-33,44!,45
Plant
Ca
(NSPS)
ca,b
(NSPS)
Gc
(NSPS)
Hd
Ic
Je
(NSPS)
Kf
(NSPS)
N9
(NSPS)
Furnace/
vessel
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
Furnace
process
Charge
Melt
Tap
Heat cycle
Charge
Melt
Tap
Charge
Melt
Tap
Heat cycle
Charge
Melt
Tap
Charge
Melt
Tap
Charge
Melt
Tap
No. of
6-minute
. averages
5
33
6
88
46
10
95
5
10
15
106
10
5
90
5
30
94
14
Maximum
6-minute
average
opacity,
percent
1.3
2.7
0.4
5.0
'
0 :
!
12
0
33
0
4.4
4.2
5.0
0
0
oi
3.3
3.1 '
1.0:
1
Average
of
6-minute
averages,
percent
0.5
•0.1
0.1
0.25
0
5.0
0
23.0
0
1.0
0.2
0.6
0
0
0
0.7
0.1
0.1
uUtilizes total furnace enclosure, canopy hood, closed roof.
^Second test at Plant C. i
^Utilizes canopy hood, DEC system, and closed roof. \
Utilizes canopy hood, DEC system, and open roof monitor.
Utilizes segmented canopy hood, DEC system, local tapping hood, closed
-roof over furnace/open roof monitor elsewhere. ;
Utilizes segmented canopy hood, side draft hood, scavenger, closed
roof. !
9Utilizes canopy hood, DEC system, scavenger ducts with cross-draft
partitions, closed roof over furnace/open elsewhere. !
4-34
-------
in the data base. The shop walls at Plant B do not extend to ground
level. This creates cross-draft problems, which adversely impact on
capture device efficiencies by causing high shop monitor opacities. As
noted above, Plants L, M, and 0 were not visited as examples of BDT.
4.6.2 Specialty Steel Shops
Source test data were obtained from eight specialty steel shop
fabric filter systems. Table 4-6 presents a summary of data relating to
the plants tested. These data include fabric filter units handling EAF
dust alone, ADD dust alone, and EAF and AOD dust combined. The tested
plants represent the range of facilities that may be regulated by any
new NSPS with regard to furnace/vessel size, ducting configuration, and
capture equipment. Scrap type, a factor when evaluating emissions from
carbon steel shops, is not a factor with specialty steel shops, because
of their almost exclusive use of clean scrap.
Figure 4-9 presents the results of measurements of particulate
concentrations at Plants P, Q, and R. Each data point represents a
separate sample collected by EPA Method 5. The maximum value for an
average of three test runs was 3.5 mg/dscm (0.0015 gr/dscf) at Plant P.
Data from Plant P and the first test at Plant Q, shown in Figure 4-9,
were obtained by EPA during the emission test program. Because the
emission test reports are incomplete or contain discrepancies, the
remainder of the data in Figure 4-9 are presented as supplementary
information. Testing was performed with hi-vol samplers, at six facilities
to determine compliance with State regulations. These data will not be
presented because the hi-vol technique provides biased results and the
technique is not recommended by the EPA.
Opacity readings were obtained by EPA personnel from the fabric
filter exhaust gas stream at Plants P, Q, and S. All of these units
have monovent exhausts except Plant Q, which has stub stacks. These
data are presented in Table 4-7. As with the carbon steel shop data,
the maximum opacity observed by Reference Method 9 was zero percent.
Table 4-8 presents visible emission data obtained from the shop
roof monitors at Plants Q and S. Emissions from the shop roof monitors
were below 5 percent opacity. Both shops have closed roof monitors over
the furnace/vessel area.
4-35
-------
TABLE 4-6. SUMMARY OF SPECIALTY STEEL PLANT DATA
No. furnaces/ ;
Furnace/vessel vessels ducted ;
type and size, to tested ; Capture
Plant
P
Q
Q
R
S
aCFH =
CH =
CR =
SCH =
SD =
Reference
46-48
13, 21,
49, 50
51
51
52-54
close-fitting hood
single canopy hood
closed roof
a
Mg (tons) fabric filter : equipment
EAF:
ADD:
ADD:
AOD:
EAF:
AOD:
EAF:
AOD:
29.0 (32)
29.0 (32)
18.1 (20)
18.1 (20)
38.1 (42)
45.4 (50)
45.4 (50)
45.4 (50)
2
2
1
1
2
1
1
1
.' SCH, CR
SCH, CR
•'• SCH, SD,
1 SCH, SD,
CH, CR
CFH, CH,
CH, CR
' CFH, CH,
'
CR
CR
CR
CR
segmented canopy hood :
scavenger duct
4-36
-------
gr/dscf mg/dscm
3
0.003 ,
o
ce
•z^
UJ
o
z
o
UJ
_J
o
0.002 -
0.001
KEY
TEST RUN
AVERAGE
Pi
PLANT
FURNACE TYPE
REFERENCE
PS - Q
EAF,AOD AOD
46 21
Q R
AOD EAF.AOD
49 51
Figure 4-9. Summary of particulate matter source data for
specialty steel shop fabric filters (Reference Method 5).
4-37
-------
TABLE 4-7. SUMMARY OF VISIBLE EMISSION DATA FROM FABRIC
FILTERS ON EAF'S AND AOD'S AT SPECIALTY STEEL SHOPS21;,46,52
Plant
P
Q
S
Length of
observations
in. minutes
438a
408b
336?
318b
Maximum VE's
based on
6-minute
average,
percent
0
0
0
0
uOne fabric filter for EAF's and AOD vessels.
DAOD vessel fabric filter.
CEAF fabric filter.
4-38
-------
TABLE 4-8. SUMMARY OF VISIBLE EMISSION DATA FROM
SHOP ROOF MONITORS ON SPECIALTY STEEL SHOPS21,52
.Plant
Q3
sb
Furnace/
vessel
AOD
EAF
and
AOD
Furnace
process
Heat cycle
Heat cycle
No. of
6-mihute
averages
69
55
Maximum
VE's based
on 6-minute
averages ,
percent
0
5
Average
of
6-minute
averages,
percent
0
0 . 1
.Utilizes canopy hood, scavenger duct, closed roof.
Utilizes close-fitting hood, canopy hood, closed roof.
4-39
-------
Plant P is an older shop and, while nominally a closed shop, has
numerous open seams between sheet metal sections on the walls| and does
not represent BDT. Visible emissions were seen exiting the building
from these open seams. The data obtained for the shop monitor emissions
at Plant P are not reported here and are not included in the data base.
4.6.3 Dust-Handling Equipment ;
Table 4-9 presents visible emission data obtained from the dust-
handling equipment at Plants B, C, P, and Q. The data for Plants B and
C were obtained using Reference Method 9 and show a maximum 6;-minute
average opacity of 0.6 and 5.0 percent, respectively. Some of the data
for Plant Q were obtained with Reference Method 9 and show a maximum
6-minute average opacity of 7.3 percent. The data for Plant P and some
data for Plant Q were obtained using Reference Method 22, and no visible
emissions were observed during a total of 654 minutes of observation at
these two facilities. All of these plants use an enclosed system of dust
collection. , :
4.6.4 Fluorides and Trace Elements in Particulate Emissions ;
The fabric filter exhaust gas particulate and fabric filter dust
catch at Plants P and Q were analyzed for fluorides, chromium, lead, and
nickel. These limited results are presented in Table 4-10. Emissions
to the atmosphere of these compounds do not appear to be significant,
with maximum emissions being only 495 kg/yr (1,090 Ib/yr)'of .flouride.
The other compounds analyzed are emitted in significantly lower quantities.
4-40
-------
TABLE 4-9. SUMMARY OF VISIBLE EMISSION DATA FROM DUST-HANDLING
SYSTEMS AT EAF AND AOD VESSEL STEEL MILL FACILITIES21,24,46,55
Plant
Ba
Ca
Pb
Qd
Qd
uEAF fabric
Length of
observations
in minutes
12
20
594
60
48
filter.
Maximum
6-minute
average,
percent
0.6
5.0
Oc
Oc
7.3e
"EAF/AOD fabric filter.
EPA Method 22 (no emissions were visible).
The operation observed did not include the
.dust transfer from the storage silo to truck.
°AOD fabric filter.
The operations observed included the dust
transfer from the storage silo to truck.
4-41
-------
TABLE 4-10. SUMMARY OF TRACE CONSTITUENT CONCENTRATIONS ANALYSIS, ppm
(kg/yr, Ib/yr)
Plant:
Constituent
Fluorides
Chromi urn
Lead
Nickel '
p46
Fabric filter
exhaust gas
31,600
(495, 1,090)
17,400C
(277, 594)
5,800C
(94, 203)
7,600C
(119, 267)
Fabric
filter
catch
47,200
39,200
8,400
16,300
Q21
Fabric filtfer
exhaust gas:
10,300
(15, 29)
7S600C
(11, 21)
<2,300C'd
(<3, <8)
6,800C
(7, 21)
Fabric
filter
catch
15,200
49,200
• 2,200
21,500
aAssumes emission concentration and flow rate of tested plant: and 4,950
.operating hours per year.
DAssumes production of 37,420 Mg/yr (41,250 tons/yr). ;
^Average of two samples. . ,
Detection limit.
4-42
-------
4.7 REFERENCES FOR CHAPTER 4
1. Fennelly, P. F., and P. D. Spawn. Air Pollution Control Techniques
for Electric Arc Furnaces in the Iron and Steel Foundry Industry.
U.S. Environmental Protection Agency. Research Triangle Park, N.C.
EPA-450/2-78-024. June 1978. 221 p.
2. Sahagian, J., P. F. Fennelly, and M. Rei. Inspection Manual for
the Enforcement of New Source Performance Standards: Steel Producing
Electric Arc Furnaces, U.S. Environmental Protection Agency.
Washington, D.C. EPA 340/1-77-077. May 1977. 73 p.
3. Brand, P. G. A. Current Trends in Electric Arc Furnace Emission
Control. Iron and Steel Engineer. 58(2):59-64. February 1981.
4. Telecon. Terry, W., Midwest Research Institute, with Eckstein, G.,
Bethlehem Steel Corporation. September 3, 1981. Information on
Bethlehem Steel-Johnstown facility.
5. Telecon. Terry W., Midwest Research Institute, with Woolen, C.,
Pennsylvania Engineering Corporation. August 19, 1981. Information
on total furnace enclosure systems.
6. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R. E., EPA/ISB. December 10, 1980. Report on November
visit to Hoeganaes Corporation Steel plant in Gal latin, Tennessee.
7. Marchisio, C. Electric Furnace Pollution Control. Iron and Steel
Maker. 8(6):36-40. June 1981.
8. Telecon. Terry W., Midwest Research Institute, with Bonistalli, R.,
Obenchain Calumet Corporation. August 18, 1981. Information on
total furnace enclosure systems.
9. Kaercher, L. T., and J. D. Sensenbaugh. Air Pollution Control for
an Electric Arc Furnace Meltshop. Iron and Steel Engineer.
51(5):47-51. May 1974.
10. Memo from Terry, W., Midwest Research Institute, to Iversen, R.,
EPA/ISB. November 18, 1981. Minutes of the November 5, 1981,
meeting with the EPA and the American Iron and Steel Institute.
11. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R,, EPA/ISB. January 6, 1981. Site visit report—North
Star Steel, Monroe, Michigan.
12. Memo and attachments from Banker, L. , Midwest Research Institute,
to Iversen, R., EPA/ISB. March 30, 1981. Source test observation
report—U.S. Steel Corporation, Baytown, Texas.
13. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R. , EPA/ISB. May 18, 1981. Source test observation
report for Carpenter Technology Corporation, Reading, Pennsylvania.
4-43
-------
14. Hutten-Czapski, L. Efficient and Economical Dust Control System
for Electric Arc Furnace. Sidbec-Dosco, Usine de Contrecoeur.
Contrecoeur, Quebec, Canada. 12 p.
15. Background Information for Standards of Performance: Electric Arc
Furnaces in the Steel Industry, Vol. I. U.S. Environmental Protection
Agency. Research Traingle Park, N.C. Publication No. EPA-450/2-74-017a.
p. 88.
16. Hot Environments—1980. U.S. Department of Health and Human Services.
Cincinnati, Ohio. Publication No. DHHS. No. 80-132.
17. Telecon. Terry W., Midwest Research Institute,
Union Carbide Corporation. September 4, 1981.
AOD utilization.
with Sarlitto, R. ,
Information about
18.
19.
20.
21
22.
23.
24.
25.
Goodfellow, H. Solving Fume Control and Ventilation Problems for
an Electric Melt Shop. Air Pollution Control Association. Hatch
and Associates, Montreal, Canada. June 1980. 22 p.
Goodfellow, H. D. Solving Air Pollution Problems in the Metallurgical
Industry. Seventh International Clean Air Congress. Adelaide,
Australia. August 24-28, 1981. 14 p.
Electric Arc Furnace Baghouse Compliance Test: Sharon Steel Corporation,
Farrell, Pennsylvania, January 6, 7, and 8, 1981. WFI Sciences
Company. Pittsburgh, Pa. WFI Science Report No. 8343.
Emission Test Report: Carpenter Technology Corporation;, Reading,
Pennsylvania. PEDCo Environmental, Inc. Cincinnati, Ohio. Contract
No. 68-02-3546, Work Assignment No. 2. July 1981.
Source Testing Report: The Babcock and Wilcox Company Electric Arc
Furnace, Beaver Falls, Pennsylvania. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. EPA Publication No. EMB
73-ELC-l. January 1973. 32 p.
Memo from Terry, W., Midwest Research Institute, to Iversen, R.,
EPA/ISB. October 31, 1980. Site visit report—Babcock; and Wilcox,
Beaver Falls, Pennsylvania. ;
Compliance Tests Under New Source Performance Standards^: Florida
Steel Corporation, Charlotte, North Carolina. Sholtes & Koogler*
Environmental Consultants. Gainesville, Florida. January 1980.
Addendum No. 1, Compliance Tests Under New Source Performance
Standards: Florida Steel Corporation, Charlotte, North Carolina.
Sholtes & Koogler, Environmental Consultants. Gainesville, Florida.
January 1980. :
4-44
-------
26.
27.
28.
29.
30.
31
32.
33.
34.
35.
36.
37.
Memo from Terry, W., Midwest Research Institute, to Iversen, R,
EPA/ISB. September 3, 1980. Site visit report—Florida Steel
Corporation, Charlotte, North Carolina.
Source Sampling Report—Hoeganaes, Inc. (83-00129-01), Gal latin,
Tennessee, Baghouse Serving the Electric Arc Furnace. Environmental
Management, Planning and Engineering, Nashville, Tennessee.
October 1981.
Stack Particulate Sampling: North Star Steel Corporation, Monroe,
Michican. Industrial Health Engineering Associates, Inc., Minnea-1
polis, Minnesota, for Ferrco Engineering, Ltd. Ontario, Canada.
Project 335-006. February 1981.
Report of Official Air Pollution Emission Tests Conducted on the
Electric Arc Furnace .Cadre Baghouse Exhaust at.Raritan River Steel,
Perth Amboy, New Jersey, on June 12 and 13, 1980. Rossnagel & Associ-
ates. Medford, N.J. Test Report No. 8132. June 17, 1980.
Compliance Sampling of Stack Emissions: Electric Arc Furnaces Baghouse
Exhaust Stack, Nucor Steel, Jewett, Texas, on November 17-18, 1981.
Southwestern Laboratories. Houston, Texas. Project No. 54~830A.
December 1981.
Compliance Tests Under New Source Performance Standards:
Corporation, Tennessee Mill Division, Jackson, Tennessee.
Koogler. Gainesville, Florida. December 1981.
Florida Steel
Sholtes and
Particulate Emission Tests for Lukens Steel
Fuller Company, Catasauqua, Pennsylvania.
Electric Melt Shop.
September 23, 1973.
Memo and Attachment from Te.rry, W., Mi dwes-t Research Institute, for EAF
files. May 17, 1982. Trip reports to Luken Steel Corporation, Coats-
vine, Pennsylvania (August 1972), and Bethlehem Steel Corporation,
Seattle, Washington (March 1973).
Letter and attachments from Lukas, A. W., J&L Steel Corporation, to
Banker, L. C., Midwest Research Institute. March 2, 1981. Submis-
sion of compliance test report for J&L Steel-Pittsburgh Works.
Memo from Banker, L., Midwest Research Institute, to Iversen, R.,
EPA/ISB. July 28, 1980. Source emission test observation report-
Jones and Laugh!in Steel Corporation, Pittsburgh, Pa.
Visible Emission Survey Report: Atlantic Steel Corporation,
Cartersville, Georgia. PEDCo Environmental, Inc. Cincinnati,
Ohio. Contract No. 68-02-3546, Work Assignment No. 2. March 1981.
Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R. E., EPA/ISB. March 23, 1981. Source test observation
report—Atlantic Steel Company, Cartersville, Georgia.
4-45
-------
38. Visible Emission Survey Report: U.S.. Steel, Baytown, Texas. PEDCo
Environmental, Inc. Cincinnati, Ohio. Contract No. 68-02-3546,
Work Assignment No. 2. April 1981.
39. Visible Emission Survey Report: Chaparral Steel Corporation,
Midlothian, Texas. PEDCo Environmental, Inc. Cincinnati, Ohio.
Contract No. 68-02-3546, Work Assignment No. 2. April 1981.
40. Memo and attachments from Banker, L., Midwest Research Institute,
to Iversen R. E., EPA/ISB. March 26, 1981. Source test observation '
report—Chaparral Steel Corporation, Midlothian, Texas. \
41. Visible Emission Survey Report: Bethlehem Steel, Los Angeles,
California. PEDCo Environmental, Inc. Cincinnati, Ohio. Contract
No. 68-02-3546, Work Assignment No. 2. May 1981.
42. Memo and attachments from Banker, L., Midwest Research Institute,
to Iversen, R., EPA/ISB. June 3, 1981. Source test observation
report—Bethlehem Steel Corporation, Los Angeles, California.
43. Memo from Banker, L., Midwest Research Institute, to Project File.
September 28, 1981. Discussion of material excluded from data
base.
44. Visible Emission Data for J£L Steel Corporation, Pittsburgh,
Pennsylvania. Allegheny County Health Department. Pittsburgh, Pa.
Undated.
45. Memo and attachments from Terry, W., Midwest Research Institute, to
Pan!, D., EPA/SDB. August 31, 1982. Source Test'Trip Report—Chaparral
Steel, Midlothian, Texas. '
46. Emission Test Report: AL Tech Specialty Steel Corporation, Watervliet,
New York. PEDCo Environmental, Inc. U.S. Environmental Protection
Agency. EPA EMB Publication No. 80-ELC-7. July 1981.
47. Memo from Banker, L., Midwest Research Institute, to Iversen, R.,
EPA/ISB. August 13, 1980. Site visit'report—AL Tech Specialty
Steel, Watervliet, N.Y. !
48. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R., EPA/ISB. May 8, 1981. Source test observation report—
AL Tech Specialty Steel Company,. Watervliet, N.Y.
49. Memo and attachment from Maxwell, W. H., Midwest Research Institute,
to EAF files. August 18, 1981. August 1978 source test report for
Cartech-Reading facility.
50. Memo from Banker, L., Midwest Research Institute, to Iversen, R.,
EPA/ISB. January 27, 1981. Site visit report—Carpenter Technology
Corporation, Reading, Pennsylvania.
4-46
-------
51. Letter and attachments from Geiser, L. H.,. Carpenter Technology
Corporation, to Banker, L. C. , Midwest Research Institute.
January 30, 1981. Submission of test data for Bridgeport and
• Reading shops.
52. Visible Emission Survey Report: Eastern Stainless Steel Company,
Baltimore, Maryland. PEDCo Environmental, Inc. Cincinnati, Ohio.
Contract No. 68-02-3546, Work Assignment No. 2. December 1980.
53. Memo from Banker, L., Midwest Research Institute, to Iversen, R.,
EPA/ISB. October 17, 1980. Site visit report—Eastern Stainless
Steel Company, Baltimore, Maryland.
54. Memo and attachments from Banker, L,, Midwest Research Institute,
to Iversen, R., EPA/ISB. December 28, 1980. Source test observation
report—Eastern Stainless Steel Company, Baltimore, Maryland.
55. Visible Emission Summary Report for Carpenter Technology Corporation,
Reading, Pennsylvania. PEDCo Environmental, Inc. April 1982.
4-47
-------
-------
5. MODIFICATION AND RECONSTRUCTION
New source performance standards (NSPS) apply to facilities whose
construction, modification, or reconstruction commenced [as defined under
40 CFR 60.2(i)] after proposal of the standards. Such facilities are
termed "affected facilities." Standards of performance are not appli-
cable to "existing facilities," which,, as defined under 40 CFR 60.2(aa),
are facilities whose construction, modification, or reconstruction com-
menced on_£r_before proposal of the standards. However, an existing
facility may become an affected facility and therefore subject to standards
if the facility undergoes modification or reconstruction.
Modification and reconstruction are defined under 40 CFR 60.14 and
60.15, respectively. The provisions of these sections are summarized in
Section 5.1 below. Section 5.2 examines the applicability of these
provisions to electric arc furnace and argon-oxygen decarburization
vessel facilities and describes conditions under which existing facilities
could become subject to standards of performance.
It is important to note that a stationary source may contain both
affected and existing facilities and that reclassifying a facility from
existing to affected status by modification or reconstruction does not
subject any other facility within that source to the standards of
performance.
5.1 SUMMARY OF MODIFICATION AND RECONSTRUCTION PROVISIONS
5.1.1 Modification
Section 40 CFR 60.14 defines modification as follows:
Except as provided under paragraph (e) and (f) of
this section, any physical or operational changes to
an existing facility which result in an increase in
emission rate to the atmosphere of any pollutant to
which .a standard applies shall be a modification.
Upon modification, an existing facility shall become
5-1
-------
an affected facility for each pollutant to which a
standard applies and for which there is an increase ;
in the emission rate.
Paragraph (e) specifies certain physical or operational changes that
are not considered modifications irrespective of any changes in;the
emission rate. These changes include:
1. Routine maintenance, repair, and replacement;
2. An increase in production rate accomplished without a capital
expenditure [as defined in Section 60.2(bb)];
3. An increase in hours of operation;
4. Use of alternate fuels or raw materials if the existing facility
were designed to accommodate the alternate fuel or raw material"•prior to
the standard [conversion to coal required for energy considerations, as
specified in Section 111 (a)(8) of the Clean Air Act, is also exempted];
5. Addition or use of any system or device whose primary function
is the reduction of air pollutants, except when an emission control
system is removed or replaced by a system considered to be less efficient;
and
6. Relocation or change in ownership.
Paragraph (f) provides that, should conflicts in provisions arise, the
provisions of each subpart supercede those of 40 CFR 60.14.
Paragraph (b) of CFR 60.14 clarifies what constitutes an increase in
emissions and the methods for determining the increase. These methods
include the use of emission factors, material balances, continuous moni-
toring systems, and manual emission tests. Paragraph (c) of 40:CFR 60.14
affirms that the addition of an affected facility to a stationary source
does not make any other facility within the source subject to the standards
of performance.
5.1.2 Reconstruction
Section 40 CFR 60.15 defines reconstruction as follows:
An existing facility, upon reconstruction, becomes an
affected facility, irrespective of any change in
emission rate. "Reconstruction" means the replace-
ment of components of an existing facility to such an,
extent that: (1) the fixed capital cost of the new
components exceeds 50 percent of the fixed capital
cost that would be required to construct a comparable
entirely new facility, and (2) it is technologically
5-2
-------
. and economically feasible to meet the applicable
standards set forth in this part.
The purpose of this provision is to ensure that an existing facility
is not perpetuated by replacing all but a few components rather than
totally replacing the facility in order to avoid becoming subject to
applicable standards of performance.
5.2 APPLICABILITY TO ELECTRIC ARC FURNACES AND ARGON-OXYGEN
DECARBURIZATION VESSELS IN THE STEEL INDUSTRY
The following components will be examined to decide whether an EAF
has been modified or reconstructed:
1. Furnace shell, including roof;
2. Transformer; and
3. Emissions Control system (capture devices, ductwork, control
device).
The following components will be examined to decide whether an AOD
vessel has been modified or reconstructed:
1. Vessel shell; and
2. Emission control system (capture devices, ductwork, control
device).
Except as noted in Section 5.1.1, any physical or operational change
to any of these items that would result in an increase in the particulate
emission rate (emissions per hour) to the atmosphere could be considered
a modification. Changes to an EAF that could increase emissions and be
considered modifications include changes to the furnace shell (including
the roof), an increase in transformer capacity, or changes to the emissions
control system. Changes to an AOD vessel that could increase emissions
include changes to the vessel shell or changes to the emissions control
system.
Replacement of more than one item may be reconstruction, depending
on the costs involved. The installation of water-cooled walls on an EAF,
an increase in the transformer capacity to raise the melt rate, or the
conversion of a normal-power EAF to ultra high power (UHP) may be recon-
struction because the cumulative capital costs of the new components
exceed 50 percent of the capital costs for a new facility.
5-3
-------
The enforcement division of the appropriate EPA regional office
should be contacted whenever a source has questions regarding modification
and reconstruction.
5-4
-------
6. MODEL PLANTS AND REGULATORY ALTERNATIVES
6.1 INTRODUCTION
Model plants are used as parametric descriptions of the general
types of EAF's and AOD vessels that are expected to be constructed and
modified/reconstructed within the industry in the future. Model plants
are used because it is impractical to evaluate environmental, economic,
and energy impacts of the various emission capture and control technologies
for every possible plant configuration. In order that the analyses be
representative of the industry as a whole, the model plants are designed
to cover the range of possible plant arrangements and are not necessarily
intended to represent any particular facility. The following sections
of this chapter describe the model plants and the alternative methods of
regulating the affected facilities.
6.2 MODEL PLANTS
Nine model plants have been developed to represent EAF's and AOD
vessels in steel mills. Brief descriptions of the plants are presented
in Table 6-1. Model plants 1 through 6 represent a range of new, modified,
or retrofit EAF's in the carbon steel shops. Model plant 7 represents a
likely retrofit EAF in either a carbon or specialty steel shop. Model „
plants 8 and 9 represent EAF's and AOD vessels in new specialty steel
shops.
Six model furnace/vessel sizes have been selected to characterize
the spectrum of EAF and AOD steel facilities. These furnaces include
four EAF sizes (three normal-power, one UHP) and two AOD vessel sizes.
The furnace sizes are based on the furnace design capacity rather than
on the melting capacity in megagrams per hour (tons per hour) because of
the variability in the grades of steel produced and the amount of refining
6-1
-------
TABLE 6-1. MODEL PLANTS
Number
la
2a
3a
4a
5a
6a
7a,b
8fa
9b
Plant description
A new shop with a
22.7-Mg (25-ton) EAF.
A new shop with a
90.7-Mg (100-ton) EAF.
A new shop with a
272.2-Mg (300-ton) EAF.
A new shop with a
136.1-Mg (150-ton) UHP EAF.
A modification of a
136.1-Mg (150-ton) normal
power EAF to a UHP EAF.
A retrofit of a
272.2-Mg (300-ton) EAF
into an existing shop.
A retrofit of a
22.7-Mg (25-ton) EAF into
an existing shop.
A new shop with a
22.7-Mg (25-ton) ADD vessel
and a 22.7-Mg (25-ton) EAF.
A new shop with a
90.7-Mg (100-ton) AOD vessel
and a 90.7-Mg (100-ton) EAF.
Steel
Mg/yr
37,420
163,300
596,000
434,800
434,800
596,000
37,420
37,420
163,300
production
(tons/yr)
(41,250)
: (180,000)
(657,000)
i
(479,300)
(479,300)
i (657,000)
(41,250)
(41,250)
(180,000)
uCarbon steel shop.
Specialty steel shop.
6-2
-------
done in the furnace, which affects the heat length and number of heats
per day. The sizes of EAF's with normal power capacities are 22.7 Mg
(25 tons), 90.7 Mg (100 tons), and 272.2 Mg (300 tons), representing
small, medium, and large EAF's, respectively. The fourth EAF size,
136.1 Mg (150 tons), represents the UHP furnaces. The ADD vessel sizes
are 22.7 Mg (25 tons) and 90.7 Mg (100 tons), representing the range of
vessel sizes that are projected to be built.
The technical parameters are presented in Table 6-2 for EAF's in
carbon steel shops and in Table 6-3 for EAF's and AOD vessel-s in specialty
steel shops. The parameters are based on design and performance data
obtained during the development of this document.1
Design and operating improvements have been made in the industry
since the development of the existing standards, of performance to achieve
higher production rates. These improvements have led to certain differences
between the parameters noted in Tables 6-2 and 6-3 and those used during
the development of the existing standards of performance, particularly
with regard to heat times. The average heat time for carbon steel,
normal-power EAF's has decreased in the past 5 years, from 3.5 to 3 hours.
The reasons for this decrease in heat times include: improved melting
efficiency and operating practices; the use of improved refractories,
water-cooled furnace walls and roofs, and higher current density, which
allows more power input a'nd faster melting; more refining performed
outside the furnace; increased computerization for greater control of
electrical input; faster and more accurate analysis and processing of
metallurgical samples; and increased competition from foreign steel
companies, which necessitates more economical production.2,3 Compared
to.a normal-power EAF, the UHP furnace has an even shorter heat time
(approximately 2 hours), which is facilitated by higher transformer
capacities and the use of water-cooled furnace walls and roofs.
The average heat time for a specialty steel EAF shop has been
reduced in the past 5 years from 7 to 3 hours. This reduction in heat
time results primarily from the increased use of duplexing, which is the
melting and refining of steel in separate vessels (the EAF functions
primarily as a metal melter and another vessel is used to refine the
molten metal). This method of operation is significantly different
6-3
-------
TABLE 6-2. MODEL FURNACE PARAMETERS—CARBON STEEL1
Furnace ID
Process information
Furnace type
Furnace capacity, Mg/heat
tons/heat
Furnace transformer rating
Heat time, h
Production rate, Mg/day
tons/day
heats/day
Operating time, h/day
days/yr
h/yr
Annual production, Mg xlO3
tons xlO3
Charge time, min
Backcharge time, min
Slag time, min
W '
Tap time, min
Operating (tap) temperature
°C
°F
Transformer capacity, kVA
Electricity use
MJ/Mg of steel
kWh/ton of steel
Shop monitor height, m
ft
Uncontrolled emissions
kg/h
Ib/h
kg/Mg
Ib/ton
Mg/yr° ,
tons/yr
A
EAF
22.7
25
Normal
3
136
150
6
24
275
6,600
37.42
41.25
3-5
3-5
5
5-7
1620
2950
10,600
1,985
500
27.4
90
114
250
15
30
561.3
618.8
B
EAF
90.7
100
Normal
3
544
600
6
24
300
7,200
163.3
180.0
3-5
3-5
5
5-7
1620
2950
42,500
1,985
500
33.5
110
454
1,000
15
30
2,499.4
2,700.0
C
EAF:
272.2
300
Normal
3
1,633
1,800
6
24
365
8,760
596.0
657.0
5-10
5-10
5
7-10
1620
2950
127,500
1 ,985
500
38.1
125
1,361
3,000
15
30
6,940.5
9,855.0
D
EAF
136.1
150
UHP
2
1,225
1,350
9
24
355
8,520
434.8
479.3
5-7
5-7
5
6-8
1620
2950
75,000
1 ,985
500
36.6
120
1,020
2,250
15
30
6,522.3
7,189.5
aThe daily production averages about six heats (furnaces A, B, and C) and
nine heats (furnace D) because of process delays, refractory' gunning, and
.furnace maintenance that is performed between heats. ,
The annual emissions are calculated by multiplying the annual production
by the emission rate.
6-4
-------
TABLE 6-3. MODEL FURNACE PARAMETERS—SPECIALTY STEEL1
Furnace/vessel ID
Process information
Furnace type
Furnace capacity, Mg/heat
tons/heat
Furnace transformer rating
Heat time, h
Production rate, Mg/day
tons/day
heats/day
Operating time, h/day
days/yr
h/yr
Annual production, Mg xlO3
tons xlO3
Charge time, min
Backcharge time, min
Slag time, min
Tap time, min
Operating (tap) temperature
°C
°F
Transformer capacity, kVA
Electricity use
MJ/Mg of steel
kWh/ton of steel
Shop monitor height, m
ft
Uncontrolled emissions
kg/h
Ib/h
kg/Mg
Ib/ton
Mg/yrD ,
'Z3r -J K
tons/yr
A
EAF
22.7
25
Normal
3
136
150
6
24
275
6,600
37.42
41.25
3-5
3-5
5
5-7
1620
2950
10,600
1,985
500
27.4
90
114
250
15
30
561.3
618.8
B
EAF
90.7
100
Normal
3
544
600
6
24
300
7,200
163.3
180.0
3-5
3-5
5
5-7
1620
2950
42,500
1,985
500
33.5
no
454
1,000
15
30
2,499.9
2,700.0
E
AOD
22.7
.25
—
1.5
136
150
6
24
275
6,600
37.42
41.25
3
« _
3
5
1620
2950
,
_ *
™" •"*
27.4
90
121
267
8
16
299.3
330.0
. F
AOD
90.7
100
—
1.5
544
600
6
24
300
7,200
163.3
180.0
3
3
5
1620
2950
--
» w
~~
33.5
110
484
1,067
8
16
1,306.4
/ 1,440.0
aFurnaces A and B have an average production of six heats per day.
Vessels E and F refine the molten "steel that was melted in Furnaces A
and B, respectively. Therefore, even though the heat time for vessels
.is half that of the furnaces, they also average six heats a day.
The annual emissions are calculated by multiplying the annual production
by the emission rate.
6-5
-------
because in the past EAF's were used for both melting and refining in
specialty steel shops. Industry sources indicate that most specialty
steel shops that are to be built will have a duplexing system, with
EAF's used to melt the scrap and do minimal refining.4-7 .
The nature of the gas stream from EAF's and AOD vessels is such
that most of the associated ductwork and gas cleaning devices can be
constructed of carbon steel. The direct fume extraction equipment, such
as the DEC on the EAF and the close-fitting hood on the AOD vessel, are
subject to very high temperatures and must either be lined with refractory
material or ceramic fiber blankets or must be water-cooled to avoid
excessive wear.6 The diverter stack or close-fitting hood above the AOD
vessel is typically made of stainless steel so that it can withstand the
heat from the vessel. The exhaust gas characteristics, such as particle
size and composition, are discussed in Chapter 3.
The existing combinations of capture technologies for the EAF's and
AOD vessels are presented in the matrix in Table 6-4. Each capture
configuration is presented as an example of the equipment that a model
plant might utilize to meet one of the regulatory alternatives that is
discussed later in this chapter. The capture technologies are based on
those used in well-controlled shops that were observed during this
study.
Table 6-5 shows the air flow rates per megagram (ton) of furnace/vessel
capacity used to establish the flows shown in Tables 6-6 through 6-9.
This information is based on that obtained from well-controlled shops.1
The fabric filter information for the EAF's in carbon steel shops is
presented in Tables 6-6 and 6-7 and for the EAF's and AOD vessels in
specialty steel shops in Tables 6-8 and 6-9. The information presented
for the fabric filters was based on worst-case conditions. In Tables 6-8
and 6-9, some of the flow rates were derived by combining the flow for
an AOD vessel with the flow for an EAF. This was done to provide adequate
!' ' " •
flow for both units when they are in their most polluting mode of operation
(i.e.,.EAF melting, AOD refining with high percentage of oxygen). These
flow figures will be used for calculating costs in Chapter 8. Other,
less conservative air flows might be possible when both the EAF and AOD
vessel are ducted to 'the same fabric filter.
6^-6
-------
TABLE 6-4. MODEL PLANTS WITH CAPTURE CONFIGURATION OPTIONS
Capture configurations'
Model plant
designation
Model
plant
No.
EAF:
DEC,
. CH,OR
AOD:
CFH,
CH.OR
EAF:
DET.SCH ,
SD.CDP,
(or CH,
TH.SH),
CR/OR
AOD:
CFH.CH,
COP, SO,
CR/OR
EAF:
DEC.SCH,
$D,CDP,
(or CH,
•TH.SH),
CR
AOD:
CFH.CH,
SO, COP,
CR
EAf:
TFE.CH
TH,CR
Carbon Steel Shop
25-ton EAF 1
TOO-ton EAF 2
300-ton EAF. 3
150-ton UHP EAF . 4
150-ton normal EAF 5
modification to UHP EAF
300-ton retrofit 6
25-ton retrofit 7
Specialty Steel Shop
25-ton EAF and AOD 8
Ducted together
Ducted separately
100-ton EAF and AOD 9
Ducted together
Ducted separately
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
COP = cross-draft partitions
CFH = close-fitting hood
CH = canopy hood
CR = closed roof
CR/OR = closed roof (over furnace)/open roof monitor (elsewhere)
DEC = direct-shell evacuation control
OR = open roof monitor
SCH = segmented canopy hood
SD = scavenger duct
SH = slagging hood
TFE = total furnace enclosure
TH = tapping hood
UHP = ultra high power
6-7
-------
TABLE 6-5. AIR FLOW RATES PER UNIT OF FURNACE/VESSEL CAPACITY1
Furnace Furnace ID
type (plant type)
EAF A (carbon, specialty)
EAF A (carbon, specialty)
EAF B (carbon, specialty)
EAF B (carbon, specialty)
EAF C (carbon)
EAF C (carbon)
EAF D (carbon)
EAF D (carbon)
ADD E (specialty)
AOD E (specialty)
ADD F (specialty)
AOD F (specialty)
*See Tables 6-10 and 6-11.
bAt 121°C (250°F).
5(m3/s)/Mg— cubic meters per second
Regulatory
alternative
1, 2
3
1, 2
3
1, 2
3
1, 2
3
1, 2
3
1, 2
3
per megagram.
Alrvfl
(raa/s)/Mgc ;
2.00
2.10
2.00
2.10
0.80
0.91
2.07
2.18
2.43 :
2.64
2.43
2.64
owb d
acfm/ton
3,850
4,050
3,850
4,050
1,550
1,750
4,000
4,200
4,700
5,100
4,700
5,100
aacfm/ton— actual cubic feet per minute per ton.
6-8
-------
TABLE 6-6. MODEL FURNACE PARAMETERS—CARBON STEEL
FABRIC FILTER INFORMATION1
(Metric units)
Furnace type u
Capture equipment
Regulatory alternative
Air flow rate, m3/sd
Total
Canopy
DEC
TFE
Scavenger
Temperature, °C
No. compartments, total
No. bags, total
Sag type
Net air-to-cloth ratio6
Gross air-to-cloth ratio
No. fans
Fan wattage, each
Net cloth area,6 m2
Gross cloth area, m2
Monitor height, m
Moisture, percent
Pressure drop, cm
Furnace type3 .
Capture equipment
Regulatory alternative
Air flow rate, m3/s
Total
Canopy
DEC
TFE
Scavenger
Temperature, °C
No. compartments , total
No. bags, total
Bag type
Net air-to-cloth ratioe
Gross air-to-cloth ratio
No. fans
Fan wattage, each
Net cloth area,8 m
Gross cloth area, m2
Monitor height, m
Moisture, percent
Pressure drop, cm
A A
1 2
1 2
45.4 45.4
31.3 28.9
14.1 14.1
..
2.4
121 121
8 8
384 384
3.18:1 3.18:1
2.79:1 2.79:1
2 2
96,980 96,980
2,825.0 2,826.0
3,210.6 3,210.6
22.9 22.9
1 1
7.6-12.7 7.6-12.7
C
1
1
219.4
186.5
32.9
—
«
121
24
1,728
3.12:1
2.99:1
5
179,040
13,845.8
14,447.8
30.5
1
7.6-12.7
A
3
3
47.7
28.9
14.1
--
4.7
121
7
A
4
3
34.2
16.5
--
17.7
--
121
12
B
1
1
181.7
125.1
56.6
— •
—
121
20
420 288 1 ,440
3.13:1 3.05:1 "3.13:1
2.68:1
Z
104,440
3,010.0
3,511.6
22.9
1
7.6-12.7
C
2
2
219.4
772.3
32.9
--.
14.2
121
24
1,728
3.12:1
2.99:1
5
179,040
13,845.8
14,447.8
30.5
1
7.6-12.7
2.80:1
2
85,190
2,210.3
2,411.2
22.9
1
7.6-12.7
C
3
3
247.7
186.5
32.9
--
28.3
121
15
1 ,980
3.16:1
2.95:1
6
179,040
15,451,1
16,554.8
30.5
1
7.6-12.7
2.97:1
5
171,580
11,437.8
12,039.8
22.9
1
7.6-12.7
D
1
1
<
283.1
192.5
90.6
—
—
121
23
2,346
3.02:1
2.84:1
6
179,040
18,461.1
19,614.9
30.5
1
7.6-12.7
B :
2
2
181.7
115.6
56.6
--
9.5
121
20
1,440
3.13:1
2.97:1
5
171,580
11,437.8
12,039.8
22.9
1
7.6-12.7
0
2
2
283.1
178.3
90.6
--
14.2
121
23
2,346
3.02:1
2.87:1
6
179,040
18,461.1
19,614.9
30.5
1
7.6-12.7
B
3
3
191.1
115.6
56.6
--
18.9
121
17
1,530
3.13:1
2.94:1
5
171,580
12,039.8
12,792.3
22.9
1
7.6-12.7
D
3
3
297.3
178.4
90.6
'--
28.3
121
23
2,346
3.12:1
2.98:1
6
186,500
18,762.1
19,614.9
30.5
1
7.6-12.7
?From Table 6-2.
1 = DEC, canopy hood, open roof.
closed roof.
2 = DEC, segmented canopy hood, scavenger duct, cross-draft partitions, open/
3 = DEC, segmented canopy hood, scavenger duct, cross-draft partitions, closed roof. 4 = TFE^
canopy hood, closed roof.
jFroni Table 6-10.
°At temperature noted.
jlOne compartment off-line for cleaning.
Fabric filter monitor height above grade.
6-9
-------
TABLE 6-7. MODEL FURNACE PARAMETERS—CARBON STEEL
FABRIC FILTER INFORMATION1
(English units)
Furnace type b
Capture equipment
Regulatory alternative
Air flow rate, acfm
Total
Canopy
DEC
TFP
llC
Scavenger
Temperature, °F
No. compartments, total
No. bags, total
Bag type
Net air-to-cloth ratio"
Gross air-to-cloth ratio
No. fans
Fan horsepower, each
Net cloth area,6 ft2
Gross cloth area, ft2
Monitor height/ ft
Moisture, percent
Pressure drop, in. •
Furnace type b
Capture equipment .
Regulatory alternative
Air flow rate, acfBi"
Total
Canopy
DEC
TFE
Scavenger
Teaperature, °F
No. conpartsents , total
No. bags, total
Bag type
Net air-to-cloth ratio*
Gross air-to-cloth ratio
No. fans
Fan horsepower, each
Net cloth area,6 ft
Gross cloth area, ft2
Monitor height/ ft
Moisture, percent
Pressure drop, in.
A /A
1 2
1 2
96,250 96,250
66,250 61,250
30,000 30,000
5,000
250 250
8 8
384 384
A
3
3
101,250
61,250
30,000
—
10,000
250
7
A
4
3
72,500
35,000
37,500
250
12 '
B
1
1
385,000
265,000
120,000
--
— .
250
20
420 288 1,440
B!
2
2
385,000
245,000
120,000
! —
20,000
250
;20
1,440
B
3
3
405,000
245,000
120,000
•"-
40,000
250
. 17
1,530
3.18:1 3.18:1
2.79:1 2.79:1
2 2
130 130
30,240 30,240
34,560 34,560
75 75
1 1
3-5 3-5
C
1
1
465,000
395,250
69,750
—
250
24
1,728
3.13:1
2.68:1
2
140
32,400
37,800
75
1
3-5
C
2
2
465,000
365,250
69,750
30,000
250
24
1,728
3.05:1
2.80:1
2
115
23,760
25,920
75
1
3-5
C
3
3
525,000
395,250
69,750
60,000
250
15
3.13:1
2.97:1
5
230
123,120
129,600
75
1
3-5
D
1
1
600,000
408,000
192,000
—
250
23
1,980 2,346
3.13:1
2.97:1
5
230
123,120
129,600
75
1
. 3-5
,0
2
600, '000
378,000
192; ooo
30,000
250
23
2,346 '
3.13:1
2.94:1
5
230
129,600
137,700
75
1
3-5
D
3
630,000
378,000
192,000
60,000
250
23
2,346
3.12:1
2.99:1
5
240
149,040
155,520
100
1
3-5
3.12:1
2.99:1
5
240
149,040
155,520
100
1
3-5
3.16:1
2.95:1
6
240
166,320
178,200
100
1
3-5
' 3.02:1
2.84:1
6
240
198,720
211,140
100
1
3-5
3.02:1
2.84:1
6
240
198J20
211,140
100
'1
3-5
3.12:1
2.98:1
6
250
201 ,960
211,140
100
1
3-5
bl * DEC, canopy hood, open roof. 2 = OEC, segmented canopy hood, scavenger duct, cross-draft partitions, open/
closed roof. 3 = OEC, segmented canopy hood, scavenger duct, cross-draft partitions, closed roof. 4 = TFE,
canopy hood, closed roof.
jJFroia Table 6-10. !
°At temperature noted.
;0ne eofsparteent off-line for cleaning.
TFabric filter monitor height above grade. :
6-10
-------
rH •
2:
o
i-ii-i
t—
OS
O
U.
1— <
Di
LU
Ll-
CS
1— t
CD
-
!«. /"•*%
HWJ (/)
CK; CD
LU 3C
LU
•i—
^J
Furnace type .
Capture equipment
Regulatory alterna
. . t .
o o «s- •*
ID CO CM
«»• to I V f".
in CO CM Hr
m CM CM
—
r-» 10 <0 in i
.... i
i— in ID cn
CO r— in
r- *- «s i i
..tii
*— • m vo
co CM m
I-- en t— r- i
r^cd^-W '
«a>CT>r-«3- t
. t • .1
S CM i— ^
*f m r— 1 1
V0^« ' '
<3- CO r—
T3 "O
v> o
•^. o
aj -r-
re t. *J
cn1*-
O ' — o! fli fl)
r— ro O > u)
t- +J c t_> m ©
e> re LU o >—
1. 1— <_> o w> o
CM
CM
r—
CM
CM
CM
^
CM
^
CM
(-_
CM
r—
,
CM
Temperature, °C
CO ID
in
CO CM
in
CO CM
in
f^» O
*~ in
O 0
CM «*•
°, 2
5
f** C3
CM
CO -«3-
^
CO >=f
co
•3
o
4->
No. compartments,
No. bags, total
_
co
CM
cn
CO
CM
r-
s
CS1
si
i.
—
5
»—
2
CV3
r—
CO
in
r.
CO
fO
m
&.
o
"o
i
o
flj 1
i; •-
cn *^
itt 01
O -C
73 t.
O
U) <->
Ift »f-
o c
i- O
o s:
_
-
i—
r--.
-
•—
r-
^-
i —
Moisture, percent
^
CM
1
ID
CM
1
^
1"-
CM
1
r~:
^
CM
1
ID
r~
CM
<£>
(>-
CM
•
l-«
CM
1
uo
1
»».
1^
CM
1
"^
P*.
r*.
CM
3
Id
Q.
O
S«
•a
0)
£-
u
in
£
a.
e-n
-------
1
u- vo
coco"
U.U1
- "CM
CO CM
U.*
ear-1"
""*«
<«
r~\
T3 IJliI lO
UJ ^-^
1
•MlJ
CQ
»*-
4->
<-> C
gj 4)
«.|S
it-fl1
*>
u t- m
10 3 *—
CW 3
Sfrff
U.OO:
COCOVOflO-— •— CO CM
•— oo vo r~ en r— S
co m tn co en •
V CM CO
*j- co 10 en i— *- i-» CM
co ed vo oo en *— M
o CM in r— en •
•» CM fl
... 1 « ^4 I"" CO
25S S ~
& C*i fl
eoiD-^r. 53 u, g
r^ o» *£ cri 5- ^ 01
o m •— CM
1
PH. Cs* r- jP*» I** f— O Q »•
e
r*» C5 «" t Is** r-" O O
. . . i . esj r- co
852 S " s
i—
^ (__ ^ ^
A A.) 'MM
O ^W CO CA ^** *""
S ^- *"" ffJ
CO CXi I in r— r— VD O
• * 1 * • CM P" CX|
csi CM en •
co r^> i i I—* »— 10 o
2c^' ' cX 2 ^ g
CM CM en «
CM •— •—
•—
^ 1-5
a in
.fo0 £ «
a) ..- aj *j
to «- *J oT 3 *>
t. aj -^ i- t.
O)1*- 3 (O -
»— •'ffl o > en *- o (0
«-4->eoi0o o> u ja
o a tu u r— Q.
t-t— oatoo ff • •
•^- v o o
< H- z =e
- ^. « o r- eo en ,
CM in 10 ^ es CM
o eo • eo r» CM
. • C9 ^>H ^^
c«) CM in - -
*" CM CM
.. • t S • • '
gin «- CM rv CM
00 • i— O CM
. .. co in en
fr> vO p*
CM CM
_. _ 10 g ,- in «
§in i— CM f- CM
00 • r— O CM
. . ro in en^
CM CM
CO CM r- JO JO CM
^p CO * yy *J "•
en <\! S *- *
§
U)
u
** en
^ i. *. VO » • •
S CO CM IO * ••
? ^- vo r*
j
ra
a
i
t i~ (•• ^ o f1^ co en
^ ^ S *n r~ «NJ
o P» • o ft CM
co CM Jn **•**•'
_ 10 r»
r— i— vo o o CM en
.... o * • •
Soo m P*- CM CM
3 « r« oo CM
. . vo 55 vo
CM CM 00 " "
- - vo o f- f- en
en CM o eo it CM
* "^ en S B
CM CM r- • •
r- in VO
S" CM S 00 ^ CM
p» • CM f» CM
. • en *— o
CM CM r^. * ••
i— in 10
o
01 •!-
•§ U NB
^.S M*-S*"-
O CJ Q) rt3 t» ^*
>-- 1 V 18 JC
V 5 j i ^ .?
5 L f £ o S
|Li.'5 c*>o'u fc
01 *J S • c *> p 'e
(d 4J U.O
CM
,
N.
- ^
CM
r~>
1
I«C
- r-
CM
i
r- r**>
1
p^
(M> (*»,
~
10
^
c^* fv
CM
1
CM
T
UD
^
f— t^
X
^'
"c §
" d
S. 2
*O
i 1 1
-8
•>.;»
cm
K §!§
«' S»
• fW »^« IQ 1 (
+J *J D.
a a
a. a.
4-> *> . .
H« <^> O) Ul
m R} w) w
t. t, OP
77 uE
t/i in
22
U, t-
o
O
«-
•I?
i. t- C > >
Q) Q) Q} (Q ft)
a) o;
SIB
O
U3 V)
si.
»r«
S§!
<- ->j
M- en
"""T Si
> vo i- .
ro a. oj cu *J •••»*•> i o>e
VO S §) &M^ W7 «S *° 3 S 01
m 01 S '*-'*- I- ui *J *3 -w
oiuinu)! i,, i v a t.r-
6-12
-------
1-1
o
i— i
i —
— 4
CO
1]
iifaii
_1
LU
LU .
^H X^S
t— w
-i -p
liimcM
IXX CO CO
CO CM CM
03-— r-
.a re
^™ ^
Furnace type
Capture equipaei
Regulatory altei
in o o *n CM tfl
f»» tft O CM
in
•
O O O O I O !•*• O
§ § § S. ' c3 - S
tn in oo i—
? CM i—
H
OOOOI O O O '
O o o o I m CM <«•
O O O O CM «T
tnino o '"•
CO CM I—
<
O 0 0 1 1 O O . O
o oo i i in CM «•
O^O^O^ CM •*_
in in o <—
oo to CM
CO CM r—
G3 CO O C3 1 C3 r** O
in m co o i m • CM
CM CM O O CM •*•
•• * *» «
yw. r»* C3 C3
o 10 ro i—
00001 o 03 3;
m m o o t m CD
CM CM 0 0 CM to
UJ r- O lO
cn to co
O O O 1 1 O 00 •«£
in in o i i in co
CM CM O CM CO
cn 10 co
"a
•a T) o
<<- o
O jCZ *
ft) V
zn u. +s •—
- c o c re
ro s- *> aT 5 *j
cn*- 3 «
2 >, C 1 *J CL Ul
S»— Q. vi &- o na
M- 4-J C O "3 O Ol U ^
O M UJ O r— Q-
*^- V O O
<: i— 2: z
<— >— CO O O
.... in to
~ ** in*
CM CM "*•
S &. " o"
CM CM "S-
r«~ *•— CO O O
w ri: 2 IS
03 m *
. . o
CM «M «f
f~ t— >f> O 0
. . . . CO C3
CO «» CM (C
CO CM CM
S
J
1
£" _ r- in O O
a co g: as
i — r- If) Q O
.... CO CM
CO f« CM r—
F— O> r
... CO
CO CM CM
1— r- CM O O
CO 03 *~ **L
*~I . CM
CO CM CO
r- •— CM O O
.... m vt
CO C7) p— CM
*O CM CO
r— i— CM O O
CO *
CO CTi r- CM
*~I '""t o"
'£> s- jr IN
ns u 4^
*" 5 S
? J= O 0)
o u s- re
r=- | 0) 01
U O 36 i-
01 0.
4J S- 0) J£
0) 1 *i— U> W +^
Q. S- It) C 5- O
$ '5 M 42 5 "3
U)
O) 4J o • cr 4J
fO CT £« O fO CU
OQ 3: o z u. z
o m
in
o in
m
O IT)
CM C»
*
in
«*•
§ (S
CO*
r~
o in
O f-
U3
CM
•"•
o in
to
S
o in
o r»
op_
^
CO
o in
to r~
in
S
o in
S ^
f.
*
*. **"
H3 i^w
Gross cloth are
Hom'tor height,
r- in
t
CO
.— in
CO
f- in
CO
^- in
i
to
<— in
i
CO
i— in
i
CO
i-* in
t
CO
i— in
i
to
t
to
•4J e:
£Z *^~
Moisture, perce
Pressure drop,
6-13
-------
I
^o
U (. R)
§5-=
1- D. C
3 It) 01
|| LL. o o;
OOOOO 0 CO CM I— i— 10
§0000 in i — co •» ••
oo o o CM in o cn
f— O CM CO •— CO CM
CTI in •— CM
OOOOO O P- CM r- r- <£>
co o o o o in i— f> ....
OOOOO CM t CO CSI
CM i— P
^
U
cu43*» L^
1. O) *r> &- £- O 1
rat- 3 (0 • +J _s-
tiScis^G i y ^ o. *j i
'5 £ z Z 'CO Z 0 =
«s* cn *— co
O CM
ro CO
o o o m *— in
to o o P-- '
P»
d
CO "*- •*•
t, * CU * CU
i- m s- +J u -
cu cu to -a *- o.
S s- o 01 o
O. +J CD "O
. c <-> o 'c •>- a)
i i£ z o ^ S o-
nJ
o
o
s-
• •
o a) a
o o
£. 0 S-
•0 I-I "3T3
CU O ^. CU
U) O C l/l
ot. cu o
"o 13 o"u
•v. cu
C U1 " *
cu o in u)
S-»-" C C
o u p o
C C [•' 4J
o o f- s-
•i— .r* IO W
4J *> O. O.
.p *J J?* 43
i. i. l|- •*-
a. a. s- s-
73 TJ
it- i*- ;tn u>
(0 (O '« W
i. !- 00
T3 -a ti-
ll !U 0
tn in
tn tn •*• *
2O 4^ -fJ
I- U U
U U 33
•v-a
jj'jj'tjj ij_ i.
u u o Icy cu
3 3 o cn cn
•a 13 s. \c. c
'^J ®
CU CU > *, » "
(0 10 "O "O ^1
ooooo
u) ,>) U U 0-—
a. a. o. i—
OOO - " - 1
c. c-a -a TJto
• 10 to o o o
•a u o j|,j2 ^_
5 || .= ..= ,c o
ro O* aj a> 4J.*J +J i
VD C CK O)*»~ •>-" "•"
03 Q) 0) <4~ f- f- U)
0) U W U> t 1 f &
r— cu a a; <—
raooo oS o o td
11 WT LJJ tJil UJ ^~ f— r— r~~
0 il II 11 11 11 II 0
I *"*
11
3 Q) 4—
6-14
-------
In genera], the air flow rates shown in Table 6-5 are higher than
those in use at the time of promulgation of the existing standards of
performance. The increased air flow rates used today are due in part to
shorter heat times (noted earlier) because the emissions are being
generated over a shorter time span, requiring increased air flow for
effective emissions capture. The increased flows have created a dilution
effect on the fabric filter exhaust gas, lowering both emission concen-
tration and opacity.
No emission control systems are required by the process or are used
specifically for product recovery or worker safety (although these are
side benefits of emissions capture). The fabric filter dust collected
from AOD vessels and EAF's in specialty steel shops can be reclaimed to
recover chrome, nickel, and iron. Carbon steel EAF shops use a variety
of methods to recycle or reclaim the fabric filter dusts. Some of the
methods include: (1) recycling the dust for the zinc content, (2) recharging
the pelletized dust into the furnace, and (3) using the dust in the
manufacture of fertilizer. Research is currently being conducted to
explore all the alternatives for recycling or reclaiming the dust to
reduce the expense of disposing of the dust in a landfill.
The EAF fabric filter dust has been classified as a hazardous waste
under the provisions of the Resource Conservation and Recovery Act
(RCRA) because of the lead, cadmium, and hexavalent chromium content.8
The dust from an AOD vessel has not been classified. However, because
the AOD vessel fabric filter dust also contains lead, cadmium, and
hexavalent chromium, it may be assumed that AOD fabric filter dust will
eventually be classified as a hazardous waste. In many cases, the dusts
from the EAF and AOD vessel are collected in the same fabric filter and
handled accordingly.
6.3- REGULATORY ALTERNATIVES
The EPA's decision on how to regulate emissions from EAF's and AOD
vessels is based in part on an examination of various combinations of
capture and control techniques. These combinations provide a basis for
establishing varying levels of emission reduction and, thus, the various
regulatory altneratives. Three such regulatory alternatives have been
6-15
-------
developed for EAF's and ADD vessels and are described below. ;Environmental,
economic, and energy impacts of the alternatives are evaluated in later
chapters.
Table 6-10 presents the regulatory alternatives for carbon and
specialty steel shop EAF's. Table 6-11 provides regulatory alternatives
for ADD vessels in the specialty steel shop facilities. Tables 6-10 and
6-11 also present the fugitive emissions capture technologies upon which
the regulatory alternatives are based. Only the fugitive emissions
capture systems and the percent emission reduction are presented in
Tables 6-10 and 6-11. This is because improvement in the efficiency of
fugitive emissions capture technology has been observed in the industry
since the development of the existing standards of performance.
Regulatory Alternative 1 is termed the baseline and provides for no
additional Federal regulatory action. The control of emissions would
rely on existing local, State, or Federal regulations. The baseline
emission limits for EAF's, dust-handling equipment, and AOD vessels are
outlined in Section 3.3. Each successive alternative beyond the baseline
(i.e., Alternatives 2 and 3) represents an increase in the level of
emissions reduction achieved through the use of additional capture
technologies and increased capture efficiency. All of the regulatory
alternatives are based on the use of a fabric filter as the control
device. " i
The fugitive emissions capture equipment for each alternative was
selected for analysis based on those used at the facilities tested. The
estimated fugitive emission reduction efficiencies of each of these
capture equipment systems were based on review of the literature, observa-
tion of these technologies at the facilities tested, and engineering
judgment. There may be other capture combinations or individual devices
applicable in a particular situation but which are not widely used or
suitable for universal application. For some furnace sizes, for example,
total furnace enclosures (TFE's) can achieve the same emissions reduction
as the equipment shown on Tables 6-10 and 6-11; however, these devices
will be analyzed on a limited basis only because of limited operating
experience with TFE's on larger furnaces.
6-16
-------
TABLE 6-10. REGULATORY ALTERNATIVES—CARBON AND SPECIALTY STEEL EAF
Reg.
alt.
Point of
emissions
Estimated
fugitive
emission
reduction,
percent
Fugitive " ,
emissions capture equipment
Shop roof
75-85
Single canopy hood, open roof
monitor; enclosed dust-handling
equipment.
Shop roof
85-95
Segmented canopy hood, scavenger
duct, cross-draft partitions
(or single canopy hood with
separate tapping and slagging
hoods), closed roof (over
furnace)/open roof monitor else-
where; enclosed dust-hand!ing
equipment.
Shop roof
95-100
Segmented canopy hood, scavenger
duct, cross-draft partitions,
closed roof; enclosed dust-
handling equipment.
Estimate based on literature review, observation of technologies, and
.engineering judgment.
DEC system used for process emissions capture on all alternatives.
Fabric filters used for process and fugitive emissions control on all
alternatives.
6-17
-------
TABLE 6-11. REGULATORY ALTERNATIVES—SPECIALTY STEEL AOD VESSEL
Reg.
alt.
1
2
Point of
emissions
Shop roof
Shop roof
Estimated
fugitive
emission
reduction,
percent
75-85
85-95
!
Fugitive . .
emissions capture equipment
Single canopy hood, open roof
monitor.
Single canopy hood, scavenger
duct, cross-draft partitions,
closed roof (over vessel )/open
roof monitor elsewhere; enclosed
dust-handling equipment.
Shop roof
95-100
Single canopy hood, scavenger
duct, cross-draft partitions,
closed roof; enclosed dust-
handling equipment.
Estimate based on literature review, observation of technologies and
.engineering judgment.
Close-fitting hood used for process emissions capture on all alternatives,
Fabric filters are used to control all captured emissions.
6-18
-------
6.4 REFERENCES FOR CHAPTER 6
]. Memo and attachments from Maxwell, W., Midwest Research Institute,
to EAF files. September 16, 1981. Parameters used in model plant
development.
2. Hess, G.W. Technology Leads the Way as Electric Furnace Steelmaking
Heads for New Heights in the U.S. 33 Metal Producing. 18(7):41-48.
July 1980.
3. Letter from Schwartz, S. M., American Iron and Steel Institute, to
Banker, L. C. , Midwest Research Institute. September 22, 1981.
Information on EAF heat times and air flow rates.
4. Telecon. Terry, W. , Midwest Research Institute, with Sarlitto, R.,
Union Carbide Corporation. September 9, 1981. Information on the
use of EAF's and ADD vessels in the steel industry.
5. Telecon. Terry, W., Midwest Research Institute, with Askins, C.,
Babcock and Wilcox. September 9, 1981. Information about Babcock
and Wilcox EAF/AOD facilities.
6. Telecon. Terry, W., Midwest Research Institute, with Bonaccorsi, A.,
Eastern Stainless Steel Company. September 9, 1981. Information
on the use of EAF's and ADD vessels in the steel industry.
7. Telecon. Terry, W., Midwest Research Institute, with Cooley, C.,
American Iron and Steel Institute. September 14, 1981. Information
on the use of EAF's and AOD vessels in the steel industry.
8. U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Part 260, Subpart D and Appendix VII, Washington, D.C.
Office of the Federal Register. May 19, 1980. pp. 33124 and
33132.
6-19
-------
-------
7. ENVIRONMENTAL IMPACTS
This chapter describes both the beneficial and adverse environmental
impacts for each regulatory alternative described in Chapter 6. The
impacts are either primary impacts, those attributable directly to the
control device (e.g., reduced emission levels), or secondary impacts,
those indirectly attributable to the operation of the control device
(e.g., solid waste disposal).
The chapter contains evaluations of the impacts of the regulatory
alternatives with respect to: (1) air pollution, (2) water pollution,
(3) solid waste disposal, (4) energy use, and (5) other environmental
impacts, such as noise pollution and the irreversible and irretrievable
commitment of resources. These impacts are discussed for each model
plant. To assess the effects of the standards of performance, the
incremental impact of each regulatory alternative beyond the baseline
level is evaluated.
7.1 AIR POLLUTION IMPACT
The most significant environmental impact of the regulatory alternatives
is the incremental reduction in air pollution below the baseline emission
levels. The following analysis discusses the impact of the annual
reduction of particulate matter emissions and the associated effects on
ambient air quality.
7.1.1 Emission Reductions
Annual particulate matter emission levels from the model furnaces
for each of the regulatory alternatives are presented in Table 7-1.
Tnese -emissions include fugitive emissions from the shop roof monitor
and the emissions in the fabric filter exhaust. Table 7-2 shows the
incremental reductions in the annual particulate matter emission levels
7-1
-------
LU
^y»
U_
_l
LU
0
Sg«
3:
o
O f—
Si
r-l LU
LU — 1 U>
< C '
ce o
LU >• -P
i— cc ^
\— o
1— CU
-
X. • '"""N
O O
cn o
^
LU . /-*.
CM m
CM CM
u_
^^
LU
I— •
U3 O
a co in
r™ r—
""'
CM
CM O
CJ l»- 0
CM CO
"^
CO ' '"^
O 0
cn o
^s
I**
CM in
CM CM
*^s
s-*.
Cn W
£ C
O
•P
?^
• »(0 «J^
CU CU O
10 (J (0
in m a.
CO C n)
> s- o
CU t- CO
0 O
tO r— tO
C CO C
S- T3 S-
3 O 3
LL. S: LL.
«tf-O <^-<5t- LOr— CO«^
1.0 CD cn CM in r*** r*~ *^m
O •=*• CM CO r- r— -*~s
CO <*• "^ v->
^~* ptm
^"^> /*"*% /***S ^~~S
coo ^^" men coco
end ijor*> coco oo
cn co •>— ' v^ ^-^
CM CO
,-S ^-1 <•> <— S
com inr*- ysto inc\i
CMCD coco r^.in i£>r^
CM co <* 10 r- co ^
in i — i — i— s^
vo r»-
ino ic, i^\ i*^ /^>
coco cOf— r*-"^ inio
r— CO CM«^ y3f>. OO
(£) r— r— r- ^~S *~J
in 10 i»-'
•a
0)
r—
p— /~*
O O 1—0) CM CO
s- c
•PC -P i — -P -P
p-O r- 0) r- r— .
'
•r~
-P
A3
C
s-
0)
4J
'fl
>>!
5.
m
* \
O)
CU '
S- •
•a .
5
cu
u :
to i
C :
t. '
3 '
CU
TJ,
O '
£
-------
a •
OO UJ CO
00 Q C
g Q _0
1 ^1 \^/
3=
Qi O O)
uj < s:
|_ UJ
< 0£
SO
u.
UJ
1— 00
C_> 111
p
ce uj
•sc z
a. i— i
1
—1 UJ
s- u
^v, 3
U O
03 i— <0
C 0) C
s- -a i.
303
u. s: u.
O CO
r«! r^
r^t* CD
csi ^j-
r— ' r—
•*~s
CD CO
, t
esj csi
cn CM
CM CO
00 00
t t
co in
!"»• CM
CO O
* A
t£J fX.
^^
/•*"N
cn co
• •
CD O
co co
I-- CO
. co cn
^«*,
co in
« • *
CO CO
cn co
CO CD
csTcsT
s-<
x»>
in r-«
00 X*
*5t" CD
in co
c
o
a>
3
•*•>
cu in
i— ^2 0)
ns >
•P C -r- r—
c o -p
eo T- « • -a
S 4J C . C
O
C CO r—
H-) J_ fO
cn cn
o csj
cn CM
CSI <"
r— r—
s^
CO ^
in co
cn csj
csj co
r^- cn
. •
st CO
«*• o
* A
tij r^
^^
S^i
r— f">»
« •
^f* f*^
CO CO
oo r-.
*> #•
co cn
^^
co cn
. .
o r»
CM CO
•si- CO
esTcsj
^~*
^s,
CD ^
«5j- r™
in r™-
m co
CM
T3
(B
O
i— CD cn co
in co co in
O CO r— i—
co «*• ^>
r— • ,—
^^
o r^» CM in
• • • •
cn cn co co
cn csi ^
CM CO
oo co cn r—
• • • «
in CM o co
r™" CO f^ P^
in i— >»-'
** *
CD r^
s^
/^*S X"*x
in i — CM i —
• • • •
r— in r-s r>~
co «*• cn o
cn co r—
0* ft S^
co cn
/*^> /***•
O CO f*«. «^
« • • «
r^ r~>. co cn
CO CM r- I--
O 00 CD CD
CO r- s^
in co
co csi
•a -a
c c
(t} (Q
O i—
r- O
co" r-
CM CO
^-/
«5i- r-
• 4
*& r*
\^
O uO
« *
Csj
<«f LO
r» ,— .
s^**
(^^
co m
• •
t^ <^»
cn i—
r- OsJ
*^«>
/^*\
«*• co
• •
co co
in in
,- ^
co m
CSJ CO
r-™ f— •
^^
co
•o
c
(0
JH.
7-3
-------
below the uncontrolled level for each of the three regulatory:alternatives
and the incremental reductions for Regulatory Alternatives 2 and 3 below
Regulatory Alternative 1.
Under all the regulatory alternatives, EAF's utilize the same
process emission capture system, and the improved emission capture is
the result of improvements in the fugitive emission capture technologies.
Argon-oxygen decarburization vessels utilize the same fugitive emission
capture for all the regulatory alternatives, and, as with EAF's, the
improved emission capture is because of improved fugitive emission
capture technologies. An additional benefit of increased capture and
control efficiencies at various regulatory alternatives is the incremental
reduction of trace metals and fluoride emissions to the atmosphere.
Table 7-3 presents the projected construction of model EAF/AOD
facilities industry wide for each year between 1983 and 1987. This
projected construction would result in an increase in industry wide EAF
capacity from 1983 to 1987 of approximately 5.2 xlO6 Mg (5.73 xlO6 tons).
Carbon steel EAF's account for 4.8 xlO6 Mg (5.29 xlO6 tons) of this
increase, and specialty steel EAF's account for 0.40 xlO6 Mg (0.44 xlO6 tons)
The growth in ADD vessel capacity during this same period, 1983 to 1987,
is approximately 0.40 xlO6 Mg (0.44 xlO6 tons).
Table 7-4 presents the industry wide particulate matter emissions
(from those facilities affected by revised standards of performance) based
on the projected growth shown in Table 7-3. Both annual and 15-year
cumulative emissions are presented for the uncontrolled case and for each
regulatory alternative.
7.1.2 Ambient Air Quality Impact :
The impact of the regulatory alternatives on air pollution is shown
by their impact on ambient air quality. Dispersion modeling was used to
predict the contribution of EAF's and ADD vessels to the ambient particulate
concentration. The dispersion model used and the results obtained are
discussed in the following subsections.
7,1.2.1 Model Description. The model used in this dispersion
analysis was the Industrial Source Complex (ISC) model in the short-term
mode (ISCST).2,3 General data and options of the ISC model used are
presented in Table 7-5. The ISC model requires input data on sources,
7-4
-------
TABLE 7-3. PROJECTED CONSTRUCTION OF MODEL EAF/AOD PLANTS
INDUSTRY WIDE> 1983-1987
Plants
Carbon steel
90. 7-Mg (100-ton) .EAF's
136.1-Mg (150-ton) EAF's
272.2-Mg (300-ton) EAF's
Specialty steel
22. 7-Mg (25-ton) EAF/AOD
90. 7-Mg (100- ton) EAF/AOD
1983
2
1
0
0
1
1984 :
0
1
1
1
0
1985;
3:
1'..
0
0
1
1986
0
1
1
1
0
1987
3
1
0
0
0
5-year
total
8
5
2
2
2
7-5
-------
0£ LLJ
0 >
U.p
||
i— i s:
co r>
CO O
LU » C
>
5- •!-"
fO -P
a> re
to 1
0
P^
CO
cn
CO
CO
en
r~
If)
CO
cn
r—
•si-
§
r—
CO
CO
cn
r_
£
re
c
s-
tt)
r—
Sc
>>
0
"re
0)
OJ
O 0
LO CO
O CM
CM CM
O O
r«- o
O CO
cn co
CO Lf)
0 CD
Lf) O
cn cn
Lf) r-
*. +t
Jf2 !^
o o
r^ o
o co
cn co
CO If)
v*-^
O O
ID O
LO i—
^ *»
r™ * ^™
"
CD 0
CM LO
«d- CD
r— CM
p— r—
CO
Q.
O
to "O
r— r—
OJ i —
QJ O
CO -P
cz
c. o
0 c
J- ZD
re
0
Lf) p- ID O CM CO
^i" LO CO G5 tD f^^
O 00 LO r- O CM
r*^ i^~* vj' f^* c\i csj
^ in csj >», f— • tjQ i^n CTi CO
U3 CT> LO CSJ CO LD
r— d- UD 00 r- r—
CO CO r- r— v^-/,
<^*s /^-*s <'*^*
CM LO CO CO CO CM
LO i— LO "5f LO l"-»
LO Cn CO O r— r-
CO CO r— CM ' — '
v^y ^^
i>~ ^ co irT cn en
co en LO CM co LO
r— .«ef CO 00 r- r—
CO CO r— r— s^«
•
CM LO lfi CO CO CM
LO cn co o i— i—
CO CO <— CM s— '
\^^ v^.^
CO CD LO LO «=j- CO
i— 00 CO O i— CM
CO CO CO LO i — r—
CM CM r— r— "— '
Q.
O
•*•
to
"i
•p
^
•p
e— CM co re
•r"
d)
cx
CO
O O CO CM i — CM OO
COLO cocn cncM LOCO
i— i— cnr^- r— to coco
CM co r- co •=*• «5h <~s
LO oo"
CO CO
oo oo oo oo
s~~\ s~^ /^S /^"s
^co CMI — cocooocn
LOI — ooo cno oo
0 00 i— CM r— ' s-/
CO 00 ^ ^
\
/~S /~N /'^
OO r— Cn CM CO; OOCM
oo co co *i" LO cn co ^
cn oo oo cn «sf «*• "— '
'*"!'"« **^>
CO <*
/^^N /^^ *^™S '^"^l
cj-co CMI — coco cocn
tor- coo cno oo
O CO r— CM r—, *~~>
OO CO s— ' v—> '
s^^
1
I
X**S /^^s /*"*% /^~N
oo i— cn CMoq OOCM
co UD co'jj- Locn co'cj'
cn co co cn •* <* s_*i
CO «3" i
V-'
l
•a
0)
^~
o
•p
CI
O r- CM CO
O
' C
ID
7-6.
-------
TABLE 7-5. GENERAL DATA AND OPTIONS OF ISCST MODEL USED
IN ATMOSPHERIC DISPERSION MODELING ANALYSIS4,5
Data item
Description/option
SOURCE DATA
Pollutant
Particle size
Particle settling
Averaging time
Special considerations
METEOROLOGICAL DATA
Year of data
Geographic locations
• Particulate matter.
EAF:
AOD:
• EAF + AOD:
Setting
0.5-1.0 urn (64%)
5.0-10.0 pro (20%)
10.0-20.0 Mm (4%)
20.0-40.0 Mm (6%)
40.0-50.0 |jm (6%)
0.1-0.5 (jm (15%)
0.5-1.0 |jm (22%)
1.0-2.5 Mm (29%)
2.5-5.0 Mm (10%)
5.0-10.0 Mm (5%)
10.0-20.0 Mm (19%)
0.1-0.5 Mm (49%)
0.5-1.0 Mm (8%)
1.0-2.5 Mm (10%)
2.5-5.0 Mm (7%)
5.0-10.0 Mm (6%)
10.0-20.0 Mm (20%)
Not included (based on insignificant
differences in modeling results).
Highest second-highest 24-hour values.
Annual geometric mean.
Downwash of the plume in the wake of a
nearby building.
1964 (last year of hourly records).
Pittsburgh., Pa. (valley terrain; 6.7%
calm winds; variable wind direction).
Oklahoma City, Ok!a. (uniform terrain;
0.'9% calm winds; predominant wind
direction).
Urban.
(continued)
7-7
-------
TABLE 7-5. (continued)
Data item
Description/option
METEOROLOGICAL DATA (continued)
Special considerations
RECEPTOR DATA
Line sources treated as one or more
buoyant point sources (low wind speeds)
and volume source centered in the cavity
region of the building (moderate to high
wind speeds).4 (ISC option allows
emission rates for each source to be
different for categories of wind speed
and stability classes.)
Locations of maximum (worst-case) ground
level concentrations.
Receptors as close as 100 m to property
boundary or three building heights
downwind, whichever is greater.
Additional polar coordinate receptor rings
at 200 m, 250 m, 300 m, 2,000 m and
20,000 m centered at the source.
Maximum concentration impact; sector only
in final model runs,.at 10° intervals:
Pittsburgh, Pa. (30° to 00°)
Oklahoma City, Okla. (310° to 10°).
Receptors at same elevation as plant
grade.
7-8
-------
meteorology, and receptors. These items are described in the following
subsections.
7.1.2.1.1 Source data. Emissions from EAF's and AOD vessels are
either vented through fabric filters or released directly through roof
monitors. Both the fabric filters and roof monitors can be modeled as
slightly buoyant line sources. A typical configuration of the melt shop
furnace building (with the roof monitor open across the length of the
furnace building)'and fabric filter is shown in Figure 7-1. This
configuration represents a worst-case arrangement for downwind concentration
impacts.
In this modeling effort, particulate emissions were considered to
be gaseous emissions, i.e., as though the particles would not settle
out. As indicated in Table 7-5, approximately 80 percent of the particles
from an EAF, AOD vessels, or EAF plus AOD vessels combination are smaller
than 10 micrometers (urn). Analysis of modeling results, with particle
settling and without particle settling, indicated that differences in
downwind concentration impacts were less than 3 percent. For this
reason, particle settling was not considered in the final computer
modeling runs.
The following data are required by the ISC model for line sources:
1. Emission height (m);
2. Exit dimensions (m);
3. Exit velocity (m/s);
4. Exit temperature (K); and
5. Particulate emission rate (g/s).
Tables 7-6 and.7-7 summarize the source characteristics of each model
furnace and regulatory alternative (as described in Chapter 6) for
fabric filter sources and roof monitor sources, respectively.
Downwash effects are also considered in the ISC model. As indicated
in Tables 7-6 and 7-7, the source emission heights are low (i.e., 22.9
to 38.1 m). £/nfssion gas streams such as those modeled (with low emission
heights, low exit velocities, and temperatures close to ambient) have
small plume^n'se and, thus, low effective emission height. These sources
will be affected by aerodynamic downwash, resulting in maximum concentration
impacts close to the source.
7-9
-------
I
I
J
U-
o
o
\
QQ
LU
CJ
> o
H- O
3
cn
7-10
-------
TABLE 7-6. MODELING DATA FOR FABRIC FILTER SOURCES
Case ,
No.§'b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Furnace
IOC
- A
A
A
B
B
B
c
c
C
D
0
D
A
A
A
B
B
B
E
E
E
F
F
F
A/E
' A/E
A/E
B/F
B/F
B/F
Reg
alt.
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Emission
Fabric
filter monitor
height, ra length, m
22.86
22.86
22.86
22.86
22.86
22.86
30.48
30.48
30.48
30.48
30.48
30.48
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
22.86
15.24
15.24
15.24
15.24
15.24
15.24
22.86
22.86
22.86
22.86
22.86
22.86
15.24
15.24
15.24
22.86
22.86
22.86
15.24
15.24
15.24
22.86
22.86
22.86
15.24
15.24
15.24
38.10
38.10
38.10
width, ro
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
3.05
Fabric
filter
width, m
7.62
7.62
7.62
13.72
13.72
13.72
15.24
15.24
15.24
15.24
15.24
15.24
7.62
7.62
7.62
15.24
15.24
15.24
7.62
7.62
7.62
15.24
15.24
15.24
13.72
13.72
13.72
28.96
28.96
28.96
Emission j
rate, g/s
0.303
0.303
0.318,
0.223
1.210
1.210
1 . 273
1.462
1.462
1.273
1.886
1.886
1.980
0.303
0.303
0.318
1.210
1.210
1.273
0.185
0.185
0.200
0.739
0.739
0.802
0.487
0.487
0.519
1.949
1.949
2.075
Operating time
h/day
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
days/yr
275
275
275
300
300
300
365
365
365
355
355
355
275
275
275
300
300
300
275
275
275
300
300
300
275
275
275
300
300
300
Exit
velo-
city
ra/s
0.98
0.98
1.03
3.91
3.91
4.11
3.15
3.15
3. 55
4.06
4.Q6
4.26
0.98
0.98
.1.03
2.60
2.60
2.74
1.40
1.24-
1.30
3.72
3.32
3.45
2.37
2.22
2.32
3.80
3.55
3.72
aCases 1 through 12 represent individual carbon steel facilities; cases 13 through 30 represent
.specialty steel shops.
The cases can be grouped as follows for modeling:
Group 1: Cases 1-6, 13-15, 19-21, and 25-27.
Group 2: Cases 7-12.
Group 3: Cases 16-18 and 22-24.
Group 4: Cases 28-30.
^Furnace ID's A, B, C, and D are EAF's; furnace ID's E and F are AOD vessels.
The exit temperature for all cases is 394°K..
7-11
-------
UJ
o
Q£
O
CO
o
o
02
O
•*
O
CD
C3
CD
••— in
O g
.
4->
Ul
!£>
1
1
1
O
in
•0
.C
in
O Ol
m -
V»
u
S.
•a
CM
c
•"<
LL.
tn
oo o
en CM CM '
us ro ro
o o o
CM CM CM
ro ro ro
o o o
o o o
ro ro ro
CM CM CM
cr>
o ^r o
— in o
— o o
us ro CD
000
^" *tf*
1 CM CM
in in
^* ^f
i ro ro
i • •
CM CM
cn cn
1 CM CM
CO CO
i ro ro
CM CM
«»• CM CM
r— \fl Lfl
cn •- r—
in in in
ro ro ro
ro ro ro
.— CM ro
LU LU LU.
co to ca
us r— co
a.
ai
co
j..
cn
o
i.
£Z.
in
ro i—
«a- ai
in
in in
a> ai
at >
ro
O
in"
.S t
'o -a
ra c
tfc- ro
i — LU
d) tn
in a.
t-H
c
O 0)
J3 'U 'S~>
J- ro tn
ro c: -^
LJ i- ' fO
= S
r— t- i.
ro ro
3 •-
•a in in
•r— - in
> LU. ' O
>r- <£ S-
*!•« dj in
i. as -a
i
(— • a>
•O to i- c ui •«
•i— ro 3 4) *^- S
tD ^uSs OJ 4-
7-12
-------
7.1.2.1.2 Meteorological data. Meteorological data required by
the ISC model include' hourly values (for an entire year) of:
1. Ambient temperature (K);
2. Wind speed (m/s);
3. Wind direction (nearest 10 degrees); and ;
4. Stability class.
Daily morning and afternoon mixing height data are also required and are
interpolated internally in the ISC model to hourly values.
EAF and ADD vessel facilities were assumed to be located in urban
areas and, therefore, include heat island effects. This option in the
ISC model converts all stable hours to neutral stability. Differences
of less than 10 percent were noted for impacts using the urban mode
versus the rural mode in sample calculations.4
In this study, 1964 climatological data for Pittsburgh, Pennsylvania,
and Oklahoma City, Oklahoma, were used for comparison purposes. Both
data sets are reasonably consistent with meteorological conditions
representing maximum or worst-case impacts. Maximum impacts occur when
the wind direction persists within a narrow angular sector and. moderate
winds are predominant.
Pittsburgh is characterized by roll ing-to-mountainous terrain,
strong-to-moderate wind speeds, some persistent wind directions, neutral
stability, and periods of calms. Oklahoma City, on the other hand, is
located on relatively flat terrain. The climatological conditions are
characterized by very persistent wind directions, moderate-to-high wind
speeds, neutral stability, and few calms.
Climatological data from 1964 were used because these data are
fairly complete on an hour-by-hour basis. These data are considered to
be meteorologically representative although no claim can be made in
terms of climatological normalcy. The ISC model rejects days with
questionable wind directions (which are often associated with light
winds).
As noted in Table 7-5, special considerations in the modeling
effort were made. Specifically, all line sources were initially treated
as nonbuoyant volume sources as recommended by the ICS model user's
guide.2 This resulted in high ground level concentration impacts at
7-13
-------
receptors close to the source. These concentrations occurred under low
wind speed and constant wind directions, particularly when the Pittsburgh,
Pennsylvania, meteorological data are used. The line sources were then
treated as one or more buoyant point sources under low wind speeds and
as a volume source centered in the cavity region of the building under-
moderate- to-high wind speeds. A more detailed description of the model
revisions is given in reference 4.
7.1.2.1.3 Receptor locations. The ISC model calculates concentration
impacts for receptors at specified radial distances from the center of
the source. Receptors were located at distances of 200 m5 250 m, 300 m,
2,000 m, and 20,000 m at 10-degree intervals radially. In addition, the
closest receptor was assumed to be at least 100 m or three building
heights downwind, whichever is greater. For some cases in this modeling
effort, this restriction meant that the first receptor ring had to be
more than 100 m away from the edge of the source.
All receptors were assumed to be at the same elevation as plant
grade. The only terrain effects included in the modeling were those
implicitly contained in the meteorological data (causing differences in
patterns of stability, wind speed, and wind direction for Pittsburgh,
Pennsylvania, and Oklahoma City, Oklahoma).
7.1.2.2 Discussion of Dispersion Calculations. Hourly concentration
impacts from each case are calculated by the ISC model. Concentrations
are summed for each receptor and midnight-to-midnight- averages are
determined for each 24-hour period. Annual arithmetic mean concentrations
are then calculated by the model for each receptor and converted to
annual geometric mean concentration using the relationship between the
arithmetic mean (m), the geometric mean (m )', the standard geometric
deviation (s ), and by assuming the concentrations are lognormally
distributed. This relationship is:
_ _ m
mg exp (0.5 In* s_)
7-14
-------
Standard geometric deviations for particulate matter range from about
1.5 to 2.2. Thus, the annual geometric mean concentration is in the
range of 75 to 95 percent of the annual arithmetic mean concentration.
Concentration estimates calculated by the ISC model are generally
within a factor of 2 of measured ambient concentrations.
The modeled particulate matter concentration impacts can be compared
to the National Ambient Air Quality Standards (NAAQS):
Averaging time
24-hour maximum (not to
be exceeded more than
once per year)
Annual geometric mean
Standard
type
Primary
Secondary
Primary
Secondary
Particulate
concentration (pg/m3)
260
150
75
60
7.1.2.3 Twenty-Four Hour Maximum Concentration Impacts. Maximum
24-hour (highest second-highest) particulate matter concentration impacts
are presented in Table 7-8 for Oklahoma City and Pittsburgh for each of
the regulatory alternatives for fabric filter sources. The maximum
highest second-highest concentration is indicated for each case; for all
cases, the maximum highest second-highest concentration impact occurred
at the closest receptor to the source. As indicated in Table 7-8,
concentration impacts for Pittsburgh and Oklahoma City ape similar. In
all cases, the Oklahoma City concentration impacts are slightly higher.
The maximum 24-hour concentration impact occurred for cases 28
through 30. These cases showed impacts of approximately 60 pg/m3; this
is well below the NAAQS. There is no significant difference between
regulatory alternatives for a specific furnace type.
Similarly, maximum 24-hour (highest second-highest) particulate
matter concentration impacts are presented in Table 7-9 for Oklahoma
City and Pittsburgh for each of the cases for roof monitor sources. The
maximum highest second-highest concentration is indicated for each case;
for all cases the maximum highest second-highest concentration impact
occurred at the closest receptor to the source.
As indicated in Table 7-7, the roof monitor emission rate for Regulatory
Alternative 3 is zero. There is a significant reduction in concentration
7-15
-------
TABLE 7-8. HIGHEST SECOND-HIGHEST 24-HOUR PARTICIPATE MATTER
CONCENTRATION IMPACTS FROM FABRIC FILTER SOURCES4
Case •
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16d
1?H
18d
19
20
21
22^
23?,
?4d
25
26
27
28
29
30
Furnace
A
A
A
B
B
B
C
C
C
D
D
D
A
A
A
B
B
B
E
E
E
F
F
F
A/E
A/E
A/E
B/F
B/F
B/F
Regulatory
alternative
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Distance
from source
center to
receptor (m)
165
165
165
165
165
165
204
204
204
204
204
204
165
165
165
165
165
165
165
165
165
165
165
165
165
165
165
209
209
209
Concentration,
Pittsburgh
14.0
14.0
14.7, 10. 3C
56.1
56.1
59.0
36.2
36.2
40.8
46.6
46.6
49.0
14.0
14.0
14.7
53.3
53.3
56.0
8.6
8.6
9.3
32.5
32.5
35.3
22.6
22.6
24.0
53.4
53.4
56.8
1'mpact (ug/m3)
Oklahoma City
14.5
14.5
15.2, 10. 7C
57.8
:57.8
:60.8
39.3
39.3
44.3
50.7
'50.7
53.2
,14.5
14.5
15.2
54.9
54.9
57.8
8.8
8.8
9.6
33.5
33.5
36.4
23.3
23.3
24.8
61.0
61.0
64.9
aCases 1 through 12 represent individual carbon steel facilities; cases 13
.through 30 represent specialty steel shops.
DFurnace ID's A, B, C, and D are EAF's; furnace ID's E and F are AOD
vessels.
Based on different air flow/rates and thus different emission rates (see
.Table 7-6).
Extrapolated from data in References 4 and 5.
7-16
-------
impacts for each regulatory alternative of a given furnace type. Also
indicated in Table 7-9 are concentration effects with and without the
buoyancy correction (as described in Section 7.1.2.1.2). For the cases
modeled with buoyancy, the Pittsburgh impacts are in all cases lower
than the same cases modeled without buoyancy at a 300-m distance. The
Oklahoma City impacts modeled with buoyancy are all approximately the
same as the cases modeled without buoyancy at a 100-m distance.
The maximum concentration impacts occurred for case 37. Combination
of case 7 and case 37 indicates a 24-hour concentration of approximately
210 ug/m3. This is below the primary NAAQS. In addition, all highest
second-highest concentration impacts decreased monotonically for downwind
distances of 250 m, 300 m, 2,000 m, and 20,000 m for all sources.4
7.1.2.4 Annual Geometric Mean Concentration Impacts. Maximum
annual arithmetic mean concentrations are presented in Table 7-10 for
Oklahoma City and Pittsburgh for each of the regulatory alternatives for
fabric filter sources. As indicated in Section 7.1.2.2, the annual
arithmetic mean concentration is slightly higher than the annual geometric
mean concentration. In addition, assumed background levels (e.g., 30 to
50 ug/m3) are not included in the values reported in Table 7-10. As in
the 24-hour concentration results, the concentration impacts are slightly
greater in Oklahoma City than in Pittsburgh.
The maximum annual concentration impacts are 14.7 ug/m3 to 15.6 ug/m3
for cases 28 through 30 for the three regulatory alternatives. These
are well below the primary and secondary NAAQS's. In all cases, the
impacts for Oklahoma City impacts are somewhat higher than those for
Pittsburgh.
Similarly, maximum annual arithmetic mean concentrations are presented
in Table 7-11 for Oklahoma City and Pittsburgh for each regulatory
alternative for roof monitor sources. As in the 24-hour concentration
results, there is a significant reduction in concentration impacts for
each regulatory alternative for a specific furnace type.
In all cases, the annual concentration impacts are less than the
primary NAAQS if a background value of 30 ug/m3 were used.
7-17
-------
'•
UJ
*^»
UJ CO
*-~ 1 t 1
PARTICULAl
[TOR SOURC!
02 Z
=30
1 f *
«3-o
CM O
02
COS
UJ O
re 02
rn 1 1
IM^P Uw
t_4
Q O
Z
c
ro
o
_j
.0
_C
.
5
w^
u
c
re
>,
o
3
•™
23
O
E
O
O
CO
"re
E
0
O
re
>i.
u'
C
re
Q*
3
ja
jf-
•r-
re
ja
+j
3
O
jr
•P
S
g:
O
O
CO
"re
0
o
re
di-P
O> r—
02 re
QJ
re
CO
S- 0
t 1
GJTO
in •
re o
0 Z
*—
LO
I—
ro
i — i
WO CO
•Si- CSJ O
CM r—
O O
•^ r-^ o
co i— o
CM r—
u
E
LO
. ^
re
cn o
CO CM O
co jo
LO CSJ 0
CM r-
o o
«4- CM O
<- CM
t— CM CO
a)
S- 10 J2
in
CO Q) O
"* "~ ^
JZ, m >r- i-
a» co -p ro
in > in
ro a) 01
u a -P
o a> o •
. •- ,
r— u_ re -p
•r- Q.T-
u -a E o
M— re ro
! r— UJ O5 O
: O) S- JZ
O) in 3 ro
+J _ JD r-
1/5 rH +J 0
c +J
O £l) »r« *» *
X3 O 0. >, >,
j_ re r— u
re c t- e
u s- . re re
3 C r- >,
r— 4- . O •!- O
ro s- •<- E 3
3 • - o -p -i- ja
•a in -P o oo
•r- - CL. O) -P
1 > U_ 0) S- 3
•r- >-c:
C S- (J U -P
• r- , re 5
+j re -P a >>
c c o -a
a> a s- ro 3 a) >>ja r—
CU T3 -P O QJ
S- C C 3 -P T3
. a. ro -P in
CM O (J T- -r- +J
•* j= - S- 3: 3: u
in co 3 ro
x: o -a t3 Q.
1 O)r— - in O) CL) E
3 Ol >r-( 0
i — ^J
ro Q. 3 «i— S! -P in
o 01 u_ a z ro ro
ro .a u -a
-------
TABLE 7-10. MAXIMUM ANNUAL ARITHMETIC MEAN PARTICULATE MATTER
CONCENTRATION IMPACTS FROM FABRIC FILTER SOURCES
Concentration, impact
Case
No. •
1
2
3
4
5
6
7
8
9
10
n
12
13
14
15
16
17
18
19
20 •
21
22
23
24
25
26
27
28
29
30
u
Furnace ID
A
A
A
B
B
B
C
C
C
D
D
0
A
A
A
B
B
B
E
E
E
F
F
F
A/E
A/E
A/E
B/F
B/F
B/F
Regulatory
alternative
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
. 3
1
2
3
excluding
Pittsburgh
2.0
2.0
2.1, 1.5
8.1
8.1
8.6
6.0
6.0
6.8
7.7
7.7
8.1
2.0
2.0
2.1
7.7
7.7
8.1
1.2
1.2
1.3
4.7
4.7
5.1
3.3
3.3
3.5
8.0
8.0
8.5
background, \ig/m3
Oklahoma City
3.2
3.2
3.3, 2.3
12.7
12.7
13.4
11.0
11.0
12.5
14.2
14.2
14.9
3.2
3.2
3.3
12.1
12.1
12.7
1.9
1.9
2.1
7.4
7.4
8.0
5.1
5.1
5.5
14.7
14.7
15.6
Cases 1 through 12 represent individual carbon steel shops; cases
,13 through 30 represent specialty steel shops.
Furnace ID's A, B, C, and D are EAF's; furnace ID's E and F are AOD
vessels.
7-19
-------
TABLE 7-11. MAXIMUM ANNUAL ARITHMETIC MEAN PARTICULATE MATTER
CONCENTRATION IMPACTS FROM ROOF MONITOR SOURCES
Concentration impact
Case
No.
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
u
Furnace ID
A
A
A
B
B
B
C
C
C
0
D
D
A/E
A/E
A/E
B/F
B/F
B/F
Regul atory
alternative
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
excluding
Pittsburgh
3.1
1.6
0
NMd
NM
0
NM
NM
0
16.5
9.9
0
NM
NM
0
9.2
7.3
0
background, M9/m
Oklahoma City
5.4
2.7
0
NM
NM
0
NM
NM
0
33,7
17.8
0
NM
NM
0
20.5
12.4
0
aCases 31 through 42 represent individual carbon steel
facilities; cases 43 through 48 represent specialty steel
.shops.
Furnace ID's A, B, C, and D are EAF's; furnace ID's E .and F
are AOD vessels.
.With buoyancy correction.
NM = Not modeled with buoyancy correction; however, modeled
without buoyancy correction, the annual concentration impacts
were well below the NAAQS even with an assumed background
concentration of 50 jjg/m3.
7-20
-------
7.2 WATER POLLUTION IMPACT
. The emission control device most likely to be used in the future is
the fabric filter rather than electrostatic precipitators or wet scrubbers.
It is.not likely that water sprays will be used to cool the exhaust gases
since ambient air is usually used to cool the gases. A heat exchanger could
be used to cool the exhausts from direct fume extraction, such as the DEC on
the EAF and the close-fitting hood on the AOD vessel.
Water-cooled ductwork for direct fume extraction and (e.g., EAF-DEC,
AOD-close fitting hood) water-cooled furnace walls and roofs is expected to
see greater use in the future, but this water use will not present a water
pollution problem because the water is in a closed-loop cooling system and
will not become contaminated.
7.3 SOLID WASTE DISPOSAL IMPACT
The quantities of solid waste (i.e., dust captured by the fabric filter)
generated per year for each model plant are-presented in Table 7-12. The
increase in the amount of EAF fabric filter.,.dust that will be captured by
improved fugitive emissions capture will be small compared to the amount of
dust already captured by present (baseline) emission capture systems.
Furthermore, these solid wastes are a small portion of total steel mill
solid wastes (e.g., slag).
Table 7-13 presents a summary of the quantities of solid waste generated
Nationwide from fabric filters controlling emissions from EAF's, For each
regulatory alternative, the quantities of solid waste generated from
emission control equipment were calculated for 1 to 5 years after proposal
of the revised standards of performance. A summary of nationwide solid
waste generated for AOD vessels is presented in Table 7-14. The incremental
increase in the fabric filter dust generated from AOD vessel operation was
also calculated for each regulatory alternative for 1 to 5 years after the
proposal of the revised standards of performance.
The chemical analyses of the exhaust gases from EAF and AOD
facilities were mentioned in Section 3.3.1. The dust collected from the
EAF fabric filter has been classified as hazardous waste due to the presence
of lead, cadmium, and hexavalent chromium.2,3 The dust collected from AOD
vessels has not been classified, but since it also contains lead, cadmium,
7-21
-------
0
"Z.
5
o.
LU
a
o
02
0
U-
yf*
0
V- 1
a:
LU
LU
:n^
O tfl
h- C
S 0
CJ 4J
OS
LU i-
t-J^>
i— i cn
U— SsZ
O
»— 1
CO
O
CO LO
O r— r—
^^
CM
CM O
r- o
O CM CO
v~>
^,
C> O
cn o
•TV r^-
r^
CM in
CM CM
>,
r"~ • • •f""
CO CO U
u) a re
u> ra a.
CO C (O
> s- u
^^•N 3 ^^1
CO 4- 0) V)
u u cn c
(O i— (O S O
C CO C -M
s- -a s- ^
303
LL.S:LL.
o o
0 0
CO 0
ID 00
I— r— •
^~f
o o
CM un
•5*- CM
r>» <~
CO ^f"
o o
O 0
00 CO
•sj-'cn" !
co p**
^"S
o o
0 O
o o
ID I*~
cn LO
Lf) tD
S^.^
0 0
o o
co o
co o
ID CO
' '
0 O
CM LT>
•sj- CM
« #*
r*—> i
CO «=t-
r_
fO • •
3 C
c: o
c. •<-
(0 •*-> >~^
U Ul
CO 3 0> C
u -a s o
mo -P
C S- - — '
s- a.
H7
LJ_
o UD cn cr»
r*"*» r*^ CD csj
r~ o cn CM
CM <*• CM «3-
^c -c
ID ID CO '""'
CM CM to ID
cn CM cn CM
CM CO CM CO
co co r-- cn
co in «*• co
r--. CM «tf- o
CO O •* r—
ID r~- ID r-»
*^*^ *^*^
/~\ /*^l
cn ID r- r-
ID O "Sj- I1*"-
co co co co
p^ tjr> co r*-*
co o*i co cn
"^ ^
ID Ln co cn
co co o r-
cn co CM .>
0 4-J
fO C
r— i- .— CM
3 CO
O) 4^
CO i—
cz: to
r— CD
to co
o co
CO "2j"
^C
o r-~
cn cn
cn CM
CM CO
CO CO
ID CM
r- CO
LT> r—
ID r--.
*^~s
^^
LO r—
r^- in
co "id-
en CO
CO cn
'""'
O CO
<3- cn
CM ,
u
CO
•r-
o
<<-
4-
O)
OJ
S-
3
a.
«
u
s.
^
•*-•
0
'i.
X5
(0
**"
c:
a)
u
s-
0)
a.
cn
•
cn
cn
O)
c:
*E
3
01
in
^C
(O
7-22
-------
LU
J—
CO
•
a; z
1— 0
oo ca
rs oc
a >
i -x.
_i oo in
-t
H- LL. i.
2 "^
M 05
u_ oi s:
o ca
>- U_
as
s o
ZD U_
oo
o
. t— 1
co i—
r— ^C
i ce
[••"• LU
•—p
LU LU
_l O
ca
^^
I—
i.
CU T-
1 fO
in r—
""t
ss
3
O
00
cn
to
00
cn
r—
t_n
00
cn
f^
oS
cn
f"~
ro
00
cn
r—
V5
C^
•r"~
•P
•r- 03
r- OJ
•r- N
U r— «r-
fO <1) in
4- "O
o a>
0 fO
.e c
. -P 5-
O •!— 3
••y ^* tr j
/"^S
oo csj in o o
O 0
cn tn
M i^
i— CM
O CM
CM CM
t*^s
co o i— o o
cn oo
tn cn
co j
O r— r- O O
•* O
CM 00
tn UD
1 C3
/^^
CO 0 r- CD 0
cn oo
LO cn
*s «S
CO «!i-
Xw/
O r- r- O O
*f O
CM 00
** A
UO iJD
r— r— ,
x*™^
CSJ CD r— O O
r- UD
CM ro
r-^CsT
r— r—
x-*^
r™~
.
•P
r~™-
CQ O C3
O O
O O
r- O
*i *>
«^- tn
O CM
CM CM
S»*/
0 O
•sj- m
co"m
o o
•— cn
*>j" cn
.LrTuD*
• ' O
_^^
0 Q
p>, f«H
M ^
ro in
^^^ /
o o1
r- CD
**J* CTl
«n *,
LO <£)
*™ ^3
y^™^
O O
«* 0
co in
I-^CM"
s^^
CVJ
»
4«>
r~-
CM CsJ
S»^
O O
1^5 P>-
r— r—
O.O
r^» cn
r— r—
/^>
0 O
m r>-
r— • r— •
s^^#
O O
r-> cn
^_* r—
•hn^*
1**"^
C3 C3
CO 5j-
^'^
O
•P
CU
3
"O
to
O)
•p
ca
-o
•r"
r™*
O
in CM
c. .
•1— .p
p—
i~—
0) o o o
-a r— o to
O '—'CO i —
g S^/ s^^y
O)
4- S O5 CD
O I £ 21
r~> i i
C . CM UD
o o r~- co
•r- cn CM f—
-p
10 II || ||
a
re ca o in
a.
X
LU
fO
7-23
-------
•
LU
O. =3
(— 1 CZ
i -~
— 1 «C->
O r-l
CO
LU LU
O UJ
1—4 (— -
3: to
S£
co j
—1 I CO
>
p^ 1 t .t ^s^
U. O 3C
>- OQ
Ol T-
i ca
in i —
u
00
cn
CD
CO
ir>
00
cn
r—
00
cn
CO
CO
cn
ca
o
"^X, «p»
CO 4->
CM CM O O O O
r- *3-
O) Ol
Q± OS
O O
CO f—
co «^-
"^
0 0
cn cn
5 •*
^~^
cn cn
r— LD
o
-P
Ol
O)
+J
fO
s
•r—
"o
CO CM
C .
Ol
LTJ CO
CO CO
0 O
0 CO
CO CO
«sf CM
cn co
P*- r—
CO «^-
m r-.
o oo
00 00
^^
•=*• CM
cn co
P- r—
co ^f
co
•£
^
C31
Ol
C£
0 O
P-- CO
"^^S
0 0
oo en"
r— r—
CO r—
CO CD
v_^
oo cn
r— r—
CO i—
oo cn
o
-P
Ol
Ol
I/)
n3
^
-a
• r-
"o •
CO CO
c .
•1— -p
Ol
ra o -p
_^; j^ |
O 1 O
Lf> O
C CM IT
•IT- V^"^>
1— O> O)
o c c
O 0
c.
0 II II
4J UJ U_
c: ^x^.
re!
-------
and hexavalent chromium, It is assumed that it will also be a hazardous
waste.4 Furthermore, dusts from both EAF and ADD vessels are, in many
cases, controlled by the same control device. Under the provisions of
the Resource Conservation and Recovery Act (RCRA), hazardous waste must
be either collected, separated, and recovered for reuse, or the
nonrecoverable residues must be disposed of in an environmentally safe
manner.3
The hazardous solid waste collected by the fabric filter can either
be transferred to a landfill or recycled to the furnace. Currently, the
use of landfills is predominant, but some recycling is performed and
research is underway to enumerate the options available for recycling.5
One plant directly injects the dust into the EAF while another forms
dust briquettes, which are charged back into the furnace. Experimental
reclamation of zinc from EAF fabric filter dust has been successfully
performed for steel mills that use galvanized scrap. The zinc concentration
has to be increased to 50 to 60 percent by weight before zinc smelters
can use the zinc.5
Several specialty steel companies in the northeastern United States
have their dust recycled at an offsite reclamation plant. The dust is
reclaimed for the nickel (Ni), chromium (Cr), and iron (Fe) content, and
the resulting product contains 12 percent Ni, 17 percent Cr, and 71 percent
Fe.6 The reclamation plant sells the reclaimed metals back to the steel
companies at a percentage of the current market price of the metals.
Recycling would reduce the solid waste disposal impact related to
the increased emission capture effectiveness of the various regulatory
alternatives. Furthermore, it would prevent permanent loss of expensive
alloys and reduce the raw material usage.
7.4 ENERGY IMPACT
Fabric filters are used and will continue to be used to meet the
emission limits for both the baseline and regulatory alternatives. The
impact of incremental increase in energy use associated with improved
fugitive emissions capture is minimal (see Table 7-15). This incremental
use is small because the regulatory alternatives (see Chapter 6) call
7-25
-------
1 1 1
t-H
a^
_J
O
oi
LU
LU
_J
f ^
1— i
C£
| —
CJ
LU
LU
LU
O
s
>_
/•y*
1 —
co
rD
z
1™ H
LO
1
LU
f¥\
^f
1—
«
CU
i- •!-
03 4->
CD (O
f^
LO £
O
f*N^
CO
cn
r—
CO
cn
Lf)
CO
cn
^J*
CO
cn
co
CO
cn
CU
£•• •!""
0 -P
4-> fO
re e
i— S-
3 QJ
CD r—
ac re
0
0
o
M
co
LO
«t
CM
LO
CO
CD
O
o
A
r—
LO
**"
O
CO
o
o
0
c^T
en
CD
co"
0
o
o
f,
f—
LO
<^
A
O
CO
0
0
o
M
1—
LO
•^
o"
CO
0
o
o
^r
cn
CD
co"
r—
o
o
o
ft
fm-.
LO
f-J-
0
CO
CD
o
0
*l
^j*
cn
co
co"
r~
CD
o
0
**
LO
CO
CO
CM
CM
0
0
o
*»
CO
cn
CO
A
CO
co
CO
o
o
o
«*
r~"
f^**
CO
o
CO
o
0
0
cn
CO
co^
o
CM
o
CD
CD
•*
r—
f^^
co
o
CO
CD
o
o
*»
cn
CO
co^
o
CM
CD
o
o
•»
CO
r~.
CD
•=d"
CM
CO
o o
o o
o o
o
o
o
o o «$•
LO LO CO
r- r- co
LO
LO
CO
a
o
o
< **
cn
LO
o
o
o
r-
o
co
CO
o
o
CD
cn
•st-
LO
o
o
o
f,
r~*
o
co'
o o
o o
o o
CO
CO
cn
LO
O O
o o
o o
o
co
co
UD
CD CD
O O
O O
cn
LO
US
o o
o . o
o o
r-.
o
co
t-O
CO
cu
cu
u
ai
a.
co
CM
ro
I
c
i-
03
CU
E
O
re
7-26
-------
for improved use of baseline equipment with partially or totally closed
furnace shop roofs, cross-draft partitions, and scavenger duct systems.
An experimental system that uses a heat exchanger to recycle AOD
vessel heat is in operation on the offgas duct of a close-fitting hood
of an'AOD vessel.7 This system is used to heat the shop during the
winter months and represents a means to reduce overall energy consumption.
Also, a European steelmaker recycles the heat from EAF water-cooled side
panels for heating purposes. The energy recovered can also be used for
low pressure steam, preheating boiler water, desalinating brackish sea
water, and various other purposes.8
7.5 OTHER ENVIRONMENTAL IMPACTS
No increase in noise level over those existing under baseline
control is expected with any of the regulatory alternatives. Somewhat .
larger fans (located outside the melt shop) might be needed to comply
with the emission levels outlined in some of the regulatory alternatives,
but the increased noise level is still low compared to the EAF shop
noise levels. On the other hand, if total or partial furnace enclosures
are used to improve emission capture, the noise level in the furnace
shop will be reduced.
7.6 OTHER ENVIRONMENTAL CONCERNS
7.6.1 Long-Term Gains/Losses
. Increased emission control of the air pollutants resulting from the
operation of the EAF's and AOD vessels would result in improved air
quality. The long-term gains achieved by the regulatory alternatives
result from reducing particulate matter, trace metals, inhalable
particulates, and fluoride emissions to the ambient air. There would be
a minimal increase in the amount of energy used. -More solid waste will--
be generated as a result of improved fugitive emissions capture. Improved
capture would improve shop internal environment as well as the ambient
air. If the larger quantities of solid waste collected were recycled,
considerable quantities of raw materials would be saved, thereby decreasing
the cost of the final product.
7-27
-------
7.6.2 Irreversible and Irretrievable Commitment of Resources
The increased quantities of fabric filter dust generated due to the
incremental improvement in fugitive emissions capture required by the
regulatory alternatives will increase the amount of solid waste generated.
The s61id waste disposal methods employed for the EAF and AOD vessel
fabric filter dust that is not recycled may lead to an irreversible
commitment of land. A majority of the discarded fabric filter dust is
landfilled, and a lesser amount is stockpiled for future consideration
of recycling.
The RCRA requires that the hazardous waste be delivered to a landfill
that conforms with proper management practices. The principal objective
of RCRA is to prevent contamination of ground water by wastes that could
leak out of a nonhazardous waste la'ndfill if the liner failed. ;The
!
impact of the requirements of RCRA will be to stimulate interest in
recycling the fabric filter dust to avoid the elevated disposal costs
and to recover the expensive alloys that are added to the EAF and AOD
vessel.
Use of the air pollution capture and control devices to achieve
increased emission reduction would result in minor increased energy
demands and represents a negligible irretrievable commitment of; energy
resources (i.e., coal, oil, natural gas, or nuclear fuel).
7.6.3 Environmental Impact of Delayed Standards
The impact of delaying the standards will be the continued discharge
of fugitive emissions. The incremental benefits discussed in Section 7.1
would not be achieved as long as the standard is delayed.
7.7 REFERENCES FOR CHAPTER 7
1. Memo from Banker, L., and W. Maxwell, Midwest Research Institute,
to Iversen, R., EPA/ISB. October 6, 1981. Discussion of final
model plant parameters.
2. Bowers, J., J. Bjorkland, and C. Cheney. Industrial Source Complex
(ICS) Dispersion Model User's Guide. Volume I. U.S. Environmental
Protection Agency. Publication No. EPA-450/4-79-030. December 1979.
3. Bowers, J., J. Bjorkland, and C. Cheney. Industrial Source Complex
(ICS) Dispersion Model User's Guide. Volume II. U.S. Environmental
Protection Agency. Publication No. EPA-450/4-79-031. December 1979.
7-28
-------
4. Stoeckenius, T., P. Gutfreund, and R. Morris. Further Dispersion
Modeling Analyses of Particulate Control Regulatory Alternatives
for Baghouse and Roof Monitor Sources in the Steel Industry.
Prepared under EPA Contract.No. 68-02-3582, Task 7. May 21, 1982.
5. Gutfreund, P., and R. Morris. Dispersion Modeling Analyses of
Particulate Control Regulatory Alternatives for Baghouse and Roof
Monitor Sources in the Steel Industry. Prepared under EPA Contract
; No. 68-02-3582, Task 1. December 15, 1981,
6. Letter and attachments from Andolina, A. V., AL Tech Specialty
Steel Corporation to Iversen, R. E., EPA:ISB. August 21, 1980.
Submittal of requested information.
7. U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Part 260, Subpart D and Appendix VII. Washington, D.C.
Office of the Federal Register. May 19, 1980. pp. 33124 and
33132.
8. Emission Test Report: Carpenter Technology Corporation, Reading,
Pennsylvania. PEDCo Environmental, Inc. Cincinnati, Ohio. Contract
No. 68-02-3546, Work Assignment No. 2. July 1981.
9. Telecon. Terry, W., Midwest Research Institute, with Porter, J.,
Lehigh University. October 1, 1981. Information on research
project to determine recycle potential of EAF dust.
10. Telecon. Banker, L., Midwest Research Institute, with Patrick, J. -,
Inmetco Corporation. September 23, 1981. Information on recycling
fabric filter dust from specialty steel EAF shops.
11. Telecon. Terry, W., Midwest Research Institute, with Sarlitto, R.,
Union Carbide Corporation. September 14, 1981. Information on AOD
vessel emission controls.
12. Junker, A. Electric Steelmaking—The Bottom Tapping Combined
Process Furnace (CPF)j Part I—Technical Inovations. Iron and
Steel Engineer. 59(12):25-28. December 1981.
7-29
-------
-------
8v COSTS
This chapter presents the process and control costs for each of the
model plants for new or retrofit and modified or reconstructed facilities.
Emphasis is placed on the incremental control cost impacts of implementing
the various regulatory alternatives presented in Chapter 6. The costs
presented in the following sections provide input for the economic
impact analysis described in Chapter 9. These control costs are dependent
upon the model furnace parameters and the fabric filter and capture
device parameters presented in Chapter 6. Any changes in those parameters,
specifically in flue gas flow rates, may significantly impact the control
costs.
8.1 COST ANALYSIS OF REGULATORY ALTERNATIVES
The capital and annualized costs for the process equipment in the
new and retrofit model plants (Table 6-1, Chapter 6) are presented in
Table 8-1. The model plant developed to, account for modification will
be described in Section 8.1.2. Model Plants 8 and 9 (Table 6-1, Chapter 6)
in the specialty steel segment are subdivided into four model plants (in
Table 8-1) to account for two control system configurations. The costs
presented in Table 8-1 include the costs for the EAF/AOD vessel, melt
shop building, transformer, cranes, ladles, teeming aisle or continuous
caster, and other integral equipment in an EAF/AOD melt shop. Inclusion
of the costs of these equipment items in the plant costs is necessary to
estimate the economic and capital availability impacts of the regulatory
alternatives (see Chapter 9). The plant costs were estimated from
information derived from several sources and do not include the'control
system costs. The annualized costs consist of direct and indirect
operating costs, including the capital recovery and scrap costs. The
8-1
-------
TABLE 8-1. CAPITAL AND ANNUALIZED COSTS FOR UNCONTROLLED MODEL PLANTS1-5
(March 1981 dollars)
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Model plant
New shop: one 22.7-Mg
(25-ton) EAF
New shop: one 90.7-Mg
(100-ton) EAF
New shop: one 272.2-Mg
(300-ton) EAF
New shop: one 136.1-Mg
(150-ton) UHP EAF
One new 272.2-Mg (300-ton)
EAF added to an existing
shop
One new 22.7-Mg (25-ton)
EAF added to an existing
shop
New shop: EAF and AOD vessel ,
22.7-Mg (25-ton) capacity
each; separately controlled
New shop: EAF and AOD vessel ,
22.7-Mg (25-ton) capacity
each; commonly controlled
New shop: EAF and AOD vessel ,
90.7-Mg (100-ton) capacity
each; separately controlled
New shop: EAF and AOD vessel ,
90.7-Mg (100-ton) capacity
each; commonly controlled
Production
Mg/yr
(tons/yr)
37,420
(41,250)
163,300
(180,000)
596,000
(657,000)
434,800
(479,300)
596,000
(657,000)
(Additional)
37,420
(41,250)
(Additional)
37,420
(41,250)
37,420
(41,250)
163,300
(180,000)
163,300
(180,000)
Capital
costs
$10*
7.0
20.0
25.6
33.2
17.0
4.5
14.6
14.6 ,
44.0
44.0
Annual i
costs
$10e
8.9
28.7
89.8
68.8
88.3
8.5
18.8
18.8
68.0
68.0
ged
f*
$/Mgc
238
176
150
158
148
227
502
502
416
416
ulncludes direct and indirect capital costs.
Capital recovery and scrap costs are included.
Costs are per megagram of steel produced.
8-2
-------
capital recovery costs are based on an interest rate of TO percent and
an equipment life of 15 years. The estimated scrap costs are $94.16/Mg
($85.42/ton) for carbon steel shops and $273.72/Mg ($248.32/ton) for
specialty steel shops.6 A steel/scrap yield of 85 percent is assumed.
Table 8-2 presents the cost factors that are used to compute the
control system capital costs. The cost factors for various items were
developed as a function of control equipment cost from the information
provided in the EPA study on the capital and operating costs of air
pollution control systems. Similarly, Table 8-3 presents the bases for
estimating the control system annualized costs.
Table 8-4 presents a summary of capital and annualized costs for
the control systems utilized by the model plants under various regulatory
alternatives. The control system capital costs include the purchase
price of the fabric filter (control device) and the auxiliary equipment
(such as capture devices, fans, ductwork, solid waste handling), and the
direct and indirect installation costs. The control system and the
capture equipment listed in Tables 6-6 through 6-9 of Chapter 6 for
various furnaces under different regulatory alternatives were used to
calculate the total capital costs of the overall control system. The
solid waste handling equipment includes the pneumatic dust conveying
system associated with the control system and the dust storage silo.
The direct installation costs consist of the costs for foundation and
supports, handling and erection, electrical wiring, piping, and painting.
Similarly, the indirect installation costs are compiled by combining the
engineering and supervision costs; construction and field expenses;
construction fee; and startup, performance test, and contingencies •
costs.
Generally, the pollution control system costs range between 15 and
20 percent of the total plant costs at baseline levels. These cost
percentages increase with increasing levels of control.
No secondary water treatment is required as a result of the air
pollution control system becaus-e fabric filters will be the principal
control technology. Therefore, no water treatment costs are included in
the determination of the control costs shown in Table 8-4.
8-3
-------
TABLE 8-2. COMPONENT CAPITAL COST FACTORS FOR FABRIC FILTERS
AS A FUNCTION OF EQUIPMENT COST-NEW FACILITIES7
(March 1981 Dollars)
Cost element
Cost
factor
Direct costs
1. Purchased equipment
Fabric filter A
Auxiliary equipment (ductwork,
canopies, continuous opacity
monitor, dust removal
system, and fan system) B
Instruments and controls - - - 0.15 (A+B)
Taxes 0.03 (A+B)
Freight 0.05 (A+B)
Base price (total of above) = 1.23 (A+B)
2. Direct installation costs
Foundation and supports
Handling and erection
Electrical wiring
Piping
Painting
Total
Indirect costs
3. Indirect installation costs
Engineering and supervision
Construction and field expenses
Construction fee
Start-up
Performance test
Contingencies
Total
Total capital costs (1+2+3)
0.04C
0.75C
0.08C
0.01C
0.02C
0.90C
0.15C
0.30C
0.10C
0.01C
0.01C
0.03C
0.60C
2.50C
8-4
-------
TABLE 8-3. BASIS FOR ESTIMATING ANNUALIZED COSTS-
NEW FACILITIES6-9
(March 1981 Dollars)
Cost element
Cost factor
Direct operating costs
1. Utilities
A. Electricity
2. Operating labor
A. Direct labor
B. 'Supervision
3. Maintenance
A. Labor (hourly rate
of 10% premium over
operating labor)
B. Material
4. Visible emissions observer
5. Flow monitoring
A. Continuous
B. Periodic
6. Replacement Parts
A. Bag replacement every
1.5 yr
B. Filters for opacity monitors
7. Solid waste transfer
$0.0463/kWh
$10.14/h; 6 h/day
15% of 2A
$11.15/h; 2 h/day
100% of 3A
$4,540
$25,000
$13,600
Cost of bags/1.5
$130
$90/ton
Indirect operating costs
8. Overhead
9. Capital charges
A. Administrative
B.
C.
D.
Property tax
Insurance
Capital recovery factor
80% of 2A+2B+-3A
2% of capital cost
1% of capital cost
1% of capital cost
0.1315
aBased on 10 percent interest rate and an equipment life
of 15 years.
8-5
-------
TABLE 8-4. SUMMARY OF CAPITAL AND ANNUALIZED COSTS OF POLLUTION CONTROL
EQUIPMENT—EAF/AOD PROCESS AND FUGITIVE EMISSIONS 10
(March 1981 dollars)
Capital costs0
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Model plant
New shop: one 22.7-Mg
(25-ton) EAF
New shop: one 90.7-Mg
(100-ton) EAF
New shop: one 272.2-Mg
(300-ton) EAF
New shop: one 136.1-Mg
(150-ton) UHP EAF
One new 272.2-Mg (300-ton)
EAF added to an existing
shop
One new 22.7-Mg (25-ton)
EAF added to an existing
shop
New shop: EAF and AOD vessel ,
22.7-Mg (25-ton) capacity
each; separately controlled
New shop: EAF and AOD vessel ,
22.7-Mg (25-ton) capacity
each; commonly controlled
New shop: EAF and AOO vessel ,
90.7-Mg (100-ton) capacity
each; separately controlled
New shop: EAF and AOD vessel ,
90.7-Mg (100-ton) capacity
each; commonly controlled
Production Percentage of
Mg/yr Regulatory total plant
(tons/yr) alternative $106. capital costs
37,420
(41,250)
163,300
(180,000)
596,000
(657,000)
434,800
(479,300)
596,000
(657,000)
(Additional)
37,420
(41,250)
(Additional)
37,420
(41,250)
37,420
(41,250)
163,300
(180,000)
163,300
(180,000)
1
2
3fl
3s
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3.
3e
1
2
3
1
2
3
1
2
3
1
2
3
1.73
1.78
1.81
1.71
4.37
4.52
4.62
4.87
5.17
5.57
6.29
6.48
6.67
5.84
6.20
6.68
2.08
2.14
2.17
2.05
3.59
3.65
3.77
2.84
2.90
2.97
9.28
9.45
9.78
8.45
8.59
8.99
19.8
20.3
20.5
19.7
17.9
18.4
18.8
16.0
16.8
17.9
16.0
16.3
16.7
25.5
26.7
28.2
31.6
32.2
32.5
31.3
19.7
20.0
20.5
16.3
16.6
16.9
17.4
17.7
18.2
16.1
16.3
17.0
Annual! zed costs ,,
c S10-5
500.8
502.7
514.9
482.2
1,411.7
1,433.2
1,453.0
2,230.5
2,282.6
2,441.3
2,316.9
2,349.0
2,404.0
2,307.0
2,360.9
2,525.0
518.0
519.9
532.6
498.7
933.7
955.6
989.0
: 809.4
813.1
835.3
2,505.8
2,545.9
2,624.9
2,371.7
2,389.7
2,479.7
$/Mg
13.38
13.43
13.76
12.87
8.64
8.78
8.90
3.74
3.83
4.10
5.33
5.40
5.53
3.87
4.15
4.44
13.84
13.89
14.23
13.33
24.95
25.54
26.43
21.63
21.73
22.32
15.34
15.59
16.07
14.52
14.63
15.18
^Includes control device, capture devices, ductwork, and solid waste handling.
"includes direct and indirect capital costs.
jTotal plant investment includes the costs of the melt shop and the pollution control equipment.
°Costs are per megagram of steel produced.
Costs are for total furnace enclosure (TFE) option. TFE costs were evaluated only for
22.7-Mg EAF's.
8-6
-------
Regulatory Alternative 1 is the baseline alternative, which corresponds
to the existing standards of performance for EAF's and the typical State
regulations for AOD vessels. Regulatory Alternatives 2 and 3 represent
more stringent levels of emission control. Facilities with EAF's regulated
under "Regulatory Alternative 1 are required to install a continuous
opacity monitor to monitor the opacity of the visible emissions from the
control device and to install flow monitors to monitor continuously the
flows through the capture hoods. Under Regulatory Alternatives 2 and 3,
EAF facilities can monitor the opacity of the visible emissions (VE)
from the control device with a continuous opacity monitor or by visual
observations made by a certified VE reader. These alternatives also
allow periodic inspections of the capture system and once-per-shift
recording of the key operating parameters (e.g., damper positions, fan
amps, furnace static pressure) in lieu of continuous monitoring of the
flows through the hoods. Adoption of the visual observation method and
periodic inspections is very likely and has been assumed in calculating
the capital costs for Regulatory Alternatives 2 and 3. The capital
costs associated with monitoring by a certified VE reader are zero. Thie
capital costs under Regulatory Alternatives 2 and 3 are higher than
capital costs under Regulatory Alternative 1 because of higher costs for
the control and capture systems, which offset the decrease in the costs
related to the use of periodic monitoring.
Tables 8-5 and 8-6 present a detailed cost breakdown of the capital
costs for fabric filter control systems for carbon and specialty steel
facilities, respectively. The estimated purchase costs for air pollution
control equipment agree well with the purchase costs of control systems
bought by the industry in recent years.1-3/11,12 The installation costs
shown on Tables 8-5 and 8-6 may differ from those incurred by the industry
because the nationwide average cost factors that were used in computing
the costs on the table differ from site-specific installation costs.
The annualized costs for the control system include direct operating
costs (e.g., utilities, labor, parts, VE observer, and solid waste
transfer) and indirect operating costs (e.g., plant overhead, capital
charges, and capital recovery costs).10 The solid waste transfer costs
for carbon steel shops are estimated at the rate of $90/ton because the
8-7
-------
co co CM co
CO
LU
O
LU
LU
O
en
o;
- f I
ooo o oooo o
ooo o oooo o
CD CD CO CD CM r— r— O CO
f— in in in i*- CM
O CM «* f» "3" CO
LO CM CO LO
o
o
CO
O O O O
O O O O
CM i— I "3" CO
r- CM
«*• 10
com
in «r
r- r- CO
OOO O OOOO O O
ooo o oooo o o
I I CO CO O t^» LTS r— in CO O CO
co CD in en o ^ co r*~ r** *3*
CMCOCM CD CM «3- f^fc O CM CO
CO i— <• CM CO CO *4*
to
—1 S~
§,2
15-5
o a
C£ CO
LU cn
u_ a
o ra
2:
OOOOO
ifl
co m o o o o
i co r- en co co o
OOOO O
ooaoo
f-ioocvi LO
g g.
ooo
ooo
i— I r- I CO I
« I "I » I
i-OO
r- CO CO
O
o
OOOO O O
oooo o o
i£>r^cor*> csj cn
o o
CM Lf)
•• ,*
co r^
o i—
i— ID
o o
s s
CM CO
co" r-T
O CO
!-^ *
o o
o o
co f>
»s to i— to
CO
o
> o
o o
o o
r» to i 10 m
10
in
ooo o oooo o
ooo o oooo o
i r- r» co co in r--in in cri^
to
10
CM
O
o
ID
U7
CM
r>. co in o
r- p» CM CO
o co O o
oo oo
O VD I IO CO
co co in o
r— f». CM fO
CM «a-o cn co K cn in CM oa
eo|5..= ao o> —CMCM in r-.
CO CO O O O O O
000 O OOOO O O
i it— «3- cn CM r^cocrii— o *•—
OICMCrt CO U3^00i— O —
cor*. j- CO-CM- S in
r~i fo
incs
Jo o o o o
OOO O CO
CO O1 r— f^ LO CM
o" cjT^rcMCMcn CM S co oo •— CM «*
CO CO^ C0 00 ^— CM Cn CM *—
CO OJI^ S> COCOCO
O O
O O
CO P"»
in co
CO r-
•3- 00
o O
C3 O
r~. co
o o
o o
O CM
CO CO
cj o
ce
s s ^
i.
CD
t~ o
o '5
«2 5
SC s- re
Q.-i-' 3 CL
1 3 O J=
^^•GS
£ T3 c: S
*J CU OJ H-
•i- r— 4-i
3: o o> o en
o o 3 c:
•a o ns "a -r-
o
o <-
s- L. -a
3 •
c o +j to
ca LU o u
CJ O I— CO I
S- O)
o c:
•/— £-
C O
>> 2
CL<— o S- ^-.
O ^* LO •," CO
CM ro
3 3 01 >i
O O -P J^ , O 0)
•*• ••- ui 3 CT —
4J 4-2 4J fc- to
c c c u co 40
O O 10 3 E O
O U LL. a LU t—
C
o
•a
c
CO
O
u
E 4-
£«-§,-*,
4J Q>
tn
U)
4->
V)
o
u
•f
CM
+
u
LU
ce
a
_l
-------
^ ^
TJ
CO
3
C
•r"
•P
O
U
i o
1
00
1 1 1
CO
1 —
Gt CM ^ CsJ
CJ CO
>> 41 «1-
o.'a. n
Ol ••— C
3 3 .— t.
ia cr ai oi
c o -o *J
E- O »—
3 41 £ «
U. t-
3 >v
+j C
£L O
10 4->
O 10
"si
o>
41
as
c
4>
S
4)
41
•in
o
CJ
0 OO OO OOO O OOOO 0
o o o • o o ooooooooo
i— CM «3- i en GO i i er> *— co r- in »— inooon
o •— r-* «*• r»*. focnincninincooo*—
en CM in CM »— co »— 10 «*^
CM CM CM
o oo oo ooooooooo
o oo oo ooooooooo
en CM «^ i cr> oo iiencMCMincoco«*incM
r*s r— f^. «cj*r*- cMCM^f-GOiocoineo*}'
en CM «** CM i— co. r— in co
CM" CM CM
o oo oooooooooooo
o mo omooooo oooo o
r- en r-» f»cn«*cMCM«*'ior-»i— CM r*. in
in co i— ID cncMcofOtOO/— u;
en CM •* CM o co •— in^ CM
CM" CM CM"
§00 o o ooo o oooo o
O O OO OOOOOOOOO
CM in *— i oo oo iit"*.^f— ior-".«3*tocoin
in • ^•r*' i — !*•» o co i— »— r— «r o 03 in
00 *— ? CM CO CM CM O
r- CM CM
§o o o o oooo o ooo o
CD CD CD CD CD CD CD CD CD C*l CD CD CD
CM «*•!•» ' r- fi ' ' en co* oo o CM o «*• » r"~ fj r*" UD CM CD 00
i— CM r—
O CD CD CO CD 0 C3 CD 0 O CD O CD CD CD
oo "S*! — t too r^.cniocDio o r-.mrom io
t*^ ^"* co ^~ in CM en ^*»
/— •* i-^ r-"
^^ in
•PC i-
i(_ O 41
•r- Ct.
o -P (-ra €
O T- O C 10
u wv ai t- •»— s- i —
§s- ro co c o
•P3Q.OP O i- ">
3 O -C 'c 4J C mu
O "O "O r— +J >> O 3 O O O
cj o •»- -P e >*^ ^— u . u o
.C TJ C S -r- P -I-
•P 4Jajajs-u5'*--o -a -a -u
c •.- 1— .P ra o c aiu
41 £. 3t 0 41 U O> CLi— O _S- ^-s 10 S E!
"3 ^ >JoLs-!_"a 33 5,"*^ >t c u
CT 4- I..C4I34IC OO +J -i! U < 41 3 C
•O lO»f->— Q-S*— 4) «i— «^-in!SCni — bin O)r- 4->
toejjss-'i-o m>4J4j*i 4Jt-ie4->4iT-io(8
1— in-r-J3 X CO4J10IA CCCO41+J WX41*Jf—
oo to i- 10 3 reujou3 o o m 3 S o c re s- o i—
O J= Q. U. ~1
LU
o:
I— 4
0 i— . CM
o o
O CD
r— co
o •—
O \£3
in" »-*
o o
0 0
f"~ r—
CM in
en • 'in
«*• r-
o o
§ §
CO O
co ' —
r» in
CO i—
in
4->
in
o
CM -P
u
+ 41
i-
*•«•' ' *O
c
1—
l/> C
O 0
O t/> *r-
t— in
u -a • ia
^N. 01 S-
C Ul O
41 O - *P
CL»— in
0 U
•P
•* « in
in tn 3
.1 .1 ^
•P P C
•i— •»— 10
10 >,
'5 '5
C3)
•P 4J . C
U U *i—
33 >,
T3 -O 4)
i. S- C
41 41 . O
O) Ol U
41 41 -P
> > in
IO 10 3
U U • 13
in in 4-
o u
O 13 T3 fc. 4->
20 o «
o o -o £
• SZ £1 41 3
c in 4)
41 >, >, O C
a. a. o.^- a.
o o o u
c: c
- ta ,
0 TJ T3 0 . 41
_c
4-1 P 41 CZ
>, e c >, -r- o
Q. 41 41 O. •!-> U
• O E E O *^*
i— c 3> en c r— s
i n 4i oi 10 • i- .01
— ocob
10 U
•— O CJ CJ LU r— •—
.n LU LU LU u_ .a cnja in
raoooK- 10 c 10 4i
II II II II P 3
S S in S <—
£i — CMCO«d- O't- O U
s- x s. c
U- U. LU LU l-l
IO XI O "O 41 1-
8-9
-------
CO
LLJ
O
UJ
o
LU
CO
z:
i
a
o
fd
t— a
O 00
oi /—
LU
I— -C
-J O
I— I J-
u_ «r
o o
o o
f*- (O t
o o
o o
en co
o o o o
o o o o
tO (O CO O
f^eo »o in
*r CM us CM
o o o o o
§
^ o o o
o 00,0
m CM CM f—
§OOO O
O O C3 O
co *J3 CO !-»
O)
«v/ «r- i. t.
T3 C fO O
4-1 C O CL-*J
S5raex: =
''l^i
5 O C1 §" 3 ^ _=
•c a
o c ra t- o
|_ »_4 >— U. H-
8-10
-------
u- cn
->^ -cn
03 U5
t3
a>
•P
o
u
00
UJ
_J
CQ
n] ^ u T3 OS
5 -,£
§00 0
CD O O
cn >—
O 3 O 'i- C
10 r- T3 +J *J «~
t- *— ' •!— t. £•-
C O CL4-»
S O ro § -i~
ro a x: e
"O *O "P >>*J Q
O *J -»- E
£ JJJ •e O ^ g
C> C 03**-
O S- *J 3 S- 3
_g (y IF-. O Q> O
*J <*- 3 O) 3
>» (0 ( C C C
Q. S CU •*- 0) *F-
e tj o c CM f—
o o o o
o o o o
co co CM cn
tn «-r to to
gggg
co co CM cn
m *•)• t£ <£
__
O O CD O
O 0 0 0
to •-" cn to
r-o cno
•3- «— Cn co
o o o o
O O CD CD
i— o cn co
«*• f— co CM
o o o o
o o o o
to p-. cn cn
•— o cn co
«*• r— CO CSI
*T *—
t.
ai
T3
C
O
•*->
3
4»> •(—
H. ^
OJ O i-
c m «*-
r^- £ *-^
T3 ,
C 4^ fc»i O
to w s- c
jc > o m
tn s cn
-M 4-> J—
3 co 3 E
Q U_ O UJ
O
o
eo
cn
CM
O
o
co
cn
CM
g
CO
p*.
CM*
|
in
cn
g
r*
i
i
CO
CM
cn
o
o
o
g
l~ in OJ
« X 0)
c m t.
o
o
m
cn
tn
CO
o
0
to
m
CO
«f
ro
O
O
CO^
cn*
ro
CO
o
o
cn
PH.
CO
r-
O
O
CO
1
—
0
o
cn
m
CO
'"I
f—
1
cn
CM
o
o
o
CM
O
o
cn
,-*
+j
o
u
•o
n>
tn
1
CL
«
2
g
cn
m
ro
CM
ro
O
o
0
CM
cn
o
CO
CD
O
CO
i
ro
0
o
f—
Cn
to
f—
0
o
r-.
I
r-.
O
0
CM
CM
CD.
r*«
1
CP»
CO
00^
o
§
s
CO
,-*
i
CM
f—1
w
J->
o
u
QJ
O
e
0
•5
'«
th
CM
0
CO
CO
U)
O
o
to
r*»
m
U3
g
(£7
CM
to
g
O
in
CM
CM"
g.
in
m
s
pj
1
CO*
in
^
CM
o
o
cn
CO
cn
CO
g
CM
CM
CO
rO
1
CM
CO
CM
+
•^
to
o
CJ
LLJ
CC
O
o
1—
h-
to
CD
O
f —
O
UJ
1— t
o
o
ro
in
CM
0
O
*^
_
CD^
CM*
O
CD
CO
CM
O
CsT
O
0
p"»
CM
P-
O
o
in
S
0
o
m
r~.
CO
\Q
O
O
cn
cr.
m
CM
o
CD
CO
ro
o
CM
<~
O
CD
CO
CO
i—
C
fo
o
o
p*.
S '
cn
CO*
0
o
o
cn
CO
.m
CO* .
o
o
^1
cn
*?
co"
o
p*-
cn
a
CM
O
g
S
cn
CM
0
o
p-
cn
ro
CO
Csj
g
in
cn
CM
m
o
o
o
U?
o
in
g
ro "
CM
cn
W
*•" s
ro
4-
CM
*
"-'
1—
8
^
O.
CJ
O
^
O
o
4.
Q) O O
VI O O
0 t- s-
(j T3 ^31*^
•*•» (U <1J O
'c tn u> o
0*0 *u ~o
^v 0)
« * C t/)
t/j tn . 1) O
c c a.-—
00 0 (J
'p- *r» in in
4-> •*-> C C
S- t. O O '
CL CL 4-* +•*
+J 4J t- i-
22 CL CL
_-i^ s'S
*t^ S- t- -W 4J
O C31 O) 3 3
Sec -o-o
> t- s-
CL o O 01 O)
o in in c: c
* - - > >
o o o u u
O O O in in
4_ * -
C C C *" J= J=
u o o cu >> >,
CL CL a,
- - - o o o
•a -o -a c c
000*000
O) oi cnjr cu a>
c c c •*-> **
. 4J 4-i 4-> d.
in
OJ •
<"
in T3
u ro
|ro
3 in
«*- a>
CT JD
C fO _
in € r
-r- O
x s-
UJ U- >•
3- OJ **-
8-11
-------
EAF dust is considered hazardous, and disposal at a certified landfill is
required. Tables 8-7 and 8-8 present a detailed cost breakdown of the
annualized costs for fabric filter control systems at carbon and specialty
steel plants, respectively. The annualized costs under Regulatory Alter-
natives 2 and 3 are higher than those costs under Regulatory Alternative 1
even though the costs fo.r periodic monitoring are less than costs for
continuous monitoring.
The total EAF capacity growth projection from 1983 to 1987 is estimated
to be 5.2 xlO6 Mg (5.73 xlO6 tons) (Chapter 7). This constitutes a
4.8 x!06-Mg (5.29 x!06-ton) increase in capacity in carbon steel shops and
a 0.4 x!06-Mg (0.44 x!06-ton) increase in capacity in specialty steel shops.
The estimated distribution of the number of plants in the various size
capacities to satisfy the growth projections are listed in Table 8-9 along
with the control system capital and annualized costs at various regulatory
alternatives. As shown, the carbon steel industry would need to spend up
to $82 million in capital costs to 'install control systems by 198'7 to meet
the regulatory alternatives and up to $29 million in annual operating and
maintenance costs in 1987. The baseline costs would be about $76.2 million
in capital costs and about $27.4 million in annual operating and maintenance
costs.
The specialty steel industry would need to spend up to $24 million in
control equipment capital costs by 1987 and about $6.7 million in annual
operating and maintenance costs in 1987 to meet the projected capacity
growth. The baseline costs would be about $22.6 million in capital costs
and about $6.3 million in annual operating and maintenance costs.
8.1.1 New Facilities
Costs for the air pollution control systems for the new facilities
are based on the model plant and the capture/control system design
parameters presented in Chapter 6. These costs were derived from the
cost factors developed for purchase and installation expenditures.7 The
Chemical Engineering Plant Cost Index was used to adjust costs given in
Reference 7.
Table 8-10 presents the annual costs of incremental (oyer baseline
levels) collection of process and fugitive dust. Total furnace enclosures
are evaluated as a process and fugitive emissions capture system option
8-12
-------
O
r—t
CO
UJ
L-,,1
I—
t— I
i •
t— 4
" H
+J 4J co
O.C C >
>,
a. o
re *J
O 10
*3
OJ
ce
c
»2J
01
V)
o
U
O CM CO
ID CM r-.
^ OOCM
PO
§»
* 00 CM
CO *~
O i— O
o roi- .
r» r». in
N." oesT
o •— o
o co •—
CM r^ to
O U3 CM
0 r— O
O roi—
in vis CM
1C r—
0 J^ 0
cw r- in
S 2™
i/> t-
o >, i. o c
U 4J 0 XJ O
•r- J3 «3 -i-
CO U « »— Irt
t?r <^> r— >r- Ot
1— 1 t- 4-J > U
1— ml *J Out- C
«s -I- 3 9)
a. -i- uj oj o.
«>
•a-
o
o
ro
o
o
ro
a
o
o
in"
CM
1
ro"
1
CO
g
IO_
ro
O
o
in
CM
monitoring
o
U.
in
o if
o in
co i o
O i—
ro •*
CM
O <"•
o «•
o i co
* t *
O1 CD
CM PO
CM
O O r—
o PO i—
o •— •«•
CM ro
CM
O CM
o •»
03 1 CM_
in
r-- CM
*. ' ^i
03 in
ro in
ro lO
i—. i en
^~ S
ro o r^
ro co CD
in
i.
o
u> a
01 gj
*- •&
'o t.
o ro at
•r- Q. t-
in £» O tn
•P J3 c
fa. ID S- rrj
ID <»- O t-
a s. *" *•*
** 0 UJ 0)
C >1- 5. +•>
O 4> U>
U ro •«-
ro m u_ TJ
'a. ^-
at • • o
O£ f CO 1/1
us r-^
I/O
I—
8
OPERATING
i —
o
UJ
(X
t-^
o
z
1—4
•S-
CM
CM
2
CM
CM
•s-
CM
CM
O
0
CO
0
0
o
PO
o
o
o
ro
o
CM
o
o
PO
o"
CM
01
ai
o <
o in in r- o
m r-» r- o —
PO r— I— CM O
CM ID ID r- PO
US «(•
r-"
o m in »± o
en en os ^— en
ro r— f— CO r-
o in in "cf PO
en ^* ^* en ro
in «a-
r—
03 cn en co en
in CM CM r- «s-
o in m en t--
l~ CO CO * r—
m «*•
r-T
<^ f*s r^» oo co
CO r— r— CO «S-
CM •— I— 0 CM
«3- r- r>- in CM
CO i— r— CM CO
CM "S-
^j> r- r- •— ro
r^. ro ro o o
CM <— i— in en
ID co co as «r
PO i — r— CO r-
CM in
ib co co o in
m r*« r** oo PO
in r~ r* PO CM
ro F— •— PO o
CM in
«3* CM CM CM in
CM •— ^- en co
PO r— r— CM O
CM in
-U
VI
o
U
>,
c at i/i
0 > r—
•i- x o oo
l/> *> IB U O
at is *•> ai o
CD ,t- at s-
L. 4-J >, U O
ft) in +J CJ i— LU
e -r- k a « 1^1
u c ai s- *J t-i
•r- D- 3 — _J
~~ S O Ul D-
-------
TJ
O)
•t-
•p
o
o
^^^
•
CO
UJ
_J
l_
O CO tf CO
.
O CM*r CM
•o
CJ CO CO CO
,
O CM CO CM
-0
*- r-
10 X) U 01
& 4-> *> O>
O.C e=£
ex ex tJ
U *3 r— t.
03 cr
c: >
*•> C
g- s
ej 2
O)
ai
(X
i
01
^
(A
c_>
1
in
00
o
0
CD
•Cf*
g
(JD
g
CO
r«.
o
o
cn
cn
CO
|
gj
CO
£
m
0 >, S-
U 4-> 0
in tj 10
1— u> *» a
<£ O> U C
Qi ••- •>
CO CO ^
000
•t «> A
CO CO •*
oo
1— 1— I
coco"
IS £-
i- I. >
O (U t-
ja *>
00
s.
3
tn c
4J
*-
£. 0 cn
^a c
as- CO
0 > 1—
•i— X O CO
o +-> to u o
1) 10 ->>> u a
a u> 4-> c >— LU
j c ai s- *-> t-i
•f- Q. 3 -r- -J
- is o in a. -l O 2;
z:
a <
j < 03 -
C S-
ai ia •
a. Q.
o
tn «f-
§3
•r- +J
-P U
: T- 3
*•> T3
(a t-
_,_ g1
^> Q>
; -f- >
If ^
tja
€C
: i- 0
c -a
a> cij
> JZ J= 01
C V)
0) X >, O
CL Q. O.I—
o o o u
c c
•a u u -a
o o
O T) T3 O
JT at ot j=
>> c c ><
a. ai a> o. o>
. o e s o u
r~ e cn o) e 10
3
r— CJ CJ CJ LU •— >—
JQ UJ ' ' ' UJ U. £i OJD
loaooi— 10 c ia
s" " " " a? •
Or-CMCO«* O"- O
t- fc. X t-
LL. U. LU LU
ro ja u ~a oi
8-14
-------
UJ CO CO CO
o
1 1 1
— j
1— I UJ CM CO M
o '
a* CM
H-
tn
___j g -
O 03
l_- ^- -o
Z: O eo "*"""
o a
o
Qi CO
LU (T)
_J
CH J= «-.-_*_
U_ U <=«x»«'
O (8
CQ
U_
(•£
g Q.CCW
5 ti .§• ^ m
S3 0 3 r- OJ
C £
. U- S-
od 3
i ^
»u «
CO <->
UJ
_J
ea . -g
«*•> »
CM \O CM (JOVO^CO ^~
O r- O CM CM O O P^
o CO*— rococo vo
ro r*-m r— r- m to CM
O «— r— •—
O f— O CM CM P-
O CO '•" CO <*) , ^C
ro r* in *—^-ii CM
i. . .. * | i •
in to CM vo ^o c
o *— ^~
O CMCO OOOO O
O in co lOtn^O CO
C3 LO CO CO P*» «*
o P*» co co m m
' ' o" o" 0*0*^1-* co*
CM *3" CM CM •«>" » rf o1 . '• CM egSgg &
CO
in f—
f-^
^ CM co coooo o ^i* omtncoco
o Lftco cncn^o o •* *G S ^ £ £!•
iotn o >n p^ p1^- in ' p^
" ' * **^ . *~l^,Lt'*^ ^.
0 jgesT. »*o' W •• « co . o inr^r;j^ .0-
, . * •
S c^2 $S 'S 8 ^ ' '••§• ' "888S--v8
CM r^In J-:S-!n§ r-f t co ^rcMCM^-cM
m. . *. * • - - *
1 i/^ . (jO CM VO *uD *J* OO P1
« ' O " ^" t** f*~ *JO CO
CM CO r— r- CM PT>
-,---. CM "«•
1 51 22 , i '.g.§' ., : ..i lisa s
» . • - i »
m 10 CM »o us tn P
<£> r- CM
^ • O rO "JO tO ^3" CM
«. CM 5
«-',-•.'
o
(fl C , '
(- O
1
4
1
•
u2 tn
to t- 4-» J
O >* t- O C CM S-
EJ i*ojao .cms
•r- ja ro T- u> «^- a
O (j fO f*- U> *— S-
7* ,f- r~- . r- Q) ro t- O 4->
i-i t- -»-> > O •r-014-lC1'
H- «••-> crus- cs-t- > •*- a
uj •»->•— +j «^- n o>
o s <: c51 u- c£
UJ
Q i— CM CO <3" lO 8
j re o) t— • ^~
- 0. 4- (A = H-
a c o -r-xotn
ot.ro i/i 4J ro o c
k- o i- cj oj ro -»-> 4> o
t*- 4-> Z C7 i. O) U
U *-i t. *J >> O O
DIA cut— n)ui4JCr-u.
- t. w ^; jz •••" t- -«
W*J OJ UJ -O — CL 3 -r- -J
3i^- s cu re f— eomci. <
rn •*•> O "U ro TT S- C fO I_
» u. -a jc +*
-------
u. ro
CO 1C*
U. CM
>,. «-Cn CM
co m
CO *t
ILU f»
•*••» *co co
ti^O CJ
•o
•eij0^) ro
CM r*« co ro i a *— CM
in" CO'CM" ro^ro ^j r^
gr— cD *O" "^ CD O CO
ro *"• tO to W CD ro
O lO CM CM CM *T CO CO
CM •— ^" *"" r~ ^~
CM
§,— * CD ^" *y O CD P*»
co^- iota «r o vo
o" to CM CM" CM" <*" .m rs." '
CM
CD i— o CO
01 •— **»
CO
t-
o
u>'5
Is
^ £?
U '
u m.
i XUO-C M t 1*
S 5*0^0 C2W£
.*- ja m — jw 'J2 a t.
^ .Ji ^2 *~ .12
2"- l*» CC < Q3
UJ
cc
O ,-! CM ro -V Lf) 1C
en p*. co oo — co
I ^ P^ CO CO O >JD
PM, en en en csj en
CM r*. co co co P^-
^- CM
U3 O CD O CO CM
en co en en in ro
r»» *•- in tn en en
CM P^ CO CO CM CO
r~ *""1 °^.
r- CM
en co "f ^ en »—
»— /•— CO
r-" CM*
§ en o r-»
i r— en ^r <• co r-
' CM* ** CM CM O P*>
CM o m m en CM
«.
r—
-a- o o o H-
C U "- X 0 j O O
O) ) — rt) trt -*-» C '— UJ
In ce uca)fc-4-> »-<
rtj yj ~Q •*- CL 3 ••" — J
•o °-= *J rti ....
oo a: o o
0
•O 4^ W E
5 3 o- o. t?
4_) *J jC JS ^
U (J *J 4-> -C
. "° "° S 2 > t. t- "a
a; m «t5 ai
a. u o en CT J-
o tn w c c s-
4)
>>>»>% o X3 ~a c:
tll£ll £
ro re c ro
• o-OOO m *J
•a T3 T3 C C i— «
ooo*n>ro
o *o "o *^, U)
,Jl .,-.,->, c e m -a o
. 4^ 4j ^j o E e uram
i 4— M— <*- nj 41 OJ i e: ro *o
m - " • CJ o O a ro c: ro j
^ " " " " " " K 5 s T3
or— CM ro ^ in to o-r- Of—
LL. U- UJ U. CO
(O ij3 U "O tt) -i-
8-16
-------
co
LU
rs
i— t
a /->
LU CO
a. en
X r~
LU
"^*
2: ca
• 1 1 -^^/
co co
>- LU
co >
l""l
O - 1—
l-l OS CO
3s o en
Z 1— r-
0 U
Z LU 03
ce s
C3 ***>
LU CO
1— 13
O O
LU 1— 1
p™3 >*v*
o <:
OS >
a.
til
1 It
en z
1 =3
CO
i
ca
r-
•P
03
C
s_
0)
-P
r- CM
03
^k
C
o
03
,_»
Z3
O)
CO i—
(O
o
s
"
•P
in
o
o
r-«
03
• pH
0.
03
O
co
ai
•P
c
0)
-P
f~m
03 Cvi
>^
i->
o
4^
(0
^-*
3
05
CD
<*- w
0 -P
c
• 03
O r-
z a.
w
^_>
c
03
r—
a.
(— '
Q)
•a
o
2:
ib
,_!
r~
in
,
(—
fHH.
CO
I1-"--11
o
•
f^»
CO
CM
CD
CO
O
LO
CO
CO
LU
«C
LU
s-^
C.
O
4J
1
O
0
r— '
0) O5
0^ ^1
P 1
co r*»-
,
c o
o en
J3
S- Q)
03 C
0 0
O
CM
f^»
•
r—
i—
CD
•
r™"
^J"
.
CO
CO
*^t"
,
CM
CO
IT)
r— •
CO
Lf>
u.
LU
c=
o
•P
o
in
r—
O5
2C
1
.
CD
co
f—
0)
c
o
en
^
CD
•
«^J»
in
fW.
•
r™f
r— '
oo
O
r~"
^
CT^
CM
LU
-------
TABLE 8-10. AVERAGE COST EFFECTIVENESS OF EAF/AOD PROCESS
AND FUGITIVE EMISSIONS CONTROL OVER BASELINE
Regulatory
Model plant alternative
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
New shop: one 22.7-Hg
(25-ton) EAF
New shop: one 90.7-Mg
(100-ton) EAF
New shop: one 272.2-Mg
(300-ton) EAF
New shop: one 136. 1-Mg
(150-ton) UHP EAF
One new 272.2-Mg (300-ton)
EAF added to an existing
shop
One new 22.7-Hg (25-ton)
EAF added to an existing
shop
New shop: EAF and AOD
vessel, 22.7-Mg (25-ton)
capacity each;
controlled separately
New shop: EAF and AOD
vessel, 22.7-Mg (25-ton)
capacity each;
controlled in common
New shop: EAF and AOO
90.7-Mg (100- ton)
capacity each;
separately controlled
New shop: EAF and AOO
vessel, 90.7-Mg (100-ton)
capacity each;
controlled in common
1
2
3K
3b
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3h
3b
1
2
3
1
2
3
1
2
3
1
2
3
Emission reduction
over baseline
Mg/yr
0
6.1
12.3
12.3
0
26.7
53.3
0
97.2
194.6
0
70.9
142.0
0
97.2
194.6
0
6.1
12.3
12.3
0
9.3
18.7
0
9.3
18.7
0
40.6
81.5
0
40.6
81.5
tons/yr
0
6.7
13.5
13.5
0
29.4
58.8
0
107.1
214.5
0
78.1
156.5
0
107.1
214.5
0
6.7
13.5
13.5
0
10.2
20.6
0
10.2
20.6
0
44.7
89.8
0
44.7
89.8
Incremental
cost over
baseline
$/yr
0
1,900
14,100
-18,600
0
21,500
41,300
0
52,100
210,800
0
32,100
87,100
0
53,900
218,000
0
1,900
14,600
-19,300
0
21,900
55,300
0
3,700
25,900
0
40,100
119,100
0
18,000
108,000
Annual cost of
i ncremental
collection
$/Mg
0
311
1,146
-1,512
0
80S
775
0
536
1,083
0
453
613
0
554
1,120
0
311
1,187
-1 ,569
0
2,355
2,957
0
398
1,395
0
988
1,461
0
443
1,325
$/ton
0
284
1,044
-1,377
0
731
702
0
486
983
0
411
556
'0
503
1,016
0
284
1,081
-1,430
0
2,147
2,684
0
363
1,257
0
897
1,326
0
403
•1 ,203
{"Costs are per megagram (ton) of particulate matter removed.
Total furnace enclosure option.
8-18
-------
under Regulatory Alternative 3 for smaller furnaces. This is because
TFE's have been used primarily on smaller furnaces, and there is very
limited operating experience related to TFE's used on medium-to-large
furnaces. As shown on Table 8-10, the TFE capture system incremental
cost i's negative, which means that better capture of fugitive emissions
(over baseline levels) is achieved at a lower cost. Total furnace
enclosures require lower air flow rates than other comparable capture
systems to achieve similar efficiencies, thus resulting in lower costs.
The incremental costs of collecting dust are higher for the smaller
furnaces. The smaller furnace/higher cost relationship is also true
with the AOD vessels.
The annual cost under Regulatory Alternatives 2 and 3 ranges from a
high of $l,146/Mg (1,044/ton) to a low of $311/Mg ($284/ton) of emission
reduction for new EAF facilities. The range is from a high of $2,957/Mg
($2,684/ton) to a low of $398/Mg ($363/ton) of emission reduction for
new specialty steel facilities with an EAF and AOD vessel.
The capital and annualized costs for retrofit facilities are also
presented in Tables 8-1 and 8-4. The control system capital costs for
the retrofit facilities are about 20 percent higher than those which
would be incurred in a greenfield plant. At the same time, the capital
cost for the retrofit plant is estimated to be about 75 percent of the
cost of a greenfield plant, taking into consideration the existing
building and availability of some of the auxiliary equipment, such as
cranes, charge buckets, ladles, casting capabilities, etc.3 Thereby,
the percentage of the control system capital costs compared to the total
plant capital costs are higher.
8.1.2 Modified/Reconstructed Faci1ities
The capital costs for modifying a 136.1-Mg (150-ton) EAF from
normal power to ultra high power (UHP) are estimated to be between 25
and 30 percent of those for a new greenfield plant.4 Thus, the costs
incurred in the modification of the furnace may range between $8.3 million
and $10 million.
The capital cost for upgrading the control system to meet the level
of emission reduction required by the standards of performance is estimated
to be about 20 percent of the cost for a new control system. This would
result in an expenditure of about $1.3 million.
8-19
-------
8.2 OTHER COST CONSIDERATIONS
In addition to the cost of control for process and fugitive emissions,
there are other regulatory costs mandated under the Occupational Safety
and Health Act (OSHA) and the Resource Conservation and Recovery Act (RCRA).
The Office of Technology Assessment estimates that investment costs
to comply with OSHA are about 14 percent of costs to comply with standards
of performance on an industry wide basis. Applying this percentage to
EAF/AOD pollution control systems gives the results presented in Table 8-11.
The EAF dust is classified as hazardous under RCRA. The annualized
costs for the control systems shown in Table 8-4 include the solid waste
handling system costs and the solid waste disposal costs. The incremental
costs are not significantly increased because the EAF dust is already
being handled as a hazardous material. As explained in Chapter 7,
research is being conducted to develop ways of recycling the EAF dust to
help reduce or eliminate the hazardous waste material disposal costs.
The collected dust from EAF/AOD combined operations and control may
be processed at a reclamation center. Therefore, no solid waste disposal
costs are included in the calculation of annualized costs for the specialty
steel shops.10
There are no costs related to water pollution control resulting
from the use of the air pollution control systems on either EAF's or ADD
vessels.
Promulgation of standards of performance for EAF's and AOD vessels is
not expected to impose major resource requirements on regulatory and enforce-
ment agencies since the agencies are already maintaining surveillance for
the sources in the EAF/AOD shops for primary emission control.
8-20
-------
TABLE 8-11. COST ESTIMATES FOR OSHA COMPLIANCE—
EAF/AOD VESSEL FACILITIES13
(March 1981 dollars)
Regulatory
Model plant . alternative
1.
2;
3.
4.
5.
6.
7.
8.
9.
10.
New shop: one 22.7-Mg
(25-ton) EAF
New Shop: one 90.7-Mg
(100-ton) EAF
New shop: one 272.2-Mg
(300-ton) EAF
New shop: one 136.1-Mg
(150- ton) UHP EAF
One new 272.2-Mg (300-ton) EAF
added to an existing shop
One new 22.7-Mg (25-ton) EAF
added to an existing shop
New shop: EAF and AOD vessel ,
22.7-Mg (25-ton) capacity
each; controlled separately
New shop: EAF and AOD vessel,
22.7-Mg (25-ton) capacity
each; commonly controlled
New shop: EAF and AOD
90.7-Mg (100-ton) capacity
each; separately controlled
New shop: EAF and AOD vessel
90.7-Mg (100- ton) capacity
each; commonly controlled „
1
2
3
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3.
3b
1
2
3
1
2
3
1
2
3
1
2
3
Total
pollution
control
investment
$106
1.73
. 1 . 78
1.81
1.71
4.37
4.52
4.62
4.87
5.17
5.57
6.29
6.48
6.67
5.84
6.20
6.68
2.08
2.14
2.17
2.05
3.59
3.65
3.77
2.84
2.90
2.97
9.28
9.45
9.78
8.45
8.59
8.99
OSHA
compliance
investment
$103
242
249
253
239
612
633
647
682
724
780
881 "
907
934
818
868
935
291
300
304
287
503
511
528
398
406
416
1,299
1,323
1,369
1,183
1,203
1 ,259
.Calculated as 14 percent of total pollution control investment.
Total furnace enclosure option.
8-21
-------
8.3 REFERENCES FOR CHAPTER 8
1. Johnstown Electric Furnace Shop Nears Completion. Iron and Steelmaker.
8:10. June 1981.
2. Telecon. Banker, L., Midwest Research Institute, with Geiser, L.,
Carpenter Technology Corp. October 29, 1981. Information on plant
and control costs.
3. Letter and attachments from Lukas, A., Jones & Laugh!in Steel
Corp., to Banker, L., Midwest Research Institute. February 17, 1981.
Submittal of requested information.
4. Telecon. Banker, L., Midwest Research Institute, with Nicholson, W.,
Chaparral Steel Corp. August 26, 1981. Information on furnace
modifications and installation of a new furnace.
5. Telecon. Terry, W., Midwest Research Institute, with Baker, D.,
Rockwood Iron & Metal Co. August 24, 1981. Information on retrofitting
emission controls on Rockwood Iron's facility.
6. Producer Prices and Price Indexes. U.S. Department of Labor,
Bureau of Labor Statistics. Data for March 1981.
7. Capital and Operating Costs of Selected Air Pollution Control
Systems. CARD, Inc. Niles, Illinois. EPA No. 450/5-80-002.
December 1978.
8. Employment and Earnings. U.S. Department of Labor, Bureau of Labor
Statistics. June 1981.
9. Memo from Pahl, D., EPA:SDB, to the files. March 31, 1980. Basis
for estimation of hazardous waste hauling and disposal costs.
10. Memo from Banker, L., Midwest Research Institute, to Pahl, D.,
EPA:SDB. September 15, 1982. Revised final tabular costs.
11. Telecon. Terry, W., Midwest Research Institute, with Woolen, C.,
Pennsylvania Engineering Corp. August 19, 1981. Information on
total furnace enclosure systems.
12. Telecon. Terry, W., Midwest Research Institute, with Bonistalli, R.,
Obenchain Calumet Corp. August 18, 1981. Information on total
furnace enclosure systems.
13. Office of Technology Assessment. Technology and Steel Industry
Competitiveness. June 1980. p. 349.
8-22
-------
9. ECONOMIC IMPACTS
9.1 SUMMARY OF IMPACTS
Three regulatory alternatives proposed for electric arc furnace (EAF)
and electric arc furnace/argon-oxygen decarburization (EAF/AOD) steel
plants are considered in this analysis. Regulatory alternative 1 is the
baseline alternative from which the impacts of the more stringent alter-
natives 2 and 3 are measured. Several impacts are computed for each
alternative; one for each of the 10 model plants. In an effort to provide
a more useful analysis, a single impact estimate is presented for each
alternative based on the premise that model plants 4 and 7 are most represent-
ative of future plant construction. Impacts are computed for 1987 in
1981 dollars.
As noted, regulatory alternative 1 is the baseline from which the
impacts of alternatives 2 and 3 are computed. It is projected that in
1987, at baseline, domestic semifinished steel output would be 119.7 million Mg
(131.7 million tons); steel industry employment would be 438,461 workers;
and steel consumers in the U.S. would import 20.3 million Mg (22.3 million
tons) of foreign steel.
It is estimated that the price of semifinished carbon steel would be
$0.12 per Mg ($0.11 per ton) higher in 1987 under alternative 2 than it
would be under baseline alternative 1. The impact on the price of specialty
steel is an estimated $0.81 per Mg ($0.74 per ton) increase. These price
increases represent impacts of 0.06 percent and 0.12 percent respectively.
Steel users could purchase less domestic steel and more foreign steel
in 1987 as a result of the increases in domestic steel prices. Domestic
steel shipments would be down an estimated 150,000 Mg (165,000 tons) from
baseline under alternative 2--an impact of 0.13 percent. Imports could be
approximately 19,000 Mg (21,000 tons) greater--an impact of 0.09 percent.
9-1
-------
The impact on domestic steel production would cause an estimated reduction
in employment opportunities of about 0.12 percent, or about 542 jobs, in
1987. The estimated total cost of regulatory alternative 2 in 1987 is
$1.381 million (in 1981 dollars).
It is estimated that regulatory alternative 3 would raise the price of
semifinished carbon steel by $0.30 per Mg ($.27 per ton) above its baseline
1987 price. The anticipated impact on the price of specialty steel of
alternative 3 is $2.09 per Mg ($1.90 per ton). These price increases
represent impacts of 0.15 percent and 0.31 percent respectively.
Domestic steel shipments would be down an estimated 370,000 Mg (407,000
tons) in 1987 under regulatory alternative 3—an impact of 0.31 percent.
This impact on domestic steel production would cause a proportionate decrease
in 1987 employment opportunities, about 1,387 jobs. Steel imports in 1987
would be approximately 40,000 Mg (44,000 tons) higher under alternative 3
than under baseline control—an impact of 0.20 percent. The estimated
total cost of alternative 3 in 1987 is $2.575 million (in 1981 dollars).
Electric arc steel making is apparently a profitable activity. This
is indicated by two kinds of evidence. First, construction of new EAF
facilities and the replacement of older steel making technologies by EAF
shops is brisk. This indicates that the industry itself views EAF steel
making as profitable. Second, the engineering cost data' for the 10 model
facilities examined in this study suggest strongly that steel can be produced
at an average total cost significantly below market price. In light of
this observation, it seems unlikely that the industry impacts reported
above would significantly affect the future growth of electric steel making.
9-2
-------
9.2 INDUSTRY PROFILE
9.2.1 Introduction
This industry profile focuses on the blast furnaces and steel mills
industry. The profile has two purposes. The first is informative in
nature; the profile should provide the reader with a broad overview of the
industry. In this sense, the industry profile should be meaningful when
read alone. The second purpose is to lend support to an economic analysis
involving the industry by helping to assess the appropriateness of using a
competitive, monopoly, or other model to analyze the industry. Further,
the profile can provide some of the necessary data that will be used in the
analysis itself.
The industry profile is comprised of five major sections. The
remainder of this introduction, which constitutes the first section,
provides a brief, descriptive, and largely qualitative look at the
industry. The remaining four sections of the profile conform with a
particular model of industrial organization analysis. This model maintains
that an industry can be characterized by its basic conditions (supply and
demand), market structure, market conduct, and market performance.1
The basic conditions in the industry, discussed in the second major
section of this profile, are believed to be a major determinant of the pre-
vailing market structure. Most important of these basic conditions are
supply and demand conditions. Supply conditions are largely technological
in nature, while demand conditions are determined by the attributes of the
products themselves.
The market structure of the blast furnaces and steel mills industry is
examined in the third section of this profile. Issues addressed include
economic aspects of production functions, cost structures, market power,
integration, and barriers to entry. Market structure is the second link in
the overall framework, and has major influence on the third link, the
conduct of market participants.
Market conduct, addressed in the fourth section, concerns price and
nonprice behavior of sellers. Of particular interest is the degree to
which the industry pricing behavior approximates the competitive pricing
model, the monopoly pricing model, or some model of imperfect competition.
9-3
-------
The fifth and finaT section of the industry profile addresses market
performance. The historical record of the industry's'financial performance
is examined, with some emphasis on its comparison with other industries.
Industry trends are noted, and projections are presented for key variables
such as price, output, and investment.
9.2.1.1 Definition of the Blast Furnaces and Steel Mills Industry.
The blast furnaces and steel mills industry (hereafter the steel
industry) is the name given to those firms classified in SIC 3312. The
industry comprises establishments primarily engaged in the production of
hot metal, pig iron, and ferroalloys from iron ore and iron and steel
scrap. SIC 3312 also includes establishments that produce coke, an impor-
tant fuel input to iron production. Establishments primarily engaged in
converting pig iron, scrap iron, and scrap steel into steel are also
classified in SIC 3312. Finally, SIC 3312 includes establishments pri-
marily engaged in hot rolling iron and steel into such basic products as
plates, sheets, bars, and tubing.2
The majority of all goods produced by plants in the steel industry are
intermediate in nature. As such, they are generally shipped to other
plants within SIC 3312 or to plants classified in other industries to be
used as inputs in the production of final goods or other intermediate
goods. The input-output relationships of SIC 3312 will be dealt with in
detail in Section 9.1.2. It is worthwhile to note here that the demand for
blast furnace and steel mill products depends only indirectly upon consumer
tastes and preferences. It will be shown later that this indirect link is
nonetheless significant.
9.2.1.2 The Steel Industry in the Macroeconomy. The value added by
all establishments in SIC 3312 in 1977 was $15,331.9 million. This repre-
sents approximately 0.8 percent of total gross national product for 1977.
Employment in the steel industry in 1977 was 441,900, approximately
0.5 percent of total U.S. employment in 1977. Capital expenditures on new
plant and equipment in SIC 3312 v/as $2,143.1 million in 1977, approximately
0.7 percent of the national total.3 4
A comparison of these shares to those of other industries provides a
greater understanding of the importance of SIC 3312 in the overall economy.
The $15,331.9 million of value added by the steel industry in 1977 was
9-4
-------
approximately 6.3 percent greater than that of SIC 2911, the petroleum
refining industry. Value added by the motor vehicle industry—SIC 3711 in
1977 was approximately 22 percent greater than that of the steel industry.5
SIC 3312 is thus seen to be among the larger manufacturing industries.
The U.S. steel industry is a major contributor to total world steel
production. In 1979, the world leader in steel production was the Soviet
Union, which produced 19.9 percent of worldwide steel output. The United
States was second, with 16.5 percent of total world production. Japan
produced 14.9 percent of total steel produced in 1977.6
9.2.2 Basic Conditions
9-2.2.1 Supply Conditions.
9.2.2.1.1 Product description. Nearly .all output of SIC 3312
consists of intermediate goods. As such, most steel mill products are sold
to producers of other goods. These goods, in turn, might be final goods
such as automobiles, or other intermediate goods such as engine pistons and
valves.
Sales of these intermediate goods can be classified into three types.
The first type consists of sales of the intermediate goods from plants
classified in SIC 3312 to plants classified in other SIC codes. In 1972,
approximately 83 percent of all sales from SIC 3312 were of this type.7
The second type of sale involves shipments from plants classified in
SIC 3312 to other plants classified in the same industry. The third type
of sale is actually implicit in nature and is represented by the transfer
of goods within a plant from one production stage to another. These two
types of intraindustry sales together accounted for approximately 17
percent of all SIC 3312 sales in 1972.7
A large number of identifiably distinct products is produced by plants
classified in SIC 3312. It is useful to categorize SIC 3312 output into
four products: coke, pig iron, raw steel, and finished steel products.
Coke is the carbon residue that results from the heating of coal "in
the absence of oxygen. Coke constitutes the chief fuel used in blast
furnaces to produce pig iron from iron ore. Coke is classified as a steel
industry product since it is nearly always produced as a secondary product
in SIC 3312 plants. While certain other fuels may be used to supplement
coke in the blast furnace, none are used to replace coke entirely.
9-5
-------
Pig iron is the chief material input in raw steel production. The
conversion of iron ore into pig iron takes place in a blast furnace, fueled
mainly by coke. Pig iron is drawn from the furnace in a molten state, and
is approximately 93 percent iron and 7 percent impurities. Once produced,
the mdlten iron is utilized in one of two ways. In an integrated steel
mill, the molten iron can be charged, with or without iron and steel scrap,
into a steel furnace. Alternatively, the molten iron is poured into molds
called "pigs" where it solidifies when cooled. In solid form, pig iron can
be stored in the plant for later conversion into steel, or can be shipped
to other facilities that produce steel.8
Raw steel is the primary manufacture of plants classified^ in SIC 3312.
Produced in one of several types of steel furnaces, steel is substantially
freer from impurities than pig iron.
The American Iron and Steel Institute identifies three major grades of
steel—carbon, alloy, and stainless.9 Carbon steel contains only small
amounts of s'uch alloying elements as vanadium, molybdenum, manganese,
silicon, and copper.10 Carbon steel is suitable for use in applications
not requiring additional strength or other special properties. As most
applications are of this type, carbon steel accounted for 85.3 percent by
weight of U.S. raw steel production in 1979.9
Alloy steel consists of steels containing specific percentages of
vanadium, molybdenum, or other elements as well as larger amounts of manga-
nese, silicon, and copper than carbon steel. Tool steel is an alloy
containing such elements as tungsten and molybdenum. Its added strength
and hardness make it well suited for use in tools and power-driven cutting
and shaping machinery. Another alloy, high-strength low-alloy steel,
contains only small amounts of alloying elements, and is harder than carbon
steel because of special processing.10 Alloy steels are gaining relative
importance in the steel market, increasing from 9.7 percent of total U.S.
raw steel production by weight in 1970 to 13.2 percent in 1979.9
Stainless steel accounted for 1.6 percent by weight of U.S. raw steel
production in 1979.9 Most stainless steel is an alloy of steel, chrome,
and nickel. Stainless steel is generally strong and highly resistant to
rust and corrosion, making it particularly useful in medical, chemical, and
aerospace applications.10
9-6
-------
Raw steel exists as solid ingots'and crude shapes such as slabs,
billets, and blooms. Slabs, billets, blooms, and other semifinished shapes
are suitable for rolling or otherwise processing into finished steel
products. These semifinished products can either be continuously cast from
molten steel direct from the furnace or can be formed from reheated steel
ingots. Approximately 15 percent of U.S. produced raw steel is contin-
uously cast; the rest is produced in ingot form.11 12
The fourth major category of output of SIC 3312 is finished steel
products. Included in this category are only very basic steel shapes such
as plates, sheets, strips, rods, bars, and tubing. Establishments that
produce these products are classified in SIC 3312 only if they are "hot-
rolled" from iron and steel. Establishments that produce very similar
products by a cold-finishing technique are classified in other SIC codes by
the type of product. For example, the total value of output by all
domestic industries of steel nails and spikes in 1977 was $359.0 million.
Of this total, 43 percent was produced in SIC 3312 while 52 percent was
produced in SIC 3315—steel wire and related products.13 14
9.2.2.1.2 Production technology.
9.2.2.1.2.1 Raw materials and other inputs. Coke, iron, and steel
production require a wide variety of material inputs. Some of these
inputs, for example coal and iron ore, are nonrenewable resources. Others-
such as steel scrap are better labeled renewable. An examination of the
goods and services that are purchased, by,, plants classified in SIC 3312
reveals some information about how steel products are produced.
The 1972 input-output (1-0) model of the United States economy identi-
fies 492 distinct goods and services that comprise all economic activity.
The national 1-0 model can be used to determine the economic and technical
interrelationships between all 492 "industries." One of the 492 sectors
identified is SIC 3312—blast furnaces and steel mills.7
One feature of the 1-0 model allows a determination of the "recipe"
for producing one dollar of output of SIC 3312. This recipe is an industry
average, and several cautions should be voiced. First, different plants
may use somewhat different mixes of inputs to produce the same kind of
product. Second, a plant that produces, for example, only coke, will use
quite a different set of inputs than a plant that produces only steel.
9-7
-------
Both plants, however, are classified in SIC 3312. Third, this industry-
average recipe is based on the U.S. economy in 1972. No more recent model
is available. To the extent that input substitution has occurred in the
steel industry since 1972, the input mix has changed.
Of the 492 input-output Sectors, the 10 representing the largest input
cost shares to the blast furnaces and steel mills industry are recorded in
Table 9-1. Most interesting is that the input having the largest cost
share in blast furnaces and steel mill products production is blast
furnaces and steel mills products themselves. Table 9-1 indicates that the
steel industry purchased 17.3 cents of its own output for every dollar of
output it produced. This observation can be explained as follows.
Producers of iron, who are classified in SIC 3312, purchase significant
quantities of coke from coke plants, which are also classified in SIC 3312.
Similarly, steel plants purchase pig iron. The relatively large size of
this intraindustry input coefficient stresses the importance of vertical
integration in the iron and steel industry.
The second most important input by value is iron ore, which represents
an input cost share of 5.8 percent. The third and fourth most significant
inputs by value are not goods at all, but services. For each dollar of
output, producers in SIC 3312 purchased 3.7 cents of railroad transpor-
tation and 3.7 cents of wholesale trade services in order to have all other
inputs delivered to their plants.
The 2.7 percent input cost share for coal is a reminder that coke pro-
ducers are classified in SIC 3312 and that some integrated steel mills
produce their own coke. Coal, of course, is the primary input in coke
production.
Note finally that the purchase of iron and steel scrap represents the
eighth largest cost share input. Few other industries purchase such a
large share of scrap. This share can be explained by the importance of
iron and steel scrap as a substitute for pig iron in raw steel production.
9.2.2.1.2.2 Production processes. As stressed earlier, coke, pig
iron, raw steel, and finished steel products are all produced by plants
classified in SIC 3312. The purpose of this section is to focus only on
the production processes employed to produce the primary output of the
industry—raw steel. A discussion of technological change and trends
follows in Section 9.1.2.1.2.3.
9-8
-------
TABLE 9-1. IMPORTANT INPUTS TO THE BLAST FURNACES AND STEEL
MILLS INDUSTRY—SIC 33127
(1972)
Input
Share of total
input cost (percent)
Blast furnace and steel mill products
Iron ore
Railroad transportation
Wholesale trade services
Coal
Industrial chemicals
Electrometallurgical products
Scrap
Electricity
Natural gas
17.3
5.8
3.7
3.7
2.7
2.0
2.0
2.0
1.7
1.4
9-9
-------
The production of raw steel generally involves the introduction of
some combination of pig iron, iron scrap, arid steel scrap into a steel
furnace. In the furnace the impurities in the input charge are oxidized at
high temperature. Steelmaking technologies differ in the type of furnace
used. ' There are three primary furnace types in use worldwide: the open
hearth furnace, the basic oxygen furnace (BOF), and the electric arc
furnace (EAF). The Bessemer type furnace, associated with the oldest
steel-making process, has virtually disappeared from use.
The open hearth furnace is the oldest furnace type currently in use in
the United States, and is the least prevalent. In 1979, only 14.1 percent
of all raw steel produced in the United States was produced using an open
hearth process.9 In this process, pig iron, scrap, or both are placed in a
long, shallow furnace made of dolomite or silica brick. Gas or oil is
burned to produce a flame that heats the input charge from above the open
furnace. Preheated combustion air, sometimes enriched with pure oxygen, is
forced into the furnace to aid the oxidation of impurities.15 1S
One advantage of the open hearth furnace is its ability to handle a
charge of all pig iron, all scrap, or any combination of both. This allows
open hearth operators to utilize inputs in an economically efficient manner
as relative prices change. The major disadvantage of the open hearth
process is that the heating cycle time, up to 8 hours, is lengthy in
comparison with the BOF or EAF process.16 Due to the heat time required,
the process consumes approximately 2.5 times as much energy (measured in
Btu's) per ton of steel than does the electric arc process. The open
hearth process is relatively labor intensive as well, requiring up to twice
as many labor hours per unit of steel as the basic oxygen process.17
The BOF process accounted for 61.1 percent of all U.S. steel pro-
duction in 1979.9 The metal charge for a BOF must be largely molten
furnace iron—seldom less than 70 percent. The remaining charge can be
cold pig iron or scrap. A largely molten metal charge is necessary because
supplemental heat is not added to a BOF as it is to an open hearth furnace.
The sole heat source in a BOF is the latent heat in the molten charge
itself.18 Even without the addition of supplemental heat to a BOF, steel
refining is achieved in a short period of time due to the introduction of
large quantities of pure oxygen. The availability in recent years of large
9-10
-------
quantities of pure oxygen at low relative prices has made BOF steelmaking
economically feasible.19
The chief advantage of BOF steel making is that pig iron is converted
to steel in a period of approximately 45 minutes. Labor and certain other
costs'per ton of steel are thus substantially lower for the BOF process
than for the open hearth process. The major disadvantage of the BOF process
is that the charge must consist mainly of molten iron. A BOF shop must
have access to a source of,, molten metal; it is therefore usually part of an
integrated steel mill.
The EAF steelmaking process accounted for 24.9 percent of all U.S.
steel production in 1979.9 In the EAF process, solid iron and scrap steel
are placed in the furnace. The EAF can accept a charge of up to 100 percent
scrap. Charged electrodes in the iron and steel charge create an electric
arc between the electrodes and the metal charge. The heat generated melts
the charge. Because no impurities are added by fossil fuels, the EAF
process is particularly well suited to the production of alloy steels and
stainless steel.20
Because the EAF may be charged entirely with scrap, it need not be
affiliated with an integrated plant. EAF facilities can be quite small
relative to open hearth or basic oxygen furnace operations. Thus, they are
relatively inexpensive.to build and market entry is not prohibited by high
start-up costs.
9.2.2.1.2.3 Technical potential for input substitution and production
process innovation. This section discusses the technical potential for
input substitution and process innovation in raw steel production. The
distinction should be made between technical potential and economic poten-
tial.
Technical potential is purely an engineering concept regarding the
physical ability to employ different inputs or processes to produce a
certain output. Technical potential does not ensure economic potential.
Innovation is economically feasible only if (1) technical innovation is
possible, and (2) market conditions are such that profit-maximizing firms
are willing to incorporate the technology. It is clear that if technical
innovation is observed in the industry, it must be both economically and
technically feasible. On the other hand, if technical innovation is not
9-11
-------
observed, only economic feasibility can be absolutely ruled out. Technical
potential is thus a necessary, but not sufficient condition for adoption of
the technology by firms.
Evidence generally indicates that considerable potential exists in a
technical sense for process innovation. However, the domestic steel indus-
try is frequently criticized for being slow to adopt new technologies.21
The very fact that significant quantities of steel are produced by
three distinctly different basic processes supports the notion that process
innovation has occured. Table 9-2 presents the percent of total raw steel
produced in the United States from 1968 through 1979 by the three steelmak-
ing processes. The increase in the proportion of steel produced by the
basic oxygen and electric arc processes at the expense of the open hearth
process is striking. BOF production increased from 37.1 percent of the
total in 1968 to 61.1 percent in 1979. Over the same period, production by
the EAF process nearly doubled.
Technological innovation within the steelmaking processes is evident
as well. Preheating scrap metal allows more of it to be used in the basic
oxygen process. BOF efficiency is also being increased by the practice of
blowing oxygen into the metal charge from below as well as from above the
surface.22
Significant technological advances of other types are also occurring.
Perhaps the most significant change currently is the adoption of continuous
casting technology. In continuous casting, freshly refined molten steel
from the furnace is poured directly into a water-cooled mold. The
hardening steel shape is mechanically pulled continuously from the bottom
of the mold and cut into desired lengths. Bypassing the steel ingot stage
and subsequent reheating saves energy, time, manpower and waste steel. EAF
operators are most rapidly employing continuous casting.11
The major material input in raw steel production is metal—molten
iron, pig iron and scrap steel. The technical potential for substitution
between pig iron and scrap steel can be characterized in two ways. First,
in the general activity labeled steelmaking, pig iron and scrap metal are
excellent substitutes for each other. Second, within a defined steelmaking
process, substitution potential varies greatly with the type of process.
9-12
-------
TABLE 9-2. RAW STEEL PRODUCTION BY PROCESS TYPE9
Year
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
Open hearth
(Percent)
14.1
15.. 5
16.0
18.3
19.0
24.4
26.4
26.2
29.5
36.5
43.1
50.1
Basic oxygen
(Percent)
61.1
60.9
61.8
62.4
61.6
56.0
55.2
56.7
53.1
48.2
42.6
37.1
Electric arc
(Percent)
24.9
23.6
22.2
19.2
19.4
19.7
18.4
17.8
17:4
15; 3
14.3
12.8
9-13
-------
When steelmakirtg is considered as a broadly defined, homogeneous
activity, input substitution is excellent. The ferrous metal input can be
anywhere between 100 percent pig iron (open hearth process) and 100 percent
scrap (electric arc process). In practice, a mixture of pig iron ..and scrap
is typical for the open hearth and BOF processes. Table 9-3 indicates that
in 1978 the open hearth process consumed a 53.5 percent pig iron/46.5 percent
scrap mixture. The pig-iron/scrap mixture was 72.1 percent/27.9 percent in
basic oxygen steelmaking and 2.7 percent/97.3 percent in electric arc
steelmaking.
Substitution-potential between pig iron and scrap within a given
steelmaking process varies, depending on the process. In, the open hearth
process» the ferrous charge can be all pig iron, all scrap, or any mixture.24
However, as indicated in Table 9-3, pig iron has steadily comprised a
little over half of the ferrous charge in open hearth process steel production
in recent years.
The basic oxygen furnace is far less tolerant of input substitution
than the open hearth furnace. The pig-iron/scrap mixture has not varied
far from 70 percent/30 percent. Technological advances, though, are beginning
to relax this constraint. By simultaneously bottom-blowing additional
oxygen into the BOF and preheating the scrap input, scrap can comprise up
to 42 percent of the ferrous charge.22
Electric arc furnaces' have historically depended on scrap as their
major input. Technically, though, the EAF can operate satisfactorily with
a wide range of variation in pig-iron/scrap content.25
Substitution among fuel inputs is possible as well. Gas and oil work
equally well in generating heat for the open hearth process. In a basic
oxygen furnace, the molten ferrous charge itself is the heat source, but
varying amounts of oxygen are utilized in the process.
9.2.2.2 Demand Conditions.
As indicated earlier, steel is
Because
9.2.2.2.1 Historical demand trends.
an intermediate good; most sales are to other producing industries.
of this, demand for it tends to increase in times of increasing business
activity and decrease in times of slackening business activity.
Apparent consumption of steel mill products in the United States is
defined as shipments from U.S. plants, plus imports, minus exports. It is
a measure of how much steel, imported and domestic, U.S. buyers purchase.
9-14
-------
TABLE 9-3. PIG IRON AND SCRAP INPUTS TO RAW
STEEL PRODUCTION23
Pig iron
(Percent)
Scrap
(Percent)
1978
Open hearth
Basic oxygen
Electric arc
1976
Open hearth
Basic oxygen
Electric arc
1974
Open hearth
Basic oxygen
Electric arc
53.5
72.1
2.7
55.7
71.6
1.5
54.2
71.5
3.0
46.5
27.9
97.3
44.3
28.4
98.5
45.8
28.5
97.0
9-15
-------
Table 9-4 depicts 11 years of U.S. apparent consumption of steel mill
products and real gross national product (GNP). Note that the decline in
real GNP from 1973 to 1975 is accompanied by a decline in apparent consump-
tion of steel mill products. The recovery from 1976 to 1978 is marked by
an increase in the demand for steel.
While apparent consumption of steel tends to rise and fall in step
with general economic conditions, it has not increased at an annual rate
equal to that of real GNP. The continuously compounded annual rate of
growth of real GNP from 1969 to 1979 was 2.8 percent. The comparable
growth rate for steel consumption was only 1.1 percent (Table 9-4).
Possible reasons for the slow growth in steel demand relative to that in
general economic activity will be discussed in Section 9.1.2.2.3.
9.2.2.2.2 Important users of steel mill products. Previously, the
1972 input-output table of the U.S. economy was used to identify major
suppliers of inputs to SIC 3312—the blast furnaces and steel mills
industry. The 1-0 table also indicates the value of output of steel mill
products sold to each of the 492 [industries. Table 9-5 identifies the
10 industries that received the most shipments from plants classified in
SIC 3312 in 1972.
Table 9-5 indicates that the most important buyer of steel mill prod-
ucts in 1972 was the steel mill industry itself. This fact emphasizes the
great degree of vertical integration in the industry. Coke producers sell
their output to producers of pig iron who in turn sell pig iron to steel
producers; all are classified in SIC 3312.
The next two most significant purchasers of steel mill products by
value of goods purchased are the motor vehicle parts and accessories indus-
try, which purchased 6.8 percent of all steel in 1972, and the automotive
stampings industry, which purchased 6.3 percent of the total in that year.
In 1979, 18.6 percent of all steel products was sold to a broadly defined
"automotive" market.29
As shown in Table 9-5, the remaining seven of the 10 most significant
purchasers of output of the steel industry are all industries that produce
intermediate goods: metal cans (used to produce canned goods), fabricated
structural metal (for building), fabricated plate work, etc. Again, the
table stresses the intermediate nature of the goods produced in the steel
industry.
9-16
-------
TABLE 9-4. U.S. REAL GROSS NATIONAL PRODUCT AND APPARENT
CONSUMPTION OF STEEL MILL PRODUCTS26 27 28
Year
Real GNP
(109 1972 $)
Apparent consumption
Of steel mill products
(10s Mg)
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1,431.6
1,399.2
1,340.5
1,273.0
1,202.3
1,217.8
1,235.0
1,171.1
1,107.5
-1,075.5
1,078.8
104,270
105,800
98,365
91,678
80,738
108,485
111,133
96,698
92,981
88,070
93,133
9-17
-------
TABLE 9-5. IMPORTANT PURCHASERS OF OUTPUT FROM THE BLAST FURNACES
AND STEEL MILLS INDUSTRY—SIC 33127
(1972)
Purchasing industry
Percent of total
SIC 3312 output purchased
Blast furnaces and steel mill products
Motor vehicle parts and accessories
Automotive stampings
Metal cans
Fabricated structural metal
Fabricated plate work
Screws, bolts, nuts, rivets, washers
Iron and steel forgings
Miscellaneous fabricated wire products
Sheet metal work
17.3
6.8
6.3
5.6
4.5
3.3
2.3
2.1
2.1
1.9
9-18
-------
9.2.2.2.3 Competing products. Section 9.1.2.2.1 Indicated that the
growth rate in the demand for steel has lagged behind the growth rate in
real GNP in recent years. Section 9.1.2.2.2 emphasized that steel mill
products are primarily intermediate goods used in the production of other
intermediate goods. It is possible that the relatively slow increase in
demand is due to the degree of substitution potential that exists between
steel and competing materials for use in the production of other goods.
This section addresses this issue.
As stated previously, 18.6 percent of all steel was purchased by
automotive producers in 1979. According to one examination:
... the longer-range probabilities for more shipments [of steel] to this
area [automotive] do not look good. Fuel economy regulations legislated
by the government have compelled automobile manufacturers to design more
efficient cars. This has meant not only smaller autos [which use less
steel], but cars that contain proportionately less steel....30
The chief materials being substituted for steel in automobile production
are aluminum and plastic, which are being well accepted.
The substitution of aluminum for steel is also having an adverse
impact on the demand for steel by beverage container manufacturers. Even
though canned beverage consumption continues to increase, steel shipments
to the container market peaked in 1974.31
Steel faces competition from materials on many fronts. In construc-
tion, steel competes with concrete and, to a lesser extent, aluminum.
Aluminum has replaced steel in some facets of ship construction. Polyvinyl
chloride pipe and tubing are major competitors to the steel counterparts.32
9.2.3 Market Structure
In 1977 there were in total 396 blast furnaces and steel mill com-
panies. This represents an increase of 64.3 percent from the 241 companies
in 1972. From 1967 to 1972, the number of companies increased only
20.5 percent, from 200 in 1967.3 This is an interesting observation in
that apparent steel consumption grew at least as quickly in this 5-year
period as in the latter (1972-1977,). The most plausible explanation is
market entry in recent years by small companies operating small, electric
arc furnace mills. This issue is addressed later in section 9.1.3.2.
9-19
-------
9.2.3.1 Geographic Distribution of Plants. The influence of basic
supply conditions on market structure is also apparent in the effect on
geographic distribution. Basic oxygen furnace plants and, to a lesser
extent, open hearth plants, favor plant integration. In turn, integrated
plants are ideally located near sources of coal and iron ore. Electric arc
furnace plants, however, utilize scrap as their major input. Thus, EAF
facilities are less tied to coal and ore producing regions of the country.
The geographic distribution of plants bears this out. In 1972, 50.0
percent of all plants were located in five states: Pennsylvania, Ohio,
Indiana, Illinois, and Michigan. These states contain coal and ore produc-
tion sites. In 1977, only 38.9 percent of all plants were located in these
same states.33 34 As EAF steel production increased from 17.8 percent of
the total in 1972 to 22.2 percent of the total in 1977 (see Table 9-2),
geographic concentration became less pronounced.
Fully integrated plants, which need coal to produce coke, and ore to
produce iron, are as expected, concentrated in the five states listed
above. In 1977, 65.7 percent of all fully integrated plants were located
in these five states.
9.2.3.2 Firm Concentration. Basic supply conditions affect industry
concentration ratios. The greater the proportion of total output produced
by a given number of the largest firms, the more concentrated the industry.
Raw steel production would be expected to become less concentrated as
more steel is produced by smaller, less integrated EAF facilities. Because
EAF facilities need not be part of an integrated operation, they are less
costly to build. One estimate places the cost (per ton of annual output)
of building an integrated steel mill at almost six times that for building
a small electric arc furnace plant.35
The four largest domestic producers of steel mill products together
shipped 53.7 percent of total market tonnage in 1974. In 1979 these same
producers, still the four largest, accounted for only 49.8 percent of all
shipments.36
Recent market entry by small, electric arc furnace steel plants is
credited in part with reducing industry concentration. One analyst
estimates that the so-called "mini-mills" accounted for 13 percent of
domestic steel shipments in 1978, and forecasts that they will capture at
least 25 percent of the market by 1990.3S
9-20
-------
9.2.3.3 Vertical Integration. As indicated, certain steel production
technologies favor some degree of integration. A plant that produces steel
using the basic oxygen process, for example, requires a ferrous charge
comprised largely of molten iron. The molten iron can most efficiently be
transferred directly from the blast furnace where it is produced, never
having cooled. Conversion of iron ore into iron in the blast furnace in
turn requires coke. Coke can be purchased from other plants, but is
frequently produced in the same establishment.
Production of steel via the electric arc furnace requires little plant
integration. As cold scrap is the usual ferrous charge, neither blast
furnace-produced molten iron nor coke is necessary.
The 1977 Census of Manufactures indicates that there were 504 plants
classified in SIC 3312 in that year. This is an increase of 38.5 percent
from the 364 plants reported by the Census in 1972. In Table 9-6, the
plants classified in SIC 3312 are further classified by degree of inte-
gration for the years 1972 and 1977. Plants that produce coke, iron, raw
steel, and finished steel are regarded as "fully integrated." Plants that
produce iron, raw steel and finished steel, or just raw steel and finished
steel may be considered "partially integrated." Plants that produce only
one of these four products are called "nonintegrated."
As shown in Table 9-6, the proportion of total plants that were fully
integrated declined from 10.7 percent in 1972 to 6.9 percent in 1977. Over
the same period, the percent of total plants that were partially integrated
declined from 23.1 in 1972 to 19.7 in 1977. Nonintegrated plants, however,
increased from 66.2 percent of the total in 1972 to 73.4 percent in 1977.
The trend towards nonintegrated plants is an example of basic supply
(technological) conditions influencing market structure. Table 9-2
indicated a substantial increase in the production of steel by the electric
arc process since 1972. This trend towards a technology requiring little
plant integration may account for the shift away from plant integration.
9.2.3.4. Horizontal Integration. Horizontal integration is defined
here as diversification into the production of nonsteel goods or services
by steel firms. Integration of this type is evidenced by firms in the
steel industry. One estimate indicates that as much as 25 percent of total
sales by U.S. steel firms are nonsteel.39
9-21
-------
TABLE 9-6. PLANT INTEGRATION37 38
Degree of integration
Percent of total plants
1972 1977
Coke, iron, raw steel, finished steel
Iron, raw steel, finished steel
Raw steel, finished steel
Single product or other combination
Total plants (no.)
10.7
,.3.6
19.5
66.2
364
6.9
3.4
16.3
73.4
504
9-22
-------
Inadequate profit margins in steelmaking, to b^ discussed later in
this profile, are frequently cited as the reason for diversification by
steel firms. Petroleum, chemicals, and financial services are among the
industries in which steel companies have invested. Diversification is
expected to continue. Investment in nonsteel enterprises by steel manufac-
turers presently accounts for 20 percent of their total investment.40
9.2.3.5 Economies of Production.
9.2.3.5.1 Long-run cost structure. An industry is said to have an
increasing (decreasing) cost structure if the expansion of industry output
over time increases (decreases) the real, average total cost of producing
one unit of output. The long-run cost structure of the steel industry is
examined here under the assumption that price equals average total cost.
From 1961 to 1979, U.S. raw steel production (in tons) increased at a
continuously compounded average annual rate of 1.8 percent.28 41 During
this same 18-year period, the nominal price of steel increased at an
average annual rate of 6.2 percent.42 That part of the nominal price
increase attributable to inflation can be removed by deflating the nominal
estimate by the increase in the general price level (as measured by the
implicit price deflator) over the same period. The implicit price deflator
was estimated at 4.8 percent per year over this period; therefore the real
price of steel increased 1.4 percent per year from 1961 to 1979. An
increasing long-run cost structure is thus indicated for the steel
industry.
9.2.3.5.2 Short-run cost structure.
9.2.3.5.2.1 Economies of scale. The presence or absence of economies
of scale is often cited as being partially responsible for determining
short-run cost structure. There is general agreement that significant
economies of scale do exist in steel production; that is, physical output
increases more than proportionately with all inputs. The Office of Tech-
nology Assessment of the U.S. Congress states that domestic steel pro-
ducers, because they have smaller, higher cost plants than their foreign
counterparts, are less able to realize economies of scale.43 The U.S.
General Accounting Office says "..-.bigger is cheaper in integrated steel-
making. Large-scale plants can more efficiently use equipment, labor, and
energy than small plants."44 The same report follows with an estimate of
9-23
-------
4 million tons of output (of raw steel) annually as a minimum economically
efficient plant scale.
Economies of scale and minimum efficient scale are however, determined
in part by the technology the plant employs. Many new steel companies are
operating electric arc furnace plants with capacities under 0.5 million Mg
annually. Far from being uneconomic, these small plants are able to sell
raw steel below prices charged by integrated mills, and still profit.35
Further advances in steelmaking technology could further reduce the
presence of economies of scale.
9.2.3.5.2.2 Production costs and plant vintage. There is a definite
link between steel production costs and plant age. In general, per unit
costs of steel production are higher for older plants than for newer
plants. The General Accounting Office maintains that the obsolescence of
U.S. steel plants is the industry's major obstacle in meeting domestic
steel demand.45
The cost-age relationship is primarily the result of technological
advance over time. Electric arc furnaces can produce steel at lower cost
than the open hearth or basic oxygen furnaces. Newer plants increasingly
adopt the electric arc process, hence newer plants are lower cost
operations.
9.2.3.6 Entry Conditions. The U.S. steel industry has long been
regarded as presenting barriers to potential market entrants.46 Entry,4.s
said to be difficult because the very large minimum efficient plant scale
makes for a substantial initial capital investment. One estimate places
the cost of a new integrated steel mill at $1,107 per Mg of capacity, in
1978 dollars.35 This would put the capital cost of a 4 million Mg plant
(the previously mentioned estimate of the minimum efficient plant scale) at
nearly $5 billion.
That barriers to entry are actually restrictive is not clear from
Census data. In 1967, 200 companies operated one or more establishments
classified in SIC 3312. The number of companies increased 20.5 percent
from 1967 to 1972, for a total of 241 companies in 1972. Market pene-
tration was even more pronounced from 1972 to 1977. A total of 396
companies operated in SIC 3312 in 1977, an increase of 64.3 percent from
1972.3
9-24
-------
Much of the dramatic increase in the number of new companies since
1972 is no doubt the result of new, small-scale, electric arc furnace plant
operations. "Mini-mills" are less expensive to build because blast furnace
and coking facilities are unnecessary. An estimate for building a new
mini-mi 11 is $192 per Mg of capacity in 1978 dollars—far under the $1,107
estimate for an integrated mill.35 Removal of the start-up cost barrier is
significant. Entry barriers of other types appear to be minimal.
9.2.4 Market Conduct
This section focuses mainly on pricing behavior in the steel industry.
The question is whether the steel industry most closely approximate the
competitive pricing model, the monopoly pricing model, or some model of
imperfect competition. Important considerations are homogeneity of
product, degree of industry concentration, barriers to entry, and observed
pricing practices.
9.2.4.1 Homogeneity of Product. The degree to which the output of an
industry is perceived fay demanders to be homogeneous is an important deter-
minant of industry pricing behavior. The more homogeneous the product, the
more likely a single market price will be observed. A perfectly homo-
geneous good is difficult to sell at some price higher than that offered by
one or more competitors. Interestingly, in his discussion on this subject,
Nicholson suggests steel girders as one tentative example of a strictly
homogeneous good.47
The degree of product homogeneity in the steel industry is not easily
determined. Section 9.1.2.1,1 of this profile suggests four major
categories of products produced by establishments classified in the blast
furnaces and steel mills industry—coke, pig iron, raw steel, and finished
steel products. These products differ greatly from one another in use.
Coke, for example, is of no interest to a construction firm demanding steel,
girder. The price for a megagram of coke would not be equivalent to that
for a megagram of steel girder; the two are not homogeneous products.
A meaningful discussion of product homogeneity cannot proceed without
some reasonable limitation of product definition. Section 9.1.2.1.2.2 of
this profile, which discusses production processes employed in SIC 3312,
concentrates on the primary output of the industry—raw steel. The same
approach will be adopted here. The relevant question is whether raw steel
is perceived by buyers to be homogeneous.
9-25
-------
In 1979, approximately 85 percent of all raw steel was carbon steel,
the remainder was alloy and stainless.9 Raw carbon steel is produced as
ingots, blooms, slabs, billets, and other semifinished shapes. All of
these shapes resemble one another in that they are, to varying degrees,
square, blocky, solid forms with rounded edges. All are similar in that
they are not useful in these forms, but must be rolled, drawn or otherwise
processed into more finished shapes such as girders, rails, bars, rods,
sheets, pipe, and plates.
Scherer compares the homogeneity of steel with that of cement and
rayon.48 Any differences in steel, he argues, are superficial. There may
be complex differences in finish, temper, packaging, etc., but steel is
fairly homogeneous.49 The Federal Trade Commission says that within each
steel product line, steel is relatively homogeneous; the product of one
plant is physically substitutable for the product of another.50
Homogeneous output alone does not ensure a single pricing model. A
strictly homogeneous product can be produced under perfect competition,
monopoly, or any other market model.51 It becomes necessary to investigate
other indicators of pricing behavior. However, recognizing the hazards of
such a generalization, product homogeneity will be assumed for raw steel.
9.2.4.2 Degree of Concentration. Industry concentration largely
determines market pricing behavior. At the one extreme, where all industry
output is produced by a single firm, "the pure monopoly pricing model is
most relevant. The other extreme is an industry characterized by many
sellers, with no one firm producing a significant share of total output. A
perfect competition pricing model is then applicable. Either extreme is
rare, and the steel industry is neither extreme.
The U.S. steel industry is clearly not a pure monopoly. In 1977 there
were 396 companies with plants classified in SIC 3312.3 Whether the exist-
ence of 396 companies justifies the use of a perfect competition pricing
model is less clear. For perfectly competitive pricing practices to
result, it is not sufficient that a large number of firms exist. No one
firm must produce a large enough share of industry output to enable it to
influence market price by its own actions. In 1979, the single largest
domestic steel firm, U.S. Steel, accounted for 20.8 percent of total
domestic steel shipments.52 This large market share of a single producer
9-26
-------
brings into question the appropriateness of the perfectly competitive
pricing model for the industry.
9.2.4.3 Barriers to Entry. The degree to which barriers to entry
effectively reduce market penetration by new firms influences industry
pricing behavior. Nicholson calls barriers to entry the source of all
monopoly power.53 Section 9.1.3.6 concludes that effective barriers to
entry do not exist in the steel industry.
The apparent lack of significant barriers to entry promotes the selec-
tion of the perfectly competitive pricing model. "Free" market entry is
necessary but not sufficient to support this model. "Free" market entry
obtains when the only costs of production incurred by new firms are those
incurred by established firms.54 In the case of the steel industry, econo-
mies of scale enjoyed by existing firms once resulted in lower per unit
cost of production. Technological advances, especially the development of
the electric arc furnace, have apparently diminished this technological
advantage, as evidenced by recent market entry.
9.2.4.4 Observed Pricing Practices. The purpose of this section is
to comment on actually observed pricing practices in the steel market as
another aid in selecting an appropriate pricing model. If market partici-
pants behave as perfect competitors, a perfect competition model would
logically be deemed appropriate.' If evidence of imperfect competition
behavior is apparent, other market models must be explored. Emphasis will
be placed on observed pricing behavior in recent times--the past 20 years.
Selection of the past 20-year period for examining pricing behavior is
not entirely arbitrary. Until the 1960's, the U.S. steel industry had a
long history of adherance to a pattern of rigid administered pricing.
During normal times in the period prior to the 1960's, U.S. Steel Corpora-
tion, with a market share exceeding one half, assumed the role of price
leader and its smaller rivals followed.55
In the 1960's and 1970's, pricing behavior has been complicated by two
factors. First is the competition provided by foreign steel supply.
Second is the industry structural Change resulting from changing basic
conditions, especially technological advance.
Pricing behavior in the domestic steel industry is most frequently
said to be characteristic of an oligopolistic market. An oligopoly is a
9-27
-------
market with relatively few firms producing a homogeneous product. A single
firm in an oligopoly can have some effect on the price it will receive for
its output because it has a significant market share of the total output.56
Smolik cites the U.S. steel industry as being a "... concentrated
oligopoly on the basis of concentration ratios and on the basis of its
highly knit social structure." The oligopolistic structure, he argues, is
further strengthened by the homogeneous nature of most steel products.57
Adams (writing in 1977) states:
The steel industry today is—structurally speaking—an oligopoly and is
dominated by a relatively few, large, integrated producers. These, taken
together, own or control about three quarters of the nation's . . . ingot
and "steel for casting" capacity . . . .5S
Unfortunately, a tentative conclusion that the American steel industry
is an oligopoly makes it more difficult, not easier,-to identify a specific
pricing behavior model. There are no generally accepted pricing behavior
models for oligopolists as there are for perfect competitors and mono-
polists. Many price-output combinations can result based on various sets
of behavioral assumptions. Observation of actual oligopoly markets
suggests that almost anything can happen.59 60
One generalization that perhaps can be made is that some form of
administered pricing behavior will emerge under oligopoly. The outcome of
a single firm in an oligopoly changing its output price depends, in part,
on the reactions of other firms. A firm stands to suffer if it cannot
accurately determine beforehand how its actions will be met by other firms.
Prices tend to be fixed until some unifying event can ensure the "appro-
priate" response by all firms.61
A system of price leadership often emerges in oligopoly. If a single
firm is recognized as the price leader, price adjustments can occur with
some assurance of market order as other firms follow in concert.
Price leadership has been an apparent characteristic of the U.S. steel
industry. U.S. Steel Corporation is the usually recognized price leader.
All other companies follow the leader's actions closely.57 62 As recently
as 1971, an announcement of a price hike for all mill products by U.S.
Steel Corporation was promptly followed by a similar declaration by other
major producers.63 When the U.S. General Accounting Office in 1979 asked
9-28
-------
steel users why they buy foreign steel, one answer given was that they, as
buyers, perceived that many U.S. mills follow suit each time a major steel
mill publishes a new price list.64
While there is near universal agreement that the domestic steel indus-
try has historically been oligopolistic with U.S. Steel Corporation acting
as price leader, trends in recent years indicate the situation is changing
or has changed. The reason for this basic change in market conduct is,
once again, changes in market structure. Three structural changes have
been observed.
First, the Federal Trade Commission reports that since 1960, U.S.
Steel Corporation has had to share the role of leader with Armco Steel and
Bethlehem Steel. Efforts to raise prices by any leader have frequently
been ignored unless they reflected basic supply and demand conditions.65
A second major structural change has been the dramatic increase in
steel output by independent "mini-mills." Mini-mills regularly engage big
producers in price competition, and are beginning to be a greater threat to
historically large steelmakers than are foreign producers.35 6S The
success of mini-mi 11s is in turn the result of a successful technological
advance—the electric arc furnace. This is an example of how changes in
basic conditions can, by influencing market structure, ultimately result in
changes in market conduct.
The third major structural change is the increasingly significant role
of steel imports. In 1961, imported steel mill products accounted for only
4.7 percent of United States consumption. This import market share
increased to 13.8 percent in 1970 and reached a peak of 18.1 percent in
1978.2S 41
Steel imports may significantly affect the price behavior of domestic
steel firms. If the U.S. steel industry is an oligopoly, U.S. Steel '.,
Corporation is the dominant firm, with steel shipments of 19.0 million Mg
in 1979.52 In that same year, total steel mill imports totaled 15.9 mil-
lion Mg—83.7 percent of the dominant firm's output.28 As foreign pro-
ducers collectively capture more of the U.S. market, domestic "firms
individually and collectively enjoy less market power.
To summarize, it is not easily concluded that currently observed
pricing behavior in the steel industry is oligopolistic. Increasingly,
9-29
-------
individual firms are setting their own price based mainly on their own cost
structures and basic underlying supply and demand conditions.
9.2.5 Market Performance
In a conditions-structure-conduct-performance industry profile, the
performance section is of special importance. Market performance is seen
as the result of the earlier causal chain. Emphasis in this section will
be on three aspects of market performance. First, a financial profile of
the steel industry will be presented with comments and comparisons to other
industries. Second, recent trends in such industry variables as price,
output, employment, and investment will be presented. Third, industry
projections will be discussed.
9.2.5.1 Financial Profile of the Steel Industry. The objective of a
firm is to maximize shareholders' wealth. It can be shown that a firm that
maximizes profit, maximizes the discounted stream of cash flows to its
shareholders.67 If this is accepted as the primary objective of firms,
some measure of success in achieving this objective is surely a relevant
measure of financial performance.
. Stockholders' equity is the portion of a corporation's assets owned by
holders of common and preferred stock. It is equal to the excess of a
firm's assets over its liabilities.68 A ratio frequently used to assess a
firm's success in maximizing shareholder's wealth is after-tax profit to
stockholders' equity. An average of this ratio for firms representing
approximately 90 percent of raw steel output over the period 1960-1979 is
presented in Table 9-7. As a comparison, the same ratio is presented for
firms representing all manufacturing over the same period.
A profit-to-equity ratio of 10 percent or higher is usually regarded
as necessary to provide dividends to the holders of stock, as well as funds
for future growth.72 The 20-year average profit-to-equity ratio for all
manufacturing, as calculated from the data presented in Table 9-7, is
13.0 percent. The 20-year average for the steel industry is considerably
lower—7.4 percent. In only one year, 1974, was the ratio for steel firms
equal to or greater than that for all manufacturing. The profitability
difference as measured by this ratio between steel and other industry is
even more pronounced in recent years. During the period 1970-1979, average
annual return to steel stockholders was only 7.2 percent, compared to
13.8 percent for all manufacturing stockholders.
9-30
-------
TABLE 9-7. AFTER-TAX PROFIT TO STOCKHOLDERS' EQUITY69 70 7i
Year
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
1966
1965
1964
1963
1962
1961
1960
Steel
(Percent)
6.8
7.3
0.1
7,8
9.8
17.1
9.3
5.8
4.3
4.1
7.0
8.2
6.9
8.9
9.4
9.0
7.3
5.3
6.5
7.9
All manufacturing
(Percent)
16.7
15.9
14.9
15.0
. 12.6
15.2
14. 9
12. 1
10.8
10.1.
12.4
13.3
12.6
14.2
13.9 V
12.6
11. 6
10.9
9.9
10.6
9-31
-------
These figures tend to support the contention by Adams that the steel
industry's profit record has been poor in recent years. Adams blames the
industry's noncompetitive pricing conduct, leading to technological
lethargy and inadequate capacity utilization, for low profitability.73
9-2.5.2 Financial Profile of Firms Owning Electric Arc Furnace
Facilities. In 1980, 80 independent companies either owned EAF facilities
or owned smaller companies that owned EAF facilities. These 80 companies
are listed in Table 9-8, along with financial data for the 42 firms for
which such data are readily available. The remaining 38 firms are believed
to be relatively small.
Several financial ratios have been calculated for each firm for which
data are available. These ratios are presented in Table 9-9.
The liquidity ratio is a measure of the firm's ability to meet current
obligations as they come due. A liquidity ratio above one indicates the
firm can meet current obligations with little trouble. A firm with a
liquidity ratio below one may be unable to pay bills on time, which may
lead to the firm's demise.75 Firms operating EAF facilities evidenced
liquidity ratios that varied from 1.25 to 3.15 in 1979.
The profit ratio measures the ability of the firm to pay dividends to
stockholders while maintaining adequate funds to ensure growth. A ratio of
10 percent or higher is usually deemed necessary to secure these ends.72
Sixteen EAF firms had profit ratios in 1979 below 10 percent and two firms,
Penn-Dixie and United States Steel, had negative profit ratios. A total of
17 firms had profit margins in excess of 15 percent.
The leverage ratio indicates the relationship between total debt and
stockholders' equity. A ratio exceeding one indicates the firm is mainly
financed by debt. While it is difficult to say what the highest satis-
factory ratio is, in general the higher the ratio, the more likely that the
firm is unable to meet its long-term obligations. Of the 42 firms, 26 had
ratios exceeding one in 1979, indicating they are substantially debt
financed.
Table 9-10 lists the simple (unweighted) means of the three ratios for
the 42 EAF companies. Also listed are industry average ratios for several
other industries for the same year. While it is difficult to draw con-
clusions, observations can be made. The mean liquidity ratio for EAF firms
9-32
-------
V*)
\£\
z.
Z3
U.
O
on
H
z <^
H~1 O*1
t— u- r~-
<^ HH en
OH rH
1 1 1 ^ s^ /
O 5
O
CO
UJ _J
^™i «cC
2 H-t
£>
CO
o
II
lfl CO
4-> CU
rti ••••
fit "f"*
tO r— +J
01 (0 •!-
O ^
i— h- ja
(t3 03
•P -r-
0 -1
-p
cr
UJ
*r"
-
i.
Q.
(A
•P -r-
C. -P
2. r^
S- T-
3 .O
0 (O
•r-
1
•P
c co
j_ (jj
i- in
3 W
**p\
c
03
O.
S
0
o
rH rH en
o o tn
00 O rH
•* rH f>*
M
rH
ijrt r^» ^D
rH"oo"rH"
CM
00 f~>
r-f Cs'to"
tfi •* IT)
CM CO
rH 00 CO
to" r»T r-T
CM 0 CO
CO rH CO
ri
to
CU
s_
r— CO
0) 3
0) -0
•P C
CO HI
>> S
4-> 3
r™" i*—
W3 "O r~™
•r- 3 CO
O — 1 CU i—
CU -P CU
O- >> CO C
CU CJ5 ^3 CD £H £»
|IBK GU S"~ tJ Tj Z3
r— r^ E E 4J "3
O
rH
CO
CM
LO
CM
O
en
• uf
CO
f^
«t
Lf)
o
co
en"
oo
CO
CM
o
LO
CM
o
en
crT
CO
O
rH
O
CM
^jT
«=r
00
r-T
X
O
u
r— r~-
•r~ CU
3: 0)
-p I—
t3 r— CO CO
C CU CU
ro 01 e -P
+J CU CO
U CU S-
(O 03 CU O
oo
rH
f
ID
P^
cn
CM"
^^
en
CO
en
en
co
5
o
o
^jT
CO
CM
O
CM
co"
p^
«tf-
c
O r—
. -r- 0)
r*— >p CU
eu tn -P
CU t- CO
-P 0
CO Q. n3
S- •!-
c o c:
3 O
s- ns la
• r™' uj • u; "^ ( ; • »su ^i <^j ^i *~j
oo o - en en co rH co. ^t* ^f* r>^
to co if) o r-» en en rH un en
LOLO cor^oo Looco«=fr^
f*°\ Q*J QQ QQ fio oo co CD co en
CO rH rH rH rH CD CM
,
JoP^ r^rHLO LnSrHOO^
oo«!fcncM«* cooocounoo
(~l C^ * f^* tji ^*J* GO 00
^^ I*" i ^}" LO rH • LO ^^ co r***" r^-
COCO Lf) LT> CO CSJOLDCOlT)
CMO ocMcn r^-oOLncor^-
^* oo co oo co co r** co co r** ^^
O)r— i— CO
l/> O «r" CO
S^! )«« *5C" CO tO
s- o -P to
O C O) r— CO CO
3 J= C 0) •—
i— O -r- CO r— r— C
COCCU<— 4JOOJ •!-
COOr- i— COOCO 03
H->S- Or- r— I— -PT3 +J
CO t— 1 0- Q£ CO CO r— COi— CO
CU CO CO 03 n3 CO CO
[ ^ r™ [ ^ ^0 ^^ 4^ S" *r™ to 3; r— tO C
COOCT3COCOS-jai-S_jQQ.l-
IrOn3n3r8COU_^='ooOJ->>rt3
CO
«*
^T
iD
• rS\
• • OO
LO
tH
CO
CM
en
CD
r-T
f^-
* §
«*"
rH
* * en
CO
0_
^jT
CM
o
LO"
co
r— to
CO Ct
CO O r—
+j co co
CO CO
T3 .p
S- C CO
CO nj
4-> 03
03 r— -a
•a -i- •—
9-33
-------
.0
cu
a
(A
CU
II
W)
-P
CU T-
tn i— -P
tn re T-
<«^
r- H- -Q
re re
cr
LU
•a
•P -r-
<= -P
- CM
CO rH
U3 LT) CTi
r- •
rH
co <*
co o
o «^
r4"i-T
O CO
co o co r-- oo
co r- CM o to
CO «* rH rH
cn
cn rH
LO CM
in co
CO
co
in
co
co LO
CM CM
CM CM
CD rH
CO in
O CO
CO iH CM <• CM
co cn co to co
r-. O rH CO rH
«S M
in co
<* r- o CM r-
«^- ITS r«. «* co
cn rH <5j- CM co
ID rH
cn CM
o co
•cn rH co r*. co
•3- CM LO LO CM
CO ID r- rH
LO
o
rH
CO
LO
LO
CM
CO
cn
CO
rH
rH
o r^- o en
o CM o to
co to r-- co
r~«. o un r-
CM rH O CM
o cn o LO
O 00 o cn
p"* cn cj" P*"
* ** •» **
cn ^* cn to
rH CM CO «*•
*t ^
CM LO
O CM O
O
CM «*• CO CO
cn rH co rH
Lf5
ro
CM
cn
r->
CM
O
CO
co
Lf)
CM
CO
*i
ID
CO
LO
r~
co
cn
o
• CM •
CO
«sh
CM
<*• CO
h- to
«^- cn
co to
rH I-.
rH
CM t~-
LO O
cn co
M *N
o co
cn co
f»» • • CO
CO
CM
CO
cn
rH
CO •
0
rH
rH
CO
rH
CO
P-.
cn
CO
LO P-.
CO CO
to co
o co
CO
co
CM
LO
co
r-.
iH
«*• •
p-
rH
O
CM
• • P^i *
LO
P-.
• O
o
CM
CM"
r<
rH
to
f-
o
o
p-
LO
CO •
to
CO
i *»
cn
to
to
to
CM
rH
co to
tO Lf)
cn r--
co co
rH tD
p- r- P^ cn LO
o LO to LO LO
CM rH r-» to o
r-T r-T to" CM" cn
CM CM ^- CO CO
CM tO rH
LO rH '
<^- LO co cn
-si- o CM «^-
CM CO CM iH
•51-
co
cn
ft
LO
to
r»- co
LO rH
LO CO
co to «3- **• to
CO CM O CO O
CO O rH Cn rH
en
*\
CM
O CO O •*
O «^- O rH
an LO •«*• co
M *. #t *l
cn CM co to
o in LO co
<*• CO rH
O CO O O
o o o co
in r-. co to
to cn to cn
cn LO rH
CO
co
CM
co
r-.
r-
rH
in
r»
CM
co
co
LO
to
CM
3?
to
CO
in
co"
to
CM
to
ro
co
CM rH
r~> to
CO
rH cn
rH LO
CO CO
cn"co
if) •=!-
cn co
r^. CM
cn cn
co to
re •—
a. cu
E O r-
O P CU TD
cu ~
cu
cu
•P
co
o co
•p
c co
c
re
s-
o
-P o
o -p
2: CU •!-
CT)'r-
-o s- re
s- o •»
o cu re
C r— r—
re cu «
Q£ CU -P
iP If} CU
re
o
u
s- s-
_ _ eu cu
Cnr— -P P I
c c c c
i— CO
o
in tJ
i- C
cu re
re
^_
J=.
i— cn
CU 3
cu re
4J 1
CO
a.
o in
in cu
in c.
cu o
in
cu
cu
p
C CO
o
in in
a in
at cu
cn«—
i. c
O •!-
<-} re
. CO
S c
CU >,
r— r—
i- in
re o
LU <~)
O)
p
CO
0
i.
+J
u
r— CU
CU i—
CU LU
4.)
00 >,
c u
O 3
(/•> P
•a c
3 CU
•a
cu
re
•o
*© c
10 o
C 5-
O M
CJ
CU
cu i—
C i —
O -r-
p >
in x
>> o
cu c.
r—
CU
CU
-p
co
cu
•a
cu
u
re
r«
CU c—
•P cu
00 CU
p
S- CO
re
P in
in c
cu cu
IS
r—
CU r—
CU CU
P CU
co P
CO
c
o J=:
Marath
Mel out
CU c—
cn cu
s- cu
O P
U_ CO
"re re
c c
0 0
re re
22 32
el & Structure
el
cu cu
p p
co to
>> s-
cu re
in P
s- co
cu
° P
O) O
25 2=
cu
$-
5
T3
c
re
rcu
cu
p
CO
c
S-
-------
r*"S
•a
CO
3
C
'•P
C
O
o
s_>
•
CO
t
en
UJ
CD
^j
.a
CO
o
11
in to
•P . CO
CO •!-
tO r— 4->
tO (tf -I—
fc
4^
•f—
3"
cr
LU
•P
4-
O
a.
f\\
vll
•P •!-
C -P
0) •!-
S- r—
t. -r-
3 X2
O (0
•r™
-P
c to
3 (f>
O »
C
to
OL
e
o
o
^-
p^
CO
*
r^
o
o
L^
*
fS»
^3"
rH
tD
CM
CM
O
^^
T*H
CO
rH
CO
rH
1 f*\
U7
co
«*
f*s»
CO
^
CO
«*
en
X
C
rs
CD-
• "=1"
r*1^
i p
"
CO
CM
rH
CM
• r^
CO
cn
^j-
p*^
*
CM
CO
« en
rH
CO
*i
CO
r~"4
*i
r"H
CM
tH
_ , \
• l~n
**
g~ ^
^^
Lrt
* LO
CM
Lf")
M
CO
• en
r—
CO
CO
+J
co
r—
S- CO
CO CO
> -P
•i- CO
O£.
U
C •<-
its r-
40 o
•i- 3
S- CL
. p^.
rH
<^-
. "
S
• en
co
M
r—4
en
• CM
r--
tn"
^J
* ^J
^t"
CO
_ /fc 1
• VXJ
CO
A
CO
^M
* ^^
rH
LO
M
CO
U
*(••
t-
-p
O i—
CO CO
r- CO
UJ -P
co
CO
O *i™"
C r—
m ja
a: cc:
*
'*
«
•
«
to
s-
o
3:
P"~
CO
CO
•P
t/5
to
a:
en Is*. «
r^ r^»
^^ *!j"
" "
S CO
CO rH
cn CM •
r~« r>-
*> *%
O CO
rH f**«
rH CO
•^
rH
O UO •
o cn
r-"o"
Lf5 C^
CM rH
CO l^} *
CD UO
CO *si"
** •*
CO U">
m co
^V% £VI •
r>3 v\f •
r*. r~.
M *t
co cn
csj r^*
co
CD 0 •
0 rH
f— t 00
M M
r-. o
in CM
r>—
r— CO
CO -P "flj
•p co co
CO -P
05 CO
O C CU
i~ O r***
fO _S 3
co co co
• • CO
rH
CM
"
S
CM
. . r-s
cn
A
K
CM
O
*t
CM
• • en
r--
cn
r~~
*sfr
r*.
IH"
* * CD
CO
en
I-H"
r~.
CO
cn
A
ps.
o
. . o
rH
U3
p**..
??J"
r^.
to
r"*
<8
+J 6
CO U
S W
ro
r- >
>v
flj 13' OJ
_S S- r-
H^ CO r—
• •
• *
• •
* t
• *
rco
CO
[ ^
CO
DJ
»r-
C35
i.
0
U_ P—
CO
CO CO
0) -P
to co
(ft
c to
C X
1— I—
fO
U"}
<**^
"
CO
CM
CM
p-j
cn
M
CM
t^J"
cn
cn
cn
CO
m
CO
r*>
rH
co
rH
M
CM
O
rH
co
o
. «*
px.
P^
r~l
CO
pH
**t"
^
irt
LO
^t*
r—
CO
CO
•p
co
C
J^/
ts
H^
« 0
CO
cO
"
CO
CO
to"
• CD
f^
cn
M
cn
C\J
o
*«
rH
rH
. O
O
CD
^jT
en
CO
m
• o
o
o
n
co
cn
CM
t
o
00
**
CD
rH
"^t*
M
CM
« O
O
("S,*
A
CD
LO
•sf
CO
r-*
CO r—
CO CO
•P (U
co -P
to
U
•l— tO
i- CO
4-» -P
U (O
CO -P
r— CO
UJ
T3
O 4^
•r— «i—
=D ZD
* 00 •
c^
CO
"
m
rH
• cn •
cn
cn
• ft
en
CD
CM
• rH «
O
CO
CO
tn
• CM •
cn
co
*•
rH
rH
•?t-
CO
»•
CO
T"""J
ip-}
* cn *
^
CO
n
CO
tO
rH
CO
f-m
r~-
«f—
i-" ' «£!
CO
CO r—
CO -P CO
4- CO CO
•r- +J
3: c co
0
C -P C
4. O) CO
3 C S
^{^ *f«~ ^y
f-» f^ 1 \
to to -P
3 3*5
9-35
-------
TABLE 9-9. FINANCIAL RATIOS FOR U.S. COMPANIES OPERATING
ELECTRIC ARC FURNACES
(1979)
Company .
Allegheny Ludlum Industries
Ameron
Armco
Bethlehem Steel
Cabot Corporation
Cameron Iron Works
Carpenter Technology
Ceco Steel
CF&I Steel
Chaparral Steel
Connors Steel
Copperweld
Crucible
Cyclops
Eastern Stainless Steel
Florida Steel
Ford Motor
Ingersol Rand
Inland Steel
Inter! ake
ITT
Oessop Steel
Jones & Laugh! in
Earle M. Jorgenson
Keystone Consolidated
Lonestar Steel
Lukens Steel
Marathon Steel
McLouth Steel
National Steel
Nucor
Penn-Dixie Steel
Phoenix Steel
Quanex
Republic Steel
Rob! in Steel
Sharon Steel
Sirtonds Steel
Teledyne Vasco
Timken Steel-
United States Steel
Washington Steel
Current
Ratio
2.01
2.64
2.07
1.78
2.03
2.62
2.29
2.65
1.58
1.82
2.02
1.97
2.55
1.93
2.66
2.71
1.25
2.45
1.43
1.73
1.25
2.24
1.71
2.72
2.17
2.35
1.82
1.59
1.44
1.74
1.85
1.51
1.59
2.14
1.62
1.65
1.84
3.15
1.83
2.57
1.43
1.36
Leverage
Ratio
1.35
0.91
0.90
0.96
1.36
0.95
0.42
0.79
0.88
1.48
1.26
1.06
1.04
1.40
1.26
0.76
1.26
0.94
1.06
1.10
1.68
2.41
4.10
0.42
1.25
1.38
0.76
3.04
1.46
1.20
0.82
1.63
1.25
1.10
0.85
2.64
3.32
0.96
0.16
0.34
1.04
2.94
Percent
Profit
14.74
8.67
12.89
10.44
17.62
18.13
17.96
12.60
7.21
32.29
16.21
15.19
17.44
3.13
10.50
20.18
11.22
13.64
9.92
'11.35
6.77
6.24
22.92
15.65
0.67
18.64
5.29
4.27
4.95
8.80
31.60
-0.06
6.14
18.70
8.14
2.55
22.78
18.59
21.28
14.47
-5.43
21.37
9-36
-------
TABLE 9-10. FINANCIAL RATIOS FOR SELECTED INDUSTRIES76
(1979)
Industry
Liquidity
ratio
Profit
rati o
Leverage
ratio
EAF steel firms3
Primary nonferrous metals, NEC—SIC 3339
Aluminum foundries—SIC 3361
Primary metal products, NEC-^-SIC 3399
Motor vehicles—SIC 3711
Petroleum refining—SIC 2911
1.99
1.86
"2.15
2.28
1.73
1.36
13
22
22
24
17
26
1.33
2.32
1.84
0.69
1.44
1.56
Means from Table 9-9.
9-37
-------
is in line with other industry averages. These firms should generally be
able to meet current obligations with little difficulty. The mean EAF
profit ratio is below that of all other industries presented. Finally,
there is no evidence that EAF firms are too heavily debt financed, based on
a comparison with the leverage ratios of the other industries.
9.2.5.3 Industry Trends. The purpose of this section is to highlight
recent trends in certain steel industry variables. The variable values
presented and the related discussion are intended only as an indication of
past industry performance. Projections of future performance are presented
in Section 9.1.5.4.
9.2.5.3.1 Physical output. The annual physical output of an industry
is perhaps the single best indicator of industry performance. In general,
a healthy industry is expected to increase physical production in pace with
general economic activity. As the population increases and real income
increases, the demand for industry output is expected to increase. An
annual index of shipments of domestically produced steel mill products
tonnage has been constructed and is presented in Table 9-11. In the same
table is the total industrial production index of the U.S. Board of Gover-
nors of the Federal Reserve System. Both are indexes of actual physical
output.
It is clear from Table 9-11 that during the 13 years ending in 1979,
steel production did not keep pace with total industrial production. The
average annual continuously compounded rate of growth-of total industrial
output was 3.5 percent, compared with only 1.5 percent for steel mill
products. Of the period covered, domestic steel shipments peaked in 1973.
The two-year decline in the total industrial output index in 1974-1975
marks a recession. Steel output declined as well during those years, but
has been far slower to recover than the overall economy.
9.2.5.3.2 Real value of output. Because of the effect of inflation,
observation of the value of industry output over time can be misleading.
Industry value of output is more meaningful when deflated to constant
dollars for some base year.
Table 9-12 presents the real value of output (deflated by the implicit
price deflator for Gross National Product) of all establishments classified
in SIC 3312 from 1967 to 1977. The annual growth rate for real output of
9-38
-------
TABLE 9-11.
STEEL MILL PRODUCTS AND TOTAL INDUSTRIAL OUTPUT
INDEXES (1967 = 100)28 41 7? 78
Year
Steel mill products Total industrial output
'1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
119.5
116.7
108.6
106.6
95.3
130. 5
132.8
109.4
103.7
108.2
111.9
109.5
100.0
152.5
146.1
137.1
129.8
117.8
129.3
129.8
119.7
109.6
107.8
111.1
106.3
100.0
9-39
-------
TABLE 9-12. REAL VALUE OF OUTPUT FOR SIC 3312
(10s 1967 $)
Year
Real value of output
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
23,487.6
23,423.5
22,157.2
28,382.8
22,679.4
18,922.7
18,081.9
23,360.0
20,319.8
20,251.8
19,620.6
9-40
-------
SIC 3312 was 1.8 percent during the period, well below the 2.4 percent
growth rate of real GNP.26,27
9.2.5.3.3 Steel mi 11 products prices. The producer price index for
steel mill products is compared with the GNP price deflator index in
Table 9-13. Steel prices increased at a rate very near the general rate of
inflation until 1973. Note that in 1973 the steel index is only four-
tenths higher than the GNP index with 1967 = 100. In 1974, however, steel
prices increased 26.8 percent over 1973 levels while the implicit price
deflator increased only 9,6 percent. Increases in the price of steel
continued to outpace inflation through 1979, the last year for which data
are available.
This trend, occurring from 1973 to 1979, can in part be attributed to
an important basic supply condition. Steel production is energy intensive.
OPEC-induced energy price increases beginning in 1974 raised the price of
this important input to production, which was in turn reflected in in-
creased real prices of steel.
9.2.5.3.4 Investment in new plant and equipment. In 1967, new
capital expenditures in SIC 3312 firms totaled about $1.7 billion. In
1977, new capital expenditures totaled only $1.2 billion in 1967 dollars.3
In order to determine the significance of this decline, an index has been
constructed of new investment by steel firms and is compared with an index
of new investment by all industry in Table 9-14.
Real investment in new plant and equipment by all industry increased
at a continuously compounded average annual rate of 1.5 percent during the
period 1967-1977. Real expenditures by firms in SIC 3312 declined over the
same period at an average annual rate of 3.3 percent.
The American Iron and Steel Institute believes the steel industry's
poor investment record is due to general economic conditions and government
policy and its effects, specifically, high inflation, excessive taxation,
and burdensome regulation, which discourage new capital formation.83 One
independent observer, however, blames the poor investment record not on the
inability of U.S. steel firms to invest but on their unwillingness to
compete with foreign steel by investing in new, lower cost techniques.84
9.2.5.3.5 Productivity. Productivity is meant to be a measure of
technological advance. The most common measure of productivity is output
9-41
-------
TABLE 9-13. STEEL PRICE INDEX AND GNP PRICE DEFLATOR
(1967 = 100)42 79 80
Year
Steel price index
GNP deflator
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
280.4
254.4
229.9
209.8
197.2
170.0
134.1
130.4
123.0
114.3
107.4
102.5
100.0
209.1
192.2
179.0
169.2
160.7
146.6
133.7
126.4
121.4
115.5
109.6
104.4
100.0
9-42
-------
TABLE 9-14. INDEXES OF REAL NEW INVESTMENT
(1967 = 100)3 81 82
Year
' 1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
SIC 3312
72.2
76.1
77.4
67.4
49.9
45.7
49.8
69.3
86.4
67.3
100.0
All industries
115.9
108.6
107.0
116.9
113.8
106.7
102.1
105.9
105.2
99.1
100.0
NOTE: Investment expenditures were deflated using the GNP
implicit price deflator before indexing.
9-43
-------
per employee-hour. Table 9-15 compares this measure for the steel industry
with that for all manufacturing for the period 1967-1978. Output per
employee-hour increased 28 percent for all manufacturing from 1967 to 1978
while output per employee-hour increased only 21.8 percent for the steel
industry.
Increases in labor productivity are often attributed to improvements
in capital stock. To the extent it is true that more advanced capital
results in greater labor productivity, the relatively slow growth in output
per employee-hour in steel is not surprising. As mentioned in Section
9.1.5.3.4, investment in new steel plant and equipment declined in real
terms from 1967 to 1977.
9.2.5.3.6 Exports^and imports. Much is said about the declining U.S.
steel trade balance. There can be little doubt that the domestic industry
is losing importance in the world market. Table 9-16 presents data on
exports of steel mill products by domestic producers and imports of steel
mill products by domestic consumers. During the period 1961-1979, exports
of steel mill products changed very little. Exports peaked at 6.4 million
Mg in 1970. Over this period imports of steel mill products by domestic
consumers increased rapidly, at an average annual rate of 9.5 percent. On
balance, the United States was a net importer of steel during the 1960's
and 70's. Net exports declined at an average annual rate of 14.1 percent
from 1961 to 1979.
9.2.5.4. Projections of Electric Arc Furnace Capacity Additions.
Projecting electric arc furnace capacity additions is complicated by
several considerations. These include uncertainties about steel demand in
future years, the ability of steel firms to raise capital at an acceptable
cost, the rate at which existing capital will be retired, and the relative
merits of various competing technologies.
Because projections of capacity additions per se are generally
unavailable, the first requirement is a projection of domestic steel
production. Then, given estimates of the shares of future total production
produced by each furnace type as well as capacity utilization rates,
capacity projections by furnace type can be estimated.
The projection of domestic raw steel production used .in this report is
presented in Table 9-17. This projection evidences a continuously com-
9-44
-------
TABLE 9-15.
INDEX OF OUTPUT PER EMPLOYEE-HOUR
(1967 = 100)85 86 87
Year
Steel industry
All manufacturing
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
121.8
116.0
114.5
107.6
123.5
123.5
112.7
106.2
101.3
104.0
103.5
100.0
128.0
127.2
124.2
121.2
129.3
128.3
121.5
115.2
107.9
107.4
104.7
100.0
9-45
-------
TABLE 9-16.
STEEL MILL PRODUCTS
(103 Mg)
TRADE28 41
Year
' 1979
1978
1977
1976
1975
1974
1973'
1972
1971
1970
1969
1968
1967
1966
1965
1964
1963
1962
1961
Exports
2,556
2,197
1,817
2,407
2,678
5,291
3,675
2,606
2,564
6,405
4,743
1,968
1,528
1,564
2,264
3,122
2,017
1,826
1,813
Imports
15,889
19,169
17,511
12,956
10,895
14,485
13,741
16,037
16,602
12,121
12,729
16,290
10,390
9,753
9,417
5,841
4,940
3,719
2,869
Net exports3
-13,333
-16,972
-15,694
-10,549
-8,217
-9,194
-10,066
-13,431 .
-14,038
-5,716
-7,986
-14,322
-8,862
-8,189
-7,153
-2,719
-2,923
-1,893
-1,056
^Exports minus imports.
9-46
-------
TABLE 9-17. PROJECTED DOMESTIC RAW STEEL PRODUCTION,
(106 Mg)88
Year
Production1
1990
1989
1988
1987
1986
1985
1984
1983
1982
124.6
122.1
120.9
119.7
119.7
118.5
118.5
118.5
118.5
A. D. Little projection is for finished products assuming the industry
meets current environmental requirements. A finished-to-raw yield of
75 percent is assumed for the construction of this table.
9-47
-------
pounded growth rate of 0.6 percent per year over the entire 9 years,, with
the greatest growth in the latter part of the period. Production is pro-
jected to increase at a rate of only 0.2 percent per year during 1982-1987.
Domestic raw steel production is projected to total 119.7 million Mg
in 1987. Estimates vary on what the share of future total production by
electric furnaces will be. In this report, a projected electric arc furnace
production share of 31.5 percent for 1987 is used.89,90 It is thus projected
that electric furnaces will produce 37.7 million Mg of raw steel in 1987.
Assuming a capacity utilization rate of 85 percent, electric furnace capacity
will total 44.4 million Mg in 1987.
It is estimated that electric furnace capacity totaled 39.2 million Mg
at year's end 1981.91,92 If none of the existing electric arc capacity
existing in 1981 is retired before 1987, 5.2 million Mg of new electric arc
furnace capacity will be added during the period 1982-1987; an annual
growth rate of 2.1 percent.
A summary of the electric arc furnace capacity projection for 1987 is
presented in Table 9-18. Of the total 5.2 million Mg new electric arc
capacity, it is projected that 4.8 million Mg will be met by the con-
struction of carbon steel shops and 0.4 million Mg will be met by the
construction of specialty steel shops.92
New additions of carbon and specialty capacity are expressed as model
plants in this same table. The model plants are those described in Chapter 6.
Table 9-18 shows, for example, that the projected 4.8 million Mg of new
carbon capacity could be satisfied by the construction of 29 new 90,7-Mg
(100 ton) vessel shops of the type represented by model plant 2. The
projected 0.4 million Mg of new specialty capacity could be met by the
construction of 11 22.7-Mg (25 ton) electric arc furnace/ argon-oxygen
decarburization (EAF/AOD) shops.
Projecting the number of each type of model plant that is likely to be
built is even more difficult than projecting total capacity additions. The
result that follows below from such an effort should be regarded as a
.. scenario as much as a projection.
Examination of characteristics of EAF shops coming on line in 1981 and
1982 reveals that a large proportion of the new shops have a total vessel
capacity of about 90.7 Mg; for example, shops with two 36.3-Mg vessels and
shops with two 54.4-Mg vessels. Also common but less prominent are shops
9-48
-------
TABLE 9-18. SUMMARY OF ELECTRIC ARC FURNACE CAPACITY PROJECTION
1981
1987
1982-1987
1982-1987
1982-1987
Electric Arc Furnace Capacity
Electric Arc Furnace Capacity
New Electric Capacity Additions
New Carbon Steel Shop Capacity
New Specialty Shop Capacity
-
-
-
39.2 x 10s Mg
44.4 x 106 Mg
5.2 x 106 Mg
4.8 x 106 Mg
0.4 x 10s Mg
New Capacity Additions Expressed As Model Plants
Model Plant(s)
1, 6
2
3, 5
4
Model Plant(s)
7, 8
9, 10
Carbon Plants
106 Mg Capacity 1982-1987
Per Plant No, of Plants
. 03742
.1633
. 596
.4348
Specialty Plants
128
29
8
11
106 Mg Capacity 1982-1987
Per Plant No. of Plants
. 03742
. 1633
11
2
No. Plants
Per Year
21
5
1
2
No. Plants
Per Year
2
<1
9-49
-------
with vessel capacity totaling 136.1 Mg and 272.2 Mg.93 Table.9-19 presents
a projection of new plant construction that is fairly consistent with the
size distribution of new plants coming on line in 1981 and 1982. Con-
struction of these 19 plants of various sizes would approximately satisfy
the projection of 5.2 million Mg Of additional electric arc furnace capacity
to be built between 1982 and 1987.
The accuracy of a projection can be known with certainty only after
the fact. Even at this writing there is some evidence that the projection
of 5.2 million Mg of new EAF capacity additions during 1982-1987 is low.
At the same time, however, the evidence is conflicting in general and this
projection results from a careful synthesis of information from several
sources.
9.2.6 Small Business Impacts
The Regulatory Flexibility Act (RFA) requires consideration of pro-
posed regulations on small "entities." This section briefly examines the
applicability of the RFA to the proposed NSPS on EAF shops.
The guidelines for conducting a regulatory flexibility analysis define
a small business as "any business concern which is independently owned and
operated and not dominant in its field as defined by the Small Business
Administration Regulations under Section 3 of the Small Business Act." The
Small Business Administration has determined that any firm classified in
SIC 3312 that employs less than 1,000 workers will be considered small in
regard to the Small Business Act. •
RFA guidelines clearly focus on regulations that apply to existing
rather than new sources. It is seldom possible to determine which indi-
vidual firms will construct new sources and thus be affected by the NSPS.
In the case of an NSPS on new electric arc steel plants, this problem is
very relevant. First, plants may be constructed in the future by firms
that are not currently involved in electric arc steelmaking or, for that
matter, in any manner of steelmaking. The employment of such unknown
potential producers can not be known. Second, even assuming that the only
firms that will build EAF plants in the future are those firms that
currently operate such plants, it does not logically follow that all or
even some of the new plants will be built by the existing small firms. It
is quite possible, however, that some of the projected new EAF plants will
be constructed by small firms.
9-50
-------
TABLE 9-19. PROJECTED CONSTRUCTION OF ELECTRIC FURNACE PLANTS,
1982-1987
Carbon Steel Model Plant No.
Model Plant 2--one 90.7-Mg EAF
Model Plant 4— one 136.1-Mg UHP EAF
Model Plant 3— one 272.2-Mg EAF
TOTAL
of Plants
8
5
2
15
10s Mg
Total Annual
Capacity
1.3
2.2
1.2
4.7
Specialty Steel Model Plant
Model Plants 7,8 —22.7-Mg EAF/AOD
Model Plants 9,10—90.7-Mg EAF/AOD
-• TOTAL
No. of Plants
2
2
4
106 Mg
Total Annual
Capacity'
0.07
0.33
0.40
9-51
-------
Of the 80 firms that currently operate one or more EAF shops, employ-
ment and financial data are available for only 42. Of these 42, none
employ fewer than 1,000 employees. It is quite likely, however, that some
of the remaining 38 firms do qualify as small entities. If it is assumed
that all of these 38 firms are small, the number of small firms that would
have to be affected by the regulation to constitute a "substantial" number
is 20 percent of these 38, or 7 firms. Because it is projected that any-
where from 10 to over 100 new EAF plants will be built by 1987, it is
possible that a substantial number of small firms will be affected by the
regulation.
Once it has been determined that a substantial number of small
entities may be affected by a regulation, the RFA requires a determination
of whether these impacts are "significant." The degree of economic impact
that a regulation will have on any firm building an electric arc furnace
plant will depend in part on the type of plant it builds. In general, the
"significance" of the impact of each proposed NSPS as measured by RFA
criteria is greater for smaller EAF plants than for larger ones. It is
not, however, logical to presume that small entities will construct small
EAF plants. To test the hypothesis that small firms tend to build small
plants, a correlation analysis was conducted using the data on employment
and plant size that is available for the 42 firms mentioned above. EAF
vessel capacity, the major determinant of annual steel capacity, was used
as a proxy for plant size. When a firm had one or more plants and a total
of two or more vessels, an average vessel capacity was used.
The null hypothesis that firm employment and average vessel capacity
are not correlated can not be rejected at any reasonable level of signi-
ficance. The Pearson correlation coefficient for a sample of this size at
a 5 percent level of significance is approximately 0.304. The coefficient
of correlation between employment and average vessel capacity for the
sample tested is 0.07. There is thus no significant support for the
hypothesis that small firms build small plants.
A second statistical test involved comparing the average vessel size
for the 42 firms known to be "large" with the average vessel size for the
38 firms assumed to be small. The average vessel size for large firms is
78 Mg. The average vessel size for small firms is 58 Mg. Whether or not
the 20 Mg difference between vessels of large and small firms is signi-
9-52
-------
fleant can be tested using a t-test for the difference between the means
from two samples. Using such a test, the calculated t-statistic for this
problem is 1.43. In order for the difference'between means to be signi-
ficant for a sample of this size at a 10 percent significance level, the
t-statistic must exceed 1.67. This test lends little support to the notion
the small firms build small plants. .
Despite the statistical evidence to the contrary cited above, the
investigation of the possible need to conduct a regulatory flexibility
analysis will continue on the premise that small firms will construct small
electric arc furnace plants. The RFA next requires an evaluation of whether
the economic impact of the proposed standards on small entities will be
"significant" as measured by several criteria. This evaluation follows
below.
The impact of a regulation on a small entity is judged to be signifi-
cant if the regulation causes the average total cost of production to
increase by 5 percent or more. The average total cost impacts reported in
Table 9-20 indicate that neither regulatory alternative 2 or 3 causes an
increase in ATC from baseline alternative 1 as high as 5 percent for any
model plant. The most severe impact is 0.31 percent, for model plant 7
under regulatory alternative 3. It can thus be safely concluded that the
impacts are not "significant" from an average total cost standpoint.
The second criteria that renders an alternative's impact "significant"
relates compliance costs to sales for small versus large entities. If .
compliance costs as a percent of sales for small entities is at least
10 percentage points higher than compliance costs as a percent of sales for
large entities, the impact is judged to be significant.
It is unclear whether "sales" refer to total sales by the entity
(firm) or total sales by the affected facility. Since firm sales data are
unavailable for many of the companies, total plant sales are used. The
1981 market price of carbon or specialty steel billets is multiplied by
each plant's estimated annual output of billets to arrive at an estimate of
each plant's annual sales. The total annualized cost of compliance for
each plant under each alternative is expressed as a percent of its own
annual sales in Table 9-21.
The total annualized cost of compliance as a percent of sales is
generally higher for small plants than for large plants. The differences,
9-53
-------
TABLE 9-20. AVERAGE TOTAL COST IMPACTS ASSUMING
6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
(1981 $/Mg)
Regulatory Impact From Basel
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
•
1
1981$
19.13
19.13
12.17
5.06
7.38
5.32
20.26
20.26
36.27
31; 00
22.17
20.83
percent
6.35
6.35
5.49
2.72
3.73
2.91
7.12
7.12
5.70
4.87
4.23
3.97
Regulatory
1981$
0.14
na
0.22
0.14
0.12
0.15
0.16
na
0.81
0.22
0.36
0.22
Alternative
2
percent
0.04
na
0.09
0.07
0,06
0.08
0.05
na
0.12
0.03
0.07
0.04
inea
3
1981$
0.59
-0.64
0.37
0.50
0.30
0.53
0.62
-0.68
2.09
1.06
1.08
1.04
percent
0.18
-0.20
0.16
0.26
0.15
0.28
0.20
-0.22
0.31
0.16
0.20
0.19
almpact of baseline regulatory alternative 1 is measured from no control.
bTotal Furnace Enclosure Option
NOTE: na = not applicable.
9-54
-------
TABLE 9-21. TOTAL ANNUALIZED COST OF COMPLIANCE ABOVE BASELINE
AS A PERCENT OF ANNUAL PLANT SALEC
Model
Plant
1
la
2
3
4
5
6
6a
7
8
9
10
Regulatory
2
0.02
NA
0.04
0.03
0.02
0.03
0.02
NA
0.10
0.02
0.04
0.02
Alternative
3
0.11
-0.15
0.07
0.10
0.59
0.11
0.12
-0.15
0.25
0.12
0.12
0.11
Total furnace enclosure option.
9-55
-------
however, are not great. The total annualized cost of compliance as a
percent of sales is never as much as one-half of one percentage point
greater for a small plant than for a large one. The small business impact
of these alternatives is certainly not significant by this measure.
A third criterion to measure the significance of an impact on small
firms compares the capital cost of compliance with the capital available to
small firms. It is extremely difficult, if at all possible, to determine
how much capital is available to a firm. With financial data on small
firms unavailable, the problem is magnified.
A reasonable approach is to recognize that the capital available to a
small firm building a new EAF plant at least equals the capital cost of the
plant .itself. The capital cost of compliance with each proposed regulatory
alternative expressed as a percent of the capital cost of the plant itself
is presented for each plant in Table 9-22. The capital cost of compliance
with regulatory alternative 2 is well under two percent of plant capital
cost for all model plants. For regulatory alternative 3, the greatest
compliance capital cost as a percent of plant cost is 3.68 percent (model
plant 5).
It is the conclusion of this examination that capital costs of com-
pliance do not represent a "significant" portion of capital available to
small entities.
A regulation is considered to have a significant impact on small
entities if it is likely to result in closures of small entities. It is
quite safe to conclude that the imposition of a New Source Performance
Standard on electric arc steelmaking will result in no closures of firms of
any size. Indeed, a more likely effect is an extension of the economic
life of existing firms by raising the market price of their output, steel,
while not increasing their costs of production (see Section 9.2.6).
Considering all of the significance criteria, it is concluded that the
proposed standards would not have a significant .impact on small entities.
Average total cost impacts are all under one-half of one percent. Total
annualized cost of compliance as a percent of sales is not significantly
higher for small plants than for large plants. Capital costs of compliance
relative to plant costs are all quite small, and the standards are unlikely
to result in closures of small firms. Because anticipated impacts on small
9-56
-------
TABLE 9-22. COMPLIANCE CAPITAL COST ABOVE BASELINE AS A PERCENT OF
BASELINE PLANT CAPITAL COST
Model
Plant
1
la
2
3
4
5
6
6a
7
8
9
10
2
0.57
NA
0.62
0.98
0.48
1.58
0.91
NA
0.33
0.34
0.32
0.27
Regulatory Alternative
3
0.92
-0.23
1.03
2.30
0.96
3.68
1,37
-0,46
0.99
0.75
0.94
1.03
Total furnace enclosure option.
9-57
-------
entitles 'are insignificant by all measures, a Regulatory Flexibility
Analysis is not conducted for these standards.
9-58
-------
9.3 ECONOMIC IMPACTS OF REGULATORY ALTERNATIVES
This section presents the estimated impacts of the regulatory alterna-
tives for electric arc process steel plants. As described in Chapter 6, 10
types of model plants are used to represent typical EAF and EAF/AOD
facilities that might be constructed by the industry in the future.
Three regulatory alternatives are considered. Regulatory alternative 1
is the baseline case from which impacts of the two proposed, increasingly
more stringent alternatives are measured. '
Section 9.2.1 summarizes the range of estimated maximum impacts that
could result from the NSPS alternatives and presents anticipated impact
estimates. Section 9.2'.2 presents the theoretical model of discounted cash
flows employed to compute both net present value and price impacts.
Section 9.2.3 presents the full range of maximum impacts on project net
present value assuming full cost absorption, and the full range of maximum
price impacts assuming full cost pass-through. Section 9.2.4 employs a
refined model to compute a single set of price impacts and associated
output, employment, and import impacts for each alternative. The issue, of
capital availability is addressed in Section 9.2.5. Section 9.2.6 presents
estimates of maximum-total costs of industry compliance.
9.3.1 Summary
Control of fugitive emissions under regulatory alternative 2 or 3
imposes additional capital costs on the construction of an EAF project and
increases its annual operating cost. It is estimated that regulatory
alternative 2 would increase the cost of producing semifinished carbon
steel by between $0.12 and $0.22 per Mg ($0.11-$0.20 per ton). These
changes represent impacts of from 0.04 to 0.09 percent. The impact of
alternative 2 on the cost of producing specialty steel is estimated to
range from $0.22 to $0.81 per Mg ($0.20 to $0.74 per ton). These
impacts are betweeen 0.03 and 0.12 percent.
It is estimated that under regulatory alternative 3, cost impacts
could vary.from $-0.68 to $0.62 per Mg ($-0.62 to $0.56 per ton) for carbon
steel and from $1.04 to $2.09 per Mg ($0.95 to $1.90 per ton) for specialty
steel. These impacts are all less than 0.32 percent.
Assuming that producers pass all increases in production costs on to
consumers, the increases in production costs would cause corresponding
impacts on 1987 steel prices. These estimated price impacts could in turn
9-59
-------
lead to Impacts on 1987 domestic steel output, industry employment, and
steel imports. These impacts are presented in Section 9.2.3.
The economic impacts that would actually result from the NSPS depend
on which projects are pursued by industry participants and the .supply and
demand conditions prevalent in the market. The conclusion of this analysis,
which is based on the premise that model plants 4 and 7 most closely typify
future EAF and EAF/AOD construction, is that impacts would occur as increases
in steel prices, reductions in industry employment opportunities, and
increases in steel imports.
It is estimated that the price of semifinished carbon steel would be
$0.12 per Mg ($0.11 per ton) higher in 1987 under alternative 2 than it
would be under baseline alternative 1. The impact on the price of specialty
steel is an estimated $0.81 per Mg ($0.74 per ton). These price increases
represent impacts of 0.06 percent and 0.12 percent respectively.
Steel users might purchase less domestic steel and more foreign steel
in 1987 as a result of the increases in domestic steel prices. Domestic
steel shipments would be down an estimated 150,000 Mg (165,000 tons) under
alternative 2--an impact of 0.13 percent. Imports could be approximately
19,000 Mg (21,000 tons) greater—an impact of 0.09 percent. The impact on
domestic steel production would cause a reduction in employment opportunities
of about 0.12 percent, or about 542 jobs, in 1987. The estimated total
cost of regulatory alternative 2 in 1987 is $1.381 million.
It is estimated that regulatory alternative 3 would raise the price of
semifinished carbon steel by $0.30 per Mg ($0.27 per ton) above its baseline
1987 price. The anticipated impact on the price of specialty steel of
alternative 3 is $2.09 per Mg ($1.90 per ton). These price increases
represent impacts of 0.15 percent and 0.31 percent, respectively.
Domestic steel shipments would be down an estimated 370,000 Mg
(407,000 tons) in 1987 under regulatory alternative 3—an impact of
0.31 percent from baseline. This impact on domestic steel production would
cause a proportionate decrease in 1987 employment opportunities, about
1,387 jobs. Steel imports in 1987 would be approximately 40,000 Mg (44,000
tons) higher under alternative 3 than under regulatory alternative 1—an
impact of 0.20 percent. The estimated total cost of regulatory alternative 3
in 1987 is $2.575 million.
9-60
-------
Electric arc steel making is apparently a profitable activity. This
is indicated by two kinds of evidence. First, construction of new EAF
facilities and the replacement of older steel making technologies by EAF
shops is brisk. This indicates that the industry itself views EAF steel
making as profitable. Second, the engineering cost data for the 10 model
facilities examined in this study suggest strongly that steel can be produced
at an average total cost significantly below market price. In light of
this observation, it seems unlikely that the industry impacts reported
above would significantly affect the future growth of electric steel making.
9.3.2 Methodology
9.3.2.1 The Discounted Cash Flows Approach. The economic impacts of
New Source Performance Standards on electric arc process furnaces are
estimated using a discounted cash flows (DCF) analysis. Under this approach,
the expected future annual net revenue flows generated by an investment in
an electric arc furnace project are discounted at an appropriate interest
rate and summed to determine the net present value of the project. This
section describes the DCF theory and methodology in some detail.
An investment is expected to generate a series of cash inflows and
outflows during its lifetime. The net cash flow in the first year (year
zero) is negative as the cash outflows of the initial investment are not
offset by any cash inflows. After the project begins production, it will
generate a stream of cash inflows in the form of revenues from the sale of
its output and depreciation of the capital equipment, and cash outflows in
the form of operating expenses. Beginning with year one and continuing
throughout the lifetime of the project, annual cash flows are expected to
be positive, but need not be. Although cash inflows and outflows may occur
at any time, it is assumed that they will take place at the end of the
year. It is also assumed that the only investment in the project takes
place at the end of year zero and is followed by a series of net cash
inflows. These assumptions guarantee a'unique rate of return for each
project.94
The cash outflow in the first year may be expressed:
9-61
-------
YQ = (FCC + WC) - (TCRED
FCC)
(9-1)
where FCC is the principal value of the capital investment, WC is the value
of the working capital and TCRED is the percentage of the capital invest-
ment resulting in a direct tax savings, or tax credit, to the firm.
The project generates its first revenues at the end of its first year
of production (year two). The net cash flows in this and succeeding years
can be expressed:
C'
Yt = (Rt - Et) (1 - T) + DtT t = 1, . . . , N . (9-2)
The first term in equation (9-2) represents the net after- tax inflows of
the project generated by the sales of the output. Total revenues in year t
can be expressed:
Rt = (P • Q)t (9-3)
where P is the per-unit price of .output and Q is the quantity of output
sold during the year. Total operating costs in year t can be expressed:
Et = ((V • Q)
F)
(9-4)
where V is the per-unit variable cost of production and F is the fixed
annual cost of operating the project. Variable costs include expenditures
on material inputs, labor and energy. Fixed costs include such expenses as
site rent (explicit or implicit), insurance, and administrative overhead.
Only net revenues are subject to corporate income taxes (T). Thus,
annual total operating cost is deducted from total revenue to yield the
taxable net revenue. The firm's after- tax net revenue in year t is thus
the first term in equation (9-2).
Federal income tax laws allow a deduction for depreciation of the
capital equipment. This deduction reduces income tax payments and is thus
treated as a cash inflow in the second term in equation (9-2). In this
analysis, the straight-line method of capital depreciation is assumed.
Thus,
DtT = (FCC/N) • T (9-5)
where N is the project life in years. The salvage value of the plant is
assumed to be zero.
9-62
-------
The net cash flows represented by equation (9-2) occur at the end-of
the first through the Nth years, where N is the life of the project. An
additional cash inflow actually occurs at the end of the Nth year when the
working capital, WC in equation (9-1), is recovered at the end of the
project. This inflow is in fact accounted for in the analysis.
The investment project is thus represented as a cash outflow in the
first year followed by N cash inflows and outflows in successive years.
Cash flows that occur over a future period must be discounted by an appro-
priate interest rate to reflect the fact that a sum of money received at
some future date is worth less than an equal sum received today. The
discounted value of this sum received in the future is called its present
value. The discount factor is a function of both time and the interest
rate, and can be expressed as:
DFt = (1 + r)"1 (9-6)
where DF is the discount factor for year t and r is the interest rate.
An understanding of the discount factor and the selection of an appro-
priate rate of interest in practice is important. The interest rate r in
equation (9-6) can be viewed as the cost to the firm of acquiring funds for
the project. The firm can acquire funds in essentially any combination of
three ways. It can issue bonds, sell stock, or utilize currently held
liquid assets. There is a cost associated with each method. Interest must
be paid on bonds, dividends on stock, and there is an opportunity cost
associated with utilizing internal funds. In the absence of specific
information on how a project would be financed, a weighted average cost of
capital can be used.
The sum of the discounted cash flows from a project over its life is
the net present value (NPV) of the project. The NPV of a project can be
expressed most simply as:
N
NPV = I
t=l
'- Y
(9-7)
where all terms are as defined above. The net present value of a project
is thus the sum of the discounted after-tax net revenues minus the initial
capital expenditure.
9-63
-------
As discussed previously, the objective of a firm is to maximize
shareholders' wealth. This is accomplished by maximizing the discounted
stream of cash flows to the firm. The NPV of an investment is in fact the
discounted value of cash flows from the project available as dividends
payable to stockholders. Thus, the objective of the firm is to invest in
any project with a positive NPV or, where funds are limited, to invest in
the project with the highest positive NPV.95596 Equation (9-7) can be used
to calculate the net present value of each plant under any given regulatory
alternative and to evaluate each plant type according to the decision
criterion.
In Section 9.1.4.4 it is concluded that recent pricing behavior in the
domestic steel industry does not strongly support either a monopoly or an
oligopoly market structure. Increasingly, steel prices appear to be more
the result of underlying supply and demand conditions than the result of
price-setting behavior of any market participant. Thus, in this analysis,
competitive conditions are assumed to prevail. In particular, firms are
assumed to be price takers.
In this analysis, two types of impacts that could result from the
proposed NSPS on EAF facilities are estimated. Equation (9-7) is utilized
as stated to estimate the NPV of each project under each regulatory alter-
native. A different form of the same equation is used to estimate the
average total cost of EAF steel production for 'each plant under each
regulatory alternative. This estimate is used in a discussion of steel
price impacts.
9.3.2.2 Net Present Value Impact Methodology. In general, an NSPS
increases average total cost (ATC) of production. This can result from
additional capital costs for pollution control equipment, increased operat-
ing costs, or both. If market demand for the project is perfectly elastic,
a situation of full cost absorption prevails—the increase in ATC cannot be
passed forward in the form of increased product price. ATC is an expense
deducted from gross revenue in equation (9-7). Thus, the NSPS will reduce
the NPV of a project if unit price remains unchanged. In a full cost
absorption situation, the effect of an NSPS is to reduce the present value
of dividends payable to stockholders.
9-64
-------
The nature of the NPV impact on a given stockholder depends on whether
the stock was purchased before or after the impact was perceived by the
market as a whole. The recipients of the loss in NPV are the owners of
stock at the time the loss information becomes known. Once it is perceived
that the increased ATC resulting from the NSPS will reduce future net
income streams, prices of outstanding stock decline until the new, lower
present value of the dividends yields the same rate of return available on
other shares of stock.97 Holders of outstanding stock at the time this
price adjustment occurs suffer a one-time wealth loss.
Once stock prices have adjusted dovmward, the dividend payments yield
the same rate of return on these shares as dividend payments on totally
unaffected stocks yield. The holders of NSPS affected stock at the time
the information is learned receive this same market rate of return on their
new, lower priced stock. Moreover, investors who purchase EAF steel firm
stock after the market has accounted for the NSPS impact suffer neither a
loss in wealth nor a loss in their rate of return. The lower price they
pay for stock compensates them for the lower dividend payments they antici-
pate.
In this analysis, equation (9-7) is utilized to calculate NPV impacts
resulting from the proposed regulatory alternatives. The net present
values of the 10 model plants under regulatory alternatives 1, 2, and 3,
are compared to the NPV's of these same plants with no fugutive emissions
control. The loss in NPV occurring as a result of the NSPS under a
situation of full cost absorption is interpreted as the impact on stock-
holders' wealth.
9.3.2.3 Steel Price Impact Methodology. An increase in the ATC of
EAF steel will be exactly reflected in higher prices in a market with
perfectly elastic supply or perfectly inelastic demand.
Price impacts of the three regulatory alternatives are calculated
assuming that the price change is exactly equal to the change in ATC. This
change in ATC is in turn calculated using a revised form of equation (9-7),
derived below. In a market situation where price equals average total
cost, the NPV of a project is equal to zero. That is, dividend payments in
excess of those required to raise capital for the project, which themselves
yield the rate r in equation (9-6), are zero. When equation (9-2) is
9-65
-------
substituted into (9-7) and (9-7) is set equal to zero, (9-7) can be written:
N
DFt = Yo •
(9-8)
If revenues and expenses are the same over all periods, equation (9-8) can
be written
(R - E) (1 - T) I OF.
t=l
N
I D,TDF,. =
* •
(9-9)
recognizing that (1 - T) is a constant. The sum of the discount factors as
t ranges from 1 to N can be written:
N
v-N-
F = X DFt = [1 - (1+ r) "]/r .
t=l r
Substituting equations (9-5) and (9-10) into (9-9) yields:
(R - E) (1 - T) F + (FCC/N) • T • F = Y,
(9-10)
(9-11)
Substituting equations (9-3) and (9-1) into (9-11) and rearranging further
yields:
_ (FCC + WC) - (TCRED
(1 - T) •
FCC) - DSL
(9-12)
where DSL = (FCC/N) • T • F, and represents the present value of the tax
savings due to straight-line depreciation of the fixed capital.
Finally, Q and E can be moved to the right-hand side of (9-12) to
yield:
- (FCC + WC) - (TCRED • FCC) - DSL E
(1 - T) F • Q Q '
(9-13)
Where P = ATC, equation (9-13) calculates average total cost. The first
term in (9-13) is capital cost per unit including allowances for the tax
credit and depreciation, and the second term is operating cost per unit.
The cost per megagram of EAFisteel is calculated using equation (9-13)
for each model project uncontrolled and under each regulatory alternative.
Price impacts are estimated assuming that any additional cost resulting
from a proposed NSPS is passed forward in a higher price.
9-66
-------
In Section 9.2.3, maximum impacts on net present value, price, and
other relevant variables are presented for all plants under each regulatory
alternative. In 9.2.4, an assessment is made of actual expected impacts
based on a somewhat refined model and certain simplifying assumptions.
9.3.3 ' Economic Impacts of Regulatory Alternatives
This section presents ranges of maximum impact estimates for the
regulatory alternatives. Maximum impacts on net present value are pre-
sented for the case of full cost absorption. The presented range of
impacts on steel prices, output, employment, and imports results from a
situation of full cost pricing (full cost pass-through to consumers).
Table 9-23 contains cost data in 1981 dollars for each model plant
uncontrolled and under each regulatory alternative. The annual operating
costs include an estimate for the value of scrap used by the steel furnace
equivalent to $94.16 per Mg (1981 dollars) of scrap used in carbon shops
and $273.72 per Mg (1981 dollars) of scrap used in specialty shops.98 A
steel/scrap yield of 85 percent is assumed. Unlike the annualized costs in
Table 8-1* the annual operating costs in Table 9-23 do not include a capital
recovery allowance. Capital costs in Table 9-23 are employed in the FCC
term of equation (9-1) while annual operating costs are employed in the E
term in equation (9-4).
Table 9-24 lists the parameter values actually employed to compute NPV
and ATC. A value of working capital equal to 50 percent of the value of
the fixed capital is assumed. The average working capital-to-fixed asset
ratio for steel firms is 83 percent.76 However, it is believed that this
average ratio is probably too high for an incremental investment, and a.,,,
50 percent ratio has been arbitrarily selected. Currently, the investment
tax credit is 10 percent of the capital cost. The federal and average
state marginal corporate tax rates are 46 percent and 6 percent, respec-
tively." Since state taxes are deductible from taxable income for federal
tax purposes, the overall effective tax rate is 49 percent. The capital
budgeting decision is based on an expected economic project life of 15 years.
Two separate interest rates are employed. The weighted average cost of ..
capital to the steel industry has been estimated to be 6.2 percent.100 An
alternative interest rate of 10 percent is employed to investigate the
sensitivity of the estimate to this parameter. The steel prices used are
Bureau of Labor Statistics estimates for steel billets.98
9-67
-------
UJ
CO
*~
cu
•r™
•P
(0
i-
•P
>
O
•p
CC
C
a
>
CO
CM
iH
£_
•P
C
O
O
o
05
n c
i™" *r~
(0 -P -P
3 to to
C S- O
c cu o
«d-,i~-cM«5j-o
^^ ^^ p*^ f^ c^ r^^ ^™i ^^ ^^ ^* if) utf
oo oo ^^ oo tjn ^^* oo co f***™ r^* »o ^o
CMCOU3CO rHrHU3«jO
• [ |>p^ ^j p^. p^ co r*^* if^ r^* r*^* oo en
co r^*1 ^G T™H co ^o t o L^ co 1^5 r*** en
CMCOCOCSJ iH t— 1 IO LO
en cncor^-toen CDCMCOO
CM cnoentDtH cocoLnih
• ••••• .•»•
co to co tD r^^ co r"*** r*^ co co
CMOOIOCO rHr-HlOlO
oo CM r*» . oo o ^* unotn en
r-. Lnr^«,toc\jto e\iLf)<*Ln
co «^-o:cn-coto 001 — CO«M
CMCO^COCM tHi-HLDtn
co r>-oocnm CMtotor-«
LT> ocsicMcnun ^tr-tcnto
CM CT> . O CTi tO rH COCO'^'SJ-
• ••»•• «•••
CMCOVOCO r- 1 r- 1 tO tO
co ' r^- r**« cr> ^i" co en "^* oo to
f~» CO «^ ^t" ' 00 m rH «* CM «T
-CMCOCOCM tHrHLrt'LT)
cs ooooo ooog
CO f-. CO CO to t— 1 OOOOr- IrH
en o-4-^oen COOOCMCM
r^^ y3 t^ «ri* t^3 f^^ IJ^ (*O C\l O\l
CMCOtOCO i—frHtOtO
O OOOOO OOOO
o o to csj o in to to o o
CMCMCOrH iHrH'^''*'
,
(O (tf O
i— Ir— ICsJCO«^-LniOtOr-»OOO"irH
cu
c >
0 0
•r- U
P a>
a. o;
o
CO (0
i- -P
3 *r*"
to Q.
O <0
r— CJ
U
c
-------
TABLE 9-24. MODEL PARAMETER VALUES76'98'99'100
Parameter
Value
Working capital (WC)
Federal investment tax credit (TCRED)
Federal corporate tax rate (FT)
State corporate tax rate (ST)
Project life (N)
Interest rate (r)
Steel price (P)
0.5 x FCC
0.1 x FCC
46 percent
6 percent
15 years
6.2 percent and 10.0 percent
$412.16 per Mg--carbon
$715.18 per Mg—specialty
9-69
-------
If the recent past is any indicator of things to come, the general
price level will rise each year through the future. This continual in-
crease in the overall level of prices is called inflation. During a period
of inflation, some prices rise more rapidly than others while the prices of
some goods actually decline. The change in the price of an individual good
is called its nominal, price change. The change in the price of an individual
good relative to the rate of inflation is called its real price change. If
the rate of inflation is 10 percent and the nominal price of, for example,
a ton of steel increases by 12 percent, the real price of steel has increased
only 2 percent. It should be clear then that nominal prices tend to be
less stable than real prices.
Historically, real prices of individual goods have risen far more
slowly than nominal prices. Referring back to Table 9-13 in section 9.1,
nominal steel prices have increased at an average annual rate of 8.6 per-
cent while real steel prices have increased at an average annual rate of
only 2.5 percent. This difference is attributed to a rate of inflation of
6.1 percent over the same period.
Project net present values depend in large part on future cash flows,
which in turn depend on prices received for output and prices paid for
inputs. Historically, the real price for a particular product has tended
to change similarly with the weighted average of the real prices of its
various inputs. For this reason, net present value calculations in real
.terms are little different when they are calculated using projected nominal
cash flows than when they are calculated assuming constant real prices for
inputs and output. In this analysis, an assumption of constant real prices
for both inputs and outputs is made.
A second important factor in net present value calculations is the
discount rate. This parameter is perversely affected by the rate of inflation,
with implications for net present value computation. Again, an important
distinction must be made between nominal and real values. The nominal rate
of interest is comprised of a real rate of interest plus an inflation
premium. The inflation premium is intended to compensate the lender for
the fact that a dollar he receives in repayment is worth less than a dollar
he initially lent. In periods of significant inflation, nominal interest
rates, which are what we commonly hear quoted in the media, are considerably
higher than real interest rates. Because future cash flows are assumed to
9-70
-------
remain constant in real terms, as described above, the net present value
computation should incorporate a real discount rate. The 6.2 percent and
10.0 percent discount rates used in this analysis are real discount rates,
which explains why they appear low relative to current market rates, which
are expressed in nominal terms.
An alternative methodology would be to employ nominal values for both
cash flows and discount rates, but the net present value result would be
little changed. While inflated cash flows would increase the nominal net
present value, the higher nominal discount rate would offset this effect,
yielding a very similar net present value estimate.
9.3.3.1 Net Present Value Impacts. Recall that the NPV of a project
is the discounted present value of total revenue minus total costs over the
life of the project, where total costs include capital recovery. The
project NPV is equivalent to stockholders' wealth attributable to the
project.
The NPV of each model project under each regulatory alternative is
presented in Table 9-25. All 10 EAF shops modeled yield positive NPV's
under all regulatory alternatives. Larger shops generally yield higher
NPV's. This is attributable to apparent plant economies of scale, to be
discussed in Section 9.2.3.2.
It is evident from Table 9-25 that plant NPV's become lower as
regalatory alternatives become more stringent. The reason for this is
simple: assuming no price changes, the control of secondary emissions—
imposes additional capital and operating costs while leaving the value of
the marketable output, and hence revenues, unchanged. It must be remem-
bered, however, that the controlled plants yield an additional output--
reduced emissions.
There is one apparent exception to the rule that more stringent
controls result in lower NPV's. Model plants 1 and 6 yield higher NPV's
under alternative 3 than at baseline when they are controlled with total
furnace enclosures. This is the result of the lower operating costs
associated with this control option.
Uncontrolled carbon shop (model plants 1, 2, 3, 4, 5 and 6) NPV's
range from a low of $16.524 million to a high of $546.409 million. When
these same plants are required to meet baseline alternative 1, NPV's range
9-71
-------
TABLE 9-25. PROJECT NET PRESENT VALUES ASSUMING
6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
(106 1981 $)'
Regulatory -Alternative'
Model
Plant
1
la
2
3
4
5
6
6a
7
8
9
10
No Control
16.524
16.524
124. 064
537.019
372.263
546.409
19.070
19.070
11.747
11.747
124.477
'124.477
1
(Baseline)
13.667
13.667
116.134
524.984
359.463
533.767
16.045
16.045
6.332
7.119
110.032
110.905
2 : •
13.646
na
115.993
524.658
359.261
533.417
16.021
na
6.211
7.086
109.796
110.765
3
13.579
13.762
115.890
523.789
358.947
532.502
15.952
16.146
6.020
6.962
109.322
110.230
aTotal Furnace Enclosure Option
NOTE: na = not applicable.
9-72
-------
from $13.667 million to $533.767 million. Under proposed alternative 2,
NPV's range from $13.646 million to $533.417 million. The range of NPV's
for alternative 3 is $13.579 million to $532.502 million.
Uncontrolled specialty steel shop (model plants 7, 8, 9 and .10) NPV'$
range'from $11.747 million to $124.477 million. When these same plants are
required to meet baseline alternative 1, NPV's range between $6.332 million
and $110.905 million. Under proposed alternative 2, NPV's are between
$6.211 million and $110.765 million. Under regulatory alternative 3 specialty
shop NPV's are between $6.020 million and $110.230 million.
Table 9-26 presents the maximum changes in NPV that could occur from
the various regulatory alternatives. The loss in NPV should be interpreted
as the reduction in wealth that affected stockholders would experience.
The impact would be felt as a one-time reduction in affected stock prices.
Moving from no control to baseline alternative 1 results in NPV losses
between $2.86 million and $12.80 million for carbon shops and between
$4.63 million and $14.44 million for specialty shops. These losses represent
changes from no control of as little as 2.24 percent for model plant 3 to
as high as 46.09 percent for model plant 7.
Both carbon and specialty EAF steel plant NPV's are little different
from baseline (regulatory alternative 1) in absolute terms under alter-
native 2 and 3. The largest NPV change from baseline is the $1.26 million
reduction experienced by model plant 5 under regulatory alternative 3. As
indicated in Table 9-26, some plant NPV's actually increase under the more
stringent controls due to reductions in operating costs of total furnace
enclosures and reductions in capital costs of common control devices.
Net present value impacts from baseline are also modest in percentage
terms. In no case does a plant's NPV change by more than 5 percent, and
reductions of less than 1 percent are indicated for most shops.
Tables 9-27 and 9-28 present analogous NPV impacts using a weighted
average cost of capital of 10.0 percent. Note that in all cases project
NPV is lower than when a weighted average cost of capital of 6.2 percent is
employed. This is because the present value of future net revenues declines
as the cost of capital rises. However, neither the magnitude changes nor
the percentage changes are significantly different when a weighted average
9-73
-------
TABLE 9-26. NET PRESENT VALUE IMPACTS ASSUMING
6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
Regulatory Impact From Basel
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
•
10b
1981$
-2.86
-2.86
-7.93
-12.04
-12.80
-12.64
-3.03
-3.03
-5.41
-4.63
-14. 44
-13.57
1
percent
-17.29
-17.29
-6.39
-2.24
-3.44
-2.31
-15.86
-15.86
-46.09
-39.39
-11.60
-10.90
inea
Regulatory Alternative
10b
1981$
-0.02
na
-0.14
-0.33
-0.20
-0.35
-0.02
na
-0.12
-0.03
-0.24
-0.14
2
percent
-0.16
na
-0.12
-0.06
-0.06
-0.07
-0.15
na
-1.91
-0.46
-0.21
-0.13
10e
1981$
-0.09
0.10
-0.24
-1.20
-0.52
-1.26
-0.09
0.10
-0,31
-0.16
-0.70
-0.68
3
percent
-0.64
t
0,70
-0.21
-0.23
-0.14
-0.24
-0.58
0.63
-4.94
-2.21
-0.64
-0.61
9
Impact of baseline regulatory alternative 1 is measured from no control.
K
Total Furnace Enclosure Option
NOTE: na = not applicable.
9-74
-------
TABLE 9-27. PROJECT NET PRESENT VALUES ASSUMING
10.0 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
(106 1981 $)
Regulatory Alternative
Model
Plant
1
la
2
3
4
5
6
6a
7
8 .
9
10
No Control
10.797
10.797
91.825
417.617
284. 386
427.908
13.644
13.644
4.496
4.496
84. 222
84.222
1
(Baseline)
7.959
7.959
84.089
406.458
272.152
415.948
10.556
10.556
-0.987
-0.114
69.694
70.661
2
7.925
na
83.928
406,101
271.929
415.551
10.517
na
-1.103
-0.161
69,451
70.504
. '"• 3
7.862
8.041
83.814
405.279
-.- 271.617
414.667
10.453
10.646
-1.294
-0.282
68. 973
69.947
Total Furnace Enclosure Option
NOTE: na - not applicable.
9-75
-------
TABLE 9-28. NET PRESENT VALUE IMPACTS ASSUMING
10.0 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
10s
1981$
-2.84
-2.84
-7.74
-11.16
-12.23
-11.96
-3.09
-3.09
-5.48
-4.61
-14. 53
-13.56
Regul
1'
percent
-26.29
-26.29
-8.42
-2.67
-4.30
-2.80
-22.63
-22.63
-121.95
-102.55
-17.25
-16.10
atory Impact From Baseline
Regulatory Alternative
2
106
1981$
-0.03
na
-0.16
-0.36
-0.22
-0.40
-0.04
na
-0.12
-0.05
-0.24
-0.16
percent
-0.42
na
-0.19
-0.09
-0.08
-0.10
-0.37
na
11.75
40.17
-0.35
-0.22
3
106
1981$
-0.10
0.08
-0.28
-1.18
-0.53
-1.28
-0.10
0.09
-0.31
-0.17
-0.72
-0.71 '
percent
-1.21
1.03
-0.32
-0.29
-0.20
-0.31
-0.98
0.85
31.18
146.64
-1.03
-1.01
almpact of baseline regulatory alternative 1 is measured from no control.
Total Furnace Enclosure Option
NOTE: na = not applicable.
9-76
-------
cost of capital of 10.0 percent is used. Note that the NPV's of model
plants 7 and 8 are negative under each control option when a weighted
average cost of capital of 10 percent is employed.
To reiterate, these estimated changes in NPV are maximum impacts.
They occur only in the extreme case of full cost absorption. The other
extreme, the case of full cost pass through, is examined, in Section 9.2.3.2.
9.3.3.2 Steel Cost Impacts. The purpose of this section is to present
NSPS impacts on the average total cost of producing steel in each of the
model plants. Equation (9-13), which is derived in Section 9.2.2.3, is
used to calculate the average (per megagram) total cost (ATC) of producing
steel billets in each plant type under each alternative. The difference in
the ATC of producing steel in a given plant under a more stringent emis-
sions control alternative is attributable to the added capital and operating
costs of the control equipment. In the extreme case where the entire
increase is passed forward in higher steel prices, these changes represent
maximum steel price impacts.
The average total cost in 1981 dollars of producing one megagram of
steel billets in each model plant under each regulatory alternative is
presented in Table 9-29. Uncontrolled carbon steel shops can produce steel
billets at a total cost per megagram of between $186.35 and $301.48. This
spread of approximately $120 per mg in ATC is an indication of significant
economies of scale to be realized in electric arc furnace "steel shops, The
highest ATC of production is realized in a 22.7-Mg vessel shop with a
molten steel capacity of only 37,420 Mg per year. One Mg of carbon steel
billets can be produced for $115.13 less in a shop with a 272.2-Mg vessel
and an annual capacity of 596,000 Mg of raw steel.
The 1981 producer price for carbon steel billets is $412.16 per Mg.
The spread between market price and the ATC of production in these shops
varies from $111 to $226. This difference is the reason that these model
plants all yield positive net present values. ,
The 1981 producer price for stainless steeT billets is $715.18 per Mg.
The specialty steel shops modeled here, if uncontrolled, can produce these
billets at a cost of $524.15 to $636.50 per Mg. Again, this spread of $78
to $191 between price and ATC accounts for the profitabilty of these plants,
as measured by NPV.
9-77
-------
TABLE 9-29. AVERAGE TOTAL COST ASSUMING 6,2 PERCENT
WEIGHTED AVERAGE COST OF CAPITAL01
(1981 $/Mg)
Regulatory Alternative
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
No Control
301. 48
301.48
221.77
186.35
197.60
182.40
284.42
284.42
636.50
636.50
524. 15
524.15
1
(Baseline)
320.62
320.62
233.94
191.41
204.97
187.72
304.69
304. 69
672.76
667.49
546.32
544.98
2
320.76
na
234. 15
191.55
205.09
187.86
304.85
na
673.58
667.71
546.68
545.20
3
321.21
319.98
234. 31
191. 92
205.27
188.25
305.31
304.01
674.86
668.55
547.40
546.02
aModel Plants 1-6 produce carbon billets; plants 7-10 produce alloy billets.
Total Furnace Enclosure Option
NOTE: na = not applicable.
9-78
-------
As expected, the ATC of producing steel inere?ses when plants are
required to meet the baseline standard. The ATC of producing carbon steel
billets under regulatory alternative 1 increases to a range of $191.41 to
$320.62 per Mg. The cost per Mg of producing alloy billets is $546.32 to
$672,76 at baseline.
Under proposed regulatory alternative 2 or 3, the ATC of production
increases only slightly from baseline for most plants and declines for all
others. The ATC of carbon steel is between $187.86 and $320.76 under
alternative 2, and between $188.25 and $321.21 under alternative 3. The
control of model plants 1 and 6 using total furnace enclosures to achieve
alternative 3 actually reduces from baseline the ATC of producing carbon
steel billets to $319.98 and $304.01, respectively.
The ATC changes associated with moving to more stringent controls are
more explicitly presented in Table 9-30. Moving from no control to base-
line increases the ATC of production by $5.06 to $20.26 per Mg for carbon
shops and $20.83 to $36.27 per Mg for specialty shops. These cost in-
creases represent impacts of 2.7 percent to 7.1 percent for carbon steel
and 4.0 percent to 5.7 percent for specialty steel.
Cost impacts measured relative to baseline alternative 1 are generally
quite small for the proposed alternatives. The impacts on ATC of alter-
natives 2 and 3 are all under $1 per Mg for all carbon plants. The impacts
are likewise small in percentage terms. The greatest cost impact is the
0.31 percent increase in ATC for model plant 7 under alternative 3.
The average total cost data presented above pertain to estimates using
a weighted average cost of capital of 6.2 percent. Analogous estimates
using a 10 percent cost of capital are presented in Tables 9-31 and 9-32.
The ATC of steel is higher when the cost of capital is higher because the
higher cost of borrowed funds must be recovered. The differences, however,
are not great.
The average total cost and net present value estimates presented in
this and the previous section indicate, that the modeled steel shops with
electric arc furnaces and, in the case of specialty shops, with argon-
oxygen decarburization vessels are profitable investments given the current
market price of steel and the cost of capital to steel firms. This holds
true for all regulatory alternatives. If the input data and assumptions
implicit in the discounted cash flows model are correct, the potential
maximum impacts of the proposed regulations on stockholders as measured by
9-79
-------
TABLE 9-30. AVERAGE TOTAL COST IMPACTS ASSUMING
'6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
(1981 $/Mg) '
Regulatory Impact From Basel
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
•
1
1981$
19.13
19.13
12.17
5.06
7.38
5.32
20.26
20.26
36.27
31.00
22.'17
20.83
percent
6.35
6.35
5.49
2.72
3.73
2.91
7.12
7.12
5.70
4.87
4.23
. 3.97
Regulatory
1981$
0.14
na
0.22
0.14
0.12
0.15
0.16
na
0.81
•0.22
0.36
0.22
Alternative
2 : •'" " '
percent
0.04
na
0.09
0.07
0.06
0.08
0.05
na
0.12
0.03
0.07
0.04
• a
me
• ' . 3 .
1981$
.0.59
-0.64
0.37
0.50
0.30
0.53
0.62
-0.68
2.09
1.06
1.08
1.04
percent
0.18
-0.20
0.16
0.26
0.15
0.28
0.20
-0.22
0.31
0.16
0.20
0.19
almpact of baseline regulatory alternative 1 is measured from no control.
Total Furnace Enclosure Option
NOTE: na = not applicable.
9-80
-------
TABLE 9-31. AVERAGE TOTAL COST ASSUMING 1Q.O PERCENT
WEIGHTED AVERAGE COST OF CAPITAL"
(1981 $/Mg)
Regulatory Alternative
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
No Control
321.01
321.01
234.55
190.83
205.57
185.38
296.97
296.97
677.23
677.23
552.28
552.28
(Baseline)
344.97
344.97.
249.51
196.75
214.45
191. 72
323.04
323.04
723. 51
716.15
580.38
578.51
... ... 2 .. .. .
345. 25
na
249.83
196.94
214.62
191.93
323.37
na
724.49
716.53
580.85
578.81
3
345.78
344. 28
250.05
197.37
214.84
192.40
323.91
322.28
726.11
717.56
581. 77
579.89
aModel Plants 1-6 produce carbon billets; plant's" 7-10 produce alloy billets."
Total Furnace Enclosure Option
NOTE: na = not applicable
9-81
-------
TABLE 9-32. AVERAGE TOTAL COST IMPACTS ASSUMING 10.0 PERCENT
WEIGHTED AVERAGE COST OF CAPITAL
Regulatory Impact
From Basel i
nea
Requlatory Alternative
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
1
1981$
- 23.96
23.96
14.96
5.91
8.89
6.34
26.07
26.07
46.28
38.92
28.10
26.23
percent
7.46
7.46
6.38
3.10
4.32
3.42
8.78
8.78
6.83
5.75
5.09
4.75
2
1981$
0.28
na
0.31
0.19
0.16
0.21
0.33
na
0.98
0.39
0.47
0.31
percent
0.08
na
0.13
0.10
0.08
0.11
0.10
na
0.14
0.05
0.08
0.05
1981$
0.81
-0.69
0.54
0.63
0.39
0.68
0.87
-0.76
2.60
1.42
1.39
1.38
3
percent
0.24
-0.20
0.22
0.32
0.18
0.35
0.27
-0.24
0.36
0.20
0.24
0.24
almpact of baseline regulatory alternative 1 is measured from no control.
Total Furnace Enclosure Option
NOTE: na = not applicable.
9-82
-------
NPV reductions and the potential maximum price imparts as measured by ATC
changes are negligible.
Several qualifications to the above statements are worthy of mention.
To begin, the observed differences between market price and average total
costs'are more substantial than would be expected in a competitive market
with free entry (see Section 9.1.3.6). This may be an indication that
inaccuracies in the cost data or elsewhere ara leading to underestimation
of ATC and a resulting overstatement of project profitabilty. In response
to this objection, however, it is noteworthy that these are new plants
incorporating the best technology to date, and are expected to have low
average costs. Not only do they employ the lowest cost steel melting
technology, but low cost continuous casters as well. Further, firms are
currently building plants of this type at a rapid rate, which is an indica-
tion that they themselves perceive rents to be earned.
Another qualification relates to the impact of these proposed regu-
lations alone as distinct from the cumulative impacts of other current and
proposed regulatory requirements. Note, for example, that the impacts of
regulatory alternative 1 (baseline) relative to no control, are the most
substantial. Other regulations governing, for example, water emissions and
plant safety^ may impose added costs. A prospective builder considers all
current and, to the best of his ability, future regulatory requirements in
his decision. Still, this analysis concludes that these modeled shops are
profitable meeting both current requirements and the proposed New Source
Performance Standards.
Finally, the discounted cash flows model employed may fail to ade-
quately account for investment uncertainties. Increases in input costs,
for example the cost of scrap, would tend to reduce plant profitability.
Steel price reductions would have the same effect. Potential changes in
the cost of capital, in corporate income tax rates, in depreciation laws
and other factors make the capital budgeting decision more difficult and
risky than this analysis indicates.101
9.3.3.3 Output, Employment and Imports Impacts. Estimates of NSPS
impacts on output and employment in the domestic steel industry and on
steel imports are presented in this section. Impacts are presented for the
year 1987 for all model plants assuming that the real market price of steel
in 1987 increases from its baseline price by an amount exactly equal to the
9-83
-------
change in ATC resulting from the regulatory alternatives. All impacts are
computed using the estimated weighted average cost of capital of 6.2 percent.
9.3,3.3.1 Impacts on steel output. Domestic steel production in 1987
is projected to total 119.7 million Mg. Of this total, 101.7 million Mg
will be carbon steel and 18.0 million Mg will be specialty steel. If the
real price of domestic steel increases as a result of the NSPS, demand for
domestic steel would be expected to decline; industry output would be below
the 119.7 million Mg projection.
For each model plant, the projected change in industry output is
computed assuming the price of carbon or specialty steel increases by an
amount equal to the change in ATC. The percentage change in ATC resulting
from each regulatory alternative is multiplied by an estimated own-price
elasticity of demand for specialty steel of -1.86.102 These percentage
changes are then multiplied by the baseline carbon or specialty steel
projections for 1987 to yield the relevant output reductions. Results of
these computations are presented in Table 9-33.
Table 9-33 indicates that 1987 domestic production of carbon semi-
finished steel would be 0.08 million Mg lower under regulatory alternative 2
than under baseline if all projected new EAF capacity was met by the
construction of plants like model plant 1. This impact represents a
reduction of under one-tenth of one percent from the baseline projection of
101.7-million Mg. The production impacts of regulatory alternatives 2 and
3 are in general quite small. Percentage impacts from baseline vary from
-0.06 to -0.22 percent for alternative 2 and from -0.58 to 0.41 percent for
alternative 3.
9.3.3.3.2 Steel industry employment impacts. Estimated impacts on
domestic steel employment are presented in Table 9-34. They are computed
by multiplying the estimated output changes in Table 9-33 by an employ-
ment-to-output coefficient of 3,663 workers per million megagrams.28
While the employment impacts of baseline alternative 1 relative to no
control are measured in tens of thousands of jobs, the impacts of the new,
proposed standards are much smaller. The impact on employment resulting
from moving from baseline to regulatory alternative 2 ranges from a loss of
40 jobs to a loss of 642 jobs, depending on the affected facility. The
impacts of alternative 3 range from a gain of 1,538 jobs to a loss of 1,963
9-84
-------
TABLE 9-33. DOMESTIC STEEL PRODUCTION IMPACTS FOR 1987C
Regulatory Impact From Basel i
neb
Regulatory Alternative
Model
Plant
1
lc
2
3
4
5
6
6C
1
8
9
10
106 Mg
-12.01
-12.01
-10.38
-5.14
-7.06
-5.51
-13.48
-13.48
-1. 91
-1.63
-1.42
-1.33
1
percent
-11.80
-11.80
-10.21
-5.05
-6.9'4
-5.42
-13.25
-13.25
-10.60
-9.06
-7.87
-7.39
106 Mg
-0.08
na
-0.18
-0.14
-0.11
-0.15
-0.10
na
-0.04
-0.01
-0.02
-0.01
2
percent
-0.08
na
-0.17
-0.13
-6.11
-0.15
-0.10
na
-0.22
-0.06
-0.12
-0.07
106 Mg
-0.35
0,38
-0.30
-0.50
-0.27
-0.54
-0.38
0.42
-0.10
-0.05
-0.07
-0.06
3
percent
-0.34
0.37
-0.30
-0.49
-0.27
-0^53
-0.38
0,41
-0.58
-0.29
-0.37
-0.35
Assuming 6.2 percent weighted average cost of capital.
Impact of baseline regulatory alternative 1 is measured from no control.
cTotal Furnace Enclosure Option
NOTE: na = not applicable.
9-85
-------
TABLE 9-34. DOMESTIC STEEL INDUSTRY EMPLOYMENT IMPACTS FOR 1987C
Model
Plant
1
lc
*
2
3
4
5
6
6C
7
8
9'
10
1
" jobs
-43,976
-43,976
-38,020
-18,817
-25,872
-20,195
-49,368
-49,368
-6,987
-5,972
. -5,186
-4,873
Impact From
percent
-11.80
-11.80
10.21
-5.05
-6.94
-5.42-
-13.25
-13.25
-10.60
-9.06
-7.87
-7.39
Baseline of Regulatory Alternative
2
jobs
-308
na
-642 •
-496
-394
-542
i
-362
na
-148
-40
-81
j -48
percent
-0.08
na
-0.17
-0.13
-0.11
-0.15 r
-0.10
na
-0.22
-0.06
-0.12
-0.07
3
jobs
-1,275
1,392
-1,102
-1,820
-1,005
— -i,%sT
-1,409
1,538
-382
-194
-241
-233
percent
-0.34
0.38
-0.30 "
-0.49
-0.27
-0.53
-0.38
0.41
-0.58
-0.29
-0.37
-0.35
aAssuming 6.2 Percent Weighted Average Cost of Capital
blmpact of baseline alternative 1 measured from no control
°Total Furnace Enclosure Option
NOTE: na = not applicable.
9-86
-------
jobs. Even the greatest Impact, that for model plant 5 under alternative 3,
represents a loss from baseline employment of only 0.53 percent.
It is important to remember that these employment losses are not
layoffs. Rather, they are jobs that will not be created by 1987 that
otherwise would have been. Also, these computed losses do not take account
of any employment increases that might result in other sectors of the
economy. For example,-because aluminum competes with steel, employment
increases might be expected in the aluminum industry. Employment gains are
expected to occur in industries that produce control equipment.
9.3.3.3.3 Impacts on steel imports. As the real price of domestic
steel rises relative to that for imported steel, domestic users are expected
to increase their purchases of foreign steel. It has been estimated that a
1 percent increase in the domestic price for finished steel results in a
1.51 percent increase in steel imports.103 This import elasticity is
multiplied by the percentage change in ATC from Table 9-30 to compute the
percentage change expected to result in steel imports from the NSPS on EAF
secondary emissions.
If the same quantity of steel imported into the U.S. in proportion to
domestic production is imported in 1987 as in 1979, steel imports will
total 20.3 million Mg in 1987. Of this total, 19.3 million Mg will be
carbon steel and 1.0 million Mg will be specialty steel.104 The NSPS
induced percentage change in steel imports estimated as explained above is
multiplied by the relevant baseline import projection to compute import
impacts. Estimates are presented in Table 9-35.
In general, more stringent regulatory alternatives result in greater
steel imports. The greatest impact on carbon steel imports relative to
baseline is the 0.08 million Mg increase resulting from imposing regu-
latory alternative 3 on model plant 5. This represents, however, only a
0.43 percent increase. The impacts on specialty steel imports are all
under 0.01 Mg.
9.3.4 Anticipated Economic Impacts
Section 9.2.3 presented ranges of impacts on NPV assuming full cost
absorption, and on price and other variables assuming full cost pricing.
The purpose of this section is to present estimates of impacts on net
9-87
-------
TABLE 9-35. STEEL IMPORTS IMPACTS FOR 1987C
Requlatory Impact From Basel i
Model
Plant
1
lc
2
3
4
5
6
6C
7
8
9
10
10s Mg
1.85
1.85
1.60
0.79
1.09
0.85
2.08
2.08
0.09
0.07
0.06
0.06
1
percent
9.58
9.58
8.29
4.10
5.64
4.40
10.76
10.76
8.60
7.35
6.39
6.00
neb
Requlatory Alternative
106 Mg
0.01
na
0.03
0.02
0.02
0.02
0.02
na
d
d
d
d
2
percent
0.07
na
0.14
0.11
0.09
0.12
0.08
na
0.18
0.05
0.10
0.06
106 Mg
0.05
-0.06
0.05
0.08
0.04
0.08
0.06
-0.06
d
d
d
d
3
percent
0.28
-0.30
0.24
0.40
0.23
0.43
0.31
-0.33
0.47
0.24
0.30
0.29
aAssuming 6.2 percent weighted average cost of capital.
^Impact of baseline regulatory alternative 1 is measured from no control.
°Total Furnace Enclosure Option
dLess than 0.005 million Mg.
NOTE: na = not applicable.
9-88
-------
present value, steel price, and other relevant variables that are con-
sidered most likely to result under the proposed alternatives. The impacts
that actually occur depend on a number of factors including the types of
plants actually built and the extent to which cost increases are passed
forward. The first factor is addressed in Section 9.2.4.1 while Section
9.2.4.2 deals with the cost-price mechanism.
9.3.4.1 Model Plant Selection. Ten model plants are described in
Chapter 6. The plants vary in certain respects but are all electric arc
process steel furnace projects. As seen in Table 9-23, each model plant
has associated with it a unique set of fixed capital and annual operating
costs. Accordingly, as evidenced in Tables 9-25 and 9-29, each project has
associated with it a unique net present value and average total cost. It
is thus clear that the economic impacts that will actually result from the
proposed NSPS will depend in part on which types of facilities are actually
constructed.
Of the model plants investigated in this report, plant 4 is believed
to best typify new EAF carbon steel shops to come on line over the next
several years. This represents a 136.1 Mg ultra high power vessel shop.
Model plant 7 is thought to be most representative of specialty steel shops
to be built in the near future.
These models are considered representative of future projects partly
because they are similar in capacity to recently constructed shops. The
separate control of emissions from the EAF and EAF/AOD vessels in model
plant 7 allows greater operational flexibility than the common control of
model plant 8.
9.3.4.2 Estimates of Anticipated Impacts. The economic impacts that
are expected to result from the proposed regulatory alternatives are esti-
mated assuming that model plants 4 and 7 typify future EAF construction
through 1987. Table 9-36 has been constructed from Tables 9-25 and 9-29.
The ATC data in Table 9-36 are a certain result of the NSPS given that
the cost data and model parameter values are correct. That is, the total
cost of producing a..megagram of steel in each model plant is sure to increase
as a result of the NSPS. The impact of this change on other relevant
variables is less certain. Two extreme results are possible. If the
market prices of carbon and specialty steel were to remain unchanged,
9-89
-------
TABLE 9-36. NET PRESENT VALUE AND AVERAGE TOTAL COST DATA
FOR MODEL PLANTS 4 AND 7a
Regulatory Alternative
1
Model
Plant
4
7
NPV
(106
1981 $)
359.463
6.332
ATC
(1981
$/Mg)
204. 97
672.76
2
NPV
(106
1981 $)
359.261
6.211
ATC
(1981
$/Mg)
205.09
673.58
NPV
(106
1981 $)
358.947
6.020
3
ATC
(1981
$/Mg)
205.27
674.86
Assuming 6.2 percent weighted average cost of capital.
9-90
-------
the proposed regulations would diminish the difference between price and
ATC, thus reducing the NPV's of these plants. The NPV of carbon shop model
plant 4 would decline from its baseline $359.463 million to $359.261 million
under alternative 2, or to $358.497 million under alternative 3. These
0.06 and 0.27 percent NPV reductions would be felt as a one-time reduction
in wealth spread across all holders of outstanding shares of stock of
affected firms.. As seen in Table 9-36, similar NPV reductions would result
from the proposed regulations on specialty shop model plant 7 assuming full
cost absorption.
In a case of full cost pricing, project NPV's would remain unchanged
from their baseline values as .market price rose by the same amount as ATC.
A most relevant1question thus becomes: what is the mechanism linking ATC
to price and what will be the result of this mechanism in this circum-
stance? . 1; 1
9.3.4.2.1 Analytical Framework
The analytic framework thatT is applied in this analysis depends heavily
upon the work of W. E. G. Salter.105 The framework is soundly based on
<-
standard microeconomic theory, employs a comparative statistics approach,
and assumes certainty in relevant markets. Price and quantity are deter-
mined by market forces, not by individual market participants.
This approach recognizes that there are two distinctly different types
of production decisions: operating decisions and investment decisions.
Operating decisions involve simply whether or not a firm with plant
and equipment already in place purchases inputs to produce output. These
are sometimes called short-run decisions since the decision period is
sufficiently short that certain inputs, namely plant and equipment, are
fixed. A profit-maximizing firm will operate existing capital as long as
the market price for its output exceeds it unit variable costs of production.
As long as market price even marginally exceeds average variable (operating)
cost, .the producing plant will cover not only the cost of its variable
inputs but will cover part of its capital cost as well. A profit-maximizing
firm will not pass-up an opportunity to recover even part of the initial
investment it made in the plant and durable equipment.
Investment decisions differ from operating decisions in that they
involve whether or not the firm should put in place new plant and/or equipment.
9-91
-------
The investment decision is sometimes called a long-run decision since the
time frame is sufficiently long that all inputs, including capital, are
variable. A firm will not invest in new capital unless current and expected
future market price is sufficient to cover both the cost of operating the
new capital (variable costs) and the cost of purchasing and owning the
capital, including a normal rate of return. Put differently, a firm will
not invest unless market price equals or exceeds average total cost.
The hypothesized supply schedule from a single existing plant is
depicted in panel (a) of Figure 9-1. Given the capital in place, the plant
owner is willing to supply output Q* as long as market price equals or
exceeds the plant's average operating cost (AOC). If market price is below
AOC, the owner is unwilling to produce even a fraction of Q* because a per
unit loss would be incurred. If market price should substantially exceed
AOC, the owner would be wi11ing to produce output beyond Q* but is unable
to do so given plant capacity.
The hypothesized supply schedule from an as yet unconstructed plant is
depicted in panel (b) of Figure 9-1. Because the plant and equipment are
not yet in place, all inputs are variable. The scale (capacity) of the
plant itself is variable. Thus, the supply schedule does not turn up at
any output rate. The assumption of a perfectly elastic plant supply curve
is probably realistic. It is unlikely that input factor prices would be
bid-up by the demands of a single plant. . -
Supply will not be forthcoming from the new plant, that is, it will
not be built, unless market price exceeds the average total cost of production.
The plant will be constructed only if the anticipated market price is
sufficiently above average operating cost to recover the capital investment
and provide a normal return on the capital.
With an understanding of plant-level supply, focus is now directed
towards market-level considerations. In panel (a) of Figure 9-2, the
conventional equilibrium determination of market price and quantity is
depicted. Market demand (D) is assumed to be downward sloping. Because an
NSPS affects new supply facilities, it is the supply schedule (S) that is
of interest here.
During any period of time, the market output of a good is the sum of
the quantities produced by individual plants. As discussed above, once a
9-92
-------
$/Q
AOC
$/Q
ATC
Q*
(a]
.Q/time
.Q/time
ibj
Figure 9-1. Supply schedules for constructed and unconstructed plants.
9-93
-------
0)
c
CO
cr
TJ
(0
O
a.
•*>j
®
jfi
CO
O
CO
•8
CO
4>
W
a,
£
3
D)
UU
CO
9-94
-------
Q Q
I i
S
O)
I
o>
LL
9-95
-------
plant is in place, it will produce as long as market price is equal to or
greater than its average operating cost. Failure to do so would involve
passing-up an opportunity to earn some return on existing fixed capital.
In general, newer plants will have capital of superior technology to that
of older plants. Thus, the average operating costs of newer plants will
generally be lower than those of older plants: Thus, each plant is willing
to produce output at a different market price.
The willingness of existing plants with different average operating
costs to produce at a different minimum market price results in the upward
slope of the supply schedule. This is illustrated in panel (b). The
newest plant, which is of vintage n-1, has the lowest average operating
cost. Thus, it is willing to supply output as long as price at least
equals P,. Plants constructed in successively earlier periods have
increasingly higher average operating costs and are willing to produce only
at higher prices. The oldest plant, produced in period n-5, has the highest
average operating cost and is the oldest existing plant that is willing to
produce at prevailing price P*; it is thus said to be a marginal plant.
Now that it is understood why the supply schedule is upward sloping to
Q*, it is time to investigate the slope of the supply schedule beyond Q*.
This is a question of long-run supply, since output in excess of Q* can be
produced only after a new plant has been constructed.- The question becomes
then, what is the market price at which a new plant will be constructed?
Recall from the discussion of panel (b) in Figure 9-1 that all costs
of an unconstructed plant are variable costs. The prospective builder will
invest in the new plant only if the anticipated market price is suffi-
ciently high to cover average total cost, which is average operating cost
plus average capital cost, including a normal return on the capital.
In panel (c) of Figure 9-2, plant n represents the as yet unconstructed
plant. Because it will incorporate the latest technology, it will have low
operating costs. In the figure, oa is the average operating cost of the
new plant. The cost component aT is the average capital cost of the new
plant. This cost component represents the return per unit output in excess
of operating cost required in order to repay the principal of the original
capital investment and earn a normal rate of return on that investment.
The firm's desire to recover the investment principal and earn a normal
9-96
-------
return holds for both existing facilities and facilities under consideration.
In the latter case though, even these costs are variable, indeed they are
avoidable. The firm has the alternative of not building at all; i.e., of
investing in another project. Thus, the new plant will be built only if
market price equals or exceeds average total cost oT. Once built, it will
supply output Q' - Q* as long as market price covers average operating
cost; its capital costs become sunk. Thus, the long-run supply schedule,
at least from Q* to Q', is elastic at price = average total cost of the
best technology plant.
The validity of assuming perfectly elastic long-run market supply is
unknown. For relatively small increases in market output resulting from
the construction of, say, only a few new plants, the assumption is probably
reasonable. If, however, the number of newly constructed plants increased
market demand for factors of production significantly, it is possible that
factor prices would be bid-up and that long-run supply would be upward-si oping.
Significantly additional quantities of iron and steel scrap, for example,
might be forthcoming at higher market prices. The implications for impact
estimates of assuming perfectly elastic long-run market supply will be
addressed later.
With an understanding of the decision criterion for new plant con-
struction, it is possible to analyze the effects of an NSPS in a market
initially in both short-run and long-run equilibrium. In panel (d), the
long-run supply schedule (S) beyond Q* is determined by the average total
cost of the design-stage plant. Prior to the NSPS, some unknown number of
new plants would have been constructed during time period n + 1. These new
plants would have been constructed until the expansion of market output
relative to demand drove market price below P*; i.e., until the next plant
to come on line would be unable to cover average total cost. In general,
an NSPS raises both the costs of building and operating a new plant. The
result is an increase in the average total cost of production. In panel (d),
the increase in average total cost results in an upward shift of the supply
schedule beyond Q*, from S to S1. The anticipated market price required to
bring the new plant on line is now P1. Only at this higher price can the
investor cover operating costs and earn a normal rate of return.
In a competitive output market, the prospective investor is a price-taker;
9-97
-------
his individual decisions do not affect market price. If market price is P*
(panel (d)) he cannot simply sell his own output at price P'. Unless
market-wide forces cause price to increase from P* to P1 , investment in the
plant will fail to yield a normal rate of return. Funds will instead be
invested in an investment that does yield a normal rate. New plant investment
will be delayed until market forces bring about market price P1. The
number of new plants constructed in period n + 1 will be lower with the
NSPS than without.
A likely circumstance is that increasing demand for the product, as a
result of economic growth, relative to the supply from existing plants will
drive up product price. This is illustrated in panel (e). When demand
shifts from D to D1, market price P' will prevail, allowing a normal rate
of return on a new, controlled plant.
Two consequences of an NSPS have already become apparent from the
model. The NSPS, by increasing the average total cost of production in the
new plant, raises the market price at which it is "profitable" to build
that plant. The result is a delay in construction until market forces,
such as an increase in demand, bring about an increase in market price.
There is thus a price impact and a new plant construction impact.
Panel (e) indicates there is a third impact - the impact on total
quantity traded. In the absence of the NSPS, a shift in demand from D to
D1 would have resulted in market trade of Q** units per period of time. In
the presence of the NSPS, only Q* units are traded given D'. There is an
output impact of Q** - Q* units.
The steel market may well be in short-run equilibrium. The market is
clearing with neither excess supply nor excess demand.
The net present value and average total cost results presented in
Section 9.2.4.2 indicate that the domestic steel market is not, however,
presently in long-run equilibrium: the average total costs of model plants 2
and 7 at baseline are considerably below the market prices of $412.16 per
Mg of carbon steel and $715.18 per Mg of specialty steel. Thus, it is
profitable for firms to build EAF plants of this type; i.e., the plant
NPV's are positive. This is in .keeping with our observation that electric
arc furnace steel shop construction is brisk and our projection of continued
growth in EAF capacity.
9-98
-------
In order to determine the economic impacts of a NSPS on these new
plants, consider first the expected market adjustments in the absence of
'new standards. Panel (a) of Figure 9-3 depicts the market for (carbon)
steel in disequilibrium. Output Q is being traded at market price P determined
by the intersection of demand and/supply (S) from existing'firms. However,
the ATC of production for the newest, unconstructed plant (model plant 2)
is below market price Pi Long-run equilibrium will be restored only when
market price falls to PQ - ATC, Given demand,;and ,for simplicity given, :
supply from existing plants, the construction of some number of new plants
will expand supply relative to demand until market price falls to' P-e, Once,
market price P is acheived there will be no further incentive.to build new
plants. .
The impact of either of the two proposed regulatory alternatives on
model plant 4 can now be analyzed. The standard increases the ATC of
production for model plant 4 from ATC to ATC'; e.g., from $203.12 to $203.35
for alternative 3, an increase of $0.23 per Mg. In panel (b) of Figure 9-3,
the ATC of the regulated new plant, ATC', is higher than the ATC of the,
unregulated new plant, ATC in panel (a). The ATC of production for the
regulated new plant is still below current market price P, $412.16 per Mg.
In the presence of the standard, construction of new plants will continue
until the long-run equilibrium price is achieved; i.e., until the NPV of a
new plant is zero. The new long-run equilibrium price is no longer Pg
however, but P^. Thus, fewer plants will be constructed before supply will
be expanded to Q' restoring long-run equilibrium.
The anticipated economic impacts of the proposed standards can thus be
summarized. There is a price and quantity impact associated with a standard
on new electric arc furnace steel facilities. However, the price impact
would be felt not as an increase in the price of steel above the current
price but rather as an increase above the (lower) market price that would
have obtained in the absence of the standard: the difference between PQ
and P1. Likewise, the output impact would not be felt as a reduction from
current output but rather as a reduction from the projected baseline output: the
amount Q - Q1. There is no net present value impact because the new long-
run equilibrium steel price is just sufficiently higher than the baseline
long-run equilibrium price to compensate producers for increased production
9-99
-------
W
*"'
1
! I
11
Q> ~
m «"
© 1
"c ®
5 .2
«o o.
I I
MM ,J5
&
OT
CO
i
I
O"
(2)
9-100
-------
costs. The higher price for domestic steel will lead to an increase in
steel imports and the reduction in domestic output will have an impact on
industry employment.
The impact estimates presented below assume perfectly elastic long-run
market supply. If long-run supply is upward sloping, different impacts
will result. An effort is made here to discuss the qualitative implications
of upward-si oping long-run supply on the impact estimates.
c,In the case of perfectly elastic long-run market supply, long-run
equilibrium price is equal to today's average total cost of the best technology
plant. However, if factor prices are bid-up as more and more of these
plants begin producing, average total cost will increase with market output
and the long-run equilibrium price will be determined by the plant supplying
the marginal output. Thus, long-run equilibrium price would be expected to
be higher with upward-si oping long-run market supply. Logically, with
downward-si oping demand, market clearing output would be lower at the
higher long-run equilibrium price.
The absolute impacts on both price and quantity are lower in the case
of upward-si oping supply. That is, the differences between controlled and
uncontrolled long-run equilibrium price and quantity are lower when supply
slopes upward. Because consumers purchase less at higher prices, market
clearing quantity falls and fewer new plants come on line. With fewer
plants producing, input prices are not bid-up as severely. Hence, there is
a smaller increase in price for the vertical supply shift the more upward
sloping is supply.
To summarize, long-run equilibrium price would increase and long-run
equilibrium output would decline as long-run supply becomes less elastic.
The differences between controlled and uncontrolled prices and quantities
would be smaller with less elastic supply.
Again, the estimate impacts presented below assume perfectly elastic
long-run market supply. Hence, the impact estimates a biased upward..
9.3.4.2.2 Impact Estimates
Table 9-37 summarizes the anticipated impacts of regulatory alter-
native 2 on electric arc carbon and specialty steel making. The impacts
computed are those expected in 1987. Impacts are measured relative to
regulatory alternative 1, which is baseline.
9-101
-------
TABLE 9-37. SUMMARY OF ECONOMIC IMPACTS FROM BASELINE OF
REGULATORY ALTERNATIVE 2a
Net Present Value
Steel Price ($/Mg)
Domestic Output (Mg)
Employment (jobs)
Imports (Mg)
Total Cost of Regulatory
Alternative (106 $)
Carbon Steel
(Model Plant 4)
0.12 ( 0.06%)
-110,000 (-0.11%)
-394 (-0.11%)
17,000 ( 0.09%)
0.608
Alloy Steel
(Model Plant 7)
0.81 ( 0.12%)
-40,000 (-0.22%)
-148 (-0.22%)
b ( 0.18%)
0.773
Model Parameters:
Cost of capital
Own-price demand elasticity
Import elasticity
1987 Baseline price
1987 Baseline output
1987 Baseline employment
1987 Baseline imports
°Less than 5,000 Mg.
GSee Section 9.2.6, Table 9-41.
6.2%
-1.86 '
1.51
$229.62/Mg - carbon
$652.70/Mg - specialty
101.7 x 106 Mg - carbon
18.0 x 10s Mg - specialty
372,527 - carbon
65,934 - specialty
19.3 x 106 Mg - carbon
1.0 x 106 Mg - specialty
9-102
-------
There is no anticipated impact on project net -resent values. The
proposed standard would have an estimated price impact of $0.12 per Mg for
carbon steel and $0.81 per Mg for specialty steel. These represent price
impacts of 0.06 percent and 0.01 percent, respectively. The price in-
creases would lead consumers to purchase 110,000 fewer Mg of carbon steel
and 40,000 fewer Mg of specialty steel in 1987. These are reductions of
0.11 and 0.22 percent from 1987 baseline production. These output re-
ductions would lead to fewer employment opportunities in 1987 totaling
approximately 542 jobs—an employment loss of about one-tenth of one
percent. The price increase of domestic steel would also cause domestic
users to substitute foreign for domestic steel. An estimated 17,000 Mg
more of carbon steel would be imported under alternative 2 than at baseline.
Table 9-38 summarizes the anticipated impacts of regulatory alter-
native 3 on electric arc carbon and specialty steelmaking. The impacts
computed are those expected in 1987. Impacts are measured relative to
regulatory alternative 1, which is baseline.
The anticipated impact on project net present values is zero. It is
expected that this alternative would have a $0.30 per Mg impact on the
price of carbon steel and a $2.09 per Mg impact on the price of specialty
steel. Thus, the price of carbon steel in 1987 would be 0.27 percent
higher than it would be in the absence of this regulation and the price of
specialty steel 0.58 percent higher. The price increases would lead c
onsuniers to purchase less domestic steel--270,000 Mg less carbon steel ?and
100,000 Mg less specialty steel. These output effects of 0.27 percent and
0.58 percent, respectively, would cause proportionate reductions in employ-
ment opportunities—an estimated 1,387 fewer jobs in the industry in 1987.
The higher prices of domestic steel would also encourage higher imports.
An estimated 40,000 Mg more of carbon steel would be imported under regula-
tory alternative 3 than at baseline.
9.3.5 Capital Availability
This section investigates how the proposed regulatory alternatives
will affect the steel industry's ability to raise capital.
9-103
-------
TABLE 9-38. SUMMARY OF ECONOMIC IMPACTS FROM BASELINE OF
REGULATORY ALTERNATIVE 3a
Net Present Value
Steel Price ($/Mg)
Domestic Output (Mg)
Employment (jobs)
Imports (Mg)
Total Cost of Regulatory
Alternative (106 $)c
Carbon Steel
(Model Plant 4)
0
.. 0.30 ( 0.15%)
-270,000 (-0.27%)
-1,005 (-0.27%)
40,000 ( 0.23%)
1.415
Alloy Steel
(Model Plant 7)
0
2.09 ( 0.31%)
•100.,000 (-0.58%)
-382 (-0.58%)
b ( 0.47%)
1.160
Model Parameters: Cost of capital = 5.2%
Own-price demand elasticity = -1.86
Import elasticity = 1.51
1987 Baseline price
1987 Baseline output
1987 Baseline employment
1987 Baseline imports
Less than 5,000 Mg.
:See Section 9.2.6, Table 9-42.
$229.62/Mg
$652.70/Mg
carbon
specialty
101.7 x 10s Mg - carbon
18.0 x 106 Mg - specialty
372,527 - carbon
65,934 - specialty
19.3 x 10s Mg - carbon
1.0 x 106 Mg - specialty
9-104
-------
To begin, the capital required to build each <-odel plant under each
regulatory alternative is compared. Table 9-39 shows capital requirements
by plant type uncontrolled, under baseline, and under the two proposed
alternatives. The percentage changes in the capital requirement from
baseline to alternatives 2 and 3 are also shown.
Note first that the additional capital required to meet alternative 1,
which is baseline, over no control is substantial in percentage terms—from
a low of 18.95 percent for model plant 4 to a high of 46.22 percent for
model plant 6. these same increases are not, however, especially high in
dollar terms. The greatest increase is the $9.3 million increase for model
plant 9.
The additional capital required over baseline to meet alternative 2
or 3 is relatively small. Regulatory alternative 2 requires an added
capital expenditure that never exceeds 1.6 percent of the baseline expendi-
ture. In dollar terms, the greatest increase is $0.36 million (model
plant 5). Regulatory alternative 3 also requires little increase in capital
expenditures. The greatest increase from baseline is $0.84 million (model
plant 5). In percentage terms, alternative 3 has the greatest impact on
model plant 5—3.68 percent.
Particular attention should be drawn to the incremental capital require-
ments for model plants 4 and 7 since these projects are utilized for the
anticipated impacts analysis. 190,000 dollars more capital is required to
build model plant 4 under alternative 2 and $380,000 more is required under
alternative 3. These represent increases from baseline of 0.48 and 0.96 percent,
respectively. Construction of model plant 7 requires $60,000 more capital
under alternative 2 and $180,000 million more under alternative 3.
One qualification should be discussed. The NSPS on electric steel-
making is only one of several environmental regulations affecting the steel
industry. Others include regulations on coking facilities and other steel-
making processes. Each is expected to impose additional capital costs on
the industry. Taken together these regulations may result in some difficul-
ties in obtaining financing for some companies. Even this is not certain.
Table 9-40 presents the "funded debts to net working capital"
9-105
-------
TABLE 9-39. CAPITAL REQUIREMENTS OF REGULATORY ALTERNATIVES
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
' No
Control
7.00
7.00
20.00
25.60
33.20
17.00
4.50
4.50
14.60
14.60
44.00
44.00
Capital requirements
(106 1981 $)
Regulatory Alternative
1
8.73
8.71
24.37
30.47
39.49
22.84
6.58
6.56
18.19
17.44
53.28
52.45
2
8.78
na
24.52
30.77
39.68
23.20
6.64
na
18.25
17.50
53.45
52.59
3
8.81
8.71
24.62
31.17
39.87
23.68
6.67
6.55
18.37
17.57
53.78
52.99
Change from baseline
(percent)
Regulatory Alternative
1
24.71
24.43
21.85
19.02
18.95
34.35
46.22
45.78
24.59
19.45
21.09
19.20
2
0.57
na
0.62
0.98
0.48
1.58
0.91
na
0.33
0.34
0.32
0.27
3
0.92
0
1.03
2.30
0.96
3.68
1.37
-0.15
0.99
0.75
0.94
1.03
almpact of regulatory alternative 1 measured from no control.
Total Furnace Enclosure Option
NOTE: na = not applicable.
9-106
-------
TABLE 9-40. INDUSTRY DEBT RATIO?76
Industry
Debt
ratio'
Blast furnaces and steel mills—.SIC 3312
Primary nonferrous metals, NEC—SIC 3339
Aluminum foundries—SIC 3361
Primary metal products, NEC—SIC 3399
Motor vehicles—SIC 3711
Petroleum refining—SIC 2911 '•'.;.
87.2
48.4
. .60.6
! 39.5
, 41.;8
. 61.2
Debt ratio =
Funded debts
Net working capital
9-107
-------
ratio (debt ratio) for several industries. The numerator of this ratio,
funded debts, is all obligations with maturities exceeding I year, including
bonds, mortgages, and term loans. These are the instruments that would.
likely be utilized to finance a new steel project. The denominator, net
workirig capital, represents the excess of current assets over current
liabilities; it represents available liquid funds. The higher the debt
ratio, the more difficult it becomes to obtain further capital through debt
issue. A ratio in excess of 100 is ordinarily considered excessive.106
The debt ratio for the steel industry, 87.2, is a little under the
recommended maximum. It is, however, significantly higher than the ratios
for other listed industries. This1 may be some indication that the steel
industry has borrowed to its financially practical limit. If this is true,
the imposition of regulatory alternatives that require further capital
investment could be financially damaging to the industry.
9.3.6 Total Cost of Regulatory Alternatives
The purpose of this section is to compute the total cost of each
regulatory alternative. A partial equilibrium analysis is employed to
examine all costs measurable in the market for steel. The cost estimates
are to be interpreted as gross total costs, not net total costs; i.e., they
are costs before the deduction of the monetary value of any environmental
benefits resulting from the regulation.
A simple competitive model of the current domestic steel market is
depicted in figure 9-4. The supply curve (Sx) from existing firms is
upward sloping for the reasons discussed in section 9.2.4.2. Current
market price (P) and output (Q) prevail given demand (D). The market is in
long-run disequilibrium because the average total cost of production for
new, electric arc furnace plants (ATCN) is below the prevailing market
price P. The long-run steel supply schedule is thus elastic (represented
by the horizontal segment SN) and;below the prevailing market price for
steel.
Owners of newly-constructed electric steel mills stand to earn economic
rents as long as the market price for steel exceeds the ATC of the new
plant. The long run equilibrium price (Pe), the price at which investors
no longer have an incentive to enter the market, is thus equal to
assuming no changes in technology, factor prices, or demand.
9-108
-------
Q/time
Figure 9-4. Current steel market disequilibrium
and long-run equilibrium.
9-109
-------
Long-run market disequilibrium is not expected to maintain indefinitely.
In time, any frictional delays or market barriers to entry should diminish,
allowing new plants to come on line. As they do so, market supply will
increase relative to demand, driving market price towards long-run equilibrium
I
price'P .
There is little reason to believe that the United States domestic
steel market will long be in long-run disequilibrium. Many of the
conventionally cited barriers to entry are absent for new electric steel
mills.107 Economies of scale are not apparent; electric arc steel mills,
which have low unit production costs, have low start-up capital costs as
well. Steel is a homogeneous product, so new firms have little difficulty
marketing their product. Electric arc mills require no inputs from sources
held captive by established firms.
Finally, construction of new electric arc plants is currently quite
active and is expected to continue to be brisk. This is evidence both that
there is economic incentive to enter the market and that barriers to entry
are not prohibitive.
The effects of a new source performance standard are illustrated in
figure 9-5. The average total cost of production for a new, electric steel
plant increases from ATCN to ATC^ as a result of the increased capital and
operating costs of the regulation. However, the market is still in a state
of long-run disequilibrium as long as ATC^ < P. Post-regulatory market
long-run equilibrium now attains at P^ and Q^.
Prior to the standard, total consumers' surplus is equal to the value
of the area P be—the area below the demand curve and above long-run
e
equilibrium market price Pe- After the standard, consumers' surplus is
defined by the area P'bd. Thus, the readily apparent effect of the standard
is to reduce consumers' surplus by the amount Pgbc - P^bd = PQP^dc. Consumers
now pay more per unit of output (P^ - Pfi) and consume less (Qe - Q^).
We are now in a position to calculate the total cost of the regu-
lation, but care must be taken to avoid double-counting and including
transfers as a cost. The intuitively most apparent cost of the regulation
is the real resource cost associated with the expenditures by investors in
new plants to meet the standard. New plants that come on line in the
presence of the standard incur higher capital and operating costs than
9-110
-------
Pe*ATCN
Pft=ATCM_
Q
Q/tim©
e
Figure 9-5. NSPS effects on U.S. steel market
long-run equilibrium
9-111
-------
would prevail in the absence of the standard. This real resource cost is
paid by consumers (loss of consumers' surplus) and is equivalent to the
rectangular area gfdh. Consumers pay a higher price per unit of output
equal to P'-P for each of the QS~QQ units produced by new firms. This
component of the cost of the regulation will hereafter be referred to as
the "new plant real resource cost".
A second readily recognizable component of the total cost of the
regulation is the deadweight loss, in consumers' surplus. Consumers' surplus
equal to the area of the triangle hdc is lost as market price increases
from P to P1 and consumers reduce their consumption accordingly.
e e
Thus far a significant portion (area gfdc) of the total reduction in
consumers' surplus (area PgPgdc) has been identified as part of the total
cost of the New Source Performance Standard. Still to be accounted for is
the area P P'fg. As seen in figure 9-5, this area is divided into two
parts: triangle kfg and area PgPgfk.
In the absence of the NSPS, old (preexisting) plants produce output
Q . In the presence of the standard, old plants produce more output—Q^.
This is so because in the presence of the standard the higher market price
makes it profitable for more old plants to continue production. Output
equivalent to Q'-Q is produced by new, low cost plants in the absence of
the standard but by old, higher cost plants with the standard. Thus, real
resources of a value equal to area kfg are used to produce Q0~Q0 units of
output with the NSPS that are saved without the standard. The value of
triangle kfg is thus a real resource cost of the regulation, and will be
referred to as the "old plant real resource cost."
The final component of the total reduction in consumers' surplus to be
accounted for is area P P'fk. This component, it will be shown, is not a
cost of the regulation for some Undetermined length of run but becomes a
real resource cost of the regulation in the very long run.
In the long-run market depicted in figure 9-5, area PgPgfk represents
a transfer from consumers' to producers' surplus. Without the standard,
owners of old plants enjoy rents equivalent to area aPgk. With the standard,
owners of old plants enjoy greater rents—equivalent to area aPQf. The
regulation thus results in a transfer from consumers' to producers' surplus
of P P'fk. This transfer cannot appropriately be considered a cost of the
e e
regulation.
9-112
-------
This situation changes, however, over a sufficiently long period of
time, the actual length of which cannot be determined in this analysis.
Over time, the number of.old plants that continue to produce diminishes.
This might occur, for example, as the unit operating costs of the deteri-
orating plants rise above market price P'.108 As old plants retire, they
are replaced by new plants subject to the NSPS. Eventually then, what had
been a transfer from consumers' to producers' surplus becomes a real
resource cost of the regulation.
To summarize, the total cost of the NSPS is comprised of three com-
ponents in the long run: the new plant real resource cost, the deadweight
loss in consumers' surplus, and the old plant real resource cost. In the
very long run, the new plant real resource cost component increases as old
plants are replaced by new ones and the total cost of the regulation in any
given year is equivalent to the total reduction in consumers' surplus
represented by area PpP'dc in figure 9-5.
The analytical framework described above is used to calculate the
total cost of each regulatory alternative for each model plant. Two esti-
mates of the total cost of each regulation are computed: (1) the long-run
estimates which include the new plant real resource cost, the deadweight
loss in consumers' surplus, and the old plant real resource cost, and (2)
the very long-run estimate which is the sum of the long-run estimate and
the real resource cost of new plants that replace retiring old capacity.
The total annual cost of the regulation is estimated for 1987 assuming
that the long-run, not the very long-run, market will have attained by that
time. The total cost of the regulation presented for 1987 is thus equivalent
to area kfdc in figure 9-5. The long-run baseline equilibrium price (P )
is the ATC of producing one Mg of steel billets in a given model plant
meeting regulatory alternative 1. This price is assumed to prevail in 1987
in the absence of new standards. The baseline equilibrium output (Qe) is
based on the steel projections in section 9.1.5.4. The new equilibrium
market price (P1) under a proposed standard is the ATC of making steel in
the EAF plant meeting the standard; A new equilibrium output (Q^) results
given a demand elasticity of -1.86. The baseline output in 1987 from
plants existing in 1981 (Q ) is based on the projections section 9.1.5.4.
This output is presumed to be forthcoming given P . An estimate of output
9-113
-------
from existing plants given the higher market price (P^) due to the proposed
standard (Q1) is calculated using a supply elasticity of 2.16.109
Using the estimates of Pe> P^, Qe> Q^, QQ and Q^, it is possible to
estimate the 1987 total regulatory cost equivalent to area kfdc for each
alternative assuming each model plant is the "price-setting" plant. The
formulas used to calculate each component of the cost of the regulation
are:
- New Plant Real Resource Cost
= area gfdh
= (Pe-Pe) (Qe-Qd);
• Deadweight Loss In Consumers' Surplus
= area hdc
= % (Pe-Pe) (Qe-Qe);
• Old Plant Real Resource Cost
= area kfg
= h (Pe-Pe) (Q6-Qo);
• Total Cost of Regulation
= area kfdc
= gfdh + hdc + kfg.
Estimates of the total cost of regulatory alternative 2 in 1987 are
reported in Table 9-41. Estimates for specialty steel shops vary from
a $0.216 million for model plants 8 and 10-to a high of $0.773 million for
model plant 7. Estimates of the total cost of alternative 2 for carbon
shops vary from a low of $0.608 million for model plant 4 to a high of
$1.096 million for model plant 2. The new plant real resource cost regular-
ly comprises the major part of the total cost of the regulation in this
long run period.
Greater total costs are associated with regulatory alternative 3.
Table 9-42 indicates that total costs for carbon plants vary .between $1.415
million and $2.666 million, excluding the negative costs for model plants
1 and 6 with a total furnace enclosure. Total costs for specialty steel
plants range from 0.802 to $1.160 million.
In the very long run, all plants that are producing steel today will
have been retired. As output frbm these existing plants declines, it will
be replaced by output from new plants. It is difficult to speculate whether
EAF plants will continue in the very long run to be the best technology.
9-114
-------
TABLE 9-41. TOTAL COST OF REGULATORY ALTERNATIVE 2
IN 1987a '
:' '•"•:• ...";..-• •'. ..'/•' :. •'•.::."•;;'• • . •..:. 106 1981 $ . , .
Model
Plant
1
lb
2
3 •
4
5
6
6b
7
8
9
10
New Plant
Real Resource
Cost
0.696
na
1.049
0.682 .
0.592
0.723
0.786
na
0.737
0.213
0.338
0.213
Deadweight
Loss In
Consumers' •', Surplus
0.011
na
0.026
0.013
0.008
0.016
'
0.015.
na
0.020 •
0.002
0.005
0.002
Old Plant
Real Resource
. Cost
: 0.006 .':-..'•
: na ;
0,021
'."'• 0:011 • ;"- ';
:0;;008 '•.--.. ;.";
:o.oi3
0;008
na
0.016 .
0.001
0.005
0.001
Total
Cost Of
Regulation
0.713; :-
.na
,1.096
0.706 ;,"
^ 0.608 ='
0.752
. 0.809\
na
0.773
0.216
0.348
0.216
aModel parameters: Q = 101.7 x 106 Mg (carbon) •
e = 18.0 x 10s Mg (specialty)
Q = 96.5 x 106 Mg (carbon)
°= 17.0 x 106 Mg (specialty)
Total Furnace Enclosure Option.
NOTE: na = not applicable.
9-115
-------
TABLE 9-42. TOTAL COST OF REGULATORY ALTERNATIVE 3
IN 1987a
106 1981 $
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
aModel
New Plant
Real Resource
Cost
2.077
0
-4.422
1.691
1.780
1.269
1.818
2.108
-4. 549
0.230
0.572
0.562
0.562
parameters: Qe =
Qo =
Deadweight Old Plant
Loss In Real Resource
Consumers' Surplus Cost
0.133
-0.122
0.056
0.150
0.051
0.178
0.158
-0.024
0.125
0.037
0.043
0.042
101.7 x 10^ Mg
18.0 x 106 Mg
96.5 x 106 Mg
-• — • »«i -t i«\ £5 ti_-
0.363
-0.426
0.061
0.260
0.095
0.272
0.400
-0.483
0.805
0.207
0.216
0.198
(carbon)
(specialty)
(carbon)
Total
Cost Of
Regulation
2.573
-4.970
1.808
2.190
1.415
2.288
2,666
-5.056
1.160
0.816
0.821
0.802
Total furnace enclosure option.
9-116
-------
However, if they do, the new plant real resource cost will increase by an
amount equal to the area PgP'fk in figure 9-5. This occurs because new,
replacement plants are (presumably) subject to the standard indefinitely.
To approximate the annual, very long-run total cost of the regulation,
area P P'dc is computed for 1987. The formula is: Total Cost In The Very
Long Run = (Pe-Pe) Qe + h (Pe-Pe) (Qe-Qe). This is an underestimate of the
long run total cost if in fact demand should shift rightward beyond 0
during the projection period. The estimates are valuable because they
emphasize that in the very long run, the total decrease in consumers'
surplus, equivalent to the entire area under the demand curve between the
regulated and the unregulated prices, becomes a real resource cost of the
regulation.
Table 9-43 reports the total cost estimates for alternatives 2 and,3
in the very long run. For regulatory alternative 2, the estimates range
from a low of $12.195 million to a high of $22.347 million for carbon steel
plants and from $3.958 million to $14.560 million for specialty steel
plants. These very long run annual costs of regulation are considerably
higher than the costs net of replacement plant real resource costs. Under
regulatory alternative 3, the maximum cost estimate is $62.896 million for
model plant 6 (without total furnace enclosure).
Model plants 4 and 7 are believed to best typify future plant con-
struction. The anticipated 1987 total cost of regulatory alternative 2 is
thus $0.608 million for carbon plants and $0.773 million for specialty
plants for a total cost of $1.381 million. The anticipated 1987 total cost
of alternative 3 is $1.415 million for carbon plants and $1.160 million for
specialty plants for a total of $2.575 million. These 1987 anticipated
regulatory cost estimates are calculated assuming long-run, not very long-
run, market conditions prevailing in that year. That is, these estimates
do not include the real resource cost incurred as new, controlled plants
eventually retire old, uncontrolled plants.
9-117
-------
TABLE 9-43. TOTAL COST OF REGULATORY ALTERNATIVES
IN THE VERY LONG RUN3
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
2
14. 228
na
22.347
14.226
12.195
15.240
16.257
na
14. 560
3.958
6.474
' 3.958
106 1981 $
Regulatory Alternative
3
59.871
-65.453
37,574
50.700
30.459
53.724
62.896
-69. 228
37.494
19.043
19.784
18.679
aBased on projected 1987 demand conditions.
Total furnace enclosure option.
NOTE: na = not applicable.
9-118
-------
9.4 REFERENCES FOR CHAPTER 9
1. Scherer, F. M. Industrial Market Structure and Economic Performance.
Chicago, Rand McNally, 1970. pp. 3-6.
2. U.S. Department of Commerce, Bureau of the Census. 1977 Census of
Manufactures. Washington, D.C., U.S. Government Printing Office,
• August 1980. p. 33A-2.
3. Reference 2, p. 33A-5.
4. U.S. Department of Commerce, Bureau of Economic Analysis. Survey
of Current Business. Washington, D.C., U.S. Government Printing
Office, July 1978. pp. 24, 48, 54.
5. Reference 2, p. 29A-5, 37A-5.
6. Annual Statistical Report, 1979. Washington, D.C., American Iron and
Steel Institute, 1979. pp. 92-93.
7. U.S. Department of Commerce, Bureau of Economic Analysis. The
Detailed Input-Output Structure of the U.S. Economy; 1972. Magnetic
Computer Tapes. Washington, D.C.
8. Rowley, Charles K. Steel and Public Policy. London, McGraw-Hill,
1971. pp. 24-25.
9. Reference 6, p. 55.
10. Steel-Coal Basic Analysis. Standard and Poor's Industry Surveys.
October 11, 1979. p. S53.
11. Continuous Casting in Steel. The New York Times. January 1, 1981.
p. 28.
12. Steel at the Crossroads: The American Steel Industry in the 1980's.
Washington, D.C., American Iron and Steel Institute, January 1980.
p. 35.
13. Standard Industrial Classification Manual, 1972. Statistical Policy
Division, Office of Management and Budget.- Washington, D.C., U.S.
Government Printing Office, pp. 145-146.
14. Reference 2, p. 33A-19.
15. Vaughan, W. J. A Residuals Management Model of the Iron and Steel
Industry: A Linear Programming Approach. Ph.D. dissertation, George-
town University, 1975. pp. 202-203.
16. Reference 8, p. 27.
17. Reference 12, pp. 36-37.
9-119
-------
18. Reference 15, pp. 203-204.
19. Reference 8, p. 28.
20. U.S. Department of Commerce, Patent and Trademark Office. Tech-
nology Assessment and Forecast. Wasington, D.C., U.S. Government
Printing Office, March 1979. p. 120.
21. Congress of the United States, Office of Technology Assessment. Tech-
nology and Steel Industry Competitiveness. Washington, D.C., U.S.
Government Printing Office, June 1980. p. 267.
22. Reference 21, p. 240.
23. Reference 6, p. 72.
24. Reference 15, p. 203.
25. Shriner, R. D. An Econometric Analysis of the Demand and Supply for
Scrap Iron and Steel. Ph.D. dissertation, Indiana University, 1974.
p. 26.
26. U.S. Department of Commerce, Bureau of Economic Analysis. Survey
of Current Business. Washington, D.C., U.S. Government Printing
Office, September 1980. p. 10.
27. U.S. Department of Commerce, Bureau of Economic Analysis. Business
Statistics, 1977. Washington, D.C., U.S. Government Printing Office,
March 1978. p. 4.
28. Reference 6, p. 8.
29. Reference 6, p. 35.
30. Reference 10, p. S55.
31. Reference 10, p. S56.
32. Alexander, W. 0. The Competition of Materials. Scientific American.
September 1977. pp. 255-266.
33. U.S. Department of Commerce, Bureau of the Census. 1972 Census of
Manufactures. Washington, D.C., U.S. Government Printing Office,
August 1976. p. 33A-10.
34. Reference 2, p. 33A-9.
35. Little Giants. The Wall Street Journal. January 12, 1981. p. 1.
36. Steel Industry Quarterly Review. Merrill, Lynch, Pierce, Fenner, and
Smith, Inc. February 1980. pp. 66-69.
9-120
-------
37. Reference 33, p. 33A-14.
38. Reference 2, pp. 33A-9 - 33A-10.
39. Mueller, Hans, and Kawahito, Kiyoshi. Steel Industry Economics—A
Comparative Analysis of Structure, Conduct, and Performance,
New York, International Public Relations Co., Ltd., 1978. p. 9.
40. U.S. General Accounting Office. New Strategy Required for Aiding Dis-
tressed Steel Industry. Washington, D.C., U.S. Government Printing
Office, January 1981. pp. 2-16 - 2-18.
41. Annual Statistical Report, 1970. Washington, D.C., American Iron and
Steel Institute, 1971. p. 8.
42. Reference 6, p. 17.
43. Reference 21, p. 133.
44. Reference 40, pp. 4-12 - 4-13.
45. Reference 40, p. 2-9.
46. Reference 8, pp. 146-147.
47. Nicholson, Walter. Microeconomic Theory, Basic Principles and Exten-
sions. Hinsdale, 111., The Dryden Press, Inc., 1972. p. 258.
48. Reference 1, p. 197.
49. Reference 1, pp. 190-191.
50. Duke, Richard M., et al. The United States Steel Industry and
Its International Rivals: . Trends and Factors Determining Inter-
national Competitiveness. Bureau of Economics, Federal Trade
Commission. Washington, D.C., U.S. Government Printing Office,
November. 1977. p. 63.
51. Reference 47, p. 260.
52. Reference 36, p. 71.
53. Reference 47, p. 292.
54. Reference 47, p. 266.
55. Adams, Walter, ed. The Structure of American Industry. New York,
Macmillan, 1977. pp. 86-89.
56. Reference 47, p. 261.
9-121
-------
57. Smolik, J. E. An Empirical Study of the Price Responsiveness of
Domestically Produced Steel Products to Import Competition: 1954-
1967. Ph.D. dissertation, University of Illinois at Urbana-Champaign,
1973. p. 5.
58. Reference 55, p. 90.
59. Henderson, James M., and Quandt, Richard E. Microeconomic Theory,
A Mathematical Approach. New York, McGraw-Hill, 1971. pp. 222-229.
60. Reference 1, p. 131.
61. Reference 47, pp. 313-315.
62. Reference 55, p. 107.
63. Reference 50, p. 160.
64. Reference 40, p. 3-4.
65. Reference 50, p. 170.
66. Reference 35, p. 1, p. 19.
67. Copeland, Thomas E., and Weston, J. Fred. Financial Theory and Cor-
porate Profit. Reading, Mass, Addison-Wesley Publishing Co., 1979.
pp. 20-21.
68. Ammer, Christine, and Ammer, Dean S. Dictionary of Business and Eco-
nomics. New York, The Free Press, 1977. p. 138.
69. Reference 21, p. 120.
70. Reference 6, p. 13.
71. Monthly Economic Letter. New York, Citibank Corp., April 1981.
p. 10.
72. Key Business Ratios. New York, Dunn and Bradstreet, 1978. p. 2.
73. Reference 55, pp. 120-122.
74. Moody's Industrial Manual. New York, Moody's Investors Service, Inc.,
1980. Vols. 1,2. Various pages.
75. Jones, Charles P. Ratio Techniques of Financial Analysis. Research
Triangle Park, N.C., Research Triangle Institute. RTI Working Paper.
1980.
76. Reference 72, pp. 65, 78, 80-81, 109.
9-122
-------
77. U.S. Department of Commerce, Bureau of Economic Analysis. Survey
of Current Business. Washington, D.C., U.S. Government Printing
Office, October 1980. p. S-2.
78. Reference 27,. p. 19.
79. Reference 27, p. 5.
80. "Reference 77, pt 15,
81. Reference 27, p. 12.
82. Reference 4, p. S-^2. ; ,
83. Reference 12, pp. 17-23, 69.
84. Reference 55, p. 89.
85; Reference 6, p. 25.
86. U.S. Department of Labor, Bureau of.Labor Statistics. Handbook of
Labor Statistics, 1975. Washington, D.C.* U.S. Government Printing
Office, 1975. p. 187. ;
87. U.S. Department of Labor, Bureau of Labor Statistics. Employment
and Earnings, December 1979. Washington* D.C.'-, U.S. Government
Printing Office, 1979. p. 105. V
88. Environmental Policy for the 1980's: Impact on the American Steel
Industry. Report to the American Iron and Steel Institute. Arthur D.
Little, 1981. Table 1-4.
89. Telephone communication. Anderson, Donald W., Research Triangle
Institute, to Avery, Peter, American Iron and Steel Institute.
June 19, 1981.
90. Technology Leads the Way as Electric Furnace Steel making Heads for New
Heights in the U.S. 33 Metal Producing. July 1980.
91. U.S. Department of Commerce, Bureau of Industrial Economics. 1981
U.S. Industrial Outlook. Washington, D.C., U.S. Government.Printing
Office, January 1981. pp. 199-205.
92. Chapter 3 of this report-
93. Letter and attachments from Banker, L., Midwest Research Institute, to
Anderson, D., Research Triangle Institute. November 12, 1981.
94. Bussey, Lynn E. The Economic Analysis of Industrial Projects. Engle-
wood Cliffs, N.J., Prentice-Hall, Inc., 1978. pp. 220-223.
95. Van Home, James C. Financial Management and Policy. Englewood
Cliffs, N.J., Prentice-Hall, Inc., 1980, pp. 115-116.
9-123
-------
96. Jones, Charles P. The Cost of Capital. Research Triangle Park, N.C.,
Research Triangle Institute. RTI working paper. 1981.
97. Kamerschen, David R., and Valentine, Lloyd M. Intermediate Microeco-
nomic Theory. Cincinnati, South-Western Publishing Co., 1977.
pp. 452-453.
98. U.S. Department of Labor, Bureau of Labor Statistics. Producer Prices
and Price Indexes. Washington, D.C., U.S. Government Printing Office,
March 1981. p. 51.
99. Tax Foundation, Inc. Facts and Figures on Government Finance.
Washington, D.C. 1981. pp. 134-135, p. 234.
100. Giberson, L., et al. Economic Impact of NSPS Regulations on Coke Oven
Battery Stacks. Research Triangle Park, N.C., Research Triangle
Institute. - May 1980. pp. 8-45 - 8-47.
101. Weaver, James B. Project Selection in the 1980's. Chemical and
Engineering News. November 2, 1981. pp. 37-46.
102. Ramachandran, V. An Econometric Model of the U.S. Steel Industry.
Research Triangle Park, N.C., Research Triangle Institute. April
1981. pp. 10.
103. Reference 102, p. 12.
104. Reference 6, p. 44.
105. Salter, W. E. G. Productivity and Technical Change. Cambridge,
Cambridge University Press, 1969, pp. 48-82.
106. Reference 72, pp. 2-3.
107. Porter, Michael E. Competitive Strategy. New York. Macmillan
Publishing Co. 1980. pp. 7-13.
108. Reference 105, p. 51. ;
109. Reference 102, p. 15.
9-124
-------
APPENDIX A.
EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
Standards of performance for electric arc furnaces (EAF's) in the
steel industry were promulgated in October 1974. The four-year review
of these standards, required by Section lll(b) of the Clean Air Act as
amended August 1977, was initiated by EPA in March 1979. Based on the
results of an initial study, the decision was reached to pursue develop-
ment of a Background Information Document (BID) to determine if revision
of .the standards was necessary.
In April 1980, an effort was begun to obtain the information needed to
develop the BID. This information gathering has included literature surveys;
canvassing of State, regional, and local air pollution control agencies;
plant visits; meetings with industry representatives and associations;
contact with engineering consultants and equipment vendors; and emission
source testing. Significant events relating to the evolution of the BID
are itemized in Table A-l.
A-l
-------
TABLE A-l. EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
Date
Company, consultant, or agency/
Location
Nature of action
10/79
04/21/80
06/13/80
06/23-
26/80
06/25/80
07/02/80
07/02/80
07/02/80
07/10/80
07/10/80
07/15/80
07/16/80
07/22/70
07/23/80
07/24/80
U.S. Environmental Protection Agency
Research Traingle Park, N.C.
U.S. Environmental Protection Agency
Research Triangle Park, N.C.
Jones and Laugh!in Steel Corporation
Pittsburgh, Pa.
Jones and Laugh!in Steel Corporation
Pittsburgh, Pa.
Florida Steel Corporation
Charlotte, N.C.
Allegheny Ludlum Steel Corporation
Brackenridge, Pa.
Armco, Incorporated
Butler, Pa.
Washington Steel Company
Washington, Pa.
AL Tech Specialty Steel Corporation
Watervliet, N.Y.
Colt Industries, Incorporated
Syracuse, N.Y.
Allegheny Ludlum Steel Corporation
Brackenridge, Pa.
Armco, Incorporated
Butler, Pa.
Washington Steel Company
Washington, Pa.
AL Tech Specialty Steel Corporation
Watervliet, N.Y.
Colt Industries, Incorporated
Syracuse, N.Y.
"Review of Standards
of Performance for
Electric Arc
Furnaces in Steel
Industry" published.
Federal Register
announcement of
notice of intent to
explore revision of
existing standards
Section 114 letter
Observation of emission
test
Section 114 letter
Section 1!4 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Site visit
Site visit
Site visit
Site visit
Site visit
(continued)
A-2
-------
TABLE.A-1. (continued)
Date
Company, consultant, or agency/
Location
Nature of action
08/05/80 Florida Steel Corporation
Charlotte, N.C.
08/28/80 American Irorr-and Steel Institute
Durham, N.C.
09/10/80 Eastern Stainless Steel Company
Baltimore, Md.
09/10/80 Ford Motor Company
Dearborn, Mich.
09/23/80 Ford Motor Company
Dearborn, Mich.
09/24/80 Eastern Stainless Steel Company
Baltimore, Md.
10/02/80 AL Tech Specialty Steel Corporation
Watervliet, N.Y.
10/02/80 Colt Industries, Incorporated
Syracuse, N.Y.
10/08/80 Colt Industries, Incorporated
Syracuse, N.Y.
10/09/80 AL Tech Specialty Steel Corporation
Watervliet, N.Y. ' :
10/15/80 Babcock and Wilcox Company
Beaver Falls, Pa.
10/15/80 North Star Steel Company
Monroe, Mich.
10/21/80 Inter!ake Incorporated
Riverdale, 111.
10/21/80 Rob!in Steel Company
North Tonawanda, N.Y.
10/23/80 North Star Steel Company
Monroe, Mich.
10/24/80 Babcock and Wilcox Company
Beaver Falls, Pa.
10/24/80 Eastern Stainless Steel Company
Baltimore, Md.
10/28/80 Eastern Stainless Steel Company
Baltimore, Md.
Site visit
Meeting to discuss
project
Section 114 letter
Section 114 letter
Site visit
Site visit
Section 114 letter
Section 114 Tetter
Pretest survey
Pretest survey
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Site visit
Site visit
Section 114 letter
Pretest survey
(continued)
A-3
-------
TABLE A-l. (continued)
Date
Company, consultant, br agency/
Location
Nature of action
11/10- Hoeganaes Corporation
11/80 Gallatin, Tenn.
11/25/80 Colt Industries, Incorporated
Midland, Pa.
11/25/80 North Star Steel Company
St. Paul, Minn.
11/25/80 Republic Steel Corporation
Cleveland, Ohio
11/25/80 Sharon Steel Corporation
Parrel!, Pa.
11/25/80 Northwestern SteeT and Wire Company
Sterling, 111.
11/25/80 Nucor Steel Corporation
Darlington, S.C. ;
11/25/80 Raritan River Steel Company
Perth Amboy, N.J.
12/09/80 Eastern Stainless Steel- Company
Baltimore, Md.
12/09/80 North Star Steel Company
Monroe, Mich.
12/16/80 AL Tech Specialty Steel Corporation
Watervliet, N.Y.
12/17- Eastern Stainless Steel Company
18/80 Baltimore, Md.
12/18- North Star Steel Company
19/80 Monroe, Mich.
01/20/81 Carpenter Technology'Corporation
Reading, Pa. !
01/22/81 Carpenter Technology Corporation
Reading, Pa.
02/17/81 American Iron and Steel Institute
Washington, D.C.
02/17/81 U.S. Steel Corporation
Pittsburgh, Pa.
02/19/81 Andersen Samplers, Incorporated
Altanta, Ga.
Observation of emission
test
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
Section 114 letter
'Visible emission
source test
Observation of emission
test
Section 114 letter
Pretest survey
Request for information
on opacity monitors
Section 114 letter
Request for information
on opacity monitors
(continued)
A-4
-------
TABLE A-1. (continued)
Date
Company, consultant, or agency/
Location
Nature of action
02/19/81 Bethlehem Steel Corporation
Los Angeles, Calif.
02/19/81 Chaparral Steel Corporation
Midlothian, Tex.
02/19/81 Contraves Goerz Corporation
Pittsburgh, Pa.
02/19/81 Datatest, incorporated
Levittown, Pa.
02/19/81 Dynatron, Incorporated
Wallingford, Conn.
02/19/81 Environmental Data Corporation
Monrovia, Calif.
02/19/81 Lear Siegler, Incorporated
Englewood, Colo.
02/24/81 Atlantic Steel Company
Atlanta, Ga.
03/02- Chaparral Steel Corporation
03/81 Midlothian, Tex.
03/05- U.S. Steel Corporation
06/81 Baytown, Tex.
03/11- Atlantic Steel Company
12/81 Cartersville, Ga.
03/25/81 Sholtes and Koogler
Gainesville, Fla.
04/06- AL Tech Specialy Steel Corporation
09/81 Watervliet, N.Y.
04/21/81 Bethlehem Steel Corporation
Los Angeles, Calif.
04/23/81 Carpenter Technology Corporation
Reading, Pa.
04/27- Carpenter Technology Corporation
30/81 Reading, Pa.
05/05- Bethlehem Steel Corporation
06/81 Los Angeles, Calif.
07/22/81 Rob!in Steel Company
North Tonawanda, N.Y.
Section 114 letter
Section 114 letter
Request for information
on opacity monitors
Request for information
on opacity monitors
Request for information
on opacity monitors
Request for information
on opacity monitors
Request for information
on opacity monitors
Section 114 letter
Visible emission
source test
Visible emission
source test
Visible emission
source test
Request for information
on opacity monitors
Emission source test
Section 114 letter
Section 114 letter
Emission source test
Visible emission
source test
Section 114 letter
(continued)
A-5
-------
TABLE A-l. (continued)
Date
Company, consultant, or agency/
Location
Nature of action
08/18/81 American Iron and Steel institute
Washington, D.C.
09/30/81 Al Tech Specialty Steel Corporation
Watervliet, N.Y.
09/30/81 Allegheny Ludlum Steel Corporation
Brackenridge, Pa.
09/30/81 American Iron and Steel Institute
Washington, D.C.
09/30/81 Armco, Incorporated
Butler, Pa.
09/30/81 Atlantic Steel Company
Atlanta, Ga.
09/30/81 Bethlehem Steel Corporation
Los Angeles, Calif.
09/30/81 Carpenter Technology Corporation
Reading, Pa.
09/30/81 Chaparral Steel Corporation
Midlothian, Tex.
09/30/81 Eastern Stainless Steel Company
Baltimore, Md.
09/30/81 Sholtes and Koogler
Gainesville, Fla.
09/30/81 Ferrco Engineering, Limited
Whitby, Ontario, Can.
09/30/81 Ford Motor Company
Dearborn, Mich.
Request for information
on EAF's
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5j and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for-comment on
draft BID Chapters 3,
4, 5, and 6
(continued)
A-6
-------
TABLE A-1. (continued)
Date
09/30/81
09/30/81 >
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
09/30/81
Company, consultant, or agency/
Location
Interlake, Incorporated
RiverdaTe, 111.
Jones and Laugh! in Steel Corporation
Pittsburgh, Pa.
North Star Steel Company
•;•*• Monroe, Mich.
" ' '" .••.•'.:• •"-"•'.
North Star Steel Company
St. Paul, Minn.
... ' ' '}
Nucor Steel Corporation
Darlington, S.C.
Obehchain Calumet Corporation
Gary, Ind. V
Pennsylvania Engineering Corporation
, Pittsburgh, Pa.
Raritan River Steel
Perth Amboy, N. J.
' '. .'..•
Roblin Steel Company
North Tonawanda, N.Y.
Union Carbide Corporation
Danbury, Conn.
Union Carbide Corporation
Tarry town, N.Y.
U.S. Steel Corporation
Pittsburgh, Pa.
Washington Steel Company
Washington, Pa.
Nature of action
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request .for comment on
draft BID Chapters, 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6 ', .
Request., for comment on
draft BID Chapters 3,
4, 5, and 6
Request for cdmment on
draft BID Chapters 3,
4, 5, and 6
Request for ; comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4 , 5 , and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Request for comment on
draft BID Chapters 3,
4, 5, and 6
(continued)
A-7
-------
TABLE A-1. (continued)
Date
Company, consultant, or agency/
Location
Nature of action
10/13/81 Colt Industries, Incorporated
Midland, Pa.
10/13- Colt Industries, Incorporated
20/81 Midland, Pa.
10/21- Rob!in Steel Company
23/81 Durkirk, N.Y.
11/05/81 American Iron and Steel Institute
Durham, N.C.
12/11/81 Kaiser Steel Corporation
Fontana, Calif.
01/06/82 Bayou Steel Corporation
LaPlace, La.
01/12/82 Hatch Associates, Limited
Toronto, Ontario, Can.
02/03/82 Carpenter Technology Corporation
Reading, Pa.
02/03/82 Jones and Laugh!in Steel Corporation
Pittsburgh, Pa.
02/10/82 Carpenter Technology Corporation
Reading, Pa.
02/18/82 Atlantic Steel Company
Atlanta, Ga.
02/24/82 Atlantic Steel Company
Cartersville, Ga.
03/09- Jones and Laugh!in Steel Corporation
10/82 Pittsburgh, Pa.
04/06/82 Dr. Sholtes, Sholtes and Koogler,
Environmental Consultants
Durham, N.C. !
07/21- National Air Pollution Control
22/82 Techniques Advisory Committee
Dallas, Texas
07/23/82 Chaparral Steel Company
Midlothian, Texas
Section 114 letter
Observation of emission
test
Observation of emission
test
Meeting to discuss
project
Request for comment on
draft BID Chapters 3,
4, 5, 6, 7, and 8
Section 114 letter
Request for comment on
draft BID Chapters 3,
4, 5, and 6
Section 114 letter
Section 114 letter
Visible emission
source test
Section 114 letter
Site visit
Visible emission
source test
Meeting to discuss
project
Meeting to discuss
recommended standard
Site visit
(continued)
A-8
-------
TABLE A-l. (continued)
Date
Company, consultant, or agency/
Location
Nature of action
08/06/82 Dr. Sholtes, Sholtes and Koogler,
Environmental Consultants
Durham, N.C.
08/12/82 Chaparral Steel Company
Midlothian, Texas
08/23- Chaparral Steel Company
24/82 Midlothian, Texas
Meeting to discuss
project
Section 114 letter
Visible emission
source test
A-9
-------
-------
APPENDIX B
INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS
This appendix consists of a reference system, cross-indexed with
the October 21, 1974,,Federal Register (39 FR 37419) containing the Agency
guidelines concerning the preparation of environmental impact statements.
This index can be used to identify- sections of the document which contain
data and information germane to any portion of the Federal Register
guidelines.
B-l
-------
TABLE B-l. CROSS-INDEXED REFERENCE SYSTEM TO HIGHLIGHT
ENVIRONMENTAL IMPACT PORTIONS OF THE DOCUMENT
Agency guidelines for preparing
regulatory action environmental
impact statements (39 FR 37419)
Location within the Background
Information Document
BACKGROUND AND SUMMARY OF
REGULATORY ALTERNATIVES
Summary of regulatory alternatives
Statutory basis for proposing
standards
Relationship to other regulatory
agency actions
Industry affected by the
regulatory alternatives
Specific processes affected by
the regulatory alternatives
2. REGULATORY ALTERNATIVES
Control techniques
The regulatory alternatives from
which standards will be chosen
for.proposal are summarized
in Chapter 1, Section 1.1.
The statutory basis for proposing
standards is summarized in
Chapter 2, Section 2.1.
The relationships between the
regulatory agency actions are
discussed in.Chapters 3, 7,
and 8.
A discussion of the industry
affected by the regulatory
alternatives is presented in
Chapter 3, Section 3.1. Further
details covering the business
and economic nature of the
industry are presented in
Chapter 9, Section 9.2.
The specific processes and
facilities affected by the
regulatory alternatives are
summarized in Chapter 1,
Section 1.1. A detailed technical
discussion of the processes
affected by the regulatory
alternatives is presented in
Chapter 3, Section 3.2.
The alternative control techniques
are discussed in Chapter 4,
Sections 4.2, 4.3, 4.4, and 4.5.
(continued)
B-2
-------
TABLE B-l (continued)
Agency guidelines for preparing
regulatory action environmental
impact statements (39 FR 37419)
Location within the Background
Information Document
Regulatory alternatives
3.
ENVIRONMENTAL IMPACT OF THE
REGULATORY ALTERNATIVES
Primary impacts directly
attributable to the regulatory
alternatives
Secondary or induced impacts
4. OTHER CONSIDERATIONS
The various regulatory alterna-
tives, including "no additional
regulatory action," are defined
in Chapter 6, Section 6.2. A
summary of the major alternatives
considered is included in
Chapter 1, Section 1.1.
The primary impacts on mass
emissions and ambient air quality
due to the alternative control
systems are discussed in
Chapter 7, Sections 7.1, 7.2, 7.3,
7.4, and 7.5. A matrix
summarizing the environmental
impacts is included in Chapter 1.
Secondary impacts for the various
regulatory alternatives are
discussed in Chapter 7,
Sections 7.1, 7.2, 7.3, 7.4, and
7.5.
A summary of the potential
adverse environmental impacts
associated with the regulatory
alternatives is included in
Chapter 1, Section 1.2, and
Chapter 7. Potential socio-
economic and inflationary impacts
are discussed in Chapter 9,
Section 9.3. Irreversible and
irretrievable commitments of
resources are discussed in
Chapter 7, Section 7.6.
B-3
-------
-------
, APPENDIX C. EMISSION TEST DATA FO;R FABRIC FILTERS
'.! ON ELECTRIC ARC FURNACES AND ARGON-OXYGEN DECARBURIZATION
; : VESSELS IN THE STEEL INDUSTRY
This appendix presents the available emission test data for steel
industry electric arc furnaces (EAF's) and argon-oxygen decarburizatioh
(AOD) vessels. The particulate matter emission data and visible emission
(VE) observations for carbon steel plants are presented in Section C.I.
Section C.2 presents the particulate matter emission data and VE observations
for EAF's and AOD vessels in the specialty steel segment of the steel
industry. Section C.3 presents the available VE data for the dust-hand!ing,
systems at both carbon and specialty steel shops.
The particulate matter emission data presented in this appendix
were obtained using Reference Method 5 and draft Reference Method 5D.
Draft Reference Method 5D involves the use of a Method 5 sampling train
to sample sites that do not meet Reference Method 1 criteria. Discussion
of this test method is presented in Appendix D. The visible emissions
data presented in this appendix were obtained using Reference Methods 9
and 22 and continuous opacity monitors. Table C-l contains a summary of
the data from, and information about, the 15 carbon steel shops and
4 specialty steel shops that are presented in this appendix.
Another test method that has been used to sample EAF and AOD vessel
emissions from fabric filters involves the use of high volume (hi-vol)
samplers, which are commonly used for ambient air sampling. However,
the hi-vol sampling method is not recommended by EPA because there are
problems with the sampling methodology. Although the hi-vol method was
used at several steel mills to determine compliance with applicable
State and Federal regulations, the hi-vol data will not be used in
determining any revisions to the standard.
C-l
-------
C.I EMISSION TEST DATA FOR FABRIC FILTERS ON ELECTRIC ARC FURNACES IN THE
CARBON STEEL INDUSTRY
C.I.I Particulate Matter and Visible Emissions
This section provides a summary of participate matter emission
sampling conducted at eight carboy steel shops. The exhaust concentration
of particulate matter emissions was measured at Plants A through H by
Reference Method 5. The emission data on these plants are summarized in
Figure C-l. In addition, the concentration of particulate matter emissions
was measured by the hi-vol sampling technique at six plants; however,
the data will not be used in support of a standard.
The VE observations were made using Method 9. The VE data from
fabric filters on plants with best demonstrated control technology,
Plants B, C, F, G, H, I, J, and K, are summarized in Table C-2. The
available VE data from shop roof monitors on Plants C, G, H, I, J, and K
are summarized in Table C-3.
C.I.1.1 Plant A. Plant A has two furnaces; one of 45.4-megagram
(Mg) (50-ton) melting capacity and the other of 68-Mg (75-ton) melting
capacity. The heat lengths vary from 4 to 8 hours depending on the
grade of steel produced. The emissions from the furnaces are captured.
by a canopy built into the shop roof. The shop roof is closed so the
hoods above the furnaces act to evacuate the melt shop fully, and the
two canopy hoods are connected to a common duct leading to the fabric
filter. A single positive-pressure fabric filter rated at 217 cubic
meters per second (ms/s) (460,000 actual cubic feet per minute [acfm])
flow rate is used to clean the exhaust gases.1 The 12-compartment
fabric filter is equipped with Type 55 Dacron® filter bags and has a net
air-to-cloth ratio of 3.2:I.2 The pressure drop across the bags is 0.87
to 1.02 kilopascals (kPa) (3.5 to 4.1 inches [in.] water gauge [w.g.]).
The fabric filter has six stub stacks of which three were sampled
for particulate matter emissions. The plant was operating near design
capacity during the tests, and the control system was operating normally.
Three 4-hour tests were performed at the inlet duct upstream of the fans
and at three of the fabric filter outlet stacks. The emissions were
C-2
-------
fir/dscf
0.005 -
0.004 -
C.003 -
s:
o
1
LU
g
O
0.002 -
0.001 .
PLANT
REFERENCE
mq/dscm
12
10 1
1
1
1
1
1
1
1
1
-8 1
1
1
1
1
1
II '
* .
• 6 M He
1 1 1
1 ,.
'!
II -
l! 1
RO i
1 1
t 1 j
-•If) p ?. '
! i « i
^ < P !
II 4 ! :
l D ^ O 1
cJ ^ 7 /! ;
•2 i fe t?
i d oo ,
i i
o c
A8BCDEFG
13456789
KEY
0 TEST RUN
1 1 AVERA.GE
J O
1
H
10
Figure C-l. Summary of particulate matter source test
data from fabric filters on EAF's at carbon steel shops.
C-3
-------
measured using Method 5 (with the single exception that the probe was
not heated).
Detailed results of the particulate matter emission tests are
presented in Tables C-4 and C-5. The inlet particulate matter concen-
trations determined from the three samples were 91.0, 136.9, and
141.7 milligrams per dry standard cubic meter (mg/dscm) (0.0397, 0.0597,
and 0.0618 grains per dry standard cubic foot [gr/dscf]), with an average
inlet-concentration of 123.1 mg/dscm (0.0537 gr/dscf). The outlet
particulate matter emission data determined from the three samples on
three of the six stacks were 4.81, 2.21, and 2.29 mg/dscm, averaging
3.11 mg/dscm (0.0021, 0.00097, and 0.001 gr/dscf, averaging 0.0014 gr/
dscf).1
C.1.1.2 Plant B. Plant B has two EAF'S, one rated at 22.7 Mg
(25 tons) and another rated at 36.3 Mg (40 tons).3 Each furnace has a
side draft hood to capture emissions during melting and refining. A
canopy hood is located directly above each furnace to capture fugitive
emissions from charging and tapping. The roof of the shop is closed;
however, the side walls on the shop extend to within 10 feet of the
ground and the two ends of the building are open. This creates cross-
drafts within the shop.
The emissions from both furnaces are controlled by two negative-
pressure fabric filters (Nos. 3 and 4). Fabric filter No. 3 serves the
side draft hoods on both furnaces during the melting and refining phases
of a heat. Fabric filter No. 4 ;is designed to control the fugitive
emissions that are captured by the canopy hoods. During charging and
tapping, the air flow through the side draft hoods and fabric filter
No. 3 is redirected automatically to the canopy hoods above each furnace,
thereby supplementing the air flow to fabric filter No. 4. Fabric filter
No. 3 has 14 compartments and is rated at 62.2 m3/s (132,000 acfm) flow
rate with a gross air-to-cloth ratio of 2.37:1. Fabric filter No. 4 is
rated at 42.9 ms/s (91,000 acfm) flow rate and has a gross air-to-cloth
ratio of 2.69:1.11 The pressure drop across the bags in both fabric
filters is 1.01 to 2.7 kPa (4 to 11 in. w.g.)11 Both fabric filters are
equipped with Dacron® polyester!bags that are cleaned with a shaker-type
mechanism.
C-4
-------
The emission tests were run on the new 36.3-Mg (40-ton) EAF, which
is subject to the existing NSPS. The existing 22.7-Mg (25-ton) EAF was
not operated during the emission tests. The fabric filters were tested
using a Reference Method 5 sampling train. The furnace was operating at
design capacity during the emission tests. During the tests, opacity
monitors on each of the fabric filters were in operation and in the
process of certification tests. Opacity readings were made of the
emissions from the shop roof monitor and the dust-handling equipment.
Detailed results of the tests are presented in Tables C-6 and C-7.
The particulate matter concentrations for fabric filter No. 4 were 1.60,
1.15, and 4.12 mg/dscm, averaging 2.29 mg/dscm (0.0007, 0.0005, and
0.0018 gr/dscf, averaging 0.0010 gr/dscf).3 : The particulate matter
concentrations for fabric filter No. 3 were 6.41, 6.64, and 4.81 mg/dscm,
averaging 5.95 mg/dscm (0.0028, 0.0029, and 0.0021 gr/dscf, averaging
0.0026 gr/dscf).4
The VE data from the fabric filters were obtained with a continuous
opacity monitor. The average opacity from the fabric filter continuous
opacity monitor at Plant B was 2.5 percent.3 The VE data for the shop
roof monitor will not be used to support a standard because the shop
walls at Plant B do not extend to .ground level. This arrangement of the
shop walls creates cross-draft problems, which reduces the capture
device efficiency and causes opacities of the emissions from the shop
roof monitor to be higher than those from shops in which the walls
extend to ground level.
C.I.1.3 Plant C. Plant C operates one 49.9-Mg (55-ton) EAF.5
Particulate matter emissions from the furnace are captured by a total
furnace enclosure (TFE), a local tapping hood, and an overhead canopy
hood. The TFE captures the melting and refining emissions and most of
the charging emissions. During charging, the TFE doors close, and an
air curtain, which aids in the capture of the charging emissions, blows
across the enclosure opening provided for the crane cables. The fugitive
emissions that escape the TFE during charging are captured by the overhead
canopy hood. The tapping hood is located below the EAF and captures the
emissions from the tapping operation. In addition, there is a tundish
lancing hood located beside the tapping area that is ducted to the same ;
C-5
-------
control device as the TFE and tapping hood. (The molten steel is poured
from the ladle to the continuous caster through a tundish. The tundish
is lanced with oxygen to remove any solidified metal.)
The emissions from the EAF are controlled by a negative-pressure
fabric filter. The fabric filter is rated at a 70.8-m3/s (150,000-acfm)
flow rate and has a gross air-to-cloth ratio of 4.5:1.12 The polyester
bags are cleaned by reverse air flow. The pressure drop across the bags
is maintained between 0.50 to 1.74 kPa (2 to 7 in. w.g.).
Grain loadings from the fabric filter outlet-were measured using
Reference Method 5 to determine compliance with the existing NSPS for
EAF's. The fabric filter operated normally and the EAF was operating at
full capacity during the emission test. Visible emission readings were
taken of the fabric filter stack and the building roof.
The building roof is closed; however, one end of the shop is open
to facilitate scrap transport. The fugitive emissions from all processes
within the building (furnace operation, tundish lancing, casting, slag
removal, etc.) mix and slowly drift out the open end of the shop.12
This created difficulties in reading VE's because of the time lag between
the release of emissions from a particular furnace operation and the
observation of visible emissions. For example, emissions from the
removal of slag by the bulldozer would mix with furnace emissions and
distort the vis'ible emission readings. Efforts were made to distinguish
between the emissions from different sources, and only opacities of the
emissions from the furnace operation were read.
Two tests were performed at Plant C. The first test was not accepted
by the State due to problems with the Reference Method 5 test. There-
fore, only the particulate data from second test will be presented. The
visible emission data obtained during both tests will be presented.
During the second test, visible emission observations were made of the
dust-handling system.
Three Reference Method 5 sampling runs were performed with the first
test run covering two full heats, and the other two test runs each covering
one full heat. The particulate matter concentrations from the three samples
were 3.05, 2.52, and 2.79 mg/dscm, averaging 2.79 mg/dscm (0.00133,
C-6
-------
0.00110, and 0.00122 gr/dscf, averaging 0.00122 gr/dscf).5 The sampling
results are presented in Table C-8.
For the first test, the opacity of the emissions from the fabric filter
outlet stack was always zero. A maximum 6-minute average opacity of
14.6 percent was observed from the end of the shop one time during a charge.
This high opacity reading may have been the result of other plant operations,
as it occurred when a front-end loader was removing slag near the furnace.
The remainder of the visible emission readings were below 5 percent opacity
for charging, melting, and tapping operations.
For the second test, no visible emissions were observed coming from the
fabric filter outlet. The highest 6-minute average opacity observed from the
melt shop was 5,2 percent, with a majority of the emissions being 0 percent
opacity. Table C-3 summarizes the VE data for the two tests.
C.I.1.4 Plant D. Plant D operates one ultra-high power (UHP) EAF rated
at a 108.8-Mg (120-ton) capacity.6 The meltdown and refining emissions from
the EAF are controlled by a direct-shell evacuation control (DEC) system that
is ducted to a positive-pressure fabric filter. The DEC system has an
independent fan that pushes the DEC exhaust to two main fans that serve the
fabric filter. Fugitive emissions (i.e., those generated during charging and
tapping and any emissions that are not captured by the DEC) are captured by
an overhead canopy hood and ducted to the fabric filter. A retractable
tundish lancing hood, which is only used periodically, is ducted to the same
fabric filter along with the exhausts from the canopy hood. The melt shop
roof at Plant D is closed.
The flow through the fabric filter is 200 m3/s (425,000 acfm).13 The
polyester bags in the fabric filter are cleaned by reverse air. The pressure
drop across the filter bags is 1.25 to 1.75 kPa (5 to 7 in. w.g.). The stack
located above the fabric filter has a fan to supplement the two main fans and
reduce the back pressure in the fabric filter compartments caused by the
stack.
The fabric filter outlet was tested for particulate matter using
Reference Method 5. The furnace was operating at design capacity during the
tests. The fabric filter was operating without the stack booster fan during
the test. The length of the EAF heat cycle varied during the tests from
1 hour and 30 minutes to 2 hours. The sampling was performed at four
C-7
-------
traverse points with 24 minutes of sampling time per traverse point, for
a total sampling time of 96 minutes. The testing was peformed so that a
sample representative of typical furnace operation was obtained; however,
the sampling did not always begin ;at the time of the furnace charge.
Detailed results of the test are presented in Table C-9. The
particulate matter concentrations from the three samples were 1.90,
3.82, and 1.99 mg/dscm, averaging 2.57 mg/dscm (0.00083, 0.00167, and
0.00087 gr/dscf, averaging 0.0011 gr/dscf).13 Visible emissions could
not be read by a certified reader during the test because of inclement
weather.
C.I.1.5 Plant E. Plant E operates one UHP EAF rated at 117.9 Mg
(130 tons), with an average heat time of 2.25 hours.7 The EAF emissions
are controlled by a positive-pressure fabric filter. The melting and
refining emissions from the furnace are captured by a DEC system. The
charging and tapping emissions are captured by an overhead canopy hood.
The design air flow rate to the fabric filter is 198.2 mVs (420,000 acfm).
The Dacron® polyester bags are exposed to a maximum temperature of 99°C
(210°F), and the gross air-to-cloth ratio is 2.82:I.14 Reverse air flow
is used to clean the bags.
The emission compliance test at Plant E was run on the 3.7 x 5.5 meter
(m) (12 x 18 foot [ft]) outlet on the side of the fabric filter. Reference
Method 5 and hi-vol sampling tests were run simultaneously on the fabric
filter outlet. The sampling times for the tests were approximately
4 hours and typically covered more than one full heat from tap to tap.
The furnace and the fabric filter were operating normally during the
test. The particulate matter emissions for the three Method 5 tests
were 3.44, 3.44, and 4.12 mg/dscm, averaging 3.67 mg/dscm (0.0015,
0.0015, and 0.0018 gr/ dscf, averaging 0.0016 gr/dscf).7 These results
are presented in Table C-10. (The particulate matter emissions for the
three hi-vol tests were 0.62, 0.73, and 0.62 mg/dscm, averaging 0.66 mg/dscm
[0.00027, 0.00032, and 0.00027 gr/dscf, averaging 0.00029 gr/dscf].)
Visible emissions from the shop roof monitor were not read during the
test. An opacity monitor was in operation during the test on the outlet
of the fabric filter. The average opacity of the exhaust gases during
C-8
-------
the test was reported as 3 percent; however, no data was presented to
substantiate this.
C.I.1.6 Plant F. Plant F operates two 36.3-Mg (40-ton) and three
45.5-Mg (50-ton) EAF's. The test information presented here is for the
three 45.4-Mg (50-ton) EAF's and the control system for these furnaces
only will be discussed. The primary emissions from the EAF's are captured
by side draft hoods and the fugitive emissions are captured by canopy
hoods built into the roof trusses. The shop roof at this facility is
closed. The side draft and canopy hoods are ducted together to a positive-
pressure fabric filter that is rated at 349.2 mVs (740,000 acfm). The
baghouse has 60 modules and a gross air-to-cloth ratio of 2.33:1.8,15
The bags are cleaned by a shaker mechanism and the clean gases are
vented out a 25.9-m (85-ft) tall metal stack with an inside diameter
of 5.5 m (18 ft).
The fabric filter stack was tested following Reference Method 5 .
procedures. Three Reference Method 5 test runs were performed that
covered heats from each of the three furnaces, which were operated on
staggered schedules. Sampling results are presented in Table C-ll. The
particulate matter concentrations from the three samples were 1.83,
2.52, and 1.83 mg/dscm, averaging 2.06 mg/dscm (0.0008, 0.0011, and
0.0008 gr/dscf, averaging 0.0009 gr/dscf).8 Eighteen minutes of visible
emission observations were made of the baghouse stack during the test
and all the opacity readings were 0 percent opacity. A continuous
opacity monitor was in use at Plant F during the test. However, the
monitor was not operating properly and no data were reported. No visible
emission observations were reported for the melt shop roof.
C.I.1.7 Plant G. Plant G operates one EAF that is rated at 108.9 Mg
(120 tons) per heat. The primary emissions from the EAF are captured by
a DEC system, while the fugitive emissions are captured by an overhead
canopy hood. The shop roof at Plant G is closed. The DEC and the
canopy hood are ducted together to a positive-pressure fabric filter.
The six-compartment fabric filter is rated at 188.8 m3/s (400,000 acfm)
and has a net air-to-cloth ratio of 2.60:1. The pressure drop across
the compartment varies from 0.30 to 0.60 kPa (1.2 to 2.4 in. w.g.).
C-9
-------
The fabric filter outlet, a 3.7 x 24.7 m (12 x 81 ft) monovent, was
tested with a modified version of Reference Method 5. The modifications
to the Method 5 procedure included: (1) traversing across the top of 2
of the 6 compartments per run (12 traverse points evenly spaced), (2) a
longer sampling time to obtain a minimum sample volume of 4.5 dscm (160 dscf)
as required by the NSPS regulation, and (3) a special procedure for setting
the isokinetic sampling rate. This procedure was necessary due to flow
disturbances in the baghouse outlet duct (i.e., a positive-pressure fabric
filter monovent since the sampling site did not meet Method 1 criteria).
The fabric filter operated normally during the test period. The particulate
matter emissions for the three modified Method 5 tests were 10.6, 5.9, and
1.14 mg/dscm, averaging 5.9 mg/dscm (0.0046, 0.0026, and 0.0005 gr/dscf,
averaging 0.0026 gr/dscf).9 These results are presented in Table C-12.
Visible emissions from the shop roof monitor were read during the test and
the results were no visible emissions for 46 six-minute averaging periods.
In addition, there was a continuous opacity monitor in operation with a
24.7-m (81-ft) path length across the monovent of the fabric filter. The
highest 6-minute average opacity for all test runs was 2.8 percent. Visible
emissions from the dust-handling equipment were also observed. However,
because of high opacity readings, the equipment failed to meet compliance
and will have to be improved. Table C-3 presents the VE data for Plant G.
C.I.1.8 Plant H. Plant H operates two 90.7-Mg (100-ton) EAF's and
one 136.1-Mg (150-ton) EAF. The 136.1-Mg (150-ton) EAF was not in operation
during the emission test. The furnaces average 4.5 hours tap-to-tap, with
30 minutes at the beginning of each heat spent on conditioning the furnace
refractories. The process emissions are captured by DEC systems while the
fugitive emissions are captured by canopy hoods above each furnace. The
shop roof monitor at Plant H is open. The exhausts from the DEC systems are
ducted to two positive-pressure fabric filters each rated at 66.1 ms/s
(140,000 acfm) at 135°C (275°F).16 The fabric filters have an air-to-cloth
ratio of 3:1 and vent the cleaned exhaust gases out a continuous roof
monitor that is 1.5 m (5 ft) wide and 14.6 m (48 ft) long. The emissions
captured by the canopy hoods are ducted to a single negative-pressure fabric
filter with an air flow rate of 162.8 mVs (345,000 acfm) that vents out
a stack. This fabric filter has an air-to-cloth ratio of 4:1.16
C-10
-------
The outlet of one DEC fabric filter that controls one EAF was
emission-tested with a Method 5 sampling train that was modified to use
a calibrated hot wire anemometer instead of a pi tot tube. Twenty-four
sampling ports were installed in the vertical sheeting of the throat of
the fabric filter roof monitor. Two sampling points were made at each
port for a total of 48 separate points. One 4-hour sampling run was
performed and the particulate matter emissions were 5.95 mg/dscm
(0.0026 gr/dscf).10 A summary of the results is presented in Table C-13.
The canopy hood fabric filter that serves all three furnaces was
also tested. The 136.1-Mg (150-ton) furnace was not in operation during
the test. The results of the Method 5 test were high due to reasons
that are assignable, and the results from this fabric filter will not be
used in the development of the standard.
Visible emissions readings were taken from the melt shop roof and
the canopy hood fabric filter. A total of 10 hours and 58 minutes of
observations were made of the melt shop roof. Eighty-six percent of the
emissions had zero percent opacity, however the charges and taps were
higher. The highest opacity based on a 6-minute average for a charge
was 12 percent and the highest for a tap was 33 percent.10 During the
test period, 2 hours of visible emission readings were made of the
canopy hood fabric filter and the result was 0 percent opacity. Table C-3
presents the summarized VE data for this plant.
C.I.1.9 Plant I. Plant I operates two EAF's each with a rated
capacity of 108.9 Mg (120 tons). The process emissions from both furnaces
are captured by a DEC system, while the fugitive emissions are captured
by an overhead canopy hood. The shop roof at Plant I is closed to
prevent fugitive emissions from escaping the building. The emissions
captured by the DEC systems on each furnace are ducted to a 12-compartment
fabric filter that operates at 34.9 m3/s (74,000 acfm). The fabric
filter is equipped-with silicone-treated glass bags that are cleaned by
reverse air. The inlet temperature to the fabric filter varies from
177° to 232°C (350° to 450°F).16 The emissions from the canopy hood are
ducted to a separate 14-compartment fabric filter. The total air flow
to this fabric filter is 330 ms/s (700,000 acfm) except during charging
and tapping operations when the control dampers are fully open and the
C-ll
-------
maximum flow is 533 mVs (1,130,000 acfm).16 This fabric filter is
equipped with Dacron® polyester bags that are cleaned by a shaking
mechanism.
Visible emission readings were taken at this plant from both fabric
filters and the shop roof. One hour of visible emissions were read from
the DEC baghouse that all exhibited 0 percent opacity. Visible emissions
from the canopy hood fabric filter were also observed for one hour.
This fabric filter exhibited puffs of approximately 5 seconds in duration
that appeared to coincide with compartment cleaning. These puffs indicate
that the fabric filter is not representative of a well-maintained and
operated fabric filter and that data will not be used to support a
standard. An hour of visible emission readings was also taken of the
melt shop roof. All the readings exhibited 0 percent opacity. Table C-3
presents the summarized VE data for this plant.
C.I.1.10 Plant J. Plant J has two UHP furnaces, and each produces
317.5 Mg (350 tons) of molten steel per heat.17 The two furnaces were
retrofitted into an existing open hearth shop, and their emissions are
controlled by a new fabric filter. The majority of the meltdown and
refining emissions are captured by a DEC system. The tapping emissions
are captured by a tapping hood situated above the tapping platform that
supports the ladle. The slagging emissions are captured by a close-fitting
slagging hood. The charging emissions and any fugitive emissions that
escape the DEC or tapping hood are captured by an overhead canopy hood.
The shop roof monitor is closed over the furnaces and open elsewhere.
The design flow rate for the fabric filter is 445.9 mVs (945,000 acfm),
and the gross air-to-cloth ratio is 3:1. The pressure drop across the
fabric filter bags is 0.75 to 1.87 kPa (3 to 7.5 in. w.g.).18 The
fabric filter is equipped with Nomex® bags, which are cleaned by a
mechanical shaker. The fabric filter exhausts through a continuous
ridge ventilator.
The emission compliance test was conducted with hi-vol samplers;
therefore, the particulate emission data are not presented.
Concurrent with the fabric filter sampling, visible emission data
were obtained from the baghouse and the shop roof monitor. The maximum
6-minute average visible emission reading during a tap that was attributable
C-12
-------
to the EAF's was 5.0 percent.19 The highest opacity reading during a
charge that was attributable to an EAF was 4.4 percent. The highest
shop roof opacity reading during EAF melting and refining was 4.2 percent.
Plant personnel have indicated that the closed roof monitor area above
the furnaces has been extended (thus eliminating the open area) in order
to reduce the noise problem and to contain the fugitive emissions to
obtain the desired shop roof opacity of 0 percent. A total of. 96 six-minute
averages that are between 0 and 5 percent opacity were recorded during
the melting phase of EAF operation. All the VE readings for the fabric
filter showed 0 percent opacity. The VE data is summarized in Table C-3.
C.I. 1.11 Plant K. Plant K operates two EAF's, each rated at a
117.9-Mg (130-ton) capacity.20 The heats are typically 4 to 5 hours in
duration. The furnaces are on a staggered schedule. The emissions
generated during meltdown and refining are captured by lateral (side)
draft hoods. Above each furnace are two adjacent canopy hoods, one
designed to capture charging emissions and the other designed to capture
.tapping emissions. Each hood has electrically controlled dampers to
direct the flow to capture the emissions from the different furnace
operations. There is a scavenger duct at this facility to capture any
emissions that escape the canopy hood. The shop roof at Plant K is closed.
The captured emissions are cleaned by a positive-pressure fabric filter
rated at 424.7 mVs (900,000 acfm). The polyester filter bags can
withstand a maximum temperature of 135°C (275°F) and the unit has an
air-to-cloth ratio of 2.93:I.21 The fabric filter exhausts through a
continuous ridge ventilator. The fabric filter bags are cleaned by
reverse air.
The emission test at Plant K was conducted with hi-vol samplers,
and, therefore, the particulate emission data .are not presented. Visible
emission data were obtained from the fabric filter outlet and melt shop
roof during the particulate matter emission sampling test. There were
no visible emissions observed from the melt shop roof (600 minutes of
observation) or the fabric filter outlet (300 minutes of observation)
during the entire test period. These data are summarized in Table C-3.
c-1-1-12 P1ant L. Plant L operates one EAF rated at a 90.7-Mg
(100-ton) capacity with an average heat time of 2.25 hours. The furnace
C-13
-------
process emissions are captured by a DEC system while the fugitive emissions
are captured by an overhead canopy hood. The DEC and canopy hood are
ducted to a positive-pressure fabric filter that is cleaned by reverse
air. The fabric filter has a design flow rate of 198.2 ms/s (420,000 acfm)
and a gross air-to-cloth ratio of 2.8:1. The pressure drop across the
polyester bags is 1.74 to 2.24 kPa (7 to 9 in. w.g.).22,23 The shop
roof at this plant is closed; however, there was a continuous opening
along the eaves of the roof and an open door on the shop wall that
permitted the fugitive emissions that bypassed the canopy hood to escape
the melt shop. For these two reasons, the VE data obtained at Plant L
will not be used in support of a standard.
C.I.1.13 Plant M. Plant M operates four EAF's, two of which
(Nos. 1 and 2) are older and have a melting capacity of 182 Mg (200 tons)
each. The two newer furnaces (Nos, 3 and 4) have a melting capacity of
204.1 Mg (225 tons) each.24 The process emissions from furnaces 1 and 2
are controlled by a wet scrubber, and the furnace fugitive emissions are
uncontrolled. The emissions from each of the two newer furnaces (Nos. 3
and 4) are captured by a DEC system and canopy hood. At Plant M the
shop roof is closed over the furnaces and the roof monitor is open
elsewhere. The DEC systems and canopy hoods are ducted in common to a
single positive-pressure fabric filter. The flow through the fabric
filter is 825.8 m3/s (1,750,000 acfm). The bags are cleaned by a mechanical
shaker, and the gross air-to-cloth ratio is 3.4:1.24,25 The pressure
drop across the bags is 0.75 to 1.25 kPa (3 to 5 in. w.g.).24,25
The plant was visited to discuss the potential heat stress problem
in warm and humid Southern locations. During the visit, a DEC malfunction
allowed fugitive VE's to escape out the melt shop roof. In addition,
two compartments in the baghouse were off-line for maintenance, and only
three of the four fans were operating. For these reasons, the VE data
obtained at Plant M will not be used to support a standard.
C.I.1.14 Plant N. Plant N operates two UHP EAF's. Furnace A has
a melting capacity of 104 Mg (115 tons) and Furnace B has a melting
capacity of 136 Mg (150 tons).26 The process emissions from melting and
refining are captured by separate DEC systems for each furnace. The
fugitive emissions are captured by two canopy hoods located above Furnace A
C-14
-------
and one large canopy hood above Furnace B. There are sheet metal partitions
suspended from the melt shop roof down to the crane level that create •---
secondary collection zones to capture fugitive emissions not captured by the
canopy hoods. These secondary collection zones are evacuated by scavenger
ducts. The process and fugitive emission capture systems are ducted
separately to a mixing chamber and then to the fabric filter control systems.
The emissions from the furnaces are controlled by a positive-pressure fabric
filter with a design flow rate of 212.4 m3/s (450,000 acfm), a net air-to-
cloth ratio of 3:1, and a pressure drop across each compartment of
2.0-2.5 kPa (8-10 in.) water and by a negative-pressure fabric filter with a
design flow rate of 354 mVs (7505000 acfm), a net air-to-cloth ratio of
5.5:1 and a pressure drop of 1.75-2.0 kPa (7-8 in.) water.26,27,28
The first visit to this plant, in March 1981, was to review potential
heat stress problems at plants in Southern locations. At the time of the
visit, the plant was undergoing changes to accommodate the installation of a
new furnace (Furnace B). A large opening in the side wall of the melt shop
created cross-drafts within the shop that prevented effective capture of the
charging and tapping emissions by the canopy hood. Thus, the VE data
obtained at Plant N during the first visit will not be used to support a
standard.27
The second visit to this plant, in August 1982, was to conduct a VE test
on the melt shop roof monitor. All plant modifications to accommodate
Furnace B, in process during the first visit, had been completed. This visit
coincided with a compliance test performed on the control system associated
with Furnace B (Furnace A was not operating due to reduced demand for Plant N
products). Visible emission data were obtained from the melt shop roof
monitor over Furnace B for 5.4 hours the first day and 8.4 hours the second
day. On the first day, the maximum 6-minute average VE readings during
charging, melting, and tapping were 3.3, 0, and 0 percent, respectively; and
on the second day, the maximum readings were 3.3, 3.1, and 1.0 percent,
respectively. Observations in the melt shop showed that these maximum
opacity levels were achieved even during furnace upset conditions when
melting emissions were not captured by the DEC system but were emitted into
the shop and captured by the canopy hood and scavenger ducts. Melt shop
observations showed that the uncontrolled emissions from the nitrogen
C-15
-------
stirring station combined with the emissions from the first charge of most
heats. Table C-3 summarizes the VE data recorded during this visit.
C.I. 1.15 Plant 0. Plant 0 operates two EAF's, each with a rated
capacity of 90.7 Mg (100 tons). (There is one additional 45.4-Mg [50-ton]
furnace in the shop, but it has not been used in over 7 years.) The process
emissions from each furnace are captured by a DEC system, while the fugitive
emissions are captured by a segmented canopy hood. The shop roof monitor at
Plant 0 is equipped with louvers that are designed to contain the emissions
within the shop for capture by the canopy hood. When the louvers are opened,
they vent heat and any emissions hot captured by the canopy hood to the
atmosphere.
Emissions captured by the DEC systems and canopy hoods are ducted in
common to a single positive-pressure fabric filter with a design flow rate of
247.7 mVs (525,000 acfm). The fabric filter is cleaned by reverse air, and
the net air-to-cloth ratio is 1.95:!.29 The pressure drop across the bags
is maintained between 0.75 and 1.75 kPa (3 and 7 in. w.g.).29
Plant 0 was visited to discuss the potential heat stress problems in
warm climates and to observe the operation of a louvered roof monitor. The
plant and the control system are old and underdesigned for capture of the
fugitive emissions. Furthermore, the fabric filter was in need of mainten-
ance during the visit. Several broken bags in some of the baghouse compart-
ments allowed uncontrolled emissions to be vented to the atmosphere. For
these reasons, the VE data obtained at the plant will not be used to
support a standard.
C.2 EMISSION TEST DATA FOR FABRIC FILTERS ON ELECTRIC ARC FURNACES AND
ARGON-OXYGEN DECARBURIZATION VESSELS IN THE SPECIALTY STEEL INDUSTRY
C.2.1. Particulate Matter and Visible Emissions
This section provides a summary of emission source test data on fabric
filters installed at four specialty steel shops. The data includes fabric
filters controlling EAF's alone, AOD vessels alone, and EAF and AOD vessels
ducted together to a fabric filter. The particulate matter emissions at
Plants P, Q, and R were measured by a Reference Method 5 sampling train and
a summary of the data is presented in Figure C-2. The particulate matter
C-16
-------
qr/dscf mg/dscm
8
0.003 -
0.002
-------
emission data for five additional plants were obtained using the hi-vol test
method; however, these data are not presented here.
Opacity readings of the shop roof monitor exhaust gases were obtained
at Plants P, Q, and S using EPA Reference Method 9. Opacity readings
from the fabric filter monovent exhaust gas stream were taken at Plants P,
Q, and S, and are summarized in Table C-14. Table C-15 presents a
summary of the shop roof opacity data from Plants Q and S. Opacity data
from Plant P will not be used to support standards because it is an
older shop with open seams in the sheet metal section of the shop walls
and does not represent best demonstrated technology.
C.2.1.1 Plant P. Plant P is a specialty steel mill that has two
EAF's, each rated at 29.0 Mg (32 tons), and two AOD vessels, each rated
at 29.0 Mg (32 tons).30 The two EAF's feed the molten steel to either
of the two AOD vessels for refining. The capture system for the EAF's
consists of two canopy hoods above each furnace, one to capture charging
and melting emissions and one to capture tapping emissions. Each canopy
hood has automatic dampers that can be closed to direct more suction to
another hood. The dampers on the charging canopy hood are always open
whereas the dampers on the tapping side only open during the tap. Two
canopy hoods are also located above each AOD. vessel. These hoods are
situated so that each canopy captures half of the refining emissions.
Both AOD vessels have venturi-type diverter stacks to direct the emissions
to the canopy hoods and to reduce the emission dispersion in the shop
due to cross-drafts. The damper on each AOD vessel canopy hood is
always open. The roof is closed over the entire melt shop at Plant P.
All the canopy hoods are ducted to the same positive-pressure
fabric filter that has a rated flow rate of 282 m3/s (600,000 acfm). The
fabric filter is equipped with polyester bags that are cleaned automatically
by a shaker mechanism. The gross air-to-cloth ratio is 2.65:1. The
pressure drop across the bags is 1 to 1.5 kPa (4 to 6 in. w.g.).34
The fabric filter was tested for particulate matter concentrations
at the inlet (two ducts) and the outlet. The tests were run to cover two
full heats on one of the EAF's, while the other EAF and the two AOD vessels
operated normally. Generally, the EAF's operate on a staggered schedule,
C-18
-------
and the average heat time (tap-to-tap) Is 3.7 hours. The emission control
system appeared to be operating normally during the test period.
Detailed results of the tests are presented in Tables C-16, C-17,
and C-18. The particulate matter emission concentration for the north
fabric filter inlet averaged 308 mg/dscm (0.134 gr/dscf) while the
particulate matter concentration for south fabric filter inlet averaged
183 mg/dscm (0.080 gr/dscf). The outlet particulate matter concentrations
determined for three samples were 3.27, 2.89, and 4.24 mg/ dscm, averaging
3.47 mg/dscm (0.00143, 0.00126, and 0.00185 gr/dscf, averaging
0.00151 gr/dscf).30
Concurrent with the fabric filter sampling, visible emission readings
were obtained from the fabric filter and the roof monitor above the
shop. This is an older shop with numerous holes in the closed roof and
in the walls of the melt shop through which the fugitive emissions
escaped. Visible emission readings on the fabric filter were taken for
a total of 438 minutes with the maximum 6-minute average of 0 percent.
In addition to the VE readings of the fabric filter and melt shop roof
using Reference Method 9, the fabric filter dust-handling system was
observed for fugitive emissions by Reference Method 22.
C.2.1.2 Plant Q. Plant Q operates five EAF's and two AOD vessels.
Each of the EAF's has an average heat size of 13.6 Mg (15 tons). One of
the AOD vessels has a 13.6~-Mg (15-ton) capacity and the other has an
18.1-Mg (20-ton) capacity.35 One emission test was conducted on the
positive-pressure fabric filter associated with the 18.1-Mg (20-ton) AOD
vessel.32 The AOD vessel emissions are directed to a canopy hood by a
movable diverter stack. The canopy hood is located above the crane in
the roof trusses. The emissions that escape the canopy hood are withdrawn
by a scavenger duct that is located at the high point of the closed
roof. (In addition, a small amount of the fabric filter flow capacity
is used to control cutting torch emissions at the continuous caster.)
The total flow rate capacity of this fabric filter is 141.6 m3/s
(300,000 acfm).35 The cleaned gases from the fabric filter are exhausted
to the atmosphere via five stub stacks. Each stack exhausts cleaned
gases from two compartments. The fabric filter is equipped with Dacron®
bags and has a gross air-to-cloth ratio of 2.68:I.35 The dirty bags are
C-19
-------
cleaned by reverse air. The pressure drop across the bags is 0.6 to
1.5 kPa (2.5 to 6 in. w.g.).
Simultaneous inlet and outlet tests for particulate matter were run
on the fabric filter. The AOD vessel was operating at design capacity
during the test, and the fabric filter appeared to be operating normally.
The actual sampling time for each of the three test runs was 450 minutes.
The average heat time was 1.6 hours, and these test runs included five ~
to six heats per sampling period. The inlet particulate matter concen-
trations determined from the three tests were 150, 141, and 211 mg/dscm,
averaging 167 mg/dscm (0.0655, 0.0617, and 0.0921 gr/dscfm, averaging
0.0731 gr/dscfm). Results of the inlet tests are presented in Table C-19.
The outlet particulate matter concentrations determined from^the three
tests were 0.836, 0.497, and 0.650 mg/dscm, with an average of 0.661 mg/dscm
(0.000365, 0.000217, and 0.000284 gr/dscf, with an average of ,
0.000289 gr/dscf).31 The results of the outlet tests are presented in
Table C-20.
Concurrent with the particulate matter sampling of the 18.1-Mg
(20-ton) AOD vessel, visible emission readings were obtained from the
shop roof and fabric filter outlets. No visible emissions were seen
from either the shop roof or fabric filter during the test period
(408 minutes of observation). In addition to these readings, fugitive
emissions observations of the fabric filter dust-handling system were
performed in accordance with Reference Methods 9 and 22.
Another Method 5 test was performed at this facility with only one
sample run. A detailed description of this test is presented in Table C-21.
The results of this test showed a particulate matter concentration of
6.87 mg/dscm (0.003 gr/dscf).35 i
C.2.1.3 Plant R. Plant R has two 38.1-Mg (42-ton) EAF's that
operate on a staggered schedule to supply molten steel to one 45.4-Mg
(50-ton) AOD vessel.33 The emissions from both the EAF's and the AOD
vessel are captured by overhead canopy hoods. Scavenger duct openings
are utilized to capture fugitive emissions that escape the canopy hood.
The emissions are cleaned by a positive-pressure fabric filter with a
flow rate of 316 m3/s (675,000 acfm).33 The gross air-to-cloth ratio is
3.26:1. The 12-compartment fabric filter is equipped with polyester
C-20
-------
bags that are cleaned by reverse air. The fabric filter exhaust is
vented out of one of two continuous ridge ventilators. The pressure
drop across the bags is 1.5 to 2.0 kPa (6 to 8 in. w.g.).
Both the inlet and outlet of the fabric filter were tested for
particulate matter concentration. The inlet was sampled with a standard
Reference Method 5 probe and sampling train, and the average of the two
inlet tests was 178.7 mg/dscm (0.07804 gr/dscf).33 A detailed summary
of the inlet test is presented in Table C-22. The outlets of 3 of the
12 compartments were sampled for 2-1/2 hours by traversing the top of
each compartment with a nozzle filter assembly of the Reference Method 5
sampling train. The outlet particulate matter concentration varied from
0.413 to 3.94 mg/dscm (0.00018 to 0.00172 gr/dscf).33 The detailed
summary of this test is presented in Table C-23.
C.2.1.4 Plant S. Plant S operates one 45.4-Mg (50-ton) EAF and
one 45.4-Mg (50-ton) ADD vessel.37 The emissions from the EAF arid AOD
vessel are controlled by two fabric filters. The EAF emissions are
captured by an overhead canopy hood and ducted to a negative-pressure
fabric filter. Fugitive emissions from the AOD vessel are eventually
controlled by this fabric filter since the shop roof is closed and the
AOD emissions drift to the EAF canopy hood. The negative-pressure
fabric filter has a design flow rate of 108.5 ms/s (230,000 acfm) and
utilizes Dacron® bags with an air-to-cloth ratio of 3.62:1. The bags
are cleaned by a shaker mechanism. This fabric filter exhausts the
cleaned gases out a large steel stack.37 The pressure drop across the
fabric filter bags is 1.25 to 2.50 kPa (5 to 10 in. w.g.).
The primary exhaust from the AOD vessel is captured by a bell-shaped
hood immediately above the vessel mouth. The primary emissions are
ducted along with the teeming aisle fumes and some of the EAF fumes to a
positive-pressure fabric filter. This fabric filter has a design flow
rate of 108.5 ms/s (230,000 acfm) and utilizes Dacron® bags with an
air-to-cloth ratio of 3:62:I.37 The bags are cleaned by reverse air.
This fabric filter exhausts the cleaned gases through a continuous ridge
ventilator.
Visible emission readings were taken at this plant at both fabric
filters and at the shop roof above both the EAF and AOD vessel.38 A
C-21
-------
total of 61 six-minute sets of visible emission readings were recorded
for the melt shop roof, 56 sets for the EAF fabric filter stack, and
53 sets for the ADD fabric filter roof monitor. All sets averaged
0 percent opacity except for two sets taken at the melt shop roof during
the EAF melting and ADD refining phases. These two 6-minute average
were 1 and 5 percent, respectively.
C.3 VISIBLE EMISSION DATA FOR DUST-HANDLING SYSTEMS FOR BOTH CARBON
AND SPECIALTY STEEL PLANTS
Visible emission data were obtained from the dust-handling systems
at four steel mills. As mentioned in Chapter 4, the dust-hand!ing
equipment includes the baghouse dust hoppers, the pneumatic system that
conveys the dust to the storage bin, the storage bin, and the dust
transfering system that conveys the dust from the storage bin to a
truck. Table C-24 summarizes the results from Plants B and C, carbon
steel plants, and Plants P and Q, specialty steel plants. The highest
recorded opacity for 734 minutes of readings with both Reference Methods 9
and 22 was a 6-minute average of 7.3 percent. This high opacity observation
was made during a storage-bin-to-truck transfer operation.
C.4 ENGLISH/METRIC CONVERSIONS
The information in the test reports was furnished in English units.
The following shows the conversion factors used to report the data in
metric units. The factors were taken from "Standard for Metric Practice,"
American Society for Testing and Materials, September 1977.
English units Multiply by Metric unit
inches, Hg
inches, H20
cubic feet
English units
gr/dscf
foot
grai ns
3,376.85
248.84
0.0283
Multiply by
2,290
0.3048
0:0648
pascals
pascals
cubic meters
Metric unit
mg/dscm
meters
grams
C-22
-------
a
LU
C£
o
LU
cn
LU
co
LU
I—
a.
>-
Q
LU
oo
LU
I—
O
>-
oo
LU
_J
CO
in
cu "O
4-5 CU
4- S~
o o
4-
•0) S-
Q. OJ
in
c:
o
4-5
CU
N
tn
O)
u
ro
C
Lu C7>
s:
^>s.
CU i—
CJ G)
ro in
c m
S- 0)
3 >
LU
in
4- (U
O CJ
ro
. c
O i~
4-
cu
ro
cn
4-5
u
-a
Q
a.
4J
ro
Q-
cn
M
LO LO
•a -a
0 O
4-5 4-J
a> cu
LO LO
r>- CM
*» *»
0 0
LO •?)•
oo r>-
CM
- CM
^1"
LO OO
VD
oo
LU LU
4-5
10
c
o
J2
CO
O
LU
cn cn
A *N
LO LO
-a -a
O 0
4J 4-5
cu cu
o o
LO CM
r—
«d- cn
LO CO
<*• o
r~
LU LU
o
1-3 ii
cn cn
Method
Method
O LO
a CM
r— CM
f-. r—
CD LU
CM CM
O CU
cu cu
4-5 4-9
in in
c. c.
0 0
ro ro
0 0
Z O
CM CM
CM CM
cn cn"
f\ **
to un
Method
Method
CM CM
OO OO
O 0
cn cn
CM CM
LU o
a>
4-5
10
4-5
•r-
u
cu
a.
oo
Q.
cn
•s
LO LO
Method
Method
O CM 0
CM sfr LO
r— r— <*•
00 CO LO
r— OO ^~
a LUQ
o oo
cy ce:
cn cn
0 0
cu cu
2: 2:
o o
LO LO
«5t- ^J-
LO LO
LU Q
-------
TABLE C-2. SUMMARY OF VISIBLE EMISSION DATA FROM
FABRIC FILTERS ON EAF'S AT CARBON STEEL SHOPS5,7-10,16,18,20
Maximum
6-minute
Plant
Ba
Ca
Fa
Gd
Hd
Id
Jd
Kc
Length of
observation,
minutes
960
596C
18
936
120
60
577
300
average
opacity,
percent
2
0
0
2
0
0
0
0
.5b
.8b
e
^Single stack? exhaust.
Not 6-minute average; average
from continuous monitor during
Reference Method 5 test.
^Combination of data from two tests.
Monovent exhaust.
eEPA Method 22.
C-24
-------
TABLE C-3. SUMMARY OF OPACITY DATA FROM SHOP ROOF MONITORS
AT CARBON STEEL SHOPS5,9,10,16,19,20,26
Plant Furnace
Ca EAF
Ca EAF
Gb EAF
Hc EAF
Furnace
process
Charge
Melt
Tap
Heat cycle
Heat cycle
Charge
Melt
Tap
No. of
6-minute
averages
5
33
6
88
46
10
95
5
Maximum
6-minute
average
opacity,
percent
1.3
2.7
0.4
5.0
0
12
0
33
Average
6-minute
average
opacity,
percent
0.5
0.1
0.1
0.25
0
5.0
0
23.0
EAF
Heat cycle
10
Ju EAF
Ke EAF
Nf EAF
Charge
Melt
Tap
Charge
Melt
Tap
Charge
Melt
Tap
15
106
10
5
90
5
30
94
14
4.4
4.2
5.0
0
0
0
3.3
3.1
1.0
1.0
0.2
0.6
'0
0
0
0.7
. 0.1
0.1
.Utilizes total furnace enclosure, canopy hood, closed roof.
Utilizes canopy hood, DEC, closed roof.
.Utilizes canopy hood, DEC, open roof.
Utilizes canopy hood, DEC, local tapping hood, closed roof over
furnaces/open elsewhere.
Utilizes canopy hood, side draft hood, tapping hood, closed
froof.
Utilizes canopy hood, DEC, scavenger ducts with cross-draft
partitions, closed roof over furnace/open elsewhere.
C-25
-------
TABLE C-4. SUMMARY OF PARTICULATE MATTER RESULTS—PLANT A1
(Fabric filter inlet)
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate" per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Particulate matter emi
Mass collected
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
|Mg
tons
ms/mi n
acfm
dscm/min
dscfm
ds cm/mi n/Mg
dscfm/ ton
°C
°F
% volume
% volume
% volume
ssions
mg
gr
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
10/18/72
240
—
113.4
125
13,349
471,699
12,204
431 ,236
--
37
98.6
0.7
<0.1
20.6
437.9
6.76
80.84
0.0353
64.9
143
—
2
10/19/72
240
"• —
113.4
125
14,066
497,016
12,875
454,932
...
38
100.4
0.6
<0.1
20.2
666.1
10.28
118.2
0.0516
99.8
220,
—
3
10/20/72
240
«. M
113.4
125
14,283
504,697
13,309
470,300
__
34.9
94.8
0.6
<0.1
20.7
732.5
11.30
129.2
0,0564
110.7
244
—
Average
240
—
113.4
125
13,899
491,137
12,796
452,156
-""•
36.6
97.9
0.6
<0.1
20.5
612
9.45
109.4
0.0478
91.8
202
—
Reference Method 5 data.
C-26
-------
TABLE C-5. SUMMARY OF PARTICIPATE MATTER RESULTS—PLANT A1
(Fabric filter outlet)3
Run number/date
Units
1 2 3
10/18/72 10/19/72 10/20/72 Average
Test data
Sampling time
240
240
240
240
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
_ b
Temperature
Water vapor
C02 dry
02 dry
Mg/h
tons/h
Mg
tons
m3/mi n
acfm
ds cm/mi n
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
—
— —
113.4
125
6,480
228,978
5,924
209,343
—
—
38.9
102
0.6
<0.1
20.6
__
— —
113.4
125
6,353
224,479
5,842
206,431
—
--
38.3
101
0.5
<0.1
20.2
--
— —
113.4
125
6,526
230,592
6,008
212,283
—
--
37.5
99.5
0.5
<0.1
20.7
—
— —
113.4
125
6,453
228,016
5,925
209,352
—
--
38.2
100.8
0,5
<0.1
20.5
Parti cul ate matter emissions
Mass collected
Dust concentration
Emission rate
Emission factor
mg
gr
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
74.6
1.15
4.81
0.0021
5.9
12.9
—
35.3
0.54
2.22
0.00097
3.4
7.41
—
36.3
0.56
2.29
0.001
4.8
10.5
__
48.7
0.75
,3.11
0.0014
4.7
10.3
—
Reference Test Method 5 data. Baghouse has six 9-ft-diameter stacks.
Three stacks sampled simultaneously. Emission rate based on inlet gas
flow.
Total through three of six stacks.
C-27
-------
TABLE C-6. SUMMARY OF PARTICULATE MATTER RESULTS—PLANT B3
(Fabric filter outlet~-Canopy hood)
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
-
Water vapor
C02 dry
02 dry
Units "
min
Mg/h
tons/h
Mg
tons
mVmi n
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
Run
1
01/22/80
240
13.8
15.2
36.3
40
2,580
91,183
2,377
83,977
65.5
2,099
25
77
1.2
0
20.9
number/date
2
01/22/80
240
22.6
24.9
36.3
40
2,496
88,190
2,328
82,270
64.1
2,057
22.2
• 72
0.8
0
20.9
3
01/23/80
240
15.5
17.1
36.3
40
2,447
86,454
2,279
80,516
62.8
2,013
22.8
73
0.8
0
20.9
Average
240
17.3
19.1
36.3
40
2,508
88,942
2,328
82,254
64.1
2,056
23.3
74
0.9
. 0
20.9
Parti cul ate matter emissions
Concentration
Emission rate
Emission factor
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1.37
0.0006
1.60
0.0007
0.22
0.49
0.016
0.032
1.'15
0.0005
1.15
0.0005
0.17
0.38
0.008
0.015
3.89
0.0017
4.12
0.0018
0.55
1.21
0.035
0.071
2.14
0.0009
2.29
0.001
0.31
0.69
0.02
0.039
Reference Method 5 data.
C-28
-------
TABLE C-7. SUMMARY OF PARTICULATE MATTER RESULTS—PLANT B4
(Fabric filter outlet—Side draft hood)3
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Units
min
Mg/h
tons/h
Mg
tons
m3/min
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
1
05/05/80
240
— —
36.3
40
3,563
125,900
2,972
105,032
81.9
2,625.8 •
66
150.9
0.008
0
2
05/06/80
240
— —
36.3
40
3,615
127,741
3,048
107,712
84.0
2,693
63
138.8
0.013
0
3
05/06/80
240
—
36.3
40
3,722
131,527:
3,090
109,200
85.1
2,730
67.6
149
0.012
0
Average
240
— —
36.3
40
3,633
128,389
3,037
107,315
83.7
2,683
65.5
146.2
0.011
0
Particulate matter emissions
Concentration
Emission rate
Emission factor
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
5.5
0.0024
6.41
0.0028
1.16
2.56
::
5.27
0.0023
6.64
0.0029
1.18
2.61
—
3.89
0.0017
4.81
0.0021
0.88
1.93
::
4.89
0.0021
5.95
0.0026
1.07
2.37
::
Reference Method 5 data.
C-29
-------
TABLE C-8. SUMMARY OF PARTICULATE MATTEF
(Fabric filter outlet)'
RESULTS—PLANT C5
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
CO dry
02 dry
C02 dry
Units
mi n
Mg/h
tons/h
Mg
tons
m3/mi n
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ton
°c
°F
% volume
% volume
% volume
% volume
1
333
18.85
20.78
49.9
55
3,609
127,527
3,279
115,867
65.7
2,107
63.1
145.5
0.26
2
179
16.46
18.14
49.9
55
3,674
129,830
3,341
118,062
70.0
2,147
62.6
144.6
0.36
3
182
17.48
19.27
49.9
55
3,580
126,487
3,316
117,179
66.5
2,131
56.3
133.4
0.03
Average
231.3
17.60
19.40
49.9
55
3,621
127,948
3,312
117,036
66.4
2,128
60.7
141.2
0.22
Parti cul ate matter emissions
Concentration
Emission rate
Emission factor
mg/m3
gr/acf
m/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
2.98
0.00130
3.05
0.00133
0.60
1.324
0.032
0.064
2.47
0.00108
2.52
0.00110
0.51
1.121
0.031
0.062
2.70
0.00118
2.79
0.00122
0.56
1.227
0.032
0.064
2.72
0.00119
2.79
0.00122
0.56
1.224
0.032
0.063
Reference Method 5 data.
C-30
-------
TABLE C-9. SUMMARY OF PARTICIPATE MATTER RESULTS—PLANT D6
(Fabric filter outlet)
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
m3/min
acfm
dscm/mi n
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
ssions
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg.
Ib/ton
Run
1
12/18/80
96
63.9
70.4
108.8
120
9,520
336,400
8,765
309,700
80.6
2,581
35.3
95.5
0.80
0.1
21.0
1.75
0.00076
1.9
0.00083
1.0
2.2
0.016
0.031
number/date
2
12/19/80
96
58.4
64.4
108.8
120
9,848
348,000
9,265
327,400
85.2
2,728
34.6
94.3
0.71
<0.1
21.0
3.60
0.00157
3.82
0.00167
2.1
4.7
0.036
0.073
3
12/19/80
96
58.8
64.8
108.8
120
10,089
356,500
9,667
341,600
88.9
2,847
29.4
85
0.70
<0.1
21.0
1 . 91
0.00083
1.99
0.00087
1.1
2.5
0.019
0.039
Average
96
60.4
66.5
108.8
120
9,819
347,000
9,232
326,200
84.9
2,719
33.1
91,6
0.74
<0.1
21.0
2.42
0.00105
2.57
, 0.0011
1.4
3.13
0.024
0.048
Reference Method 5 data.
C-31
-------
TABLE C-10.
SUMMARY OF PARTICULATE MATTER RESULTS3—PLANT E7
(Fabric filter outlet)
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
mVmi n
acfm
ds cm/mi n
dscfm
dscm/min/Mg
dscfm/ton
,°c
°F
% volume
% volume
% volume
ssions
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
06/12/80
240
35.2
38.8
117.9
130
15,448
545,875
13,267
468,785
112.5
3,606
68.3
155
1.5
0.1
19.3
2.98
0.0013
3.44
0.0015
2.7
5.9
0.08
0.15
2
06/13/80
240
62.8
69.2
117.9
130
15,390
543,800
13,186
475,936
111.8
3,661
67.8
154
T.8
0.1
17.1
2.98
0.0013
3.44
0.0015
2.7
5.9
0.04
0.09
3
06/13/80
240
37.2
41.0
117.9
130
15,360
542,765
13,258
468,494
112.5
3,604
66.1
151
1.5
0.1
18.3
3.66
0.0016
4.12
0.0018
3.3
7.2
0.09
0.18
Average
• 240
45.1
49.7
117.9
130
15,400
544,150
13,237
471,070
112.3
3,624
67.4
153.3
1.6
0.1
18,2
3.21
0.0014
3.67
0.0016
2.9
6.3
0.07
0.14
Reference Method 5 data.
C-32
-------
TABLE C-ll.
SUMMARY OF PARTICULATE MATTER RESULTS3—PLANT F8
(Fabric filter outlet)
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
Q£ dry
Particulate emissions
Concentration
Emission rate
Emission factor9
Units
min
Mg/h
tons/h
Mg
tons
mVmi n
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
11 /1 7/81
288
67.7
74.6
136.1
150
19,901
702,878
17,638
622,750
129.6
4,152
47.2
117
1.6
--
—
1.62.
0.0007
1.83
0.0008
1.94
4.27
0.03
0.06
2
11 /1 7/81
288
67.7
74.6
136.1
150
20,256
715,400
17,957
635,750
131. 9
4,238
46.7
116
1.6
'•--
--
2.23
0.0010
2.52
o.oon
2.71
5.98
0.04
0.08
3
11 /1 8/81
288*
63.3
69.8 •
136.1
150
19,651
694,022
17,083
603,200
125.5
4,021
51 .1
124
1.8
—
1.59
0.0007
1.83
0.0008
1.88
4.14
0.03
0.06
Average
288
66.2
73.0
136.1
150
19,936
704,100
17,559
620,567
129.0
4,137
48.3
119
1.7
1.81
0.0008
2.06
0.0009
2.18
4.80
0.033
0.067
Reference Method 5 data.
C-33
-------
TABLE C-12.
SUMMARY OF PARTICULATE MATTER RESULTS—PLANT G9
(Fabric filter outlet)
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate emissions
Concentration
Emission rate
Emission factor3
Units
1
1
i
min
Mg/h
tons/h
: Mg
tons
mVmi n
acfm
dscm/mi n
dscfm
dscm/mi n/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
;mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
12/15/81
390
27.7
30.5
108.9
120
11,404
402,965
10,953
387,030
100.6
3,225
32.2
90
0.4
10.1
0.0045
10.5
0.0046
6.98
15.39
0.25
0.50
2
12/17/81
276
38.7
42.7
108.9
120
9,347
330,280
8,832
312,086
81.1
2,601
34.4
94
0.5
5.7
0.0024
5.90
0.0026
3.08
6.78
0.08
0.16
3
12/17/81
270
41.0
45.2
108.9
120
9,567
338,370
8,913
314,943
81.8
2,625
40.6
105
0
1.0
0.0004
1.1
0.0005
0.56
1.24
0.014
0.03
Average
312
35.8
39.5
108.9
120
10,109
357,205
9,566
338,020
87.8
2,817
35.7
96.3
0.3
5.6
0.0024
5.9
0.0026
3.54
7.80
0.115
0.23
Reference Method 5 data.
C-34
-------
TABLE C-13. SUMMARY OF PARTICULATE MATTER RESULTS—PLANT H10
(DEC fabric filter outlet)3
Test data
Sampling time
Furnace production
Nomi nal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit of
furnace capacity
Temperature
Water vapor
C02 dry
02 dry
Particulate emissions
Concentration
Emission rate
Emission factor3
. Units
min
Mg/h
tons/h
Mg
tons
mVmi n
acfm
ds cm/mi n
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
Run
number/
date
1
08/28/73
240
—
—
99.8
no
4,379
154,650
3,675
129,800
36.8
1,180
41.7
107
3.16
—
--
5.04
0.0022
5.95
0.0026
2.60
2.87
—
"
Modified Reference Method 5.
C-35
-------
TABLE C-14. SUMMARY OF VISIBLE EMISSION DATA
FROM FABRIC FILTERS AT SPECIALTY STEEL SHOPS30,31,38
Plant
P
Q
S
Length of
observations,
minutes
438a
408b
336?
318b
Maximum
6-minute
average
opacity,
percent
0
0
0
0
j*0ne fabric filter for EAF and ADD vessels.
AOD vessel fabric filter.
CEAF fabric filter.
C-36
-------
TABLE C-15. SUMMARY OF VISIBLE EMISSION DATA FROM
SHOP ROOF MONITORS ON SPECIALTY STEEL SHOPS31,38
Plant
Qa
Sb
Furnace/vessel
AOD
EAF and AOD
Furnace process
Heat cycle
Heat cycle
No. of
6-minute
averages
69
55
Maximum
6-minute
average
opacity,
percent
0
5
bUtilizes diverter stack, canopy hood, scavenger duct, and closed roof.
Utilizes close-fitting hood, canopy hood, and closed roof.
C-37
-------
TABLE C-16.
SUMMARY OF PARTICULAR MATTER RESULTS—PLANT P30
(South fabric filter inlet)
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
*
Units
min
Mg/h
tons/h
! Mg
tons
ms/mi n
acfm
ds cm/mi n
dscfm
dscm/min/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
ssions
mg/rn3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
04/07/81
360
—
__
--
""**
8,442
298,100
7,794
275,200
--
--
46.7
116
0.65
_-
-.—
183
0.080
198
0.0867
93
205
—
2
04/08/81
360
--
•"•"
—
•• ••
8,466
298,900
7,758
273,900
—
--
47.8
118
0.69
— —
*._
220
0.096
240
0.1047
111
246
--
3
04/09/81
360
—
*• — >
—
*• ""
8,634
304,800
7,824
276,300
—
— —
45.6
114
1.29
— —
** *•
101
0.044
m
0.0485
52
115
—
Average
360
--
« MM
__
*** "™
,8,514
300,600
7,792
275,100
—
.._
46.7
116
0.88
"• ""
*™ "™
168
0.073
183
0.080
86
188
—
Reference Method 5 data.
C-38
-------
TABLE C-17.
SUMMARY OF PARTICULATE MATTER RESULTS—PLANT P30
(North fabric filter inlet)
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
mVmin
acfm
dscm/min
dscfm
dscm/mi n/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
ssions
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
04/07/81
360
__
--
__
—
8,376
295,800
7,800
275,500
—
—
43.3
110
0.81
--
--
202
0.088
217
0.0948
102
224
—
..
2
04/08/81
360
--
—
__
—
7,980
281,900
7,308
258,200
-- '
--
47.2
117
0.89
--
—
166
0.0727
182
0.0794
80
176
--
""" **
3
04/09/81
360
—
—
__
--
8,304
293,300
7,632
269,600
— .
—
41.7
107
1.25
--
--
482
0.2103
523
0.2288
240 .
529
...
"* •"
Average
360
__
--
— w
—
8,220
290,300
7,580
267,800
__
__
44
m
0.98
—
283
0.124
307
0.1343
141
310
__
"""
Reference Method 5 data.
C-39
-------
TABLE O18.
SUMMARY OF PARTICULATE MATTEE
(Fabric filter outlet)'
RESULTS—PLANT P30
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
mVrni n
acfm
ds cm/mi n
dscfm
dscm/min/Mg
dscfm/ton
°C
I °F
[
% yolume
% volume
% volume
ssions
img/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
1 kg/Mg
Ib/ton
1
04/07/81
330
28.9
31.9
116
128
16,542
584,100
15,594
550,700
134
4,302
42
108
0.41
— —
— —
3.09
0.00135
3.27
0.00143
3.06
6.75
0.11
0.21
2
04/08/81
330
30.2
33.3
116
128
16,380
578,600
15,066
532,100
130
4,157
48
119
0.68
— —
"" "™
2.66
0.00116
2.88
0.00126
2.61
5.75
0.09
0.17
3
04/09/81
330
28.2
31.1
116
128
16,920
597,600
15,456
545,900
133
4,265
46
114
1.17
"" —
w •_
3.87
0.00169
4.24
0.00185
3.94
'8.68
0.14
0.28
Average
330
29.1
32.1
116
128
16,614
586,800
15,372
542,900
132
4,241
45.3
114
0.75
*— »
—
3.21
0.0014
3.46
0.00151
3.20
7.06
0.11
0.22
{"Reference Method 5 data.
Basis: Three of four furnaces operating at the same time.
C-40
-------
TABLE C-19.
SUMMARY OF PARTICULATE MATTER RESULTS—PLANT Q31
(Fabric filter inlet)3
Run number/date
Test data
Sampling time
Furnace production
.Nominal vessel
capacity
Shop effluent
Flow rate
Flow rate. per unit
of furnace
capacity
Temperature
•
Water vapor
C02 dry
02 dry
Part icu late matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
; tons/h
Mg
tons
mVniin
acfm
ds Cm/mi n
dscfm
dscm/min/Mg
dscfm/ton
°C
op
% volume
% volume
% volume
ssions
mg/m3
gr/acf
mg/dscm
gr/dscf
. kg/h
Ib/h
kg/Mg
Ib/ton
1
04/28/81-.
450 .
9.5
10.5
18.1
20
9,000
317,800
7,836
276,800
433
13,840
57
135
1.1
0.6
19.4
131
0.057
150
0.0655
70.5
155.4
7.4
14.8
2
- 04/29/81
450
8.6
9.5
18.1
20
9,486
335,000
8,322
293,800
460
14,690
51
124
1.7
0.6
19.4
124
0.0541
141
0.0617
70.5
155.3
8.2
16.3
3
04/30/81
450
9.3
10.3
18.1
20
8,994
317,600
7,998
282,400
442
;r 14,120
50
122
0.9
0.6
19.4
188
0.0819
211
0.0921
101.2
223.0
10.9
21.6
Average
450
9.1
10.1
18.1
•20
9,160
323,500
8,052'
284,300
445
14,215
53 :
'f27
1.2 .
0.6
19,4
148
0.0643
167
0.0731
80.7
177.9
8.8
17.6
Reference Method 5 data.
C-41
-------
TABLE C-20.
SUMMARY OF PARTICULATE MATTEj
(Fabric filter outlet)'
RESULTS—PLANT Q
31
Run number/date
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
ms/mi n
acfm
dscm/min
dscfm
ds cm/mi n/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
ssions
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
1
04/28/81
480
9.5
10.5
18.1
20
3,990
141,000
3,540
125,100
196
6,255
55
130
0.3
0.6
19.4
0.742
0.000324
0.827
0.000365
0.390
0.859
0.041
0.082
2
04/29/81
480
8.6
9.5
18.1
20
4,320
152,800
3,780
133,400
209
6,670
55
131
0.3
0.6
19.4
0.433
0.000189
0.493
0.000217
0.246
0.542
0.029
0.057
3
04/30/81
480
9.3
10.3
18.1
20
4,566
161,300
4,110
142,200
227
7,110
49
119
0.5
0.6
19.4
0.573
0.000250
0.650
0.000284
0.312
0.687
0.034
0.067
Average
480
9.1
TO;!
18.1
20
4,292
151,700
3,810
133,600
210
6,680
53
127
0.37
0.6
19.4
' 0.583
0.000254
0.651
0.000289
0.316
0.696
0.035
0.069
Reference Method 5 data.
C-42
-------
TABLE C-21.
SUMMARY OF PARTICULATE MATTEJ
(Fabric filter outlet)'
RESULTS-PLANT Q32
Test data
Sampling time
Furnace production
Nominal furnace capacity
Shop effluent
Flow rate
Flow rate per unit of
furnace capacity
Temperature
Water vapor
C02 dry
02 dry
Parti cul ate matter emissions
Concentration
Emission rate
Emission factor
Units
*'
min
Mg/h
tons/h
Mg
tons
mVmin
acfm
dscm/mi n
dscfm
dscm/mi n/Mg
dscfm/ton
°C
°F
% volume
% volume
% volume
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
Run No. lb
08/22/78
240
11.6
12.8
18.1
20
811
28,655
719
25,394
39.7
1,270
57.2
135
0.016
—
6.18
0.0027
6.87
0.003
0.30
0.67
0.026
0.052
^Reference Method 5 data.
This test involved only one run.
C-43
-------
TABLE C-22.
SUMMARY OF PARTICULATE MATTEf
(Fabric filter inlet)c
RESULTS—PLANT R33
Run number/date
Test data
Sampling time
Gas velocity
Flow rate
Temperature
Water vapor
C02 dry
02 dry
CO dry
Parti cul ate matter
Concentration
Emission rate
Units
min
m/s
ds cm/mi n
dscfm
°C
, °F
% volume
% volume
% volume
% volume
emissions
mg/dscm
gr/dscf
kg/h
ilb/h
1
02/10/78
251
25.09
17,717
626,038
35.4
95.7
0.0
1.6
19.0
0.0
187.2
0/08174
198.9
438.6
2
02/13/78
189
25.88
17,880
631 ,804
40.3
104.6
0.5
1.1
19.6
0.0
170.3
0.07435
182.6
402.63
Average
220
25.48
17,798
628,921
37.9
100.2
0.25
1.4
19.3
0.0
178.7
0.07804
190.8
420.62
a
Reference Method 5 data.
C-44
-------
CO
CO
ce:
t—
- U_
^> fv. p_
CM S *~
^* CO Ul * CO
*" . . CM "~
cnco i—
CO TJ-
v G§ °co
i— co r~ '<— •—
. . CM
co in >—
CO CO
? oo en "^ co
r— r— f^ f~* f~
• . CM
r*- cn r—
CM CM
* «a- in vo «*•
• • CM
CM CM
«*• o 10 in «
& <3> V> O
Q
i?
u
re
o.
o S
4-> 01
u u
§3 <0 J->
•o c c
•/- 0 t- 0)
*J S- 3 3
Q. **- t—
O) >*-
C tli r— tf«
•r- O ID 01
r— 03 C
tC -r- Q.
t. S 0
re 3 o j=
v) u. s:
CMCousen i < cncM inocnco
CO "Cj* CM CO f^ O O O CO r™
.... CO O ^~
cj r^> r^>ooooo
.... CO O CM
r— in r- CO t—
in in
mcMo^- i i cncM coocnco
incovcM f^ o r- & cr\ o
.... CO O >—
^» ,..
ocor^o i i cncM CMOCOCO
r>.«»-tocn r-o i— o co i—
.... COO •—
r— •— "— cn i—
(O in
r— ^- co CM i i cn CM o o m in
«coco«a- 1^.0 o o p~ CM
.... CO O i—
CMCOCMCM <-
•—«»•«*• cn i i cn CM -co«>co r~ o o o co r—
.... CO O i —
*— o *— cn i —
to in
CM o m eo i i cn CM in o •* o
P>. CM 10 CO l«~ 0 OOCnr-
.... ~ CO O r—
CM ID CM CM r—
cn cn
cnoesicM i i cn CM CM o «s- o
CM "3" r— co r^ o o o cn *«
.... CO C3 *"•
CO *3* CO r^ r—
r— i— •
r>** r>*
C E C E QIC O U- OJ O) <3J 0)
« ETa->-e oooo
B U S>>>
t/i
C
o
•r-
U>
M- B
O S
•U >> S-
c •- 5
3 O 4->
t- Q- S
U 4J rtj •—
n IB TJ (D > ^s 3
1- t. C t- >> S- O
s- N C.
F— •— 41 (8 O NO <0
u- U- t— 3r CJ O O Q-
co if) if> tD co cn i i
r^ CM cn CM ID co i i
m o in o ..
dodo *~
do
CO O CO CM O CM 1 1
•0*0 • •
'-§'-§ ~-*
o o
Cn r— CM CM CO t^ 1 1
CO '^* ID ^* O CM 1 1
cn o en o . .
. O • O F— CM
oooo
d d
r— CO •* Cn r— ID 1 1
<^ CM 'r1 CM r*"» m i i
ID O 1C O . .
. o « O O f—
oooo
d d
CM CM CO CO •«• f~ 1 1
^ ^ O O • •
do o o °
dd
in CM in CM cn in i i
CMOCM O ^ •
• o • o o o
oooo
d d
f»- us r- to in CM i i
r— O •— o • •
oooo
o o
co eo co co o co i i
C5 C3 CO ^~* CM ^J* 1 1
. o *"• C3 d d
oooo
d d
in cn in cn 10 CM i i
^r o «*• o . • *
. o • o o r—
oooo
d d
cn r^. cn r^ r— F— i i
roo co o • .
• o • O O O
OOOO
d d
me"u u 'u ^^ szo
es {_, *\^ *%^ ->s jj5
O* O) S- r-
E 0)
•"^
o
C CD •*->
O 4-> U
•r- m (O
S- C C
*J O O
Q) ui tn
U 01 wi
C -I— •»—
o e e
CJ UJ UJ
TJ
0)
,2
(J
IB
u
o
c
o
+J
u
M-
c
o
C/)
tn
'S
at
•o
c
(0
0
u
•a
Ok
01
u
re
c:
,3
c*.
C
3
S.
01
• Q.
4J 4)
C 4-»
Jre
Z
<~ -s
ro o
£ U_
U
t- T3
o a>
o. tn
o ai
4-> 4->
in -w
£S
O E
ra 4^
s.
•a re
a o.
" 0
ai u
fl)
-------
TABLE C-24. SUMMARY OF VISIBLE EMISSION
DATA FROM DUST-HANDLING SYSTEMS AT EAF AND
AOD VESSEL STEEL MILL FACILITIES3,30,31,36
Plant
Ba
C
Pb
Qd
Qd,e
Length of
observations
minutes
12
20
594
60
• 48
Maximum
6-minute
average
opacity,
percent
0.6
5.0
Oc
Oc
7.3
*EAF fabric filter.
DEAF/AOD fabric filter.
cReference Method 22 (no emissions were
visible). The operation observed did not
include the dust transfer from the storage
.silo to truck.
QAOD fabric filter.
The operations observed included the dust
transfer from the storage silo to truck.
C-46
-------
C.5 REFERENCES FOR APPENDIX C
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Source Testing Report: The Babcock and Wilcox Company Electric Arc
Furnace, Beaver Falls, Pennsylvania. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. EPA Publication No. EMB
73-ELC-l. January 1973. 32 p.
Memo from Terry, W., Midwest Research Institute, to Iversen, R.,
EPA/ISB. October 31, 1980. Site visit report—Babcock and Wilcox,
Beaver Falls, Pennsylvania.
Compliance Tests Under New Source Performance Standards: Florida
Steel Corporation, Charlotte, North Carolina. Sholtes & Koogler,
Environmental Consultants. Gainesville, Florida. January 1980.
Addendum No. 1, Compliance Tests Under New Source Performance
Standards: Florida Steel Corporation, Charlotte, North Carolina.
Sholtes & Koogler, Environmental Consultants. Gainesville, Florida.
January 1980.
Source Sampling Report, Hoeganaes, Inc. (83-00129-01), Gallatin,
Tennessee. Environmental Management Planning and Engineering.
Nashville, Tennessee. October 1981.
Stack Particulate Sampling: North Star Steel Corporation, Monroe,
Michican. Industrial Health Engineering Associates, Inc., Minnea-
polis, Minnesota, for Ferrco Engineering, Ltd. Ontario, Canada.
Project 335-006. February 1981.
Report of Official Air Pollution Emission Tests Conducted on the
Electric Arc Furnace Cadre Baghouse Exhaust at Raritan River Steel,
Perth Amboy, New Jersey, on June 12 and 13, 1980. Rossnagel & Associ-
ates. Medford, New Jersey. Test Report No. 8132. June 17, 1980.
Compliance Sampling of Stack Emissions: Electric Arc Furnace Baghouse
Exhaust Stack, Nucor Steel, Jewett, Texas, on November 17-18, 1981.
Southwestern Laboratories. Houston, Texas. Project No. 54-830A.
December 1981.
Compliance Tests Under New Source Performance Standards: Florida
Steel Corporation, Tennessee Mill Division, Jackson, Tennessee.
Sholtes and Koogler. Gainesville, Florida. December 1981.
Particulate Emission Tests for Lukens Steel Electric Melt Shop.
Fuller Company. Catusauqua, Pennsylvania. September 23, 1973.
Memo from Terry, W., Midwest Research Institute, to Iversen, R.,
EPA/ISB. September 3, 1980. Site visit report—Florida Steel
Corporation, Charlotte, North Carolina.
C-47
-------
12. Memo from Terry, W., Midwest Research Institute to Iversen, R. ,
EPA/ISB. December 10, 1980. Site Visit Report—Hoeganaes Corporation,
Gallatin, Tennessee.
13. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R., EPA/ISB. January 6, 1981. Site visit report—North
Star Steel, Monroe, Michigan.
14. Letter from Christiansen, J., Raritan River Steel Co., to Iversen, R. E.,
EPA/ISB. September 24, 1981. Response to request for information
about the Perth Amboy, New Jersey, plant.
15. Telecon. Terry, W., Midwest Research Institute, to Bottoms, P.,
Nucor Steel. September 30, 1981. Information on control system and
compliance test at the Jewett, Texas, plant.
16. Memo and attachment from Terry, W., Midwest Research Institute,
to EAF files. May 17, 1982. Trip reports to Lukens Steel, Coatsville,
Pennsylvania (August 1972), and Bethlehem Steel Corporation, Seattle,
Washington (March 1973).
17. Memo from Banker, L., Midwest Research Institute, to Iversen, R.,
EPA/ISB. July 28, 1980. Source emission test observation report--
Jones and Laugh!in Steel Corporation, Pittsburgh, Pennsylvania.
18. Letter and attachments from Lukas, A. W., J&L Steel Corporation, to
Banker, L. C., Midwest Research Institute. March 2, 1981. Submis-
sion of compliance test report for J&L-Pittsburgh Works.
19. Visible Emission Data for J&L, Pittsburgh, Pennsylvania. Allegheny
County Health Department. Pittsburgh, Pennsylvania. Undated.
20. Electric Arc Furnace Baghouse Compliance Test: Sharon Steel Corpora-
tion, Parrel!, Pennsylvania, January 6, 7, and 8, 1981. WFI Sciences
Company. Pittsburgh, Pennsylvania. WFI Sciences Report No. 8343.
21. Letter from Heintz, J. K., Sharon Steel Corp. to Iversen, R. E., ..
EPA/ISB. February 2, 1981.' Response to request for information
about the Sharon, Pennsylvania, plant.
22. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R., EPA/ISB. March 23, 1981. Source test report—Atlantic
Steel, Cartersville, Georgia.
23. Letter from Hawkins, J. B.,|Atlantic Steel Company, to Iversen, R. E.,
EPA/ISB. April 13, 1981. Response to request for information about
the Cartersville, Georgia, plant.
24. Memo and attachments from Banker, L. , Midwest Research Institute, to
Iversen, R., EPA/ISB. March 30, 1981. Source test report—U.S.
Steel, Baytown, Texas.
C-48
-------
25. Letter from Carnes, D. S., U.S. Steel, to Iversen, R. , EPA/ISB.
September 24, 1981. Response to request for information about the
Baytown, Texas, plant.
26. Memo and attachments from Terry, W., Midwest Research Institute, to
Pahl D., EPA/SDB. August 31, 1982. Source Test Trip Report—Chaparral
Steel, Midlothian, Texas.
27. Memo and attachments from Banker, L., Midwest Research Institute,
to Iversen R., EPA/ISB. March 26, 1981. Source Test Trip Report--
Chaparral Steel, Midlothian, Texas.
28. Letter from Nicholson, W. T., Chaparral Steel, to Iversen, R.,
EPA/ISB, December 3, 1981. Response to request for information
about the Midlothian, Texas, plant.
29. Memo arid attachment from Banker, L , Midwest Research Institute, to
Iversen, R., EPA/ISB,. June 3, 1981. Source Test Report—Bethlehem
Steel, Los Angeles, California.
-30. Emission Test Report: AL Tech Specialty Steel Corporation, Watervliet,
New York. EPA/EMB No. 80-ELC-7. July 1981.
31. Emission Test Report: Carpenter Technology Corporation, Reading,
Pennsylvania. PEDCo Environmental, Inc. Cincinnati, Ohio. Contract
No. 68-02-3546, Work Assignment No. 2. July 198T.
32. Memo and attachment from Maxwell, W., H., Midwest Research Institute,
to EAF files. August 18, 1981. August 1978 source test report for
the CarTech-Reading facility.
33. Letter and attachments from Geiser, L. H. •, Carpenter Technology
Corporation, to Banker, L. C., Midwest Research Institute.
January 30, 1981. Submission of test data for Bridgeport and
Reading, Pennsylvania, shops.
34. Memo and attachments from Terry, W., Midwest Research Institute, to
Iversen, R., EPA/ISB. May 8, 1981. Source test observation report—
AL Tech Specialty Steel Company, Watervliet, New York.
35. Memo and attachments from Terry, W./Midwest Research Institute, to
Iversen, R., EPA/ISB. May 18, 1981. Source test.observation
report—Carpenter Technology Corporation, Reading, Pennsylvania.
36. Visible Emission Summary Report for Carpenter Technology Corporation,
Reading, Pennsylvania. PEDCo Environmental, Inc. April 1982.
37. Memo and attachments from Banker, L., Midwest Research Institute,
to Iversen, R. , EPA/ISB. December 28, 1980. Source test observation
report—Eastern Stainless Steel Company, Baltimore, Maryland.
C-49
-------
38. Visible Emission Survey Report: Eastern Stainless Steel Company,
Baltimore, Maryland. PEDCo Environmental, Inc. Cincinnati, Ohio.
Contract No. 68-02-3546, Work Assignment No. 2. December 1980.
".-5
-------
APPENDIX D
EMISSION MEASUREMENT AND
CONTINUOUS MONITORING
D-l
-------
D.I EMISSION MEASUREMENT METHODS
Particulate matter emissions from electric arc furnaces
(SAP's) and argon-oxygen decarburization (ADD) vessels were mea-
sured at-two specialty steel mills. At Plant P the emissions
from two EAF's and two AOD vessels were ducted to a single
positive-pressure fabric filter (PPFF). At Plant Q the emis-
sions from a single AOD vessel were routed to one PPFF. Both
controlled and uncontrolled emissions of particulate matter were
measured simultaneously at each plant.
Because the multiple-furnace arrangement at Plant P pre-
cluded testing for an integral number of heats at all sites,
representative operation of all four units was used as the test-
ing guideline. Tests were interrupted if one of the EAF's expe-
rienced an operational delay of longer than 20 minutes. Periods
of no production that lasted from 1 to 1.5 hours were tolerated
for the AOD vessels, bedause'that amount of downtime normally
occurred at this plant when an AOD vessel was waiting for the
next metal charge from an EAF.
At Plant Q the significance of testing for an integral
number of heats was minimized by the large number of heats sam-
pled (approximately five heats per run). Also, because the
interval between heats, was s|hort (between 5 and 10 minutes),
sampling continued until all traverses were completed and was
not interrupted except for port changes, AOD vessel operational
delays, and incidents of unrepresentative conditions. •
D-2
-------
Visible emissions at the control device outlet and the melt
shop roof were determined at Plants P, Q, and S and at four
carbon steel shops (Plants L, M, N, and 0). Visible emissions,
from the melt shop roof also were determined at Plant J.
Fugitive emissions from dust handling systems were evalu-
ated at Plants P, Q, and J.
The following sections describe the methods used to measure
particulate matter at all inlet sites and the outlet at Plant Q,
modifications made to EPA Reference Method 5* for testing the
PPFF outlet at Plant P, and the methods used to determine visible
i
emissions. The last section discusses the selection of .methods.
D.1.1 Reference Methods
Environmental Protection Agency Reference Methods 1, 2* 3,
and 5* were used to measure volumetric flow rates and particu-
late-, emissions at the inlet sites at'both plants and the outlet
! -' • ' ' '
sites at Plant Q. •• \
Sampling points were located in the duct cross sections
according to Method I, except in the case of the inlet site at
Plant Q, which did not meet minimum specifications for distance
from flow disturbances. This site was considered to be: accept-
able for sampling after an evaluation of the velocity profile
and a comparison of measured flows with design flows.
A "type-S" pitot tube and an inclined draft gauge manometer
were used to measure the gas velocity pressures at each sampling
40 CFR 60, Appendix A, July 1, 1980.
D-3
-------
point across the duct (according to the procedures outlined in
EPA Reference Method 2) to determine an average value. A thermo-
couple and potentiometer were used to measure the temperature at
each sampling point.
Flue gas composition was determined by the procedures
described in EPA Reference Method 3. Integrated bag samples of
gas collected during preliminary runs were analyzed for oxygen
and carbon dioxide by use of an Orsat Gas Analyzer. Since these
results verified that the gas streams were essentially air,
additional samples were not collected.
Reference Method 5 was used to measure particulate con-
centrations. Tests were conducted isokinetically by traversing
the cross-sectional area of the stack and regulating the sample
flow rate relative to the gas 'velocity in the duct as measured
by the pitot tube and thermocouple attached to the sample probe.
The Plant Q outlet sampling train (shown in Figure D-l) con-
sisted of a heated, glass-lined probe, heated 87-mm (3-in.)
diameter glass fiber filter (Gelman Type AE), and a series of
Greenburg-Smith impingers followed by an umbilical line and
metering equipment. The inlet sampling train (shown in Figure
D-2) was similar except that the probe was lined with 316 stain-
less steel and a Teflon sample line was used between the filter
and first impinger. At the end of each test, the nozzle, .probe,
and filter holder portions of the sample train were acetone-
rinsed. The acetone rinse and filter media were dried at room
temperature, desiccated to a constant weight, and weighed on an
D-4
-------
o
c
to
CL
E
re
to
(O
«
a.
Q
O)
3
cn
— O
CO
D-5
-------
GJ
*c
•f*»
o-
c
-------
analytical balance. Total filterable, particulate matter was
determined by adding the net weights of the two sample frac-
tions. The amount of water collected in the impinger section of
the sampling train was measured (any condensate in the sample
line used at the inlet was first drained into the impingers).
The impinger contents of the -Plant Q outlet train were recovered
and analyzed for organic and inorganic condensible matter by
ether-chloroform extraction.
The outlet site at Plant Q consisted of five stacks, which
were all traversed during each particulate test run. The actual
minimum sampling time and volume for these tests were 8 hours
and 10.8 dscm (380 dscf), which exceeded the respective minimum
requirements of 4 hours and 4.5 dscm (160 dscf) specified for
EAF's in Subpart AA of the Federal Register.*•
D.1.2 Modified Reference Methods .
Controlled emissions at Plant'P exited a fabric filter
through a top-mounted monovent in a configuration typical of
PPFF's. This exhaust configuration required that several modifi-
cations be made to EPA reference methods so they could be used
in the outlet tests... •
Figure D-3, a top view of the fabric filter arrangement,
shows the general location of sampling points. Figure D-4, an
end view of the fabric filter, shows the location of the sam-
pling plane used in each compartment with respect to the site
configuration.
*40 CFR 60, Subpart AA, July 1, 1980.
D-7
-------
WP
/
\[
5
UJ
Si
t*»
•M»^(
.-
>—
_ — o — -
:
«
u.
•Ii4*
lo 1
1 ^^ |
K/l
§i
4.|8.T.
ac o.
5 5 J. .
_..x.
«C «n
^
. i
«
•
!
*
o
i
-\
o!
i
i
*
i
»
j
K
3
*j
K
ttJ
5
«g>
OJ
4J
o
s_
ID
4J
U
•r-
i.
oo
t
o
HI
s_
3
D-8
-------
(O
a.
4->
«3
-M
CD
3
o
i-
OJ
O)
C
O
u
o
a.
t
o
O)
en
u_
D-9
-------
Reference Method 5 equipment and modified sampling pro-
cedures were used. The sampling train was similar to those used
at the inlet sites (shown in Figure D-2) except for the lack of
a pitot tube and the use of a glass-lined probe. Tests were
conducted at a constant sampling rate based on the estimated
average velocity of the entire sampling area. This average
velocity was calculated by fiirst converting the total flow rate
measured at the inlet to outlet conditions of temperature,
pressure, and moisture, and then dividing by the total'outlet
sampling area. The resultant average velocity was assumed to
represent each sampling point and was used to calculate an
average isokinetic sampling rate. The heated probe and filter
assembly was suspended from the center of the ridge beam in a
fabric filter compartment, as shown in Figure D-4. The nozzle
was positioned at each of the four sampling points in a compart-
ment by rotating the probe and filter assembly. Compartment
cleaning cycles were sampled as they occurred. Compartment gas
flows were interrupted while test equipment was moved from one
compartment to another, but conditions were allowed to equili-
brate for several minutes before sampling resumed. Outlet
samples were recovered and analyzed in a manner identical to
that used for inlet particulate samples.
As shown in Figure D-3, four compartments were sampled per
run. By the end of three runs, each compartment had been tested
at least once. The actual minimum'sampling time and volume were
t
5.3 hours and 5.9 dscm (208 dscf), which exceeded the respective
minimum requirements of 4 hours and 4.5 dscm (160 dscf).
D-10
-------
D.I.3 Visible and Fugitive Emission Methods
Certified observers recorded visible emissions from melt
shop roofs and fabric filter exhausts according to procedures
described in EPA Method 9.* Data were taken in 6-minute sets,
and individual readings were recorded in percent opacity at 15-
second intervals. Intermittent rest periods were taken to
prevent eye fatigue; however, as long as emissions were visually
detectable, readings were continued until a break was absolutely
necessary. The emission points were casually monitored during
break periods, and readings were resumed if emissions greater
than zero opacity were noticed.
In most cases the configuration of melt shop roofs and
fabric filter exhausts presented a potentially large area in
which visible emissions could occur. These large areas were
separated into two or more smaller segments that were labeled
and simultaneously observed. Then, the magnitude of any emis-
sions that were visible during the observation period was
matched with the segment from which they were emitted. This -
procedure facilitated data collection and comparison of emis-
sions with -furnace operations.
The opacity data typically represent visible emissions
emanating from an area much smaller than the total potential
exhaust area. • For example, at one plant melt shop emissions
escaped through an opening in the side of the building. The
40 CFR 60, Appendix A, July 1, 1980.
D-ll
-------
average opacity recorded on the data sheet represented that
small area, not what the average opacity would have been for the
entire melt shop. Another example would be the typical long and
narrow exhaust of a PPFF. Emissions from the entire exhaust
might be 0 percent opacity until a compartment cleaning cycle
occurs. At that time the opacity from one-tenth (for a 10-com-
partment baghouse) of the exhaust area may be 5 percent opacity
while the rest is 0 percent. The recorded opacity would be 5
percent.
Because the intent was to record only emissions directly
related to operation of the furnaces being evaluated, visible
emissions from nearby furnaces or casting areas were disregarded
with the aid of a process observer inside the melt shop.-- Visible
emission data at Plants P and Q were obtained simultaneously
with those for particulate matter tests.
Fugitive emissions from fabric filter dust-handling systems
i
at Plants P and Q were observed according to proposed Method
22.* Emissions were recorded as the cumulative amount of time
during a 20-minute observation period that any fugitive emis-
sions were visually detectable. Between three and thirty obser-
vation periods were recorded during each test series.
Emissions from the dust-handling systems at Plants Q and J
were observed according to EPA Method 9 by a certified observer.
A visual inspection was made of the dust hoppers beneath the
main fabric filter (s) and the ductwork associated with the •«-
"Federal Register, Vol. 45, No. 224, November 18, 1980
D-12
-------
pneumatic transfer system while dust was being transferred from
the hoppers to the storage bin. No fugitive emissions were
visually detected in these systems at either plant. Between two
and six 6.-minute sets of visible emission data were then re-
corded at the outlet of the small fabric filter, controlling
emissions from the pneumatic transfer system. Between two and
eight 6-minute sets were recorded during the transfer of dust
from the storage bin to a truck. Individual opacity readings .
were taken at 15-second intervals. The amount of data collected
was dependent on the length and frequency of the different
operations.
D.I.4 Discussion of Methods Selected
The inlet sites at Plants P and Q and the outlet site at
Plant Q were amenable to sampling by EPA Reference Methods 1, 2,
3, and 5; therefore, these methods were selected. Long sample
times and large sample volumes were used for the outlet tests to
minimize potential sample handling and weighing errors caused by
low particulate loadings.
The outlet site at Plant P was not directly amenable to
sampling by EPA Reference Methods 1, 2, and 5. In the past, a
typical approach to sampling PPFF's has been to use high-volume
•(hi-vol) type samplers. In general,' particulate concentration
data obtained from EAF installations by hi-vol techniques are
lower than concentrations measured by Method 5 at similar instal-
lations. Specifically, a comparison of particulate concentra-
tions obtained simultaneously at one plant by a hi-vol technique
D-13
-------
and a modified version of Method 5 indicated that the average_
hi-vol particulate concentrations were less than 20 percent of
the corresponding modified Method 5 results.1'2 Because of this
apparent .unresolved discrepancy, the use of a modified Method 5
was preferable to any hi-vol technique for obtaining emission
data that might be used in the; standard setting process.
The modified test methods used at Plant P., which generally
would apply when testing PPFF'-s without stacks, are discussed in
the following sections.
Although they could not be analyzed precisely, the effects
of method variations on outlet particulate concentration results
at Plant P were considered to be relatively minimal. The two
deviations from Method 1 were : the use of a sampling location
less than two equivalent duct diameters downstream from the
nearest disturbance and sampling at fewer than the minimum
number of points. The three deviations from Method 5 were the
lack of velocity monitoring at individual sampling points, the
use of a constant sampling rate at all points during a given
test, and the traversing of only half of the large exhaust area
for each test.
The outlet site configuration offered no sampling location
capable of meeting minimum Method 1 criteria. Therefore, the
throat of the monovent was chosen as an optimum sampling loca-
tion because it represented the smallest cross-sectional area.
D-14
-------
This small area would not only provide the highest gas veloci-
ties, but also offer less chance for bias due to faulty bags.
Sampling at the selected location was supported by a previous
independent study.
Because moving from one traverse point to another during
sampling required test personnel to enter a compartment while
the fabric filter was operating, extraneous dust could have been
stirred up by the accidental bumping of the probe against nearby
beams or by personnel activity, and results could have been
biased if this dust had entered the nozzle during sampling.
Therefore, only four sampling points per compartment were used
so as to lessen the possibility of biasing results. The sam-
pling of only four compartments and the use of extreme care
during point changes precluded the occurrence of these potential
problems in all of the tests.
A better sampling approach for this type of fabric filter
configuration would be to sample from the outside (on the bag-
house roof) and to use ports located in the monovent throat.
This would reduce the possibility of sampling extraneous dust
and shorten the time required to change traverse locations.
More points and compartments per run could then be sampled.
Because safe access to the roof was not readily available, such
ports were not installed at Plant P.
A constant sampling rate was used because accurate measure-
ments could not be taken of individual point velocities to make
isokinetic sampling rate adjustments. The preliminary traverse
D-15
-------
verified the inaccuracy of velocity measurement attempts. Thus,
the average velocity at the outlet sampling location was calcu-
lated by the use of measured inlet data (assuming no air leak-
age) .
At Plant P the inlet and outlet tests were conducted simulta-
neously. Therefore, previously measured inlet velocity data
were used to estimate an average velocity for each outlet test.
Measured data from the simultaneous inlet tests were later used
to calculate real average velocities.
As previously stated, the conversion of inlet flow to out-
let flow assumed no air leakage. At Plant P there were two
sources of air inleakage: the induced draft fans and the open
gratings at the bottom level of the fabric filter. A comparison
of measured versus design flow indicated that the leakage at the
fans was minimal. Entry of dilution air through the open grating
was minimized by covering theiopenings during the tests. In'as
much as a significant inflow of ambient air would have caused a
temperature decrease, the close agreement between inlet and
outlet gas temperatures indicated that any entry of dilution air
was negligible.
The average isokinetic sampling rates indicated for the
outlet tests at Plant P were all within the acceptable range
(100 4- 10 percent). Several factors and assumptions affected
these isokinetic rate calculations: the inlet flow equaled the
outlet flow (previously discujssed); the effect of fabric filter
i
cleaning cycle time and frequency on average outlet velocity
• D-16
-------
could be calculated; the average velocity for the four compart-
ments sampled per run was approximately equal to the average
velocity for all eight compartments; and the average velocity at
each traverse point was approximately equal to the overall
average velocity. As discussed in the following paragraphs, it
is believed that these factors were satisfactorily resolved.
The effect of the fabric filter cleaning cycle could be
determined by first calculating the fraction of time that a
compartment was on line, and then multiplying this fraction by
the total exhaust area to determine the average effective ex-
haust area and the corresponding average effective exhaust
velocity.
The average velocity for the four compartments sampled per
run should have been reasonably close (+10 percent) to the
average velocity for all eight compartments, even if the gas
flow was not evenly distributed between all of the compartments.
This is because each sample run included half of the total
number of compartments.
Although average outlet isokinetic rates were acceptable,
• local gas velocities could have influenced values at individual
traverse points and the results could have been much different.
: The range of actual velocity variation was difficult to gauge
without valid point data; however, the overall isokinetic rates
were within specified limits, and any point-specific biases
should tend to be averaged. ,.
In summary, the results of the modified Method 5 tests at
the Plant P outlet site were fairly consistent and appeared to
D-17
-------
be representative. It is believed that Method 5 equipment can
be used to yield acceptable particulate concentration results at
sites that do not meet Method 1 criteria and where point veloci-
ties cannot be accurately measured. The precision of a method
using a constant sampling rate could be less than that of Method
5, which requires isokinetic sampling at each point.
D.2 MONITORING SYSTEMS AND DEVICES
D.2.1 Opacity Monitoring
Transmissometers suitable for continuously monitoring the
opacity of a confined gas stream, such as in the exhaust stack
of a suction-type fabric filter, are readily available. A
complete continuous opacity monitoring system (COM) is described
by Performance Specification 1 (PS-1) in Appendix B of the
Federal Register.* In a case where condensed moisture would be
present in the gas stream, such as after a wet scrubber, a COM
would not be applicable; therefore, another operating parameter
would need to be monitored as an indication of emissions.
Equipment and installation costs for a COM meeting the re-
quirements of PS-1 are estimated to be approximately $30,000 per
unit, including the cost of the initial Performance Specifica-
tion Test. Annual operating costs are estimated to be approxi-
mately $10,000 per unit including data handling, but not in-
cluding any reporting requirements.
Currently available COM's have not proven to be effective
when applied to exhaust gas streams from PPFF's typically used
'40 CFR 60, Appendix B, July 1, 1980.
D-18
-------
at EAF and/or AOD vessel installations. Since PPFF's frequently
have multiple, exhaust points, COM's would have to be installed
at each point.,, thereby greatly increasing capital and annual
costs. The othe^r _.f reguent type of PPFF exhaust configuration is
a long, narrow ven-t. There are several technical problems
associated with positioning a single COM to monitor several
small emission points or one long emission path. These include:
1. The narrow path of the COM measurement beam does not
always cover a representative portion of the fabric
filter emissions. Stratification of particulate
emissions can occur when bags leak or tear, causing a
significant increase in localized opacity. However,'
because of stratified flow, the gas stream with in-
creased opacity may not intercept the path of COM
measurement beam.
2.
3.
4.
.The signal response of the COM measurement beam cross-
ing the exhaust of many fabric filter compartments is
relatively insensitive to the performance of any
individual fabric filter compartment. A significant
emissions increase from a single compartment due to an
equipment, failure may appear as only a small increase
in the opacity signal of a COM that has a long measure-
ment path and may not be sufficient indication to
plant personnel that additional maintenance is re-
quired.
, The opacity standard is based on emissions from much
shorter;. .path. _l.engths than the lengths of many roof
monitors, requiring a correction factor for the COM to
relate £j^e J-cing path COM reading to the emission
standard." The correction factor magnifies the defi-
ciency described in Item 2.
Unless the COM measurement beam is enclosed in the
spaces iSetween multiple emission points, interference
from the ^aipbient air (e.g., dust, fog) can indicate
higher e'm'i'ssions than are actually present in the
exhaust. .
stream.
Thus, a COM may^ not be a reliable indicator of good emis-
sions control and prQger. operation and maintenance techniques
D-19
-------
when used on a PPFF that does not have a single discharge stack.
However, Reference Method 9 measurements can be made by certi-
fied plant personnel in lieu of COM data where PPFF1s without
single stacks are used.
The approach would require that a visible emissions ob-
server walk the length of the fabric filter and briefly observe
the emissions from each compartment or exhaust to determine if
emissions occur. If emissions do occur, the observer would
conduct a longer-term visible emission test of the specific
exhaust(s) to document the magnitude of emissions.
The walk-through observation would be required once per day
(five times per week) during the melting and refining phase of a
heat cycle. Three 6-minute periods of observation would be
required for each emission point exhibiting visible emissions.
This approach is based on the concept that minor fabric filter
malfunctions generally manifest themselves gradually.
The preceding paragraphs on technical problems associated
with COM's on PPFF's and the proposed use of daily Method 9
observations in place of a COM are based on EPA analysis dis-
4
cussed in an EPA report.
The annual cost of an approach using daily Method 9 obser-
vations is estimated to be approximately $5,000 per control
device, including the cost of observer certification.
D.2.2 Flow Monitoring Systems
Subpart AA contains several continuous flow monitoring
requirements for EAF emission capture systems: the flow in each
D-20
-------
separately ducted emission capture hood must be monitored; the
internal static pressure of the. furnace free space must be moni-
tored when process emissions are captured by a direct shell
evacuation control (DEC) system; and no monitoring is necessary
on systems that capture all emissions (i.e., building evacua-
tion) . , ' •
These monitoring requirements were included in Subpart AA
as an enforceable means .of ensuring good capture of EAF emis-~
sions. The originally proposed concept of imposing opacity
limits for visible emissions from melt shops was- considered to
be unenforceable, principally'because of problems associated
with multifurnace shops. The monitoring provisions were in-
tended to work in the following way. The source was initially
required to demonstrate that its emission capture system effec-
tively captured EAF emissions during all phases of operation.
As an example, this might mean that "X" amount of flow was
required at Hood A and "Y" amount was needed at.Hood B during'
the melt phase. During tapping or charging operations, the
capture system might have to be adjusted so that 0.5X amount of
flow was required at Hood A and 3Y was needed at Hood B. These
flow adjustments would usually be made by changing various
damper positions and maintaining constant fan conditions. Once
the emission capture system was operating effectively/ the
various flow rates required in each hood during each phase of
operation would be measured (e.g., X, 0.5X, Y, 3Y). Velocity
probes .would then be installed in appropriate ducts, and one or
D-21
-------
more recorders would be used to provide continuous output of the
flow in each hood. Anytime the monitoring system indicated a
flow rate less than the predetermined amount required for
effective capture of emissions,, operation could be deemed to be
unacceptable.*
Since the promulgation of Subpart AA, compliance with the
flow and pressure monitoring provisions therein has been at-
tempted in several different ways. One plant has a simple
system that records static pressure in the combined duct Of an
emission, capture system consisting of a DEC and canopy hood
ducted to one control device. Another plant has velocity probes
in three of four separate ducts that continuously record flow
rates. Other EAF installations have tried various combinations
of duct static pressures, fan amperages, damper position indi-
cators, and furnace mode indicators. Some sources have used
such systems for local monitoring requirements or simply to
ensure proper operation of the capture system.
Three general approaches to satisfying the intent of the
flow monitoring requirements are discussed: 1) periodic opera-
tional status inspections, 2) monitoring of system parameters
(other than velocity), and 3) flow monitoring. These discus-
sions are followed by a discussion of performance specifica-
tions, costs, and availability of equipment for the different
approaches.
Federal Register, Vol. 40, No. 185, pp. 43851-43852, September
23, 1975.
D-22
-------
A simple means of monitoring the operation of an emission
capture system is to use periodic operational status inspec-
tions. This approach is based on two premises: 1) most emis-
sion capture systems that need to alternate air flow between one
or more hoods (or DEC ducts) will have a microprocessor or
equivalent set up to make appropriate damper adjustments (auto-
matically or semiautomatically) according to furnace mode, and
2) normal operation of an emission capture system should not
vary except during maintenance or equipment malfunctions.
Initial demonstration of effective capture would still be re-,
quired. At that time the various combination of operating
parameters (e.g., damper position, fan amps, furnace static
pressure) that produced effective capture would be recorded.
Periodic (e.g., monthly) inspections of the system would include
giving specific attention to these parameters.
The second approach to flow monitoring actually consists of
two alternatives. The first alternative includes maintaining a
log of key operating parameters (e.g., damper position, fan
amps, furnace static pressure) on a once-per-shift basis. This
would essentially be a more frequent operational-status inspec-
tion (as just described). The second alternative would include
recording these key parameters on strip charts so they could be
compared with initial values. This entire second approach is
based on the concept that the effectiveness of the emission
capture system can be evaluated by observing parameters that are
either indicative, proportional, or correlated to flow, as
opposed to actually measuring flow.
D-23
-------
The third approach would involve installation of velocity
probes and chart recorders, which is essentially the same as
what is currently required by Subpart AA.
Only the third approach would require any type of per-
formance specification. The monitoring provisions of Subpart AA
include several general equipment specifications, but no appli-
cable Performance Specification is included in Appendix B of the
Federal Register.* The EPA is in the process of developing a
draft versi9n of a Performance Specification applicable to flow
monitoring systems. Although intended.for use specifically on
the emission control systems of basic oxygen furnaces, the
specification should be generally applicable to any flow monitor
now required by other New Source Performance Standards (NSPS).
Therefore, an additional Performance Specification especially
for EAF's is not necessary.
No additional equipment would be required for the periodic
operational status inspections. The annual cost of monthly
inspections is estimated to be $600 per furnace.
No additional equipment would be needed to maintain logis of
key operating parameters because these parameters will have
indicators already. The annual cost of maintaining once-per-
shift logs is estimated to be no more than $13,000 per furnace or ressel.
The cos.ts of the monthly operational status inspections or
once-per-shift parameter logs can be absorbed by plant operating
*40 CFR 60, Appendix B, July 1, 1980.
D-24
-------
budgets with little or no impact because the work can be done
without hiring additional personnel.
Recorders are readily available for continuous monitoring
of operating parameters. Assuming two multipoint recorders
would be needed per furnace, equipment costs are estimated to be
$6,000, including installation. Annual operating and mainte-
nance costs, including data handling, are estimated to be $25,000
per -furnace.
Velocity probes suitable for use in EAF emission capture
systems are readily available. Operational history of velocity
monitoring systems on EAF's appears to be limited, but probes
from at least two different manufacturers have been used with
relative success. One plant indicated that two of three probes
(multi-point, averaging pitot tubes) used in a system that had
been operating for 2 years required monthly cleaning, whereas
the third probe needed cleaning every tw.o weeks. Another plant
using a different type of velocity device (continuous purge
pitot tube) had been using two probes for over a period of a
year with minimal attention. The reliability of a third type of
velocity probe (heavy duty anemometer) has not been demonstrated
on EAF's, but there is evidence to indicate that it should be
suitable for this application without requiring more than
monthly cleaning.
A typical EAF and/or AOD vessel installation is estimated
to require either two or three velocity probes and one or two
multipoint strip chart recorders per furnace. Equipment and
D-25
-------
installation costs (including the initial Performance Specifica-
tion Test) would be between $15,000 and $20,000 depending on the
number and type of devices selected. Annual operating and
maintenance costs would be between $25,000 and $30,000 per
furnace, with the major portion of these costs applied to opera-
tion and data handling. Equipment costs were obtained from
vendors-or vendor catalogs; installation costs are based on the
work being done by plant personnel; and operating and mainte-
nance costs are baseWon comments^from several users.
D.3 PERFORMANCE TEST METHOD'S
Reference Method 5 is recommended for determining particu-
late concentration at any type of control device outlet capable
of providing a sampling site that meets minimum Reference Method
1 criteria (e.g., a single or multiple stack that is more than
i
2.5 duct diameters long).
At other sites (e.g., PPFF's with monovents, ridge vents,
or stub stacks less than 2.5 duct diameters long), proposed
Reference Method 5D is preferred over any type of hi-vol tech-
nique. It may be necessary to determine specific modifications
on a case-by-case basis, depending on the particular facility.
The procedures discussed in Sections D.I.2 and D.I.4 can be used
as guidelines for alternative procedures.
Reference Method 9 can be used to determine visible emis-
sions from melt shops, control^device outlets, and dust handling
equipment. These methods are consistent with the methods used
to gather emission data for the development of this document.
D-26
-------
Subpart A of the Federal Register* requires (emphasis
added) that facilities 'subject to tiSPS be. constructed so as to
provide sampling ports, platforms, access, and utilities to
conduct performance tests by the designated methods. It is
recommended that- these details be discussed by the plant and
appropriate control agency during permit application prior to
construction to avoid costs of retrofitting.
Depending On the site configuration, a performance test
consisting of three Method 5 or Method 5D sampling runs on one '
control system may require between 4 .and.8 man-days of field
work. Total cost of an. initial performance .test on one control
system is estimated to range from $5,000 to $8,000 including the
test report.
40 CFR 60, Appendix A, July 1,- 1980.
P-27
-------
REFERENCES
1.
2.
3
4.
Rossnagel and Associates. Report of Emission Tests on an
Electric Arc Furnace Baghouse Exhaust at Raritan River
Steel. Report Ho. 8132, June 17, 1980.
MMT Environmental. Compliance on a Baghouse at Raritan
River Steel. Report No. 0302, June 18, 1980.
Fluidyne Engineering Corporation. Effective Sampling
Techniques for Particulate Emissions From Atypical Sources.
Preoared for U.S. Environmental Protection Agency under
Contract No. 68021796. EPA 600/2-80-034, January 1980.
U.S. Environmental Protection Agency. Emission Standards
and Engineering Division Recommendations for Emission Mea-
surement of Positive Pressure Baghouses. Prepared by Peter
Westlin, December 1981.
D-28
-------
TECHNICAL REPORT DATA
{Pisase read Instructions on the reverse before completing
EPA 450/3-82-020a
2.
13. RECIPIENT'S ACCESSION NO.
"LE ANP SUBTITLE
Electric Arc Furnaces and Argon-Oxygen Decarburization
Vessels in the Steel Industry—Background Information
for Proposed Revisions to Standards
|5. REPORT OATS
July 1983
6. PERFORMING ORGANIZATION COOS
8. PERFORMING ORGANIZATION REPORT NO.
-- • -RFORMING ORGANIZATION NAME ANO ADQRESS
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.
68-02-3059
G AGENCY NAME ANO ADDRESS
Director for Air Quality Planning and Standards
Office of Air, Noise, and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
EPA/200/04
Standards of Performance for the control of emissions from electric arc furnaces and
argon-oxygen decarburization vessels in the steel industry are be nVproposed under
authority of Section 111 of the Clean Air Act. These standards would Ip£°y to those
2p^eC?hSrUqted' °r mod1f1ed elec^ic arc furnaces and argon-oxygen SScaVSrlStlon
vessels that commence construction on or after the date of proposal of the requ ations
asses^ntf nV;!!ta1nS p?***™* information and environmental and economic fmjact
assessments of the regulatory alternatives considered in developing the proposed standards
7.
WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
ir pollution
Dilution control
tandards of performance
tee! plants
lectric arc furnaces.
rgon-oxygen decarburization vessels
3. DISTRIBUTION STATEMENT
nlimited
b.lDENTIFIERS/pPEN ENDED TERMS
Air pollution control
19. SECURITY CLASS /This Report)
Unclassified
EPA Form 2220-1 (Rev. 4-77) PREVIOUS eb.noN is ossousre
CLASS fThis page!
Unclassified
c. COSATI 'Field/Croup
21. NO. OF PAGES
412
-------
-------
-------
rSf| =
! ~3?3
iw = oG
"HI!
fiSgi
ill
§3 OTJ
-sg-f
111!
.^is
o=So
S.I-' S
5| I
II I
O C
fl
® «•
co CD 3»
83 =n
Q) O
'gz
"Eo.
------- |