vvEPA
          United States
          Environmental Protection
          Agency
          Office of Air Quality
          Planning and Standards
          Research Triangle Park NC 27711
  June 1982
          Air
Electric Arc
Furnaces and
Argon-Oxygen
Decarburization
Vessels in
Steel Industry —
Background
Information for
Proposed Revisions
to Standards

Preliminary Draft
Draft
EIS

-------
                          NOTICE

This document has not been formally released by EPA and should not now be construed to represent Agency
policy. It is being circulated for comment on its technical accuracy and policy implications.
            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

-------
                            TABLE OF CONTENTS

Section      Title                                                 Page
List of Figures	„	      v

List of Tables	    vii

1             SUMMARY	    1-1

1.1          Regulatory Alternatives 	    1-1

1.2          Environmental Impact  	    1-2

1.3          Economic Impact	    1-2

2            INTRODUCTION  	    2-1

2.1          Background and Authority for Standards  	    2-1

2.2          Selection of Categories of Stationary Sources .  .  .    2-5

2.3          Procedure for Development of Standards of
               Performance	    2-6

2.4          Consideration of Costs	    2-9

2.5          Consideration of Environmental  Impacts  	    2-10

2.6          Impact on Existing Sources  	    2-11

2.7          Revision of Standards of Performance  	    2-11

3            ELECTRIC ARC FURNACES AND ARGON-OXYGEN
               DECARBURIZATION VESSELS IN THE STEEL INDUSTRY:
               PROCESSES AND POLLUTANT EMISSIONS 	    3-1

3. 1          General	    3-1

3.2          Process Facilities and Their Emissions  	    3-15

3.3          Emissions	    3-35

3.4          References for Chapter 3	    3-43

4            EMISSION CAPTURE AND CONTROL TECHNIQUES 	    4-1

4.1          Introduction	    4-1

4.2          Capture of EAF Process and Fugitive Emissions .  .  .    4-2

4.3          Capture of ADD Vessel Process and Fugitive
             Emissions	    4-19


                                   ii

-------
                      TABLE OF CONTENTS (continued)

Sectlon      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 .  .    .  .     ...      .  .    6-13

6.4          References for Chapter 6	    6-17

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

7.7          References for Chapter 7	      7-28

8            COSTS   ...       	         	    8_]

8.1          Cost Analysis of Regulatory Alternatives   .     .      8-1

8.2          Other Cost Considerations  ...           ...    8-12
                                    11

-------
TABLE OF CONTENTS (continued)
Section
8.3
9
9.1
9.2
9.3
9.4
APPENDIX A-

APPENDIX B:
APPENDIX C:
C.I
C.2
C.3
C 4
C.5
APPENDIX D:
D. 1
D.2
D.3
Title
References for Chapter 8 	
ECONOMIC IMPACTS 	
Summary of Impacts 	
Industry Profile 	
Economic Impacts of Regulatory Alternatives . . .
References for Chapter 9 	
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 	
References for Appendix C .... 	
EMISSION MEASUREMENT AND CONTINUOUS MONITORING . .
Emission Measurement Methods 	
Monitoring Systems and Devices 	
Performance Test Methods 	
Page
8-14
9-1
9-1
9-2
9-58
9-119
A-l

B-l
c-i
C-2
C-16
C-21
C-21
C-47
D-l
D-2
D-18
D-26
             IV

-------
                             LIST OF FIGURES

No.   Title                                                          Paqe
	  	                                                          	2—

3-1   Flow Diagram of an Iron and Steel Plant	            3-2

3-2  Schematic Flowchart for Integrated and Noni integrated
       Steelmaking   ....     .             	    3-3

3-3  Steel Production by Furnace Type	    3-14

3-4  Electric Arc Steel Furnace	        -  •    3-17

3-5  Material Balance of Electric Arc Furnace Based on  1,000  kg
       of Steel Produced	      .         ...      .    .  .    3-26

3-6  Argon-Oxygen Decarburization Vessel  .       	      3-29

3-7  Carbon-Chromium Equilibrium Curves  .       	    3-32

4-1  Direct-Shell Evacuation Control (Two Views)  	    4-4

4-2  Side Draft Hood (Two Views)     	    4-6

4-3  Partial Furnace Enclosure      	      .  .      4-8

4-4  Total Furnace Enclosure at Lone Star Steel  Company       .      4-10

4-5  Canopy Hood Capture System     	      4-13

4-6  Diverter Stack With Canopy Hood            .    ....      .    4-20

4-7  Close-Fitting Hood With Canopy Hood .  .     .      ...          4-22

4-8  Summary of Particulate Matter  Source Data  for Carbon Steel
       EAF Fabric Filters (Reference Method 5)  	    4-31

4-9  Summary of Particulate Matter  Source Data  for Specialty
       Steel Shop Fabric Filters (Reference Method 5)    ...      4-37

9-1  Supply Schedules for Constructed and Unconstructed Plants      9-93

9-2  NSPS Effects on Market Price and Quantity      ....    9-94

9-3  NSPS Effects on the Average Total Cost of  New Plants,
       Equilibrium Price and Equilibrium Quantity   	    9-100

9-4  Current Steel Market Disequilibrium and  Long-Run
       Equilibrium          .     .               .        _            9-109

9-5  NSPS Effects on U.S. Steel Market Long-Run  Equilibrium   .  .    9-111

-------
                       LIST OF FIGURES (continued)

No.   Title

C-l   Summary of Particulate Matter Source Test Data From Fabric
       Filters on EAF's at Carbon Steel Shops 	   C-3

C-2  Summary of Particulate Matter Source Test Data for
       Specialty Steel Shop Fabric Filters  	   C-17

D-l   Particulate Sampling Train at Plant Q Outlet 	   D-5

D-2  Particulate Sampling Train at Plants A and Q Inlets  ....   D-6

0-3  Fabric Filter Outlet at Plant P	   0-8

0-4  Sampling Location at the Fabric Filter Outlet at Plant P .  .   0-9

-------
                             LIST OF TABLES

No.    Title

1-1    Environmental and Economic Impacts of Various Alternatives
        Compared to Alternative 1 (Baseline) in the Fifth Year
        (1987)	        	          1-3

1-2   Matrix of Environmental and Economic Impacts for
        Regulatory Alternatives  ...      	    1-4

3-1    Electric Arc Furnaces in the United States--1981    .    .  .    3-6

3-2   Location of Steel Plants With EAF's and ADD Vessels--
        1981 .   	    3-9

3-3   Industry Percentages of EAF's and AOD Vessels by
        Capacity--1981	      	      3-10

3-4   Argon-Oxygen Decarburization Vessels in the United
        States--1981 	        3-11

3-5   Steel Industry Data and Projections  ....      	    3-13

3-6   Raw Materials and Products-~EAF and AOD Operations  .  .  .      3-27

3-7   Particulate Matter Emission Factors (Uncontrolled)  .    .  .    3-37

3-8   Trace Constituent Emission Factors (Uncontrolled)   .      .    3-37

3-9   Chemical Analysis of Electric Arc Furnace Dust by Phase of
        Furnace Operation  	    3-38

3-10  Exhaust Gas Particulate Matter Composition 	    3-39

3-11  Size Distribution of Particulate Matter Emissions From
        Steelmaking EAF and AOD Facilities	    3-40

3-12  Summary of State Air Pollution Regulations  .  .      ....   3-42

4-1   Fugitive Emissions Capture Technology Combinations
        (Capture and Specialty Steel EAF)  	    4-23

4-2   Fugitive Emissions Capture Technology Combinations
        (Specialty Steel AOD)	    4-25

4-3   Summary of Carbon Steel Plant Data	             .    4-30

4-4   Summary of Visible Emission Data from Fabric Filters  on
        EAF's at Carbon Steel Shops    .              ...      4-33

4-5   Summary of Opacity Data From Shop Roof Monitors on  Carbon
        Steel EAF Shops            	          .        .      4-34


                                   vi i

-------
                       LIST OF TABLES (continued)

No.   Title

4-6   Summary of Specialty Steel Plant Data  	   4-36

4-7   Summary of Visible Emission Data From Fabric Filters on
        EAF's and AOD's at Specialty Steel Shops	   4-38

4-8   Summary of Visible Emission Data From Shop Roof Monitors
        on Specialty Steel Shops 	   4-39

4-9   Summary of Visible Emission Data From Dust-Handling
        Systems at EAF and ADD Vessel Steel Mill Facilities  .   .   4-41

4-10  Summary of Trace Constituent Concentrations Analysis, ppm .   4-42

6-1   Model Plants	   6-2

6-2   Model Furnace Parameters — Carbon Steel	   6-4

6-3   Model Furnace Parameters—Specialty Steel	   6-5

6-4   Model Plants With Capture Configuration Options  	   6-7

6-5   Air Flow Rates Per Unit of Furnace/Vessel Capacity ....   6-8

6-6   Model Furnace Parameters—Carbon Steel Fabric Filter
        Information (Metric Units) 	   6-9

6-7   Model Furnace Parameters—Carbon Steel Fabric Filter
        Information (English Units)  	   6-10

6-8   Model Furnace Parameters—Specialty Steel Fabric Filter
        Information (Metric Units) 	   6-11

6-9   Model Furnace Parameters—Specialty Steel Fabric Filter
        Information (English Units)  	   6-12

6-10  Regulatory Alternatives—Carbon and Specialty Steel EAF   .   6-15

6-11  Regulatory 'Alternatives —Specialty Steel  AOD Vessel  .  .   .   6-16

7-1   Annual Particulate Matter Emissions for Each Model  Furnace
        and Regulatory Alternative 	  ....   7-2

7-2   Annual Particulate Emission Reduction Below Uncontrolled
        and Baseline Levels for Each Model Furnace/Vessel Size  .   7-3

7-3   Projected Construction of Model EAF/AOD Plants Industry
        Wide,  1983-1987	   7-5
                                   VI 1 1

-------
                       LIST OF TABLES (continued)

No.    Title                                                        Łag_e

7-4   Total Participate Matter Emissions for New Shops Added
        in 1983-1987 and 5-Year Cumulative 	   7-6

7-5   General  Data and Options of ISCST Model Used in Atmospheric
        Disperson Modeling Analysis  	   7-7

7-6   Modeling Data for Fabric Filter Sources  	   7-11

7-7   Modeling Data for Roof Monitor Sources	-   7-12

7-8   Highest Second-Highest 24-Hour Particulate Matter
        Concentration Impacts From Fabric Filter Sources ....   7-16

7-9   Highest Second-Highest 24-Hour Particulate Matter
        Concentration Impacts From Roof Monitor Sources       .  .   7-18

7-10  Maximum Annual Arithmetic Mean Particulate Matter
        Concentration Impacts From Fabric Filter Sources ....   7-19

7-11  Maximum Annual Arithmetic Mean Particulate Matter
        Concentration Impacts From Roof Monitor Sources  ....   7-20

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 Fi1ters--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 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   Projected Nationwide Control System Expenditures Under
        Various Regulatory Alternatives (by  1987)  	   8-9
                                   IX

-------
                       LIST OF TABLES (continued)

No.   Title                                                        Page
8-6   Average Cost Effectiveness of EAF/AOD Process and Fugitive
        Emissions Control Over Baseline  	   8-10

8-7   Cost Estimates for OSHA Compliance--EAF/AOD Vessel
        Facilities	   8-13

9-1   Important Inputs to the Blast Furnaces and Steel Mills
        Industry--SIC 3312	   9-8

9-2   Raw Steel Production by Process Type	   9-13

9-3   Pig Iron and Scrap Inputs to Raw Steel Production  ....   9-14

9-4   U.S. Real Gross National Product and Apparent Consumption
        of Steel Mill Products	   9-16

9-5   Important Purchasers of Output From the Blast Furnaces and
        Steel Mills Industry--SIC 3312	   9-18

9-6   Plant Integration  	   9-22

9-7   After-Tax Profit to Stockholders' Equity 	   9-31

9-8   U.S. Companies Operating Electric Arc Furnaces:   Financial
        Data if Available (1979)	   9-33

9-9   Financial Ratios for U.S.  Companies Operating Electric
        Arc Furnaces (1979)	   9-36

9-10  Financial Ratios for Selected Industries (1979)  	   9-38

9-11  Steel Mill Products and Total Industrial Output Indexes   .   9-39

9-12  Real Value of Output for SIC 3312	   9-41

9-13  Steel Price Index and GNP Price Deflator 	   9-42

9-14  Indexes of Real New Investment	   9-43

9-15  Index of Output per Employee-Hour	   9-45

9-16  Steel Mill Products Trade	   9-46

9-17  Projected Domestic Raw Steel Production  	   9-48

9-18  Summary of Electric Arc Furnace Capacity Projection  .  .   .   9-49

9-19  Projected Construction of Electric Furnace Plants,
        1982-1987	   9-51

-------
                       LIST OF TABLES (continued)

No.    Title                                                        Page

9-20  Average Total  Cost Impacts Assuming 6.2 Percent Weighted
        Average Cost of Capital	   9-54

9-21  Total Annualized Cost of Compliance Above Baseline
        as a Percent of Annual Plant Sales   	9-56

9-22  Compliance Capital Cost Above Baseline as a Percent of
        Baseline Plant Capital Cost  	 9-57

9-23  Model Plant Cost Data	   9-67

9-24  Model Parameter Values 	  .   9-69

9-25  Project Net Present Values Assuming 6.2 Percent Weighted
        Average Cost of Capital	  9-72

9-26  Net Present Value Impacts  Assuming 6.2 Percent Weighted
        Average Cost of Capital	9-73

9-27  Project New Present Values Assuming 10.0 Percent Weighted
        Average Cost of Capital	  9-75

9-28  Net Present Value Impacts  Assuming 10.0 Percent Weighted
        Average Cost of Capital	9-76

9-29  Average Total  Cost Assuming 6.2 Percent Weighted
        Average Cost of Capital	  9-78

9-30  Average Total  Cost Impacts Assuming 6.2 Percent Weighted
        Average Cost of Capital	  9-80

9-31  Average Total  Cost Assuming 10.0 Percent Weighted
        Average Cost of Capital	  9-81

9-32  Average Total  Cost Impacts Assuming 10.0 Percent Weighted
        Average Cost of Capital	  9-82

9-33  Domestic Steel Production  Impacts In 1987 Assuming'
        6.2 Percent Weighted Average Cost of Capital  	  9-85

9-34  Domestic Steel Industry Employment Impacts for 1987 ....  9-86

9-35  Steel Imports Impacts for  1987	    9-88

9-36  Net Present Value and Average Total Cost Data for Model
        Plants 4 and 7	9-90
                                    XI

-------
                       LIST OF TABLES (continued)

No.    Title                                                        Page

9-37  Summary of Economic Impacts From Baseline to Regulatory
        Alternative 2	   9-102

9-38  Summary of Economic Impacts From Baseline to Regulatory
        Alternative 3	   9-104

9-39  Capital Requirements of Regulatory Alternatives 	   9-106

9-40  Industry Debt Ratios	9-107

9-41  Total Cost of Regulatory Alternative 2 in 1987	9-115

9-42  Total Cost of Regulatory Alternative 3 in 1987	9-116

9-43  Total Cost of Regulatory Alternatives in the Very Long Run   9-118

A-1    Evolution of the Background Information Document  	   A-2

B-l    Cross-indexed Reference System to Highlight Environmental
        Impact Portions of the Document 	   B-2

C-l    Summary of Plants Tested and Type of Tests Performed  .  .  .   C-23

C-2   Summary of Visible Emission Data From Fabric Filters on
        EAF's at Carbon Steel Shops	   C-24

C-3   Summary of Opacity Data From Shop Roof Monitors at Carbon
        Steel EAF Shops	C-25

C-4   Summary of Particulate Matter Results--Plant A (Fabric
        Filter Inlet) 	     C-26

C-5   Summary of Particulate Matter Results — Plant A (Fabric
        Filter Outlet)  	     C-27

C-6   Summary of Particulate Matter Results--Plant B (Fabric
        Filter Outlet—Canopy Hood))	     C-28

C-7   Summary of Particulate Matter Results — Plant B (Fabric
        Filter Outlet—Side Draft Hood) 	     C-29

C-8   Summary of Particulate Matter Results—Plant C (Fabric
        Filter Outlet)  	     C-30

C-9   Summary of Particulate Matter Results—Plant D (Fabric
        Filter Outlet)  	     C-31

C-10  Summary of Particulate Matter Results—Plant E (Fabric
        Filter Outlet)  	     C-32


                                   xi i

-------
                       LIST OF TABLES (continued)

No.    Title                                                        Page

Oil   Summary of Particulate Matter Results--Plant F (Fabric
        Filter Outlet)      ...      ...     ....     .     O33

C-12  Summary of Particulate Matter Results--Plant G (Fabric
        Filter Outlet .        	      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                       ....         O36

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)	O38

C-17  Summary of Particulate Matter Results--Plant P (North
        Fabric Filter Inlet)    	           	O39

C-18  Summary of Particulate Matter Results--Plant P (Fabric
        Filter Outlet)	        ....          O40

C-19  Summary of Particulate Matter Results--Plant P (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

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
                                   XI 1 1

-------
                                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 Alternative 2 show a decrease from the baseline
levels.  These savings in capital and the annualized costs are due to
revision of the existing standards for EAF's to permit 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 AOD vessel in lieu of continuous flow monitoring and of
continuous opacity monitoring.  Under Regulatory Alternative 3, however,
these  savings are overcome by additional expenditures in the capital and
annualized costs of larger control and capture systems than under Regulatory
Alternative  1.
     Cost impacts on product price are generally quite small for the
regulatory alternatives.  Compared to Regulatory Alternative 1, Regulatory
Alternative 2 would result in net savings 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 all under $1.50 per megagram (Mg).  Regulatory
                                 1-2

-------
             TABLE 1-1.   ENVIRONMENTAL AND ECONOMIC  IMPACTS  OF  VARIOUS  ALTERNATIVES  COMPARED
                         TO ALTERNATIVE 1  (BASELINE) IN  THE  FIFTH  YEAR  (1987)
Emission reduction Solid
Reg.
alt.
Mg/yr (tons/yr)
Per-
cent Mg/yr
waste increase Energy increase Cost increase
(tons/yr)
Per-
cent kWh/yr
Per- Capital,
cent $
Annual-
lized, $
Carbon steel
2
3
2
3
760 (850)
1,530 (1,690)
100 (110)
200 (220)
48 760
96 1,530
48 100
96 200
(850)
(1,690)
Specialty
(110)
(220)
1 No increase
2 5.5 xlO6
Steel
1 No increase
2 1.1 xlO6
(130,000)b
1.5 2,660,000
(140,000)b
2.7 1,020,000
(63,000)b
742,000
(52,000)b
212,000
.Costs are reported in March 1981  dollars.
 Represents decrease in costs below baseline.

-------
             TABLE  1-2.   MATRIX  OF  ENVIRONMENTAL AND ECONOMIC
                    IMPACTS  FOR  REGULATORY ALTERNATIVES

Regul atory
alternative
1
2
3

Air
impact
0
+3**
+4**

Water
impact
0
0
0
Solid
waste
impact
0
-1**
- -| **

Energy
impact
0
0
_ 1 **

Noi se
impact
0
0
0

Economic
impact
0
+ 1**
_ i **
Key:   +     Beneficial  impact
            Adverse  impact
      0     No  impact
      1      Negligible  impact
      2     Small  impact
      3     Moderate impact
      4     Large  impact
      *     Short-term  impact
      **    Long-term  impact
      ***   Irreversible  impact
                                    1-4

-------
Alternative 3 would result in a worst-case product price increase of
approximately 0.25 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.  100 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 m 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 and 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

-------
merits 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 Regi ster announcement of the proposed regulation
the public is invited to 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 second volume 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.

-------
2.4  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
     Secti.on 111 of the Act defines a new source as ".  .  .  any stationary
source, the construction or modification of which is commenced ..."
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 energy 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

-------
SINTER
IRON ORE
  COKE OVEN
LIMESTONE
                 SLAG
                                SCRAP
                                            ELECTRIC-AR
                                              FURNACE
                                                 0

                                                               CONTINUOUS CASTING

                                                                           BiLLETS/BLOOMS/SLABS
                                             ARGOII-OXYGEN
                                            DECARBURIZATION
                                               VESSEL
                                                                   INGOT TEEMING
       6
 | |  SOAKING
raTiOE"  PIT
INGOTS
                                                                                      PRIMARY
                                                                                      MILLS
         Figure 3-1.   Flow  diagram  of an iron and  steel plant.1

-------
                                                     STEELMAKING
                                                     FURNACE
                                                  _  BASIC OXYGEN
                                                     OPEN HEARTH
                                                  _  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 .^
                                     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 ADD  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 100 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, ADD 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 322 EAF's in the United States that were operated
by 87 companies in 125 locations (Table 3-1).   These plants are in 36 States
with 62 percent of the plants located in 8 States (Table 3-2).   Of the
322 EAF's approximately 10 percent or 31 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).9,10  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).11  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

-------
TABLE 3-1.   ELECTRIC ARC FURNACES IN THE UNITED  STATES--19819>10>1:
P 1 ant/ location
Al ffCII SPECIALTY STEEL CORP.
Watervliet, N V.

Al 1 EGlltNY LUUtllM S1EEL CORP.
Brackenridye, Pa

AMIRON STEEL AND WIRE D1V.
E I iwanda , Calif.
ARHCO, INC.
Ba 1 t imore Works
Baltimore, Md.
Butler Works
Butler Pa.
Houston Works
Houston, Tex.
• Kansas City Works
Kansas City, Mo.
Marion Works
Mai-ion, Ohio
• Sand Springs Works
Sand Spr i ngs , Ok.
National Supply Division
Torrance , Cal i f .

ATLANTIC STEEL CO.
Atlanta Works
Atlanta, Ga.
Cartersvi 1 le Works
Cartersvi 1 le, Ga.
AUBURN SILEL CO.
Auburn, N.Y.
BABCOCK & WILCOX CO.
Beaver falls, Pa.



BAYOU STEEI CORP.
New Orleans , La.
No of
fur-
naces

2

1
2

2
1

1
1
3

2
4
2
4
2

2
1
1


2
1

1
2
2
2
3

2a
Operating
capacity
(Hg)

29.0

19.9
63 5

72.6
108.8

45.4
36.3
149.7

106 1
158.8
113.4
136. 1
27.2

68.0
9. 1
20.0


77.1
90.7

54.4
22.7
45.4
68.0
90.7

54.4
Plant/location
BETHLEHEM STEEL CORP.
Bethlehem Plant
Bethlehem, Pa.
• Steelton Plant
Steelton, Pa.
• Johnstown Plant
Johnstown, Pa.
• Los Angeles Plant
Los Angeles, Calif.
• Seattle Plant
Seattle, Wash.
BORDER STEEL MILLS, INC.
El Paso, Tex.
BRAEBURN ALLOY STEEL DIV.
Lower Burrel 1 , Pa.
CABOT CORPORATION
Stell ite Div.
Kokomo, Ind.
CALIFORNIA STEEL CO.
Chicago, 111.

CAMULET STEEL CO.
Chicago Heights, 11 1 .

CAMERON IRON WORKS, INC.
Houston, Tex.
CARPENTER TECHNOLOGY CORP.
Steel Division
Bridgeport, Conn.
• Reading Plant
Reading, Pa.
CASCADE ROLLING MILLS
McMinnville, Oreg.
THE CECO CORP.
• Lemont Manufacturing Co.
Lemont, 111.
No. of
fur-
naces

1
1
4
3

2a
2

2


2
2

1
1
1


2

2
2

5 a b
(1)a,b
2

3

Operating
capacity
(Hg)

6.4
25.4
45.4
154.2

163.3
90.7

108.8


22.7
9.1

4 5
13.6
38.1


27.2

54.4
38. 1

13.6
27.2
22.7

36.3

Plant/location
THE CECO CORP. (cont. )
Milton Manufacturing Co.
Milton, Pa.
• Southern Elec. Steel Div
Birmingham, Ala.

CF&I STEEL CORP.
Pueblo, Colo.
CHAPARRAL STEEL CO.
Midlothian, Tex.

COLUMBIA TOOL STEEL CO.
Chicago Heights, 111.

CONNORS STEEL CO.
Birmingham Works
Birmingham, Ala.
• Huntington Works
Huntington W. Va.

COPPERWELD STEEL CO.
Warren, Ohio
CRUCIBLE, INC.
Stainless Steel Div
Midland, Pa.

Specialty Metals Div.
Syracuse , N.Y.

CYCLOPS CORP.
Empire Detroit Steel Oiv
Mansfield, Ohio
• Universal Cyclops
Specialty Steel
Bridgeville, Pa.
EASIERN STAINLESS STEEL CO.
Baltimore, Md.
EDGEWATER STEEL CORP.
Oakmont, Pa.
No. of
fur-
naces

3

2

2

' a,b
(1)

1
1
2
2


5

1
4,
2a
1
1
1
2

2
1


1
1
Operating
capaci ty
(Hg)

18 1

13.6

108.8

104 3
136. 0

4.5
7. 3
40.8
45.4


81.6

40.8
90.7
154.2
13.7
18 2
31.9
90.7

18. 1
36.3


45.4
45.4
                                                                                    (continued)

-------
TABLE  3-1.  (continued)
Plant/location
ELCCTRAILOY CORP.
Oi 1 City, Pa.
FINKLE & SONS CO.
Chicago, 111.
FLORIDA STEEL CORP.
Charlotte Mill
Charlotte, N.C.
• Indiantown Mil 1
Indiantown, Fla.
• Baldwin Mill
Baldwin, Fla.
• Tampa Mi 1 1
1 ampa , Fla.

Jackson Mi 11
Jackson, Tenn.
FORD MOTOR STEEL OIV.
Dearborn, Mich.

GEORGETOWN STEEL CORP.
Georgetown, S.C.
GEORGETOWN TEXAS STEEL CORP.
Beaumont, Texas
HAWAIIAN WESTERN STEEL LTD.
Ewa, Hawaii
INGFRSOL JOHNSON STEEL CO.
New Castle, hid.

1NGERSOL RAND-OIL FIELD PROD.
Pampa, Texas
INLAND STECI CO.
East Chicago, Ind.

INIERCOASTAL STEEL CORP.
Chesapeake, Va.



No. of
fur-
naces

1

2

1
la
1
1
la

1
1
1

la

2


3
2

1
2
2
1
2


2



Operating
capacity
(Mg)

31.8

59.0

22.7
36.3
27.2
31.8
68.0

18.1
27. 2
31 8

120.0

181.4


68.0
99.8

31.8
8.2
10.9
36.3
108.8


18.1



Plant/location
INTERLAKE, INC.
Hoeganaes Corporation
Gallatin, Tenn.
ITT HARPER
Morton Grove, 111.

JESSOP STEEL CO.
• Washington Works
Washington, Pa.
• Green River Steel
Owensboro, Ky.
JONES & LAUGHLIN STEEL CORP.
- Cleveland Works
Cleveland, Ohio
• Pittsburgh Works
Pittsburgh, Pa.
• Warren Works
Warren, Mich.

EARLE M. JORGENSEN CO.
Seattle, Wash.
JOSLYN STAINLESS STEELS
Fort Wayne, Ind.

JUOSON STEEL CORP.
Emeryville, Calif.
KENTUCKY ELECTRIC STEEL CO.
Ashland, Ky.
KEYSTONE CONSOLIDATED
INDUSTRIES, INC.
Keystone Group Steel Works
Peoria, 111.

KNOXVILLE IRON CO.
Knoxville, Tenn.



No. of
fur-
naces

la


1
1

3
2


2

2a
5



2
2
1
1

2a

2


2



Operating
capacity
(Mg)

49.9


4.5
9. 1

18.1
54.4


167.8

317.5
72.6



36.3
15.4
18. 1
40.8

45.4

154.2


31.7



Plant/location
LACLEDE STEEL CO.
Alton, 111.
LONESTAR STEEL CO.
Lone Star, Tex.
LUKENS STEEL
Coatsville, Pa.

MARATHON LE TOURNEAU CO.
Longview, Tex.

MARATHON STEEL CO.
Tempe, Ariz.
McCLOUTH STEEL CORP.
Trenton, Mich.
MISSISSIPPI STEEL DIV.
Flowood Works
Flowood, Miss.

NATIONAL FORGE CO.
• Erie Plant
Erie, Pa.
Irvine Forge Division
Irvine, Pa.
NATIONAL STEEL GREAT LAKES
STEEL DIV
Ecorse, Mich.
NEW JERSEY STEEL 4 STRUCTURE
CORP.
Sayerville, N.J.
NEWPORT STEEL
Newport, Ky.
NORTH STAR STEEL CO.
St. Paul Plant
St Paul Minn

• Monroe Plant
Monroe, Mich.
Wilton Plant
Wilton, Iowa
No. of
fur-
naces

2

2a

2
2
2


3

2
2
1


1
2
I
1

2

2
3
2


la

1

Operating
capaci ty
(Mg)

204. 1

54.4

136.1
163.3
22.7


22.7

181.4
12.7
31.7


31.7
68.0
18. 1
40.8

136.4

59.0
77. 1
54.4


108.8

54.4

                                                               (continued)

-------
                                                      TABLE  3-1.    (continued)
F' lant/ local ion
NORIHWESI SILlt ROI LING
MILLS, INC.
Kent. Wash.
NORIIIWESTERN STEEL & WIRE CO
Sterling, 111.

NUCUR CORP
Uarl i ngton Mill
Darl ington, S.C.
Jewett Mill
JeweLt, lex.
Norfolk Mill
Norfolk, Nebr.

Plymouth Mill
Co Plymouth, Utah
'
03 OREGON STEEL MILLS
Portland, Oreg.

OWENS ELECTRIC STEEL COMPANY
Cayce , S.C.


PENN-OIXIE STEEL CORP.
Kokonio Plant
Kokomo, Ind.
PHOENIX SIEEL CORP.
• Plate Div.
Claymont, OB).

QUANTEX CORP.
MacSteel Uiv.
Jackson, Mich.

RARIIAN RIVER STEEL CO.
Perth Amboy, N. J.

RtPUBUC SfEEL CORP.
Central Alloy Works
Canton, Ohio
Number
of fur-
naces


2

1
2


3a
2a
f'
22'
(l)3'b

2a



2

1
1
1


2

2



2



la


3
4
Operating
capac i ty
(Hg)


31. 7

226.8
362.9


31.7
31.7
36.3
45.1
36.3
36.3
36 3

54.4



68.0

9.1
22.7
31.7


158.8

136. 1



36.3



117.9


77. )
181.4
PI ant/location
REPUBLIC STEEL CORP. (cont. )
• South Chicago Works
South Chicago, 111.
Southern District
Gulf Steel Works
Gadsden, Ala.

ROANOKE EtECTRIC STEEl CORP.
Roanoke, Va.

ROBtIN STEEt CO.
Dunkirk Works
Dunkirk, N.Y.
ROSS STEEt WORKS
Amite, La.


SHARON STEEL CORP.
Sharon, Pa.

SIMONDS STEEL DIV.
Wallace Murray Corp.
Lockport, N.Y.
SOULE STEEL CO.
Carson Works
Carson, Calif.
STANDARD STEEL DIV. OF TIMET
Burnham, Pa.




STRUCTURAL METALS, INC.
Sequin, Tex.


TEtEOYNE VASCO
Latrobe, Pa.

TENNESSEE FORGING STEEt CORP.
• Harriman Works
Harriman, Tenn.
No. of Operating
fur- capacity
naces (Mg)

3 181.4

2 167.8



1 5.4

2 27.2
la 27.2
1 9.7
1 36.3


2a 117.9


3 13.6

2 20.0

16.3
36.3
40.8
63.5


45.4
3 81.6


1 13.6


3 22.7

Plant/location
TENNESSEE FORGING STEEL CORP
• Newport Works
Newport, Ark
TEXAS STEEt CO.
Fort Worth, Tex.



TIMKIN CO.
Steel and Tube Div.
Canton, Ohio
• Latrobe Steel
Latrobe, Pa.

UNION EtECTRIC STEEL CORP
Burgettstown, Pa.

UNITED STATES STEEL CORP.
Fairless Work
Fairless Hills, Pa
- Johnstown-Center Works
Johnstown, Pa
• National Ouquesne Works
Duquesne, Pa.

South Works
South Chicago, 111.
• Texas Works
Baytown, Tex.

WASIIBURN WIRE CORP.
Phillipsdale, R. I.

WASHINGTON STEEL CO.
Fitch Works
Houston Pa.

WITTEMAN STEEL MILES
Fontana, Calif.


to. of
fur-
naces
(cont. )
2


1
1
)
1


1
4
3
1
2


1


2

2
1
1
1
3
1
2
2
2


2


2



1


Operating
capac i ty
(Mg)

22 7


7.3
10.9
22 7
27.2


54.4
90.7
136 1
4 5
27 2


31.8


181.4

4.5
27.2
18. 1
63.5
77. 1
90.7
181.4
181.4
204. 1


36.3


40.8



22.7


^Subject  to existing NSPS.
 Completion date Spring 1982.

-------
     TABLE  3-2.   LOCATION  OF  STEEL  PLANTS  WITH
        EAF'S  AND AOD  VESSELS--198110,xl,13
State
Alabama
Arizona
Arkansas
Cal i form' a
Colorado
Connecticut
Delaware
Florida
Georgia
Hawai i
Illi no is
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


EAF
3
1
1
6
1
1
1
3
2
1
11
5
1
3
2
2
6
1
1
1
1
1
2
5
6
1
2
27
1
3
4
12
1
2
3
1
125
Plants
AOD
vessel





1




1
3



2
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--1981.10,11,13
EAF's
Capacity
range (Mg)
Under 50
50-99.9
100-149.9
150-199.9
200-299.9
Greater than 300

Percentage
51
25
9
13
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

-------
                         TABLE  3-4.   ARGON-OXYGEN  DECARBURIZATION  VESSELS  IN THE UNITED  STATES--!981 1l
CO
I
Plant and location
AL 1ECH SPECIALTY STEEL CORP.
Watervliet, N.Y.
ALLEGHENY LUDLUM STEEL CORP.
Brackenridge, Pa.
ARMCO, INC.
• Bal timore Works
Baltimore, Mel.
• Butler Works
Butler, Pa.
BABCOCK AND WILCOX CO.
Beaver Falls, Pa.
CABOT CORP.
Stellite Div.
Kokomo, Ind.
CARPENTER TECHNOLOGY CORP.
• Steel Div.
Bridgeport, Conn.
• Reading Plant
Reading, Pa.
CRUCIBLE, INC.
• Stainless Steel Oiv.
Midland, Pa.
- Specialty Metals Div.
Syracuse, N.Y.
CYCtOPS CORP.
• Universal Cyclops
Specialty Steel
Bridgeville, Pa.
Vessel
No. of capacity
vessels (Mg)

2 29.0

1 90.7

1 36.3

1 158.8


1 22.7
1 3.6
1 13.6

1 45.4

1 13.6
1 18.1

1 90.7

1 36.3


1 45. 4a


Plant and location
EASTERN STAINLESS STEEL CO.
Baltimore, Md.
ELECTRALLOY CORP.
Oil City, Pa.

INGERSOL JOHNSON STEEL COL.
New Castle, Ind.
JESSOP STEEL CO.
Washington Works
Washington, Pa.
JONES AND LAUGHLIN STEEL CORP.
Warren Works
Warren, Mich.
EARL M. JORGENSEN STEEL CO.
Seattle, Wash.
JOSYLN STAINLESS STEELS
Fort Wayne, Ind.
McLOUTH STEEL CORP.
Trenton, Mich.
REPUBLIC STEEL CORP.
Central Alloy Works
Canton, Ohio
STANDARD STEEL DIV.
Burnharo, Pa.

U.S. STEEL CORP.
South Works
Chicago, 111.
No. of
vessels

1

1
1

1

1


1

1

1
h
or

i


1 (used)
1 (idle)

1

Vessel
capacity
(Mg)

45.4

15.4
27.2

13.6

18.1


72.6

31.8

17.2

90.7

90.7


13.6
18.1

90.7

          Empire-Detroit  Div.
          Mansfield, Ohio
                                                           90.7
WASHINGTON STEEL CO.
  Fitch Works
  Houston, Pa.
36.3
      ^Facility has second vessel  (22.7 Mg)  that can be used in the single trunion ring.
       lias not operated since installation in  1977.

-------
and railroad equipment).12   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 1ighter-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. 12,13
Industry predictions for 1981 call for  94.3  xlO6 Mg (104 xlO6 tons) in
apparent domestic consumption.14  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.12,14  The
projected apparent domestic consumption of steel for 1985 is 108 xlO6 Mg
(119 xlO6 tons).12  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.14   This figure assumes that imported steel  will
capture 15 percent of the market.14  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 steelmaking has been
steady for the past 18 years except for the  recession period of  1974 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).   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

-------
                                           TABLE 3-5.   STEEL  INDUSTRY  DATA AND PROJECTIONS12 >ll+
CO
I
Steel industry data
Raw steel production, 106 Mg
(106 tons)
Apparent domestic consumption, 106 Mg
(106 tons)
Steel mil) products
Industry shipments, 106 Mg
(106 tons)
Value, 106 dollars
Average value, $/Mg
($/ton)
Exports
Quantity, 10G Mg
(10f> tons)
1976
116
(128)
91.7
(101.1)
81
(89)
32,762
404.47
(368.11)

2.4
(2.7)
1977
113
(125)
98.7
(108.8)
83
(91)
36,302
437.37
(398.92)

1.8
(2.0)
1978
124
(137)
105.8
(116.6)
89
(98)
42,828
481.21
(437.02)

2.2
(2.4)
1979
123
(136)
104.3
(115.0)
91
(100)
48,373
531.6
(483.7)

2.5
(2.8)
1980
101
(111)
82.5
(90.9)
74
(82)
40,381
545.7
(492.4)

3.7
(4.1)
1981
lU
(126)
94
(104)
83
(91)
49,358
594.7
(542.4)

2.3
(2.5)
1985a
N/AC
N/A
108
(119)
94
(104)
N/A
N/A
N/A

2.3
(2.5)
1988b
143
(158)
N/A
N/A
106
(117)
N/A
N/A
N/A

N/A
N/A
          Value,  106 dollars
1,255
1,037
1,329
1,878
2,604
1,832
                                                         N/A
          N/A
        Imports

          Quantity,  10" Mg
                  (106 tons)

          Value,  10B dollars
   13
  (14)

4,025
   17
  (19)

5,531
   19
  (21)

6,917
   16
  (18)

6,967
   14
  (15)

6,585
 14.1
(15.5)

7,444
N/A
N/A

N/A
N/A
N/A

N/A
        Total  employment, 103 persons
  469
                                                                      470
                                                                                  472
                                                                                             477
                                                                                                         429
                                                                                                                    472
                                                                     N/A
                                                                   N/A
        . Projections.l2
        ^Projections, M
         N/A = not available.

-------
106  Mg      106  tons

    160
 o
 o
      10
               —I      I      I       j      I      I      I       I


               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.13  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 1985.15  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 1990.16,17  The most optimistic prediction
calls for an even more prolific future with EAF's producing half the
nation's raw steel by 1990.15
     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
will 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 1987.14  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 ADD vessels is expected to be small in the
stainless steel industry since most of the existing steel mills that
produce stainless steel already have ADD vessels.   Any new ADD 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.18
This number would be higher if the use of AOD vessels was increased in
the shops that produce alloy steels and no stainless steels.19
3.2  PROCESS FACILITIES AND THEIR EMISSIONS
3.2.1  Electric Arc Furnace Components
     The electric arc furnace process for steelmaking 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 1906.20
                                 3-15

-------
     The direct electric arc furnace used today is a refractory-1ined,
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 and can  be
lowered and  raised  through holes in the furnace roof.  The electrodes
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
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 to 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.15  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.15   The new UHP furnaces use larger electrodes
                                 3-16

-------
                                  RETRACTABLE
                                 ' ELECTRODES
            REMOVABLE
            FURNACE
            ROOF
DOOR FOR
SLAGGING.^
SAMPLING
& ADDITIONS
   SCRAP
   CHARGE
   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 100-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.1S
     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.15  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 steel foundries
and forging shops.   The bottoms of basic-lined arc furnaces used in the
steel industry consist of  a burned magnesite brick subhearth with a
working surface approximately 30.5 centimeters (cm) (12 inches [in.])
thick of magnesia rammed material.  Basic  arc furnace  roofs are generally
constructed of high-alumina brick with high-alumina rammed or castable
materials for the center section around the electrodes.20,21
     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 lance can be
inserted through the  partially open slag  door or through a separate
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.15,20
     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.22
     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.20
     A large  variety of  scrap is charged to EAF's.   According to the
Institute of  Scrap Iron  and Steel,  all  grades of scrap are to 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.21
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.21,22  Charging emissions are made up of particulate matter,
carbon monoxide, hydrocarbon vapors, and soot.22  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.20  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.20  Melting is accomplished by the heat
supplied by direct radiation from the arcs formed between the electrodes
of the furnace and the metallic charge, 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.20,24  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 to 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
of scrap may be added  to the furnace (in some shops there may 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 bath and gas
temperatures, gas evolution, and generation of particulates.  Oxygen is
now used almost universally  (instead of iron ore or mill  scale) for
"boiling'1 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, manganese,  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 either
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 ferrosi1 icon, 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.21  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.22
     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
ferrosil icon.20  The purposes of the reducing sl-ag 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.20,22  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.22  Fluoride and trace constituents,  such as nickel, hexavalent
chromium, lead, cadmium, and arsenic, are emitted from EAF's.25,26  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.20  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.20,21  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 during meltdown.21
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 Germany, 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 MetalIgewinnung of West Germany to enable the
installation of water-cooled  panels further down the furnace sidewall 15
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, when 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 impact of 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.15
                                 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.27-35  The amount lost as dust varies
from 0.6 to 2.9 percent of the metal charged.36  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.35  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 of electrode consumption is approximately 5 kg/Mg (10 Ib/ton)
of steel produced.36  Many factors affect the rate of electrode consump-
tion, including the operating conditions, the amount of refining performed
in the EAF, and the pressure within the furnace.  Electrode consumption
is a source of hydrocarbon, carbon monoxide, and carbon dioxide emissions.
     3.2.1.3  Energy Considerations.  Unless scrap preheating is employed,
only electrical energy is used in EAF operation.  The largest percentage
of energy is used during the melting and refining of the scrap charge,
and some electricity is used to operate cranes, fans, pumps, and emission
control equipment.   Natural gas may be used to keep the refractories on
idle ladles and furnaces at the desired temperature.
     The amount of energy consumed by the EAF varies from 1 ,647 to
2,614 megajoules per megagram (MJ/Mg) (416 to 660 kilowatt-hours/ton
[kWh/ton]) of steel produced.31-35,37-41  The amount of electricity
consumed depends upon the makeup of the scrap, the size of the furnace,
the final temperature desired, and the quality of the steel produced.
The amount of energy consumed per megagram of ste.el produced will not
change significantly even if scrap preheating is used.   Preheating scrap
is done primarily to raise scrap temperature so that the meltdown time
is decreased.   Preheating also drives off superficial water and burns
oily material  and other combustible contaminants so that fewer emissions
are generated during charging and less slag is formed.
                                 3-25

-------
                                                                                            CLEANED GAS  LOUVERS
CO
I
IX)
CTl
      ELECTRICITY—425 kW TO MELT

                -100 kW TO REFINE

      ELECTRODE CONSUMPTION-^ kg


SCRAP--1075 kg
                                ALLOYS—VARIABLE

                                LIME--38-50 kg
                            CHARGING
                            BUCKET
                                                                 DIRECT SHELL
                                                                 EVACUATION
                                                                (WATER-COOLED)
                                                            /_  COMBUSTION AIR GAP

                                                                STEEL —1000 kg
                                   STEEL LADLE
                                   SLAG
                               OXYGEN--4.1-25 scm
                                                                                                                COLLECTED
                                                                                                              DUST (10-15 kg)
                                               rmo^
, CONTINUOUS CASTER


INGOT MOLDS
                 Figure  3-5.   Material balance  of electric  arc furnace  based  on 1000  kg of  steel  produced.
                                                                                        36

-------
                           TABLE 3-6.  RAW MATERIALS AND PRODUCTS—EAR AND ADD OPERATIONS27-31
                                                      (Factor)
OJ
Steel type
Furnace type
Raw Materials
Metal 1 ics
Additives





TOTAL
Products
Steel





Common
EAF

Scrap (0.93)
(Nos. 1 & 2 bundles, rail-
scrap, butt ends,
shredded, pit scrap, etc.)
Lime (0.05)
Carbon
FeSi
SiMn
HiCFeMn
Other


Steel

(Rebar,
(0.002)
(0.002)
(0.01)
(0.005)
(0.005)
(1-0)

(0.88)

tubing, plates,
Specialty
EAF

AOD

Scrap (0.64) Molten metal
(Stainless steel, chrome-
nickel, low carbon steel,
low phosphorus steel, iron
ore, etc.)
Lime (0.05) Lime, carbon
HiCCr (0.19)
Ni (0.05)
Other (0.07)


- (0.86)
>
aluminum, iron,
18-8, HiCCr,
MoOx, FeSi,
Nickel, MnSi
(0.14)

>
HiCMn, Mn, other
(1-0)

Molten (0.88)
metal

pipe, structural shapes,



By-products

TOTAL
rounds ,


Slag
Dust

etc. )


(0.10)
(0.02)
(1.0)



Slag (0.10)
Dust (0.02)
(1-0)


Steel

(Stainless ,
tool and die
alloy, high
alloys, high
valve, etc.)
Slag
Dust

(1-0)

(0.91)

sil icon,
, low
temp.
speed,

(0.08)
(0.01)
(1-0)
       5Factor designates portion of total in material balance (i.e., raw material -» product + by-product).

-------
     The kilovolt-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.15
3.2.2  Argon-Oxygen  Decarburization Vessels and Their Operation
     The argon-oxygen decarburization process  was first developed by the
Linde Division of  the Union Carbide Corporation at their Niagara 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  in  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 bath chemistry.
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  the 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.31-33  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 steel  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 from 9.9 to
19.6 cubic meters/megagram (m3/Mg) (390 to 770 cubic feet per ton [ft3/ton])
of steel produced, nitrogen from 2.8 to 13.7 m3/Mg (112 to 540 ft3/ton),
and oxygen around 18.8 m3/Mg (700 ft3/ton).31,33,41  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 ADD 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
ADD 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).27,31,33  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
 OJ
 CJ
 s_
 OJ
o
CQ
     0.002
                                                     CO
                             S       12

                         CHROMIUM  (Percent'
                                    16       20
Figure 3-7.   Carbon-chromii
                                        equilibrium  curves.5
                                3-32

-------
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.25,26
     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 ADD 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.42  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.43  In
addition to this, some of the heat from AOO 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 ADD 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.15  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 argon 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.43  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.43
Holding the steel  under low pressure with argon stirring serves to
promote the oxidation of unwanted carbon and to purge the hydrogen gas.
This 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.43  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.22  The majority
of the emissions from EAF's are particulates, both ferrous and nonferrious
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.22  Increases in
electrical power to the furnace and the use of oxygen lancing 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 refining) 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 (e.g., charging) or when the
                                 3-35

-------
primary emission capture device cannot operate (e.g., charging and
tapping) are considered to be fugitive emissions.
     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 the 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 for 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, and the tap
occurs after the carbon content has been greatly reduced.   Table 3-7
presents a summary of the emission factors  for EAF furnaces and AOD
vessels.  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 the 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
AOD vessels is presented in Table  3-10.   The distribution of the particulate
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)1,22,25,26,35

Melt and refine
Charge and tap
3Draft test data,
EAF's
kg/Mg
7.5-20
0.8-1.0
includes charge

Ib/ton
15-40
1.6-2.0
and tap emi
AOD
kg/Mg
8.0
--
ssions.
vessels3
Ib/ton
16.0
--

TABLE 3-8.  TRACE CONSTITUENT EMISSION FACTORS
           (UNCONTROLLED)1,25,26,35
Constituent
Carbon monoxide
Nitrogen oxides
Sulfur oxides
Fl uoride
Chromium
Lead
Nickel
EAF's
kg/Mg Ib/ton
0.26-3.3 0.52-6.5
0.05 0.1
0.005 0.01
--
__
__
__
AOD
kg/Mg
--
--

0.13
0.43
0.019
0.19
vessel s
Ib/ton
--
--

0.27
0.87
0.039
0.38
                     3-37

-------
               TABLE  3-9.   CHEMICAL ANALYSIS OF  ELECTRIC ARC FURNACE DUST BY PHASE OF FURNACE OPERATION21
CO
I
CO
Dust composition (percentage)
Phase Si02 CaO MgO Fe2of
Melting 9.77 3.39 0.45 65.75
Oxidizing 0.76 6.30 0.67 66.00
Oxygen lancing 2.42 3.10 1.83 65.37
Reduction Tr. 35.22 2.72 36.00
A1203 MnO Cr203 S02 P203
0.31 10.15 1.32 2.08 0.60
0.17 5.81 1.32 6.00 0.59
0.14 0.17 0.86 1.84 0.76
0.45 0.70 0.53 7.55 0.55
       The iron content was determined as total  iron and coverted  to  Fe203.

-------
   TABLE  3-10.   EXHAUST  GAS  PARTICIPATE
     MATTER  COMPOSITION21,25,26,31,45
                 (Percent)
Constituent
Fe203
CaO
A1203
Si02
MgO
Mn203
ZnO
NiO
Cr203
CuO
MnO
WO 3
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

ADD1
--
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.
                   3-39

-------
    TABLE 3-11.   SIZE DISTRIBUTION OF PARTICIPATE MATTER
   EMISSIONS FROM STEELMAKING EAF AND ADD FACILITIES25,26
Particle size
range, pm
<0.5
0.5-1.0
1.0-2.5
2.5-5.0
5-10
10-20
20-40
>40
Size
EAF
--
57-73
--
--
8-38
3-8
2-15
0-18
distribution, percent
EAF + ADD
49
8
10
7
6
20b
--
--
by weight
AGO
15
22
29
10
5
19b
--
--
 Particles  smaller than  15  pm  are  considered to be
, respirable or in  the  respirable  categories.
 Larger than 10 pm.
                            3-40

-------
     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 (based on
a 6-minute average).
     Process weight regulations are commonly used to determine regulations
for sources of particulate matter emissions such as ADD vessels.   The
emission level for ADD 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 P0-67 for P ^30
         E = 55 po.n_4o 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 ADD 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.
                                 3-41

-------
TABLE  3-12.   SUMMARY OF  STATE AIR POLLUTION  REGULATIONS
[.PA
region State
1 1 Connecticut

V Illinois


V Indiana
III Maryland

V Michigan
II New York

V Ohio

III Pennsylvania


X Washington

III West Virginia

al:E = 3.59 P"-6X P
E = 17.31 Pu- 1(i F
Particulate regulation
Equation set I
Controlled to extent reasonable
Equation set 2
Equation set 3
No visible emissions
Equation set 3
PM Ł0.03 gr/dscfh
Controlled to extent feasible
Equation set 3
PM Ł0. 15 gr/dscf
PM <0.05 gr/dscf
Equation set 2
Controlled to extent reasonable
Equation set 4
PM §0.04 gr/dscf , exception
No emissions
PM <0. 10 gr/dscf

Table"1
Controlled to extent reasonable
§30
>30
Appl i-
cation
t
c
f
b
c
g
g
c
g
b
f
g
e
b
9
c
g

b
c


where E = particulate emission rate in Ib/hour
P = process weight
Existing source.
Fug i ti ve emissions.
2:E = 2.54 P»-S-T1 F
E = 24.8 P"- IG F
63:E = 4. 10 P°-450
Ł30
>30
n 1-







Emission limit (Ib/h)
for indicated Opacity
vessel size (tons/h) regulation Air pollution
16.7 66.7 (percent) regulation reference
20.6 33.9 <20, exception Sec. 19-508-18, 1980
0 at lot line
11.4 23.9 Rules 202 and 203, 1980
27.0 47.3 §30, exception
0 at lot line
27.0 47.3 §40, exception Regulations APC 3 and 5, 1980
i i COMAR 10. 18. 10, 1980

27.0 47.3 Ł20, exception Part 3, 1980
i i §20 Sec. 216, 1979
i i §20
11.4 23.9 <20, exception Regs. 3745-17-07, 3745-17-08,
Same 1977
11.6 20.9 <20. exception Sees. 123.13, 123..41, 1979
1 i

i i Ł20, exception Regs. WAC 173-400-040,
WAC 173-400-060, 1976
'24.0 34.3 §1 on Ringleman Ad. Reg. 16-20
scale
. PM = particulate concentration in gr/dscf.
.Depends on gas flow rate.
j-One AOD in Michigan has not been operated since 1977.
4:E = 0.76 R°-42
where R = F x W
F = process factor, 40 Ib/ton of product
W = production or charging rate, units/hour
, E = allowable emissions in Ib/hour
PM = 6,000 E for 150,000 
-------
3.4  REFERENCES FOR CHAPTER 3

 1.   Background Information for Standards of Performance:  Electric Arc
     Furnaces in the Steel Industry.  Vol I:  Proposed Standards.  U.S.
     Environmental Protection Agency.  Research Triangle Park, 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.

 3.   New EF Technology Stretches Steelmaking 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.  1:79-89.   1978.

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.   ADD:  The New Common Denominator in Stainless Steel  Making.   World
     Steel Industry Data Handbook/United States.   J_:109-lll.   1978.

12.   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.

13.   Developments in the Iron and Steel Industry.   U.S.  and Canada--1979.
     Iron and Steel Engineer _58(2):D1 to D22.   February  1982.

14.   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.
                                 3-43

-------
15   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
17
18.
     Telecon.   Terry,  B.,  Midwest Research Institute
     Union Carbide Corporation.   September 14, 1981.
     growth projections  for EAF's.
                                                with Brown, J. V
                                                 Information  on
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.

Telecon.   Terry,  B.,  Midwest Research Institute with Sarlitto, R. ,
Union Carbide Corporation.   September 4, 1981.   Information on the
use  of AOD vessels  in the  steel  industry.
19.   Telecon.   Terry,  B.,  Midwest Research Institute
     Union Carbide Corporation.   September 14,  1981
     use of AOD vessels in the steel  industry.
                                                with Sarlitto, R. ,
                                                 Information on  the
20.   The Making,  Shaping and Treating of Steel.   United States Steel,
     Pittsburgh,  Pa.   December 1970.   pp.  403,  551, 553, 574.

21.   Sahagian,  H.  ,  P  F  Fennelly,  and M.  Rei.   Inspection Manual for
     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.

22.   Fennelly,  P.  F.  and P.  D.  Spawn.   Air Pollutant Control Techniques
     for Electric Arc Furnaces in  the Iron and  Steel Foundry Industry
     U.S.  Environmental  Protection Agency.   Research Triangle Park, N.C.
     Publication  No.  EPA 450/2-78-024.   pp.  2-1  through 2-9, 2-16.

23.   Hogan,  W.  T.   Does Direct Reduction Have a  Future?  Iron and Steel
     Engineer   59(2)57-58.   February 1981

24.   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.

25.   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.

26.   CarTech test report (not finalized).

27   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.
                                 3-44

-------
28.   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.

29.   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.

30.   Memo and attachments from Terry, W., Midwest Research Institute, to
     Iversen, R. E.,  EPA/ISB.  May 8, 1981.   Report of source test trip
     to AL Tech Specialty Steel,  Watervliet,  N.Y.

31.   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.

32.   Letter and attachments1from Mauris,  F. C., Allegheny Ludlum Steel
     Corporation,  to Iversen, R.  E.,  EPA/ISB.  September 24, 1980.
     Submittal of  requested  information.

33.   Letter and attachment from Clouse, R. L. , Armco, Incorporated, to
     Iversen, R. E.,  EPA/ISB.  Submittal  of requested information.

34.   Letter and attachments  from Heintz,  J. R., Sharon Steel Corporation,
     to Iversen, R. E. ,  EPA/ISB.   February 2, 1981.  Submittal  of  requested
     information.

35.   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.

36.   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.

37.   Letter and attachments  from Dauksch, W.  E., Nucor Corporation, to
     Iversen, R. E.,  EPA/ISB.  December 26, 1980.  Submittal of requested
     information.

38.   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.

39.   Letter and attachments  from Sussman, V.  H., Ford Motor  Company,  to
     Iversen, R. E. ,  EPA/ISB.  October 24, 1980.  Submittal  of  requested
     information.

40.   Letter and attachments  from Hilton,  A.,  Florida Steel Corporation,
     to Iversen, R. E.,  EPA/ISB.   Submittal of requested information.
                                 3-45

-------
41.   Letter and attachments from Petrella, P., Babcock and Wilcox Tubular
     Products Division,  to Iversen,  R.  E.  , EPA/ISB.   November 24, 1980.
     Submittal  of requested information.

42.   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.

43.   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

44.   Gray, R.D.   Vacuum  Refining and Secondary Steelmaking.   Iron and
     Steel Engineer.   58(12):44-47.   December 1981.

45.   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.
                                3-46

-------
                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 pol1utants--primarily particulate mattei—
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 plant in 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 system or
closed roof shop is the total furnace enclosure (TFE).   Several  TFE's
have been installed in carbon shops in the past 5 years.  These 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 meet 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 AOD 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

-------
FURNACE  ROOF
TAP_
SPOUT
                             A.  PLAN
         ELECTRODES (3)
ELECTRIC ARC FURNACE
                                 DUCT TO
                               /CONTROL DEVICE
                                               REFRACTORY LINED
                                               OR WATER-COOLED
                                               DUCT
                                       z
                           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
         FURNACE
         ROOF
                         A. PLAN
                         nna
                                 .ELECTRODES (3]
TAP
SPOUT
ELECTRIC ARC FURNACE
                       B.  ELEVATION
                                        \
                      SMALL GAP TO
                      FACILITATE
                      ROOF MOVEMEN"
            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

-------
                                        OPERATING FLOOR
         SLAGGING HOOD





                    SLAG POT






Figure 4-3.  Partial furnace  enclosure.





                  4-8

-------
with a new 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 melting/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

-------
                         Roof
                                                  Roof Vent
     j/V/X
                      Roor
                -15'
     Concrete

     floor
                                    rgc Door
                              Main Ł.xricusf Ouc*
                                  -. n
                                   •Mil
                                             cr
               -20'
      noo noor
                     Rear Enclosure Door
                           Slag Pit
                                            L=a'le '
                                                       'Front C'narge Doors
                                                             Addition Cnure
                                                     • i cpping ex
                                                                  s* Oucf
                                   SIDE view
                    Front C'narge Doon


                 Top Cnarge Door —-r
To Control Cevics


Air Curtain Fan—

Al loy Ada'i t ion
Chute r ~)

/^~:



'
"X-








/




"'
Fur n

%,
c
c
/



i^j
ace
'••'
i '
i
_z:




,
— [ — ' — Main Exnausf Duct
\. ''Damper
1

1
1
1
Concrete Floor
s 	 ^- -! !
"udilj
	 [
^Topping Exnaust Duct
Snoo Floor
F3CNT VIEW
Figure 4-4.   Total  furnace  enclosure  at  Lone  Star  Steel  Company.
                                     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-1arge
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 100 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 used, typically
in combinations with other capture technologies discussed in other
sections of this chapter.
     The thermal currents from the hot furnace help the fumes 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

-------
                                     ROOF MONITOR
                                  CANOPY HOOD
-p=»

I—'
GO
                                   JQL
FABRIC FILTER
                                                               WWW
                              Figure  4-5.  Canopy hood  capture system.

-------
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 and 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
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
clearance between the crane  and the roof, or  the roof configuration
itself may not be adaptable  to a canopy installation.   Also, space
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 ducts 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
be a problem is the adjustable or louvered roof monitors described
above.   These monitors can be closed during periods of high emissions
and opened at other times to permit natural ventilation.  The
effectiveness of this type of system may not be as great for multifurnace
                                  4-17

-------
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 particulate removal
efficiency is 99 percent.1   The  emissions capture rate,  however,  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)  ADD 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

-------
I
rv>
o
                                         ROOF MONITOR
                                      CANOPY HOOD
                               DIVERTER
                                STACK
                                                                            FABRIC  FILTER
                                                                      WWW
                                Figure 4-6.   Diverter stack  with  canopy  hood.

-------
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 ADD 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

-------
                                          ROOF MONITOR
r\j
ro
                                       CANOPY HOOD
                             CLOSE-FITTING
                             HOOD
FABRIC FILTER
                                                                       \7WVW
                             Figure 4-7.   Close-fitting hood with canopy  hood.

-------
      TABLE 4-1.  FUGITIVE EMISSIONS CAPTURE TECHNOLOGY COMBINATIONS
                     (CARBON AND SPECIALTY"STEEL EAF)
Fuguti ve
errn ssion
capture
system
combination Capture equipment
Estimated
fugitive
emission ,
reduction ,
percent
               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
u    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 and AOD vessels is the fabric filter.   Only
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
     Fabric filters have many advantages that make them suitable for
control of  EAF and AOD vessel emissions.   Fabric filters use less 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
combi nation
Capture equipment
Estimated
 fugitive
 emission .
reduction,
  percent
                  Single canopy hood, open roof monitor.
                                        75-85
                  Single canopy hood, closed roof (over

                  vessel)/open roof monitor.
                                        85-95
                  Single canopy hood, scavenger duct,

                  cross-draft partitions, closed roof  (over

                  vessel)/open roof monitor.
                                        90-95
                  Single canopy hood, scavenger duct,

                  cross-draft partitions, closed roof.
                                        95-100
 Close-fitting hoods are used for process emissions capture on all
 combi nations.
 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 on 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! impaction,  diffusion,  direct interception, and sieving.
The first three  mechanisms prevail  only briefly during the first few
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 ADD 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 ADD 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.
     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 with 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 T2.7 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
when the sampler is turned off and handled.   EPA prefers a Reference
Method 5 sampling train with the probe inserted into the compartment of
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 EPA Reference
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 participate 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, ADD, 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 to 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

-------
                                   TABLE  4-3.   SUMMARY  OF  CARBON STEEL PLANT DATA
OJ
o
No. furnaces
ducted to
Furnace size ^ ^^

h
Plant Reference Mg (tons) fabric filter Capture equipment"
A 22, 23 45.4, 68.0 (50, 75) 2
B 24-26 36.3, 22.7 (40, 25) 2
C 6, 27 49.9 (55) 1
D 11, 28 108.8 UHP (120) 1
E 29 117.9 (130) 1
F 30 45.4 (50) 3
G 31 108.9 (120) 1
H 32, 33 90.7 (100) 2
I6 33 108.9 (120) 2
J 34, 35 317.5 UHPd (350) 2
K 20 117.9 (130) 1
L 36, 37 90.7 (100) 3
M 12, 38 204.1 (225) 2
N 39, 40 104.3 (115) 1
0 41 , 42 90.7 (100) 2
.Normal power unless noted.
BE = building evacuation CH = single canopy hood
CR = closed roof TFE = total furnace enclosure
TH = tapping hood SCH - segmented canopy hood
SH = slagging hood SDH = side draft hood
CR/OR - closed roof (over furnace)/open roof (elsewhere)
Type of scrap is relative. Clean scrap has low proportion of
BE, CR
SDH, CH, CR
TFE, CH, TH, CR
CH, DEC, CR
CH, DEC, CR/OR
CH, SDH, CR
CH, DEC, CR
CH. DEC, OR
CH, DEC, CR
CH, DEC, TH, SH, CR/OR
SCH, SDH, SD, CR
SCH, DEC, CR
SCH, DEC, CR/OR
CH, DEC, CR/OR
SCH, DEC, LR

Type c
scrap
Unknown
Dirty
Dirty
Clean
Unknown
Unknown
Unknown
Dirty
Unknwon
Dirty
Unknown
Dirty
Dirty
Dirty
Dirty

DEC = direct shell evacuation control
LR = louvered roof monitor
SD - scavenger duct


oil and rust and is usual
making higher quality grades of steel. Facilities making common grades of steel (e.g.,
bars, structural shapes) will use any type of scrap, although
UHP = ultra high power
eOnly the DEC fabric filter was tested.
most plants tend to avoid





ly used in shops
rei nf ore ing
very oily scrap



-------
CAPTURE EQUIPMENT

      REFERENCE
•VdSCf



0.005 -



0.004


C.003




0.002



0.001


ENT
mq/dsci!
-12




|C
1
-10 I
1
!
!
I
1
1
- 3 |
1
1
1
1
!
1! '
i 1 i
- 5 M—i i-O
1 ' 1
M I
1 I 1
y '
i
"
4 " q P '
"up ^ ^ :
^ !j || w i
1 1 ' 1 Łn 1 '
! : -I 'o ^ fl I
00 fr 'l ,LL !
!! § - ;
L1 1
Sc CH SD TFE, DEC, DEC, SDH, DEC


1
KEY
0 TEST RUN
i — i AVERAGE







-i O









, DEC ,
                    22
                        24
                            25
                                CH   CH   CH    CH   CH   CH
                                27   23   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 on the
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 suction-type
fabric filters with stacks,  while Plants G, H, I, J, K, L, M, N, and 0
have positive-pressure fabric filters with monovent exhausts.  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
zero 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, and K represent best demonstrated technology (BDT).
Plants L, M, N, and 0 were visited primarily to discuss the worker heat
stress problem at steel facilities located in  the Southern United States.
Although some visible emission data were obtained, they are not being
reported here nor used in the data base.  The  fabric filters at these
plants are not considered BDT because of plant age, level  of emission
limit required, degree of maintenance, and design considerations.43
     Table 4-5 presents visible emission data  obtained from the shop
roof monitors at  Plants C, G, H, I, J, and K.   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 monitor at Plant J is
closed over the furnace and  open elsewhere.
     Visible emission data have also been obtained from the shop roof
monitors at Plants B, L, M,  N, and 0.  The overall emissions capture
technology at these plants do not represent BDT, and the  data obtained
are not used 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, N, and 0 were not visited as
examples of BDT.   In addition to the reasons noted earlier, construction
                                  4-32

-------
TABLE 4-4.  SUMMARY OF VISIBLE EMISSION DATA FROM FABRIC
FILTERS ON EAF'S AT CARBON STEEL SHOPS15,25,27,29-34,44
Plant
Ba,b
cb,d
cb,d
Fb
Qa,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
0
0
of
0
            From continuous monitor; not
           .Reference Method 9.
            Single stack exhaust.
           CNot 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



Plant
ca
(NSPS)

ca,b
(NSPS)
GC
(NSPS)

Hd




Furnace/
vessel
EAF


EAF

EAF


EAF




Furnace
process
Charge
Melt
Tap
Heat cycle

Charge
Melt
Tap
Charge
Melt
Tap

No. of
6 -mi nute
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
of
6~mi nute
averages ,
percent
0.5
0. 1
0. 1
0.25


0

5.0
0
23.0
             EAF
Heat cycle
10
Je EAF
(NSPS)

Kf EAF
(NSPS)

Charge
Melt
Tap
Charge
Melt
Tap
15
106
10
5
90
5
4.4
4.2
5.0
0
0
0
1.0
0.2
0.6
0
0
0
.Utilizes  total  furnace  enclosure,  canopy  hood,  closed roof.
^Second test at  Plant  C.
dUtilizes  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
                                4-34

-------
of a large hole in the wall that permitted excessive cross-drafts was
underway at Plant N.
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 ADD 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.
     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
                                  4-35

-------
TABLE 4-6.   SUMMARY OF SPECIALTY STEEL PLANT DATA
Plant
P
Q
Q
R
S
3CFH =
CH =
CR =
SCH =
SD =
Reference
45-47
13, 21,
48, 49
50
50
51-53
close-fitting hood
single canopy hood
closed roof
No. furnaces/
Furnace/vessel vessels ducted
type and size, to tested Capture
Mg (tons) fabric filter equipment
EAF:
ADD:
ADD;
ADD:
EAF:
ADD:
EAF:
ADD:

29.0 (32) 2
29.0 (32) 2
18.1 (20) 1
18.1 (20) 1
38.1 (42) 2
45.4 (50) 1
45.4 (50) 1
45.4 (50) 1

SCH, CR
SCH, CR
SCH, SD, CR
SCH, SD, CR
CH, CR
CFH, CH, CR
CH, CR
CFH, CH, CR

segmented canopy hood
scavenger duct
                       4-36

-------
            gr/dscf   mg/dscm

                    - 3
              0.003  -
                    . 6
0Ł
I—
•z.
LlJ
O
•z.
O


UJ
         a:
         cC
         Q_
              0.002  -
             0.001 .
                                             KEY


                                             TEST RUN

                                             AVERAGE
                    . 4
                 b
                 i i
                 O
                    .2
              PLANT
       FURNACE TYPE
          REFERENCE
                 P     0     Q      R
              EAF,AOD  ADD    AOD  EAF,AOD
                45    21    48     50
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 SHOPS21,45,51


Plant
P
Q
S
.One fabric
ADD vessel
CEAF fabric


Length of
observations
in minutes
438a
408b
336h
318b
filter for EAF's
fabric filter.
filter.
Maximum VE1 s
based on
6-minute
average ,
percent
0
0
0
0
and AOD vessels.
                           4-38

-------
             TABLE 4-8.   SUMMARY OF VISIBLE EMISSION DATA FROM
             SHOP ROOF MONITORS ON SPECIALTY STEEL SHOPS21,51



Plant
Qa
sb




Furnace/
vessel
ADD
EAF
and
ADD


Furnace
process
Heat cycle
Heat cycle



No. of
6-minute
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

-------
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
col lection.
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 FACILITIES21,24,45,54



Plant
Ba
Ca
Pb
Qd
Qd

Length of
observations
in minutes
12
20
594
60
48
Maximum
6-minute
average,
percent
0.6
5.0
Oc
0C
7.3e
         aEAF fabric filter.
         °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.
         aAOD 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
p45

Fabric filter
exhaust gas
31 ,600
(495, 1,090)
17,400°
(277, 594)
5,800C
(94, 203)
7,600C
(119, 267)

Fabric
filter
catch
47,200
39,200
8,400
16,300
Q21

Fabric filter
exhaust gas
10,300
(15, 29)
7,600C
(11, 21)
<2,300C'd
(<3, <8)
6,800C
(7, 21)

Fabric
filter
catch
15,200
49,200
2,200
21 ,500
 Assumes emission  concentration  and  flow rate  of tested plant and 4,950
.operating hours  per year.
 Assumes 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, B., Midwest Research Institute, with Eckstein, G. ,
     Bethlehem Steel Corporation.  September 3, 1981.  Information on
     Bethlehem Steel-Johnstown facility.

 5.  Telecon.   Terry B., Midwest Research Institute, with Woolen, C.,
     Pennsylvania Engineering Corporation.  August 19, 1981.   Information
     on total  furnace enclosure systems.

 6.  Memo and attachments from Terry, B., 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 B., 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. L  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.
     Cincinnatti, Ohio.   Publication  No.  DHHS.   No.  80-132.

17.   Telecon.  Terry  W.,  Midwest Research Institute, with Sarlitto, R.,
     Union Carbide Corporation.   September 4, 1981.   Information about
     ADD utilization.

18.   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.

19.   Goodfellow, H.  D.   Solving Air Pollution Problems in the Metallurgical
     Industry   Seventh  International Clean Air Congress.  Adelaide,
     Australia.   August  24-28,  1981.   14 p.

20.   Electric Arc Furnace  Baghouse Compliance Test:   Sharon Steel Corporation,
     Parrel 1, Pennsylvania,  January 6,  7, and 8, 1981.   WFI Sciences
     Company   Pittsburgh,  Pa.   WFI Science Report No.  8343.

21   Emission Test Report:   Carpenter Technology Corporation, Reading,
     Pennsylvania.   PEDCo  Environmental,  Inc.   Cincinnati, Ohio.   Contract
     No.  68-02-3546,  Work  Assignment  No.  2.   July 1981.

22.   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.

23.   Memo from Terry,  W.,  Midwest  Research Institute,  to Iversen, R. ,
     EPA/ISB.  October 31,  1980.   Site  visit report--Babcock and Wilcox,
     Beaver Falls,  Pennsylvania.

24.   Compliance Tests Under New Source  Performance Standards:   Florida
     Steel Corporation,  Charlotte,  North Carolina.  Sholtes & Koogler,
     Environmental  Consultants.   Gainesville, Florida.   January 1980.

25.   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.   Memo from Terry, W.,  Midwest Research Institute, to Iversen, R,
     EPA/ISB.  September 3, 1980.  Site visit report—Florida Steel
     Corporation, Charlotte, North Carolina.

27.   Source Sampling Report—Hoeganaes, Inc.  (83-00129-01), Gallatin,
     Tennessee, Baghouse Serving the Electric Arc Furnace.   Environmental
     Management, Planning and Engineering, Nashville, Tennessee.
     October 1981.

28.   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.

29.   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.

30.   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.

31.   Compliance Tests Under New Source Performance Standards:   Florida Steel
     Corporation, Tennessee Mill Division, Jackson,  Tennessee.   Sholtes and
     Koogler.  Gainesville, Florida.  December 1981.

32.   Particulate Emission Tests for Lukens Steel Electric Melt Shop.
     Fuller Company.  Catasauqua, Pennsylvania.  September  23,  1973.

33.   Memo and Attachment from Terry, W., Midwest 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).

34.   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.

35.   Memo from Banker, L. , Midwest Research Institute,  to Iversen, R.,
     EPA/ISB.  July 28, 19&0.  Source emission test observation report--
     Jones and Laughlin Steel Corporation, Pittsburgh,  Pa.

36..  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.

37.   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.   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.

46.   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

47.   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

48.   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.

49.   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

-------
50.   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.

51.   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.

52.   Memo from Banker,  L. ,  Midwest Research Institute, to Iversen, R.,
     EPA/ISB.   October 17,  1980.  Site visit  report—Eastern Stainless
     Steel Company,  Baltimore, Maryland.

53.   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.

54.   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 or 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 ADD
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 Mg/h (tons/h)  because of the variability in the grades
of steel produced and the amount of  refining done in the furnace, which
                                 6-1

-------
                         TABLE 6-1    MODEL PLANTS
                                                     Steel production
Number             Plant description              Mg/yr(tons/yr)


  la         A new shop with  a                    37,420        (41,250)
               22.7-Mg (25-ton)  EAF.

  2a         A new shop with  a                   163,300       (180,000)
               90.7-Mg (100-ton) EAF

  3a         A new shop with  a                   596,000       (657,000)
               272.2-Mg (300-ton) EAF.

  4a         A new shop with  a                   434,800       (479,300)
               136.1-Mg (150-ton) UHP EAF.

  5a         A modification  of a                 434,800       (479,300)
               136.1-Mg (150-ton) normal
               power EAF to  a UHP EAF.

  6a         A retrofit of a                      596,000       (657,000)
               272.2-Mg (300-ton) EAF
               into an existing  shop.

  7a'b       A retrofit of a                       37,420        (41,250)
               22.7-Mg (25-ton)  EAF into
               an existing shop.

  8b         A new shop with  a                    37,420        (41,250)
               22.7-Mg (25-ton)  ADD vessel
               and a 22.7-Mg  (25-ton) EAF.

  9b         A new shop with  a                   163,300       (180,000)
               90.7-Mg (100-ton) AOD vessel
               and a 90.7-Mg  (100-ton)  EAF.

.Carbon steel shop.
 Specialty steel  shop.
                                    6-2

-------
affect 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, respec-
tively.  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 ADD vessels 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 allow more power input and
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
because in the past, EAF's were used for both melting and refining in
specialty steel   shops.   Industry sources indicate that most specialty
                                 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
day/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
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
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
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
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
The daily production  averages  about  6  heats  (furnaces  A,  B,  and C) and
9 heats (furnace D)  because  of process  delays,  refractory gunning, and
furnace maintenance  that is  performed  between  heats.
                                   6-4

-------
          TABLE 6-3.  MODEL FURNACE PARAMETERS—SPECIALTY STEEL1
Process information
Furnace type
Furnace capacity
Mg/heat
tons/heat
Furnace transformer rating
Heat time, h
Production rate
Mg/day
tons/day
heats/day3
Operating time
h/day
day/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
Furnace/Vessel ID
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
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
E
ADD

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
F
ADD

90.7
100
--
1.5

544
600
6

24
300
7,200

163.3
180.0
3
--
3
5

1620
2950
--

—
—

33.5
no

484
1,067
aFurnaces A and B have an average production of 6 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 6 heats per day.
                                    6-5

-------
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 ADD 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 ADD 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 ADD
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
ADD 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 are
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 wel1-control 1ed 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.
     In general, 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
                                 6-6

-------
        TABLE 6-4.   MODEL PLANTS WITH CAPTURE CONFIGURATION OPTIONS
                                             Capture configurations'
Model plant
designation
Model
plant
No.
EAF:
DEC,
CH.OR
ADD:
CFH,
CH,OR
EAF:
OEC,SCH,
(or CH,
TH,SH) ,
CR/OR
AOO:
CFH,
CH,
CR/OR
EAF:
3EC,SCH,
(or CH,
TH.SH) ,
SD.CR
AOO:
CFH,CH,
SO, COP,
CR
EAF:
TFE.CH
TH.CR
Carbon Steel Shop
25-ton EAF
100-ton EAF
300-ton EAF
150-ton UHP  EAF
150-ton normal  EAF
  modification to UHP EAF
300-ton retrofit
25-ton retrofit
Specialty Steel Shop
25-ton EAF and ADD
  Ducted together
  Ducted separately
100-ton EAF and ADD
  Ducted together
  Ducted separately
1
2
3
4
5
6
7
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
   CDP = 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
type
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
AOD
AOD
AOD
AOD
A
A
B
B
C
C
D
D
E
E
F
F
Furnace ID
(plant type)
(carbon,
(carbon,
(carbon,
(carbon,
(carbon)
(carbon)
(carbon)
(carbon)
specialty)
specialty)
specialty)
specialty)




(specialty)
(special
(special
(special
ty)
ty)
ty)
Regulatory

Air fl
alternative0 (m3/s)/Mg"'
1,
3
1,
3
1,
3
1,
3
1,
3
1,
3
2

2

2

2

2

2

2.
2.
2.
2.
0.
0.
2.
2.
2.
2.
2.
2.
00
10
00
10
80
91
07
18
43
64
43
64
ow
~i
acfm/tonu
3
4
3
4
1
1
4
4
4
5
4
5
,850
,050
,850
,050
,550
,750
,000
,200
,700
,100
,700
,100
     Tables 6-10 and 6-11.
ucf\t 121°C (250°F).
cj(m3/s)/Mg--cubic meters per second per megagram.
 acfm/ton--actual cubic feet per minute per ton.

-------
              TABLE  6-6.   MODEL FURNACE PARAMETERS—CARBON STEEL  FABRIC FILTER  INFORMATION1
                                                    (Metric units)
Furnace type .
Capture equipment
Regulatory alternative
Air flow rate, nr'Vs
Total
Canopy
DEC
TFE
Scavenger
Temperature, °C
No. compartments, total
No. bags, total
,x, Ba?J type
Net air-to-cloth ratio
Gross air-to-cloth ratio
No. fans
Fan wattage, each
Net cloth area, m2
Gross cloth area, m2
Monitor height, m
Moisture, percent
Pressure drop, cm
A
1, 2
1 , 2

45.4
31.3
14. 1
--
--
121
8
384
3. 18: 1
2.79: 1
2
96,980
2,826.0
3,210.6
22.9
1
7.6-12.7
A
3
3

47.7
28.9
14.1
--
4.7
121
7
420
3.13:1
2.68: 1
2
104,440
3,010.0
3,511.6
22.9
1
7.6-12.7
A
4
3

34.2
16.5
17.7
--
121
12
288
3.05: I
2.80: 1
2
85,190
2,210.3
2,411.2
22.9
1
7.6-12.7
B
1, 2
1, 2

181.7
125. 1
56.6
--
--
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
B
3
3

191.1
115.6
56.6
--
18.9
121
17
1,530
3.13: 1
2.94: 1
5
171,580 1
12,039.8 13
12,792.3 14
22.9
1
7.6-12.7 7.
C
1, 2
1 , 2

219.4
186.5
32.9
--
--
121
24
1,728
3. 12:1
2.99: 1
5
79,040
,845.8
,447.8
• 30.5
1
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
D
1, 2
1, 2

283.1
192.5
90.6
--
--
121
23
2,346
3.02:1
2.84: 1
6
179,040
18,461. 1
19,6)4.9
30.5
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.
 I  = DEC, canopy hood,  open roof.
 2  = DEC, segmented canopy hood, open/closed roof.
 3  = DEC, segmented canopy hood, scavenger duct, closed  roof.
 4  = TFE, canopy hood,  closed roof.
^From Table 6-10.
 At temperature noted.
;0ne compartment off-line  for cleaning.
 Fabric  filter monitor  height above  grade.

-------
                TABLE  6-7.   MODEL  FURNACE PARAMETERS—CARBON STEEL  FABRIC  FILTER  INFORMATION1
                                                     (English units)
r a
r urnace type
Capture equipment
Retaliatory alternative
Air flow rate, acfm
Total
Canopy
DEC
TFE
Scavenger
Temperature, °F
No. compartments, total
No bags, total
^ Bag type
j_[j 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.
A
1 , 2
1 , 2

96,250
66,250
30,000
--
--
250
8
384
3. 18: 1
2. 79: 1
2
130
30,210
31,560
75
1
3-5
A
3
3

101 ,250
61 ,250
30,000
--
10,000
250
7
120
3. 13: 1
•2.68: 1
2
110
32,100
37,800
75
I
3-5
A
1
3

72,500
35,000
--
37,500
--
250
12
288
3.05: 1
2.80: 1
2
115
23,760
25,920
75
1
3-5
B
1 , 2
1 , 2

385,000
265,000
120,000
__
--
250
20
1 ,110
3.13:1
2.97: I
5
230
123,120
129,600
75
1
3-5
B
3
3

105,000
215,000
120,000
--
10,000
250
)7
I ,530
3. 13: 1
2.91: I
5
230
129,600
137,700
75
1
3-5
C
1 , 2
1, 2

465,000
395,250
69,750
--
--
250
21
1,728
3. 12: 1
2.99: 1
5
210
119,010
155,520
100
1
3-5
C
3
3

525,000
395,250
69,750
--
60,000
250
15
1 ,980
3. 16: 1
2.95: 1
6
210
166,320
178,200
100
1
3-5
D
1 , 2
1 , 2

600,000
408,000
192,000
--
--
250
23
2,316
3.02: I
2.81: I
6
210
198,720
211 , 140
100
1
3-5
D
3
3

630,000
378,000
192,000
--
60,000
250
23
2,346
3. 12: 1
2.98: 1
6
250
201 ,960
211,140
100
1
3-5
 From Table 6-2.
 1  =  DEC, canopy hood, open  roof.
 2  =  DEC, segmented canopy hood, open/closed roof.
 3  =  DEC, segmented canopy hood, scavenger duct,  closed roof.
 4  =  IFE, canopy hood, closed roof.
^f-rom Table 6-10.
 At temperature noted.
J)ne  compartment off-line for cleaning.
 fabric filter' monitor height above  grade.

-------
             TABLE 6-8.   MODEL FURNACE PARAMETERS—SPECIALTY STEEL FABRIC FILTER  INFORMATION1
                                                   (Metric  units)
Furnace type .
Capture equipment
Regulatory alternative
Air flow rate, nrVs
lota)
Canopy
DEC
Scavenger
Close-fitting hood
Temperature, °C
No. compartments, total
No. bags, total
Net air-to-cloth ratio
Gross air-to-cloth ratio
No. fans
Fan wattage, each
Net cloth area,6 in2
Gross cloth area, m2
Monitor height, m
Moisture, percent
Pressure drop, cm
['From Table 6-3.
A
1, 2
1, 2

45.4
31.3
14. 1
--
--
121
8
384
3. 18: 1
2.79:1
2
96,980
2,813.,0
3,214.9
22.9
1
7.6-12.7

A
3
3

47.7
28.9
14. 1
4.7
--
121
7
420
3.13:1
2.68:1
2
104,440
3,014.0
3,516.3
22.9
1
7.6-12.7

B
1, 2
1. 2

181.7
125. 1
56.6
--
--
121
20
1,440
3.13:1
2.97:1
5
171,580
11,453.2
12,056.0
22.9
1
7.6-12.7

B
3
3

191.1
115.6
56.6
18.9
--
12)
17
1,530
3.13:1
2.94:1
5
171,580
12,056.0
12,809.5
22.9
1
7.6-12.7

i
1, 5
1, 2

55.4
30.7
—
--
24.7
121
8
512
2.89: 1
2.57:1
3
104,440
3,781.5
4,253. 1
22.9
1
7.6-12.7

E
6
3

60. 1
30.7
—
4.7
24.7
121
8
576

2.46:1
3
111,900
4,219.6
4,822.4
22.9
1
7.6-12.7

F
4, 5
1, 2

221.8
122.7
—
--
99.1
121
16
1 ,920
2.89: 1
2.72:1
6
179,040
15,128.7
16,074.7
22.9
1
7.6-12.7

F
6
3

240.7
122.7
--
18.9
99.1
121
16
2,112
2.86: 1
2.68: 1
6
186,500
16,577.0
17,682.2
22.9
1
7.6-12.7

A, E
1,2,1,
I , 2

100.7
61.9
14.1
--
24.7
121
10
880
3.01: 1
2.71:1
4
156,660
6,605.7
7,337.3
22.9
1
7.6-12.7

A, t
5 3, 6
3

107.8
59.6
14.1
9.4
24. 7
121
15
900
3.03: 1
2.82: 1
4
164,120
7,023.4
7,546.2
22.9
1
7.6-12.7

B, F
1, 2, 4, 5
1, 2

403.4
247.7
56.6
—
99. 1
121
17
3,332
3.00:1
2.85:1
6
343,160
26,512.1
27,907.5
22.9
, 1
7.6-12.7

B. f
3, 6
3

431.8
238.3
56.6
37.8
99.1
121
18
3,582
3.02: 1
2.85: 1
6
350,620
28,184.7
29,970.8
22.9
1
7.6-12.7

;AL temperature noted.
K0ne compartment off-line for cleaning.
 Fabric filter monitor height above grad

-------
                         TABLE  6-9.    MODEL FURNACE  PARAMETERS—SPECIALTY  STEEL  FABRIC  FILTER  INFORMATION1
                                                                      (English units)
        f urn,ice  type
        C.ipl ure  tM|ii ipmunt
        HCMJU I a t ory a I terna t

        Air fluw rate,
          loldl
          Canopy
          OtC
          bt avenger
          Close-f i ttIng

        1cmperaIure, °F

        Nn compartments,

        Nu bagb, lota

        B.HJ type
        Nn.  fans

        Kin  horsepower

_*^      Net  cloth area,

^      Cross cloth area,

        Moni tor height,

        Mo i sture,  percent

        Pressure drop,
emat i ve
d
ac fin




3 hood

Ls, total
1
_h ratio
loth ratio

each
e a*
•a. ft*
f ft
:nt
i n.
A
1 , 2
1 . 2

96.250
66,250
30,000
--
--
250
8
384
3. 18: 1
2 79: 1
2
130
30,210
34.560
75
I
3-5
A
3
3

I0l ,250
61 .250
30,000
10,000
--
250
7
420
3. 13: 1
2.68: 1
2
140
32.400
37,800
75
1
3-5
B
1 , 2
1 . 2

385,000
265,000
120,000
--
--
250
20
1 ,440
3 13: 1
2.97: 1
5
230
123,120
129,600
75
1
3-5
B
3
3

105,000
215,000
120,000
40,000
--
250
17
1 ,530
3. 13: 1
2 94: 1
5
230
129,600
137,700
75
1
3-5
E
4, 5
1 , 2

1 17,500
65,000
--
--
52,500
250
8
512
2.89: 1
2.57: 1
3
140
40,650
45,720
75
1
3-5
i
6
3

127,500
65,000
--
10,000
52,500
250
8
576
2.81. 1
2.46: 1
3
150
45,360
51 ,840
75
1
3-5
F
4, 5
1, 2

470,000
260,000
--
--
210,000
250
16
1 ,920
2.89: 1
2.72: 1
6
240
162,630
172,800
75
1
3-5
1
6
3

510,000
260,000
--
40,000
210,000
250
16
2,112
2.86: 1
2.68:1
6
250
178,200
190,080
75
1
3-5
A, t
1 , 2, 4, 5
1 , 2

213,750
131 ,250
30,000
--
52,500
250
10
880
3.01: 1
2. 71: 1
4
210
71 ,010
78,875
75
1
3-5
A, t
3, 6
1

228,750
126,250
30,000
20,000
52,500
250
15
900
3.03: 1
2.82: 1
4
220
75,500
81 ,120
75
1
3-5
B, f
1 , 2, 4, 5
1 , 2

855,000
525,000
120,000
--
210,000
250
17
3,332
3.00: 1
2.85: 1
6
460
285,000
300,000
»
1
3-5
B, I
3, I
i

915,000
505,000
120,000
80,000
210,000
250
18
3,582
3.02: 1
2.84: 1
6
470
302.980
322,180
75
1
3-5
        1 From 1able  6-3.
         I - DEC,  canopy hood,  open roof.
         2 - UEC,  segmented canopy hood,  open/closed roof.
         3 - DEC,  segmented canopy hood,  scavenger duct, closed roof.
         4 - close-fitting hood, canopy hood, open roof.
         5 = close-fitting hood, canopy hood, open/closed roof.
         Ci = clobe-fitting hood, canopy hood, scavenger duct,  closed roof.
        C| rom ladles 6-10 and 6-11.
         At temperature noted.
        ,0ne compartment off-line  for cleaning.
         fabric filter monitor  height above grade.

-------
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 ADD 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 ADD vessel has not been classified.   However, because
the ADD vessel fabric filter dust also contains lead, cadmium, and
hexavalent chromium,  it may be assumed that ADD fabric filter dust will
eventually be classified as a hazardous waste.   In many cases, the dusts
from the EAF and ADD 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 ADD
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
developed for EAF's and AOD vessels and are described below.  Environ-
                                 6-13

-------
mental,  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 ADD 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.
     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.
                                 6-14

-------
   TABLE 6-10.   REGULATORY ALTERNATIVES—CARBON AND SPECIALTY STEEL EAF


Reg.
alt.


Point of
eim ssions
Estimated
fugitive
emi ssi on
reduction,3
percent


Fugitive
emissions capture



Q
equipment
          Shop roof
75-85
Single canopy hood,  open roof
monitor; enclosed dust-handling
equipment.
          Shop roof
85-95
Segmented canopy hood (or single
canopy hood with separate
tapping and slagging hoods),
closed roof (over furnace)/open
roof monitor elsewhere;  enclosed
dust-handling equipment.
          Shop roof
95-100
Segmented canopy hood (or single
canopy hood with separate
tapping and slagging hoods),
scavenger duct,  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.
                                   6-15

-------
      TABLE 6-11    REGULATORY ALTERNATIVES — SPECIALTY STEEL AOD VESSEL
Reg.
alt.
Point of
emi ssions
Estimated
fugitive
emission
reduction,'
 percent
          Fugitive
emissions capture equipment
 1
Shop roof
  75-85
Single canopy hood, open roof
monitor.
          Shop roof
                  85-95
                Single canopy hood, closed roof
                (over vessel)/open roof monitor
                elsewhere; enclosed dust-
                handling 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.
 Close-fitting hood used  for process  emissions  capture on all  alternatives
 Fabric filters are used  to control  all  captured emissions.
                                   6-16

-------
6.4  REFERENCES FOR CHAPTER 6

 1.  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, B., 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, B., Midwest Research Institute, with Askins, C.,
     Babcock and Wilcox.  September 9, 1981.   Information  about Babcock
     and Wilcox EAF/AOD facilities.

 6.  Telecon.   Terry, B., 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, B., 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.
                                 6-17

-------
                         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.
These 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

-------
                TABLE 7-1.  ANNUAL PARTICULATE MATTER EMISSIONS FOR EACH MODEL FURNACE
                                      AND REGULATORY ALTERNATIVE1
                                               Mg (tons)
Furnace/vessel •
Model furnace:
Furnace capacity: Mg
(tons)
Reg. Alt. 0
(Uncontrol led)
Reg. Alt. 1
(Basel ine)
Reg. Alt. 2
Reg. Alt. 3
EAF
A
22.7
(25)
561.3
(618.8)
12.8
(14.1)
6.7
(7.4)
0.5
(0.6)
B
90.7
(100)
2,499.4
(2,700.0)
55.8
(61.5)
29.1
(32.1)
2.4
(2.7)
C
272.2
(300)
8,940.5
(9,855.0)
203.6
(224.4)
106.4
(117.3)
9.0
(9.9)
D
136.1
(150)
6,522.3
(7,189.5)
148.5
(163.7)
77.6
(85.6)
6.5
(7.2)
E
22.7
(25)
299.3
(330.0)
6.7
(7.4)
3.5
(3.9)
0.3
(0.3)
ADD
F
90.7
(100)
1,306.4
(1,440.0)
29.4
(32.4)
15.5
(17.1)
1.3
(1.4)
Refer to Chapter 6 for explanation of model furnace and regulatory alternative designation.

-------
t
co
                      TABLE 7-2.  ANNUAL PARTICIPATE EMISSION REDUCTION BELOW UNCONTROLLED AND
                                 BASELINE LEVELS FOR EACH MODEL FURNACE/VESSEL SIZE
                                                      Mg (tons)
Furnace/vessel :
Model furnace:
Furnace capacity: Mg
(tons)
Incremental
reduction between
alternatives
0 and 1
0 and 2
0 and 3
1 and 2
1 and 3
EAF
A
22.7
(25)
548.5
(604.7)
554.6
(611.4)
560.8
(618.2)
6.1
(6.7)
12.3
(13.5)
B
90.7
(100)
2,393.6
(2,638.5)
2,420.3
(2,667.9)
2,447.0
(2,697.3)
26.7
(29.4)
53.4
(58.8)
C
272.2
(300)
8,736.9
(9,630.6)
8,834.1
(9,737.7)
8,931.5
(9,845.1)
97.2
(107.1)
194.6
(214.5)
D
136.1
(150)
6,373.8
(7,025.8)
6,444.7
(7,103.9)
6,515.8
(7,182.3)
70.9
(78.1)
142.0
(156.5)
AOD
E
22.7
(25)
292.6
(322.6)
295.8
(326.1)
299.0
(329.7)
3.2
(3.5)
6.4
(7.1)
vessel
F
90.7
(100)
1,277.0
(1,407.6)
1,290.9
(1,422.9)
1,305.1
(1,438.6)
13.9
(15.3)
28.1
(31.0)

-------
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 5-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 AOD 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 1983
Carbon steel
90.7-Mg (100-ton) EAF's 2
136.1-Mg (150-ton) EAF's 1
272.2-Mg (300-ton) EAF's 0
Specialty steel
22.7-Mg (25-ton) EAF/AOD 0
90.7-Mg (100-ton) EAF/AOD 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

-------
TABLE 7-4.   TOTAL PARTICULATE MATTER EMISSIONS FOR
NEW SHOPS ADDED IN 1983-1987 AND 5-YEAR CUMULATIVE
                  Mg/yr (tons/yr)
Regulatory Alternative
Carbon steel shops
Uncontrol led
1
2
3
Specialty steel shops
Uncontrol led
1
2
3
1983
11,472.0
(12,645.0)
261.3
(288.0)
136.5
(150.5)
11.4
(12.6)
3,798.0
(4,186.0)
86.1
(94.9)
45.2
(49.8)
3.8
(4.2)
1984
15,595.0
17,190.0)
355.2
(391.5)
185.6
(204.6)
15.6
(17.2)
805.4
(887.8)
18.2
(20.1)
9.6
(10.6)
0.8
(0.9)
1985
13,907.0
(15,330.0)
316.7
(349.1)
165.6
(182.5)
13.9
(15.3)
3,798.0
(4,186.0)
86.1
(94.9)
45.2
(49.8)
3.8
(4.2)
1986
15,595.0
(17,190.0)
355.2
(391.5)
185.6
(204.6)
15.6
(17.2)
805.4
(887.8)
18.2
(20.1)
9.6
(10.6)
0.8
(0.9)
1987
13,907.0
(15,330.0)
316.7
(349.1)
165.6
(182.5)
13.9
(15.3)
0
0
0
0
0
0
0
0
5-year
cumul ati ve
206,556.0
(227,685.0)
4,704.5
(5,185.7)
2,458.5
(2,710.0)
206.2
(227.3)
35,213.0
(38,815.0)
798
(879.8)
419.2
(462.1)
35.2
(38.8)

-------
         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
  Setti ng
Participate matter.

EAF:          0.5-1.0 pm (64%)
             5.0-10.0 pm (20%)
            10.0-20.0 pm (4%)
            20.0-40.0 pm (6%)
            40.0-50.0 pm (6%)

AOD:          0.1-0.5 pm (15%)
             0.5-1.0 pm (22%)
             1.0-2.5 pm (29%)
             2.5-5.0 pm (10%)
             5.0-10.0 pm (5%)
            10.0-20.0 pm (19%)
EAF + AOD:    0. 1-0.5 pm (49%)
             0.5-1.0 pm (8%)
             1.0-2.5 pm (10%)
             2.5-5.0 pm (7%)
             5.0-10.0 pm (6%)
            10.0-20.0 pm (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, Okla.  (uniform terrain;
0.9% calm winds; predominant wind
direction).

Urban.
                                                               (conti nued)
                                    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 90°)
  Oklahoma City, Okla.  (310° to 10°).

Receptors at same elevation as plant
grade.
                                    7-i

-------
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).   Emission gas streams such as those modeled (with low emission
heights, low exit velocities, and temperatures close to ambient) have
small plume rise 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

-------
     FURNACE BUILDING
                                                     FABRIC  FILTER
                     ROOF MONITOR
                                                    20 m APART
Figure 7-1.   Typical  furnace building and fabric filter layout
      (with  roof monitor open across furnace building).

-------
             TABLE  7-6.   MODELING DATA  FOR FABRIC FILTER  SOURCES
Case . Furnace
No.S'b ID8
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
A
A
A
B
3
B
C
C
C
0
D
D
A
A
A
3
8
B
E
E
E
F
F
F
A/E
A/E
A/E
B/F
B/F
B/F
Reg Emissio
alt. height,
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
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
Fabric
n filter monitor
Fabric
filter
m length, m width, m width, m
15
15
15
.24
.24
.24
15.24
15.24
15.24
22.86
22.86
22.86
22
22
22
15
15
15
22
22
22
15
15
15
22
22
22
15
15
15
38
38
38
.36
.85
.86
.24
.24
.24
.86
.86
.86
.24
.24
.24
.86
.86
.86
.24
.24
.24
.10
.10
. 10
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
aCases 1 through 12 represent individual carbon steel
.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 0 are EAF's; furnace ID's
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
facilities;
E and F are
Emission ,
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
1.386
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
; cases 13 t
AOO vessels
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
hrough 30
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
represent
Exit
velo-
city
m/s
0.98
0.98
1.03
3.91
3.91
4.11
3.15
3.15
3.55
4.06
4.06
4.25
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

The exit  temperature  for ail cases is  394°K.

-------
                                  TABLE 7-7.  MODELING DATA FOR ROOF MONITOR  SOURCES
I
ro
(.dbJ lul
•MdCe Key.
No.B ID1' alt.
.11
32
13
11
i5
36
37
3U
39
40
41
42
43
44
45
46
4 /
48
['Width is
br .. ,
t(.us-es 31
,1 uriuice

A
A
A
B
11
11
C
C
C
1)
1)
0
A/E
A/E
A/E
li/F
IVI
B/E
3.05 in
through
ID'S A,

1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Inr
42
B

1 ength of enn ss ion
1) i stance
between
Emission
lit, m
27.4
27.4
2/.4
33.5
33.5
33.5
38. 1
38. 1
38. 1
36.6
36.6
36.6
27.4
27.4
27.4
33.5
33.5
33.5
all cases.


Open 1
45. 72
10. 6/
10.67
68.58
15.24
15.24
106.68
25.91
25.91
76.20
18.29
18.29
53.34
7.62
7.62
91.44
15.24
15.24

Roof

Space
__
13.72
13.72
-_
21.34
21.34
_-
27.43
27.13
-_
22.86
22.86
--
13.72
13.72
--
21.34
21.34

represent individual carbon
C, and 1) are

area.
EAf's


; furnace


monitor' length, m

1 Open 2 Space 2
-
21.34
21.34
-_
32.00
32.00
__
53.34
53:34
__
35.05
35.05
__
10.67 13.72
10.67 13.72
_-
18.29 21.34
18.29 21.34

steel facilities; cases
ID's E and F are AOD ves




Open 3 rate, g/s
0.515
0.258
0.00
2.06
1.03
0.00
6.18
3.09
0.00
4.64
2.32
0.00
0.515/0.275
7.62 0.258/0.137
7.62 0.00/0.00
2.06/1.10
15.24 1.03/0 549
15.24 0.00/0.00

43 through 48 represent
sels.


Exit
Operating times temp.
h
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24

days
275
275
275
300
300
300
365
365
365
355
355
355
275
275
275
300
300
300

specialty steel






K
320
320
320
320
320
320
320
320
320
320
320
320
320/690
320/320
320/320
320/690
320/320
320/320

shops.



Exit
velocity
m/s
3.38
3.38
3.38
3. 74
3. 71
3. 74
4.00
4.00
1.00
3.90
3.90
3.90
3.38/3.90
3.38
3.38
3.74/4.3I
3.74
3.74





emission areas.
1 miss ions .ire uniformly spread
across
area(s).







-------
     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 AOD 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 m, 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  =        m
                          g   exp (0.5 In"1 s )
                                 7-14

-------
Standard geometric deviations for participate 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):
                                     Standard            Particulate
Averaging time                         type          concentration (ug/m3)
24-hour maximum (not to              Primary                 260
be exceeded more than                Secondary               150
once per year)
Annual geometric mean                Primary                  75
                                     Secondary                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 are 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 ug/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 PARTICULATE MATTER
             CONCENTRATION  IMPACTS  FROM FABRIC FILTER SOURCES4
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
^
17H
18d
19
20
21
22H
23H
24d
25
26
27
28
29
30
Furnace
ID
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
Regul atory
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
impact (ug/m3)
Oklahoma City
14.5
14.5
15.2, 10.7
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
 Cases 1  through  12  represent  individual  carbon steel  facilities;  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.
 Based on different  air flow/rates  and thus  different emission rates (see
dTable 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 NAAQS1s.   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

-------
                     TABLE 7-9.  HIGHEST SECOND-HIGHEST 24-HOUR PARTICULATE MATTER
                            CONCENTRATION IMPACTS FROM ROOF MONITOR SOURCES
Concentration impact (|jg/m3)
Pittsburgh
Case
No.
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Cases 31
Furnace
IDb
A
A
A
B
B
B
C
C
C
D
D
D
A/E
A/E
A/E
B/F
B/F
B/F
through
, special ty steel
Furnace ID's A,
^Distance
'-'KIM — Ki^t
Reg.
Without buoyancy
alt. at 100 m at 300 m
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
42 represent
44.0
22.0
0
113
52.8
0
226
111
0
199
118
0
65.9
32.7
0
167
79.8
0
individual
25.3
12.6
0
71.5
35.5
0
170
84.7
0
139
84.3
0
38.8
19.4
0
107
53.3
0
carbon
shops.
B, C, and D are EAF's; Turnace
from source center to
receptor.
n Di +
Without
With buoyancy at 100 m
23.9|
12.0 at
0 J
NMd
NM
0
NM
NM
0
96.9?
57. 3( at
0 J
NM
NM
0
67.7?
44.5) at
o J
steel facili
ID's E and F

23.6
165 mc 11.9
0
68.4
30.0
0
124
62.3
0
112
204 m 68.7
0
35.0
17.4
0
83.3
209 m 39.8
0
ties; cases 43
are AOD vessels

Oklahoma
buoyancy
at 300 m
14.0
7.0
0
38.8
18.5
0
86.7
40.7
0
74.4
43.3
0
21.5
13.3
0
56.7
27.4
0
through 48


City

With buoyancy
24.6?
12.3 at 165 m
0 J
NM
NM
0
NM
NM
0
111 ?
62.4 (at 204 m
0 J
NM
NM
0
83.9]
49.6 /at 209 m
0 J
represent


at 300 m modeled without buoyancy.   Similarly, Oklahoma City impacts are estimated to be about the same
as the 100 m impacts modeled without buoyancy.

-------
      TABLE 7-10.   MAXIMUM ANNUAL ARITHMETIC MEAN PARTICULATE MATTER
             CONCENTRATION IMPACTS FROM FABRIC FILTER SOURCES

Case
No.3
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 ID
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

Regul atory
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
Concentrati
on impact
excluding background, ug/m3
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
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.
3Furnace  ID's A,  B,  C,  and  D  are  EAF's;  furnace  ID's  E and  F  are  ADD
 vessels.
                                   7-19

-------
TABLE 7-11.   MAXIMUM ANNUAL ARITHMETIC MEAN PARTICULATE MATTER
         CONCENTRATION IMPACTS FROM ROOF MONITOR SOURCES
Case
No.3
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Furnace ID
A
A
A
B
B
B
C
C
C
D
D
D
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
Concentration impact
excluding background, ug/m3
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
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
 Cases 31  through 42 represent individual  carbon steel
 facilities;  cases 43 through 48 represent specialty steel
 shops.
'Furnace ID's A,
 are AOD vessels.
shops.
Furnace ID's  A,  B,  C,  and D are EAF's;  furnace ID's E and F
 .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 ug/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

-------
                                                                                          a,1
            TABLE 7-12.  SOLID WASTE (FABRIC FILTER CATCH) GENERATION FOR EACH MODEL PLANT '
                                             Mg/yr (tons/yr)
Furnace/vessel :
Model furnace:
Furnace capacity:
Mg
(tons)
Furnace annual
production:
Mg
(tons)
^ Regulatory
^ alternative
ro
1

2

3


A

22.
(25)


37,420
(41,250)



548.
(604.
554.
(611.
560.
(618.



7








5
7)
6
4)
8
2)







163
(180



2
(2
2
(2
2
(2
EAF
B

90.7
(100)


,300
,000)



,393.6
,638.5)
,420.3
,667.9)
,247.0
,697.3)
ADD






596
(657



8
(9
8
(9
8
(9
C

272.2
(300)


,000
,000)



,736.9
,630.6)
,834.1
,737.7)
,931.5
,845.1)






434
(479



6
(7
6
(7
6
(7
D

136.1
(150)


,800
,300)



,373.8
,025.8)
,444.7
,103.9)
,515.8
,182.3)
E

22.7
(25)


37,420
(41,250)



292.6
(322.6)
295.8
(326.1)
299.0
(329.7)






163
(180



1
(1
1
(1
1
^(1
F

90.7
(100)


,300
,000)



,277.0
,407.6)
,290.9
,422.9)
,305. 1
,438.6)
aAssuming 99.9 percent fabric filter capture efficiency.

-------
--J
I
ro
                              TABLE 7-13.   SUMMARY OF POTENTIAL INDUSTRY WIDE SOLID WASTE
                                 GENERATION FROM FABRIC FILTERS—CARBON STEEL INDUSTRY
                                                    Mg/yr (tons/yr)
No. of facilities
with model
furnace size3
B
C
D
Reg. Alt. 1
Reg. Alt. 2
Increase in solid wastes due to
Reg. Alt. 2
Reg. Alt. 3
Increase in solid wastes due to
Reg. Alt. 3
1983
2
0
1
11,210
(12,360)
11,340
(12,500)
130
(140)
11,460
(12,630)
250
(270)
1984
0
1
1
15,240
(16,800)
15,410
(16,990)
170
(190)
15,580
(17,170)
340
(370)
1985
3
0
1
13,590
(14,980)
13,740
(15,150)
150
(170)
13,900
(15,320)
310
(340)
1986
0
1
1
15,240
(16,800)
15,410
(16,990)
170
(190)
15,580
(17,170)
340
(370)
1987
3
0
1
13,590
(14,980)
13,740
(15,150)
150
(170)
13,900
(15,320)
310
(340)
5-year
cumulative
8
2
5
201,900
(222,500)
204,100
(225,000)
2,200
(2,500)
206,400
(227,500)
4,500
(5,000)
        Explanation of model furnaces in Chapter 6:
            B = 90.7-Mg (100-ton) EAF
            C = 272-Mg (300-ton) EAF
            D = 136-Mg (150-ton) EAF

-------
I
ro
                              TABLE 7-14.  SUMMARY OF POTENTIAL INDUSTRY WIDE SOLID WASTE
                               GENERATION FROM FABRIC FILTERS—SPECIALTY STEEL INDUSTRY
                                                    Mg/yr (tons/yr)
No. of furnaces/
vessel combination
A/E
B/F
Reg. Alt. 1
Reg. Alt. 2
Increase in solid waste due to
Reg. Alt. 2
Reg. Alt. 3
Increase in solid waste due to
Reg. Alt. 3
1983
0
1
3,711
(4,091)
3,752
(4,136)
41
(45)
3,794
(4,182)
83
(91)
1984
1
0
787
(868)
796
(877)
9
(9)
805
(887)
18
(19)
1985
0
1
3,711
(4,091)
3,752
(4,136)
41
(45)
3,794
(4,182)
83
(91)
1986
1
0
787
(868)
796
(877)
9
(9)
805
(887)
18
(19)
1987
0
0
0
0
0
0
0
0
0
0
0
0
5-year
cumul ati ve
2
2
34,410
(37,940)
34,790
(38,350)
380
(410)
35,180
(38,780)
770
(840)
        Explantion of model plants in Chapter 6:
            A/E = one each 22.7-Mg (25-ton) EAF and AOD
            B/F = one each 90.7-Mg (100-ton) EAF and AOD

-------
and hexavalent chromium, it is assumed that it will also be a hazardous
waste.4  Furthermore, dusts from both EAF and AOD 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 zi nc. 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 (Mi), 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

-------
     TABLE  7-15.   INDUSTRY WIDE ELECTRICAL ENERGY REQUIREMENTS FOR 1983-1987 AND 5-YEAR CUMULATIVE
                                               (kWh/yr)
Regulatory
al ternati ve
Carbon steel
]
2
3
Specialty steel
1
2
3
1983
23,657,000
23,657,000
24,076,000
16,307,000
16,307,000
16,661 ,000
1984
18,694,000
18,694,000
20,839,000
4,549,000
4,549,000
4,766,000
1985 1986 1987
30,451,000 18,694,000 30,451,000
30,451,000 18,694,000 30,451,000
30,871,000 20,839,000 30,871,000
16,307,000 4,549,000
16,307,000 4,549,000
16,661,000 4,766,000
5-year
cumulative
352,253,000
352,253,000
368,898,000
157,750,000
157,750,000
161,884,000
Cumulatives are in kWh.

-------
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 ADD
vessel heat is in operation on the offgas "duct of a close-fitting hood
of an AOO 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 solid 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
landfi1 led,  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 landfill 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.  Morn's.   Further Dispersion
     Modeling Analyses of Participate 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  ADD
     vessel emission controls.

12.   Junker, A.   Electric Steelmaking--The  Bottom Tapping Combined
     Process Furnace (CPF), Part I—Technical  Inovations.   Iron and
     Steel  Engineer.   59(12):25-28.   December  1981.
                                 7-29

-------
                                8.  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 EAF/AOD vessel,  the 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

-------
        TABLE 8-1    CAPITAL AND ANNUALIZED  COSTS  FOR  MODEL  PLANTS1-5
                          (March  1981  dollars)
Model plant
1 . New shop: one 22.7-Mg
(25-ton) EAF
2. New shop: one 90. 7-Mg
(100-ton) EAF
3. New shop: one 272.2-Mg
(300-ton) EAF
4. New shop: one 136.1-Mg
(150-ton) UHP EAF
5. One new 272.2-Mg (300-ton)
EAF added to an existing
shop
6. One new 22.7-Mg (25-ton)
EAF added to an existing
shop
7. New shop: EAF and ADD vessel,
22.7-Mg (25-ton) capacity
each; separately controlled
8. New shop: EAF and ADD vessel ,
22.7-Mg (25-ton) capacity
each; commonly controlled
9. New shop: EAF and AOD vessel,
90. 7-Mg (100-ton) capacity
each; separately controlled
10. 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 Annualized
costs3 costs
$10b $10b $/Mgc
7.0 8.9 238
20.0 28.7 176
25.6 89.8 150
33.2 68.8 158
17.0 88.3 148
4.5 8.5 227
14.6 18.8 502
14.6 18.8 502
44.0 68.0 416
44.0 68.0 416
blndudes  direct  and  indirect  capital  costs.
Capital  recovery and scrap  costs  are  included.
 Costs are per megagram  of  steel produced.
                                    5-2

-------
capital recovery costs are based on an  interest rate of 10 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 the 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, 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.
     Regulatory Alternative 1 is the baseline alternative,  which corresponds
to the existing standards of performance for EAF's and the typical  State
regulations for ADD vessels.   Regulatory Alternatives 2 and 3 represent
more stringent levels of emission control.   Several  of the model plants
show lower total capital  costs under Regulatory Alternative 2 than under
Regulatory Alternative 1.   This is because 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
                                  8-3

-------
TABLE 8-2.   COMPONENT CAPITAL COST FACTORS FOR FABRIC FILTERS
       AS A FUNCTION OF EQUIPMENT COST—NEW FACILITIES7
                    (March 1981 Dollars)

                                                       Cost
              Cost element                            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)                C

2.   Direct installation costs

    Foundation and supports                           0.04C
    Handling and erection                             0.75C
    Electrical                                        0.08C
    Piping                                            0.01C
    Painting                                          0.02C

  Total                                               0.90C

Indirect costs

3.   Indirect installation costs

    Engineering and supervision                       0.15C
    Construction and field expenses                   0.30C
    Construction fee                                  0.10C
    Start-up                                          0.01C
    Performance test                                  0.01C
    Contingencies                                     0.03C

  Total                                               0.60C

Total capital costs (1+2+3)                       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
    A.   Plant and payroll

9.   Capital charges
    A.   Administrative
    B.   Property tax
    C.   Insurance
    D.   Capital  recovery factorc
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.
                               3-5

-------
        TABLE 8-4.
              EQUIPMENT
CAPITAL AND ANNUALIZED  COSTS OF POLLUTION  CONTROL
     EAF/AOD PROCESS  AND  FUGITIVE  EMISSIONS3  10
          (March 1981  dollars)
Capital costs
Model plant
1 Sew shoo: one 22.7-Mg
(25-ton) EAF

2- New shoo: one 90 7-Mg
(100- ton) EAF

3. New snop: one 272.2-Mg
(300-ton) EAF
4. New shop: one 136.1-Mg
i. 150- ton) UHP EAF

5 One new 272.2-Mg (300-ton)
EAF added to an existing
snop
6 One new 22- 7-Mg (25-ton)
EAF added to an existing
snoo

7 New snoo: EAF and ADD vessel ,
22.7-Mg (25-ton) capacity
each; separately controlled
3. New shoo- EAF and AOO vessel,
22.7-Mg (25-ton) capacity
eacn; commonly controlled
9 New shop: EAF and AOO vessel,
90. 7-Mg (100-ton) capacity
each; separately controlled
10. New shoo: EAF and AOO vessel,
90. 7-Mg (100-ton) capacity
eacn; commonly controlled
Producti on
Mg/yr Re
(tons/yr) al
37,420
(41 ,250)

163,300
(180,000)

596,000
(657,000)
434,800
(479,300)

596,000
(657,000)
(Addi ti ona 1 )
37,420
(41 ,250)
(Additional )

37,420
(41 ,250)

37,420
(41 .250)

163,300
(180,000)

163,300
(180,000)

Percentage of
sgulatory total plant
ternative 310s capital costs
1
2
3
3e
1
2
3
1
2
3
]
2
3
1
2
3
1
2
3
3e
1
2
3
1
2
3
1
2
3
1
2
3
1.73
1, 70
1.74
1 . 71
4.37
4.36
4. 47
4.87
4.87
5.30
6.29
6.28
6.49
5.84
5.85
6.36
2.08
2.04
2.09
2.05
3.59
3.56
3.72
2.84
2.81
2.92
9.28
9.27
9.67
8.45
3.41
8.88
19.8
19.6
20.0
19.7
17.9
17.9
18.3
16.0
16.0
17. 1
16.0
15.9
16.4
25.5
25.6
27.2
31.5
31.2
31. 7
31 .3
19.7
19.6
20.3
16.3
16.2
16.7
17.4
17. 4
18.0
16. 1
16.0
16.8
Annual i zed costs
c sio-
500.8
488.9
502.9
482.2
1 ,41 1 . 7
1 ,405.0
1 ,428.3
2,230.5
2,232.4
2,395.4
2,316.9
2,314.2
2,372.3
2,307.0
2,309.0
2,477.5
518.0
505. 7
520.2
498.7
933. 7
939.3
980.5
809.4
796.8
826.9
2,505.8
2,514.7
2,605.7
2,371.7
2,358 4
2,460.4
S/Mg"
13. 38
13 06
13.44
12.37
3.54
8.50
8. 75
3. 74
3.75
4,02
5.33
5.32
5.46
3.87
3.87
4. 16
13.34
13 51
13.90
13.33
24 95
25. 10
26.20
21.63
21.29
22. 10
15.34
15.40
15.96
14.52
14.44
15.07
^ncl^aes control  device, capture devices, ductwork,  and solid waste handling.
3Incljdes direct and indirect capital,costs.
 Total  ol ant investment includes the costs of the melt shop  and the pollution control equipment.
 Costs  are per megagram of steel produced.
 Casts  are 'or total furnace enclosure {TFE) option.  TFE costs were evaluated only for
 22 7-^g EAF15.

-------
flows through the capture hoods.  Under Regulatory Alternatives 2 and 3,
EAF facilities can monitor the opacity of the visible emissions from the
control device with a continuous opacity monitor or by visual observations
made by a certified smoke reader.  These alternatives also allow monitoring
of the flows through the hoods by periodic inspections of the capture
system and once-per-shift recording of the key operating parameters
(e.g., damper position, fan amps, furnace static pressure).  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 smoke reader are zero.  Even though the capital costs for
fugitive emissions capture equipment are higher under Regulatory Alterna-
tive 2 than under Regulatory Alternative 1, these higher costs are
offset by the decrease in cost related to the use of a certified smoke
reader.  The capital costs under Regulatory Alternative 3 are higher
than capital costs under Regulatory Alternative 1 because of higher
costs for the control and capture systems.   The annualized costs for a
majority of the model plants are lower under Regulatory Alternative 2
than under Regulatory Alternative 1 because of the reasons discussed
above.  The annualized costs under Regulatory Alternative 3 are higher
than costs under Regulatory Alternative 1.
     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
EAF dust is considered hazardous and disposal at a certified landfill is
requi red.
     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 because 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.
                                  3-7

-------
     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,:x,12  The installation costs shown on
Table 8-4 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 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 xl06-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 8-5 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 $79  million in capital  costs to install  control
systems by 1987 to meet the  regulatory alternatives and up to $28  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 $6.6 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 8-6 presents the  annual costs of incremental  (over baseline
levels) collection of process  and  fugitive dust.   Total  furnace enclosures
are evaluated as a process and fugitive emissions capture system option

-------
 I
ID
                            TABLE 8-5.   PROJECTED NATIONWIDE CONTROL SYSTEM EXPENDITURES
                                   UNDER VARIOUS REGULATORY ALTERNATIVES (BY 1987)
                                                (March 1981 dollars)
Capital costs, $106
No. of
Model plants plants
Carbon Steel
One 90.7-Mg (100-ton) EAF
One 136.1-Mg (150-ton) EAF
One 272.2-Mg (300-ton) EAF
Total
Specialty Steel
One 22.7-Mg (25-ton) EAF/AOD shop
One 90.7-Mg (100-ton) EAF/AOD shop
Total
8
5
2
15
2
2
4
Regulatory alternative
1 2 3
35.
31.
9.
76.
5.
16.
22.
0
5
7
2
7
9
6
34.
31.
9.
76.
5.
16.
22.
9
4
7
0
6
8
4
35.8
32.4
10.6
78.8
5.8
17.8
23.6
Annual ized costs^$106
Regulatory alternative
1 2 3
11.
11.
4.
27.
1.
4.
6.
3
6
5
4
6
7
3
11.
11.
4.
27.
1.
4.
6.
2
6
5
3
6
7
3
11.4
11.9
4.8
28.1
1.7
4.9
6.6

-------
          TABLE  8-6.   AVERAGE COST  EFFECTIVENESS OF  EAF/AOD  PROCESS
                 AND FUGITIVE  EMISSIONS CONTROL OVER  BASELINE
Regulatory
Model plant alternative
1. New shop: one 22.7-Mg
(25-ton) EAF

2. New shop: one 90.7-Mg
(100-ton) EAF

3. New shop: one 272. 2-Mg
(300-ton) EAF

4. New shop: one 136.1-Mg
(150-ton) UHP EAF

5. One new 272. 2-Mg (300-ton)
EAF added to an existing
shop
5. One new 22.7-Mg (25-ton)
EAF added to an existing
shop

7. New shop: EAF and ADD
vessel , 22. 7-Mg (25-ton)
capacity each;
controlled separately
8. New shop: EAF and ADD
vessel , 22. 7-Mg (25-ton)
capacity each;
controlled in common
9. New shop: EAF and ADD
90.7-Mg (100-ton)
capacity each;
separately controlled
"0. New shop: EAF and AOD
vessel, 90.7-Mg (100-ton)
capacity each;
controlled in common
1
2
3,
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
i 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)
(20.5)

0
(44.7)
(89.3)

0
(44.7)
(89.8)

Incremental Annual cost of
cost over incremental
baseline collection
S/yr
0
-11 ,900
2,100
-18,600
0
-6,700
16 ,600
0
1 ,900
164,900
0
-2,700
55,900
0
2,000
170,600
0
-12,300
2,200
-19,300
0
5,600
46,800

0
-12,600
17,500

0
8,900
99,900

0
-13,300
88,700

S/Mg
0
-1 ,951
170
-1 ,512
0
-251
311
0
20
347
0
-38
394
0
21
877
0
-2,016
179
-1 ,569
0
602
2,503

0
-1 ,355
936

0
219
1 ,226

0
-.328
1 ,088

($/ton)
0
(-1,776)
(155)
(-1,377)
0
(-228)
(282)
0
(18)
(769)
0
(-35)
(357)
0
(19)
(795)
0
(-1 ,836)
(163)
(-1 ,430)
0
(549)
(2,272)

0
(-1,235)
(850)

0
(199)
(1,112)

0
(-298)
(988)

"Costs are oer negagram (ton) of particulate matter removed.
"otal furnace enclosure option.
                                       8-10

-------
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 8-6, the TFE capture system incremental cost
is 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 incre-
mental costs of collecting dust are higher for the smaller furnaces.
The smaller furnace/higher cost relationship is also true with the ADD
vessels.
     A majority of the model facilities show a net savings in the annual
costs of incremental collection of the particulate matter under Regulatory
Alternative 2 compared to the baseline levels of control.   For those
model facilities that do not show a savings, the annual cost under
Regulatory Alternatives 2 and 3 range from a high of $847/Mg (769/ton)
to a low of $20/Mg ($18/ton) of emission reduction for new EAF facilities.
The range is from a high of $2,503/Mg ($2,272/ton) to a low of $219/Mg
($199/ton) of emission reduction for new specialty steel facilities with
an EAF and ADD 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 Facilities
     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.
                                  8-11

-------
     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.
8.2  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 8-7.
     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 AOD
vessel s.
     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.
                                  8-12

-------
             TABLE 8-7.   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 ADD 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,
h
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
pol 1 ution
control
investment
$106
1.73
1.70
1.74

1.71
4.37
4.36
4.47
4.87
4.87
5.30
6.29
6.28
6.49
5.84
5.85
6.36
2.08
2.04
2.09
2.05
3.59
3.56
3.72
2.84
2.81
2.92
9.28
9.27
9.67
8.45
8.41
8.88
OSHA
compl iance
investment
$103
239
238
244

239
612
610
626
682
682
742
881
879
909
818
819
890
291
286
293
287
503
498
521
398
393
409
1 ,299
1 ,298
1 ,354
1 ,183
1 ,177
1 ,243
^Calculated as 14 percent of total  pollution control  investment.
3Total  furnace enclosure option.
                                   8-13

-------
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 & Laughlin  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.  GARD, Inc.   Miles,  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 Iversen,  R.,
     EPA:ISB.   June 10,  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 Bom'stalli, 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.

-------
                           9.  ECONOMIC IMPACTS

9.1  SUMMARY OF IMPACTS
     Three regulatory alternatives proposed for electric arc furnace
(EAF) and electric arc furnace/argon-oxygen decarburization vessel (EAF/AOD
vessel) steel plants are considered in this analysis.  Regulatory
Alternative 1 is the baseline alternative from which the impacts of the
more stringent Alternatives 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 representative 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 xlO6 Mg
(131.7 xlO6 tons); steel industry employment would be 438,461 workers;
and steel consumers in the U.S.  would import 20.3 xlO6 Mg (22.3 xlO6 tons)
of foreign steel.
     It is estimated that the price of semifinished carbon  steel would be
$0.03 per Mg (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 $0.16 per Mg ($0.14 per ton) increase.  These price
increases represent impacts of 0.02 percent and 0.02 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 38,000  Mg (42,000 tons)
from baseline under Alternative 2--an impact of 0.03 percent.   Imports
could be approximately 5,400 Mg (6,000 tons) greater—an impact of
0.03 percent.
                                  9-1

-------
The impact on domestic steel  production would cause an estimated reduction
in employment opportunities  of about 0.03 percent,  or about 137 jobs, in
1987   The estimated total  cost of Regulatory Alternative 2 in 1987 is
$0.313 million (in 1981  dollars).
     It is estimated that Regulatory Alternative 3  would raise the price
of semifinished carbon steel  by $0.23 per Mg ($0.22 per ton) above its
baseline 1987 price.   The anticipated impact on the price of specialty
steel  of Alternative 3 is $1.79 per Mg ($1.62 per ton).   These price
increases represent impacts  of 0.11  percent and 0.27 percent,  respectively.
     Domestic steel  shipments  would be down an estimated 297,000 Mg
(327,000 tons) in 1987 under  Regulatory Alternative 3--an impact of
0.25 percent.   This impact on  domestic steel  production would  cause a
proportionate decrease in 1987 employment opportunities, about 1,088 jobs.
Steel  imports in 1987 would  be approximately 36,000 Mg (39,600 tons)
higher under Alternative 3 than under baseline control—an impact of
0.17 percent.   The estimated  total  cost of Alternative 3 in 1987 is
$2.766 million (in 1981  dollars).
     Electric arc steelmaking  is apparently a profitable activity.   This
is indicated by two kinds of  evidence.   First, construction of new EAF
facilities and the replacement of older steelmaking 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
steelmaki ng.
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
                                  9-2

-------
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 organizational 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
prevailing market structure.  Most important of these basic conditions
are supply and demand conditions.  Supply conditions are  largely techno-
logical 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.
     The fifth and final 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
                                  9-3

-------
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 important
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 9.2.2.  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
represents approximately 0.8 percent of total gross national product for
1977   Employment in the steel  industry in 1977 was 441,900, approximate-
ly 0.5 percent of total  U.S.  employment in 1977.   Capital  expenditures on
new plants and equioment in  SIC 3312 was $2,143.1 million  in 1977, approxi-
mately 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
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
                                  9-4

-------
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.
     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 molten  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
                                  9-5

-------
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 such 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
manganese,  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  lowralloy 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
     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 continuously cast;  the rest  is  produced in ingot  form.11,12
                                  9-6

-------
     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
identifies 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.
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
                                  9-7

-------
  TABLE 9-1.   IMPORTANT INPUTS TO THE BLAST FURNACES AND STEEL
                    MILLS INDUSTRY—SIC 33127
                                (1972)

                                                     Share of
                                                    total input
              Input                                cost, percent
Blast furnace and steel  mill  products

Iron ore

Railroad transportation

Wholesale trade services

Coal

Industrial  chemicals

Electrometal1urgical 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-J

-------
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
producers 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 9.2.2.1.2.3.
     The production of raw steel generally involves the introduction of
some combination of pig iron,  iron scrap, and 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
                                  9-9

-------
arc furnace (EAF).   The  Bessemer-type furnace,  associated with the oldest
steelmaking 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.15,16
     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 1S   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 quantities of pure oxygen at low relative prices
has made BOF steelmaking economically feasible.19
     The chief advantage of  BOF  steelmaking 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
                                  9-10

-------
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 startup 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
potential.
     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
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
industry is frequently criticized for being slow to adopt new technologies.21
                                  9-11

-------
     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 steelmaking 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. ll
     The major material input in raw  steel production is metal--mo!ten
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.
     When  steelmaking 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/
                                  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

-------
     TABLE  9-3.   PIG  IRON  AND  SCRAP  INPUTS  TO  RAW
                   STEEL  PRODUCTION23
                         Pig  iron,
                          percent
                   Scrap,
                  percent
1978
  Open hearth
  Basic oxygen
  Electric arc
53.5
72.1
 2.7
46.5
27.9
97.3
1976
  Open hearth
  Basic oxygen
  Electric arc
55.7
71.6
 1.5
44.3
28.4
98.5
1974
  Open hearth
  Basic oxygen
  Electric arc
54.2
71.5
 3.0
45.8
28.5
97.0
                       9-14

-------
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.
     9.2.2.2.1  Historical demand trends.   As indicated earlier, steel  is
an intermediate good; most sales are  to other producing industries.
Because 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.
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.
                                  9-15

-------
       TABLE  9-4.   U.S.  REAL  GROSS  NATIONAL
               PRODUCT  AND  APPARENT
      CONSUMPTION  OF  STEEL  MILL  PRODUCTS26-28

                                        Apparent
                                      consumption
                 Real GNP,            of  steel mill
Year             1972 $109          products,  103 Mg
1979              1,431.6              104,270
1978              1,399.2              105,800
1977              1,340.5                98,365
1976              1,273.0                91,678
1975              1,202.3                80,738
1974              1,217.8              108,485
1973              1,235.0              111,133
1972              1,171.1                96,698
1971              1,107.5                92,981
1970              1,075.5                88,070
1979              1,078.8                93,133
                     9-16

-------
     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 9.2.2.2.3.
     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
products 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
industry, 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 signifi-
cant 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.
     9.2.2.2.3  Competing products.   Section 9.2.2.2.1 indicated that the
growth rate in the demand for steel  has lagged behind the growth rate in
real GNP in recent years.   Section 9.2.2.2.2 emphasized that steel mill
products are primarily intermediate goods used in the production of other
                                  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

-------
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  propor-
          tionately 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 9.2.3.2.
     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
                                  9-19

-------
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 Scontain coal  and ore
production 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 Slisted 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
     9-2.3.3  Vertical Integration.   As indicated, certain  steel production
technologies favor some degree  of integration.  A plant that produces
                                  9-20

-------
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 Manufacturers 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
Coke, iron, raw steel, finished steel
Iron, raw steel, finished steel
Raw steel, finished steel
Single product or other combination
Total No. of plants
Percent
total pi
1972
10.7
3.6
19.5
66.2
364.0
of
ants
1977
6.9
3.4
16.3
73.4
504.0
             9-22

-------
     Inadequate profit margins in steel making, to be 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 exam'ined 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
steelmaking.  Large-scale plants can more efficiently use equipment,
labor,  and energy than small  plants."44  The same report follows with an
                                  9-23

-------
estimate of 4 xlO6  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 xlO6 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 is
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 xlO6 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 startup 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 approximates 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 by demanders to be homogeneous is an important
determinant of industry pricing behavior.   The more homogeneous the
product,  the more likely a single market price will be observed.   A
perfectly homogeneous 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 9.2.2.1.1  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 9.2.2.1.2.2 of
this profile,  which discusses production processes employed in SIC 3312,
                                  9-25

-------
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.
     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 existence  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
                                  9-26

-------
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 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 9.2.3.6 concludes that effective barriers to
entry do not exist in the steel industry.
     The apparent lack of significant barriers to entry promotes the
selection 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, economies 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
participants 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 adherence to  a pattern of rigidly administered pricing.
During normal times in the period prior to the 1960's, U.S. Steel Corporation
with a market share exceeding one-half, assumed the role of price leader
and its smaller rivals followed.55
                                  9-27

-------
     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
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--structural ly 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   .  .  .58
     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 "appropriate" response  by  all firms.61
                                  9-28

-------
     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 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
industry 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.35,66
The success of mini-mills is, in turn, the result of a successful t
echnological 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 1978.28,41
                                  9-29

-------
     Steel  imports  may significantly affect the price behavior of domes-
tic 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 xlO6 Mg in
1979.52  In that same year, total  steel  mill imports totaled 15.9 xlO6 Mg-
83.7 percent of the dominant firm's  output.28  As foreign producers
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,
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.
                                  9-30

-------
TABLE 9-7.   AFTER-TAX PROFIT TO
   STOCKHOLDERS' EQUITY69-71
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
manufac-
turing,
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
12.6
11.6
10.9
9.9
10.6
            9-31

-------
     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--?.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, com-
pared to 13.8 percent  for  all  manufacturing  stockholders.
     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
Faci1ities.  In  1980,  80 independent companies either owned EAF facilities
or owned smaller companies  that owned EAF facilities.   These SO 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 1  indicates
the firm can meet current  obligations with  little trouble.   A firm with a
liquidity ratio  below  1  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.
                                  9-32

-------
              TABLE 9-8.  U.S. COMPANIES OPERATING ELECTRIC ARC FURNACES:
                                                       (1979)
FINANCIAL DATA IF AVAILABLE
co
CO
Company
AL Tech Specialty Steel
Allegheney Ludlum Industries
Ameron
Armco
Atlantic Steel
Auburn Steel
Babcock and Wi Icox
Bayou Steel
Bethlehem Steel
Border Steel
Braeburn Steel
Cabot Corporation
California Steel
Calumet Steel
Cameron Iron Works
Carpenter Technology
Cascade Rolling Mills
Ceco Steel
CF&I Steel
Chaparral Steel
Columbia Tool Steel
Connors Steel
Copperweld
Crucible
Cyclops
Eastern Stainless Steel
Edgewater Steel
Finkel and Sons
Florida Steel
Current
assets
NAa
526,123
107,806
1,361,643
NA
NA
NA
NA
1,844,200
NA
NA
476,205
NA
NA
373,762
184,374
NA
133,694
130,424
65,289
NA
175,994
153,773
854,182
256,988
126,347
NA
NA
65,041
Current
liabilities
NA
261,973
40,849
656,741
NA
NA
NA
NA
1,035,900
NA
NA
234,074
NA
NA
142,678
80,618
NA
50,543
82,552
35,819
NA
87,257
78,099
335,534
133,378
47,578
NA
NA
24,039
Profit
NA
71,527
8,719
221,040
NA
NA
NA
NA
275,700
NA
NA
70,445
NA
NA
58,412
40,740
NA
13,273
15,371
30,258
NA
23,950
23,078
111,375
5,263
8,183
NA
NA
14,468
Equity
NA
485,154
100,555
1,715,271
NA
NA
NA
NA
2,639,800
NA
NA
399,784
NA
NA
322,181
226,872
NA
105,383
213,123
93,695
NA
147,741
151,883
638,479
168,037
77,904
NA
NA
71,696
Total assets=
total
1 iabil ities
NA
1,140,165
191,968
3,260,163
NA
NA
NA
NA
5,165,900
NA
NA
942,973
NA
NA
627,862
322,477
NA
188,978
400,217
232,455
NA
333,657
312,816
1,301,619
402,580
175,874
NA
NA
126,158
Debt
NA
655,011
91,413
1,544,892
NA
NA
NA
NA
2,526,100
NA
NA
543,189
NA
NA
305,681
95,605
NA
83,595
187,094
138,760
NA
185,916
160,933
663,140
234,543
97,970
NA
NA
54,462
                                                                                                       (continued)

-------
TABLE 9-8.   (continued)
Company
Ford Motor Company
Georgetown Steel
Hawaiian Steel
Ingersol Rand
Inland Steel
Inlercoastal Steel
Interlake
III
Jessop Steel
Jones & Laugh! in
Earle M. Jorgenson
Joslyn Stainless Steels
Judson Steel
Kentucky Electric Steel
Keystone Consolidated
Knoxville Iron
Laclede Steel
Lonestar Steel
Lukens Steel
Marathon Steel
Mclouth Steel
National Forge
National Steel
New Jersey Steel & Structure
North Star Steel
Northwestern Steel and Wire
Nucor
Oregon Steel Mills
Owens Electric Steel
Perm-Dixie Steel
Phoenix Steel
Current
assets
11 ,571 ,300
NA
NA
1 ,557,139
813,687
NA
383,477
7,026,244
104,805
1,934,223
106,149
NA
NA
NA
142,918
NA
NA
961,500
95,703
561,800
196,630
NA
1,275,417
NA
NA
NA
117,362
NA
NA
89,479
69,528
Current
1 iabi 1 ities
9,263,000
NA
NA
634,447
567,441
NA
221,207
5,621,157
46,767
1,132,659
39,055
NA
NA
NA
65,934
NA
NA
409,900
52,543
353,400
136,814
NA
732,133
NA
NA
NA
63,536
NA
NA
59,311
43,859
Profit
1 ,169,300
NA
NA
149,342
131 ,108
NA
39,735
380,685
3,337
173,527
20,174
NA
NA
NA
757
NA
NA
172,200
7,467
57,000
9,368
NA
126,466
NA
NA
NA
42,265
NA
NA
-47
3,767
Equity
10,420,700
NA
NA
1 ,095,039
1 ,321 ,351
NA
349,944
5,621,157
53,470
757,242
128,887
NA
NA
NA
113,693
NA
NA
923,600
141,262
1 ,335,000
189,226
NA
1,436,825
NA
NA
NA
133,737
NA
NA
72,478
61 ,344
Total assets=
total
1 iabi 1 ities
23,524,600
NA
NA
2,128,410
2,725,473
NA
733,559
15,091 ,321
182,141
3,864,757
182,858
NA
NA
NA
255,310
NA
NA
2,196,700
247,989
5,392,400
465,795
NA
3,160,279
NA
NA
NA
243,112
NA
NA
190,952
138,307
Debt
13,103,900
NA
NA
1 ,033,371
1 ,404,122
NA
383,615
9,470,164
128,671
3,107,515
53,971
NA
NA
NA
141,617
NA
NA
1,273,100
106,727
4,057,400
276,569
NA
1,723,454
NA
NA
NA
109,375
NA
NA
118,474
76,963
                                                       (continued)

-------
                                               TABLE 9-8.   (continued)
I
co
en
Company
Quanex
Raritan River Steel
Republic Steel
Roanoke Electric
Roblin Steel
Ross Steel Works
Sharon Steel
Simonds Steel
Soule Steel
Timet
Structural Metals
Teledyne Vasco
Tennessee Forging Steel
Texas Steel
Timken Steel
Union Electric Steel
United States Steel
Washburn Wire
Washington Steel
Witteman Steel Mills
Current
assets
79,987
NA
923,525
NA
38,514
NA
597,106
250,810
NA
NA
NA
747,610
NA
NA
455,413
NA
3,456,700
NA
158,379
NA
Current
1 iabi 1 ities
37,355
NA
570,117
NA
23,312
NA
323,763
79,742
NA
NA
NA
407,697
NA
NA
177,031
NA
2,410,800
NA
116,846
NA
Profit
13,131
NA
121,158
NA
642
NA
58,603
35,455
NA
NA
NA
371,960
NA
NA
102,131
NA
-293,000
NA
11,392
NA
Equity
70,226
NA
1,489,198
NA
25,172
NA
257,300
190,695
NA
NA
NA
1,747,979
NA
NA
705;859
NA
5,394,600
NA
53,301
NA
Total assets=
total
liabilities
147,600
NA
2,749,872
NA
91,589
NA
1,110,779
373,172
NA
NA
NA
2,027,197
NA
NA
942,912
NA
11,029,900
NA
209,999
NA
Debt
77,374
NA
1,260,674
NA
66,417
NA
853,479
182,477
NA
NA
NA
279,218
NA
NA
237,053
NA
5,635,300
NA
156,698
NA
     NA = Not available.

-------
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
Copperwel d
Crucible
Cyclops
Eastern Stainless Steel
Florida Steel
Ford Motor
Ingersol Rand
Inland Steel
Interlake
ITT
Jessop Steel
Jones & Laughlin
Earle M. Jorgenson
Keystone Consolidated
Lonestar Steel
Lukens Steel
Marathon Steel
McLouth Steel
National Steel
Nucor
Penn-Dixie Steel
Phoenix Steel
Quanex
Republic Steel
Roblin Steel
Sharon Steel
Simonds 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

-------
     The  leverage ratio  indicates the relationship between total debt and
stockholders' equity.  A ratio exceeding  1  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  1 in 1979, indicating they are substantially debt
fi nanced.
     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 conclusions, observations can be made.  The mean liquidity ratio for
EAF firms 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 9.2.5.4.
     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
Governors 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
                                  9-37

-------
          TABLE 9-10.   FINANCIAL RATIOS FOR SELECTED INDUSTRIES76
                                  (1979)

                                         Liquidity      Profit    Leverage
Industry                                   ratio        ratio       ratio


EAF steel  firms3                            1.99         13         1.33

Primary nonferrous metals,  NEC--SIC 3339    1.86         22         2.32

Aluminum foundries--SIC 3361                 2.15         22         1.84

Primary metal  products, NEC—SIC 3399       2.28         24         0.69

Motor vehicles--SIC 3711                     1.73         17         1.44

Petroleum refining--SIC 2911                 1.36         26         1.56

aMeans from Table 9-9.
                                  9-38

-------
  TABLE 9-11.   STEEL MILL PRODUCTS AND
TOTAL INDUSTRIAL OUPUT INDEXES28,41,77,78
                 (1967 = 100)
Year
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
Steel
mill products
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
Total
industrial
output
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

-------
products.   Of the period covered,  domestic steel  shipments peaked in
1973.   The 2-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
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 mill product 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 increased 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
                                  9-40

-------
TABLE 9-12.   REAL VALUE OF OUTPUT FOR SIC 3312
                  (1967 $106)
Year
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
Real value
of output
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
                    9-41

-------
TABLE 9-13.   STEEL PRICE INDEX AND
    GNP PRICE DEFLATOR42,79,80
           (1967 = 100)
Year
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
Steel price index
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
GNP deflator
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 INVESTMENT3,81,82
                    (1967 = 100)
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
    Investment expenditures were deflated
     using the GNP implicit price deflator
     before indexing.
                       9-43

-------
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
     g.2.5.3.5  Productivity.  Productivity is meant to be a  measure of
technological advance.  The most common measure of productivity is output
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  9.2.5.3.4,  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 xlO6 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
                                   9-44

-------
TABLE 9-15.   INDEX OF OUTPUT
   PER EMPLOYEE-HOUR85-87
        (1967 = 100)
Year
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
Steel industry
121.8
116.0
114.5
107.6
123.5
123.5
112.7
106.2
101.3
104.0
103.5
100.0
All
manufac-
turing
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  TRADE28,41
                        (103  Mg)
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
. a
exports
-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

-------
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 s_e 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
compounded 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
projected 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 xlO6 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 xlO6 Mg of raw
steel in 1987.   Assuming a capacity utilization rate of 85 percent,
electric furnace capacity will total 44.4 xlO6 Mg in 1987.
     It is estimated that electric furnace capacity totaled 39.2 xlO6 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 xlO6 Mg of new electric arc
furnace capacity will be added during the period 1982-1987, an annual
growth rate1 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 xlO6 Mg new electric arc
capacity, it is projected that 4.8 xlO6 Mg will be met by the construction
of carbon steel shops and 0.4 xlO6 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 xlO6 Mg
                                  9-47

-------
     TABLE 9-17.   PROJECTED DOMESTIC RAW
              STEEL PRODUCTION88
                  (106  Mg)


Year                               Production3
1990                                  124.6

1989                                  122.1

1988                                  120.9

1987                                  119.7

1986                                  119.7

1985                                  118.5

1984                                  118.5

1983                                  118.5

1982                                  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.
                   9-48

-------
TABLE 9-18.   SUMMARY OF ELECTRIC ARC FURNACE CAPACITY PROJECTION3


                New capacity additions expressed as model plants
Model           106 Mg capacity1982-1987No. plants
plant(s)           per plant          No.  of plants    per year


                           Carbon plants
1, 6
2
3, 5
4

7, 8
9, 10
a!981
1987
1983-1987
1983-1987
1983-1987
0.03742 128
0.1633 29
0.596 8
0.4348 11
Specialty plants
0.03742 11
0.1633 2
Electric arc furnace capacity:
Electric arc furnace capacity:
New electric capacity additions:
New carbon steel shop capacity:
New specialty shop capacity:
21
5
1
2

2
<]
39.2 xlO6 Mg
44.4 xlO6 Mg
5.2 xlO6 Mg
4.8 xlO6 Mg
0.4 xlO6 Mg
                             9-49

-------
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 xlO6 Mg of new specialty capacity could be met by the
construction of eleven 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 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.
Construction of these 19 plants of various sizes would approximately
satisfy the projection of 5.2 xlO6 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 xlO6 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.
                                  9-50

-------
 TABLE 9-19.   PROJECTED CONSTRUCTION OF ELECTRIC FURNACE PLANTS
                            1982-1987


                                                          Total
                                                         annual
                                               No.  of  capacity,
                                               plants    106 Mg
Carbon steel model  plant

  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

Specialty steel model plant

  Model Plants 7, 8 —22.7-Mg EAF/AOD

  Model Plants 9, 10— 90.7-Mg EAF/AOD

     TOTAL
 5

 2

15



 2

 2

 4
1.3

2.2

1.2

4.7



0.07

0.33

0.40
                             9-51

-------
     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 cannot 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.
     Of the 80 firms that currently operate one or more EAF shops, employment
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 anywhere 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  are available for the 42 firms mentioned
above.  EAF vessel  capacity, the major determinant of annual steel capacity,
                                  9-52

-------
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 cannot 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-
ficant 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 significant
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 nor 3 causes an increase
in ATC from baseline Alternative 1 as high as 5 percent for any model
plant.   The most  severe impact is 0.27 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.
                                  9-53

-------
             TABLE 9-20.   AVERAGE TOTAL COST IMPACTS ASSUMING
                 6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL




Regu
latory impact from baseline
Regulatory alternative
Model
pi ant
1
lb
2
3
4
5
6
6b
7
8
9
10

1981
18.
18.
11.
5.
7.
5.
20.
20.
35.
30.
22.
20.
1
$/Mg
.18
. 18
.94
.00
.29
.44
.96
96
82
45
24
90


Percent
6.
6.
5.
2.
3.
2.
7
7.
5.
4.
4.
3.
03
03
39
68
69
98
40
40
64
79
24
99

1981
0

0
0
0
0
0

0
0
0
0
2
$/Mg
.05
NAC
.06
.03
.03
.04
.04
NA
.16
.01
.07

3
Percent
0

0
0
0
0
0

0
0
0
0
.02
NA
.02
.02
.02
.02
.01
NA
.02

.01

1981
0.
-0.
0.
0.
0.
0.
0.
-0.
1.
1.
0.
0.
$/Mg
57
15
27
41
23
45
64
14
79
18
91
95
Percent
0. 18
-0.05
0. 12
0.21
0.11
0.24
0.21
-0.05
0.27
0.18
0.17
0.17
blmpact of baseline Regulatory Alternative 1  is measured from no control
 Total  furnace enclosure  option.
 NA = not applicable.
                                  9-54

-------
     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,
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 1 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
2.32 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.
                                  9-55

-------
TABLE 9-21.   TOTAL ANNUALIZED COST OF
   COMPLIANCE ABOVE BASELINE AS A
    PERCENT OF ANNUAL PLANT SALES
Model
plant
1
la
2
3
4
5
6
6a
7
8
9
10
Regulatory
alternative
2
0.016
NAb
0.013
0.008
0
0.009
0.019
NA
0.027
0.007
0.010
0.001
3
0.127
-0.036
0.055
0.089
0.048
0.097
0. 143
-0.029
0.216
0. 144
0.105
0.107
   .Total  furnace enclosure option.
    NA = not applicable.
               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
-0.11
NAb
0.04
0.03
0
0.09
-0.30
NA
-0.11
-0.11
0.02
-0.04
Regulatory
alternative
3
0.34
o.
0.49
1.44
0.53
2.32
0.46
-0.15
0.77
0.52
0.77
0.86
   ^Total furnace enclosure option.
    NA = not applicable.
               9-57

-------
     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 9.3.6).
     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 entities are insignificant by all measures, a regulatory flexibility
analysis is not conducted for these standards.
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 9.3.1 summarizes the range of estimated maximum impacts that
could result from the NSPS alternatives and presents anticipated impact
estimates.  Section 9.3.2 presents the theoretical model of discounted
cash flows employed to compute  both net-present value and price impacts.
Section 9.3.3 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 passthrough.  Section 9.3.4 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 9.3.5.  Section 9.3.6
presents estimates of maximum total costs of industry compliance.

                                  9-58

-------
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 $0.03 and $0.06 per Mg ($0.03 to $0.05 per ton).   These
changes represent impacts of from 0.01 to 0.02 percent.  The impact of
Alternative 2 on the cost of producing specialty steel is estimated to
range from zero to an increase of $0.16 per Mg ($0.14 per ton).  These
impacts are betweeen 0 and 0.02 percent.
     It is estimated that under Regulatory Alternative 3, cost impacts
could vary from $-0.14 to $0.64 per Mg ($-0.13 to $0.58 per ton) for
carbon steel and from $0.91 to $1.79 per Mg ($0.82 to $1.62 per ton) for
specialty steel.  These impacts are all less than 0.30 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, lead to impacts on 1987 domestic steel output, industry employment,
and steel imports.  These impacts are presented in Section 9.3.3.
     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 oppor-
tunities, and increases in steel imports.
     It is estimated that the price of semifinished carbon steel would be
$0.03 per Mg ($0.03 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 $0.16 per Mg ($0.14 per ton).   These price increases
represent impacts of 0.02 percent and 0.02 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 38,000 Mg (42,000 tons)
under Alternative 2--an impact of 0.03 percent.   Imports could be  approxi-
                                  9-59

-------
mately 5,400 Mg (6,000 tons) greater—an impact of 0.03 percent.  The
impact on domestic steel  production would cause a reduction in employment
opportunities of about 0.03 percent, or about 137 jobs, in 1987   The
estimated total cost of Regulatory Alternative 2 in 1987 is $0.313 million.
     It is estimated that Regulatory Alternative 3 would raise the price
of semifinished carbon steel by $0.23 per Mg ($0.22 per ton) above its
baseline 1987 price.   The anticipated impact on the price of specialty
steel of Alternative 3 is $1.79 per Mg ($1.62 per ton).  These price
increases represent impacts of 0.11 percent and 0.27 percent, respectively.
     Domestic steel shipments would be down an estimated 297,000 Mg
(327,000 tons) in 1987 under Regulatory Alternative 3--an impact of
0.25 percent from baseline.  This impact on domestic steel production
would cause a proportionate decrease in 1987 employment opportunities,
about 1,088 jobs.  Steel  imports in 1987 would be approximately 36,000 Mg
(39,600 tons) higher under Alternative 3 than under Regulatory Alternative 1
an impact of 0.17 percent.   The estimated total cost of Regulatory Alterna-
tive 3 in 1987 is $2.766 million.
     Electric arc steelmaking is apparently a profitable activity.  This
is indicated by two kinds of evidence.   First, construction of new EAF
facilities and the replacement of older steelmaking 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
steelmaki ng.
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 appro-
priate interest rate and summed to determine the net present value of the
project.   This section describes the DCF theory and methodology in some
detail

                                  9-60

-------
     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 1 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:
                     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
investment 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 2).  The net cash flows in this and succeeding
years can be expressed:
            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)t                      (9-4)
                                  9-61

-------
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,
                              D T = (FCC/N) •  T                       (9-5)

where N is the project life in years.   The salvage value of the plant is
assumed to be zero.
     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
appropriate 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)-t                         (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
appropriate rate of  interest in practice is important.  The interest rate
                                  9-62

-------
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 =1  (Y  • DF.) - Y                      (9-7)
                              t=l               °
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.
     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.95,96  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 9.2.4.4 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.
                                  9-63

-------
     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
alternative.  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
operating costs, or both.   If market demand for the project is perfectly
elastic, a situation of full cost absorption prevai Is — 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.
     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 downward,  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 anticipate.
                                  9-64

-------
     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 NPVs 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 substituted into (9-7) and (9-7) is set equal  to
zero, (9-7) can be written:
                  N
                  I  [(Rt - Et)(l - T) + DtT] DFt = YQ .               (9-8)
                 t1" I
If revenues and expenses are the same over all periods,  equation (9-8)
can be written:
                                   N        N
                   (R - E) (1 - T) I  DF  + I  DtTDF  = Y             (9-9)
                                  t=l      t=l
recognizing that (1 - T) is a constant.   The sum of the discount factors
as t ranges from 1 to N can be written:
                          N                    _N
                      F = I  DF.  = [1 - (1 + r) N]/r .                (9-10)
                         t=l
Substituting equations (9-5) and (9-10) into (9-9) yields:
                   (R - E) (1 - T) F + (FCC/N) • T • F = Y  .         (9-11)
                                  9-65

-------
Substituting equations (9-3) and (9-1) into (9-11) and rearranging further
yields:
             ,D   ^   P - (FCC + WC) - (TCRED • FCC) - DSL           ,q_,?,
             (.K'^J-t-           (i _ y) . p                      v. 3  i i j

where DSL = (FCC/N) •  T •  F, and represents the present value of the  tax
savings due to straight-1ine depreciation of the fixed capital.
     Finally, Q and E can be moved to the right-hand side of (9-12) to
yield:
                 D - (FCC + WC) - (TCRED •  FCC)   DSL   E             ,
                 P ~         (1 - T) F •  Q               Q '            ^  UJ
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 EAF steel 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.
     In Section 9.3.3, maximum impacts on net present value, price, and
other relevant variables are presented for  all plants under each regulatory
alternative.  In 9.3.4, 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 passthrough 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 S94.16 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
                                  9-66

-------
                                         TABLE 9-23.   MODEL PLANT COST DATA
                                                     (1981 $106)
I
cr>




No control
Plant
1
la
2
3
4
5
6
6a
7
8
9
10
Capital
cost
7.00

20.00
25.60
33.20
17.00
4.50

14.60
14.60
44.00
44.00
Annual
operating
cost
7.980

26.050
86.430
64.460
86.080
7.880

16.850
16.850
62.240
62.240
Capital
cost
8.71

24.35
30.46
39.48
22.83
6.56

18.18
17.43
53.26
52.43
Regulatory
1
Annual
operating
cost
8.229

26.863
87.994
65.924
87.684
8.151

17.302
17.273
63.543
63.514
Alternative

Capital
cost
8.70

24.36
30.47
39.48
22.85
6.54

18.16
17.41
53.27
52.41
2
Annual
operating
cost
8.232

26.869
88.009
65,936
87.698
8.156

17.310
17.277
63.551
63.518

Capital
cost
8.74
8.71
24.47
30.90
39.69
23.36
6.59
6.55
18.32
17.52
53.67
52.88
3
Annual
operating
cost
8.241
8.224
26.877
88.115
65.966
87.809
8.165
8.148
17.331
17.293
63.591
63.560
        Total  furnace enclosure option.
       3Does not include capital recovery.

-------
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,
respectively."  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
     If  the recent past is any indicator of things to come, the general
price level will rise each year through the future.   This continual
increase 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 9.2,
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
                                  9-68

-------
                TABLE 9-24.  MODEL PARAMETER VALUES76,98-i-oc
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

-------
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 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.
                                  9-70

-------
     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 9.3.3.2.
     It is evident from Table 9-25 that plant NPV's become lower as
regulatory 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 are two apparent exceptions to the rule that more stringent
controls result in lower NPV's.  First, 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.  Second, the common control of
emissions in specialty shops yields cost savings that make Model  Plant 10
more profitable under Regulatory Alternative 2 than at baseline.
     Uncontrolled carbon shop (Model Plants 1, 2, 3, 4, 5 and 6)  NPV's
range from a low of $16.629 million to a high of $548.940 million.   When
these same plants are required to meet baseline Alternative 1, NPV's
range from $13.902 million to $535.946 million.   Under proposed Alternative 2
NPV's range from $13.895 million to $535.856 million.   The range  of NPV's
for Alternative 3 is $13.816 million to $534.863 million.
     Uncontrolled specialty steel shop (Model Plants 7, 8, 9 and  10)
NPV's range from $12.003 million to $125.021 million.   When these same
plants are required to meet baseline Alternative 1, NPV's range between
$6.629 million and $111.339 million.  Under proposed Alternative  2, NPV's
are between $6.605 million and $111.340 million.   Under Regulatory Alterna-
tive 3 specialty shop NPV's are between $6.361 million and $110.718 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.
                                  9-71

-------
             TABLE  9-25.   PROJECT  NET  PRESENT VALUES ASSUMING
               6.2  PERCENT WEIGHTED  AVERAGE COST OF CAPITAL
                               (1981 $106)
Model
Plant
1
la
2
3
4
5
6
6
7
8
9
10
No control
16.629
16.629
124.793
539.626
373.956
548.940
19.328
19.328
12.003
12.003
125.021
125.021
Regul
1
(Basel ine)
13.902
13.902
116.975
527.681
361.246
535.946
16.183
16.183
6.629
7.434
110.464
111.339
atory Alternative
2
13.895
NAb
116.937
527.601
361.191
535.856
16.177
NA
6.605
7.433
110.416
111.340

3
13.816
13.816
116.797
526.701
360.854
534.863
16.087
16.204
6.361
7.257
109.866
110.718
a_
.Total  furnace  enclosure  option.
 NA = Not applicable.
                                 9-72

-------
             TABLE 9-26.   NET PRESENT VALUE IMPACTS ASSUMING
              6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL


Regul
atory impact from baseline
Regulatory Alternative
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10

1981
$106
-2.727
-2.727
-7.818
-11.944
-12.710
-12.994
-3.145
-3.145
-5.374
-4.568
-14.557
-13.688
1
Percent
-16.40
-16.40
-6.27
-2.21
-3.40
-2.37
-16.27
-16.27
-44.77
-38.06
-11.64
-10.94
2
1981
$106
-0.007
NAC
-0.038
-0.081
-0.056
-0.091
-0.006
NA
-0.024
-0.002
-0.048
0.002
Percent
-0.05
NAC
-0.03
-0.02
-0.02
-0.02
-0.04
NA
-0.36
-0.03
-0.04
0
1981
$106
-0.086
0.022
-0.178
-0.981
-0.393
-1.083
-0.096
0.021
-0.268
-0.177
-0.598
-0.621
3
Percent
-0.62
0.16
-0.15
-0.19
-0.11
-0.20
-0.60
0.13
-4.04
-2.39
-0.54
-0.56
3T Ł u -i • n 1 rtij. 4. • -i •
Total furnace enclosure option.
NA = Not applicable.
                                9-73

-------
     Moving from no control  to baseline Alternative 1  results in NPV
losses between $2.727 million and $12.994 million for  carbon shops and
between $4.568 million and $14.557 million for specialty shops.   These
losses represent changes from no control of as little  as 2.21 percent for
Model  Plant 3 to as high as  44.77 percent for model plant 7
     Both carbon and specialty EAF steel plant NPVs are little different
from baseline (Regulatory Alternative 1) in absolute terms under Alter-
natives 2 and 3.   The largest NPV change from baseline is the $1.083 million
reduction experienced by Model Plant 5 under Regulatory Alternative 3.
As indicated in Table 9-26,  some plant NPVs 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 carbon 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 cost of capital  of 10.0 percent is used.  Note that the
NPV of Model  Plant 7 is 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 9.3.3.2.
     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 9.3.2.3,
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 emissions control  alternative is attributable to the added
                                  9-74

-------
        TABLE 9-27.   PROJECT NET PRESENT VALUES ASSUMING
          10.0 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
                           (1981 $106)
Regulatory Alternative
Model
Plant
1
la
2
3
4
5
6
6a
7
8
9
10
No control
10.880
10.880
92.403
419.685
285.729
429.916
13.848
13.848
4.699
4.699
84.653
84.653
1
(basel i ne)
8.152
8.152
84.763
408.602
273.570
417.680
10.672
10.672
-0.748
0.139
70.044
71.012
2
8. 149
NAb
84.729
408.535
273.526
417.602
10.674
na
-0.760
0.144
70.002
71.020
3
8.074
8.170
84.581
407.679
273.189
416.646
10.586
10.692
-1.007
-0.032
69.434
70.371
aTotal  furnace enclosure option
 NA = not applicable.
                            9-75

-------
              TABLE  9-28.   NET PRESENT VALUE IMPACTS ASSUMING
               10.0  PERCENT WEIGHTED  AVERAGE COST OF CAPITAL


Regul
atory impact from baseline
Regulatory Alternative
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10

1981
$106
-2.729
-2.729
-7.640
-11.083
-12.159
-12.236
-3.176
-3.176
-5.447
-4.560
-14.610
-13.642
1
Percent
-25.08
-25.08
-8.27
-2.64
-4.26
-2.85
-22.93
-22.93
-115.92
-97.04
-17.26
-16.11
2
1981
$106
-0.002
NAC
-0.034
-0.067
-0.044
-0.078
0.002
NA
-0.012
0.005
-0.041
0.008
Percent
-0.03
NA
-0.04
-0.02
-0.02
-0.02
0.02
NA
1.65
3.69
-0.06
0.01
1981
$106
-0.078
0.018
-0.181
-0.923
-0.381
-1.034
-0.086
0.020
-0.259
-0.170
-0.610
-0.641
3
Percent
-0.96
0.22
-0.21
-0.23
-0.14
-0.25
-0.81
0.19
34.62
122.74
-0.87
-0.90
^Impact of baseline  Regulatory Alternative  1  is  measured from no control.
 Total  furnace enclosure  option.
 NA = not applicable.
                                 9-76

-------
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 $182.42 and $301.32
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
megagram of carbon steel billets can be produced for $118.90 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 $93 to $224.  This difference is the reason that these model
plants all yield positive net present values.
     The 1981 produc-er price for stainless steel billets is $715.18 per
Mg.  The specialty steel shops modeled here, if uncontrolled, can produce
these billets at a cost of $524.22 to $635.17 per Mg.  Again, this spread
of $80 to $191 between price and ATC accounts for the profitabilty of
these plants, as measured by NPV.
     As expected, the ATC of producing steel increases 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
$187.86 to $319.49 per Mg.   The cost per Mg of producing alloy billets is
$545.12 to $670.99 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 $187.90 and $319.54 under
Alternative 2, and between $188.31  and $320.07 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 $319.34 and $304.15, respectively.
                                  9-77

-------
      TABLE 9-29.   AVERAGE TOTAL COST ASSUMING 6,2 PERCENT
                WEIGHTED  AVERAGE COST OF CAPITAL
                           (1981 $/Mg)
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
No control
301.32
301.32
221.55
186.32
197.63
182.42
283.33
283.33
635.17
635.17
524.22
524.22
Regul
1
(basel ine)
319.49
319.49
233.49
191.32
204.92
187.86
304.29
304.29
670.99
665.62
546.45
545.12
atory Alternati
2
319.54
NAC
233.55
191.35
204.96
187.90
304.33
NA
671.15
665.64
546.53
545.11
ve
3
320.07
319.34
233.76
191.73
205.15
188.31
304.93
304.15
672.78
666.81
547.37
546.06
 Model  Plants 1-6 produce carbon billets;  Plants 7-10 produce
^alloy billets.
 Total  furnace enclosure option.
 NA = not applicable.
                            9-78

-------
     The ATC changes associated with moving to more stringent controls
are more explicitly presented in Table 9-30.  Moving from no control to
baseline increases the ATC of production by $5.00 to $20.96 per Mg for
carbon shops and $20.90 to $35.82 per Mg for specialty shops.  These cost
increases represent impacts of 2.7 percent to 7.4 percent for carbon
steel and 4.0 percent to 5.6 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 plants except Model
Plants 7 and 8.  The impacts are likewise small in percentage terms.   The
greatest cost impact is the 0.27 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 differ-
ences, 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 NPV reductions and the potential maximum price impacts 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 9.3.3.6).   This may be an indication that
inaccuracies in the cost data or elsewhere are leading to underestimation
of ATC and a resulting overstatement of project profitabi1ty.  In response
to this objection,  however, it is noteworthy that these are new plants
                                  9-79

-------
             TABLE 9-30.  AVERAGE TOTAL COST IMPACTS ASSUMING
                6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL
                                 (1981 $/Mg)
Regulatory impact from baseline3
Regulatory Alternative
1 2
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
1981
$106
18.18
18.18
11.94
5.00
7.29
5.44
20.96
20.96
35.82
30.45
22.24
20.90
Percent
6.03
6.03
5.39
2.68
3.69
2.98
7.40
7.40
5.64
4.79
4.24
3.99
1981
$106
0.05
NAC
0.06
0.03
•0.03
0.04
0.04
NA
0.16
0.01
0.07
0
Percent
0.02
NA
0.02
0.02
0.02
0.02
0.01
NA
0.02
0
0.01
0
1981
$106
0.57
-0.15
0.27
0.41
0.23
0.45
0.64
-0.14
1.79
1.18
0.91
0.95
3
Percent
0.18
-0.05
0.12
0.21
0.11
0.24
0.21
-0.05
0.27
0.18
. 0.17
0.17
.Impact of baseline  Regulatory  Alternative  1  is  measured  from no  control
 Total  furnace  enclosure  option.
CNA = not applicable.

-------
      TABLE 9-31.   AVERAGE TOTAL COST ASSUMING 1Q.O PERCENT
                WEIGHTED AVERAGE COST OF CAPITAL
                           (1981 $/Mg)
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
No control
320.75
320.75
234.27
190.78
205.57
185.39
295.82
295.82
675.71
675.71
552.21
552.21
Regul
1
(base.l ine)
343.68
343.68
248.98
196.63
214.36
191.84
332.50
332.50
721.46
714.01
580.33
578.47
atory Alternative
2
343.70
NAC
249.04
196.67
214.39
191.88
322.48
NA
721.57
713.97
580.41
578.46

3
344.33
343.53
249.33
197.12
214.63
192.39
323.22
322.33
723.64
715.45
581.51
579.71
aModel  Plants 1-6 produce carbon billets;  Plants  7-10 produce
.alloy billets.
 Total  furnace enclosure option.
CNA = not applicable.
                            9-81

-------
       TABLE  9-32.   AVERAGE  TOTAL COST IMPACTS ASSUMING 10.0 PERCENT
                     WEIGHTED  AVERAGE COST OF CAPITAL


Regul
atory impact from baseline
Regulatory Alternative
Model
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10

1981
$106
22.93
22.93
14.71
5.85
8.79
6.45
26.68
26.68
45.76
38.31
28.13
26.26
1
Percent
7.15
7.15
6.28
3.06
4.28
3.48
9.02
9.02
6.77
5.67
5.09
4.76
2
1981
$106
0.02
NAC
0.06
0.04
0.03
0.04
-0.01
NA
0.10
-0.04
0.08
-0.02
Percent
0.01
NA
0.03
0.02
0.01
0.02
0
NA
0.01
-0.01
0.01
0
1981
$106
0.65
-0.15
0.35
0.49
0.28
0.55
0.73
-0.17
2.18
1.43
1.17
1.23
3
Percent
0.19
-0.04
0.14
0.25
0.13
0.28
0.23
-0.05
0.30
0.20
0.20
0.21
^Impact of baseline  Regulatory  Alternative  1  is  measured from no control.
 Total  furnace  enclosure  option.
 NA = not applicable.
                                 9-82

-------
 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
 indication 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 oerformance 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 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 xlO6 Mg.  Of this total, 101.7 xlO6 Mg
will  be  carbon steel  and 18.0 xlO6 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 xlO6 Mg projection.
                                  9-83

-------
     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 0.03 xlO6 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 xlO6 Mg.  The production impacts of Regulatory Alternatives 2
and 3 are, in general, quite small.   Percentage impacts from baseline
vary from -0.05 to 0 percent for Alternative 2 and from -0.50 to 0.09 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 gain of 1 job to a loss of 173 jobs, depending on the affected
facility.  The impacts of Alternative 3 range from a gain of 325 jobs to
a loss of 1,671 jobs.  Even the greatest impact, that for Model Plant 5
under Alternative 3, represents a loss from baseline employment of only
0.40 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
                                  9-84

-------
          TABLE 9-33.   DOMESTIC STEEL PRODUCTION IMPACTS IN 1987
            ASSUMING 6.2 PERCENT WEIGHTED AVERAGE COST OF CAPITAL


Recjul
atory impact from baseline3
Regulatory Alternative
Model
Plant
1
1b
2
3
4
5
6
6b
7
8
9
10

10b Mg
-11.41
-11.41
-10.20
-5.08
-6.98
-5.64
-14.00
-14.00
-1.89
-1.61
-1.42
-1.33
1
Percent
-10.09
-10.09
-9.11
-4.75
-6.42
-5.25
-12.10
-12.10
-9.49
-8.19
-7.31
-6.90

10b Mg
-0.03
NAC
-0.05
-0.03
-0.03
-0.04
-0.03
NA
-0.01
0
0
0
2
Percent
-0.03
NA
-0.05
-0.03
-0.03
-0.04
-0.03
NA
-0.04
0
-0.02
0

10b Mg
-0.34
0.09
-0.22
-0.41
-0.21
-0.46
-0.40
0.09
-0.09
-0.06
-0.06
-0.06
3
Percent
-0.33
0.09
-0.22
-0.40
-0.20
-0.45
-0.39
0.09
-0.50
-0.33
-0.31
-0.32

"Total  furnace enclosure option.
 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
, Assumi
Impact
CT~J- _ T


Jobs
-41,803
-41,803
-37,348
-18,590
-25,563
-20,655
-51,262
-51,262
-6,916
-5,880
-5,202
-4,889
Impact from
1
Percent
-10.09
-10.09
-9.11
-4.75
-6.42
-5.25
12.10
-12.10
-9.49
-8.19
7.31
-6.90
ng 6.2 percent weighted
of baseline Alternative
baseline of regulatory alternati
2
Jobs
-105
NAd
-173
-122
-108
-140
-95
NA
-29
-2
-16
1
average cost
1 measured

Percent
-0.03
NA
-0.05
-0.03
-0.03
-0.04
-0.03
NA
-0.04
0
-0.02
0
of capi
from no

Jobs
-1,239
325
-809
-1,486
-761
-1 ,671
-1 ,464
320
-327
-218
-205
-213
tal.
control .
ve
3
Percent
-0.33
0.09
-0.22
-0.40
-0.20
-0.45
-0.39
0.09
-0.50
-0.33
-0.31
-0.32

.
NA = not applicable.
                                9-86

-------
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 xlO6 Mg in 1987.   Of this total,  19.3 xlO6 Mg will be carbon
steel and 1.0 xlO6 Mg will be specialty steel.104  The  NSPS induced
percentage change in steel imports estimated as explained above is multi-
plied 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 0.070 xlO6 Mg increase resulting from imposing Regulatory
Alternative 3 on  Model Plant 5.   This represents, however,  only a 0.36 percent
increase.   The greatest impact on specialty  steel imports  results from
imposing Regulatory Alternative 3 on Model  Plant 7.   The 0.40 percent
increase represents increased imports of 0.004 xlO6  Mg of specialty
steel.
9.3.4  Anticipated Economic Impacts
     Section 9.3.3 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
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  9.3.4.1  while
Section 9.3.4.2 deals with the cost-price mechanism.
                                  9-87

-------
               TABLE  9-35.   STEEL IMPORT IMPACTS FOR 1987a


Regul
atory impact from baseline
Regulatory Alternative
Model
Plant
1
1C
2
3
4
5
6
6C
7
8
9
10

10b Mg
1.758
1.758
1.571
0.782
1.075
0.869
2.156
2.156
0.085
0.072
0.064
0.060
1
Percent
10.02
10.02
8.86
4.22
5.90
4.71
12.58
12.58
9.31
7.80
6.84
6.41
2
10b Mg
0.004
NAd
0.007
0.005
0.005
0.006
0.004
NA
0
0
0
0
Percent
0.02
NA
0.04
0.03
0.02
0.03
0.02
NA
0.04
0.03
0.02
0
10b'Mg
0.052
-0.014
0.034
0.063
0.032
0.070
0.062
-0.013
0.004
0.003
0.003
0.003
3
Percent
0.27
-0.07
0.18
0.32
0.17
0.36
0.32
-0.07
0.40
0.27
0.25
0.26
, Assuming 6.2 percent weighted average cost of capital.
clmpact of baseline  Regulatory Alternative 1  is measured from no control.
^Total  furnace enclosure  option.
 NA = not applicable.
                                 9-f

-------
     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
estimated assuming that Model Plants 4 and 7 typify future EAF construc-
tion 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
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 $361.246 million to
$361.191 million under Alternative 2,  or to $360.854 million under
Alternative 3.   These 0.02 and 0.11 percent NPV reductions would be felt
                                  9-89

-------
       TABLE 9-36.   NET PRESENT VALUE AND AVERAGE TOTAL COST DATA
                        FOR MODEL PLANTS  4 AND 7a
Regulatory Alternative
Model
Plant
4
7

NPV
(1981
$106)
361.246
6.629
1
ATC
(1981
$/Mg)
204.92
670.99
2
NPV
(1981
$106)
361.191
6.605
ATC
(1981
$/Mg)
204.96
671.15
NPV
(1981
$106)
360.854
6.361
3
ATC
(1981
$/Mg)
205.15
672.78
Assuming 6.2 percent  weighted  average  cost  of capital.
                                9-90

-------
.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 NPVs would  remain  unchanged
 from  their baseline values as  market price rose  by the same  amount  as
 ATC.  A most relevant question thus  becomes:  what is  the mechanism
 linking ATC  to price and what  will be the result of  this  mechanism  in
 this  circumstance?
      9.3.4.2.1   Analytical framework.   The analytic  framework that  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 determined  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 its  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.   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
                                  9-91

-------
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 wi11 ing 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
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
                                  9-92

-------
           S/Q
          AOC
           $/Q
          ATC
                                                 .Q/time
                                                 .Q/time
                                  [b]
Figure  9-1. Supply schedules for constructed and unconstructed plants
                              9-93

-------
$/Q
 P'
                                     $/Q
  O
                                _Q/time
Q*
                                         n-1 n 2  n-3   n 4   n-5
Q/time
                 la)
                                                      Ib!
                      Figure 9-2.   NSPS effects on market price and quantity.

-------
10
en
        $/Q
Tr
r
i
i
a| 	
1
1
1 n
,-* 1 -
S

.*^^
^^




I
1
1
1
1
1
1
i
                      Q*
Q'
                                   Q/t
           $/Q
Q*
                   $/Q




                     P

                     P*
                                      .Q/t
                                               Q/t
Q*     Q*
                                                        Id)
                                                           le)
                                              Figure 9-2.  Continued.

-------
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
pi ant.
     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
sufficiently 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 uncon-
structed 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  return holds for both existing facilities and facilities
under consideration.  In the latter case,  though, even these costs  are
                                  9-96

-------
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 Q1  - 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 Q1, 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-sloping.   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-ine 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.
                                  9-97

-------
     In a competitive output market,  the prospective investor is a price-
taker;  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 P', 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 P'   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  D',  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 9.3.4.2 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 an 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 P.  Long-run equilibrium will
be restored only when market price falls to P  = 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 .   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, ATC1,  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 P  however, but P1.  Thus, fewer plants will  be constructed before
supply will be expanded to Q1, 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 P  and P'.   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
                                  9-99

-------
O
O
                                                                AOC	
                                                                    I	
                                     [a]
Ibl
                                 Figure 9-3.  NSPS effects on the average total cost of new plants,

                                             equilibrium price and equilibrium quantity.

-------
compensate producers for increased production 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-sloping, long-run supply on the impact estimates.
     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-sloping, long-run market
supply.  Logically, with downward-sloping 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-sloping 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 Alternative 2 on electric arc carbon and specialty
steelmaking.  The  impacts computed are those expected ip 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
                                    Carbon steel
                                  (Model  Plant 4)
                            Alloy steel
                          (Model  Plant 7)
Net present value

Steel  price ($/Mg)

Domestic output (Mg)

Employment (jobs)

Imports (Mg)

Total  cost of regulatory
  alternative ($106)
      0

   0.03  (0.02%)

-30,000  (-0.03%)

   -108  (0.03%)

  5,000  (0.02%)


      0.154
     0

  0.16 (0.02%)

-8,000 (-0.04%)

   -29 (-0.04%)

   358 ( 0.04%)


 0.159
 Model  parameters:   Cost  of  capital              =
                    Own-price  demand  elasticity  =
                    Import elasticity           =
                    1987  baseline  price          =

                    1987  baseline  output         =

                    1987  baseline  employment     =

                    1987  baseline  imports        =

:See Section 9.3.6,  Table 9-41.
                 6.2%
                 -1.86
                 1.51
                 $229.62/Mg  (carbon)
                 $652.70/Mg  (specialty)
                 101.7  xlO6  Mg (carbon)
                 18.0 xlO6 Mg  (specialty)
                 372,527  (carbon)
                 65,934 (specialty)
                 19.3 xlO6 Mg  (carbon)
                 1.0  xlO6 Mg (specialty)
                                 9-102

-------
     There is no anticipated impact on project net present values.  The
proposed standard would have an estimated price impact of $0.03 per Mg
for carbon steel and $0.16 per Mg for specialty steel.  These represent
price impacts of 0.02 percent and 0.02 percent, respectively.  The price
increases would lead consumers to purchase 30,000 fewer Mg of carbon
steel and 8,000 fewer Mg of specialty steel in 1987.   These are reductions
of 0.03 and 0.04 percent from 1987 baseline production.   These output
reductions would lead to fewer employment opportunities in 1987 totaling
approximately 137 jobs — an employment loss of less than one-tenth of
one percent.   The price increase of domestic steel would also cause
domestic users to substitute foreign for domestic steel.   An estimated
5,000 Mg more of carbon steel and 358 Mg more of specialty 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 $0.23 per Mg impact on the
price of carbon steel and a $1.79 per Mg impact on the price of specialty
steel.   Thus, the price of carbon steel  in 1987 would be 0.11 percent
higher than it would be in the absence of this  regulation and the price
of specialty steel  0.27 percent higher.   The price increases would lead
consumers to purchase less domestic steel--208,000 Mg less carbon steel
and 89,000 Mg less specialty steel.   These output effects of 0.20 percent
and 0.50 percent,  respectively," would cause proportionate reductions in
employment opportunities — an estimated 1,088 fewer jobs  in the industry
in 1987.   The higher prices of domestic steel  would also encourage higher
imports.   An estimated 32,000 Mg more of carbon steel and 4,000 Mg more
of specialty steel  would be imported under Regulatory 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
                                    Carbon steel
                                  (Model  Plant 4)
                             Alloy steel
                           (Model  Plant 7)
Net present val ue

Steel  price ($/Mg)

Domestic output (Mg)

Employment (jobs)

Imports (Mg)

Total  cost of regulatory
  alternative ($106)
       0

    0.23  (0.11%)

-208,000  (-0.20%)

    -761  (-0.20%)

  32,000  (0.17%)


   1. 145
      0

   1.79 (0.27%)

-89,000 (-0.50%)

   -327 (-0.50%)

  4,000 ( 0.40%)


  1.621
 Model  parameters:   Cost  of capital              =
                    Own-price  demand  elasticity =
                    Import elasticity           =
                    1987  baseline  price          =

                    1987  baseline  output         =

                    1987  baseline  employment     =

                    1987  baseline  imports        =

DSee Section 9.3.6,  Table 9-42.
                  6.2%
                   1.86
                  1.51
                  $229.62/Mg  (carbon)
                  $652.70/Mg  (specialty)
                  101.7  xlO6  Mg  (carbon)
                  18.0 xlO6 Mg  (specialty)
                  372,527  (carbon)
                  65,934 (specialty)
                  19.3 xlO6 Mg  (carbon)
                  1.0  xlO6 Mg (specialty)
                                 9-104

-------
     To begin, the capital required to build each model 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 18.92 percent for Model Plant 4 to a high of 45.78 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 0.09 percent of the baseline
expenditure.   In dollar terms, the greatest increase is $0.02 million
(Model Plant 5).  Regulatory Alternative 3 also requires little increase
in capital expenditures.   The greatest increase from baseline is $0.51 million
(Model Plant 5).  In percentage terms, Alternative 3 has the greatest
impact on Model Plant 5—2.32 percent.
     Particular attention should be drawn to the incremental capital
requirements for Model Plants 4 and 7 since these projects are utilized
for the anticipated impacts analysis.   No extra capital is required to
build Model Plant 4 under Alternative 2 and $0.21 million more is required
under Alternative 3.   These represent increases from baseline of 0 and
0.53 percent, respectively.  Construction of Model Plant 7 requires
$0.02 million less capital under Iternative 2 and $0.14 more million
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 steelmaking processes.   Each is expected to impose additional
capital costs on the industry.  Taken together, these regulations may
result in some difficulties in obtaining financing for some companies.
Eve.n this is not certain.  Table 9-40 presents the "funded debts to net
                                  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
(1981 $106)
Regulatory Alternative
1
8.71
8.71
24.35
30.46
39.48
22.83
6.56
6.56
18.18
17 43
53.26
52.43
2
8.70
NAC
24.36
30.47
39.48
22.85
6.54
NA
18.16
17,41
53.27
52.41
3
8.74
8.71
24.47
30.90
39.69
23.36
6.59
6.55
18.32
17.52
53.67
52.88
Change from baseline3
(percent)
Regulatory Alternative
1
24.43
24.43
21.75
18.98
18.92
34.29
45.78
45.78
24.52
19.38
21.05
19.16
2
-0.11
NA
0.04
0.03
0
0.09
-0.30
NA
-0.11
-0.11
0.02
-0.04
3
0.34
0
0.49
1.44
0.53
2.32
0.46
-0.15
0.77
0.29
0.77
0.86
.Impact of Regulatory Alternative  1  measured from no control.
 Total  furnace enclosure  option.
 NA = not applicable.
                                 9-106

-------
                TABLE 9-40.   INDUSTRY DEBT RATIOS76
                                                          Debt
          Industry                                        ratio


Blast furnaces and steel mi 11s —SIC 3312                  87.2

Primary nonferrous metals, NEC--SIC 3339                  48.4

Aluminum foundries--SIC 3351                              60.6

Primary metal products, NEC--SIC 3399                     39.5

Motor vehicles--SIC 3711                                  41.8

Petroleum refining--SIC 2911                              61.2


aDebt ratio =    funded debts
              net working capital
                             9-107

-------
working capital"  ratio (debt ratio) for several  industries.   The numerator
of this ratio, funded debts, is all obligations  with maturities exceeding
1  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 working 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.  This 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 environ-
mental benefits resulting from the regulation.
     A simple competitive model of the current domestic steel market is
depicted in Figure 9-4.   The supply curve (S ) from existing firms is
                                            X
upward sloping for the reasons discussed in Section 9.3.4.2.   Current
market price  (P)  and output (Q) prevail  given demand (0).   The market is
in long-run disequilibrium because the average total cost  of production
for new, electric arc furnace plants (ATC.J is below the  prevailing
market price  P.   The long-run steel supply schedule is thus  elastic
(represented  by the horizontal segment S ) 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 (P  ), the  price at which
investors no  longer have  an incentive to enter the  market,  is thus equal
to ATC^ assuming  no changes in technology, factor  prices,  or demand.
                                  9-108

-------
$/Q
                              Q
Qe
                                                         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 fractional  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
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 conven-
tionally  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 startup 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 ATC., 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  P1 and Q'.
     Prior to the standard, total  consumers'  surplus is equal to the
value of  the area P bc--the area below the demand curve and above long-run
equilibrium market price P .   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 P be   P'bd = P P'dc.
                                                        e      e      e e
Consumers now pay more per unit of output (P1    P ) 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

-------

                                              N
                                          j	Q/time
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 Q'-Q'  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 P'  and consumers  reduce their consumption accordingly,
                e     e
     Thus far, a significant portion  (area gfdc) of the total reduction
in consumers'  surplus (area P P'dc) 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 P P'fk.
     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
Q'-Q  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 P P'fk repre-
sents a transfer from consumers' to producers' surplus.   Without the
standard, owners of  old plants enjoy  rents equivalent to area aP k.   With
the standard,  owners of old plants enjoy greater rents--equivalent to
area aPgf   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 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 p1.108  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 dead-
weight 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 P P'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
estimates 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 1  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
(Q ) is based on the steel projections in Section 9.3.5.4.   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 (Q1) results given  a demand elasticity  of -1.86.  The baseline
output in 1987 from plants existing in 1981 (Q ) is based  on the projections
                                  9-113

-------
Section 9.3.5.4.   This output is presumed to be forthcoming given P .   An
estimate of output from existing plants given the higher market price
(P1) due to the proposed standard (Q^) is calculated using a supply
elasticity of 2.16.109
     Using the estimates of Pg]  P^,  QgJ 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:
     1.   New plant real resource cost:
         = area gfdh
         - (Pe-Pe) (Qe-Qd);
     2.   Deadweight loss in consumers' surplus:
         = area hdc
         = h (Pe-Pe) (Qe-Qe);
     3.   Old plant real resource cost:
         = area kfg
         - h (Pe-Pe) (Qo'-Qo);
     4.   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
zero for Model Plant 10 to a high of $0.159  for Model Plant 7.   Estimates
of the total cost of Alternative 2 for carbon shops vary from a low of
$0.154 million for Model Plant 4 to  a high of $0.310 million for Model
Plant 2.  The new plant real  resource cost regularly 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
$1.145 million and $3.059 million, excluding the negative costs for Model
Plants 1 and 6 with a total furnace  enclosure.  Total costs for specialty
steel plants range from $0.855 million to $1.621 million.
     In the very  long run, all plants that are producing steel today will
have been retired.  As output from these existing plants declines, it
will be replaced  by output from new  plants.   It is difficult to speculate
                                  9-114

-------
            TABLE 9-41.   TOTAL COST OF REGULATORY ALTERNATIVE 2

                                 IN 1987a
1981 -$106
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10
New plant real
resource cost
0.257
NAC
0.306
0.154
0.154
0.205
0.206
NA
0.157
0.010
0.070
0
Deadweight
loss in
consumer's surplus
0.001
NA
0.002
0
0
0.001
0.001
NA
0.001
0
0
0
New plant real
resource cost
0.001
NA
0.002
0.001
0
0.001
0.001
NA
0.001
0
0
0
Total
cost of
regulation
0.259
NA
0.310
0.155
0.154
0.207
0.208
NA
0.159
0.010
0.070
0
aModel  parameters:   Q  = 101.7 xlO6 Mg (carbon)
                       =  18.0 xlO6 Mg (specialty)
                    Q° =  96.5 xlO6 Mg (carbon)
,                       =  17.0 xlO6 Mg (specialty)
 Total  furnace enclosure option.
 NA = not applicable.
                                 9-115

-------
         TABLE  9-42.   TOTAL  COST  OF  REGULATORY  ALTERNATIVE  3

                                   IN  1987a
Plant
1
lb
2
3
4
5
6
6b
7
8
9
10

New plant real
resource cost
2.559
-0.738
1.279
1.779
1.095
1.908
2.790
-0.755
1.450
1.027
0.801
0.836
1981.$!
Deadweight
loss in
consumer's surplus
0.097
-0.007
0.030
0.084
0.024
0.104
0.128
-0.006
0.081
0.035
0.027
0.029
O6
New plant real
resource cost
0.105
-0.008
0.032
0.092
0.026
0.113
0. 141
-0.007
0.090
0.041
0.027
0.029

Total
cost of
regulation
2.761
-0.753
1.341
1.955
1. 145
2.125
3.059
-0.768
1.621
1.103
0.855
0.894
aModel  parameters:   Qe  =  101.7  x  106Mg  (carbon)
                       =   18.0  x  106Mg  (specialty)
                    Q   =   96.5  x  106Mg  (carbon)
b                      =   17.0  x  106Mg  (specialty)
 Total  furnace  enclosure  option.
                                 9-116

-------
whether EAF plants will continue in the very long run to be the best
technology.  However, if they do, the new plant real resource cost will
increase by an amount equal to the area P P'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
      e e
very long run = (Pe-Pe) Qe + Jj (Pe-Pe) (Qe-Qe).   This is an underestimate
of the long-run total cost if in fact demand should shift rightward
beyond D 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 $3.051 million to a high of $6.101 million for carbon steel
plants and from zero to $2.879 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 $64.960 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 $0.154 million for carbon plants and $0.159 million for specialty
plants, for a total cost of $0.313 million.   The anticipated 1987 total
cost of Alternative 3 is $1.145 million for carbon plants and $1.621  million
for specialty plants, for a total of $2.766 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
1981
Regulatory
2
5.084
,NAC
6.101
3.051
3.051
4.067
4.067
NA
2.879
0.180
1.260
0
$106
Alternative
3
57.872
-15.197
27.429
41.613
23.367
45.662
64.960
-14.244
32.139
21.205
16.353
17.072
.Based on projected 1987 demand conditions.
 Total furnace enclosure option.
 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
     Manufacturers.  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,  pp.  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.  0.   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.  TO.

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, p. 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.

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 Steelmaking 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-1.
                                  A-l

-------
      TABLE A-l.   EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
Date
Company,  consultant,  or agency/
Location
Nature of action
10/79       U.S.  Environmental  Protection Agency
              Research Traingle Park,  N.C.
04/21/80    U.S.  Environmental  Protection Agency
              Research  Triangle Park,  N.C.
06/13/80    Jones  and  Laughlin  Steel  Corporation
              Pittsburgh,  Pa.
06/23-      Jones  and  Laughlin  Steel  Corporation
  26/80       Pittsburgh,  Pa.

06/25/80    Florida Steel  Corporation
              Charlotte,  N.C.
07/02/80    Allegheny  Ludlum  Steel  Corporation
              Brackenridge,  Pa.
07/02/80    Armco,  Incorporated
              Butler,  Pa.
07/02/80    Washington Steel  Company
              Washington,  Pa.
07/10/80    AL Tech Specialty Steel Corporation
              Watervliet,  N.Y.

07/10/80    Colt Industries,  Incorporated
              Syracuse,  N.Y.
07/15/80    Allegheny  Ludlum  Steel  Corporation
              Brackenridge,  Pa.
07/16/80    Armco,  Incorporated
              Butler,  Pa.

07/22/70    Washington Steel  Company
              Washington,  Pa.

07/23/80    AL Tech Specialty Steel Corporation
              Watervliet,  N.Y.
07/24/80    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 114 letter


                                        Section 114 letter


                                        Section 114 letter

                                        Section 114 letter


                                        Section 114 letter


                                        Site visit


                                        Site visit


                                        Site visit


                                        Site visit

                                        Site visit
                                                               (conti nued)
                                  A-2

-------
                          TABLE A-l.   (continued)
Date
Company,  consultant,  or agency/
Location
Nature of action
08/05/80    Florida Steel  Corporation
              Charlotte, N.C.

08/28/80    American Iron  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    Interlake 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 letter


                                        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,  or 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  Steel  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 1etter

                                        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-l.   (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-1.   (continued)
Date
08/18/81
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
American Iron and Steel Institute
Washington, D.C.
Al Tech Specialty Steel Corporation
Watervliet, N.Y.
Allegheny Ludlum Steel Corporation
Brackenridge, Pa.
American Iron and Steel Institute
Washington, D.C.
Armco, Incorporated
Butler, Pa.
Atlantic Steel Company
Atlanta, Ga.
Bethlehem Steel Corporation
Los Angeles, Calif.
Carpenter Technology Corporation
Reading, Pa.
Chaparral Steel Corporation
Midlothian, Tex.
Eastern Stainless Steel Company
Baltimore, Md.
Sholtes and Koogler
Gainesville, Fla.
Ferrco Engineering, Limited
Whitby, Ontario, Can.
Ford Motor Company
Dearborn, Mich.
Nature of action
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, 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-6

-------
TABLE A-l.   (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
Riverdale, 111

Jones and Laughlin Steel Corporation
Pittsburgh, Pa.

North Star Steel Company
Monroe, Mich.

North Star Steel Company
St. Paul , Minn.

Nucor Steel Corporation
Darlington, S.C.

Obenchain Calumet Corporation
Gary, Ind.

Pennsylvania Engineering Corporation
Pittsburgh, Pa.

Raritan River Steel
Perth Amboy, N.J.

Rob! in Steel Company
North Tonawanda, N.Y.

Union Carbide Corporation
Danbury, Conn.

Union Carbide Corporation
Tarrytown, 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 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
                                     (continued)
        A-7

-------
                          TABLE A-l    (continued)
Date
Company,  consultant,  or agency/
Location
Nature of action
10/13/81     Colt Industries,  Incorporated
              Midland,  Pa.

10/19-      Colt Industries,  Incorporated
  20/81        Midland,  Pa.
10/21-      Roblin 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  Laughlin 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
              Cartersvi1le,  Ga.

03/09-      Jones and  Laughlin Steel  Corporation
  10/82       Pittsburgh, Pa.

04/06/82     Dr.  Sholtes,  Sholtes and Koogler,
              Environmental  Consultants
              Durham,  N.C.
                                        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

-------
                              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
1.
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.
                                                               (conti nued)
                                   5-2

-------
                          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 FOR 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 decarburization
(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-handling
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 carbon 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 (m3/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 3.2:1.2  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

-------
nr/dscf


0.005 .



0.004 -





C.003 -
0
Qi
1—
UJ
C.J
O
t_J

0.002 -




0.001


PLANT
REFERENCE
TO/dscm
-12
KEY
0 TEST RUN
I 	 1 AVERAGE
C\
-10 |
1
|
l '
i
1 '
1 1
1 1
-8 i '
1 '
|!
§
i i i
i! '
ii i
n i
p, y ;
"4 ^ p> A
J-l l! °° '1
1 1 q i i 1
i | P 1 1 i §

i b **~p o i '
co j-t 1 1 ; ! | ]
-> U.1 ^L-L> i
-2 D i'|l
6 oo i
°l 1
O 0
A 3 3 C 0 E ' S H
1 34 5 6 7 S 9 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 participate matter emission tests are
presented in  Tables O4 and O5.   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.I.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 m3/s (91,000 acfm)  flow rate and has a gross air-to-cloth
ratio of 2.69:I.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.
                                  04

-------
     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:I.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 visible 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.
T^is 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.
     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 ms/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 m3/s (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

-------
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 m3/s (740,000 acfm).  The
baghouse has 60 modules and a gross air-to-cloth ratio of 2.33:I.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 Oil.   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.
     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 m3/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 m3/s (345,000 acfm) that vents out
a stack.  This fabric filter has an  air-to-cloth ratio of 4:I.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.
     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 mVs (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 m3/s (700,000 acfm) except during charging
and tapping operations when the control dampers are fully open and the
maximum flow is 533 m3/s (1,130,000 acfm).16  This fabric filter is
                                  C-ll

-------
equipped with Daemon® 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 we!1-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.
     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 m3/s (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.
     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 m3/s (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.
     C.I.1.12  Plant 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 m3/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:I.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 one UHP EAF that has a rated
capacity of 104.3 Mg (115 tons).26  The process emissions from melting
and refining are captured by a DEC system, and the furnace fugitive
emissions are captured by an overhead canopy hood.  All  the furnace
emissions are ducted to a single positive-pressure fabric filter.  The
reverse-air cleaning fabric filter has a design flow rate of 212.4 m3/s
                                  C-14

-------
(450,000 acfm), of which 78 percent is directed to the canopy hood.  The
fabric filter is equipped with Dacron® bags and has a net air-to-cloth
ratio of 3:I.26,27  The pressure drop across the bags is 2.0 to 2.5 kPa
(8 to 10 in.  w.g.).
     The plant was visited to review the 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.
These changes include the installation of scavenger ducts above the
canopy hoods for both the new and old furnaces.   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 will not be used to
support a standard.
     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 not 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 m3/s (525,000 acfm).   The fabric filter is cleaned by
reverse air,  and the net air-to-cloth ratio is 1.95:!.28  The pressure
drop across the bags is maintained between 0.75 and 1.75 kPa (3 and
7 in.  w.g. ).28
     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 maintenance during the visit.   Several broken bags in some of the
baghouse compartments 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-15

-------
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
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).29  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.
                                  C-16

-------
            ar/dscf
mg/dscm

0.003 -

z 0.002 .
o
I— i
i
'Z.
UJ
o
^y
0

0.001 -


PLANT
FURNACE TYPE
REFERENCE
8
O TEST RUN
HH AVERAGE
O
. 6
.«!!
1 1
i
i '
b1
b

i
0
b'
P Q Q R
EAF,AOD ADD ADD EAF,AOD'
29 30 31 32
Figure C-2.   Summary of particulate matter source test data
         for specialty steel  shop fabric filters.
                             C-17

-------
     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.).33
     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, 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, O17,
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).29
     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.34  One emission test was conducted on the
positive-pressure fabric filter  associated with the 18.1-Mg (20-ton) AOD
vessel.31  The AOD vessel emissions are directed to a canopy hood by a
                                  C-18

-------
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).34  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 2.68:1.34  The dirty bags are
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).30  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.
                                  C-19

-------
The results of this test showed a particulate matter concentration of
6.87 mg/dscm (0.003 gr/dscf).34
     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.32  The emissions from both the EAF's and the ADD
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).32  The gross air-to-cloth ratio is
3.26:1.   The 12-compartment fabric filter is equipped with polyester
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).32  A detailed summary
of the inlet test is presented in Table O22.   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).32  The detailed
summary of this test is presented in Table O23.
     C.2.1.4  Plant S.   Plant S operates one 45.4-Mg (50-ton) EAF and
one 45.4-Mg (50-ton) AOD vessel.36  The emissions from  the EAF and 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 m3/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.36  The pressure  drop across the
fabric filter bags is 1.25 to 2.50 kPa (5 to 10 in. w.g.).
                                  C-20

-------
     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.36  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.37  A
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 AOD 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 AOD 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-handling
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.
                                  C-21

-------
English units                 Multiply by               Metric unit

inches, Hg                     3,376.85                 pascals
inches, H20                      248.84                 pascals
cubic feet                       0.0283                 cubic meters

English units                 Multiply by               Metric unit
gr/dscf                           2,290                 mg/dscm
foot                             0.3048                 meters
grains                           0.0648                 grams
                            C-22

-------
TABLE C-l.   SUMMARY OF PLANTS TESTED AND TYPE OF TESTS PERFORMED
Plant
A
B
C
D
E
F
G
H
I
i
rsi i
GO J
K
L
M
N
0
P
Q
R
S
Product grade
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Carbon steel
Specialty steel
Specialty steel
Specialty steel
Specialty steel
No. of
furnaces
2
2
1
1
1
3
1
2
2
2
1
3
2
1
2
2
2
1
2
1
1
1
Furnace/
vessel
EAF
EAF
EAF
UHP EAF
EAF
EAF
EAF
EAF
EAF
UHP EAF
EAF
EAF
EAF
EAF
EAF
EAF
ADD
ADD
EAF
AOD
EAF
AOD
Furnace
Mg
45.4, 68
36.3, 22.7
49.9
108.8
117.9
45.4
108.9
90.7
108.9
317.5
117.9
90.7
204.1
104.3
90.7
29.0
29.0
18.1
38.1
45.4
45.4
45.4
size
tons
50, 75
40, 25
55
120
130
50
120
100
120
350
130
100
225
115
100
32
32
20
42
50
50
50
Type of test
performed
Method 5
Method 5, 9
Method 5, 9
Method 5
Method 5
Method 5, 9
Method 5, 9
Method 5, 9
Method 9
Method 9
Method 9
Method 9
Method 9
Method 9
Method 9
Method 5, 9, 22
Method 5, 9, 22
Method 5, 9
Method 5
Method 9
Method 9

-------
      TABLE C-2.   SUMMARY OF  VISIBLE EMISSION DATA FROM
FABRIC FILTERS  ON  EAF'S  AT CARBON STEEL SHOPS5,7-10,16,18,20

Plant
Ba
Ca
Fa
Gd
Hd
Id
Jd
KC
Length of
observation,
minutes
960
596C
18
936
120
60
577
300
Maximum
6-minute
average
opacity,
percent
2.
0
0
2.
0
0
oe
0
5b


8b




               .Single  stack exhaust.
                Not 6-minute average;  average
                from continuous  monitor during
                Reference  Method 5  test.
                .Combination  of  data from two  tests.
                Monovent exhaust.
                EPA Method 22.
                            C-24

-------
   TABLE C-3.   SUMMARY OF OPACITY DATA FROM SHOP ROOF MONITORS
             AT CARBON STEEL SHOPS5,9,10,16,19,20



Plant
Ca


Ca
Gb
HC





Furnace
EAF


EAF
EAF
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


Charge
Melt
Tap
Charge
Melt
Tap
15
106
10
5
90
5
4.4
4.2
5.0
0
0
0
1.0
0.2
0.6
0
0
0
aUtilizes total furnace enclosure, canopy hood, closed roof.
 Utilizes canopy hood, DEC, closed roof.
^Utilizes canopy hood, DEC, open roof.
 Utilizes canopy hood, DEC, closed roof over furnaces/open
 elsewhere, local tapping hood.
eUtilizes canopy hood, side draft hood, tapping hood, closed
 roof.
                            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
mVmin
acfm
dscm/min
dscfm
dscm/min/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
--
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 PARTICULATE MATTER RESULTS—PLANT A1
                         (Fabric filter outlet)
                                             Run number/date
                             Units
    1          2         3
10/18/72  10/19/72  10/20/72  Average
Test data
Sampling time
Furnace production
Nominal furnace
capacity
Shop effluent
Flow rate
Flow rate per unit
of furnace
min
Mg/h
tons/h
Mg
tons
mVmin
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ton
240
--
113.4
125
6,480
228,978
5,924
209,343
--
240
—
113.4
125
6,353
224,479
5,842
206,431
::
240
_-
113.4
125
6,526
230,592
6,008
212,283
__ —
240
—
113.4
125
6,453
228,016
5,925
209,352
--
   capacity
Temperature

Water vapor
C02 dry
02 dry
Parti cul ate matter emi
Mass collected

Dust concentration

Emission rate

Emission factor

°C
°F
% volume
% volume
% volume
ssions
mg
gr
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
38.9
102
0.6
<0.1
20.6

74.6
1.15
4.81
0.0021
5.9
12.9
--

38.3
101
0.5
<0.1
20.2

35.3
0.54
2.22
0.00097
3.4
7.41
--

37.5
99.5
0.5
<0.1
20.7

36.3
. 0.56
2.29
0.001
4.8
10.5
--

38.2
100.8
0.5
<0.1
20.5

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  06.   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
Parti cul ate matter emi
Concentration



Emission rate

Emission factor

Units

mi n
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
s s i o n s
mg/m3
gr/acf
mg/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
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

1.37
0.0006
1.60
0.0007
0.22
0.49
0.016
0.032
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

1.15
0.0005
1.15
0.0005
0. 17
0.38
0.008
0.015
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

3.89
0.0017
4. 12
0.0018
0.55
1.21
0.035
0.071
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

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 PARTICIPATE 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
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
m3/min
acfm
dscm/min
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 '
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
5.5
0.0024
6.41
0.0028
1.16
2.56
—
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
5.27
0.0023
6.64
0.0029
1.18
2.61
—
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
3.89
0.0017
4.81
0.0021
0.88
1.93
--
Average
240
--
36.3
40
3,633
128,389
3,037
107,315
83.7
2,683

65.5
146.2
0.011
0
4.89
0.0021
5.95
0.0026
1.07
2.37
--
Reference Method 5 data.
                                C-29

-------
       TABLE C-8.   SUMMARY OF  PARTICULATE MATTER RESULTS—PLANT C5
                         (Fabric  filter outlet)3
Run number/date

Test data
Samp! ing 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
Parti cul ate matter emi
Concentration



Emission rate

Emission factor

Units

min
Mg/h
tons/h
Mg
tons

m3/min
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ton

°C
°F
% volume
% volume
% volume
% volume
ssions
mg/m3
gr/acf
m/dscm
gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton
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
__
--
--

2.98
0.00130
3.05
0.00133
0.60
1.324
0.032
0.064
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
—
--
--

2.47
0.00108
2.52
0.00110
0.51
1.121
0.031
0.062
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
--
--
--

2.70
0.00118
2.79
0.00122
0.56
1.227
0.032
0.064
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
--
—
--

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 PARTICULATE MATTEE
                         (Fabric filter outlet)'
RESULTS—PLANT O6

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/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 RESULTSa--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
Particulate matter emi
Concentration



Emission rate

Emission factor

Units

min
Mg/h
tons/h
Mg
tons

m3/min
acfm
dscm/min
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

1 5 , 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
1.8
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 PARTICIPATE 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
02 dry
Parti cul ate emissions
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
m3/min
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
11/17/81
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
11/17/81
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
0.0011
2.71
5.98
0.04
0.08
3
11/18/81
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  PARTICIPATE  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
Particulate emissions
Concentration



Emission rate

Emission factor3

Units

min
Mg/h
tons/h
Mg
tons

m3/mi 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
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
                                   Units
  Run
 number/
  date
   1
08/28/73
Test data
  Sampling time
     240
Furnace production

Nominal 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 factor

Mg/h
tons/h
Mg
tons

m3/min
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
—
--
99.8
110

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 SHOPS29,30,37




Plant
P
Q
s



Length of
observations ,
minutes
438a
408b
336!:
318b
Maximum
6-minute
average
opacity,
percent
0
0
0
0
    ,0ne fabric filter for EAF and ADD vessels.
     AOD vessel fabric filter.
     EAF fabric filter.
                      C-36

-------
            TABLE C-15.   SUMMARY OF VISIBLE EMISSION DATA FROM
             SHOP ROOF MONITORS ON SPECIALTY STEEL SHOPS30,37


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
^Utilizes 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  PARTICULATE  MATTER  RESULTS—PLANT P29
                      (South fabric  filter inlet)
Run number/date

Test data
Sampl ing 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/mi n
acfm
dscm/min
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
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
111
0.0485
52
115
--
""
Average

360
__
—
--
— —

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 PARTICIPATE MATTER RESULTS—PLANT P29
                      (North fabric filter inlet)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
Parti cul ate matter emi
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
mVmin
acfm
dscm/min
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
04/07/81
360
—
--
8,376
295,800
7,800
275,500
--

43.3
no
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
--
--
8,220
290,300
7,580
267,800
__

44
111
0.98
283
0.124
307
0.1343
141
310
—
Reference Method 5 data.
                                C-39

-------
      TABLE  C-18.   SUMMARY  OF  PARTICULATE  MATTER RESULTS—PLANT P29
                         (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

m3/mi n
acfm
dscm/min
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
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
--
— —

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  Q30
                         (Fabric  filter  inlet)
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

m3/min
acfm
dscm/min
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
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
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
127
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 MATTEF
                         (Fabric filter outlet)'
RESULTS—PLANT Q30
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
Concentration



Emission rate

Emission factor

Units

min
Mg/h
tons/h
Mg
tons

m3/min
acfm
dscm/min
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
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
0.493
0.000217 0
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
.000250 0
0.650
.000284 0
0.312
0.687
0.034
0.067
Average

480
9.1
10.1
18.1
20

4,292
151,700
3,810
133,600
210
6,680

53
127
0.37
0.6
19.4

0.583
.000254
0.651
.000289
0.316
0.696
0.035
0.069
Reference Method 5 data.
                                C-42

-------
 TABLE  C-21.   SUMMARY  OF  PARTICULATE  MATTER
                     (Fabric  filter  outlet)'
RESULTS—PLANT Q31

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 emissions
Concentration
Emission rate
Emission factor
Units
min
Mg/h
tons/h
Mg
tons
nrVmin
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
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.
DThis test involved only one run.
                            C-43

-------
      TABLE C-22.   SUMMARY OF PARTICIPATE MATTER RESULTS — PLANT R32
                          (Fabric filter inlet)a
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
dscm/min
dscfm
°C
°F
% volume
% volume
% volume
% volume
emissions
mg/dscm
gr/dscf
kg/h
Ib/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
Reference Method 5 data.
                                C-44

-------
                                     TABLE C-23.    SUMMARY OF  PARTICIPATE MATTER  RESULTS—PLANT  R28

                                                           (Fabric filter  outlet)
o
•
-P>
en

Sampling time
Furnace production

Nominal furnace capacity

Shop effluent
Flow rate



Flow rate per unit, of
furnace capacity
lemperature

Water vapor
CO
02 dry
C02 dry
Participate matter emissions
Concentration



Emission rate

Emission factor

Units
min
Mg/h
tons/h
Hg
tons

nrVmin
acfm
dscm/min
dscfm
dscm/min/Mg
dscfm/ ton
°C
°F
% volume
% volume
% volume
% volume

mg/m3
gr/acf
mg/dscm
.gr/dscf
kg/h
Ib/h
kg/Mg
Ib/ton

1
02/10/78
2
144
37.120
40.926
121.5
134

3,239
114,460
3,152
111,372
--
--
37.9
100.2
0.2
0.0
19.4
1.0

0.389
0.00017
0.389
0.00017
0.41
0.91
--


2
02/11/78
2
144
24.684
27.215
121.6
134

2,722
96,200
2,613
92,348
—
--
37.9
100.2
0.5
0.0
19.4
1.0

0.435
0.00019
0.435
0.00019
0.46
1.02
--

Run number/date
3
02/13/78
2
144
27.188
29.796
121.6
134

1,722
60,845
1,680
59,377
--
--
37.9
100.2
0.0
0.0
18.3
1.8

0.183
0.00008
0. 183
0.00008
0.20
0.43
—

4
02/13/78
8
144
33.367
36.788
121.6
134

2,630
92,933
2,517
88,936
--
--
37.9
100.2
0.3
0.0
19.2
0.8

0.137
0.00006
0.137
0.00006
0.15
0.32
--

568
02/14/78 02/14/78 03/06/78
Compartment number
8
144
39.688
43.757
121.6
134

1,311
46,314
1,244
43,959
--
--
37.9
100.2
0.4
0.0
17.8
2.5

0.275
0.00012
0.275
0.00012
0.29
0.65
--

8
144
28.482
31.402
121.6
134

2,491
88,011
2,333
82,422
--
--
37.9
100.2
0.0
0.0
17.5
2.5

0.412
0.412
0.00018
0.00018
0.44
0.97
--
"
11
144
35.695
39.355
121.6
134

1,740
61,468
1,697
59,940
--
--
37.9
100.2
1.2
0.0
18.3
1.3

0.641
0.00028
0.664
0.00029
0.71
1.56
--
"
9
03/06/78
11
120
40.154
44.271
121.6
134

3,575
126,322
3,460
122,251
--
--
37.9
100.2
1.3
0.0
19.9
0.6

0.939
0.00041
0.962
0.00042
1.03
2.26
—
"
10
03/07/78
11
144
35.932
39.616
121.6
134

1,562
55,212
1,520
53,704
--
--
37.9
100.2
0.9
0.0
20.0
0.2

1.83
0.00080
1.88
0.00082
2.00
4.42
--
""
Average

141.3
33.572
37.014

134

2,332
82,418
2,246
79,369
—
--
37.9
100.2
0.5
0.0
18.9
1.3

0.573
0.00025
0.595
0.00026
0.63
1.39
--
"
       ^Reference Test Method 5 data.  Traversed across top of compartments.

        Twelve-compartment baghouse; three compartments tested.  Flow rate per  unit of furnace production and emission factor not calculated.

-------
  TABLE 024.   SUMMARY OF VISIBLE EMISSION
 DATA FROM DUST-HANDLING SYSTEMS AT EAF AND
 ADD VESSEL STEEL MILL FACILITIES3,29,30,35


Plant
Ba
C
Pb
Qd
Qd,e


Length of
observations
minutes
12
20
594
60
48
Maximum
6-mi nute
average
opacity,
percent
0.6
5.0
0C
0C
7.3
     fabric filter.
°EAF/AOD fabric filter.
 Reference 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.
                  C-46

-------
C.5  REFERENCES FOR APPENDIX C

 1.   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.

 2.   Memo from Terry, W.,  Midwest Research Institute, to Iversen, R.,
     EPA/ISB.  October 31, 1980.  Site visit report--Babcock and Wilcox,
     Beaver Falls, Pennsylvania.

 3.   Compliance Tests Under New Source Performance Standards:  Florida
     Steel Corporation, Charlotte, North Carolina.  Sholtes & Koogler,
     Environmental Consultants.  Gainesville, Florida.  January 1980.

 4.   Addendum No.  1, Compliance Tests Under New Source Performance
     Standards:  Florida Steel Corporation, Charlotte, North Carolina.
     Sholtes & Koogler, Environmental Consultants.  Gainesville, Florida.
     January 1980.

 5.   Source Sampling Report, Hoeganaes, Inc.  (83-00129-01), Gallatin,
     Tennessee.  Environmental Management Planning and Engineering.
     Nashville, Tennessee.  October 1981.

 6.   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.

 7.   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.

 8.   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.

 9.   Compliance Tests Under New Source Performance Standards:   Florida
     Steel Corporation, Tennessee Mill Division, Jackson, Tennessee.
     Sholtes and Koogler.   Gainesville, Florida.  December 1981.

10.   Particulate Emission Tests for Lukens Steel Electric Melt Shop.
     Fuller Company.   Catusauqua, Pennsylvania.  September 23, 1973.

11.   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 Laughlin 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 1,  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—Atl antic
     Steel, Cartersvi11e,  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 Cartersvi1le, 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 Banker, L. , Midwest  Research  Institute,
     to Iversen R. , EPA/ISB.  March 26,  1981.   Source Test Report—Chaparral
     Steel, Midlothian, Texas.

27.  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.

28.  Memo and attachment from Banker, L. , Midwest  Research Institute, to
     Iversen, R. , EPA/ISB.  June  3, 1981.  Source Test  Report—Bethlehem
     Steel, Los Angeles, California-.

29.  Emission Test Report:  AL Tech Specialty Steel Corporation, Watervliet,
     New York.  EPA/EMB No. 80-ELC-7.  July 1981.

30.  Emission Test Report:  Carpenter Technology Corporation,  Reading,
     Pennsylvania.  PEDCo Environmental,  Inc.   Cincinnati, Ohio. Contract
     No. 68-02-3546, Work Assignment No.  2.  July  1981.

31.  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.

32.  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.

33.  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.

34.  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.

35.  Visible Emission Summary'Report for  Carpenter Technology  Corporation,
     Reading, Pennsylvania.  PEDCo Environmental, Inc.  April  1982.

36.  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.

37.  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.
                                  C-49

-------
       APPENDIX D

EMISSION MEASUREMENT AND
 CONTINUOUS MONITORING
           D-l

-------
D.I  EMISSION MEASUREMENT METHODS



     Particulate matter emissions from electric arc furnaces




(EAF's)  and argon-oxygen decarburization (AOD) 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,  because 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 short  (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



emissions.  The last section discusses the selection of methods.



D.I.I  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 1, 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

-------
D
I
                     1.9-2.5 cm
                    (0.75-1 in.)
               1.8 cm (0.75-1 in.)
                                    _ZL
-v

— -1 '.__L ' L^JLT •




"*
THERMOCOUPLE
PROBE
                                             HEATED AREAv        /FILTER HOLDER
                                                                                                   THERMOMETER
                      NOZZLE

               "S" TYPE
                 PITOT
                 TUBE

              THERMOCOUPLE
                                                                               ICE WATER BATH
                                       TEMPERATURE
                                        INDICATOR      THERMOMETERS
             ^-100 ml  OF WATER
                                                                         BY-PASS
                                                                          VALVE
                             VACUUM GAUGE
                                                                                                       VACUUM LINE
                                                                       VACUUM PUMP
                                 Figure  D-l.  Particulate sampling train  at Plant Q outlet.

-------
D
I
CTl
               1.9-2.5 cm
              (0.75-1 irO
         1.8 cm (0.75-1 In.)
    -THERMOCOUPLE
    - PROBE
     PITOT  TUGE
                          STACK WALL


                             U   PROBE
         'S" TYPE
          PITOT
           TUBE
        THERMOCOUPLE
  THERMOMETER
HEATED V
F1LTERV*
  '-"1":
    TEELON
     HOSE
                                TEMPERATURE
                                 INDICATOR
                                 CALIBRATED
                                  ORIFICE
THERMOMETERS
 o
                                 ^..,

                             MANOMETER
                 THERMOMETER





\/


S


ILICA
(ICE
B'ATH
GEL^
                  100 ml. OF WATER
  ^CONTROL
   |^ VALVES
rtxh
   VACUUM\LINE

\VACUUM
                                                                                     +
                         Figure D-2.   Particulate  sampling train at Plant A and Q  inlets.

-------
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.I.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

-------
o
I
oo




1
1
1
1
1
_J
i
1
T^

1
8 1 7
1

-35 4 m (116 ft)



1 1
1 1
| HONOVENT y
i ; \\
• , \
i i
"1 r -| r
OUTLINE OF EXHAUST AREA^/
AT TOP OF COMPARTMENTS
L_ i _, J. ^k.
I H
|
6 i 5
1
COVERED WALKWAY







r H
SAMPLE
POINTS ~
'

2








4 1
-»
3 2


CMPT.
NO. 1
1











1


^•~~-
^^-
A


SEE SE(
IN F1GI


A
^ 1
                                                        TOP VIEW OF FABRIC FILTER
1  ACCESS DOOR PER

   COMPARTMENT
                                                      RUM I - COMPARTMENT NOS.  I  THROUGH 4

                                                      RUN 2 - COMPARTMENT NOS.  3  THROUGH 6

                                                      RUN 3 - COMPARTMENT NOS.  S  THROUGH 8
                                           Figure D-3.    Fabric  filter outlet  at Plant p.

-------
O
I
MAlKMAt
 WITH
ACCESS
 DOOR
                  D6
                             SEE SECTION B-B
                             IN FIGURE 4.3-3
GAS FLOW
      METHOD S
  PROBE AND HEATED
     FILTER BOX
   SUSPENDED FROM
CENTER OF RIDGE BEAM
                                    AND METERING EQUIPMENT
                                                                      SAMPLE
                                                                       LINE
                                                    RAISED
                                                    CENTER
                                              1	~1 WALKWAY
                                                        GRATING LEVEL ABOVE BAGS
                                                       	13.6 m  (44.7 ft)	
                         H • 3.4 m (11  ft)
                         U - WIDTH OF FXHAUST OPENING AT SAMPLING  PLANE • 3.51 m (11.5 ft)
                                                                                           HONOVENT
                                                                                               GAS FLOW
                                            SECTION A-A.  FABRIC FILTER END VIEW FACING SOUTH
                      Figure D-4.   Sampling  location at  the  fabric filter  outlet at  Plant  P.

-------
     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 first 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 par'ticulate 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



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



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. '    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 point's 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 the openings 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 _+ 10 percent).  Several factors and assumptions affected



these isokinetic rate calculations:  the inlet flow equaled the




outlet flow  (previously discussed); the effect of fabric filter




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 EAT and/or ADD 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 other frequent type of PPFF exhaust configuration is

a long, narrow vent.  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.   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.

     3.   The opacity standard is based on emissions from much
          shorter path lengths than the lengths of many roof
          monitors, requiring a correction factor for the COM to
          relate the long path COM reading to the emission
          standard.  The correction factor magnifies the defi-
          ciency described in Item 2.

     4.   Unless the COM measurement beam is enclosed in the
          spaces between multiple emission points, interference
          from the ambient air  (e.g., dust, fog) can indicate
          higher emissions than are actually present in the
          exhaust gas stream.

     Thus, a COM may not be a reliable indicator of good emis-

sions control and proper 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 PPFF's 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  EAT 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 version 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 logs 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  costs  of the monthly operational status  inspections or



once-per-shift parameter logs  can be absorbed by plant operating
*40  CKR 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 two 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 the.re 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 based on comments from several users.






D.3  PERFORMANCE TEST METHODS



     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



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 NSPS 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.
                                D-27

-------
                      REFERENCES
Rossnagel and Associates.  Report of Emission Tests on an
Electric Arc Furnace Baghouse Exhaust at Raritan River
Steel.  Report No. 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.
Prepared 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

-------