&EPA
(MM State
Enrvifcriinmial ProtocSotn
Age**
National Emission Standards for Hazardous
Air Pollutants (NESHAP) for Iron and Steel
Foundries - Background Information for
Proposed Standards
-------
1
v>EPA
(MM State*
Environ msilaJ ProtocSon
National Emission Standards for Hazardous
Air Pollutants (NESHAP) for Iron and Steel
Foundries - Background Information for
Proposed Standards
-------
(This page is intentionally blank)
-------
EPA-453/R-02-013
December 2002
National Emission Standards for Hazardous Air Pollutants (NESHAP) for
Iron and Steel Foundries--
Background Information for Proposed Standards
Prepared by:
RTI International
Research Triangle Park, NC
Prepared for:
Kevin Cavender, Project Leader
Emission Standards Division
Contract No. 68-DO1-73
Work Assignment No. 1-14
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emission Standards Division
Metals Group
Research Triangle Park, NC
-------
Disclaimer
This report has been reviewed by the Emission Standards Division of the Office of Air Quality
Planning and Standards of the United States Environmental Protection Agency and approved for
publication. Mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use. Copies of this report are available through OAQPS
Library Services, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711, or
from the National Technical Information Services, 5285 Port Royal Road, Springfield, VA
22161.
11
-------
TABLE OF CONTENTS
Page
LIST OF TABLES vii
LIST OF FIGURES ix
1.0 INTRODUCTION 1-1
1.1 STATUTORY BASIS 1-1
1.2 SELECTION OF SOURCE CATEGORY 1-2
1.3 HAP HEALTH EFFECTS 1-3
1.4 REFERENCES 1-4
2.0 INDUSTRY DESCRIPTION 2-1
2.1 BACKGROUND 2-2
2.2 INDUSTRY SIZE AND GEOGRAPHIC DISTRIBUTION 2-5
2.3 ECONOMIC TRENDS 2-6
2.4 REFERENCES 2-10
3.0 FOUNDRY PROCESSES AND EMISSIONS 3-1
3.1 GENERAL OPERATIONS 3-1
3.2 PATTERN MAKING 3-3
3.3 MOLD AND CORE MAKING 3-4
3.3.1 Sand Mold and Core Making 3-5
3.3.2 Permanent and Centrifugal Mold Preparation 3-8
3.3.3 Investment Casting Mold Making 3-9
3.3.4 Expendable Pattern Making 3-10
3.4 SCRAP PREPARATION 3-11
3.5 METAL MELTING 3-12
3.5.1 Cupolas 3-14
3.5.1.1 HAP Emissions From Cupolas 3-14
3.5.1.2 Factors Affecting Emissions From Cupolas 3-16
3.5.2 EIFs 3-17
3.5.3 EAFs 3-19
3.6 POURING, COOLING, AND SHAKEOUT 3-22
3.6.1 Sand Casting 3-22
3.6.1.1 HAP Emissions from PCS 3-23
3.6.1.2 Summary of Research Findings on Organic HAPs 3-23
3.6.1.3 Summary of Research Findings on Metal HAP Emissions
from PCS 3-26
3.6.1.4 Factors Affecting HAP Emissions from PCS 3-26
3.6.2 Centrifugal and Permanent Mold Casting 3-27
3.6.3 Investment Casting 3-27
3.6.4 Expendable Pattern Casting 3-28
3.7 SAND HANDLING 3-28
iii
-------
TABLE OF CONTENTS (continued)
3.8 MECHANICAL FINISHING 3-29
3.9 CLEANING AND COATING 3-30
3.10 REFERENCES 3-31
4.0 CONTROL TECHNOLOGY AND PERFORMANCE OF CONTROLS 4-1
4.1 MOLD AND CORE MAKING 4-2
4.2 MOLD AND CORE COATING 4-6
4.3 SCRAP PREPARATION 4-7
4.4 METAL MELTING 4-10
4.4.1 Cupola Controls 4-10
4.4.1.1 Wet Scrubbers 4-11
4.4.1.2 Fabric Filters 4-14
4.4.1.3 Afterburners 4-16
4.4.2 EIF Controls 4-19
4.4.3 EAF Controls 4-26
4.4.4 EAF and EIF Capture Systems 4-28
4.4.4.1 Side Draft Hoods 4-30
4.4.4.2 Direct Evacuation Control (DEC) System 4-33
4.4.4.3 Fume Rings 4-33
4.4.4.4 Close-Fitting Hoods 4-33
4.4.4.5 Canopy Hoods 4-33
4.4.4.6 Total Furnace Enclosure 4-36
4.4.4.7 Building and Bay Evacuation 4-38
4.5 POURING, COOLING, AND SHAKEOUT 4-39
4.6 SUMMARY OF FEDERAL AND STATE REGULATIONS 4-48
4.6.1 PM Emission Limits 4-48
4.6.2 Opacity Emission Limits 4-49
4.6.3 CO Emission Limits 4-49
4.7 REFERENCES 4-50
5.0 BASELINE EMISSIONS AND CONTROL OPTIONS 5-1
5.1 GENERAL APPROACH FOR ESTIMATING HAP EMISSIONS 5-1
5.2 SUMMARY OF EMISSION FACTORS FOR PRIMARY FOUNDRY
OPERATIONS 5-2
5.2.1 Emission Factors for Mold and Core Making and Coating 5-2
5.2.2 Emission Factors for Melting Operations 5-3
5.2.3 Emission Factors for PCS 5-8
5.3 BASELINE EMISSIONS 5-11
5.3.1 Baseline Emissions for Mold and Core Making and Coating 5-13
5.3.2 Baseline Emissions for Melting 5-17
5.3.3 Baseline Emissions for PCS 5-17
5.4 CONTROL OPTIONS 5-20
5.4.1 Control Options for Mold and Core Making and Coating 5-20
5.4.1.1 TEA Scrubber 5-20
iv
-------
TABLE OF CONTENTS (continued)
~Jage
5.4.1.2 Methanol Replacement in Binder Systems 5-20
5.4.1.3 Reduction of Naphthalene Content in Binder System
Formulations 5-22
5.4.1.4 Mold- and Core-Coating Replacements 5-22
5.4.2 Control Options for Melting 5-22
5.4.2.1 Replacement of Cupola Wet Scrubbers with Fabric Filters .. 5-22
5.4.2.2 Afterburners for Cupolas without Afterburning 5-23
5.4.2.3 Fabric Filters for Uncontrolled EIF 5-23
5.4.3 Control Options for PCS 5-23
5.4.3.1 Mold Vent Light-off 5-23
5.4.3.2 Fabric Filters for Uncontrolled Automated PCS Lines 5-24
5.5 REFERENCES 5-24
6.0 CONTROL COSTS 6-1
6.1 GENERAL APPROACH FOR ESTIMATING CONTROL COSTS 6-1
6.2 CUPOLA MELTING FURNACE CONTROL SYSTEMS 6-1
6.2.1 Baghouse Control Costs for Cupola Melting Furnaces 6-2
6.2.2 Venturi Scrubber Control Costs for Cupola Melting Furnaces 6-4
6.2.3 Net Metal HAP Control Cost for Cupola Melting Furnaces 6-4
6.2.4 Sample Calculation of Metal HAP Control Cost for Cupola Melting
Furnaces 6-5
6.2.5 Afterburning Control Cost for Cupola Melting Furnaces 6-5
6.2.6 Sample Calculation of Organic HAP Control Cost for Cupola
Melting Furnaces 6-7
6.3 ELECTRIC INDUCTION, SCRAP PREHEATER, AND POURING
STATION CONTROL SYSTEMS 6-7
6.3.1 Baghouse Control Costs for EIFs and Scrap Preheaters 6-8
6.3.2 Baghouse Control Costs for Pouring Stations 6-11
6.4 MOLD- AND CORE-MAKING CONTROL SYSTEMS 6-13
6.4.1 Acid/Wet Scrubber Control Costs 6-14
6.4.2 Naphthalene-Depleted Solvent Pollution Prevention Costs 6-14
6.5 MONITORING, REPORTING, AND RECORDKEEPING 6-15
6.5.1 Continuous CO Monitoring Systems 6-15
6.5.2 Continuous VOC Monitoring Systems 6-16
6.5.3 Bag Leak Detection Systems 6-16
6.5.4 Parameter Monitoring Systems 6-16
6.5.4.1 Parameter Monitoring Systems for Venturi (PM) Wet
Scrubbers 6-17
6.5.4.2 Parameter Monitoring Systems for Acid/Wet Scrubbing
Systems 6-17
6.5.5 Foundry Recordkeeping, Reporting, and Compliance Costs 6-17
6.5.5.1 Performance Tests 6-18
6.5.5.2 Scrap Selection and Inspection 6-19
6.5.5.3 Low-HAP-Emitting Binder Evaluation 6-19
-------
TABLE OF CONTENTS (continued)
Page
6.5.5.4 Miscellaneous Recordkeeping Costs 6-19
6.6 TOTAL NATIONWIDE COSTS 6-19
6.7 REFERENCES 6-20
VI
-------
LIST OF TABLES
Table Page
2-1 FACILITY SIZE DISTRIBUTION FOR THE IRON AND STEEL FOUNDRY
INDUSTRY 2-7
2-2 PRODUCTION DATA FOR IRON AND STEEL FOUNDRIES BY STATE 2-8
2-3 CAPACITY UTILIZATION PROJECTIONS FOR 1998 2-9
3-1 ANNUAL PRODUCTION BY CASTING OPERATION TYPE 3-4
3-2 REPORTED ANNUAL PRODUCTION BY TYPE OF FURNACE 3-13
3-3 TYPES OF MELTING FURNACES REPORTED BY FOUNDRIES 3-13
3-4 METALS POURED BY TYPE OF CASTING OPERATION 3-22
3-5 EMISSION TEST RESULTS FOR A PRE-PRODUCTION FOUNDRY 3-25
3-6 EMISSION TEST RESULTS FOR A PRODUCTION FOUNDRY 3-25
3-7 MELTING AND COATING CAPACITIES 3-31
4-1 USE OF CONTROLS FOR TEA EMISSIONS FROM COLD-BOX MOLD-
AND CORE-MAKING LINES 4-3
4-2 SOURCE TEST DATA FOR TEA ACID WET SCRUBBERS 4-4
4-3 CONTROL CONFIGURATIONS FOR SCRAP PREHEATERS 4-8
4-4 SPECIFIC CONTROL DEVICES ON SCRAP PREHEATERS 4-9
4-5 CONTROLS FOR MELTING EMISSIONS FROM CUPOLAS 4-11
4-6 PRESSURE DIFFERENTIALS OF VENTURI SCRUBBERS USED ON CUPOLAS 4-12
4-7 A/C RATIOS FORFABRIC FILTERS ON CUPOLAS 4-14
4-8 CUPOLA AFTERBURNER CO OUTLET CONCENTRATION AND EMISSIONS
DATA 4-17
4-9 OPERATING CONDITIONS FOR CUPOLA AFTERBURNERS 4-18
4-10 SOURCE TEST DATA FOR ORGANIC HAP EMISSIONS FROM CUPOLA
AFTERBURNERS 4-19
4-11 CONTROL CONFIGURATIONS 4-21
4-12 SPECIFIC CONTROLS ON EIFs AT IRON FOUNDRIES 4-22
4-13 CONTROL CONFIGURATIONS FOF EIF AT STEEL FOUNDRIES 4-23
4-14 SPECIFIC CONTROLS ON EIFs AT STEEL FOUNDRIES 4-24
4-15 A/C RATIOS FOR FILTERS ON EIF 4-24
4-16 CONTROL CONFIGURATIONS FOR EAFs AT IRON FOUNDRIES 4-26
4-17 SPECIFIC CONTROLS ON EAFs AT IRON FOUNDRIES 4-27
4-18 CONTROL CONFIGURATIONS FOR EAFs AT STEEL FOUNDRIES 4-27
4-19 SPECIFIC CONTROLS ON EAFs AT STEEL FOUNDRIES 4-28
4-20 A/C RATIOS FOR FABRIC FILTERS ON EAFs 4-28
4-21 USE OF CAPTURE SYSTEMS ON EIFs AT IRON AND STEEL FOUNDRIES ... 4-31
4-22 USE OF CAPTURE SYSTEMS ON EAFs AT IRON AND STEEL FOUNDRIES ..4-31
4-23 CONTROL DEVICES USED ON SHAKEOUT STATIONS 4-41
4-24 A/C RATIOS FOR FABRIC FILTERS ON SHAKEOUT STATIONS 4-41
4-25 PRESSURE DROPS FOR WET SCRUBBERS ON SHAKEOUT STATIONS 4-41
4-26 PCS FABRIC FILTER OUTLET CONCENTRATION AND SERVICE DATA .... 4-45
4-27 PCS WET SCRUBBER OUTLET CONCENTRATION AND SERVICE DATA . . . 4-47
5-1 AVERAGE CHEMICAL-TO-SAND RATIOS AND EMISSION FACTORS
USED IN MOLD- AND CORE-MAKING EMISSION ESTIMATES 5-4
vii
-------
LIST OF TABLES (continued)
Table Page
5-2 SUMMARY OF PM EMISSION FACTORS FOR
MELTING FURNACE OPERATIONS 5-6
5-3 IRON FOUNDRY HAP METAL CONTENT OF PM 5-7
5-4 SUMMARY OF ORGANIC HAP EMISSION SOURCE TEST
RESULTS FOR PCS 5-9
5-5 EMISSION FACTORS DEVELOPED FOR PCS LINES ASSOCIATED WITH
EPC OPERATIONS 5-10
5-6 PM EMISSION FACTORS FOR PCS LINES 5-11
5-7 HAP CONTENT OF PM FROM PCS COMPONENTS 5-11
5-8 EMISSIONS FROM MOLD- AND CORE-MAKING LINES AT MAJOR
SOURCE IRON FOUNDRIES 5-14
5-9 ASSIGNED ANNUAL PRODUCTION AND METAL HAP EMISSIONS
FORMODEL MELTING FURNACE 5-18
5-10 EMISSION ESTIMATES FORMODEL PCS LINES 5-19
5-11 SUMMARY OF ENVIRONMENTAL IMPACTS 5-21
6-1 SUMMARY OF CONTROL COSTS FOR BAGHOUSES AND WET
SCRUBBERS: 1998 $ 6-3
6-2 ESTIMATING EXHAUST AIR FLOW RATES FOR CONTROL COSTS
ESTIMATES 6-3
6-3 SUMMARY OF CONTROL COSTS FOR ACID/WET SCRUBBING SYSTEMS:
1998 $ 6-14
6-4 NATIONWIDE COST ESTIMATES FOR IRON FOUNDRY MACT: 1998 $ 6-20
Vlll
-------
LIST OF FIGURES
2-1 Uses of U.S. Cast Metal Products (EPA, 1997) 2-3
2-2 Types of Metal Cast (EPA, 1997) 2-4
3-1 Process Flow for Typical Green Sand Foundry (EPA, 1997a) 3-2
3-2 Conventional and Water-Cooled Cupolas 3-15
3-3 Types of EIFs 3-18
3-4 Side and Top View of an EAF 3-20
4-1 Filterable PM Emissions (gr/dscf) from Wet Scrubbers on Cupolas at Iron Foundries 4-13
4-2 Filterable PM Emissions (gr/dscf) from Fabric Filters on Cupolas at Iron Foundries .4-15
4-3 Filterable PM Emissions (gr/dscf) from Fabric Filters on EIFs at Iron and Steel
Foundries 4-25
4-4 Filterable PM Emissions (gr/dscf) from Fabric Filters on EAFs at Iron and Steel
Foundries 4-29
4-5a Side Draft Hood on EIF 4-32
4-5b Side Draft Hood with Blower on EIF 4-32
4-6 Side Draft Hood on an EAF 4-32
4-7 Direct Evacuation System on an EAF 4-34
4-8 Fume Ring on an EIF 4-35
4-9 Close-Capture Hood System on an EAF 4-35
4-10 Canopy Hood System 4-36
4-11 Schematic of a Total Furnace Enclosure 4-37
4-12 Building Evacuation System 4-38
4-13 Schematic of a Bay Evacuation System 4-39
4-14 Filterable PM Emissions (gr/dscf) from Fabric Filters on PCS at Iron Foundries .... 4-43
4-15 Filterable PM Emissions (gr/dscf) from Scrubbers on PCS at Iron Foundries 4-44
6-1 Control Cost Curves for Cupola Afterburners 6-7
6-2 Control Cost Curves for EIF/Scrap Preheater Baghouses 6-9
6-3 Control Cost Curves for Pouring Station Baghouses 6-12
Appendices
A Iron and Steel Foundries Reporting in the 1998 EPA Survey A-l
B Estimated HAP Emissions From Mold and Core Making Operations B-l
C Development of Emission Factors for Foundry Processes C-l
D Source Test Particulate Matter Data for Cupola Baghouses D-l
E Source Test Particulate Matter Data for Electric Induction Furnace Filters E-l
F Source Test Particulate Matter Data for Electric Arc Furnace Baghouses F-l
G Source Test Particulate Matter Data for Pouring, Cooling and Shakeout G-l
IX
-------
LIST OF ACRONYMS
AB Afterburner
A/C Air-to-cloth [ratio]
AFS American Foundrymen's Society
APCD Air pollution control device
BID Background information document
BLS Bureau of Labor Statistics
CAA Clean Air Act
CAS Chemical Abstract Service
CEMS Continuous emission monitoring system
CERP Casting Emission Reduction Program
CIST Casting Industry Supplier Association
CO2 Carbon dioxide
CO Carbon monoxide
CPMS Continuous parameter monitoring system
CRF Capital recovery factor
DEC Direct evaluation control
D/F Dioxin and furan
DMEA Dimethylethylamine
EAF Electric arc furnace
BID Electric induction furnace
EPA U.S. Environmental Protection Agency
ESP Electrostatic precipitator
FCCU Fluid catalytic cracking unit
HAP Hazardous air pollutant
ICR Information collection request
Ib/ton pound(s) per ton
MACT Maximum achievable control technology
MSDS Material safety data sheet
x
-------
LIST OF ACRONYMS (continued)
NAICS North American Industry Classification System
NESHAP National Emission Standard for Hazardous Air Pollutants
ng/dscm nannograms per dry standard cubic meter
NOX Nitrogen oxide
NSPS New source performance standard
O2 Oxygen
OAQPS Office of Air Quality Planning and Standards
PCS Pouring, cooling, and shakeout
PeCDF Pentachlorinated dibenzofuran
PCDD/PCDF Polychlorinated dibenzo-p-dioxins/Polychlorinated dibenzofurans
PH Preheater
PM Particulate matter
POHC Principal organic hazardous compound
POM Polycyclic organic matter
ppmb Parts per billion by volume
ppmv Parts per million by volume
QA Quality assurance
RATA Relative accuracy test audit
SIC Standard Industrial Classification Code
SO2 Sulfur dioxide
SOX Sulfur oxide(s)
TAG Total annualized cost
TCI Total capital investment
TEA Trimethylamine
TEF Toxic equivalency factor
THC Total hydrocarbon concentration
VOC Volatile organic compound
XI
-------
(This page is intentionally blank)
xn
-------
1.0 INTRODUCTION
This document summarizes the basic background information used in the development of
maximum achievable control technology (MACT) standards for the iron and steel foundries
source category. References and technical memoranda in the docket provide supplementary
information on those steps in the standards development process not covered within this
document.
The balance of this chapter summarizes the statutory basis for MACT standards and the
selection of the source category. Chapter 2 provides an overview of the industry. Chapter 3
discusses foundry production processes in detail and describes hazardous air pollutant (HAP)
emissions from each operation. Emission control technologies and their performance are
summarized in Chapter 4. Chapter 5 presents baseline emissions and control options. Control
costs (for use in estimating potential impacts) and options for emission control and monitoring are
discussed in Chapter 6. Nationwide environmental and energy impacts are estimated in
Chapter 7.
Appendix A lists the foundries reporting in a 1998 survey conducted by the U.S.
Environmental Protection Agency (EPA) to obtain the information needed to develop the
National Emission Standards for Hazardous Air Pollutants (NESHAP) for iron and steel
foundries. Estimated HAP emissions from mold- and core-making operations appear in
Appendix B. Appendices C through G provide detailed information on the development of
9
emission factors for foundry processes and source test data for cupola baghouses, electric
induction furnaces (EIFs), electric arc furnace (EAF) baghouses, and pouring, cooling, and
shakeout (PCS) lines, respectively.
1.1 STATUTORY BASIS
Section 112 of the Clean Air Act (CAA) requires the development of NESHAP for the
control of HAPs from both new and existing major sources or area sources. The statute requires
the standard to reflect the maximum degree of reduction in emissions of HAPs that is achievable,
1-1
-------
taking into consideration the cost of achieving the emission reduction, any non-air-quality health
and environmental reduction, and energy requirements. This level of control is commonly
referred to as MACT.
Reductions in HAP emissions may be achieved by applying a variety of measures,
processes, methods, systems, or techniques, including, but not limited to:
Implementing process changes, substituting materials, or making other
modifications to reduce the volume of, or to eliminate emissions of, such
pollutants;
• Enclosing systems or processes to eliminate emissions;
• Collecting, capturing, or treating such pollutants when released from a process,
stack, storage, or fugitive emissions point;
• Implementing design, equipment, work practice, or operational standards
(including requirements for operator training or certification) as provided in
subsection (h); or
• Employing a combination of the above [section 112(d)(2)].
1.2 SELECTION OF SOURCE CATEGORY
Section 112 specifically directs the EPA to develop a list of all categories of all major and
area sources that emit one or more of the HAPs listed in Section 112(b). The EPA published an
initial list of source categories on July 16, 1992 (57 FR 31576) and may amend the list at any
time. An original schedule for promulgation of standards for each source category was published
on December 3, 1993 (58 FR 63941).
Iron foundries and steel foundries are two of the categories on the initial list. As defined
by the EPA, the iron foundries category consists of plants engaged in producing final shape
castings from grades of iron (EPA, 1992). This source category includes the following
production steps: (1) raw materials handling and preparation, (2) metal melting, (3) mold and
core production, and (4) casting and finishing. The steel foundries category includes any facility
engaged in producing final shape steel castings by the melting, alloying, and molding of pig iron
and steel scrap. This source category also includes raw materials handling, metal melting, mold
and core production, and casting and finishing. Because of the similarity in processes, emissions,
and controls, we are presenting information for these two categories together under the rubric of
"Iron and Steel Foundries."
1-2
-------
The listings for iron foundries and steel foundries were based on the EPA Administrator's
determination that these plants may reasonably be anticipated to emit several of the listed HAPs
in sufficient quantity to be designated as major sources. The EPA schedule for promulgation of
the Section 112 emission standards required MACT rules for the iron and steel foundries source
category be promulgated by November 15,2000. If MACT standards for this source category
were not promulgated by May 15, 2002 (18 months after the promulgation deadline), Section
112(j) required States or local agencies with approved permit programs to issue permits or revise
existing permits containing either an equivalent emission limitation or an alternative emission
limitation for HAP control.
1.3 HAP HEALTH EFFECTS
Several HAPs have been identified that may be present in air emissions in significant
enough quantities to be of concern. The metal HAPs emitted from melting furnaces include
cadmium, chromium, lead, manganese, and nickel. Aromatic organic HAPs produced by mold-
and core-making lines, melting furnaces, and PCS lines contain acetophenone, benzene, cumene,
dibenzofurans, dioxins, naphthalene, phenol, pyrene, toluene, and xylene. The nonaromatic
organic HAPs emitted are formaldehyde, methanol, and triethylamine. The known health effects
of these substances are described in the EPA Health Effects Notebook for Hazardous Air
Pollutants-Draft,(EPA, 1994).
Of the HAPs listed above, benzene is a known human carcinogen of moderate
carcinogenic hazard. Cadmium, 2,3,7,8-TCDD (dioxin), formaldehyde, lead, and nickel are
classified as probable carcinogens. Chromium can exist in two valence states. Chromium VI is a
known human carcinogen of high carcinogenic hazard if inhaled. (Note: Chromium III and
Chromium VI by oral pathways are classified as Group D "not classifiable as to carcinogenicity
in humans.") Acute effects of some of the HAPs listed above include eye, nose, and throat
irritation, nausea, vomiting, drowsiness, dizziness, central nervous system depression, and
unconsciousness. Chronic effects include respiratory symptoms (such as coughing, asthma,
chronic bronchitis, chest wheezing, respiratory distress, altered pulmonary function, and
pulmonary lesions); gastrointestinal irritation; liver injury; and muscular effects. Reproductive
effects include menstrual disorders, reduced incidence of pregnancy, decreased fertility,
impotence, sterility, reduced fetal body weights, growth retardation, slowed postnatal
neurobehavioral development, and spontaneous abortions.
1-3
-------
Note that the degree of adverse effects on health experienced by exposed individuals can
range from mild to severe. The extent and degree to which the health effects may be experienced
depends on:
• Pollutant-specific characteristics (e.g., toxicity, half-life in the environment,
bioaccumulation, and persistence);
The ambient concentrations observed in the area (e.g., as influenced by emission
rates, meteorological conditions, and terrain);
The frequency and duration of exposures; and
Characteristics of exposed individuals (e.g., genetics, age, pre-existing health
conditions, and lifestyle), which vary significantly with the population.
1.4 REFERENCES
U.S. Environmental Protection Agency, 1992. Documentation for Developing the Initial Source
Category List. Office of Air Quality Planning and Standards, Research Triangle Park,
NC. EPA-450/3-92-030.
U.S. Environmental Protection Agency, 1994. EPA Health Effects Notebook for Hazardous Air
Pollutants-Draft. Air Risk Information Support Center, Office of Air Quality Planning
and Standards, Research Triangle Park, NC. EPA-452/D-94-00, PB 95 -503579.
December. Available online at: http://www.epa.gov/ttn/uatw/hapindex.html.
1-4
-------
2.0 INDUSTRY DESCRIPTION
A foundry is a facility that makes metal castings, which are near final shape products that
may be complex in form. A casting is made by pouring molten metal into a cavity that has the
shape of the product. Many metal products can be made much more effectively by casting than
by other methods such as machining or forging.
The metal casting industry makes an enormous variety of products according to end users'
specifications. Castings are used in virtually every industry that is critical to the nation's
economic health and strategic capability. Castings can be made from a wide variety of metals,
including iron, steel, aluminum, brass, bronze, and superalloys. Approximately 3,000 foundries
currently operate in the United States in virtually every State. Of these, approximately 750 pour
iron or steel. More than $25 billion worth of castings, used in 90 percent of manufactured goods,
are produced annually. A primary feed material for foundries is scrap metal, 15 to 20 million tons
of which are consumed annually. The foundry industry is thus a major recycler of primary
metals. The size of foundries varies widely, from facilities that employ more than 1,000 persons,
to those that employ fewer than 10. A significant number of foundries are operated by companies
that employ 500 or fewer persons and are therefore small businesses.
This chapter presents a brief overview of the metal-casting industry, with a specific focus
on the iron and steel foundry sector. Facilities are typically categorized by the type of metal used
in the castings as either ferrous (iron and steel) or nonferrous (e.g., aluminum, copper, zinc, brass,
and bronze). The source categories that are the subject of this report (Iron and Steel Foundries)
are ferrous foundries categorized by the Office of Management and Budget (OMB) under the
general North American Industry Classification System (NAICS) code of 33151. They are also
categorized under the general Standard Industrial Classification (SIC) code of 332. More
specifically, iron and steel foundries are categorized by the NAICS code according to the type of
iron or steel casting operations as follows:
2-1
-------
NAICS 331511 - Iron Foundries,
NAICS 331512 - Steel Investment Foundries, and
NAICS 331513 - Steel Foundries (Except Investment).
The specific SIC codes that apply are:
• SIC 3321 - Gray and Ductile Iron Foundries,
SIC 3322 - Malleable Iron Foundries,
• SIC 3324 - Steel Investment Foundries, and
SIC 3325 - Steel Foundries Not Elsewhere Classified.
2.1 BACKGROUND
The foundry or metal-casting industry is an old industry, with bronze castings dating back
to 3,200 B.C. Iron was discovered in 1,500 B.C., and the first iron casting was made in 600 B.C.
In North America, the first iron foundry started producing castings in 1642. The first steel
castings date back to 500 A.D., but the technology was lost and did not reappear until 1750. The
first U.S. cast steel foundry started production in 1818 (Lessiter and Kotzin, 1996).
About 13 million tons of castings are produced every year in the United States. Most of
these castings are produced from recycled metals. Thousands of different cast metal products are
made, many of which are incorporated into other products. Almost 90 percent of all
manufactured products contain metal castings. It is estimated that, on average, every home
contains over a ton of castings in the form of pipe fittings, plumbing fixtures, hardware, and
furnace and air conditioner parts. Automobiles and other transportation equipment use 50 to 60
percent of all castings produced. Castings for this purpose include engine blocks, crankshafts,
camshafts, cylinder heads, brake drums and calipers, transmission housings, differential casings,
universal joints, suspension parts, flywheels, engine mount brackets, front wheel steering
knuckles, hubs, ship propellers, hydraulic valves, locomotive undercarriages, and railroad car
wheels. The defense industry also uses a large portion of U.S.-produced castings. Typical cast
parts used by the military include tank tracks and turrets and the tail structure of the F-16 fighter.
Other common castings include pipes and pipe fittings, valves, pumps,
2-2
-------
Rail Road
4%
Other Transportation
2%
Industrial Machines
14%
Farm Equipment
7%
Electric Power
4%
Motor Vehicles
31%
Ingot Molds
17%
Construction
4%
Figure 2-1. Uses of U.S. Cast Metal Products (EPA, 1997).
pressure tanks, manhole covers, and cooking utensils (EPA, 1997). Figure 2-1 shows the uses of
various types of castings produced in the United States.
Most foundries manufacture castings for sale to other companies (EPA, 1997). These are
referred to as jobbing foundries. Important exceptions are the relatively few (but large) captive
foundries operated by original equipment manufacturers such as General Motors, Ford, Chrysler,
John Deere, and Caterpillar. Captive foundries account for a large portion of all castings
produced and employ a significant portion of the industry's workforce.
Gray and ductile iron make up almost 75 percent of all ferrous and nonferrous castings by
weight (see Figure 2-2). Gray iron contains carbon in the form of flake graphite and has a lower
ductility than other types of iron. It is used extensively in the agricultural, heavy equipment,
engine, pump, and power transmission industries. Ductile iron has magnesium or cerium added to
change the form of the graphite from flake to nodular, resulting in increased ductility, stiffness,
and tensile strength (EPA, 1997).
2-3
-------
Gray Iron
44%
Ductile Iron
28%
Malleable Iron
2%
Other Nonferrouj
3%
Copper
2%
Aluminum
11%
Figure 2-2. Types of Metal Cast (EPA, 1997).
Malleable iron foundries produce only about 2 percent of all castings. Malleable iron
contains small amounts of carbon, silicon, manganese, phosphorus, sulfur, and metal alloys to
increase strength and endurance. Malleable iron has excellent machinability and a high resistance
to atmospheric corrosion. It is often used in electrical power, conveyor and materials handling,
and railroad industry equipment.
Compared to steel, forms of iron are all relatively inexpensive to produce and easy to
machine, and they are widely used where the superior mechanical properties of steel are not
required (EPA, 1997). Steel castings make up about 10 percent of all ferrous foundry products.
In general, steel castings have better strength, ductility, heat resistance, durability, and weldability
than iron castings. There are a number of different classes of steel castings (based on carbon or
alloy content) with different mechanical properties. A large number of different alloying metals
2-4
-------
can be added to steel to increase its strength, heat resistance, or corrosion resistance (EPA, 1997).
The steel investment casting method produces high-precision castings, usually smaller products.
Examples of steel investment castings range from machine tools and dies to golf club heads.
2.2 INDUSTRY SIZE AND GEOGRAPHIC DISTRIBUTION
According to the 1992 Census of Manufactures, there were approximately 2,813 metal-
casting facilities operating under SIC codes 332 and 336 in that year. The payroll for 1992
totaled $5.7 billion for a workforce of 158,000 employees, and the value of shipments totaled
$18.8 billion (U.S. Census Bureau, 1992). The industry's own estimates of the number of
facilities and employment for 1994 are somewhat higher, at 3,100 facilities employing 250,000.
The industry predicted shipments of 14.5 million tons valued at $28.8 billion in 1999.
The Census of Manufactures data indicate that the industry is labor intensive. The value
of shipments per employee, a measure of labor intensity, is $119,000, which is less than half of
the value for the steel manufacturing industry ($245,000 per employee) and less than 7 percent of
the value for the petroleum refining industry ($1.8 million per employee).
The EPA surveyed the industry in 1998 for information needed to develop the NESHAP
for Iron and Steel Foundries (EPA, 1998a). A comprehensive questionnaire, the MACT
Standards Development Questionnaire for Iron and Steel Foundries, was submitted to
approximately 750 foundries. A total of 595 facilities responded with detailed information on
size and type of operations, number of employees in the facility and its parent company, and
descriptions of air pollution control measures employed, including whatever information was
available on pollutants emitted and control efficiencies. A complete list of facilities submitting
information and a summary description of their type and scope of operation is compiled in
Appendix A.
According to the survey data, most ferrous metal-casting facilities in the United States are
small. About 70 percent of the facilities reporting employed 200 or fewer people (see Table 2-1).
These smaller facilities were generally jobbing foundries. Captive foundries tended to be larger
and to have correspondingly higher production. As seen in Table 2-1, approximately 50 percent
of the ferrous metal castings in the United States are produced by roughly 10 percent of the
facilities, which have the highest number of employees. Smaller facilities also appear more likely
to produce both iron and steel castings than do larger facilities.
2-5
-------
The geographic distribution of the metal-casting industry resembles that of the iron and
steel industry. Historically, locations for metal-casting establishments were selected for their
proximity to raw materials (iron, steel, and other metals), coal, and water (for cooling, processing,
and transportation). Traditional metal-casting regions included the Monongahela River Valley
near Pittsburgh, PA, and along the Mahoning River near Youngstown, OH. The geographic
concentration of the industry is changing as facilities are built where scrap metal and electricity
are available at a reasonable cost and where there are local markets for cast products (EPA,
1997).
A summary of the number of facilities and the metal production rate for each State,
according to the EPA survey of iron and steel foundries, is provided in Table 2-2 (EPA, 1998b).
The top States by number of facilities, in order, are Ohio (61), Wisconsin (55), Pennsylvania (54),
Alabama (39), Indiana (38), Michigan (37), and California (35). Approximately 30 percent of the
iron and steel foundries in the United States are in the top three States (Ohio, Wisconsin, and
Pennsylvania). The top States by annual production rate (tons metal poured) are Ohio (17 percent
of the nationwide total), Wisconsin (14 percent), Michigan (11 percent), Indiana (9.3 percent),
and Alabama (8.5 percent). These five States account for over 60 percent of the iron and steel
castings produced in the United States.
2.3 ECONOMIC TRENDS
Between the 1970s and 1990s, the metal-casting industry suffered a long-term decline in
production. In an 18-year period, the industry witnessed a production decrease of 10.6 million
tons, from a high of 21.9 million tons in 1973 to a low of 11.3 million tons in 1991. In these
years, over 1,000 metal-casting facilities closed due to the loss of the ingot market, which resulted
from rising steel production, the lightening of cars (shift to smaller cars), and product substitution
(use of aluminum castings, plastics, ceramics, and other composites) (EPA, 1997). The metal-
casting industry is now growing at a modest rate for a mature industry. By 2007, shipments are
expected to increase to 16.3 million tons, for an annual growth rate of 1.4 percent. Sales are
projected to grow at a rate of 4.2 percent per year to $38.7 billion, reflecting increased sales of
lighter and higher priced castings (Kirgin, 1998). Sales of aluminum castings are expected to
reach $10.6 billion in 2006, or 29 percent of the total metal-casting revenue.
2-6
-------
TABLE 2-1. FACILITY SIZE DISTRIBUTION FOR THE IRON AND STEEL FOUNDRY INDUSTRY
'<• Number of
employees at
the facilities
1-20
21-50
51-100
101-200
201-300
301-400
401-500
j 501-1,000
1,001 or more
| Not reported
Total
Number
of
facilities
67
124
101
122
56
43
30
38
12
2
595
Percentage
of facilities,
%
11
21
17
20
9
7
5
6
2
0.3
100
Combined annual
production rate,
tons/yr
28,488
239,821
471,640
1,949,896
2,190,761
1,806,949
2,672,478
3,906,555
4,222,334
10,439
17,499,360
Percentage of
annual
production
rate, %
0.2
1.4
2.7
11
13
10
15
22
24
0.06
100
Number of
facilities
casting iron
only
39
64
53
63
39
23
21
26
10
1
339
Number of
facilities
casting steel
only
15
36
31
35
14
13
8
10
2
1
165
Number of
facilities casting
both iron and
steel
13
24
17
24
3
7
1
2 ]
0
0
91
Source: EPA, 1998b.
2-7
-------
TABLE 2-2. PRODUCTION DATA FOR IRON AND STEEL FOUNDRIES BY STATE
State
AL
AR
AZ
CA
i CO
CT
FL
GA
IA
IL
IN
KS
KY
LA
MA
MD
ME
MI
; MN
MO
MS
MT
NC
NE
NH
NJ
NV
1 NY
1 OH
OK
OR
PA
RI
SC
SD
i TN
TX
UT
VA
VT
WA
WI
WV
Totals
Percentage of
No. of facilities facilities, %
39
4
5
35
5
5
6
4
19
21
38
7
2
6
10
3
2
37
15
12
4
1
5
6
5
5
1
7
61
11
9
54
3
8
1
17
33
5
9
3
14
55
3
595
6.6
0.7
0.8
5.9
0.8
0.8
1.0
0.7
3.2
3.5
6.4
1.2
0.3
1.0
1.7
0.5
0.3
6.2
2.5
2.0
0.7
0.2
0.8
1.0
0.8
0.8
0.2
1.2
10.3
1.9
1.5
9.1
0.5
1.3
0.2
2.9
5.6
0.8
1.5
0.5
2.4
9.3
0.3
100.0
Iron,
tons/yr
1,246,361
31,402
86
201,359
30,634
2,480
59,635
156,779
591,608
748,254
1,516,791
96,533
86,021
21,112
41,706
12,503
2,702
1,866,899
164,796
34,097
3,433
1,161
195,598
37,636
1,600
335,058
0
73,396
2,619,454
144,879
17,952
748,062
3,055
88,517
3,975
881,272
496,066
103,694
546,660
22,484
24,235
2,313,289
21,995
15,595,229
Steel,
tons/yr
242,366
0
29,892
53,202
1,400
2,552
836
0
177,307
105,430
105,265
132,506
0
5,534
1,984
1,099
0
63,881
34,590
14,196
28,422
806
0
12,608
14,394
0
730
1,434
305,924
7,717
81,390
144,773
0
4,962
0
17,450
102,783
2,109
517
163
31,854
170,055
4,000
1,904,131
Total metal, Percentage of
tons/yr total, %
1,488,727
3 1 ,402
29,978
254,561
32,035
5,032
60,471
156,779
768,915
853,684
1,622,056
229,039
86,021
26,646
43,690
13,602
2,702
1,930,780
199,385
48,293
31,855
1,967
195,598
50,244
15,994
335,058
730
74,830
2,925,378
152,596
99,342
892,835
3,055
93,479
3,975
898,722
598,848
105,803
547,177
22,647
56,089
2,483,341
25,996
17,499,360
8.5
0.2
0.2
1.5
0.2
0.0
0.3
0.9
4.4
4.9
9.3
1.3
0.5
0.2 ;
0.2 :
0.1
0.02
11 j
l.l ;
0.3 i
0.2
0.01
1.1
0.3
0.1
1.9
0.004
0.4
17
0.9
0.6
5.1
0.02
0.5 i
0.02
5.1
3.4 ;
0.6
3.1
0.1 |
0.3
14
0.1
100.0
Source: EPA, 1998b.
2-8
-------
Increases in production have come primarily from increased capacity utilization at
existing facilities rather than from an increase in the number of facilities. In fact, the American
Foundrymen's Society (AFS) estimates that the number of metal-casting facilities decreased by
over 200 between 1990 and 1994 (EPA, 1997). Table 2-3 shows the projected capacity
utilization estimates for 1998.
TABLE 2-3. CAPACITY UTILIZATION PROJECTIONS FOR 1998
Capacity,
Metal tons/yr Utilization, %
Iron
Steel
. „_____
Aluminum
Copper-base
Magnesium
Zinc
Other nonferrous
Investment casting
Total
12,592,00
1,650,000
2,110,000
400,000
45,000
420,000
50,000
210,000
17,467,000
86
84
84
82
82
88 !
92 |
81
85
Source: Kirgin, 1998.
Ferrous casting shipments, which dropped to their lowest level (9.5 million tons) in 50
years in 1991, were expected to grow in the short term to 11.5 million tons in 1997 and 12.2
million tons in 1998. Shipments of gray iron castings were expected to increase slightly, to 6
million tons in 1997, and then to peak in 1998 and 1999 to annual levels of 6.4 million tons. If
current trends hold, ductile iron is expected to pass gray iron in sales in 2004 and to become the
shipment leader for ferrous metals (Kirgin, 1998).
In 1972, only 5 percent of all castings were aluminum. Today, aluminum accounts for
over 11 percent of the market. Aluminum castings are steadily comprising a larger share of the
market as their use in motor vehicle and engine applications continues to grow. To produce
lighter weight, more fuel-efficient vehicles, the automobile industry is in the process of
redesigning the engine blocks, heads, and other parts of passenger cars and light trucks for
aluminum. Cast aluminum is expected to increase from 140 pounds per vehicle in 1995 to 180
2-9
-------
pounds per vehicle in 2004. This increase is primarily at the expense of gray iron, which will
decrease from 358 pounds per vehicle in 1995 to 215 pounds in 2004 (EPA, 1997).
2.4 REFERENCES
Kirgin, Kenneth H., 1998. "Solid Economy Continues to Fuel Casting Growth," Modern
Casting, American Foundrymen's Society, Des Plaines, IL, vol. 89 (January), No. 1, pp.
30-33.
Lessiter, Michael, J., and Ezra L. Kotzin, 1996. "Timeline of Casting Technology," Modern
Casting, American Foundrymen's Society, Des Plaines, IL, vol. 87 (November) No. 11,
pp. 64-67.
U.S. Census Bureau, 1992. 1992 Census of Manufactures. Industry Series, Ferrous and
Nonferrous Foundries. U.S. Census Bureau, Department of Commerce, Washington,
DC.
U.S. Environmental Protection Agency, 1997. Profile of the Metal-casting Industry. Sector
Notebook Project, Office of Enforcement and Compliance, Washington DC, September.
EPA-310/R-97-004.
U.S. Environmental Protection Agency, 1998a. Detailed Information Collection Request for
Iron Foundries and Steel Foundries Source Category: MACTStandards Development
Questionnaire for Iron and Steel Foundries. Office of Air Quality Planning and
Standards, Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1998b. Compilation of Information from Questionnaire
Forms Submitted by Iron and Steel Foundries to the U.S. EPA Office of Air Quality
Planning and Standards. Office of Air Quality Planning and Standards, Research
Triangle Park, NC.
2-10
-------
3.0 FOUNDRY PROCESSES AND EMISSIONS
This chapter discusses the production processes used at iron and steel foundries and the
HAPs they emit. It also addresses the factors affecting emissions (where information is
available). The chapter first provides a brief overview of the production of ferrous castings, and
then describes the processes and emissions associated with the primary foundry operations,
which include:
• Pattern making;
• Mold and core making;
• Scrap preparation;
• Metal melting;
• Pouring, cooling, and shakeout (PCS);
• Sand handling;
• Mechanical finishing; and
• Cleaning and coating.
3.1 GENERAL OPERATIONS
Figure 3-1 is a diagram of the process flow at a typical ferrous foundry that uses sand
molds to produce castings. It shows potential emission points of HAPs, which are almost
exclusively in the forms of metals or organic compounds, including volatile organic compounds
(VOCs). Some of the operations illustrated in Figure 3-1 are present in all foundries, whereas
others depend largely on the casting specifications and the type of casting process used. The
basic operations in all ferrous foundries are pattern and mold making, metal melting, pouring of
the molten metal into some type of mold, cooling of the casting, and separation of the solid
casting from the mold. Other operations may include scrap preparation, finishing and cleaning,
sand handling, and metallurgical treatment of the molten metal such as nodularization and
inoculation.
3-1
-------
Mak
^ Particulates }
^- — 'l
\ 1
:-up sand Raw Materia|s |nputs
• Binders
I
Sand & Binder
Mixing
/
*
Core Forming
1 -
.
r *
Mniri Mold & Core / " "
Makina Assembly ' Par"cl
Particulates, \ maning J , fum&s
HAPs, VOCs x; J
X ^
""""»
Sand
^ v. Preparation {
'Wetscrubber\ Treatment
vastewaterwth .Lump Knockou
highpH / .screening
\-___-^/^ -Metal Removal
A 'Thermal Treatm
; -Wet Scrubbing
: -Other
f
~ ~ -- " \ smoke '
*
VnCs, HAPs ~* ) .,
. .;•:.
-t- ' ' Metal Melting ' ^««™'<"«. \
A -Cupola Furnace / "^oSen oxides, ^
-'• -Electric-Arc Furnace [ carbon monoxtdes, ,
-Induction Furnace > ^^ide fumes, ,
•Reverberatory Furnace ^ ^Ifurdtoxtde i
•Crucible Furnace XN /
ilates, metal oxide
carbon monoxide,
'OCs, HAPs '
1
~~t"'
Rem<
± ' Particulates \ ]
Sand
^ nnsfinn
Shakeout
ent I
Ricor Piitnff fL
Gate Removal
|
nes and ^^
tais ^ Cleaning, Finishing,
^ & Coating
i
Inspection &
Shipping
A ^^ Slag, a
^^^^ spent refractory11**^
^s. material ^^
r ^ " ~~ ^
/ N
reatment, ' «umgen oxides, ^
P« carbon monoxides,
uross metal oxide fumes, '
'Val v sulfur dioxide !
• ' ^^. /
r -"^^
ross, spent ^*s.
fry material ^^
Particulates >
'• ^
^^ jX"^ Scrap r
: — ?TxL
/ ^
l Particulates, VOCs }
netal, spent tools, ^s.
abrasives .,1^
A f Waste cleaning water witfi\
• ( solvents, oil & grease, )
^^ xuspe
^^^'"^^^
nded solids ^/
/ \
sS Spent solvents, abrasives, N.
x. coatings, wastewater ^S
^^ treatment sludge -^
^^^f Off-spec castings, >^
^^V. packaging materials ^
Figure 3-1. Process Flow for Typical Green Sand Foundry (EPA, 1997a).
3-2
-------
The first step in the production of castings is making a pattern, which is a replica of a
finished casting. Patterns are typically made of metal, wood, or plaster, and are used to create
molds into which molten metal is poured. Molds are made from a variety of materials, including
clay-bonded or chemically bonded sand, metal, or refractory material. A mold gives the casting
its basic exterior shape, while cores are used to form the internal shape, e.g., the cylinders in an
engine block. Cores are made from chemically bonded sand, plaster, collapsible metal, or
soluble salts.
The next step in the production of castings is preparing and melting metal. Foundries
typically use recycled scrap metals as their primary source of metal. They employ metal ingots
as a secondary source when enough scrap is not available or when specifications for the metal
are strict. Scrap metals typically undergo some type of preparation prior to melting such as
sizing, cleaning, and drying.
After preparation is complete, the scrap is charged to a furnace for melting. The molten
metal is poured from the furnace (tapped) into either a holding furnace or a transfer ladle. In
some operations, particularly the production of ductile iron, inoculants and/or nodularization
substances are then added. The molten metal is transported in a ladle, generally by overhead
rail, to the pouring location.
Upon reaching the pouring area, the molten metal is poured into a mold. After the metal
has solidified and cooled, it is separated from the mold. The casting is then transferred to a
finishing and cleaning area. Specific finishing and cleaning operations will vary depending on
the type of metal cast, the type of mold used to produce the casting, and casting specifications.
Finishing typically involves mechanical operations such as abrasive cleaning (shot, sand, or
tumble blast), torch cutoff, air-carbon arc cleaning, chipping, core knockout, and grinding.
Cleaning usually involves the use of organic solvents to remove scale, rust, oxides, oil, grease,
and dirt from the surface of the casting. In addition to finishing and cleaning, some castings may
be given a coating to inhibit oxidation, resist deterioration, or improve appearance.
3.2 PATTERN MAKING
A pattern is a replica of a finished casting. Patterns are used to create hollow molds into
which molten metal is poured. Most patterns are reusable and are typically made of metal,
wood, or plaster. Other patterns are made from wax or polystyrene. These types of patterns are
expendable in that they are used only once per casting.
3-3
-------
Because the level of activity required for making permanent/reusable patterns (i.e, those
made from metal, wood, or plaster) is very low, HAP emissions produced during their
manufacture are not significant. Emissions arise primarily from the use of solvents and are
almost entirely a workplace consideration. Emissions of HAPs from the manufacture of
expendable patterns are discussed in Section 3.3.4.
3.3 MOLD AND CORE MAKING
The predominant casting operations at ferrous foundries include sand mold, centrifugal,
permanent mold, investment, and expendable pattern casting. Sand molds are bonded using
resin-like chemicals or clay plus other materials. Permanent and centrifugal casting operations
use metal molds, and investment casting operations use molds made from refractory material.
Expendable pattern casting uses molds of unconsolidated sand. A variety of cores can be used
with each type of mold. Most cores are made from chemically bonded sand. Others are made
from plaster, collapsible metal, or soluble salts.
The EPA surveyed the iron and steel foundries industry in 1998 when the MACT
Standards Development Questionnaire for Iron and Steel Foundries (EPA, 1998a) was submitted
to approximately 750 foundries. A total of 595 foundries reported their production based on type
of casting operation. Table 3-1 lists the annual production of iron and steel by casting operation
type as reported in responses to the survey (EPA, 1998b).
TABLE 3-1. ANNUAL PRODUCTION BY CASTING OPERATION TYPE
(tons of metal poured)
Casting operation
Sand casting
Green sand molds *
Chemically bonded sand molds *
Total sand
Centrifugal casting*
Permanent casting *
Investment casting
Expendable pattern (lost-foam) casting
Totals
Iron
11,612,540
1,422,995
13,035,535
1,639,963
578,192
679
63,609
15,317,978
Steel
610,244
564,358
1,174,602
48,929
614,930
57,748
7,922
1,904,131
'May involve the use of chemically bonded sand cores.
Source: EPA, 1998b.
3-4
-------
The following sections briefly describe mold-making operations associated with sand
mold, permanent mold, centrifugal, investment, and expendable pattern casting processes. Core-
making processes are also described, but the discussion is limited to cores that use chemically
bonded sand, the most prevalent type of material used to make cores. Potential HAP emissions
associated with each mold- and core-making process are also discussed, as well as factors
affecting these emissions.
3.3.1 Sand Mold and Core Making
In a typical sand-casting line, molding sand is shaped around two pattern halves in metal
boxes, or flasks. The pattern halves are then removed, leaving two mold halves. If the mold
halves are made of chemically bonded sand, additional steps are needed to harden the sand.
Hardening, or curing, occurs through a chemical reaction that takes place at ambient
temperature, at elevated temperature, or by catalysis. After the mold halves are formed, cores, if
used, are placed inside the halves and then the upper half (the cope) and the bottom half (the
drag) are fastened together. A continuous mold-making line operates in a similar manner,
except that the two halves of the mold are joined in a vertical rather than a horizontal plane, and
the molds are assembled in a continuous line without being enclosed by flasks.
To direct molten metal into the mold, vertical channels called sprues are cut into the
mold. Runners connect the sprues to the bottom of the mold cavity. Risers are often cut into the
mold above the cavity to provide a reservoir of molten metal to areas of the casting that solidify
last and also to collect gas and debris such as loose sand.
Most sand molds are made from clay-bonded sand, which is commonly called green
sand. The term "green" denotes the presence of moisture in the molding sand and indicates that
the mold halves are not baked or dried prior to assembly. Green sand consists of approximately
85 to 95 percent sand, 4 to 10 percent bentonite clay, 2 to 5 percent water, and 2 to 10 percent
carbonaceous materials such as seacoal (powdered bituminous coal), petroleum products, corn
starch, or wood flour (EPA, 1997a). The clay and water act as the binder, holding the sand
grains together. Carbonaceous materials reduce mold wall movement and create a reducing
atmosphere that prevents the metal from oxidizing while it solidifies (EPA, 1992).
Carbonaceous materials also facilitate the separation of the mold and the casting, which
promotes good surface finish.
3-5
-------
Some sand molds and virtually all sand cores are made from chemically bonded sand.
Chemical bonding systems work by polymerization reactions that occur at ambient temperature
or are induced by heat or catalysis. The major types of binder systems used for core making are
the oil-bake, shell, hot-box, warm-box, no-bake, and cold-box systems. The major system used
for mold making is the shell system (EPA, 1997a).
The oil-bake system is an older method used to produce cores. The system uses oil and
cereal binders mixed with sand. The core is shaped in a core box and then baked in an oven to
harden it. Oils used can be natural, such as linseed oil, or synthetic resins, such as phenolic
resins. The oil-bake system was used almost exclusively before 1950 but has now been almost
entirely replaced by other chemical binding systems (EPA, 1981).
The shell core system uses sand mixed with synthetic resins and a catalyst. The resins
are typically phenolic or furan resins or mixtures of the two. Often the shell core sand is
purchased as precoated sand. The catalyst is a weak aqueous acid such as ammonium chloride.
The sand mixture is shaped in a heated metal core box. Starting from the heated surface of the
core box, the heat cures the sand mix into a hard mass. When the outside 1/4 to 5/8 inches of
sand has been cured, the core box is inverted, and the uncured sand is poured out, leaving a hard
sand shell behind. The shell is then removed from the core box and allowed to cure for an
additional few minutes, after which it is ready for placement in the mold (EPA, 1997a). The
system has the advantage of using less sand and binders than other systems; however, precoated
sand is more expensive than sand used in other mold-making processes.
The shell mold system is similar to the shell core system, but it is used to construct molds
instead of cores. In this process, metal pattern halves are preheated, coated with a silicone
emulsion release agent, and then covered by the resin-coated sand mixture. The heat from the
pattern halves cures the sand mix, and the mold is removed after the desired thickness of sand is
obtained. The silicone emulsion acts as a mold release, facilitating removal of the shell from the
pattern after curing (EPA, 1997a).
The hot-box system uses sand mixed with a phenolic or furan resin and a weak acid
catalyst. The major difference between this system and the shell system is that the core box is
heated until the entire core solidifies. The system has the advantage of very fast curing times
and a sand mix consistency that allows the core boxes to be filled and packed quickly. The
3-6
-------
system is therefore ideal for automation and the mass production of cores. The disadvantage is
that more sand and binder are used in this system than in the shell core system (EPA, 1997a).
The warm-box system is essentially the same as the hot-box system, but it uses a catalyst
that allows the resin binders to cure at a lower temperature (300 to 400 °F, compared to 450 to
550 °F for the hot-box system). As with the hot-box system, this system uses phenolic and furan
resins. Either copper salts or sulfonic acids are used as catalysts. The advantage of the warm-
box system over the hot-box system is that the former uses less energy for heating, which
translates into lower costs (EPA, 1997a).
The cold-box system is relatively new to the foundry industry. It uses a catalytic gas to
cure the binder at room temperature. A number of different systems are available, including a
phenolic urethane binder with dimethylethylamine or triethylamine gas as the catalyst. Other
systems include a sodium silicate binder with carbon dioxide (CO2) gas as the catalyst and epoxy
or furan binders with sulfur dioxide (SO2) gas as the catalyst. Compared to other chemical
systems, cold-box systems have a short curing time (less than 10 seconds) and therefore are well
suited to mass production techniques (EPA, 1997a). In addition, the absence of costly oven
heating can result in substantial energy savings. Because they are not consumed in the chemical
reactions, the catalytic gases must be collected after they are purged from the core box.
The no-bake or air-set binder system allows curing at room temperature without the use
of reactive gases. The no-bake system uses either acid catalysts or esters to cure the binder. The
acid catalysts are typically benzene, toluene, or sulfonic or phosphoric acids. Binders are either
phenolic resins, furan resins, sodium silicate solution, or alkyd urethanes. This type of system
has the advantage of substantial savings in energy costs, but it typically requires more curing
time than the other systems (EPA, 1997a).
Green sand mold making is not a source of significant HAP emissions because the
process does not involve heating or curing. In chemically bonded sand mold- and core-making
processes, however, the mixing and curing of the binder may generate substantial HAP
emissions. The potential for HAP emissions varies between binder systems, depending on the
amounts of HAP used in the formulation and the extent to which they react in the curing process.
For example, certain HAPs, such as methylene diphenyl diisocyanate (CAS No. 101-68-8),
phenol and formaldehyde in the original binder formulations, are polymerized during the
reaction process. Other HAP chemicals may be present as solvents, stabilizing agents, or
3-7
-------
catalysts that do not participate in the polymerization reaction. Portions of these chemicals
evaporate during the mold- and core-making process; the unevaporated portion remains in the
chemically bonded mold or core sand. This remaining portion may either (1) pyrolyze as molten
metal is poured into the mold (true particularly for chemical near the inner mold surface); (2)
evaporate during cast cooling as the temperature of the mold sand further from the molten metal
increases; or (3) evaporate during shakeout and subsequent sand handling, when a greater
surface area of sand is exposed to the atmosphere.
After chemically bonded molds and cores are cured, they are often coated with a finely
ground refractory material to provide a smoother surface finish on the casting. The refractory
material is applied as a slurry. After coating, the liquid component of the slurry is either allowed
to evaporate or, if it is a flammable substance such as alcohol, is eliminated by ignition (the
light-off process).
Little information is available on emissions from the mold- and core-coating process. If
molds or cores are coated after forming and curing, the liquid component of the slurry will either
evaporate or be destroyed, to some extent, by incineration, if the light-off procedure is used.
Most coatings used by foundries do not contain HAPs because they are either water based or
isopropanol based. One HAP commonly used in coating slurries is methanol. If the coating is
simply dried, all of the HAP in the liquid will be emitted. If the coating is ignited, emissions will
be reduced by the amount of HAP consumed by the light-off flame unless other HAPs are
generated as combustion products.
3.3.2 Permanent and Centrifugal Mold Preparation
Permanent mold and centrifugal casting operations use reusable molds made from cast
iron, graphite, or steel. Although the molds eventually deteriorate, they can be used to make
thousands of castings before being replaced. These operations may also incorporate sand cores.
The amount of sand used, however, is small compared with the amount used in a sand mold with
the same amount of metal poured. Permanent molds offer advantages over sand molds,
including a more uniform shape, a higher degree of dimensional accuracy, and a more consistent
quality of finish on the castings. The process though is more expensive than using sand molds
and is generally not employed for very large castings. Some of the largest steel foundries use
this process to make castings for the railway industry, such as wheels for railcars.
3-8
-------
In centrifugal casting, the molten metal is introduced into a mold that is rotated during
solidification of the casting. The centrifugal force shapes and feeds the molten metal as it is
forced into the designed crevices and details of the mold. This process is ideally suited to the
casting of cylindrical shapes such as pipe.
Metal molds undergo specific preparation steps prior to pouring, including an initial
cleaning of the mold followed by preheating and the spraying or brushing on of a mold coating.
Coatings are typically mixtures of sodium silicate and either vermiculite, talc, clay or bentonite
(EPA, 1997a). They may also consist of acetylene soot. The coatings insulate the molten metal
from the relatively cool, heat-conducting mold. This allows the mold to be filled completely
before the metal begins to solidify. The coatings also help produce a good surface finish, act as a
lubricant to facilitate casting removal, and allow any air in the mold to escape via space between
the mold and coating.
Emissions of HAPs from permanent and centrifugal mold-making preparation come
primarily from the making of sand cores where used. These cores are often made using the shell
system described previously. Materials used to coat permanent and centrifugal molds generally
do not contain significant amounts of HAPs.
3.3.3 Investment Casting Mold Making
The investment casting process uses a pattern around which a mold made of a refractory
material is formed. Pattern materials are most commonly wax or polystyrene. Waxes can be
synthetic, natural, or a combination of materials.
The mold-making process begins with the production of the patterns, which are usually
mass produced by injecting liquid or semiliquid wax or plastic into a die (a metal mold).
Multiple patterns are attached to a gating system (a sprue and runners) constructed of the same
material to form a tree assembly. The assembly is coated with a specially formulated heat-
resistant refractory slurry mixture that is allowed to harden around the assembly, thus forming
the mold (EPA, 1991 a).
In the flask molding method, the assembly is placed in a flask and then covered with a
refractory slurry, which is allowed to harden. In the more common shell method, the assembly is
dipped in a refractory slurry, then coarser grained refractory is sifted onto the slurry-coated
pattern assembly and the slurry is allowed to harden. This two-step process is repeated until the
desired shell thickness is reached (EPA, 1997a). In both methods, the wax assembly is then
3-9
-------
melted out of the shell and the shell is subsequently heated to remove any residual pattern
material and to further cure the binder system. The shell is then ready for the pouring of molten
metal into the central sprue; the metal is channeled through the runners into the individual molds.
Although usually not necessary, cores can be used in investment casting for complex
interior shapes. The cores are inserted during the pattern-making step. The cores are placed in
the pattern die and pattern wax or plastic is injected around the cores. After the pattern is
removed from the die, the cores are removed. Cores used in investment casting are typically
collapsible metal assemblies or soluble salt materials, the latter of which are leached out with
water or a dilute solution of hydrochloric acid or citric acid.
The refractory slurries used in investment casting are comprised of binders and refractory
materials. Refractory materials include silica, aluminum silicates, zircon, and alumina. Binders
include silica sols (very small silica particles suspended in water), hydrolyzed ethyl silicate,
sodium and potassium silicate, and gypsum type plasters. Ethyl silicate is typically hydrolyzed
at the foundry by adding alcohol, water, and hydrochloric acid to the ethyl silicate as a catalyst
(EPA, 1997a).
Emissions of HAPs from investment casting include polystyrene vapors from the melting
of wax in making patterns, pyrolysis products of wax formed during pattern meltout and shell
curing, and hydrochloric acid fumes emitted from core-leaching operations.
Vapors from wax melting and acid leaching are, at most, a workplace consideration.
Emissions from meltout are commonly incinerated by an afterburner. Wax remaining in the
shell after meltout is about 20 percent of the total at most and may typically be less than 10
percent. Limited data show that emissions of paraffin are less than 0.1 percent of the wax input
to the furnace, and emissions of particulate (not characterized) may be as high as 1.5 percent
(Investment Casting Institute, 1995). One foundry estimates its annual emission of VOCs at 0.05
tons and its annual emission of particulate matter at 0.07 tons. Emissions from this process
therefore do not appear to be significant enough to warrant further consideration.
3.3.4 Expendable Pattern Making
Expendable pattern casting, also called the lost-foam process, is a relatively new process
that is gaining increased use. A one-piece expendable pattern is made by assembling
polystyrene forms, which are made from polystyrene beads blown into a cast aluminum mold
and consolidated by heating. The mold for the casting is created by placing the pattern into a
3-10
-------
container, pouring sand around the pattern, and compacting the sand by vibrating the container.
When hot metal is poured into the mold, it replaces the foam and creates a casting of the same
shape. The foam is converted into vapor, which escapes through the sand. Emissions of HAPs
from expendable pattern making consist of polystyrene vapors. Expendable patterns are
typically made outside the foundry and therefore do not constitute a source of foundry emissions.
3.4 SCRAP PREPARATION
Foundries use recycled scrap metals as their primary source of metal and resort to metal
ingots as a secondary source when scrap is not available. Scrap metals typically require some
type of preparation prior to melting such as cutting or sizing, shot or sand blasting to remove
coatings, cleaning with organic solvents to remove oils and grease, and drying. The degree of
scrap preparation is generally dictated by furnace type. Most cupolas and electric arc furnaces
(EAFs) require minimal scrap preparation (typically only sizing) (EPA, 1981). The presence of
water or oil can cause an explosion in electric induction furnaces (EIFs); therefore, scrap is
frequently cleaned and preheated before being charged to these furnaces. Oily or wet scrap does
not cause explosions in cupolas or EAFs (EPA, 1981). A total of 117 foundries responding to
the 1998 industry survey reported using preheaters, and all of these foundries used EIFs. Most
(111 of 117) used EIFs exclusively as their melting furnace. The six other foundries used EIFs
in conjunction with EAF or cupola melting furnaces. However, these foundries generally
indicated in their survey responses that the scrap preheater was specifically associated with the
EIF (EPA, 1998b).
The use of scrap preheaters is tied not only to the type of furnace used. It also depends
on the type of scrap metal processed. Approximately 98 percent of the foundries that reported
using preheaters in 1998 melted iron. Roughly 80 percent of all iron melted in EIFs was first
preheated, whereas only about 10 percent of the steel melted in EIFs was first preheated (EPA,
1998b).
Mechanical processes associated with scrap preheaters (e.g., loading of scrap) generate
particulate matter (PM) emissions that are of concern only in the work area. Scrap preheating
itself can produce both PM and organic emissions. Over 90 percent of preheaters are direct-fired
with natural gas. Metal HAP content of the PM is expected to be a function of the composition
of the scrap. Data presented by Shaw (1982) indicate that manganese was the major HAP from
preheaters used in iron foundries in the early 1980s. Shaw reported that manganese was about 2
3-11
-------
percent of the PM from a top-firing preheater and 0.1 percent from a bottom-firing preheater.
The only other HAP metal reported at a significant level was chromium at 0.5 percent of the PM
from bottom firing and 0.1 percent from top firing. Organic HAP emissions, which arise from
oil and grease contaminants, are assumed to include products of incomplete combustion, but
these organic HAP emissions have not been characterized.
3.5 METAL MELTING
Table 3-2 shows the annual production of iron and steel reported in the 1998 survey
(EPA, 1998b). While the number of EIFs was nine times greater than the number of cupolas and
eight times greater than the number of EAFs, approximately 60 percent of the total annual
production of iron was melted in cupolas. Approximately 83 percent of the total annual
production of steel was melted in EAFs. Based on the survey responses, foundries that
exclusively used furnaces other than cupolas or EAFs tended to be smaller in terms of tons of
metal melted. Large production iron foundries typically melted with cupolas, while large
production steel foundries typically melted with EAFs.
Table 3-3 lists the usage of melting furnaces reported at ferrous foundries (EPA, 1998b).
The types of melting furnaces were, in decreasing order of number of foundries using them: EIFs
(445 foundries), cupolas (111 foundries), EAFs (81 foundries), reverberatory furnaces (5
foundries), crucible furnaces (2 foundries), and electrical resistance furnaces (2 foundries). A
total of 594 foundries identified melting furnace type; 545 (92 percent) of these foundries used
only one type of furnace. All of the remaining 49 foundries used only two types.
In addition to melting furnaces, ferrous foundries also used holding furnaces and
duplexing furnaces. A holding furnace is an EAF or EIF used to maintain the molten metal in
the proper condition until the foundry is ready to pour. A duplexing furnace is used in malleable
iron production to increase the temperature of the metal in the absence of slag. Duplexing is
necessary when a cupola is used as the primary melting unit.
The following sections briefly describe the predominant types of melting furnaces at
ferrous foundries (cupola, electric induction, and electric arc) and emissions associated with each
type.
3-12
-------
TABLE 3-2. REPORTED ANNUAL PRODUCTION BY TYPE OF FURNACE
| Type of melting furnace
Cupola
Electric induction
Electric arc
Other
Total
Number of
furnaces
143
1,397
163
15
1,718
Production, tons/yr
Iron
9,175,505
5,564,270
571,525
6,679
15,317,979
Steel
10,298
319,998
1,573,441
394
1,904,131
Total metal
9,185,803
5,884,268
2,144,966
7,073
17,222,110
Percentage of annual production, %
Iron
59.9
36.3
3.7
0.04
100
Steel
0.5
16.8
82.6
0.02
100
Total metal
53.3
34.2
12.5
0.04
100
Source: EPA, 1998b.
TABLE 3-3. TYPES OF MELTING FURNACES REPORTED BY FOUNDRIES
Type(s) of melting furnaces
Cupola only
Electric arc only
Electric induction only
Other furnace type (reverberatory; crucible) only
Total number of foundries with one furnace type
Electric induction and cupola
Electric induction and electric arc
Electric induction and other
I Total number of foundries with multiple furnace types
Total number of foundries
Number of foundries with furnace type(s)
Iron only
foundries
93
3
216
6
318
16
3
9
21
339
Steel only
foundries
0
38
104
1
143
0
20
2
22
165
Iron and steel
foundries
1
8
74
0
83
0
8
0
8
91
Total No. of
foundries
94
49
394
7
544
16
31
4
51
595
Source: EPA, 1998b.
3-13
-------
3.5.1 Cupolas
Figure 3-2 shows a schematic of a cupola. The cupola is a hollow vertical refractory-
lined or water-cooled steel cylinder. Hinged doors at the bottom allow the furnace to be emptied
when not in use. When charging the furnace, the doors are closed and a bed of sand is placed at
the bottom of the furnace, covering the doors. A charge consisting of coke for fuel, scrap metal,
alloying materials, and flux is the loaded into the furnace.
Flux, often chloride or fluoride salts, is added to the furnace to remove impurities. The
flux unites with impurities to form dross or slag, which rises to the surface of the molten metal
and helps to prevent oxidation of the metal. The presence of coke in the melting process raises
the carbon content of the metal to the casting specifications. Heat from the burning coke melts
the scrap metal and flux, which drip to the bottom. A hole that is level with the top of the sand
bed allows molten metal to be drawn off, or tapped. A higher hole allows slag to be tapped.
Additional charge is added as needed (EPA, 1997a).
3.5.1.1 HAP Emissions From Cupolas. While emission factors for PM (all three
furnace types) and VOCs (EAF only) are well documented (EPA, 1995), little information is
available for HAP emissions. To partially fill this information gap, in 1997 the EPA conducted
source tests on exhaust gases from two cupolas, one controlled by a baghouse and one by a wet
scrubber. Both cupolas were also equipped with afterburners, used primarily to combust carbon
monoxide (CO), a major reaction product of burning coke. The afterburners also serve to
incinerate other organic emissions such as products of incomplete combustion of oil and grease
contaminants on the scrap metal. In both tests, PM and HAP metals were measured at the
control device inlets and outlets using EPA Method 29. Semivolatile HAPs were measured at the
outlets using SW-846 Methods 0010 and 8270, and dioxin and furan (D/F) emissions were
measured at the outlets using Method 23. Also, volatile organic HAPs were measured at the wet
scrubber outlet using a direct interface gas chromatography/mass spectroscopy method.
A summary of the test on the cupola controlled by a baghouse can be found in the EPA's
published report on the test (EPA, 1997b). Prior to entering the baghouse, the exhaust passed
through a solids settler, afterburner, and heat recuperator. At the baghouse inlet, metal HAPs
were on average 4.08 percent of the PM, for which the average mass flow rate was 322 pounds
per hour (Ib/hr) and the PM emission factor was 7.26 Ibs/ton of metal melted. Manganese and
lead represented 51 and 47 percent, respectively, of the metal HAP content. The total of
3-14
-------
Skip-hoist rail
Brick
Cast iron lining
Charging door
Wind box
Stack
•» Refractory lining
Stack
Blast duct
Skip-hoist rail
Brick lining.
Cast iron lining
Charging door
\Werouflet.
Steel outer shd
Sted inner shell
Witer inlet
Inn trough
•Taphole
Sand bed
Door
Charging deck
Water flow between
1 inner and outer shell
Water cooled
tuyere
Carbon block
Sand bed
Door
CONVENTIONAL CUPOLA
Prop
WA1ERCOOLED CUPOLA
Figure 3-2. Conventional and Water-Cooled Cupolas.
semivolatile HAPs, of which only acetophenone, bis(2-ethylhexyl)phthalate, naphthalene,
phenol, and 2,4,6-trichlorophenol were detected in amounts above the quantitative limit, was on
average 0.00311 Ib/hr at the baghouse outlet. The total D/F at the baghouse outlet was on
average 275 micrograms of toxic equivalency (|ig TEQ) per hour.
A summary of the test on the cupola controlled by a wet scrubber can be found in the
EPA's published report on the test (EPA, 1997c). Prior to entering the scrubber, the cupola
exhaust passed through an afterburner, recuperator, and quencher. At the scrubber inlet, metal
3-15
-------
HAPs were, on average, 5.69 percent of the PM, for which the mass flow rate was 123 Ib/hr and
the PM emission factor was 4.16 Ib/ton. Manganese and lead represented, on average,
83 percent and 16 percent, respectively, of the metal HAP content. No volatile organic HAPs
and no significant quantities of semivolatile organic HAPs were measured at the outlet. The
total D/F at the scrubber outlet was on average 18.8 ng TEQ per hour.
Cowen (undated) indicated that manganese ranged from 1 to 2 percent of the cupola
baghouse catch, and the EPA (1990) reported manganese as the major HAP at 4.5 percent of the
catch. Data for total metal HAP included 5.2 percent (EPA, 1990) and 6.5 percent (Euvrard and
Jackson, 1992) of the PM.
Measurements of poly cyclic organic matter (POM) after controls were reported as
0.0035 Ib/ton of metal melted, and D/F measurements (after a baghouse) were on the order of
10"7to 10"10 Ib/ton (Emcom, 1990; Normandeau Associates, Inc., 1992). Benzene (after an
electrostatic precipitator [ESP]) was 0.0003 Ib/ton. Baldwin (1982) reported 0.18 Ib/ton of total
organics from a baghouse controlling a cupola. Information was unavailable on the HAP content
of the organics.
3.5.1.2 Factors Affecting Emissions From Cupolas. Organic vapors from cupolas
vary with the oil and grease content of the scrap and with the efficiency of afterburning.
Particulate emissions will vary according to the type of coke burned, type of metal melted,
melting temperature, and a number of operating practices. The following factors affect
particulate (and thus metal HAP) emissions from cupolas (EPA, 1981):
• Unlined furnaces generally have higher emissions than lined furnaces.
• Screening charge materials and other precautions to limit the amount of loose
sand, rust, and coke fines charged to the furnace result in a 40- to 60-percent
reduction in emissions.
The use of briquettes and oily scrap increases emissions.
• Oxygen enrichment increases the PM concentration, but any increase in emissions
may be offset by shorter melting times.
• Melting metal at a higher rate produces a higher loading of fine particles.
• An increase in the blast rate increases emissions.
A major factor in reducing organic substance emissions is the use of efficient
afterburning. When properly designed and operated, afterburners provide high destruction
efficiencies (typically more than 99 percent) of organic compounds (EPA, 1991).
3-16
-------
In summary, emissions of HAP metal compounds, particularly manganese and lead, from
cupolas can be substantial. Emissions of organic HAPs from cupolas with efficient afterburners
appear to be low.
3.5.2 EIFs
An EIF operates by passing an electric current through a coil either around or below the
main body of the furnace. Furnaces with the coil around the furnace body are called coreless
induction furnaces, and those with the coil below the body are called channel induction furnaces.
Both types are shown in Figure 3-3. An alternating electric current through the coils generates
an alternating magnetic field, which in turn creates a current in the metal charge. The metal is
melted by resistance heating produced by the current. Consequently, EIFs may also be referred
to as electric resistance furnaces. The coils carrying the electric current are typically cooled with
water. An EIF requires cleaner scrap input than EAF, but an EIF can make more precise
adjustments to the metallurgical properties of the metal (EPA, 1997a).
The coreless furnace can melt cold charges; however, most foundries maintain a heel of
molten metal in the furnace to increase efficiency and to lower thermal shock to the refractory
lining. Power inputs for furnaces used in the foundry industry range from less than 100 to
17,000 kilowatts for coreless furnaces and up to 4,000 kilowatts for channel furnaces. As
previously noted, the presence of water or oil can cause an explosion in EIFs, therefore, metal
scrap (specifically iron scrap) is frequently preheated to drive off these substances before being
charged to these furnaces.
Shaw (1982) provides a review of emission test results for uncontrolled PM emissions
from EIFs. The review includes numerous studies that were performed to characterize these
emissions and the factors that affect them. The emission factors ranged from 0.26 to 1.5 Ib/ton,
and most were in the range of 0.26 to 0.77 Ib/ton.
Another study (EPA, 1981) reported a range of 0.12 to 1.5 Ib/ton, with a best estimate of
1.0 Ib/ton. These emission factors include emissions from charging, melting, superheating, and
pouring, and in some cases, emissions from nodularization using magnesium alloy. One of the
studies found that 45 percent of the emissions came from melting, 25 percent from charging, and
30 percent from pouring and slagging. The AP-42 (EPA, 1995) emission factor is given as 0.9
Ib/ton for uncontrolled emissions and 0.2 Ib/ton after baghouse control.
3-17
-------
molten
iron
cooling
air
or water
Corelesslnduction furnace
molten
iron
air
cooled
inductor
power
coils
crucible
wall
refractory
lining
crucible
wall
refractory
lining
melting
channels
Channel induction furnace
Figure 3-3. Types of EIFs.
The EPA (1990) reported that manganese was the major metal HAP in dust from EIFs, at
1.7 percent, followed by lead, at 0.5 percent, with total metal HAPs at 2.4 percent. Shaw (1982)
reported that EIFs producing malleable and ductile iron generated dust that contained chromium
(0.5 to 0.75 percent) and manganese (0.5 percent), with total metal HAPs at 1.1 to 1.3 percent
(nickel, lead, and cobalt were also detected). The dust generated by EIFs melting steel is
expected to contain more HAP metals than the dust from iron melting because of the use of
alloys, especially for stainless steel.
3-18
-------
The following factors have been found to affect particulate emissions from EIFs (EPA,
1981):
• Whether or not the metal scrap is preheated prior to charging to the furnace (cold
charging produces more emissions than hot charging);
• The type of metal used (nodular iron produces greater emissions than malleable
iron); and
• The presence of alloying metals such as chromium and nickel (PM emissions will
contain higher fractions of these metals).
3.5.3 EAFs
Figure 3-4 shows a schematic of an EAF. EAFs are used almost exclusively to melt
steel. The EAF is a refractory-lined cylindrical vessel made of heavy welded steel plates and
having a bowl-shaped hearth and a domed-shaped roof. For alternating current furnaces, three
graphite electrodes are mounted on a superstructure above the furnace and can be raised and
lowered through holes in the furnace roof. A direct current furnace uses only one electrode and
provides stable electrical current to the metal scrap with less electrode consumption. Of the 168
EAFs described in EPA's 1998 questionnaire, 139 were alternating current and 25 were direct
current furnaces; 4 furnaces were unspecified (EPA, 1998b).
Melting is accomplished in EAFs by heat from direct radiation from arcs formed between
the electrodes of the furnace and the metallic charge, by direct radiation from the furnace lining,
and by the resistance of metal between the arc paths. Metal-melting operations in an EAF may
include: (1) furnace charging, in which metal, scrap, alloys, carbon, and flux are added to the
furnace; (2) melting, during which the furnace remains closed; (3) backcharging, which is the
addition of more metal and possibly alloys after the initial charge is melted; (4) refining by
single-slag (oxidizing) or double-slag (oxidizing and reducing) operations; (5) oxygen lancing by
injecting oxygen into the molten steel to adjust the chemistry of the metal and speed up the melt;
and (6) tapping the molten metal into a ladle or directly into molds. Raw materials may be
charged to an EAF by removing the roof and adding the materials via a bucket suspended from
an overhead crane, through a chute opening in the roof, or through a door in the side of the
furnace.
3-19
-------
direct evacuation
system
furnace shell
rocker tilt
tilt cylinder
graphite electrodes
(during charging)
molten steel
outlet for
tapping
teeming ladle
water cooled
roof
working •--•"
platform
roof
suspension
beam
water cooled
cables
power conducting arms
Figure 3-4. Side and Top View of an EAF.
3-20
-------
In steel foundries, PM may contain varying amounts of metal HAPs such as zinc, lead,
nickel, cadmium, and chromium. Carbon steel dust can be high in zinc as a result of the use of
galvanized scrap, while stainless steel dust is high in nickel and chromium. Painted scrap can
result in PM high in lead.
Test data for HAP emissions from a baghouse on an EAF showed total metal HAP
emissions at 0.0047 Ib/ton, with the major HAPs as lead (0.0029 Ib/ton) and manganese
(0.00066 Ib/ton) (Ecoserve, Inc., 1990). HAP metals reported in lower concentrations included
antimony, arsenic, cadmium, chromium, mercury, nickel, and selenium.
Calspan Corporation (1978) reported manganese as the major HAP in dust from an EAF
melting steel, at 3.4 to 4.3 percent of the dust, with lower levels of lead (0.8 to 2.4 percent) and
chromium (0.1 to 0.27 percent). The sum of the metal HAPs ranged from 4.5 to 6.8 percent.
The EPA (1990) has also reported manganese as the major HAP (8.7 percent) with total metal
HAPs at 12.3 percent for an EAF melting steel. Bates and Scheel (1974) reported that the fumes
from five alloy and stainless steel heats in an EAF contained 8.8 percent chromium compounds
(Cr2O3) and 2.8 percent manganese compounds (MnO), with total metal HAPs at 12.2 percent.
The analysis of dust at three other plants showed manganese to be the major HAP, with a range
from 0.2 to 5.0 percent, followed by chromium at 0.1 to 3 percent. Total metal HAPs were
found to range up to 3.6,4.8, and 9 percent for the three plants (Bates and Scheel, 1974).
In 1980, the EPA (1980b) presented data for dust analyses from EAFs melting iron.
Manganese was the major metal HAP at 2 percent, and total metal HAPs for three foundries
were 2.5,3, and 4.2 percent, respectively. Other metal HAPs included lead, nickel, and
chromium. These data on dust analyses indicate that the composition of EAF dust is affected by
the type of metal that is produced, with higher HAP concentrations reported for steel than for
iron, especially alloy and stainless steel.
No information was found for organic HAP emissions from EAFs. Baldwin (1982)
reported 0.35 Ib/ton of total organics from the baghouse of an EAF, but information was
unavailable on the HAP content of the organics. The EPA (1993) reports total VOC from an
EAF as 0.06 to 0.3 Ib/ton, but again there is no indication of the HAP component.
The following factors have been found to affect particulate emissions from EAF:
• Emissions are higher from scrap that contains oil, oxidation (rust), and sand
particles from casting returns (EPA, 1981).
3-21
-------
• Oxygen lancing, used for adjustment of chemistry, speeding up of the melting
process, and superheating, increases emissions (EPA, 1981).
• Alloys such as chrome may be added to the furnace just prior to tapping. The
addition of the alloys increases particulate emissions during tapping (EPA, 1983).
• Backcharging produces a large eruption of fumes with a strong upward thermal
driving force. The emissions during backcharging are higher than during
charging due to the heat of the molten steel bath in the furnace (EPA, 1983).
• Adding raw materials to the furnace by removing the roof generates more
emissions than adding the materials through a chute or side door (EPA, 1983).
3.6 POURING, COOLING, AND SHAKEOUT
According to the 1998 survey, most metal is poured into sand molds as shown in
Table 3-4, which lists the amounts of metal poured in the different types of casting lines.
Pouring operations vary widely depending upon the type of mold used and the degree of
mechanization in a particular foundry.
TABLE 3-4. METALS POURED BY TYPE OF CASTING OPERATION
Casting Operation Metal Poured, tons/yr
Sand Casting
Green sand molds * 12,222,784
Chemically bonded sand molds * 1,987,353
Centrifugal Casting * 1,688,892
Permanent Casting * 1,193,122
Investment Casting 58,428
Expendable Pattern (lost-foam) casting 71,531
*May involve the use of chemically bonded sand cores.
Source: EPA, 1998b.
3.6.1 Sand Casting
The two principal types of pouring operations are (1) floor or pit pouring, in which ladles
are moved to stationary molds, and (2) pouring stations, in which the ladle is held at one place
and the molds are moved to the station on conveyors. After molten metal has been ladled into
the mold and begins to solidify, the molds are transported to a cooling area where the casting
3-22
-------
solidifies before being separated from the mold. Larger, more mechanized foundries generally
use automatic conveyor systems to transfer the casting and mold through a cooling tunnel on the
way to the shakeout area, where castings are separated from sand or refractory molds. Less
mechanized foundries and foundries that produce very large castings allow the castings to cool
on the shop floor. In the shakeout area, molds are typically placed on vibrating grids or
conveyors to shake the sand loose from the casting. In some foundries, the mold may be
separated from the casting manually (EPA, 1981).
3.6.1.1 HAP Emissions from PCS. The majority of HAP emissions from PCS
operations are organic HAP emissions created by incomplete combustion of organic material in
the mold and core sand. Metal fume emissions may occur during pouring, and PM, primarily
sand with trace amounts of metal HAPs, is emitted during shakeout. Results of investigations on
organic and metal HAP emissions are discussed in the following sections.
3.6.1.2 Summary of Research Findings on Organic HAPs. In one laboratory
experiment (Scott, Bates, and James, 1976), researchers analyzed pouring and cooling emissions
during the manufacture of 40-kilogram castings. Metal was poured into 12 different types of
sand molds, most of which were made using chemical bonding systems. Emission sampling was
started approximately 1 minute after pouring and continued for 1 hour. Measured HAP included
hydrogen cyanide, formaldehyde, acrolein, C2-C5 aldehydes, benzene, toluene, xylene,
naphthalene, and phenol. (Note: Other HAPs may have been present that were not analyzed.)
Prior work by these researchers showed that hydrocarbon emissions peaked approximately 6
minutes after pouring; a second peak occurred during shakeout (castings were cooled for
approximately 25 minutes prior to shakeout). The results suggest that organic emissions during
shakeout can be of the same order of magnitude as those generated shortly after pouring. During
shakeout, hot metal and sand contact cooler sand that contains binder material and other
organics, which can result in additional volatilization and thermal decomposition.
Baldwin (1979) measured total organics at an operating foundry using green sand molds
with phenolic isocyanate cores and phenol-formaldehyde shell molds. The resulting emission
factors for total organics were 0.14 Ib/ton of metal poured from pouring and cooling combined,
1.2 Ib/ton from shakeout before a scrubber, and 1.0 Ib/ton after a scrubber (Baldwin, 1979; EPA,
1980a).
3-23
-------
Euvrard and Jackson (1992) reported measurements of total organic emissions from a
gray iron foundry using green sand molds with various types of cores (oil, phenolic isocyanate,
phenolic ester, furan hot box). Based on their data, the emission factor for total volatile and
semivolatile organic emissions from PCS was 0.37 Ib/ton, of which 0.32 Ib/ton were HAPs. The
single major HAP detected was benzene at 0.2 Ib/ton. The organic emissions measured after the
shakeout scrubber were about twice the emissions measured during the pouring and cooling
steps.
Wingra Associates (1992) report emissions of organic HAPs from PCS at two foundries
using green sand molds and various types of chemically bonded sand cores. Total organic HAP
emission factors for PCS combined ranged from 0.42 to 1.6 Ib/ton at the two foundries.
Emission factors for the single major HAP (benzene at one plant and acrolein at another) ranged
from 0.13 to 1.6 Ib/ton.
The results obtained by Baldwin and by Euvrard and Jackson suggest that volatile
organics are the primary component of the organic emissions from pouring and cooling and that
semivolatiles (e.g., polycyclic aromatic hydrocarbons) are minor constituents. These higher
boiling organic compounds may tend to condense on the sand as they migrate through the mold
or core. However, as the mold is broken up in the shakeout process, the potential for
semivolatile HAP emissions increases. Recent test data submitted with the survey responses
indicated that emissions of methylnaphthalene (a semivolatile HAP) accounted for two-thirds of
all HAP emissions from shakeout (EPA, 1998b).
The Casting Emission Reduction Program (CERP), a cooperative initiative involving
several industry and government stakeholders to reduce air emissions and improve casting
efficiency, performed testing in a "pre-production" foundry to measure emissions from PCS
(CERP, 1999). The pre-production foundry is a general purpose manual foundry that has been
adapted and instrumented to allow the measurement of emissions using EPA protocols. The
report cautioned that the results are not suitable for use as general emission factors; however, the
test results (summarized in Table 3-5) are consistent with those described earlier. The
background baseline results with no known organics in the molds or cores show HAP emissions
that are over an order of magnitude less than those when organics are present. The green sand
baseline and core baseline both show about the same level of HAP emissions (0.32 Ib/ton), even
though there was over 20 times more mold sand than core sand. When organics are present in
3-24
-------
both the mold and core sand (green sand and core baseline), the results are roughly the sum of
what was emitted during the green sand baseline and core baseline.
CHRP also performed testing in a production foundry to measure emissions from PCS
(CERP, 2000). Green sand molds with seacoal and phenolic urethane cold-box cores were used.
The ratio of mold sand to metal was 8.3, and the ratio of core sand to metal was 0.36. Results
are summarized in Table 3-6 and are consistent with other reported values.
TABLE 3-5. EMISSION TEST RESULTS FOR A PRE-PRODUCTION FOUNDRY
Parameter
Description
Mold sand-to-
metal ratio
Core sand-to-
metal ratio
SumofHAPs
(Ib/ton of
metal)
Benzene
(Ib/ton of
metal)
Background
baseline
Molds and cores
with no known
organics
6.0
0.27
0.025
0.006
Green sand
baseline
Green sand mold *
and core with no
known organics
5.7
0.27
0.32
0.12
Core baseline
Molds with no
known organics and
phenolic urethane
cold-box core
5.4
0.24
0.32
0.14
Green sand and
core baseline
Green sand mold *
and phenolic
urethane cold-box
core
5.5
0.26
0.54
0.22
' Contained seacoal.
Source: CERP, 1999.
TABLE 3-6. EMISSION TEST RESULTS FOR
A PRODUCTION FOUNDRY
Analyte
Sum of HAP
Sum of POM*
Benzene
Toluene
PCS emissions, Ib/ton of metal
0.49
0.06
0.23
0.072
* Polycyclic organic matter (primarily naphthalene, 1- and 2-methylnaphthalene).
Source: CERP, 2000.
3-25
-------
3.6.1.3 Summary of Research Findings on Metal HAP Emissions from PCS. The
available references provided little data on the composition of PM from PCS. Because of a large
proportion of sand, emissions from shakeout are expected to contain a lower percentage of metal
HAPs than the levels found in dust from melting furnaces. In one study, manganese was found
at a level of 2.9 percent of the PM in one of two plants, but there was no indication whether the
manganese resulted primarily from pouring or shakeout emissions (Wingra Associates, 1992).
Potas and Blair (1993) reported manganese as the major HAP in dust captured by a
baghouse; their analysis of captured dust (probably largely sand) indicated that the metal HAP
comprised only a fraction of a percent of the PM. Measurements reported at a gray iron foundry
indicated that manganese was the major metal HAP from pouring and cooling (shakeout
emissions were not analyzed), and that total metal HAPs were 3 percent of the PM (Euvrard,
1992).
Although limited, the above data indicate that metal HAPs comprise a few percent of the
total PM from pouring and cooling and that the percentage is likely to be much less for shakeout
emissions.
3.6.1.4 Factors Affecting HAP Emissions from PCS. Emissions from PCS are
expected to be affected by factors such as the composition of molds and cores, mold size, sand-
to-metal ratio, surface area of the sand/metal interface, metal temperature, pouring rate, and
cooling rate. Benzene emissions, in particular, are a byproduct of the decomposition of seacoal
and seacoal supplements in green sand molds; supplements include anthracite, gilsonite,
causticized lignite, and ground coke. In one research study, the amount of benzene emitted
during the pouring and subsequent cooling of molten metal into green sand molds was found to
be directly proportional to the volatile matter content of the seacoal and seacoal supplements in
the mold (LaFay and Neltner, 1998). This study also found that for a fixed casting weight, the
quantity of benzene emitted due to decomposition of seacoal and seacoal supplements decreased
as the sand-to-metal ratio increased. Considering the large number of factors that affect
emissions, the available data on organic emissions from pouring, cooling, and shakeout are very
limited.
3-26
-------
3.6.2 Centrifugal and Permanent Mold Casting
Centrifugal castings are cylindrically symmetric shapes such as pipe that are made by
pouring metal into a mold that is spun about its axis to hold the metal against its walls by
centrifugal force. Permanent molds are completely filled with metal so that no motion is
necessary; the metal is kept in place by gravity and pressure. Both types of molds are typically
water cooled.
Upon solidifying, the metal shrinks slightly, facilitating separation of the casting from the
mold. Little mechanical finishing is required for these types of castings because they can be
produced with the desired surface finish and with minimal or no separation of sprues, runners,
and risers required.
Emissions from centrifugal casting have not been measured, but they are assumed to
consist of metal fumes and some organic compounds that arise from the sand cores normally
used. The amount of sand used in these types of casting is much less than that used in sand
casting. For centrifugal and permanent mold casting, the sand-to-metal ratio by weight is 0.05 or
less (EPA, 1998b; Ductile Iron Pipe Research Association, 2000). By contrast, the ratio is
typically 4 to 5 for sand casting (EPA, 1998b). However, the sand used in centrifugal and
permanent mold casting is primarily for cores with chemical binders, whereas most sand casting
uses green sand molds.
As shown in Table 3-5, cores with chemical binders can emit as much HAP as green sand
molds, even when the amount of mold sand is 20 times the amount of core sand. The lower sand
usage (sand-to-metal ratio of 0.05) for centrifugal and permanent molds suggests that HAP
emissions may be somewhat lower than those from sand casting. However, at present no HAP
emission measurements have been performed and there are no metal or organic HAP emission
factors specific to centrifugal or permanent mold casting operations.
3.6.3 Investment Casting
Investment casting consists of simply pouring metal into the molds previously described.
After the metal has solidified, the mold is broken away from the tree and individual castings are
cut off the tree. Sometimes the shell does not separate cleanly from the tree and must be
removed by leaching in molten salt (e.g., the Kolene® process).
3-27
-------
Although no emission data are available for investment casting, no substantial HAP
emissions seem possible because of the nature of the processes and materials involved, nor have
significant emissions been observed during these operations.
3.6.4 Expendable Pattern Casting
Expendable pattern castings are made by pouring metal into a sprue that leads to the
bottom of the polystyrene pattern and allowing the metal to simultaneously volatilize the pattern
and replace it, forming a casting of the same shape as the pattern. Vapors generated in the
process escape through the sand that surrounds first the pattern and then the casting. Castings
are removed from the loose sand and then finished in much the same manner as those made by
sand casting.
Emissions consist of metal fumes and pyrolysis products from the vaporized polystyrene.
A discussion of organic emissions is presented by Twarog (1991). The consensus of available
data suggests that emissions contain predominately styrene, along with benzene, ethyl benzene,
and toluene. The presence of poly cyclic aromatic hydrocarbons is indicated in some but not all
studies.
3.7 SAND HANDLING
Shakeout operations generate a substantial volume of sand. Many foundries reuse a large
portion of this sand and only remove a small portion as waste, which is primarily fine grains that
result from abrasion of sand. Most foundries have a large multistep sand-handling operation for
reclaiming the reusable sand. Large foundries often have conveyor sand-handling systems
working continuously. Smaller, less mechanized foundries often use heavy equipment (e.g.,
front-end loaders) in a batch process (EPA, 1992).
Sand-handling operations receive sand directly from the shakeout step or from an
intermediate sand storage area. A typical first step in sand handling is lump knockout. Sand
lumps occur when the binders used in sand cores only partially degrade after exposure to the heat
of molten metal. The lumps, or core butts, may be crushed and recycled into molding sand
during this step. They can also be disposed of as waste material. A magnetic separation
operation is often used to remove pieces of metal. Other steps involve screening to remove fines
and cooling by aeration. In addition, some foundries thermally treat chemically bonded mold
and core sand to incinerate binders and organic impurities (EPA, 1992). Emissions from sand
3-28
-------
handling include PM such as sand, metal particles, condensed organics, and residual binder. If a
thermal treatment is used to reclaim chemically bonded sand, organic HAPs may be emitted.
In the 1998 survey, a total of 284 foundries reported sand reclamation processes, 35 of
which used thermal reclamation. The sand-processing capacities of thermal and nonthermal
reclamation processes were 260 and 8,600 tons/hr, respectively (EPA, 1998b). The majority of
sand reclamation processes (both thermal and nonthermal) were controlled by filters; one
thermal reclamation process was controlled by an incinerator.
Limited data are available on the analysis of waste sand destined for disposal. In 1978,
Calspan Corporation reported that HAPs detected in the sand in trace quantities included lead
(54 ppm), manganese (53 ppm), nickel (28 ppm), chromium (4.8 ppm), and phenol (1.1 ppm).
Considering the above information, emissions of HAPs from sand-handling operations do
not appear to be significant. Some of the factors affecting HAP emissions from sand handling
include:
• Type of sand processed (chemically bonded sand versus clay-bonded sand);
• Type of metal cast;
• Type of sand-handling operation (e.g., thermal treatment); and
• Workplace practices.
3.8 MECHANICAL FINISHING
All castings typically undergo some type of mechanical finishing. Finishing operations
begin once the casting is removed from the mold and cooled. Hammers, band saws, abrasive
cutting wheels, flame cut-off devices, and air-carbon arc devices may be used to remove the
risers, runners, and sprues of the metal transfer system. Metal fins at the parting lines (lines on a
casting corresponding to the interface between the cope and drag of a mold) are removed with
chipping hammers and grinders. Residual refractory material and oxides are typically removed
by sand blasting or steel shot blasting, which can also be used to give the casting a uniform and
more attractive surface appearance (EPA, 1992).
Finishing operations generate PM, which may contain metal HAPs. From their tests at a
gray iron foundry, Potas and Blair (1993) reported manganese as the major metal HAP from
finishing, with smaller quantities of chromium and relatively insignificant levels of lead and
cadmium also present. Uncontrolled manganese emission factors for the six emission points
sampled ranged from 0.045 to 0.21 Ib/ton (average of 0.1 Ib/ton) compared to emission factors
3-29
-------
for chromium that ranged from 0.0036 to 0.032 Ib/ton (average of 0.02 Ib/ton). Uncontrolled
emissions of manganese totaled 15 Ib/hr for all six points associated with the finishing operation
compared to chromium emission rates of 2.9 Ib/hr for all six points. Manganese comprised 1.1
to 1.2 percent of the PM from the blast/grind and spotblast/reblast operations and ranged from
0.3 to 0.4 percent of the PM for the other emission points. Total HAPs were on the order of 1.4
to 1.6 percent from the blast/grind and shotblast/reblast operations and ranged from 0.3 to 0.5
percent of the PM for the other points (Potas and Blair, 1993).
Data presented by Euvrard and Jackson (1992) showed total metal HAPs from grinding
to be 0.8 percent of the PM, compared to 1.1 percent for PM from shotblasting. These results
appear to be consistent with typical manganese content of cast iron and steel, which generally
ranges from 0.25 to 1.0 percent (Gschwandtner and Fairchild, 1992).
The PM produced by mechanical finishing is anticipated to be mainly coarse material
that would not remain airborne. That is, uncontrolled PM produced by mechanical finishing
would not generally escape the foundry building or be transported outside the facility
boundaries.
3.9 CLEANING AND COATING
The cleaning of castings precedes any coating operations to ensure that the coating will
adhere to the metal. Scale, rust, oxides, oil, grease, and dirt can be chemically removed from the
surface of a casting using organic solvents (typically chlorinated solvents, although naphtha,
methanol, and toluene are also used), emulsifiers, pressurized water, abrasives, alkaline agents
(caustic soda, soda ash, alkaline silicates, and phosphates), or acid pickling. The pickling
process involves the cleaning of the metal surface with inorganic acids such as hydrochloric,
sulfuric, or nitric acid. Castings generally pass from the pickling bath through a series of rinses.
Molten salt baths are also used to clean complex interior passages in castings (EPA, 1992).
Castings are often given a coating to inhibit oxidation, resist deterioration, or improve
appearance. Common coating operations include painting, electroplating, electroless nickel
plating, hard facing, hot dipping, thermal spraying, diffusion, conversion, porcelain enameling,
and organic or fused dry-resin coating (EPA, 1992). Table 3-7 compares coating capacities (in
tons/hr of castings coated) to melt capacities (tons/hr of metal poured) at foundries responding to
the 1998 questionnaire (EPA, 1998b).
3-30
-------
TABLE 3-7. MELTING AND COATING CAPACITIES
All foundries '
Total metal poured, tons/hr
1 8,944
Foundries with coating operations 2
Total castings coated, tons/hr
1,344
Total metal poured, tons/hr
2,373
' A total of 590 foundries reported melt capacities.
2 A total of 128 foundries reported coating operations; however, only 78 of these foundries reported information on
coatings and melt capacities.
Source: EPA, 1998b.
Cleaning and coating operations may generate organic HAPs from painting; coating and
solvent cleaning and acid and metal ion mists from anodizing; plating; polishing; hot-dip
coating, etching; and chemical conversion coating. HAP emissions from cleaning and coating
were not assessed under this study because this assessment is being made under the development
of a national emission standard for metal parts coating operations.
3.10 REFERENCES
Baldwin, V. H., 1979. "Environmental Assessment of Decomposition Products from Cores and
Molds," American Foundrymen 's Society (AFS) Transactions, vol. 87, p. 79-101.
Baldwin, V. H., 1982. "Environmental Assessment of Melting, Inoculation, and Pouring,"
American Foundrymen 's Society (AFS) Transactions, vol. 90, p. 82-153.
Bates and Scheel, 1974. "Processing Emissions and Occupational Health in the Ferrous Foundry
Industry," Journal of the American Industrial Hygiene Association, August, p. 452-462.
Calspan Corporation, 1978. Alternatives for Hazardous Waste Management in the Metals
Smelting and Refining Industries. Prepared for U.S. Environmental Protection Agency,
Office of Solid Waste, Washington, DC, by E. Isenberg and Richard P. Leonard of
Calspan Corporation. PB-278-800.
Casting Development Centre, 1997. Report on Environmental Emission Testing Using Foseco
Mold Coating Cereamol 915-9. Prepared for Foseco, Inc., by the Casting Development
Centre, Sheffield, England. March.
Casting Emission Reduction Program, 1999. Baseline Testing Emission Results -Pre-production
Foundry. Report prepared by J. Schifo. November.
Casting Emission Reduction Program, 2000. Baseline Testing Emission Results - Production
Foundry. Report prepared by J. Schifo. February.
Cowen. Cupola Collection Systems. Source unknown. Gray and Ductile Iron Founders'
Society, Inc.
3-31
-------
Ductile Iron Pipe Research Association, 2000. Letter with enclosure to James H. Maysilles, U.S.
Environmental Protection Agency, dated May 3, 2000.
Ecoserve Environmental Services, Inc., 1990. Determination of EPA Combined Metals and
Cadmium Emissions from Arc Furnace Baghouse at Barbary Coast Steel Corporation,
Emeryville, CA.
Emcon Associates, December 1990. Compliance Testing to Quantify Emissions at U. S. Pipe
and Foundry Company, Union City, CA.
Euvrard and Jackson, 1992. "Case Study: Air Audit at a Medium Size Gray Iron Foundry,"
Paper presented at the 5th Annual AFS Environmental Affairs Conference.
Gschwandtner and Fairchild, 1992. Controlling Odorous Emissions From Iron Foundries. U.S.
Environmental Protection Agency. EPA-600/R-92-058. PB92-166925.
Investment Casting Institute, 1995. Letter with enclosures to J. H. Maysilles, U.S.
Environmental Protection Agency dated November 10,1995.
Kirgin, Kenneth H., 1998. "Solid Economy Continues to Fuel Casting Growth," Modern
Casting, American Foundrymen's Society, Des Plaines, IL, vol. 89 (January), No. 1, pp.
30-33.
LaFay, V. S., and Nelrner, S. L., 1998. "Insight Gained into Green Sand's Benzene Emissions,"
article adapted from research papers presented at the 1997 (97-107) and 1998 (98-011
and 98-003) AFS Casting Congresses, available through the American Foundrymen's
Society Library.
Lessiter, Michael, J., and Ezra L. Kotzin. 1996. "Timeline of Casting Technology," Modern
Casting, American Foundrymen's Society, Des Plaines, IL, vol. 87 (November) No. 11,
pp. 64-67.
Normandeau Associates, Inc., 1992. Report on Emission ofDioxins From Cupola at U.S. Pipe
and Foundry Company, Union City, CA. Normandeau Associates, Inc., Bedford, NH.
March.
Potas and Blair, 1993. Paniculate Air Toxics Characterization at a Gray Iron Foundry. AWMA
86th Annual Meeting and Exhibition. Denver, CO.
Scott, Bates, and James, 1976. "Foundry Air Contaminants from Green Sand Molds." Journal
of the American Industrial Hygiene Association, April, p. 335-344.
Shaw, P.M., 1982. "CIATG Commission 4 Environmental Control: Induction Furnace
Emissions" (commissioned by F. M. Shaw, British Cast Iron Research Association, Fifth
Report), Cast Metals Journal, vol. 6, p.28.
3-32
-------
Twarog, Daniel L., 1991. Identification of Emissions and Solid Wastes Generated From the
EPC Process: Literature Review and Analysis. American Foundrymen's Society
Research Report, June 4.
U.S. Environmental Protection Agency, 1980a. Environmental Assessment of Iron Casting.
Cincinnati, OH. EPA-600/2-80-021.
U.S. Environmental Protection Agency, 1980b. Electric Arc Furnaces in Ferrous Foundries -
Background Information for Proposed Standards. Office of Air Quality Planning and
Standards, Research Triangle Park, NC. EPA-450/3-80-020a.
U.S. Environmental Protection Agency, 1981. Summary of Factors Affecting Compliance by
Ferrous Foundries. Office of General Enforcement, Washington DC, EPA-340/1-80-
020.
U.S. Environmental Protection Agency, 1983. Electric Arc Furnaces and Argon-Oxygen
Decarburization Vessels in the Steel Industry - Background Information for Proposed
Revisions to Standards. Office of Air Quality Planning and Standards, Research Triangle
Park, NC. EPA-450/3-82-020a.
U.S. Environmental Protection Agency, 1990. Air Emissions Species Manual. Volume I:
Volatile Organic Compound Species Profiles. Volume II: Paniculate Matter Species
Profiles. 2nd ed. Office of Air Quality Planning and Standards, Research Triangle Park,
NC. EPA-450/2-90-001aandb.
U.S. Environmental Protection Agency, 1991. Control Technologies for Hazardous Air
Pollutants. Office of Research and Development, Washington, DC. EPA-625/6-91-014.
U.S. Environmental Protection Agency, 1992. Guides to Pollution Prevention, The Metal-
casting and Heat Treating Industry. Office of Research and Development, Cincinnati,
OH. EPA-625/R-92-009.
U.S. Environmental Protection Agency, 1993. Compilation of Information from the EPA XATEF
Data Base.
U.S. Environmental Protection Agency, 1995. AP-42 Sections 12.13: Steel Foundries and
12.10: Gray Iron Foundries. Office of Air and Radiation, Washington, DC.
U.S. Environmental Protection Agency, 1997a. Profile of the Metal-casting Industry. Sector
Notebook Project, Office of Enforcement and Compliance, Washington DC, September.
EPA-310/R-97-004.
U.S. Environmental Protection Agency, 1997b. Test Report of Emission Source Test at
Waupaca - Tell City Foundry, Indiana on September 5, 8, 9, and 10, 1997.
U.S. Environmental Protection Agency 1997c. Test Report of Emission Source Test at Saginaw
Metal-casting Operations, Saginaw, Michigan on September 23-25, 1997.
3-33
-------
U.S. Environmental Protection Agency, 1998a. Detailed Information Collection Request for
Iron Foundries and Steel Foundries Source Category: MACTStandards Development
Questionnaire for Iron and Steel Foundries. Office of Air Quality Planning and
Standards, Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1998b. Compilation of Information from Questionnaire
Forms Submitted by Iron and Steel Foundries to the U.S. EPA Office of Air Quality
Planning and Standards. Office of Air Quality Planning and Standards, Research
Triangle Park, NC.
Wingra Associates (Klafka), 1992. "Air Toxics Control Alternatives for Iron Foundry Pouring,
Cooling, and Shakeout Operations," Paper presented at the 85th Annual AWMA Meeting
and Exhibition, Pittsburgh, PA.
3-34
-------
4.0 CONTROL TECHNOLOGY AND PERFORMANCE OF CONTROLS
This chapter describes emission control technologies currently in use at iron and steel
foundries and the performance of these controls. Unless otherwise noted, compilations of type
and frequency of use of specific emissions capture and control technologies and performance
data for those technologies are based on industry responses to the 1998 EPA survey
questionnaire (EPA, 1998).
Emissions capture and control technologies are discussed for the following operations:
• Mold and core making,
• Scrap preparation,
• Metal melting, and
• Pouring, cooling, and shakeout.
These operations, discussed in Chapter 3, may produce substantial HAP emissions and,
therefore, will be further assessed in this document.
The primary purpose of this document is to compile information for use in developing
NESHAP for two of the source categories, namely "Iron Foundries" and "Steel Foundries,"
listed by the EPA as required by section 112(c) of the CAA. As discussed in Chapter 3, the EPA
conducted a survey in 1998 of all known ferrous (i.e., iron and steel) foundries to collect
information to assist in the development of regulations for these source categories. Of the 595
foundries reporting information in the 1998 industry survey, 339 foundries poured iron only, 165
poured steel only, and 91 poured both types of metal. To illustrate the similarities and
differences between iron and steel foundries, process and emission control data are presented
separately for iron foundries and steel foundries. The 91 foundries pouring both iron and steel
were categorized as either iron foundries or steel foundries, for presentation in this document,
depending on the relative amounts of iron and steel poured. If a foundry poured 50 percent or
more of its iron and steel combined as steel, it was categorized in this document as a steel
4-1
-------
foundry. Thirty-six of the 91 foundries were placed in the iron foundry category and 55 in the
steel foundry category, for totals of 375 iron foundries and 220 steel foundries contributing
information to EPA's data base. The decision to categorize the foundries as described is
somewhat arbitrary, but the distinction is clear for most of the 91 foundries in question because
of the relative amounts of each type of metal poured by the foundry.
4.1 MOLD AND CORE MAKING
Most equipment for mold and core making does not include well-defined stacks.
Exceptions are baking ovens and cold-box machines that use catalyst gases. For emissions that
are not emitted through well-defined stacks, emissions control systems must include a hood or an
engineered exhaust system to capture the exhaust stream. These capture mechanisms are then
connected by ductwork to an air pollution control device (APCD).
Capture systems for mold- and core-making emissions were not specified in the survey
responses, but many of the capture systems described for electric furnaces (i.e., canopy hoods,
enclosures, building evacuation, etc.) were likely to be the systems used, if any, for these
operations. Except for cold-box operations, mold- and core-making operations were mostly
uncontrolled or controlled by filters, cyclones, and PM scrubbers. These PM controls reduce
dust (i.e., sand) emissions that arise from the sand mullers. They are not effective in reducing
emissions of organic vapors that arise from the chemical binder system. However, when
considering the potential for HAP emissions, it is emissions of organic vapors, not PM (or metal
HAPs), that are of concern from mold- and core-making operations. Meaningful HAP emission
controls for mold- and core-making operations were almost entirely controls on cold-box lines
aimed at reducing emissions of triethylamine (TEA).
TEA is a gaseous HAP that is frequently used as a catalyst to initiate the polymerization
reaction for cold-box mold- and core-making lines. Table 4-1 shows the use of controls for cold-
box mold- and core-making lines in which the catalyst gas was TEA. Most cold-box operations
that used TEA were controlled by packed-bed scrubbers that used a sulfuric acid solution to
absorb and react with the TEA gas. Packed-bed scrubbers operate on the principle of absorption,
in which one or more components of a gas mixture are selectively transferred into a relatively
nonvolatile liquid. Absorption of a gaseous component by a liquid occurs when the liquid
contains less than the equilibrium concentration of the component. The difference between the
actual concentration in solution and the equilibrium concentration provides the driving force for
4-2
-------
TABLE 4-1. USE OF CONTROLS FOR TEA EMISSIONS FROM
COLD-BOX MOLD- AND CORE-MAKING LINES' 2
Iron foundries
Type of control
Wet scrubbing
with acid solution
Incineration
Condensation
No control
Number
of lines
using
control
380
7
3
49
Number of
foundries
using control3
79
3
2
25
Steel foundries
Type of control
Wet scrubbing
with acid
solution
Condensation
No control
Number of
lines using
control
15
2
13
Number of
foundries
using control
10 !
2
7
1 A few of these lines may use dimethylethylamine (DMEA) for the catalyst; some
questionnaire responses did not distinguish between TEA and DMEA.
2 The use of controls for PM is not considered control for TEA.
Three iron foundries operated a combination of controlled and uncontrolled lines.
absorption. The absorption rate depends on the physical properties of the gas/liquid system (e.g.,
diffusivity, viscosity, and density) and the absorber operating conditions (e.g., temperature and
flow rates of the gas and liquid streams). Absorption is enhanced by lower temperatures, greater
contacting surface, higher liquid-to-gas ratios, and higher concentrations in the gas stream (EPA,
1991).
Sulfuric acid wet scrubbers are very effective for TEA emissions control because the
sulfuric acid reacts with the TEA, virtually eliminating dissolved TEA in the scrubbing solution.
Therefore, the driving force for absorption is not limited by the amount of TEA removed,
provided there is an adequate supply of unreacted sulfuric acid in the scrubbing media. Table 4-
2 gives a summary of the available source test data for TEA emission controls.
TEA emissions (or concentrations) were generally too low to be quantified in the outlet
gas streams from the acid wet scrubbers, for which source test data were available. As such,
precise removal efficiencies could not be determined, though most of these scrubbers achieved a
TEA removal efficiency of 99 percent or higher. The only other information on TEA control
consists of control efficiency information reported without supporting test data in the responses
4-3
-------
TABLE 4-2. SOURCE TEST DATA FOR TEA ACID WET SCRUBBERS
Foundry
ID
GA-1
MI-33
NC-5
VA-8
WI-01
WI-42
Test date(s)
10/21/96
through
11/1/96
6/13/96
3/16/95
through
3/17/95
11/18/96
2/7/95
Test method /
Conditions of Test
Method not reported /
pH= 0.32
AP= 1.75"H,O
Qair = 8,500 acfm
Qsw=105.6gpm
Method not reported /
QJ]r = 4,520 dscfrn
NIOSH Method 20 103
Qair = 4,886 acfm
T = 73°F
EPA Methods 2, 3,4,
and 18
NIOSH Method 221
Qa,r= 16,500 acfm
EPA Method 18
pH < 1 .0
Qajr- 18,000 acfm
5.18 tons cores/hr
Method not reported
pH - 2.0
Qair = 32,000 acfm
Run
No.
1
2
3
1
2
3
1-1
2-1
3-1
1-2
2-2
3-2
1
2
3
1
2
3
1
2
3
TEA measured at
scrubber inlet
Mass
rate,
Ib/hr
7.55
5.99
2.58
15.6
16.3
17.5
0.796
0.527
0.710
0.521
0.604
0.110
33.89
27.23
21.71
12.265
12.265
12.265
Cone.,
ppmv1
55. 82
44.6:
18.92
209
230
255
9.94
6.94
8.94
6.5'
7.64
1.44
133.48
105.34
85.00
24.04
24.04
24.04
TEA measured at
scrubber outlet
Mass
rate,
Ib/hr
<0.20
<0.18
<0.18
0.022
0.031
0.023
< 0.00250
< 0.00253
< 0.00256
< 0.00252
< 0.00256
< 0.00257
<0.02
<0.02
<0.02
< 0.256
<0.215
<0.210
<0.14
<0.14
<0.14
Cone.,
ppmv1
<1.52
<1.42
< 1.42
0.29
0.43 i
0.34
<0.034
<0.034
<0.034
<0.034
<0.034
<0.034
<0.07
<0.07
<0.07 '
< 0.904
<0.764
< 0.744
<0.3:
<0.32 I
<0.3:
' Parts per million by volume.
2 Concentration in ppmv was calculated from the test value as reported in milligrams per cubic meter (mg/m3).
3 Concentration was calculated from reported TEA mass flow rate and air flow rate.
4 NIOSH = National Institute of Occupational Safety and Health.
5 Mass flow rate was calculated from TEA usage rate and estimated control system capture efficiency.
4-4
-------
to the 1998 industry survey. Reported control efficiencies for wet scrubbers were 99.9 percent
for two lines controlled by two scrubbers at one foundry; 99 percent for 42 lines controlled by 13
scrubbers at six foundries; and 98 percent or lower for 54 lines controlled by 13 scrubbers at four
foundries. Efficiencies were not reported for 61 lines controlled by 11 scrubbers at five other
foundries. The basis for the values was usually given as design efficiency. Scrubber design was
described as a vertical packed bed for 10 of the 15 scrubbers with design collection efficiencies
of 99 percent or higher. In most cases, the scrubber descriptions included the fact that acid
solution was used as the collection medium.
Controls other than wet scrubbing used to control TEA emissions from cold-box mold-
and core-making lines are thermal oxidation, which was reported as being used on seven cold-
box lines, and condensation, which was used on three lines. Thermal oxidation would control
emissions of organic HAPs other than TEA more effectively than scrubbing, but no information
is available to indicate that either thermal oxidation or condensation was more effective for TEA
than acid scrubbing.
For binder systems other than the TEA-catalyzed cold-box system, there are no emission
control devices that effectively reduce HAP emissions from mold- and core-making lines.
However, pollution prevention methods are possible for certain binder chemical systems.
Referring to the data summarized in Appendix B, several systems produce relatively high
emissions compared with others. HAPs emitted by these systems include cumene, dimethyl
phthalate, methanol, methyl ethyl ketone, and phenol. The HAP content of each of the systems
can be varied, but the HAPs cannot always be eliminated or reduced below certain thresholds,
which depend on the conditions under which the systems are used (e.g., temperature or the
strength requirements of the molds or cores). Discussions with industry suppliers indicate that
methanol can be eliminated from the furan warm-box system, but HAP reductions in the other
high-emitting systems cannot be prescribed. The furan warm-box system was used in 55 mold-
and core-making lines in iron foundries and in 3 lines in steel foundries. At least 23 lines in iron
foundries used formulations that did not contain methanol. Complete information on
formulations is not available because the formulations were not reported in the 1998 survey. A
sample of larger foundries, however, was contacted after the survey to obtain this information.
Results of the sampling effort indicate that use of the furan warm-box system without methanol
is easily achievable.
4-5
-------
Two relatively low-emitting systems in common use, for which reductions are possible,
are the phenolic urethane cold-box and phenolic urethane no-bake systems. Both systems use
petroleum distillate solvents that contain naphthalene and lesser amounts of cumene and xylene.
These solvents commonly contain about 10 percent naphthalene, but products are also available
that contain 3 percent or less naphthalene. Naphthalene-depleted solvents can be identified in
Material Safety Data Sheets (MSDS) through their Chemical Abstract Service (CAS) numbers,
which are 70693-06-0 for a lower boiling point product (about 150 °C) and 68477-31-6 for a
higher boiling product (about 200 °C); the latter of these two products is used in binder chemical
formulations. The solvent with higher naphthalene content is CAS number 64742-94-5. Use of
naphthalene-depleted solvents may result in substantial reductions of naphthalene emissions; 705
mold- and core-making lines in iron foundries and 205 lines in steel foundries used phenolic
urethane cold-box or no-bake systems. The use of naphthalene-depleted solvent in these lines is
not known completely because this information was not reported in the 1998 survey. Several
larger foundries contacted after the survey supplied MSDS for the binder chemicals they used.
The use of these solvents varied considerably; some foundries used chemicals containing
depleted solvents in some lines but not in others. The use of these solvents seems to be
constantly increasing; however, industry sources suggested that the availability of the
naphthalene-depleted solvent may be limited (Brown, 2000; Stone, 2000).
Except for the furan warm-box and phenolic urethane systems, no HAP substitution
opportunities that can be prescribed have been found. However, other binder systems, such as
the furan no-bake, phenolic no-bake, and the Shell (Novolak flake) systems, can be formulated
without methanol. According to an industry representative, methanol replacement in these
binder systems cannot be prescribed because the substitute binder formulations may not be
compatible for a specific foundry's operations (i.e., the substitute binder formulation may lack an
essential characteristic of the methanol-containing formulation for a given application).
However, use of different (low-HAP-emitting) binder formulations appears to be a potential
means to reduce HAP emissions from mold- and core-making operations.
4.2 MOLD AND CORE COATING
For mold- and core-coating operations, in which HAPs may be present in the coating
material as liquid constituents that evaporate, one form of control that is often used is the light-
off procedure, in which the coating is ignited after application to dry it. This procedure can be
4-6
-------
used only if the coating material is flammable and if the adhesive properties of the binder
chemicals are not degraded by the heat generated by the procedure. No information is available
on emissions where the light-off process is used for coatings containing HAPs, but one such
study was made for a coating containing a solvent based on isopropyl alcohol (Castings
Development Centre, 1997). This study concluded that 70 percent or more of the solvent was
destroyed when the drying period was minimized and that the major combustion product was
carbon dioxide. One HAP for which emissions can be reduced by this procedure is methanol.
The efficiency of reducing methanol emissions by this process is not known because the only
emissions data available for a light-off process are for a coating with isopropanol as a
constituent. Based on this study, a methanol destruction efficiency of roughly 70 percent can be
expected.
In addition to the light-off process, pollution prevention methods can be used to reduce
HAP emissions from mold- and core-coating operations. The primary pollution prevention
method available is the substitution of a HAP-containing coating material (such as a methanol-
based coating) with a non-HAP-containing coating material (isopropanol for example). Of 861
mold- and core-coating lines at iron foundries, 12 used a coating containing methanol and 339
used a coating containing isopropanol. Of 474 mold- and core-coating lines at steel foundries,
17 used a coating containing methanol and 226 used a coating containing isopropanol.
Discussions with industry sources indicate that methanol can be replaced by isopropanol in
coating formulations without forcing process changes. Coating formulations based on water
instead of alcohol were also commonly used (by 382 coating lines at iron foundries and 191 lines
at steel foundries).
4.3 SCRAP PREPARATION
Scrap metal typically undergoes some type of preparation before melting, which may
include cutting or sizing, shot or sandblasting to remove coatings, cleaning with organic solvents
to remove oils and grease, and preheating. Except for preheaters, the survey questionnaire did
not ask about specific emissions capture or control technologies for scrap preparation, and no
controls were reported for other phases of scrap preparation. Preheaters were used
predominantly with EIFs, as noted in Chapter 3. Tables 4-3 and 4-4 summarize survey responses
to APCDs used for loading, heating, and discharging scrap from preheaters.
4-7
-------
TABLE 4-3. CONTROL CONFIGURATIONS FOR SCRAP PREHEATERS
Type of control used1
Loading
No control
Filter
No control
No control
Filter
Filter
Filter
No control
No control
Cyclone
Cyclone
Cyclone
Scrubber
No control
No control
No control
Heating
No control
Filter
Filter
No control
Filter
No control
Cyclone
Filter
Afterburner (AB)
Cyclone/ AB
Cyclone
Cyclone
Scrubber
Cyclone
No control
Scrubber
Discharging
No control
Filter
Filter
Filter
No control
Filter
Filter
No control
No control
No control
Cyclone
Filter
Scrubber
No control
Scrubber
No control
TOTALS
Number of foundries
with control
configuration
Iron
68
17
8
5
2
2
1
1
2
1
2
1
1
1
1
113
Steel
7
1
1
9
Number of
preheaters with
control
configuration2
Iron
76
24
19
7
7
3
2
1
6
6
2
6
1
1
1
157
Steel i
18
1
1
20
1 Blank responses
2 Blank responses
to type of control were classified as "no control."
to number of preheaters were assigned a value of
1."
As shown in Table 4-4, most preheaters were uncontrolled. Preheaters that did use
controls used mostly fabric filters, which controlled HAP metals contained in PM. Because the
fabric filters were also commonly used to control the EIFs served by the preheaters, these
devices will be discussed in the section on EIF controls.
Three foundries used afterburners, which constituted control for organic HAPs. Another
form of organic HAP control was direct gas-fired preheating, which was the mode used by most
foundries. Specifically, 171 of the 177 preheaters were direct gas-fired preheaters. Discussions
with a sample of operators revealed that preheater gas burners operated at various temperature
4-8
-------
TABLE 4-4. SPECIFIC CONTROL DEVICES ON SCRAP PREHEATERS
Controls and use
Preheaters with no control'1 2
Facilities with no control
Preheaters with filter
Facilities with filter
Preheaters with afterburner (AB)
Facilities with AB
Preheaters with cyclone and AB
Facilities with cyclone and AB
Preheaters with cyclone
Facilities with cyclone
Preheaters with scrubber
Facilities with scrubber
Total number of preheaters
Total number of facilities
Loading
Iron
111
86
31
22
14
4
1
1
157
113
Steel
20
9
20
9
Heating
Iron
86
75
46
28
6
2
6
1
11
5
2
2
157
113
Steel
19
8
1
1
20
9
Discharging
Iron
93
76
61
34
2
2
1
1
157
113
Steel
IS ;
7 ,
1
1
1
1
20 1
9
1 Blank responses to type of control were classified as "no control."
2 Blank responses to number of preheaters were assigned a value of
1."
ranges sfrom 800 to 1,300 °F. No emission tests have been conducted for organic species, so the
relative efficiencies of afterburning versus direct gas firing cannot be determined.
In addition to the techniques described above, another form of scrap preparation that is
commonly used is specification of quality. Of the 595 iron and steel foundries that provided
survey responses, 360 (or 60 percent) of iron and steel foundries indicated that they used some
type of scrap selection, cleaning, or inspection program to ensure the quality of scrap metal used
by the foundry. The percentage of respondents that specified scrap selection as a work practice
to reduce emissions was relatively consistent across foundries operating different furnace types:
45 percent of cupola foundries, 61 percent of EAF foundries, and 65 percent of EIF foundries.
The scrap selection, cleaning, or inspection programs included specifications on the types
or grades of scrap used, limits or bans on oil, grease, and/or paint in the scrap, and restrictions on
4-9
-------
lead, galvanized metals (a source of cadmium), and certain alloys (a source of chromium, nickel,
or high manganese). These scrap specifications could, in principle, result in reduced HAP
emissions from preheating and melting, which is a pollution prevention procedure. No data,
however, are available to quantify emission reductions that may result from eliminating or
reducing organic contaminants and HAP metals in the scrap fed to preheaters and melting
furnaces.
The size of foundries that used scrap specifications varied substantially, representing
almost the entire range of foundry production. For example, annual production of foundries that
had organic substance and/or HAP metal specifications for cupola, EIF, and preheater feed
ranged from more than 100,000 to less than 1,000 tons per year in each case.
4.4 METAL MELTING
As noted in Chapter 3, the predominant types of furnaces at iron and steel foundries are
cupolas (used only at iron foundries), EAF (used mainly at steel foundries), and EIF (commonly
used at both iron and steel foundries). Emissions from these furnaces are predominately metal,
but may include organic HAPs, especially in the case of cupolas. The following sections
describe existing capture and control technologies for emissions from these three types of
melting furnaces.
4.4.1 Cupola Controls
Emissions from cupolas arise from three operations: charging, melting, and tapping.
Combustion air is blown through the base of the cupola and travels upward through the charge.
Melting emissions are contained in this forced air flow, which is routed to an APCD. Cupolas
have an opening (a charging door) in the shaft of the furnace above the charge level. The
disposition of charging emissions depends on whether the exhaust gas takeoff to the melting
APCD is above or below the charging door. When materials are not being charged to the
furnace, the draft of the melting APCD creates sufficient negative gauge pressure inside the
furnace to prevent release of emissions through the charging door. For cupolas with above-
charge gas takeoff, the periodic addition of charge material (usually via a vibratory or belt
feeder) momentarily alters the exhaust stream flow in the cupola shaft. If the flow alteration is
significant, which could be caused by adding a large amount of charge material suddenly, a brief
burst of emissions, or "puffing," may occur from the charging door. Puffing generally does not
occur for cupolas with below charge gas takeoff because the exhaust is drawn from below the
4-10
-------
level of the charge, and the addition of charge material does not interrupt the exhaust stream
flow within the cupola shaft. Charging and tapping emissions from cupolas are minimal
compared with melting emissions and, hence, are typically uncontrolled (EPA, 1998).
For melting emissions, most cupolas have a wet scrubber or a fabric filter for PM control
and also employ an afterburner, which is located upstream of the filter or scrubber, for control of
organic substances. Table 4-5 summarizes the use of controls as reported in 1998. Wet
scrubbers and fabric filters are briefly described in the following sections. The use of
electrostatic precipitators is so infrequent in this industry that no discussion of this device is
presented here, but the electrostatic precipitator is generally considered to be less effective than a
fabric filter for fine PM control (Buonicore, 1992, p.l 14).
TABLE 4-5. CONTROLS FOR MELTING EMISSIONS FROM CUPOLAS
Controls
Afterburner plus fabric filter
Afterburner plus wet scrubber*
Afterburner plus electrostatic precipitator
Wet scrubber*
Fabric filter
None
Totals
Number of foundries
42
36
1
17
6
8
110
Number of furnaces
56
49
1
22
6
9
143
* Most wet scrubbers were venturi scrubbers.
4.4.1.1 Wet Scrubbers. Venturi scrubbers are the most common type of wet scrubber
used to control PM emissions, and thus metal HAP emissions, from cupolas. The primary
collection mechanisms inherent in a venturi scrubber are the impingement of particles on
droplets and the condensation of liquid on the particles. Impingement is attained by accelerating
the gas stream to velocities of 200 to 600 feet per second (ft/sec) in the venturi throat. When
water is introduced into the high-velocity stream, it is atomized into tiny droplets. Because these
droplets are at a relatively low velocity with respect to the gas stream, the particles are collected
on these droplets through impaction. Particle collection through condensation also occurs when
saturated streams are cooled in the venturi.
4-11
-------
Pressure differential is a key factor affecting the efficiency of a scrubber in removing
particulate matter and therefore metal HAPs. High-energy (i.e., high-pressure differential)
scrubbers are capable of reducing particulate loadings from cupolas to about 0.05 grains per dry
standard cubic foot (gr/dscf) (EPA, 1981, p. 73). As a rule of thumb, a high-efficiency scrubber
is one with a pressure differential greater than 50 inches of water column. Table 4-6 summarizes
pressure differentials for venturi scrubbers as reported in the survey. Of the 55 pressure
differentials compiled in the table, 16 were equal to or greater than 50 inches of water.
Scrubbers with pressure differentials in the range of 50 to 70 inches of water can reduce
emissions from cupolas to 0.05 gr/dscf, and scrubbers with pressure differentials of 100 inches of
water can reduce emissions to 0.03 gr/dscf (depending on the quality of the scrap) (EPA, 1981,
p. 73). Scrubbers are generally designed to give constant removal efficiencies for a given
pressure differential; thus, outlet grain loadings are expected to be dependent on inlet grain
loadings.
Figure 4-1 shows the results of source tests for PM measured in exhaust gases from 19
wet scrubbers on cupolas. Average outlet PM concentrations for 15 of the 19 scrubbers ranged
from 0.01 gr/dscf to 0.07 gr/dscf. As seen in Figure 4-1, the performance of the remaining four
scrubbers is significantly inferior, ranging from 0.09 to 0.20 gr/dscf.
TABLE 4-6. PRESSURE DIFFERENTIALS OF
VENTURI SCRUBBERS USED ON CUPOLAS
Pressure differential,
inches of water column
<8
20 to 29
30 to 39
| 40 to 49
50 to 59
60 to 70
Number of scrubbers
9
5
14
11
9
7
4-12
-------
0.24
0.22
0.20
0.18
0.16
- Single run value
• Average of run values
(A
5 0.1
(0
I 0.12
'in
i§ 0.10
a.
0.08
0.06
0.04
0.02
0.00
i ' i !
MI-33 NC-5 IA-5 NJ-5 MI-13 WI-24 OH-46 NJ-4 MI-30 MI-33 OH-46 WI-50 OH-46 IN-12 OH-46 OH-12 WI-18 OH-11 OH-44
Foundry ID
Figure 4-1. Filterable PM Emissions (gr/dscf) from Wet Scrubbers on Cupolas at Iron Foundries.
4-13
-------
4.4.1.2 Fabric Filters. In a fabric filter, or baghouse, the particulate-laden gas stream
passes unidirectionally though a woven or felt-type fabric, which screens out the PM. Particles
greater than 1.4 micrometers (|o,m) in diameter are collected at nearly 100-percent efficiency via
impaction, while particles 0.1 to 0.2 u.m in diameter are collected by diffusion. The major design
factor that affects the efficiency of a fabric filter is the air-to-cloth (A/C) ratio, which is the ratio
of gas volume entering the filter (cubic feet per minute) to the total surface area of the filtering
fabric (square feet) (EPA, 1981, p. 85), or, more simply, the velocity of the gas through the filter
(feet per minute [ft/min]). The ratio chosen is generally dependent on the particle size of the
emissions, with lower A/C ratios used for emissions streams with fine particulate. For woven
fabrics, A/C ratios are typically 3.0 ft/min or less (Buonicore, 1992; p. 128). For felt-type
materials, A/C ratios of 10 are common. Table 4-7 summarizes reported A/C ratios of fabric
filters used on cupolas at ferrous foundries in 1998.
TABLE 4-7. A/C RATIOS FOR FABRIC FILTERS ON CUPOLAS
A/C ratio, ft/min
<2
2 to 2.99
3 to 3.99
4 to 4. 99
5 to 5.99
Number of filters
20
8
7
2
1
Figure 4-2 shows the results of source tests for PM measured in exhaust gases from
twelve fabric filters on cupolas. Repetitive foundry listings (Figure 4-2) indicate that baghouses
were tested multiple times. A summary of the test data referenced in this figure is given in
Appendix D. Average PM concentrations ranged from less than 0.001 gr/dscf to 0.005 gr/dscf.
These concentrations were lower than those achieved by wet scrubbers.
The two cupolas that achieved average outlet PM concentrations of less than 0.001
gr/dscf both employed a novel pulse-jet baghouse with horizontally supported bags rather than
the traditionally designed vertically hanging bags. According to an operator of one of these
novel baghouses, a lighter weight fabric can be used when the bags are horizontally supported.
4-14
-------
0.007
0.006
0.005
- Single run value
• Average of run values
c-
o
(0
0.004
10
E 0.003
0.002
0.001
0.000
WI-35 WI-35 WI-35 IN-01 NJ-3 MI-26 IN-34 NC-05 VA-8 FL-6 IA-19 NJ-03 IN-35 SD-1 WI-49
Foundry ID
Figure 4-2. Filterable PM Emissions (gr/dscf) from Fabric Filters on Cupolas at Iron Foundries.
4-15
-------
When bags hang vertically (as in traditional baghouses), the tops of the bags must be strong
enough to hold up the weight of the entire bag (generally 2 or 3 ft long), and the entire filter cake
on that bag. A light-weight bag would not be able to support the weight, and would tear. By
having the bags supported horizontally, they are able to reduce the weight the bag material
supports to only the small amount under the horizontal support (typical bags are 4 to 6 inches in
diameter). The light-weight bag is easier to clean and is more permeable, which allows for a
more even distribution of the air flow. Heavier weight bags tend to get more material caught in
the bag material, and as a result need to be cleaned more frequently and more vigorously. The
contact indicated that, "since 80% of emissions are associated with cleaning," by lowering the
cleaning frequency, the baghouse emissions are lowered. The light-weight bag is also more
permeable, so that pressure drop is reduced, and air flow is more evenly distributed. This, along
with the low A/C ratio for these baghouses, allows more of the PM material to be collected on
the bag surface, rather than becoming impregnated into the fabric, making it easier to clean the
bags.
4.4.1.3 Afterburners. Afterburners are thermal incinerators that employ heat and
oxygen to oxidize (combust) organic chemicals, converting them primarily to carbon dioxide and
water. A typical cupola exhaust will contain CO at levels of 10 percent or higher. In
applications associated with cupolas, afterburners are installed primarily to combust this CO, but
they also act to incinerate any organic compounds present in the cupola exhaust. In general,
combustion temperature and residence time are two important design parameters for
afterburners. For a 98-percent destruction efficiency of nonhalogenated organics in an emissions
stream, suggested values for combustion temperature and residence time are 1,600 °F and
0.75 seconds, respectively (EPA, 1991, p. 4-5). For a 99-percent destruction efficiency of a
nonhalogenated emissions stream, suggested values for combustion temperature and residence
time are 1,800 °F and 0.75 seconds, respectively (EPA, 1991, p. 4-5).
From the 1998 survey responses, most afterburners used to control cupola emissions
reported design efficiency in terms of CO destruction. Table 4-8 presents the shows the
relationship between PM outlet concentration and A/C ratio for the filters tested and also shows
the filter materials used. The composition of the material is important in that it must be
4-16
-------
TABLE 4-8. CUPOLA AFTERBURNER CO OUTLET
CONCENTRATION AND EMISSIONS DATA
Foundry
NJ-03
VA-08
IN-34
NC-05
I A- 19
AL-37
NJ-05
TX-18
MI- 13
NJ-04
Wl-42
WI-24
OH- 13
CO outlet concentration (ppmv)
Run 1
13
17
52
51
92
50
75
136
287
155*
1.800*
Run 2
5.9
16
30
72
98
141
293
184
116
322*
300*
Run 3
7
13
37
119
103
104
18
137
334*
380*
Average
9
15
27*
40
81
98
98
129
160
180
270*
320
827*
CO emission
rate
(Ib/hr)
2.2
3.3
4.2
9.3
16
14.8
26
36
26
35
22.7
26.7
66.5
*CO concentrations calculated from reported CO emissions and volumetric flow rates.
Table 4-9 offers a summary of combustion temperatures and residence times of
afterburners used with cupolas. Note that discussions with foundry operators subsequent to
questionnaire responses revealed that not all reported temperatures were for the same zone of the
afterburner. This is because combustion of CO often continues in the exhaust stream from the
afterburner, so that combustion temperatures and residence time in the afterburner combustion
chamber itself may not represent the complete combustion characteristics. Some foundry
operators considered only the afterburner combustion chamber in providing this information,
while others considered the entire flue gas vent prior to heat recovery as the afterburner. As
such, information collected from the 1998 industry survey with respect to temperature and
residence time for cupolas is difficult to correlate with the afterburner destruction efficiency.
4-17
-------
TABLE 4-9. OPERATING CONDITIONS FOR CUPOLA AFTERBURNERS*
! Parameter
and range
Temperature, °F
> 1,000 to 1,300
> 1,300 to 1,600
> 1,600 to 1,800
> 1,800
Residence time, sec
<0.75
>0.75
Number of recuperative hot-
blast cupolas operating in this
range
7
25
3
0
20
15
Number of nonrecuperative
cupolas operating in this range
17
21
3
2
17
9
Includes only those facilities that reported the requested information; data were not provided for all of the 143
cupolas.
compatible with the temperature of the gas and resistant to any conditions of wear, corrosion,
and humidity that exist.
The EPA acquired speciated HAP data from the two tests for which PM and metal HAP
data are summarized in Table 4-10. In these tests, the average cupola combustion zone
temperatures were 1,670 and 1,560 °F. Three sampling runs were made in one test and four in
the other. Test methods used were EPA Method 23, Determination of Polychlorinated Dibenzo-
p-Dioxins and Polychlorinated Dibenzofurans (PCDDs/PCDFs) From Stationary Sources, and
SW-846 Methods 0010 (sampling) and 8270 (analysis), which are applicable to the
determination of semivolatile principal organic hazardous compounds (POHCs) from
incineration systems. Because the latter method measured 70 HAP compounds, and a cupola
with an afterburner acts as an incineration device, we believe that this combination of methods is
appropriate and that it analyzed a sufficient number of compounds to adequately assess organic
HAP emissions from a cupola. The results of these source tests indicate that organic HAP
emissions from cupola afterburners are very low. Most of the analytes were not present above
the quantitation limits of the analytical methods. Those that were detected, were present at
concentrations of less that 2 parts per billion by volume (ppbv).
4-18
-------
TABLE 4-10. SOURCE TEST DATA FOR ORGANIC HAP EMISSIONS
FROM CUPOLA AFTERBURNERS
Run No.
PCDD/PCDF concentration in offgas
adjusted by the 2,3,7,8-TCDD TEF,
ng/dscm
Total PCDD/PCDF
2,3,4,7,8-
PeCDF
Semivolatile organic HAP concentration
in offgas, ppbv \
Acetophenone
Pyrene
Foundry 1N-34 (combustion temperature: 1,650 °F)
1
2
3
{1.82}'
3.65
{5.47}
1.01
1.93
2.95
1.79
0.651
1.09
NDf
ND
ND
Foundry MI-33 (combustion temperature: 1,550 °F)
1
2
3
4
{0.85}
{0.54}
{0.18}
{0.17}
0.40
0.25
0.06
0.07
0.82:
0.411
0.45
0.291
0.070
0.046
0.021
0.019
Values in brackets indicate that at least some of the species contributing to the total value were detected in levels
below the quantitative limit.
ND - Not detected.
Sample catch was less than five times the estimated laboratory blank value.
4.4.2 EIF Controls
Unlike cupolas, electric furnaces do not include well-defined stacks. Control systems for
these furnaces must therefore include hoods or other types of capture mechanisms ducted to the
control devices. Also, the charging, melting, and tapping phases of the melting cycle occur in
sequence, whereas in a cupola these operations occur simultaneously. Charging emissions from
these furnaces may be significant. Charging and melting emissions may be captured by different
systems because the configuration of the furnace is different for the two operations (i.e., the
furnace cover is removed for charging). The two exhaust streams may be ducted to separate
control devices or to the same device. Depending on the capture systems used, tapping
emissions may also be captured, usually incidentally because these emissions are relatively
insignificant and no system dedicated to these emissions is normally used.
Tables 4-11 through 4-15 summarize the use of control devices reported in 1998 on EIFs.
Most EIFs were not controlled. The vast majority of EIFs with controls used fabric filters or
cartridge filters. As seen in Tables 4-11 and 4-13, although a great variety of combinations of
4-19
-------
controls exists, most of those combinations include only filters. Also, most preheaters (PHs)
were used in conjunction with EIFs. As shown earlier in Table 4-4, 60 PHs were equipped with
filters for at least one phase of the melting operation. Forty-eight of these were employed in
conjunction with EIFs that also were equipped with filters. Of those 48 PHs, 30 were controlled
by the same filters as their associated EIF.
Table 4-15 summarizes A/C ratio data for EIF filters. In general, baghouses used to
control EIF emissions had higher A/C ratios than baghouses used to control cupola emissions.
Source test data for induction furnace and preheater PM emissions were available for 19
fabric filters (17 baghouses and 2 cartridge filters) used to control emissions from 57 EIFs and
16 scrap PHs, from 1 venturi scrubber on 2 electric induction furnaces, and from 1 cyclone on 2
EIFs. Figure 4-3 diagrams outlet gas PM concentration to illustrate the effectiveness of these
control systems. Note, repetitive foundry listings for WI-47 in Figure 4-3 indicate separate
fabric filter control systems; otherwise, repetitive foundry listings in Figure 4-3 indicate that the
baghouses were tested multiple times. Detailed information on the source tests summarized in
Figure 4-3 can be found in Appendix E.
4-20
-------
TABLE 4-11. CONTROL CONFIGURATIONS FOR EIFs AT IRON FOUNDRIES1
1 Operation and type of control
Charging
No control
Filter3
Filter
No control
Filter
No control
No control
No control
Wet scrubber
Wet scrubber
Filter
No control
No control
Filter
Totals
Melting
No control
Filter
Filter
Filter
No control
Filter
No control
Wet scrubber
Wet scrubber
Wet scrubber
No control
Cyclone
No control
Wet scrubber
Tapping
No control
Filter
No control
No control
No control
Filter
Filter
No control
No control
Wet scrubber
Filter
No control
Wet scrubber
Wet scrubber
Number of
furnaces with
configuration2
438
210
43
17
11
8
6
5
5
4
2
2
2
1
754
Number of
foundries with
configuration
181
69
14
7
3
2
2
1
2
1
1
1
1
1
286
1 Blank responses were interpreted to mean no control and thus were classified as "no control."
2 Information is ranked by the number of furnaces controlled with the given configuration.
3 Filter = fabric filter or cartridge filter (the former make up the majority).
4-21
-------
TABLE 4-12. SPECIFIC CONTROLS ON EIFs AT IRON FOUNDRIES'
Type of control
Furnaces with filter
Foundries with filter
Furnaces with wet scrubber
Foundries with wet scrubber
Furnaces with cyclone
Foundries with cyclone
Total number of furnaces with control
Total number of foundries with control
Total number of furnaces with no control
Total number of foundries with no control
Total number of furnaces
Total number of foundries
Number of operations
Charging Melting
267 278
88 92
9 15
3 5
2
1
276 295
91 98
478 459
195 188
754 754
286 286
controlled
Tapping
226
74
7
3
233
77
521
209
754
286
1 Blank responses were interpreted to mean no control and thus were classified as "no control."
2 Filter = fabric filter or cartridge filter (the former make up the majority).
4-22
-------
TABLE 4-13. CONTROL CONFIGURATIONS FOF EIF
AT STEEL FOUNDRIES'
Operation and type of control
Charging
No control
Filter3
Filter
No control
Cyclone
No control
Wet scrubber
No control
Argon gas cover
Filter
No control
No control
Totals
Melting
No control
Filter
Filter
Filter
Cyclone
Filter
Wet scrubber
Electrostatic oil
collection
Argon gas cover
No control
No control
Wet scrubber
Tapping
No control
Filter
No control
No control
Cyclone
Filter
Wet scrubber
No control
No control
No control
Filter
No control
Number of
furnaces with
configuration2
509
81
14
11
6
4
4
4
3
3
2
2
643
Number of
foundries with j
configuration
144
23
3
5
2 i
3
1
1
1
1
1
1 |
186
1 Blank responses were interpreted to mean no control and thus were classified as "no control.
2 Information is ranked by the number of furnaces controlled with the given configuration.
3 Filter = fabric filter or cartridge filter (the former make up the majority).
4-23
-------
TABLE 4-14. SPECIFIC CONTROLS ON EIFs
AT STEEL FOUNDRIES1
Type of control
Furnaces with filter
Foundries with filter
Furnaces with wet scrubber
Foundries with wet scrubber
Furnaces with other
Foundries with other
Total number of furnaces with control
Total number of foundries with control
Total number of furnaces with no control
Total number of foundries with no control
Total number of furnaces
Total number of foundries
Number of operations controlled
Charging
98
27
4
1
9
3
111
31
532
155
643
186
Melting
110
34
6
2
13
4
132
43
514
146
643
186
Tapping
87
27
4 i
1
6
2
98
31
546 ]
156
643
186
1 Blank responses were interpreted to mean no control and thus were classified as "no control."
2 Filter = fabnc filter or cartridge filter (the former make up the majority).
TABLE 4-15. A/C RATIOS FOR FILTERS ON EIFs
A/C ratio
<2
2 to 2.99
3 to 3.99
4 to 4. 99
5 to 5.99
>6
Filters in iron foundries
3
12
17
8
6
17
Filters in steel foundries
1
6
1
4
5
3
4-24
-------
0.010
0.009
0.008
0.007
0.006
- Run data
• Average of data
_
u
w
w
o 0.005
(0
| 0.004
°~ 0.003
0.002
0.001
0.000
i-t
!_!' I
CA-01 IN-13 MI-04 WI-47 WI-43 WI-47 MN-7 MN-12 PA-06 PA-06 IN-24 MN-12 CA-09 OH-43 IN-11 TX-11 Ml-28 IN-11 IN-29 IN-12 PA-46
Foundry ID
Figure 4-3. Filterable PM Emissions (gr/dscf) from Fabric Filters on EIFs at Iron and Steel Foundries
4-25
-------
Other available data include measurements on actual HAP emissions. Data for HAP
metals are available from one test on a wet scrubber and one test on a cyclone device, termed a
skimmer, that serves a PH. The latter test was conducted in 1973, when the quality of scrap was
not as closely controlled as it is today, and therefore emissions would not be characteristic of
those expected in present operations. Emissions tested in either case are not representative of the
best controlled EIF/PH emissions from current foundries. Organic HAP emission data are
available for only two wet scrubbers controlling three EIFs at two foundries; one of these
scrubbers also controls a pouring and cooling line. Collectively, these data are not sufficient to
establish a basis for estimating HAP emissions.
4.4.3 EAF Controls
The use of controls for EAFs in iron and steel foundries for melting is similar to that for
EIFs. The number of EAFs used in ferrous foundries is much smaller than the number of EIFs
used. Arc furnaces are more common in steel than in iron foundries. Tables 4-16 through 4-20
summarize the use of controls on EAFs in 1998. Fabric filters were by far the most common
devices used. Table 4-20 summarizes A/C ratio data for EAF baghouses.
TABLE 4-16. CONTROL CONFIGURATIONS FOR EAFs AT IRON FOUNDRIES'
Operation and type of control
Charging
No control
Filter
Filter
No control
Totals
Melting
Filter3
Filter
Filter
Filter
Tapping
No control
Filter
No control
Filter
Number of
furnaces with
configuration2
10
8
6
4
28
Number of i
foundries with j
configuration
3
4
2
2 i
11
1 Blank responses were interpreted to mean no control and thus were classified as "no control.
2 Information is ranked by the number of furnaces controlled with the given configuration.
3 Filter = fabric filter or cartridge filter (the former make up the majority).
4-26
-------
TABLE 4-17. SPECIFIC CONTROLS ON EAFs AT IRON FOUNDRIES1
Type of control
Furnaces with filter2
Foundries with filter
Furnaces with no control
Foundries with no control
Total number of furnaces
Total number of foundries
Number of operations controlled
Charging
14
6
14
5
28
11
Melting
28
11
0
0
28
11
Tapping
12
6
16 i
5 :
28
11
' Blank responses were interpreted to mean no control and thus were classified as "no control."
: Filter = Fabric filter or cartridge filter (the former make up the majority).
TABLE 4-18. CONTROL CONFIGURATIONS FOR EAFs AT STEEL FOUNDRIES'
Operation and type of control
Charging
No control
Filter
Filter
No control
No control
Totals
Melting
Filter3
Filter
Filter
Filter
No control
Tapping
No control
No control
Filter
Filter
No control
Number of
furnaces with
configuration2
48
34
33
17
3
135
Number of |
foundries with
configuration
23
15
21
9
3
71
' Blank responses were interpreted to mean no control and thus were classified as "no control.
2 Information is ranked by the number of furnaces controlled with the given configuration.
3 Filter = fabric filter or cartridge filter (the former make up the majority).
4-27
-------
TABLE 4-19. SPECIFIC CONTROLS ON EAFs AT STEEL FOUNDRIES'
Type of control
Furnaces with filter2
Foundries with filter
Furnaces with no control
Foundries with no control
Total number of furnaces
Total number of foundries
Number of operations controlled
Charging
67
36
68
35
135
71
Melting
132
68
3
3
135
71
Tapping
50
30
85
41 I
135
71
1 Blank responses were interpreted to mean no control and thus were classified as "no control.
2 Filter = fabric filter or cartridge filter (the former make up the majority).
TABLE 4-20. A/C RATIOS FOR FABRIC FILTERS ON EAFs
A/C ratio
<2
2 to 2.99
3 to 3.99
4 to 4.99
5 to 5.99
>6
Filters in iron foundries
0
9
1
1
0
0
Filters in steel foundries
5
39
10
0
2
2
Source test data for arc furnace PM emissions are available for 10 baghouses used to
control the emissions from 23 EAFs operated by iron and steel foundries. Figure 4-4 is a chart of
outlet gas PM concentration data; repetitive foundry listings in Figure 4-4 indicate a baghouse
that was tested multiple times. Information on the source tests from which data in this figure are
derived is summarized in Appendix F. Average outlet PM concentrations for the ten baghouses
tested ranged from 0.0005 to 0.0044 gr/dscf, except for one baghouse that had a measured
concentration of 0.0080 gr/dscf and another baghouse for which the result of one of two tests
was 0.0066 gr/dscf.
4.4.4 EAF and EIF Capture Systems
Emissions from the different operations in the melting cycle (charging, melting, and
tapping) require different capture techniques. For example, melting emissions can be captured
by a close-fitting lid or hood equipped with a duct, which can be connected to a control device.
4-28
-------
0.012
0.010
0.008
w
§ 0.006
'w
w
UJ
P 0.004
0.002
- Single run value
• Average of run values
0.000
IN-07 IA-09 IA-09 IA-09 AL-11 MN-03 TX-19 OH-01 MI-09 WI-45 IA-17 OH-01 PA-11
Foundry ID
Figure 4-4. Filterable PM Emissions (gr/dscf) from Fabric Filters on EAFs at Iron and Steel Foundries.
4-29
-------
This lid must be removed for charging when the top of the furnace is open and for tapping when
the furnace is tilted to pour the molten metal. Capture systems consist of two general types:
close capture and general capture. Close-capture systems, which are more effective, use
techniques such as side draft hoods, direct evacuation systems, fume rings, and close-fitting
hoods that capture emissions before they escape from the immediate vicinity of the furnace.
These systems require only a small volume of air flow, which is drawn through attached
ductwork to a control device that can be dedicated to specific operations. General-capture
systems employ (1) canopy hoods or total enclosures, both of which can be used with dedicated
control devices but require a higher volume of air flow than close-capture systems, or
(2) building or bay evacuation systems, which also require large volumes of air and must serve
the entire building or a large segment of it.
Information on the use of capture systems in 1998 is given in Tables 4-21 and 4-22.
Most EIF emissions were not captured. Melting emissions from most EAFs were captured,
mainly with close- capture systems, probably because arc furnaces produce more emissions.
Comparing this information with the information on use of controls given previously, most
emissions (from both types of furnace) that were captured were also controlled. The following
sections describe some of the capture mechanisms identified above.
4.4.4.1 Side Draft Hoods. Side draft hoods are used on both EIFs and EAFs. For EIFs,
the side draft hood is located to the side of the furnace (near the top), where it controls emissions
from charging and melting operations and from the tapping spout (see Figures 4-5a and 4-5b).
For EAFs, the side draft hood is mounted on the roof of the furnace to control melting emissions
(see Figure 4-6). The capture system in Figure 4-6 requires a tight fit of the furnace roof so that
emissions can escape only through the spaces between the electrodes and the hood. The roof
hood is not effective when it is removed during charging and tapping. Particulate capture
efficiency ranges between 90 and 100 percent for melting emissions, with a typical efficiency of
99 percent (EPA, 1981). Side draft hoods on EAFs may also be placed to the side of the furnace
to control emissions from charging operations and from the tapping spout.
4-30
-------
TABLE 4-21. USE OF CAPTURE SYSTEMS ON EIFs
AT IRON AND STEEL FOUNDRIES
Capture system type
Close capture1:
Number of furnaces
Number of foundries
Other type2:
Number of furnaces
Number of foundries
No capture3:
Number of furnaces
Number of foundries
Melting furnace operation serviced
Charging
211
66
185
69
1001
334
Melting
261
78
200
84
936
315
Tapping
160 !
53
169
63
1068
353
Total number of furnaces: 1,397 Total number of foundries: 4454
1 Close capture includes side draft hood, fume ring, close-fitting hood, and direct evacuation (melting).
: Other includes canopy hood, draft system or ventilation to a baghouse, area ducting, suction tube, and building
evacuation to a baghouse.
3 No capture includes not reported, roof vent, exhaust fan, lid or cover, or general ventilation.
' The number of foundries in the table totals over 445 because some foundries had multiple configurations.
TABLE 4-22. USE OF CAPTURE SYSTEMS ON EAFs
AT IRON AND STEEL FOUNDRIES
Capture system type
Close capture1:
Number of furnaces
Number of foundries
Other type2:
Number of furnaces
Number of foundries
No capture3:
Number of furnaces
Number of foundries
Melting furnace operation serviced
Charging
32
20
41
18
92
46
Melting
120
62
26
9
17
10
Tapping
33
19
17
11
113
52
Total number of furnaces: 168 Total number of foundries: 814
1 Close capture includes side draft hood, fume ring, close-fitting hood, and direct evacuation.
2 Other includes canopy hood, draft system or ventilation to a baghouse, area ducting, suction tube, and building
evacuation to a baghouse.
3 No capture includes not reported, roof vent, exhaust fan, lid or cover, or general ventilation.
4 The number of foundries in the table totals over 81 because some foundries had multiple configurations.
4-31
-------
i
'rl
p r\Exhaust hood
n
i — i
Figure 4-5a. Side Draft Hood
on EIF (Shaw, 1982).
Exhaust hood
Figure 4-5b. Side Draft Hood
with Blower on EIF (Shaw, 1982).
ELECTRODES (3)
SIDE DRAFT HOOD
FURNACE ROOF
TAP SPOUT
2=3
ELECTRIC ARC FURNACE
SMALL GAP TO
FACILITATE ROOF
MOVEMENT
Figure 4-6. Side Draft Hood on an EAF (EPA, 1983).
4-32
-------
4.4.4.2 Direct Evacuation Control (DEC) System. The DEC system draws exhaust
gases from beneath the roof of an electric furnace. The system consists of a water-cooled or
refractory-lined duct that attaches to the furnace roof, and, when the roof is in place, joins a duct
that is connected to an emission control device (see Figure 4-7 for an example of a DEC on an
EAF). At the connecting point of the two ducts, there is a small gap that allows dilution air to
enter the duct. The gap also allows room for the furnace roof to be elevated and rotated to the
side for charging and for the furnace to be tilted for tapping. The DEC system is only effective
when the furnace roof is in place.
The DEC system provides good emission control with a minimum of energy because the
air volume withdrawn is the lowest of the process emission capture devices (EPA, 1983, p. 4-3).
During melting, a slight negative pressure is maintained within the furnace to effectively
withdraw the emissions through the DEC system. The DEC system withdraws between 90 and
100 percent of the melting emissions from the furnace. A typical particulate capture efficiency
with a properly operated DEC system is estimated to be 99 percent (EPA, 1981).
4.4.4.3 Fume Rings. As shown in Figure 4-8, a suction ring (also known as a fume ring
or lip extraction ring) can be fixed to the top of an EIF to capture emissions during melting. A
fume ring works well when the furnace lid is in place for melting and holding; however, when
the lid is removed for charging, capture is poor. During pouring, capture may not be good even
though the exhaust connection is still in use with the furnace tilted. Consequently, some
facilities use the fume ring for melting emissions in combination with a canopy hood for
emissions during charging and pouring (Shaw, 1982).
4.4.4.4 Close-Fitting Hoods. A close fitting hood is a broad term for capture
mechanisms that are located closer to their emissions sources than canopy hoods, but that do not
fall under the specific categories of side draft hoods and DEC systems. Figure 4-9 shows an
example of two close- fitting hoods on an EAF. In this figure, a rectangular hood that
completely surrounds the electrodes is used to evacuate melting and refining emissions using
minimum exhaust volumes.
4.4.4.5 Canopy Hoods. Figure 4-10 provides an example of a canopy hood system.
The hood is placed as close above the furnace as possible, but allowing clearance for a monorail
or crane charging system and for the vertical electrodes of an EAF, including the upward
4-33
-------
PLAN VIEW
ELECTRODES (3)
DUCT TO CONTROL
DEVICE
—,—,-_ AIR GAP
FURNACE ROOF
TAP SPOUT
REFRACTORY LINED OR
WATER COOLED DUCT
ELEVATION
Figure 4-7. Direct Evacuation System on an EAF (EPA, 1983).
4-34
-------
Figure 4-8. Fume Ring on an EIF (Shaw, 1982).
HOOD
EXHAUSTING
SLAG DOOR
ELECTRODE AREA
ENCLOSED WITH
RECTANGULAR
HOOD
SWIVEL JOINT
HOOD ENCLOSING
TAP SPOUT
TO BAGHOUSE
ANNULAR RING HOOD
SWINGS OVER
FURNACE TOP DURING
CHARGING
ANNULAR RING HOOD IN
PLACE TO COLLECT
CHARGING EMISSIONS
HOOD
ENCLOSING
TAP SPOUT
TO BAGHOUSE
Figure 4-9. Close-Capture Hood System on an EAF (EPA, 1981).
4-35
-------
TO DUST
COLLECTOR
Figure 4-10. Canopy Hood System (EPA, 1981).
movement of the electrodes when the furnace roof is removed. The hood may run only during
charging and tapping stages or may run through the complete melting cycle, and it must be
physically large enough and draw through a large enough volume of air to ensure effective
capture of emissions. Impingement on overhead equipment and cross drafts in the shop can
lower the collection efficiency. Devices such as curtain walls and air curtains may be used to
reduce cross drafts. The particulate capture efficiency of a canopy hood can be 80 to 90 percent,
with the lower figure considered a more typical value when considering potential crossdrafts
(EPA, 1981, p. 86).
4.4.4.6 Total Furnace Enclosure. A total furnace enclosure completely surrounds a
furnace with a metal shell that acts to contain all the charging, melting and refining, and tapping
emissions, as well as to reduce furnace noise and heat radiation outside the enclosure (see Figure
4-11). 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. Particulate capture efficiencies
for total furnace enclosures are estimated to range from 90 to 100 percent (EPA, 1983, p. 4-2).
4-36
-------
Roof
Roof Vent
-S V/>, Roof Trusses /•*x
r-Top Charge Door
\
1 1
Concrete Floor I
A
*— Main Exhaust Duct \
nfln 1
^-" *""^
i Furnace , ^r
\ ' O
Z^**—__ — -•••*' ir^^r-^"
Rear Enclosure Door i ft i
! Ladle i
i i
Shop Floor
_J__-__L
Slag Pit Sr -7.^
~ /
-. Front Charge doors
f
1 ^— Alloy Addition Chute
r^
•~»^^Tapping Exhaust Duct
^\<>'>'v<>'>'>'>'>'>^^S. vLwK^VAMV.*
SIDE VIEW
Front Charge Doors
Top Charge Door
_
nT
To Control Device
Air Curtain Fan .x**
Alloy Addition Chute
(J ~}
^
^^
^
•*» 1
/
ii
ii
Furi
/
JH
•-*.
ace
4-
/
•«»
*^_ '
* Ladle
3
I ,
.1
Z_"
>,
1
i
t
— 1— — Main Exhaust Duct
i,'VDamper
^T
1
1
1
1
1
1
1 Concrete Floor
i
. J i
' ~7 ~ -*
^.Tapping Exhaust Duct
Shop Floor
FRONT VIEW
Figure 4-11. Schematic of a Total Furnace Enclosure (EPA, 1981).
4-37
-------
4.4.4.7 Building and Bay Evacuation. A building or bay evacuation system involves a
closed shop roof with ductwork at the peak of the roof to collect all emissions from the shop
operations (see Figure 4-12). The system requires a large volume of air flow but has a
paniculate capture efficiency of 95 to 100 percent (the typical maximum particulate removal
efficiency is 99 percent) (EPA, 1981, p. 86). Bay evacuation systems can produce an emission
capture efficiency similar to building evacuation (EPA, 1981, p. 86). In bay evacuation systems
(see Figure 4-13), each shop bay is separated from other bays by air locks and/or soundproof
doors, and each bay is evacuated separately (EPA, 1981, p. 86).
PARTITION
TO SHIELD
AREA
FROM
CROSS
DARFTS
TO DUST
COLLECTOR
TO DUST
COLLECTOR
Figure 4-12. Building Evacuation System (EPA, 1991).
4-38
-------
TO
CONTROL
DEVICE
TO
CONTROL
DEVICE
ADDITIONAL
SHOP BAYS FOR
LADLES, CASTING,
OTHER
ACTIVITIES
LADLE CAR DOOR
Ml
SLAG CAR
LADLE CAR
Figure 4-13. Schematic of a Bay Evacuation System (EPA, 1980).
4.5 POURING, COOLING, AND SHAKEOUT
Controls for organic compound emissions are rarely used. Most reductions for these
compounds are achieved by ignition of the mold vents, which occurs spontaneously on
automated lines and is commonly done by manual ignition in floor or pit pouring stations.
In 1998, one foundry employed a thermal oxidizer on one of its pouring/cooling lines.
This line used exclusively sand molds that were chemically bonded by the phenolic urethane
cold-box process. According to the foundry operator, fumes from this line were substantially
greater than fumes from their green sand line. The oxidizer was operated at 1,500 °F. VOC
emissions in the exhaust gases were reported to be 2.3 ppm. No measurements were taken for
inlet gases. By comparison, emissions of benzene measured at various locations in a
pouring/cooling line in an EPA test by a portable test unit that used a gas chromatography/mass
spectrography analysis system were typically 3 ppm (EPA, 1999). Benzene was the primary
HAP detected in this test. VOCs were not measured, so no direct comparisons in emissions are
4-39
-------
possible between the green sand lines for which mold vent ignition was used and the chemically
bonded mold line in which thermal oxidation was used.
Additionally, two foundries employed a carbon adsorption system to control emissions
from their pouring/cooling lines. One of these lines used chemically bonded sand molds; the
other line used green sand molds with chemically bonded cores.
In contrast to organic emissions, controls for PM emissions are almost universal,
especially in shakeout processes, where most of these emissions are produced. Shakeout PM
emissions are mainly large particles that are mostly sand. These emissions are almost always
controlled, usually by fabric filters, but often by wet scrubbers or other devices. Shakeout
usually occurs in a partially enclosed area or in a device designed for sand/casting separation
such as a rotating cylinder or vibrating conveyor in which the sand is screened from the castings.
Emissions of PM from pouring are quite different from shakeout emissions, consisting
mainly of metal and metal oxide fumes. Emissions of PM from cooling are generally low and
consist mainly of condensible byproducts (i.e., soot) generated by the incomplete combustion of
organic material (seacoal, chemical binders, and other additives) contained in the mold and core
sand. Pouring and cooling emissions are often captured (by such devices as canopy or side draft
hoods), but they are not always controlled.
Control devices for PCS operations may be dedicated to one or more of these operations
but also may serve a variety of emission sources. Their use varies greatly from one foundry to
another. Emission data from control devices will therefore reflect the fact that many types of PM
are present in inlet streams and also that inlet PM loadings will vary substantially depending on
air flow requirements in serving the various emission sources.
As shown in Table 4-23, fabric filters and cartridge filters were the most common control
devices used for shakeout and controlled approximately half of the stations. Table 4-24 provides
the A/C ratios for these filters and shows that the ratios were generally higher than those for
filters used on melting furnaces. Similarly, the pressure drops for wet scrubbers (Table 4-25)
show levels much lower than those used on cupolas.
4-40
-------
TABLE 4-23. CONTROL DEVICES USED ON SHAKEOUT STATIONS
Control device
Fabric or cartridge filter
No control
Wet scrubber
Other1
Total
Number of shakeout stations
602
384
161
9
1,156
Number of foundries
360
225
79
7
569:
1 Other includes cyclone, rotoclone, and "wet system."
1 Total number of foundries reporting shakeout stations. The sum for each type of control is greater than 569
because several foundries had multiple configurations.
TABLE 4-24. A/C RATIOS FOR FABRIC FILTERS
ON SHAKEOUT STATIONS
A/C ratio, ft/min
<2
2 to 2.99
3 to 3.99
4 to 4.99
5 to 5.99
6 to 6.99
^ 7
Number of filters
38
37
34
23 I
61
57 '
91
TABLE 4-25. PRESSURE DROPS FOR WET SCRUBBERS ON
SHAKEOUT STATIONS
Pressure drop, inches of water
2 to 4.9
5 to 5.9
6 to 6.9
7 to 7.9
8 to 9. 9
10 to 13.5
Number of scrubbers
7
22
21
15
11
19
4-41
-------
Data for PCS PM emissions consist of outlet concentration measurements on 33 fabric
filters and 8 wet scrubbers at 21 foundries. The data for the respective devices are shown in
Figures 4-14 and 4-15. The repetitive foundry listings in Figure 4-14 indicate separate fabric
filter control systems, except for one fabric filter at WI-43 that was tested twice. Repetitive
foundry listings in Figure 4-15 indicate separate wet scrubber control systems, except for one
wet scrubber at TN-9 that was tested three times. A summary of the data is also given in Tables
4-26 and 4-27, which also identify other emission sources served by the devices.
4-42
-------
0.014
0.012
0.01
- Single run value
• Average of runs
s
o
0.008
I 0.006
0.004
0.002
f
. I . « •
. . ! I : I
r
i
•
Foundry ID
Figure 4-14. Filterable PM Emissions (gr/dscf) from Fabric Filters on PCS at Iron and Steel Foundries.
(does not include OH-22 at 0.03 gr/dscf- see Appendix G for details)
4-43
-------
u.uio -
fl 014 -
0 017 -
fi m
u
W5
T3
A
en
5 0 008 -
*5
»a
'E
w
§
&< 0 006 -
0 OOd.
OO09
n -
— Single run value
• Average of run values
•
•
•
i
•
• '
•
•
•
1
•
; i • !
•
•
•
•
1 1
•
•
•
•
1
•
•
TN-9 TN-9 TN-9
TN-9 WI-28 WI-47 TN-9 OH-22 WI-47 OH-22
Foundry ID
Figure 4-15. Filterable PM Emissions (gr/dscf) from Scrubbers on PCS at Iron and Steel Foundries
4-44
-------
TABLE 4-26. PCS FABRIC FILTER OUTLET CONCENTRATION AND SERVICE DATA
Foundry
1 ID
| IN- 13
WI-43
WI-43
WI-43
IN- 13
WI-43
WI-43
WI-43
WI-42
WI-42
WI-1
| WI-43
IA-17
MN-12
WI-1
SC-7
IN-29
OH-48
WI-43
WI-1 5
MN-12
OH- 13
TX-19
PM
(gr/dscf)
0.00029
0.0005
0.0005
0.0005
0.00055
0.0008
0.0009
0.0009
0.001
0.001
0.001
0.0012
0.0013
0.0015
0.0015
0.00185
0.0019
0.0019
0.0022
0.0024
0.0027
0.0028
0.0030
Design information for fabric filters
acfm
150,000
143,000
60,000
101,000
85,500
60,000
60,000
60,000
150,000
150,000
198,000
180,000
50,000
12,400
51,000
60,000
51,000
10,000
96,000
75,000
10,000
65,000
30,000
A/C ratio, ft/min
4.5
5.6
7.1
7.1
4.4
7.1
7.1
7.1
6.6
6.5
NR
7.1
1 .4 (cartridge)
9.4
NR
NR (cartridge)
5.5
9.1
8.9
7.0
5.8 (cartridge)
6.5
6.5
Material
polyester
polyester felt
polyester felt
polyester felt
polyester
polyester felt
polyester felt
polyester felt
polyester felt
polyester felt
NR*
polyester felt
cellulose
polyester
NR
NR
polyester felt
polyester
polyester
polyester
felt
polyester
polyester
Cleaning type
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
NR
pulse jet
pulse jet
shaker
NR
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
Operations served
shakeout lines 1 & 2; return sand system
cooling and grinding
shakeout and grinding '
cooling and sand handling
shakeout lines 3 & 4
cooling, shakeout, grinding
cooling, shakeout, grinding
cooling, shakeout, grinding - line 1
2 pouring/cooling lines; 2 cast cooling lines
shakeout
5 shakeout/cast cooling lines
cooling, shakeout, grinding - line 6
shakeout and sand transfer with line 802
pouring
pouring/cooling lines 1 & 5; sand mullor
pouring/cooling
shakeout
shakeout
cooling and shakeout
pouring, cooling, shakeout
cooling
pouring, cooling, shakeout, miscellaneous
shakeout
(continued)
4-45
-------
Table 4-26. (continued)
Foundry
ID
IA-17
Wl-1
SC-7
IA-17
OH-43
IN- 11
IA-17
! IA-17
WI-1
MI-4
OH- 13
TX-11
IA-17
OH-43
IN- 11
AZ-04
OH-22
PM
(gr/dscf)
0.0031
0.0032
0.0038
0.0038
0.0039
0.0041
0.0043
0.0043
0.0044
0.0047
0.0053
0.0056
0.0065
0.0066
0.0076
0.0109
0.0294
Design information for fabric filters
acfm
140,000
198,000
20,000
375,000
50,000
174,000
1 10,000
1 10,000
101,000
75,000
65,000
170,300
35,600
50,000
180,000
45,000
80,000
A/C ratio, ft/min
7.4
NR
NR (cartridge)
5.0
5.2
3.6
5.9
5.9
NR
5.1
6.5
2.0
7.4
5.2
7.4
6.1
6.4
Material
singed polyester
NR
NR
polyester
polyester
polyester
polyester
polyester
NR
polyester
polyester
polyester
singed polyester
polyester
polyester
polyester
polypropylene
Cleaning type
pulse jet
NR
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
NR
pulse jet
pulse jet
shaker
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
Operations served
shakeout and sand transfer with Line 803
pouring/cooling lines 2 & 4; sand handling
shakeout
shakeout and sand transfer with line 801
cooling; bond and sand storage
shakeout
shakeout and sand transfer with line 802
shakeout and sand transfer with line 802
pouring/cooling Lines 2 & 4; sand handling i
shakeout
pouring, cooling, shakeout; miscellaneous
pouring, cooling, shakeout
shakeout and sand transfer with line 803
shakeout and sand cooling
shakeout
shakeout
shakeout
* NR = Not reported.
4-46
-------
TABLE 4-27. PCS WET SCRUBBER OUTLET CONCENTRATION AND SERVICE DATA
Foundry
ID
TN-9
! TN-9
TN-9
TN-9
WI-28
WI-47
TN-9
OH-22
WI-47
OH-22
PM, gr/dscf
0.0011
0.0023
0.0024
0.0043
0.0047
0.0052
0.0055
0.0055
0.0064
0.012
Design information for wet scrubbers
acfm
75,000
75,000
45,000
54,000
60,000
73,500
75,000
99,000
32,000
104,000
Type
cyclonic
cyclonic
centrifugal
centrifugal
venturi
venturi
cyclonic
venturi
venturi
venturi
Ap, inches water
column
13.5
13.5
6.0
5.8
6.0
13
13.5
3.5
13
3.2
Liquid-to-gas
ratio,
gal/1,000 acf
8
8
2.5
2.5
5
2
8
not reported
2
not reported
Operations served
shakeout
shakeout
shakeout
pouring and cooling
cooling, shakeout \
induction furnace, pouring, cooling
shakeout
shakeout
shakeout
shakeout
4-47
-------
4.6 SUMMARY OF FEDERAL AND STATE REGULATIONS
The Federal government has not established NSPS or other air standards specific to the
metal- casting industry. Depending on the wastes produced or managed at the foundry,
foundries may be subject to the hazardous waste rules under the Resource Conservation and
Recovery Act.
Applicable State regulations were reviewed for the six states with the highest foundry
metal melting rates. These State regulations are summarized below.
4.6.1 PM Emission Limits
Michigan has PM standards specific to foundry melting furnaces and sand handling.
Existing "production" (captive) cupolas have a PM concentration limit ranging from 0.40 Ib
PM/1,000 Ib gas (approximately 0.2 gr/dscf) for cupolas with melting capacities less than 10
tons/hr to 0.15 Ib PM/ 1,000 Ib gas (approximately 0.08 gr/dscf) for cupolas with melting
capacities greater than 20 tons/hr. Existing "jobbing" cupolas have a PM concentration limit of
0.40 Ib PM/1,000 Ib gas (approximately 0.2 gr/dscf). New cupolas have an emission factor PM
limit ranging from 1.8 to 0.7-lb PM/ton metal charged, with the 0.7 Ib/ton factor applying to all
cupolas with melting capacities over 15 tons/hr. EAF melting and sand handling both have a
concentration PM limit of 0.10 Ib PM/1,000 Ib gas (approximately 0.05 gr/dscf).
Wisconsin also has PM standards specific to foundry melting furnaces. Cupolas have a
PM concentration limit of 0.45 Ib PM/1,000 Ib gas (approximately 0.24 gr/dscf). Both EAFs and
EIFs have a PM concentration limit of 0.10 Ib PM/1,000 Ib gas (approximately 0.05 gr/dscf).
Indiana has PM standards specific to foundry operations. Cupolas are provided a PM
concentration limit of 0.15 gr/dscf. All other foundry operations cannot discharge into the
atmosphere any gases with PM concentrations exceeding 0.07 gr/dscf.
Ohio has mass PM emission limits based on the process weight rate capacity of a generic
PM emission source. These PM emission limits can be converted into emission factor limits
based on the melting capacity of the furnace. The emission factors vary widely based on the
furnace melting capacity: a 1-ton/hr melting furnace would have an effective PM emission
factor limit of approximately 4 Ib/ton, a 10-ton/hr melting furnace would have an effective PM
emission factor limit of approximately 2 Ib/ton, and a 100-ton/hr melting furnace would have an
effective PM emission factor limit of approximately 0.5 Ib/ton.
4-48
-------
Illinois has similar generic PM emission limits based on process weight rates. These PM
emission limits can be converted into emission factor limits based on furnace melting capacities.
The emission factors are identical to the Ohio State PM limits for sources constructed prior to
1972. For sources constructed or modified since 1972, examples of the PM emission factor
limits follow: A 1-ton/hr melting furnace would have an effective PM emission factor limit of
approximately 2.6 Ib/ton, a 10-ton/hr melting furnace would have an effective PM emission
factor limit of approximately 0.87 Ib/ton, and a 100-ton/hr melting furnace would have an
effective PM emission factor limit of approximately 0.3 Ib/ton.
Alabama also has PM limits based on process weights. Examples include a limit of
approximately 4.7 Ib/ton for a 1-ton/hr melting furnace, 2.5 Ib/ton for a 10-ton/hr melting
furnace, and 0.4 to 0.5 Ib/ton for a 100-ton/hr melting furnace (depending on the county).
4.6.2 Opacity Emission Limits
Opacity emission limits were found for five states (Alabama, Wisconsin, Michigan,
Indiana, and Ohio). These limits generally apply to general roof vents that may contain fugitive
emissions from various sources throughout the foundry. Four of the states (Alabama, Wisconsin,
Michigan, and Ohio) have 20-percent opacity limits. Indiana has a 30-percent or 40-percent
opacity limit, depending on the location of the source (by county). Most of these limits allow
one 6-minute average per hour to be above the specified limit, but must be below a secondary
opacity limit (60 percent for most states except 40 percent for Alabama).
4.6.3 CO Emission Limits
Four states (Alabama, Michigan, Wisconsin, and Indiana) have CO limits that
specifically address cupola emissions. Michigan requires cupolas with a melting capacity of
20 tons/hr or more to be equipped with an afterburner control system that reduces CO emissions
from the cupola by 90 percent. Alabama and Wisconsin require that cupola emissions be
incinerated at 1,300 °F for 0.3 seconds. Indiana requires gas streams from cupolas with a
melting capacity of 10 tons/hr or more to be burned using either a direct-flame afterburner, a
boiler, or equivalent control system.
4-49
-------
4.7 REFERENCES
Brown, R., 2000. Exxon Chemical Corporation, Houston, TX. Private communication to J. H.
Maysilles, U.S. Environmental Protection Agency, September.
Buonicore, A. J., and W. T. Davis (eds.), 1992. Air Pollution Engineering Manual. New York:
Van Nostrand Reinhold.
Casting Development Centre, 1997. Report on Environmental Emission Testing Using Foseco
Mold Coating Cereamol 915-9. Prepared for Foseco, Inc., by the Casting Development
Centre, Sheffield, England.
Shaw, P.M., 1982. "CIATG Commission 4 Environmental Control: Induction Furnace
Emissions." Commissioned by F. M. Shaw, British Cast Iron Research Association, Fifth
Report, Cast Metals Journal, vol. 6, p. 10-28.
Stone, J.A., 2000. Delta Resins and Refractories, Delta-HA. Private communication to J. H.
Maysilles, U.S. Environmental Protection Agency, September.
U.S. Environmental Protection Agency, 1980. Electric Arc Furnaces in Ferrous Foundries -
Background Information for Proposed Revisions to Standards. Office of Air Quality
Planning and Standards, Research Triangle Park, NC. EPA-450/3-80-020a.
U.S. Environmental Protection Agency, 1981. Summary of Factors Affecting Compliance by
Ferrous Foundries. Office of General Enforcement, Washington DC. EPA-340/1-80-
020.
U.S. Environmental Protection Agency, 1983. Electric Arc Furnaces and Argon-Oxygen
Decarburization Vessels in the Steel Industry - Background Information for Proposed
Revisions to Standards. Office of Air Quality Planning and Standards, Research Triangle
Park, NC. EPA-450/3-82-020a.
U.S. Environmental Protection Agency, 1991. Control Technologies for Hazardous Air
Pollutants. Office of Research and Development, Washington, DC. EPA-625/6-91-014.
U.S. Environmental Protection Agency, 1998. Compilation of Information from Questionnaire
Forms Submitted by Iron and Steel Foundries to the U.S. EPA Office of Air Quality
Planning and Standards. Office of Air Quality Planning and Standards, Research
Triangle Park, NC.
U.S. Environmental Protection Agency, 1999. Iron and Steel Foundries Manual Emissions
Testing of Cupola Wet Scrubber at General Motors Corp., Saginaw, Michigan. Office of
Air Quality Planning and Standards, Research Triangle Park, NC. EPA-454/R-99-025A
andEPA-454/R-99-025B. July.
4-50
-------
5.0 BASELINE EMISSIONS AND CONTROL OPTIONS
This chapter describes the methodology used to develop nationwide HAP emission
estimates for iron and steel foundries.
5.1 GENERAL APPROACH FOR ESTIMATING HAP EMISSIONS
Facility-specific data were available for both iron and steel foundries as a result of the
detailed information collection request (ICR) conducted by the EPA (EPA, 1998a). With the
availability of these data, which covered a large number of different foundry operations and a
large number of processes for each operation, model plants were not used. Instead, HAP
emissions were estimated for each foundry based on its unique configuration (melting furnace
type, type of metal melted, mold type, etc.). Average HAP emissions factors (emissions
normalized by metal melting rate, which is also assumed to be the pouring rate) were developed
for the different types of processes used by the foundries. This modeling approach accounts for
differences in the emissions based on the type of metal melted, the type of processes used (e.g.,
type of melting furnace, use of scrap preheating or metal treatment, use of cores, etc.), and the
type (or absence) of APCD for each process.
By applying average emission factors to facility-specific data (production rates and
process sequences), a direct correspondence (or weighting) of emissions to process types is
achieved. A given foundry may have actual emissions that vary significantly (e.g., by a factor of
2 or more) from the emissions predicted using the average emission factors. There are many
factors that can influence the process-specific emissions, such as size and configuration of
castings, that are not accounted for in the emission estimation methodology described in this
chapter. These factors can contribute to inaccuracies in the emissions estimated for a specific
plant. However, when summing all individual foundry emission estimates to calculate the
nationwide emission estimates, the inaccuracies at the foundry level (high and low facility
emission estimates) tend to cancel out. Thus, the methodology described in this chapter is
anticipated to provide the most accurate and technically defensible estimate of nationwide
5-1
-------
emissions from the foundry industry. This method also provides a reasonable estimate of the
HAP emissions at the foundry level, although the foundry level emission estimates are likely to
have a greater uncertainty than the nationwide emission estimates.
5.2 SUMMARY OF EMISSION FACTORS FOR PRIMARY FOUNDRY
OPERATIONS
The following sections provide a summary of the emission factors used for each of the
primary foundry processes. Appendix B contains more detailed documentation of the
development of emission factors for mold- and core-making operations. Appendix C contains
more detailed documentation of the development of emissions factors for melting and pouring,
cooling, and shake-out operations from source test data or published literature.
5.2.1 Emission Factors for Mold and Core Making and Coating
The primary sources of HAPs in the mold- and core-making operations are the chemical
binders used to help "set" the molds and cores and coatings applied to the molds and cores to
improve surface finish of the cast and to aid in the separation of the cast part from the mold.
Therefore, mold- and core-making emissions are estimated only for foundries that use coatings
or binders that contain HAPs in their mold- and core-making processes. Green sand systems that
do not use cores are expected to have minimal HAP emissions during the mold-making process.
Emission factors for mold and core making were developed based on the chemical make-
up of the different binder systems and estimates of the percentage of those chemicals that are
emitted during the mold- or core-curing process. Few direct emission measurement data were
available to estimate emissions from mold and core making; the percentage of chemicals emitted
during the mold- or core-curing process was estimated primarily based on guidance provided by
the AFS and the Casting Industry Suppliers Association (CIS A) (AFS and CIS A, 1998). The
emission factors are most directly related to the amount of each HAP added to the mold or core.
Where available, actual chemical usage rates reported by a foundry and submitted on the MSDS
(which contained HAP concentration information) were used to determine the amount of each
HAP used at the foundry. Generally, this detailed information was not available, and emission
estimates had to be made based on typical binder system characteristics (e.g., the typical binder-
to-sand ratio for a given binder system and its typical HAP contents). Chemical usage rates were
generally estimated based on the reported sand usage rates for a given mold or core line and the
average binder chemical-to-sand ratio for the binder system used with that mold or core-making
5-2
-------
line. Emission factors were also developed based on the typical HAP content for a given binder
system. More detail regarding the development of these emission factors are provided in
Appendix B. The average chemical-to-sand ratio and the default emission factors for each
binder system are provided in Table 5-1.
The solvents used for mold and sand coatings are assumed to be 100 percent emitted. For
some flammable organic solvent coatings, the solvent is ignited and allowed to bum off (through
the light-off process). This process should effect some reduction in coating solvent emissions,
but few data are available to quantify the emission reduction it achieves. Consequently, for the
purposes of estimating baseline emissions, 100 percent of the HAP in the liquid portion of the
coating solvent is assumed to be emitted during the coating process.
5.2.2 Emission Factors for Melting Operations
Metal HAP emissions from melting operations (which include scrap preheating, melting
furnaces, and inoculation) are estimated from PM emission data along with estimates of metal
composition. PM test data reported in the literature, EPA source test data, and test data reported
by the industry in response to EPA's detailed ICR were compiled, and emission factors, based on
tons of metal melted (or tons of metal poured), were calculated (see Appendix C). A summary
of the PM emission factor data for melting operations is provided in Table 5-2. A summary of
the HAP content of the various PM emission sources appears in Table 5-3. Values used from
these tables to estimate baseline emissions are presented in bold.
Organic HAP emission data for cupolas that use afterburners indicate negligible organic
HAP emissions, with most HAPs present below analytical detection limits (see EPA source test
data in Appendix C). Consequently, no organic HAP emissions are estimated for cupolas that
employ afterburning for the purposes of the baseline emission estimate.
There are no data for organic HAP emissions from cupolas that do not use afterburning.
However, it is anticipated that coke combustion with inadequate oxygen (as indicated by percent
levels of CO in the exhaust gas of cupolas prior to afterburning) will generate benzene and other
organic HAP emissions. Certain fluid catalytic cracking units (FCCUs) in the petroleum refinery
industry employ "partial" combustion of coke as they regenerate the FCCU catalyst. These
regenerators burn coke under similar oxygen-deficient conditions and temperatures, resulting in
similar percent levels of CO in the exhaust gas as found in cupolas that do not use afterburning.
The estimated organic HAP emission factor reported for FCCU incomplete combustion
5-3
-------
TABLE 5-1. AVERAGE CHEMICAL-TO-SAND RATIOS AND EMISSION FACTORS
USED IN MOLD- AND CORE-MAKING EMISSION ESTIMATES
Binder system
Acrylic/epoxy/SO-,
Furan hot box
Furan no-bake
Furan/SO:
Furan warm box
Phenolic baking
Phenolic ester
nobake
Phenolic ester cold
box
Phenolic hot box
Phenolic no-bake
Phenolic-Novolac
flake
Phenolic urethane
no-bake
Chemical-
to-sand
ratio, Ib
chemical/
ton sand
34.
40.
24.
30.
32.
30.:
33.
32.
30.:
27.
50.
25.
Emission estimates, Ib chemical emitted/100 Ib binder1 chemicals
Phenol
0.
0.4
0.08
0.08
0.25
0.164
3.6
0.066
Formal-
dehyde
0.15
0.0014
0.022
0.02
0.05
0.01
0.01
0.1
0.0068
0.
0.0011
Naphtha-
lene
0.16
Cumene
5.0
0.051
Xylene
0.016
Methanol
10.2
1.1
10.
0.
10.7
0.
Glycol
ethers
0.
Dimethyl
phthalate
20.2
Methyl
ethyl
ketone
0.9
TEA
(continued)
5-4
-------
Table 5-1. (continued)
Binder system
Phenolic urethane
cold box - resin
plus coreactant
Phenolic urethane
cold box - gas
Urea formaldehyde
Chemical-
to-sand
ratio, Ib
chemical/
ton sand
30.
3.
30. 2
Emission estimates, Ib chemical emitted/100 Ib binder1 chemicals
Phenol
0.066
Formal-
dehyde
0.0011
0.02
Naphtha-
lene
0.09
Cumene
0.029
Xylene
0.009
Methanol
Glycol
ethers
Dimethyl
phthalate
Methyl
ethyl
ketone
TEA
100.
1 Based on typical chemical composition data and fraction emitted estimates as described in Appendix B.
: Values are based on nominal values for other systems; not enough information is available for these systems to establish values.
Source: EPA, 1998a.
5-5
-------
TABLE 5-2. SUMMARY OF PM EMISSION FACTORS FOR
MELTING FURNACE OPERATIONS*
Emission category/
source of data
Basis of
reported values
Range of
emissions
factors,
Ib/ton
Cupolas controlled with wet scrubbers
GM - Saginaw (EPA, 1999b)
ICR PM Tests
AP-42 (EPA, 1995)
4 Run EPA source test
1 1 Source tests
4 Values for different scrubber types
0.038-0.21
0.090- 1.46
0.08-5.0
Cupolas controlled with fabric filters
Waupaca - Tell City
(EPA, 1999a)
ICR PM Tests
AP-42 (EPA, 1995)
2 of 3 Run EPA source test
3 Source tests
As reported
.010-0.017
0.030-0.082
0.70
Cupolas uncontrolled (or prior to controls)
Waupaca - Tell City
(EPA, 1999a)
GM -Saginaw (EPA, 1999b)
ICR - Baghouse catch
Kearney (1971)
3 Run EPA source test
4 Run EPA source test
Data for 17 cupola
Data for 24 cupola
3.45-9.7
3.6-4.9
8.14-64.1
7.5-66.3
EAF melting controlled with fabric filters
ICR PM Tests
EAF -BID (EPA, 1980)
4 Source tests
Data for 1 1 EAF
0.037-0.56
0.052-0.69
EAF melting uncontrolled
ICR PM Tests
ICR - Baghouse catch
Kearney (1971)
1 Source test (3 runs)
Data for 13 EAF
Data for 19 EAF
20.2-25.9
3.3-29.5
4.0-40.0
EAF charging & tapping uncontrolled
EAF -BID (EPA, 1980)
EAF steel - BID (EPA, 1983)
As reported
As reported
1.4 iron, 1.6
steel
1.6-2.0
Induction furnace with PM control
ICR PM Tests
AP-42 (EPA, 1995)
5 Source tests
As reported
0.080-0.67
0.20
Induction furnaces uncontrolled
ICR PM Tests
ICR - Baghouse catch
BCIRA(Shaw, 1982)
2 Source tests
Data for 8 furnaces
Data for 1 4 furnace tests
0.44 - 8.94
0.33-4.0
0.26-3.3
Median
emissions
factor,
Ib/ton
0.110
0.56
0.077
7.7
4.1
24.0
21.9
0.15
0.15
23.9
8.4
12.7
1.4
0.13
1.75
0.62
Average
emissions
factor,
Ib/ton
0.117
0.580
3.0
0.014
0.063
0.70
7.0
4.3
26.1
30.2 |
0.22
0.23
25.7
11.0 i
13.8
1.6
1.8
0.30 \
0.20
4.7
2.0
0.9
Emissions factors selected for estimating baseline emissions are presented in bold.
5-6
-------
TABLE 5-3. IRON FOUNDRY HAP METAL CONTENT OF PM
Source
Reference
Cupola Melting Furnace
Uncontrolled
Uncontrolled
Wet scrubber
Wet scrubber
Wet scrubber
Wet scrubber
Wet scrubber
Baghouse
EPA source test at GM (EPA, 1999b)
EPA source test at Waupaca (EPA, 1999a)
EPA source test at GM (EPA, 1999b)
Euvrard (EPA methods) - source test (Euvrard, 1992)
Auburn Foundry (#31)- ICR
CM1 Cast Parts (#77) - ICR
Blanchester(#140)-ICR
EPA source test at Waupaca (EPA, 1999a)
EAT Melting Furnace
Dust analysis
1980 EPA EAF proposal BID (ref. 48)
Induction Melting Furnace
Uncontrolled
Uncontrolled
Uncontrolled
Dust analysis
Dust analysis
CERP source test report (CERP, 1998)
Auburn Foundry (#31)- ICR
British study - source test (Shaw, 1982)
EPA Air Emissions Species Manual (1990)
Brillion Iron Works (#201 ) - ICR
Scrap Preheater
Uncontrolled
Cyclone
Brillion Iron Works (#201) - ICR
Brillion Iron Works (#201) - ICR
Inoculation/Metal Treatment
Uncontrolled
Baldwin and Westbrook (1983)
Mn, %
4.7
2.1
7.3
3.3
2.2
0.9
2.0
1.3
0.7
0.5
1.7
0.4
0.91
1.3
0.06
Pb, %
0.8
1.9
1.1
0.9
0.3
2.2
1.1
0.5-2
0.5
0.3
0.5
0.8
0.38
0.69
>0.1
HAP2
metals, %
5.5
4.1
8.5
6.52
5.9
2.5
5.1
2.5-4.1 [3.3]
2.0
1.0
1.1-1.3
2.4
1.2
1.4
2.7
>0.17
1 HAP contents used to develop baseline emission estimates are presented in
2 Vast majority was manganese.
bold.
regeneration is 0.0020 Ib HAP/lb coke burned (EPA, 1998b). Based on the similarities in these
coke combustion units, the 0.0020 Ib HAP/lb coke burned emission factor was used to estimate
the organic HAP emission rate for cupolas not controlled by afterburning. The coke combustion
rate for a typical cupola was estimated from process data collected during an EPA source test
(EPA, 1999a) and is 140 Ib coke/ton of metal melted. Consequently, the organic HAP emission
factor for cupolas that do not use afterburning is 0.28 Ib HAP/ton metal melted (0.0020 x 140).
This emission factor (0.28 Ib/ton metal) was used to estimate the total organic HAP emission rate
from cupolas without afterburning. Because benzene emissions are a potential concern from
mold PCS operations, a benzene emission factor was also estimated for cupolas without
afterburners. The benzene content of the total organic HAP emissions for partial combustion
5-7
-------
FCCU catalyst regeneration was always less than 10 percent, so an emission factor of 0.028
Ib/ton was used, a relative high-end estimate for benzene emissions from cupolas without
afterburning.
5.2.3 Emission Factors for PCS
Emissions data reported in the literature (primarily Scott, Bates, and James, [1976]; EPA,
1998a; EPA, 1998c; and CERP, 1998) were compiled to develop emission factors for PCS in
terms of tons of metal poured. When emissions were reported separately for pouring, for
cooling, and for shakeout, the emissions were summed to develop an emission factor for PCS as
a system; however, some of the data compiled may represent emissions from a single component
of the PCS system. Table 5-4 provides a summary of the organic HAP emission factors for PCS.
The CERP emission factors were selected for use in estimating nationwide emissions because the
CERP data appeared to provide generally median HAP emission estimates for individual and
total HAP.
The organic HAP emission factors presented in Table 5-4 are based on data from
automated PCS lines. Automated PCS lines are the most prevalent, accounting for over 95
percent of the production capacity of the anticipated major source iron foundries. The mold
vents in automated PCS lines generally ignite spontaneously (auto-ignite). Floor and pit molding
is generally used at foundries that produce very large cast parts. The vent gases from these large
molds do not always auto-ignite. It is typical industry practice to ignite the mold vents that do
not auto-ignite. This mold vent "flame" is anticipated to destroy organic HAPs in the vent gases,
but the destruction efficiency is unknown. Consequently, the organic emissions factors
presented in Table 5-4 from pouring and cooling are considered to be representative of emissions
from operations where the mold vents either spontaneously ignite or are manually ignited.
Additionally, the emission factors for PCS systems presented in Table 5-4 were
developed from data of sand mold systems and may be overly conservative for certain casting
operations that do not use much sand, e.g., permanent mold casters, centrifugal casters, and
investment casters. The organic HAP emissions from PCS lines at permanent and centrifugal
casting operations are expected to be less than the organic HAP emissions from sand mold
systems because of the lower sand-use rates for permanent and centrifugal casting operations.
However, no data are available to develop emission factors that may be more relevant to these
systems. Additionally, the sand that is used in permanent and centrifugal
5-8
-------
TABLE 5-4. SUMMARY OF ORGANIC HAP EMISSION SOURCE TEST
RESULTS FOR PCS
Source/HAP
Emission factor, Ib/ton metal poured
CERP
(CERP, 1998)
GM-
Saginaw1
(EPA, 1998c)
WI study
green sand
(RMT, 1995)
WI study
no-bake
(RMT, 1995)
Scott, Bates,
& James
Pouring
Benzene
Toluene
Formaldehyde
Total organic HAPs
0.00219
0.00105
0.000138
0.0050
0.0054
0.0011
0.000354
0.0082
0.0057
0.0032
0.011
0.0059
Cooling
Benzene
Toluene
Formaldehyde
Total organic HAPs
0.0349
0.0189
0.00173
0.078
0.14
0.033
0.0035
0.22
0.045
0.0014
0.032
0.0031
0.0932
0.0272
0.00152
0.1522
Shake-out
Benzene
Toluene
Formaldehyde
Total organic HAPs
0.0268
0.0221
0.0257
0.20
0.022
0.012
0.0026
0.065
0.0083
0.0039
0.0053
0.0008
PC&S Combined
Benzene
Toluene
Formaldehyde
Total organic HAPs
0.0639
0.0421
0.0276
0.283
0.167
0.046
0.064
0.294
0.059
0.0085
0.048
0.0098
0.0932
0.0272 ;
0.00152
0.1 522
1 Values reported by General Motors using their gas chromatography method.
2 Median values from Scott, Bates, and James (1976). Includes pouring/cooling emissions only.
5-9
-------
casting operations is typically chemically bonded (the sand is used for cores). Even with the low
sand-use rates, there is adequate organic material in the sand, due to the chemical binders, to
generate emissions of the magnitude measured for sand mold systems. Therefore, the emission
factors presented in Table 5-4 were applied to all sand mold systems except expendable pattern
casting (EPC, or the lost-foam process).
Limited test data are available to characterize organic HAP emissions from PCS lines
associated with EPC operations (Twarog, 1991). These data suggest that the organic HAP
emissions from EPC operations may be significantly higher than from other casting operations.
The emission factors developed for PCS lines associated with EPC operations from these limited
data are summarized below in Table 5-5.
TABLE 5-5. EMISSION FACTORS DEVELOPED FOR PCS
LINES ASSOCIATED WITH EPC OPERATIONS
HAP
Benzene
Styrene
All other HAPs
Emission factor, Ib/ton
0.33
0.58
0.11
Particulate matter (PM) and metal HAP emission estimates were developed for PCS lines
by evaluating emissions data for captured vent streams that were dedicated to only one of the
three components of the PCS line (i.e., pouring only, cooling only, or shake-out only). Many
systems combine two or more of the PCS component emissions to a single exhaust vent or
control device and combine PCS component emissions with other foundry operation emissions
(e.g., sand-handling, shot-blasting, or grinding operations). The PM emission factors developed
for the individual components provided PM emission estimates that compared reasonably well
with the PM emissions measured from combined PCS component vent streams. The primary
reason for limiting the PM emission data to individual PCS components for estimating baseline
emissions is that the metal HAP content of the different component emissions vary widely. The
metal HAP content of the emissions from each PCS component was determined from measured
HAP emissions or captured baghouse dust analysis from control devices dedicated to a single
PCS component. The metal HAP content was assumed to be representative of both controlled
and uncontrolled HAP emissions (as a percentage of PM). The PM emission factors are
provided in Table 5-6. The metal HAP content is shown in Table 5-7. Note: These data are for
5-10
-------
automated PCS lines. We have no data specific for floor or pit molding emissions. To provide
baseline emissions estimates, the emissions factors and HAP content values presented in
Tables 5-5 and 5-6 (for automated lines) are also used to estimate emissions from floor and pit
molding operations.
TABLE 5-6. PM EMISSION FACTORS FOR PCS LINES
PCS Component
Pouring
Cooling
Shakeout
Uncontrolled component PM
emission factor,
Ib/ton metal melted
0.0873
0.29
79.3
Controlled component
PM emission factor,
Ib/ton metal melted
0.026
0.038
0.30
TABLE 5-7. HAP CONTENT OF PM FROM PCS COMPONENTS
PCS Component
' Pouring
Cooling
Shakeout
Manganese Content,
percentage of PM
3.20%
0.11 %
0.021 %
Total Metal HAP Content, 1
percentage of PM
5.08 %
0.22 %
0.024 %
Because the PM concentration of pouring and cooling emissions vent stream are very low
(due to low PM emissions and large volume of air used to capture the emissions and to effect
cooling of the molds), PM reduction efficiencies for pouring and cooling emission control
devices are limited. Conversely, due to the high PM loading and generally coarse nature of the
PM from shakeout operations (primarily sand), PM controls for shakeout generally effect greater
than 99-percent removal of the PM in the vent stream.
5.3 BASELINE EMISSIONS
Baseline emission estimates for iron and steel foundries were estimated using the facility-
and process-specific information developed from the detailed industry survey responses (EPA,
1998a) and best estimates of HAP emission factors (see Section 5.2). The survey provided a
snapshot of the amount of metal melted in 1997. To account for fluctuations in production, the
5-11
-------
production capacity of each foundry was estimated from the annual production rates reported in
the survey responses combined with reported capacity utilization rates. Capacity utilization rates
reported for the iron foundry industry for the 4 years between 1997 and 2000 ranged from 80 to
88 percent (Kirgen, 1997,1998, and 1999). The 80-percent capacity utilization value was
selected for the purposes of estimating each foundry's capacity. Therefore, the production rates
reported in the survey were scaled by a factor of 1.25, and preliminary emission estimates were
made using the emission factors presented in Section 5.2 to identify foundries with the potential
to emit more than 10 tons/yr of a single HAP or more than 25 tons/yr total of all HAPs (i.e.,
"major sources" of HAP emissions).
Using this information, HAP emission estimates were developed for all 595 foundries
that responded to the survey. This preliminary analysis indicates that essentially all (over 99.9
percent) of foundry HAP emissions originated from three "primary" operations:
• Mold and core making and coating
• Melting, and
• Pouring, cooling, and shakeout.
In addition to making preliminary emission estimates, we contacted State and Regional
environmental regulators to identify additional foundries that reported to be major sources of
HAP emissions. This effort was conducted to identify foundries that either have higher
emissions than projected using the average emission factors, or that have a potential to emit that
is much greater than expected based on the foundry production capacity estimates (e.g., a
foundry that operates 2,000 hr/yr could have a capacity 4 times greater than the capacity
projected for the foundry based on potential operating hours in a year).
Based on the emission modeling and additional data gathering efforts, 96 facilities that
responded to the detailed industry survey were projected to be major sources of HAP emissions.
Two of these facilities operate two adjacent foundries that each responded to the industry survey,
so that 98 of the 595 respondents to the EPA survey are actually included in the pool of major
sources. However, based on the definition of a facility in the CAA, these adjacent foundries are
considered one facility and the emissions from these adjacent foundries are summed to determine
if the facility's emissions exceed the major source emission threshold levels. The major source
foundries are generally large production sand casters with production capacities of 100,000
tons/yr or higher. Several foundries appear to be major sources based on use of chemical binders
5-12
-------
in their mold- and core-making processes; these foundries typically have much lower production
rates.
Baseline emission estimates provided in this section include only the emission estimates
for those iron and steel foundries projected to be major sources of HAP emissions. As described
previously, the emission factors employed are average values, and some foundries may have
higher or lower emissions than are estimated using these emission factors. Consequently, some
foundries projected to be just above the major source emission threshold may not, in fact, be a
major source of HAP emissions. Conversely, some foundries that are projected to be just below
the major source emission threshold may actually be a major source of HAP emissions. Thus,
the precise list of major source foundries may differ slightly from the foundries used in assessing
the baseline emissions for the major source foundries. Therefore, rather than presenting facility-
specific emission estimates, we present the baseline emissions for each of the three primary
sources of HAP emissions separately using general classes of operations.
5.3.1 Baseline Emissions for Mold and Core Making and Coating
Table 5-8 presents the baseline emission estimates for mold- and core-making operations
based on the binder system type and total chemical usage rates for the mold lines. The number
of mold lines per type of binder system within the anticipated major source foundries is also
provided.
Nearly all mold and core coatings used at iron foundries are either water based or
isopropanol based. These coating solvents do not contain HAP and do not produce HAP
emissions. Using reported coating solvent hourly use rates, and assuming (1) 100 percent
solvent emissions during the coating drying process and (2) the coating process operates 8,000
hours per year, the following HAP emissions are estimated for mold- and core-coating
operations from the major source foundries:
• Methanol-Based Coatings (used at eight mold or core lines)
- 223 tons/yr of methanol used
- 223 tons/yr of methanol emitted
5-13
-------
TABLE 5-8. EMISSIONS FROM MOLD- AND CORE-MAKING LINES AT MAJOR SOURCE IRON FOUNDRIES'
Chemical sand binder
system2
Phenolic urethane cold box:
TEA is controlled3
Phenolic urethane cold box:
i TEA is not controlled
Shell (phenolic Novolac
flake): Foundry coats sand
Shell (phenolic Novolac
liquid): Foundry coats sand
Shell (phenolic Novolac):
Foundry uses precoated
sand
Furan hot box
Phenolic hot box
Phenolic urethane no-bake
No. of
core lines
214
57
1115
IIP
221 5
70'
69 7
109
Chemicals
used,
tons/yr
10,480'
2,130''
3,750s
3,750s
7,500 5
4,3507
4,3507
5,530
M^» nf
iNO. 01
mold
lines
5
0
326
32 6
?6
0
0
18
Chemicals
used,
tons/yr
4,390'
4,200 6
4,200"
930 6
14,620
Estimated emissions, tons/yr
Phenol
6.9
2.9
1.4
135
151
49
55
10.8
3.65
9.65
Formal-
dehyde
0.12
0.05
0.02
0.94
1.05
6.53
4.35
0.06
0.16
Naph-
thalene
9.4
3.9
1.9
8.85
23.4
Cumene
3.0
1.3
0.62
2.82
7.46
Xylene
0.94
0.39
0.19
0.89
2.34
Methanol
0.
0.
188
210
TEA
10.5
4.4
213
(continued)
5-14
-------
TABLE 5-8. (CONTINUED)
Chemical sand binder
system2
Furan no-bake
Furan warm box
Phenolic baking (warm box)
Phenolic no-bake
Acrylic/epoxy/SO-,
Phenolic ester no-bake
Phenolic ester cold box
Other systems, mostly
CO2-catalyzed systems
Core oil
Alkyd urethane
No. of
core lines
5
24s
24 8
7
2
5
4
20
23
•}
3
Chemicals
used,
tons/yr
180
1,360s
1,3608
330
400
360
480
2,810
1,630
1,480
No. of
mold
lines
5
1
0
12
0
3
0
4
0
0
Chemicals
used,
tons/yr
13,590
130
8,120
3,310
940
Estimated emissions, tons/yr
Phenol
10.9
0.53
13.3
0.29
2.65
0.38
Formal-
dehyde
0.003
0.19
0.27
0.03
4.35
0.02
0.55
0.04
0.33
0.05
Naph-
thalene
Cumene
19.9
Xylene
Methanol
18.6
1,386
136
13
35
869
TEA
(continued)
5-15
-------
TABLE 5-8. (CONTINUED)
I Chemical sand binder
system2
Free radical/SO,
No. of
core lines
9
Chemicals
used,
tons/yr
480
No. of
mold
lines
0
Chemicals
used,
tons/yr
TOTALS (556.1 tons/yr for all HAPS)
Estimated emissions, tons/yr
Phenol
448.
Formal-
dehyde
15.6
Naph-
thalene
47.5
Cumene
35.1
Xylene
4.8
Methanol
2,170.
TEA
228
No emission estimates were made for the shell system using precoated sand or for the core oil, alkyd urethane, free radical/SO;, or CCX-catalyzed systems
because no data for these emissions exist. Amounts of chemicals were not determined for the latter four systems because their use is very limited.
Systems are listed in order of chemical usage, where known, and by the number of lines using the system otherwise.
TEA control is assumed to be 99 percent.
Amounts given are for the liquid (resin) components only. Weight of catalyst gas used is approximately 10 percent of the weight of the resin.
Assumes shell core systems are 25% flake, 25% liquid, and 50% precoated.
Assumes shell mold systems are 45% flake, 45% liquid, and 10% precoated.
Survey requested data for hot-box systems in general: assumes 50% furan hot box and 50% phenolic hot box.
Survey requested data for warm-box systems in general: assumes 50% furan warm box and 50% phenolic baking.
5-16
-------
Naphtha- or Aliphatic-Petroleum-Distillates-Based Coatings (used at five mold or core
lines)
— 622 tons/yr of naphtha used
- 62 tons/yr of HAPs emitted (assumes naphtha contains 10 percent HAPs)
- HAPs emitted may include benzene, toluene, xylene, and naphthalene
Water-Based with 1 percent #2 Fuel Oil (used at one core line)
- 2,716 tons/yr phenol resin used
- 2.7 tons/yr HAPs emitted (assumes #2 fuel oil contains 10 percent HAPs)
Phenol Resin (used at one mold line)
- 1.8 tons/yr phenol resin used
— 1.8 tons/yr phenol emitted (this is actually a dry-coating material, phenol content
of the resin is not known; estimate is a worst-case assumption)
5.3.2 Baseline Emissions for Melting
Table 5-9 presents the baseline metal HAP emissions for the melting processes by
furnace type. Roughly 70 percent of the baseline metal HAP emissions from melting processes
are metal HAPs emitted from cupolas that are controlled with a wet scrubber. In addition to the
metal HAP emissions presented in Table 5-9, two foundries have cupolas that do not use
afterburning. Organic HAP emissions from these cupolas are estimated to be 4.4 tons/year, but
these estimates have large uncertainty due to the lack of emissions data available for cupolas that
do not operate with afterburning.
5.3.3 Baseline Emissions for PCS
The baseline emissions for PCS lines are presented in Table 5-10. Based on the capacity
of the PCS lines, the fraction of each PCS component that was controlled was determined for
anticipated major source iron foundries. These fractions were applied to the annual production
rate for the major source foundries to establish production rates attributed to controlled and
uncontrolled processes. The majority of pouring and cooling lines are not controlled for PM,
whether they are automated pouring and cooling lines or are stationary floor or pit molds. All
automated shakeout lines for the anticipated major source foundries are controlled. Most floor
mold shakeout operations are uncontrolled. One of the areas of uncertainty in the PCS emissions
is the fact that the emission factors were developed only using data from automated PCS lines.
5-17
-------
TABLE 5-9. ASSIGNED ANNUAL PRODUCTION AND METAL HAP EMISSIONS
FOR MODEL MELTING FURNACE
Type of melting furnace &
control device1
Cupolas
Cupola w/ WS
Cupola \v7 BH
Total for cupolas
Electric Arc
Total for EAFs (w/ BH)3
Electric Induction
EIFw/BHorCF
EIF \v/ WS
EIF \v/ no APCD
Total for EIFs
TOTAL
Number
of
furnaces
Production, tons/yr
Iron
Steel
Total
production
PM
emission
factor,
Ib/ton
Annual emissions, tons/yr
Manganese2
Total metal
HAPs2
35
28
63
4,337,000
3,355,000
7,692,000
4,337,000
3,355,000
7,692,000
0.60
0.06
42.9
3.3
46.3
76.8
5.9
82.7
50
534,000
751,000
1,285,000
0.23
3.0
4.9
122
12
88
222
335
2,395,000
357,000
910,000
3,662,000
11,888,000
8,000
2.000
43,000
53,000
804,000
2,403,000
359,000
953,000
3,715,000
12,692,000
0.30
0.30
2.00
4.7
0.7
12.4
17.8
67.0
7.2
1.1
19.1
27.3
114.9
' WS = Wet scrubber; BH = Baghouse; CF = Cartridge filter; APCD = air pollution control device.
2 Assumptions: Cupola PM: Mn = 3.3%, total = 5.9%; EAF PM: Mn = 2.0%, total = 3.3%; EIF PM: Mn = 1.3%, total = 2.0%.
3 All major source foundries that use EAF melting furnaces control PM emissions using baghouses.
5-18
-------
TABLE 5-10. EMISSION ESTIMATES FOR MODEL PCS LINES
Model line
Emissions from pouring controlled for
PM
Emissions from pouring uncontrolled
for PM
Emissions from cooling controlled for
PM
Emissions from cooling uncontrolled
forPM
Emissions from shakeout controlled
forPM
Emissions from shakeout uncontrolled
for PM
Totals for all PCS lines
Metal
poured,
tons/yr
5,660,000
7,072,000
7,889,000
4,843,000
12,598,000
134,000
12,732,000
Emissions, tons/yr
Benzene
6.2
7.7
137.7
84.5
168.8
1.8
406.7
Total
organic
HAPs
14.2
17.7
307.7
188.9
1,259.8
13.4
1,801.6
PM
74
309
150
702
1,890
5,313
8,437
Mn
2.35
9.88
0.16
0.77
0.40
1.12
14.68
Total metal
HAPs
3.74
15.68
0.33
1.54
0.45
1.28
23.02
However, because 95 percent of production at the anticipated major source iron foundries is
processed in automated lines, the uncertainty in PCS emissions for floor and pit molding
operations lines is not expected to significantly affect the estimated baseline emissions.
As discussed previously, automated PCS lines are generally auto-ignite, and the organic HAP
emission factors are considered to be applicable to these "controlled" units. For floor and pit
molds that do not auto- ignite, it is typical industry practice to ignite these vents. Consequently,
it is assumed, for the purposes of estimating baseline emissions, that all mold vents either auto-
ignite or are manually ignited. Additionally, one pouring and cooling line at one foundry uses a
thermal oxidizer to control odor (and presumably effects some organic HAP emissions). This
line represents 0.1 percent of the major source foundry production rate. With no "additionally
controlled" emission factors available, and this line representing such a small fraction of the
overall production, little error is introduced in the baseline emissions estimate in applying only
the "controlled" organic HAP emission factors to this line.
5-19
-------
5.4 CONTROL OPTIONS
Various control options were considered for each of the primary sources of HAP
emissions. Table 5-11 summarizes the environmental impacts of the control options identified.
Additional discussion of the control options is provided in the following sections.
5.4.1 Control Options for Mold and Core Making and Coating
The primary add-on control device used in conjunction with mold and core making is an
acid wet scrubber used to control TEA emissions from phenolic urethane cold-box systems. No
other effective add-on control options were identified to reduce the emissions from mold and
core making. However, there appear to be substitute binder formulations that can be used to
achieve HAP emission reductions for certain binder formulations. The control options evaluated
for mold and core making are:
• Acid wet scrubber to reduce TEA emissions;
• Binder reformulations to reduce methanol content and emissions;
• Solvent substitution to reduce naphthalene content and emissions; and
• Coating formulations that do not contain HAP.
5.4.1.1 TEA Scrubber. Acid wet scrubbers are used to control TEA emissions from all
but four phenolic urethane cold-box mold- and core-making lines at the anticipated major
sources. The four uncontrolled lines have emissions of 146 tons/yr; emission reductions of over
99 percent are anticipated. The addition of a TEA scrubber requires pumps and fans that
consume energy and produce a spent-acid waste water stream that requires treatment and
disposal.
5.4.1.2 Methanol Replacement in Binder Systems. Methanol is used as a carrier
solvent for certain binder formulations. Alternative binder formulations are currently available
that use a non-HAP carrier solvent for furan warm-box systems. These alternative furan warm-
box binder formulations are already replacing the methanol-containing systems at many
foundries and are available at no additional cost to the foundry. Other binder systems, such as
the furan no-bake, phenolic no-bake, and the Shell (Novolak flake) systems, can be formulated
without methanol. Methanol replacement in binder formulations is estimated to effect a HAP
emission reduction of between 150 and 2,860 tons/yr, depending on the current use of non-
methanol-containing systems and the compatibility of the non-methanol-containing systems for a
specific foundry's operations.
5-20
-------
TABLE 5-11. SUMMARY OF ENVIRONMENTAL IMPACTS
I Control Option
Acid (TEA) scrubber
Non-methanol binder formulations
Use of naphthalene-depleted
solvents
Non-HAP coatings
Replace cupola wet scrubber and
fabric filter
Afterburner for cupola
Fabric filter for EIF
Mold vent light-off
Fabric filter for automated pouring
lines
Fabric filter for automated cooling
lines
Fabric filter for automated
shakeout lines
Fabric filter for floor and pit
shakeout
HAP
emission
reduction,
ton/yr
146
340'
32
303
64
4
31
?
22
1
0
2
voc
emission
reduction,
ton/yr
146
340'
32
Up to 303
0
43
0
9
0
0
0
0
PM
emission
reduction,
ton/yr
0
0
0
0
1,085
0
1,293
0
386
822
0
6,536
Energy
consumption
rate,
MW-hy/yr
510
0
0
0
(129,463)
300
17,658
0
7,387
60,447
0
Natural gas
consumption
rate,
MMcf/yr
0
0
0
0
0
27
0
0
0
0
0
0
Water
consumption
rate,
MMgal/yr
4
0
0
0
(14,633)
0
0
0
0
0
0
0
Solid waste
disposal,
tons/yr j
0
0
0
0
0 i
0
1,293
0
386 i
822
0
6,536
Best estimate; actual emission reduction dependent on the current use of non-methanol containing systems and compatibility issues.
5-21
-------
5.4.1.3 Reduction of Naphthalene Content in Binder System Formulations.
Naphthalene-containing solvents are used primarily in phenolic urethane binder systems, either
the phenolic urethane no-bake system or the phenolic urethane cold-box system. Except for the
shell system, these are the most commonly used binder systems, especially by the higher
production rate foundries. Phenolic urethane binder systems typically use a solvent that contains
approximately 10 percent naphthalene. The naphthalene content of these systems can be reduced
by using a naphthalene-depleted solvent that contains 3 percent or less naphthalene, effectively
reducing the air emissions of naphthalene from these mold- and core-making lines by
approximately 70 percent. Based on the baseline emissions of naphthalene for both the phenolic
urethane no-bake and the phenolic urethane cold-box systems, this control option is estimated to
achieve an emission reduction of 32 tons/yr of naphthalene.
5.4.1.4 Mold- and Core-Coating Replacements. Methanol, naptha, and other HAP-
containing solvents are used as carrier solvents for some mold- and core-coating operations.
Alternative coating systems are currently available that use non-HAP carrier solvents. These
non-HAP alternative systems already dominate the foundry market, representing over 90 percent
of mold- and core-coating lines. Therefore, HAP emissions can be eliminated from foundries by
replacing HAPs containing coating solvents with non-HAP solvents (e.g., water or isopropanol).
The total organic HAP emission reductions are estimated to be 303 tons/yr. Depending on the
replacement coating system used, these HAP emission reductions may or may not indicate a
reduction in VOC emissions.
5.4.2 Control Options for Melting
Control options considered for melting furnaces include:
• Replacement of existing venturi scrubbers on cupolas with fabric filters; and
• Addition of fabric filters to control captured, but uncontrolled emissions from
induction furnaces.
5.4.2.1 Replacement of Cupola Wet Scrubbers with Fabric Filters. The replacement
of cupola wet scrubbers is estimated to yield a 64-ton/year reduction in metal HAP emissions
and a 1,085 ton/year reduction in particulate matter (PM). Additionally, the baghouses operate
at lower pressure drops, and realize a savings in both energy usage and water consumption.
Although more mass of particulates is expected to be collected by the baghouse, the dust is
collected dry. The particulates collected by the wet scrubber retain some water, therefore the
5-22
-------
mass of solids requiring disposal is greater for the wet scrubber than for the baghouse, given a
common mass of dry particulate collected. It is assumed that the higher mass of particles
collected by the baghouse is approximately equal to the wet mass of particles currently requiring
disposal from the wet scrubbers.
5.4.2.2 Afterburners for Cupolas without Afterburning. Two cupolas at the major
source foundries do not use afterburning. Afterburning is estimated to effect a 98-percent
emission reduction of organic HAPs, based on general design characteristics of afterburners.
Table 5-11 summarizes the emission reductions and secondary impacts anticipated from
installing afterburners on these cupolas.
5.4.2.3 Fabric Filters for Uncontrolled EIF. PM control options are available for
EIFs. Table 5-11 summarizes the emission reductions anticipated if all currently uncontrolled
EIFs added capture and fabric filter controls.
5.4.3 Control Options for PCS
Control options considered for PCS lines include:
Requiring light-off of mold vents that do not spontaneously ignite; and
• Adding fabric filters to control captured PCS emissions from automated PCS
lines.
5.4.3.1 Mold Vent Light-off. Automated PCS lines are the most prevalent, and they
account for over 95 percent of the production capacity of the anticipated major source iron
foundries. These mold vents generally ignite spontaneously. Floor and pit molding is generally
used at foundries that produce very large cast parts. The vent gases from these large molds do
not always spontaneously ignite. It is typical industry practice to ignite these vents. All current
organic HAP emission estimates from pouring and cooling lines were made at automated lines
where the mold vents spontaneously ignite. This "vent flame" is anticipated to destroy organic
HAPs in the vent gases, but the destruction efficiency is unknown. Because all organic
emissions estimates are based on data from pouring and cooling lines that have this vent flame,
and because the light-off process is generally used for floor and pit molding, no additional
emission reduction is attributed to this control option.
One foundry has a thermal oxidizer on one pouring and cooling line, and two foundries
have carbon adsorption systems on their pouring and cooling line to eliminate odors and VOC
(and presumably organic HAP). These pouring and cooling lines are expected to have higher
5-23
-------
organic emissions than typical sand casting pouring and cooling lines based on the amount of
chemically bonded sand used at these foundries. For typical mold pouring and cooling lines, the
organic HAP content in the ventilation gas stream is very low, so that incineration, carbon
adsorption, and other known control technologies are inefficient and lead to significant adverse
secondary impacts.
5.4.3.2 Fabric Filters for Uncontrolled Automated PCS Lines. PM control options
are available for PCS lines. All automated shakeout at the anticipated major source iron
foundries is controlled. Floor and pit mold PCS emissions are generally uncontrolled due to the
difficulties associated with capturing emissions from these large area sources. Table 5-11
summarizes the maximum emissions reductions anticipated if all currently uncontrolled PCS
lines added capture and fabric filter controls.
5.5 REFERENCES
AFS and CISA, 1998. Form R Reporting of Binder Chemicals Used in Foundries. 2nd Edition.
Prepared by American Foundrymen's Society (AFS) and the Casting Industry Suppliers
Association (CISA).
Baldwin, V. H., and C. W. Westbrook, 1983. Environmental Assessment of Melting, Pouring,
and Inoculation in Iron Foundries, prepared by Research Triangle Institute for the U.S.
Environmental Protection Agency, Industrial Environmental Research Laboratory under
EPA contract nos. 68-02-3152 and 68-02-3170, revised May 1983.
CERP, 1998. Foundry Process Emission Factors: Baseline Emissions from Automotive
Foundries in Mexico, Casting Emission Reduction Program, McClellan Air Force Base,
California, November 24,1998.
Euvrard and Western Associates, 1992. "Case Study: Air Audit at a Medium Size Gray Iron
Foundry." 5th Annual AFS Environmental Affairs Conference, August 1992.
Kearney, A. T., 1971. Systems Analysis of Emissions and Emissions Control in the Iron Foundry
Industry, vol. I-III, prepared by A. T. Kearney and Company for EPA Air Pollution
Control Office under Contract No. CPA 22-69-106, NTIS PB 198 348-50, February.
Kirgin, K. H., 1997. "1997 Metalcasting Forecast and Trend - Solid Casting Markets Fuel 1997
Expansion," Modern Casting, American Foundrymen's Society, Des Plaines, IL, vol. 88
(January), No. 1. Available at
http://www.moderncasting.com/archive/featurejan_01.html
Kirgin, K. H., 1998. "Solid Economy Continues to Fuel Casting Growth," Modern Casting,
American Foundrymen's Society, Des Plaines, IL, vol. 89 (January), No. 1, pp. 30-33.
5-24
-------
Kirgin, K. H., January 1999. "1999 Contraction to Cause Demand to Dip to 14.5 Million Tons,"
Modern Casting, American Foundrymen's Society, Des Plaines, IL, vol. 90 (January),
No. 1, pp. 34-38.
Lafay, V. S., and S. L. Neltner. September 1998. "Insight gained into green sand's benzene
emissions." Modern Casting. Pp. 58-60.
RMT, 1995. Wisconsin Cast Metals Association Group Source Emission Testing, prepared by
RMT, Inc., for Wisconsin Cast Metals Association, revised June 1995.
Scott, Bates, and James, 1976. "Foundry Air Contaminants from Green Sand Molds." Journal
of the American Industrial Hygiene Association, p. 335-344.
Shaw, F., 1982. "Induction Furnace Emissions," AFS International Cast Metals Journal, June,
pp. 10-27.
Twarog, 1991. Identification of Emissions and Solid Wastes Generated from EPC Process.
American Foundrymen's Society Research Report (June 4, 1991).
U.S. Environmental Protection Agency. 1980. Electric Arc Furnaces in Ferrous Foundries
Background Information for Proposed Standards, Office of Air Quality Planning and
Standards, Research Triangle Park, NC. EPA-450/3-80-020a, May, pp. i-x.
U.S. Environmental Protection Agency. 1983. Electric Arc Furnaces and Argon-Oxygen
Decarburization Vessels in Steel Industry - Background Information for Proposed
Revisions to Standards, Office of Air Quality Planning and Standards, Research Triangle
Park, NC. EPA-450/3-82-020a, July, pp. i-xvii.
U.S. Environmental Protection Agency. 1990. Air Emissions Species Manual, Office of Air
Quality Planning and Standards, Research Triangle Park, NC. 2nd ed., vol. I: Volatile
Organic Compound Species Profiles and vol. II: Particulate Matter Species Profiles,
EPA-450/2-90-001, January, pp. vii-ix, xi-xiii, iv-v, vii-ix, xi-xii.
U.S. Environmental Protection Agency. 1995. Compilation of Air Pollutant Emission Factors,
Office of Air Quality Planning and Standards, AP-42, vol. 1, 5th ed., sections 12.10 and
12.13. Research Triangle Park, NC. January.
U.S. Environmental Protection Agency. 1998a. Compilation of Information from Information
Collection Request Forms Submitted by Iron and Steel Foundries to the U.S. EPA Office
of Air Quality Planning and Standards, Research Triangle Park, NC.
U.S. Environmental Protection Agency. 1998b. Petroleum Refineries - Background
Information for Proposed Standards, Catalytic Cracking (Fluid and Other) Units,
Catalytic Reforming Units, and Sulfur Recovery Units. Washington, DC: Government
Printing Office. EPA-453/R-98-003.
5-25
-------
U.S. Environmental Protection Agency. 1998c. Iron and Steel Foundries Direct Interface
GCMS Testing of No. 4 Mold Line, General Motors, Saginaw Metals Castings
Operations, Saginaw, Michigan. Prepared by Emission Monitoring, Inc., under EPA
requisition DAC 164QT-RT-98-0000468, for Office of Air Quality Planning and
Standards, Research Triangle Park, NC. May.
U.S. Environmental Protection Agency, 1999a. Iron and Steel Foundries Manual Emissions
Testing of Cupola Baghouse at Waupaca Foundry in Tell City, Indiana. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. EPA-454/R-99-017A and
EPA-454/R-99-017B. June.
U.S. Environmental Protection Agency, 1999b. Iron and Steel Foundries Manual Emissions
Testing of Cupola Wet Scrubber at General Motors Corp., Saginaw, Michigan. Office of
Air Quality Planning and Standards, Research Triangle Park, NC. EPA-454/R-99-025A
and EPA-454/R-99-025B. July.
5-26
-------
6.0 CONTROL COSTS
This chapter describes the methodology used to develop nationwide control costs
associated with the installation and operation of HAP emission control equipment for iron and
steel foundries.
6.1 GENERAL APPROACH FOR ESTIMATING CONTROL COSTS
EPA conducted a detailed survey of the iron and steel industries in 1998 to gather
information regarding the types of processes and control devices used by each foundry. This
survey information was used to identify the specific processes within each foundry that would
need to be upgraded or to have new control equipment added. The control costs were estimated
using the cost algorithms described in the OAQPS Control Cost Manual (EPA, 1996) and the
Handbook: Control Technologies for Hazardous Air Pollutants (EPA, 1991). The control costs
were estimated in fourth-quarter 1998 dollars.1 Costs of the control systems were driven
primarily by the flow rate of the exhaust gas requiring treatment. Typical vent stream
characteristics (e.g., flow rate,s per unit capacity or throughput, temperature) were developed
from data reported in response to the detailed questionnaires. Costs also were included for
monitoring devices, such as CO monitors for cupola afterburners, VOC monitors for scrap
preheaters, and bag leak detection systems for fabric filters (baghouses). Finally, costs were
included for recordkeeping and reporting requirements. More details regarding the control costs
estimated for specific processes are provided in the following sections.
6.2 CUPOLA MELTING FURNACE CONTROL SYSTEMS
One control option available for cupolas is to require all cupolas to operate using a
baghouse. To estimate the costs of replacing venturi scrubbers with baghouses, essentially two
cost estimates were made. First, the capital investment costs and the annual operating and
1 Cost estimates were calculated in 1998 dollars because the detailed industry survey provided a snapshot of the
industry in 1998.
6-1
-------
maintenance costs (AOCs) of a new baghouse system were estimated. Second, the AOCs of a
venturi wet scrubber system were estimated (because these costs were already being incurred by
the foundries and offset the operating cost of the baghouse). Additionally, the proposed MACT
establishes CO emission limits (as a surrogate for organic HAPs) for cupolas. Therefore, costs
were also developed for installing an afterburner to cupola furnace exhaust streams that currently
do not use afterburning.
6.2.1 Baghouse Control Costs for Cupola Melting Furnaces
Baghouse or fabric filter control costs were estimated using the CostAir program (EPA,
1996). The fabric filters were designed as pulse-jet modular systems with an A/C ratio of
3.0 ft/min using Nomex bags. Auxiliary equipment included a new fan, a motor, two dampers,
300 feet of ductwork, and a new stack. The cost of the additional damper and ductwork (300 ft
versus a typical value of 100 ft) and an additional system pressure drop of 4 inches of water were
included in the cost analyses to roughly simulate the added capital and operating costs associated
with cooling the cupola exhaust stream prior to the baghouse by using the hot exhaust gases to
preheat the cupola blast air. These costs would be incurred because a baghouse control system
cannot operate at as high an inlet gas temperature as a venturi scrubber control system. A retrofit
cost factor of 2 was applied to the total capital investment cost estimate to capture the costs of
removing existing control equipment and of dealing with other difficulties anticipated with a
system retrofit of this nature. All cost values were calculated in fourth-quarter 1998 dollars
(Vatavuk Air Pollution Control Cost Index = 110.9).
Control costs for six different sizes of baghouses were calculated based on the anticipated
range of vent stream flow rates. The baghouse flow rates considered ranged from 20,000 to
280,000 actual cubic feet per minute (acfm), which covers an approximate range of cupola
melting furnace capacities of 10 to 140 tons/hr. The total capital investment and the AOCs for
these model baghouse systems are summarized in Table 6-1. The calculated control costs for
these systems were essentially linear over the flow rates investigated; a linear regression analysis
of the capital and the operating and maintenance control costs had R2 values of 0.999 and 0.993,
respectively. Consequently, a simple linear expression was derived to estimate the total capital
investment (TCI) and the AOC based on the system exhaust flow rate as follows:
6-2
-------
where
AOCRH =
BH
TCIBH = 463,700 + 25.03 QBH
AOCBH = 93,970 + 3.312QBH
total capital investment for a baghouse ($ fourth-quarter 1998);
AOC for a baghouse (1998 $/yr); and
design exhaust vent flow rate based on cupola-baghouse (acfm).
(6.1)
(6.2)
TABLE 6-1. SUMMARY OF CONTROL COSTS FOR BAGHOUSES
AND WET SCRUBBERS: 1998 $
Flow rate through
control device —
baghouse (wet
scrubber), acfm
20,000(15,400)
40,000 (30,800)
80,000(61,500)
120,000(92,300)
200,000(154,000)
280,000(215,000)
Baghouse
total capital,
S103
$870
$1,454
$2,527
$3,544
$5,503
S7.406
Baghouse
annual
capital,
$103*
$80
$133
$231
$323
$499
$671
Baghouse
annual
operating,
$103/yr
$137
$202
$395
$518
$761
$1,001
Baghouse
total
annual,
$103/yr
$217
$335
$626
$841
$1,260
$1,672
Wet scrubber
annual
operating,
$103/yr
$199
$308
$527
$745
$1,181
$1,617
* Reflects capital recovery based on a 20-year life and a 7-percent interest rate.
The TCI and AOC for each cupola baghouse were calculated using these equations and the
maximum anticipated flow rate based on the cupola melt capacity and the cupola exhaust system
design (above or below gas takeoff). Table 6-2 provides the flow rate factors used to estimate
the exhaust stream flow rate based on the cupola melting capacity and exhaust system design.
TABLE 6-2. ESTIMATING EXHAUST AIR FLOW RATES FOR
CONTROL COSTS ESTIMATES
Cupola charge position and type of air pollution control device
Above-charge takeoff /fabric filter
Above-charge takeoff /wet scrubber
Below-charge takeoff /fabric filter
Below-charge takeoff /wet scrubber
Flow rate factor, acfm/tons/hr*
3,000
2,200
1,500
1,200
Adjusted for typical operating temperatures of approximately 500 °F.
6-3
-------
A capital recovery factor (CRF) of 0.0944 was used for baghouses to annualize the
capital investment on the basis of a 20-year equipment life and an annual interest rate of
7 percent. The total annualized cost (TAG) was calculated as the sum of the annualized capital
investment cost and the annual operating and maintenance cost (e.g., TCI x CRF + AOC).
6.2.2 Venturi Scrubber Control Costs for Cupola Melting Furnaces
The costs of operating a venturi scrubber with a pressure drop of 40 inches of water were
estimated using the cost algorithms described in the Handbook: Control Technologies for
Hazardous Air Pollutants (EPA, 1991). The cost of waste water disposal was assumed to be the
same as the water consumption cost, an assumption that likely understates the operating cost of
the venturi scrubber. The control costs were converted from 1989 to 1998 dollars using the
Chemical Engineering Plant Cost Index (from 355 for base year to 389.5). As with baghouses,
the operating costs for venturi scrubbers were calculated for six different exhaust vent flow rates,
and a linear regression analysis was performed. Because the operating and maintenance costs
for venturi scrubbers are driven by (1) the fan electrical usage, (2) the water consumption, and
(3) the waste water treatment costs, the AOC for venturi scrubbers is linear with exhaust stream
flow rate (R2 = 0.99999). The resulting AOC equation for venturi scrubbers is
AOCVS = 90,250 + 7.090 Qvs (6.3)
where
AOCVS = AOC for a venturi scrubber (1998 $/yr), and
Qvs = design exhaust vent flow rate based on cupola venturi scrubber (acfm).
As shown in Table 6-2, the average flow rate per furnace capacity is approximately 30 percent
higher when baghouse systems are employed than when venturi scrubbers are used. This is
thought to be caused primarily by additional air sucked into the exhaust system (the vent is at a
negative pressure with respect to the atmosphere) when the exhaust stream is cooled prior to the
baghouse. Consequently, the AOC for venturi scrubbers are provided in Table 6-1 for flow rates
ranging from 15,400 to 215,000 acfm, because these costs are more comparable to the baghouse
costs reported in Table 6-1 on the basis of cupola melting capacity.
6.2.3 Net Metal HAP Control Cost for Cupola Melting Furnaces
The net control costs for replacing a venturi scrubber with a baghouse control system
were calculated using the following equations:
6-4
-------
(6.4)
AOCVS_BH = AOCBH-AOCVS (6.5)
TACVS.BH = CRF x TCIVS,BH + AOCVS,BH (6.6)
where
TCIVS,BH = TCI for replacing a venturi scrubber with a baghouse, (1998 $);
AOCVS-BH = net AOC for replacing a venturi scrubber with a baghouse (1998 $/yr);
TACVS_BH = total annualized cost for replacing a venturi scrubber with a baghouse
(1998 $/yr); and
CRF = capital recovery factor = 0.0944 (20 years; 7 percent interest).
6.2.4 Sample Calculation of Metal HAP Control Cost for Cupola Melting Furnaces
The AOC of an existing venturi scrubber system was first calculated based on the melting
capacity of the furnace. For example, given an above-charge takeoff cupola with a melting
capacity of 50 tons/hr, the design vent stream flow rate for a cupola-venturi scrubber system was
calculated to be Qvs = 50 tons/hr x 2,200 acfm/tons/hr (flow rate factor from Table 6-2) =
110,000 acfm. The AOC of the existing venturi scrubber system was then calculated using
Eq. (6.3) to yield an AOCVS - $870,000/yr.
The design exhaust flow rate of the new fabric filter system was then calculated as QBH =
50 tons/hr x 3,000 acfm/tons/hr (factor from Table 6-2) = 150,000 acfm. The control costs for
the new fabric filter system were then calculated using this revised design exhaust flow rate as
shown in Eqs. (6.1) and (6.2), yielding TCIBH = $4,220,000 and AOCBH = $591,000/yr.
The net control costs were then calculated using Eqs. (6.4) through (6.6) to yield
TCIVS.BH = $4,220,000; AOCvS.BH = ($279,000/yr); and TACVS,BH = $118,000/yr.
6.2.5 Afterburning Control Cost for Cupola Melting Furnaces
Afterburning control costs were estimated using the CostAir program for incinerator
systems (EPA, 1996). The incinerators were designed to operate at a minimum of 1,300 °F.
From data collected during EPA source tests, it was assumed that the temperature of the cupola
exhaust stream entering the incinerator/afterburner was 500 °F. This inlet gas stream was
assumed to contain adequate oxygen, coming from air entering the cupola exhaust stream
through the charge door opening. The CO concentration after dilution with the charge door
ventilation air was assumed to be 5 percent. The incinerator/ afterburner was assumed to operate
without heat recovery, and a retrofit cost factor of 1.2 was applied to the TCI cost estimate.
6-5
-------
Control costs for 10 different sizes of incinerators were calculated based on the
anticipated range of vent stream flow rates because of a shift in the cost curves identified for gas
flows less than 40,000 acfm. The shift in the cost curve function can be seen in Figure 6-1.
Subsequently, two control cost equations were developed for each cost parameter (TCI and
AOC): one for systems of less than 40,000 acfm and one for systems of 40,000 acfm or more. A
log-log correlation was used for the TCI cost curves. The calculated control cost equations are:
For systems OAP < 40.000 acfm
TClAB = 1,000 x exp[2.997 + 0.2355 ln(QAB)] (6.7)
AC-CAB = 36,360 + 2.113 QAB (6.8)
For systems QAB > 40,000 acfm
TClAB = 1,000 x exp[3.339 + 0.2355 ln(QAB)] (6.8)
AC-CAB = 60,430 + 2.040 Q^ (6.9)
where
TCIAB = TCI for an afterburner, ($ fourth quarter 1998);
AC-CAB = AOC for an afterburner (1998 $/yr); and
QAB = design exhaust vent flow rate at afterburner inlet (acfm).
A linear regression analysis of these control cost equations had R2 values of 0.9999 or greater.
The cupola inlet gas flow rate was estimated using the flow rate factors presented in
Table 6-2. These flow rate factors were developed from systems that had afterburners, but the
flow rate measurements were made downstream of the cupola afterburner. Thus, use of the flow
rate factors in Table 6-2 is expected to yield cost estimates that are biased high. Nonetheless,
because cupola afterburners generally operate with a minimum of auxiliary fuel (CO in the
exhaust stream being the primary fuel), applying the flow rate factors in Table 6-2 should result
in a reasonable estimate of the flow rate at the inlet to the afterburner.
A capital recovery factor of 0.1424 was used for baghouses to annualize the capital
investment on the basis of a 10-year equipment life and an annual interest rate of 7 percent. The
total annualized cost was calculated as the sum of the annualized capital investment cost and the
annual operating and maintenance cost (e.g., TCI x CRF + AOC).
6-6
-------
$600
20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000
Row rate (acfm)
Figure 6-1. Control Cost Curves for Cupola Afterburners.
6.2.6 Sample Calculation of Organic HAP Control Cost for Cupola Melting Furnaces
Continuing the example of an above-charge takeoff cupola with a melting capacity of
50 tons/hr, the design exhaust flow rate was estimated as Q^ = 50 tons/hr x 3,000 acfm/tons/hr
(factor from Table 6-2) = 150,000 acfm. The control costs for the new afterburner system were
then calculated using Eqs. (6.8) and (6.9), yielding TCIAB = $467,000 and AOCBH = $366,000/yr.
Using the capital recovery factor of 0.1424, TACAB = $433,000/yr.
6.3 ELECTRIC INDUCTION, SCRAP PREHEATER, AND POURING STATION
CONTROL SYSTEMS
Control options for EIFs, scrap preheaters, and pouring stations generally entail the
installation of a new emission control system rather than replacing an existing control system as
in the cupola metal HAP control option. All EAFs at the 96 foundries expected to be major
sources of HAP emissions were controlled using baghouses, so no additional control costs are
estimated for EAFs. Some EIFs, scrap preheaters, and pouring stations, however, are not
6-7
-------
controlled for PM (metal HAPs). Costs were estimated for adding new PM control systems
(assumed to be baghouses for costing purposes) for those sources currently not operating a
control system.
6.3.1 Baghouse Control Costs for EIFs and Scrap Preheaters
As with the baghouse costs developed for cupolas, baghouse control costs for the control
of EIFs and scrap preheater PM emissions were estimated using the CostAir program (EPA,
1996). However, the fabric filters in service for these emission sources, based on the
information in the detailed industry survey, generally operate at much lower temperatures and at
significantly higher A/C ratios than cupola baghouses. The EIF/scrap preheater fabric filters
were designed as pulse-jet modular systems with an A/C ratio of 7.6 acfm/ft2 using polyester
bags. Auxiliary equipment included the cost of a new fan, a motor, one damper, 40 feet of
ductwork, and a new stack. A retrofit cost factor of 1.2 was applied to the total TCI to estimate
the retrofit costs for all scrap preheaters and for EIFs that already have a capture system (but no
control device). This retrofit cost factor of 1.2 was used to develop the EIF/scrap preheater cost
curves. The total capital investment cost was subsequently multiplied by 1.1 if no capture
system was reported for the EIF. Essentially, this equates to a retrofit factor of 1.32 for EIF
systems that do not already capture their emissions. As before, all cost values were calculated in
fourth-quarter 1998 dollars (Vatavuk Air Pollution Control Cost Index - 110.9).
Control costs for 10 different sizes of baghouses were calculated based on the anticipated
range of vent stream flow rates. There is a noticeable shift in the operating cost curve for gas
flows between 40,000 and 50,000 acfm (see Figure 6-2). Subsequently, two control cost
equations were developed for each cost parameter (TCI and AOC): one for systems of less than
50,000 acfm and one for systems of 50,000 acfm or more. Overall, the baghouse flow rates
considered ranged from 5,000 to 180,000 acfm. A linear regression analysis of the capital and
the operating and maintenance control costs resulted in R2 values exceeding 0.999 for each size
range for each cost parameter.
6-8
-------
o
o
o
(f)
O
O
1,200
1,000
800
600
400
200
0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000
Flow Rate (acfm)
Figure 6-2. Control Cost Curves for EIF/Scrap Preheater Baghouses.
From the linear regression analysis, the following control cost equations were developed:
For systems QEIF/SPH < 50.000 acfm
TCIEIF/SPH = 63,840 + 6.221 QE1F/SPH (6.10)
AOCEIF/SPH = 63,870+ 1.427 QEJF/SPH (6.11)
For systems QE1F/SpH ^ 50,000 acfm
TCIEIF/SPH = 99,090 + 5.492 QEIF/SPH (6.12)
AOCEIF/SPH = 133,900 + 1.398 QE1F/SPH (6.13)
where
TCIEIF/SPH
AOCEIF/SPH
QEIF/SPH
TCI for an EIF/scrap preheater baghouse (1998 $);
AOC for an EIF/scrap preheater baghouse (1998 $/yr); and
design exhaust vent flow rate based on EIF/scrap preheater (acfm).
6-9
-------
Again, a capital recovery factor of 0.0944 was used for baghouses to annualize the capital
investment on the basis of a 20-year equipment life and an annual interest rate of 7 percent.
The design exhaust flow rate of the EIF control system was estimated based on the
melting capacity of the EIF. The design exhaust flow rate for a scrap preheater control system
was estimated based on the number of scrap preheaters requiring control. If a foundry needed to
add controls for both its EIFs and its scrap preheaters, then a single baghouse would need to be
designed to control the combined system flow rate. Therefore, the EIF/scrap preheater control
system flow rate was calculated as:
QEIF/SPH = QEIF + QSPH (6-14)
where
QEIF = design exhaust vent flow rate based on EIF capacity (acfm); and
QSPH ~ design exhaust vent flow rate based on number of scrap preheaters (acfm).
The EIF control system flow rate was calculated from EIF melting rate capacity.
Additionally, if pouring station emission controls were also required at the foundry, then the EIF
control system would need to be designed to include the flow rate from the pouring station
capture system, as well. Specifically, the EIF exhaust flow rate was estimated as:
QEIF = 5,000 x (MeltCapEIF)°-667 + QPourSt (6.15)
where
MeltCapEIF = melting rate capacity of the EIF (tons/hr); and
Qpourst ~ design exhaust vent flow rate based on number of pouring stations
(acfm) (see Section 6.3.2).
The factor of 5,000 was assigned based on a 5-ft x 5-ft canopy hood with an entrance design
velocity of 200 ft/min for a 1-ton/hr EIF. The capture system exhaust flow rate from other EIF
melting furnaces was assumed to be proportional to the cross-sectional area of the furnace (or to
the 2/3 power of the capacity of the furnace). If pouring stations also required control at the
foundry, the exhaust from the pouring station capture systems was assumed to be added to the
EIF exhaust stream prior to the control device.
The flow rate of a scrap preheater control system was calculated as a simple function of
the number of scrap preheaters requiring PM controls, as follows:
6-10
-------
Number of Scrap Preheaters Requiring Control: Exhaust Flow Rate:
1 QSPH = 20,000 acfm
2, 3, or 4 QSPH = 60,000 acfm
5 or more QSPH = 100,000 acfm
These scrap preheater exhaust flow rates were used directly as the control system flow rates in
Eqs. (6-10) through (6-13) if the EIFs at a given foundry did not need control. That is, pouring-
station exhaust flows were only combined with EIF/scrap preheater control systems when the
EIFs required control. Otherwise, costs for separate control systems were developed when a
foundry required control of a scrap preheater and a pouring station, but not an EIF.
6.3.2 Baghouse Control Costs for Pouring Stations
Baghouse control costs for the control of PM emissions from pouring stations were again
estimated using the CostAir program (EPA, 1996). The design used for pouring-station control
systems was based on baghouses used to control PM emissions from PCS lines. As such, the
cost curves presented in this section can be used to estimate baghouse costs for controlling
pouring, cooling, or shakeout PM emissions. However, these equations were employed only for
pouring-station emission control, and then only when no additional EIF emission control was
required at the foundry. The pouring-station baghouses were designed as pulse-jet modular
systems with an A/C ratio of 10 acfm/ft2. Auxiliary equipment included the cost of a new fan, a
motor, one damper, 40 feet of ductwork, and a new stack. A retrofit cost factor of 1.2 was
applied to the total capital investment cost estimate. Again, all cost values were calculated in
fourth-quarter 1998 dollars (Vatavuk Air Pollution Control Cost Index = 110.9).
Control costs for nine different sizes of baghouses were calculated based on the
anticipated range of vent stream flow rates. As with the EIF/scrap preheater operating cost
curve, there is a noticeable shift in the operating cost curve for gas flows between 40,000 and
50,000 acfm (see Figure 6-3). Subsequently, two control cost equations were developed for each
cost parameter (TCI and AOC): one for systems less than 50,000 acfm and one for systems of
50,000 acfm or more. Overall, the baghouse flow rates for which direct cost estimates were
developed ranged from 5,000 to 180,000 acfm. A linear regression analysis of the capital and
the operating and maintenance control costs resulted in R2 values exceeding 0.999 for both size
ranges and for each cost parameter evaluated. The resulting control cost equations follow:
6-11
-------
1,200
1,000
800
o
o
o
V)
o
O
200
0 20,0$ 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000
Row Rate (acfrn)
Figure 6-3. Control Cost Curves for Pouring Station Baghouses.
For systems 0P?urS. < 50.000 acfm
TCIPourSt = 63,830 +5.481 QPourSt
AOCPourSt = 63,860+1.513 QPourSt
For systems QEiF/SPH ^ 50,000 acfm
TCIPourSt = 99,100 +4.752 QPourSt
AOCPourSt= 133,900+1.484 QPourSt
where
(6.16)
(6.17)
(6.18)
(6.19)
TCIPourSt = TCI for pouring-station baghouse (1998 $);
AOCPourSt - AOC for pouring-station baghouse (1998 $/yr); and
QpourSt = design exhaust vent flow rate based on number of pouring stations
requiring additional control (acfm).
6-12
-------
Again, a capital recovery factor of 0.0944 was used for baghouses to annualize the capital
investment on the basis of a 20-year equipment life and an annual interest rate of 7 percent.
The flow rates for the pouring-station control systems were calculated assuming each
pouring-station capture system was a 4-ft x 4-ft canopy hood with an entrance design velocity of
200 ft/min, so that each pouring station requiring control contributed 3,200 acfm to the pouring-
station control system. However, if only one pouring-station control system required control at a
foundry, the pouring station baghouse was designed for a flow rate of 5,000 acfm (essentially a
5-ft x 5-ft canopy hood with an entrance design velocity of 200 ft/min).
Number of Pouring Stations Requiring Control; Exhaust Flow Rate:
1 Qpours, = 5,000 acfm
2 or more Qpourst = 3,200 acfm x # Pouring Stations
As discussed in Section 6.3.1, the pouring station emissions were assumed to be
combined with the EIF emissions if the foundry was required to add a control system for the
EIFs. Even though the cost of an EIF/SPH baghouse at any given flow rate is higher than the
cost of a similar-sized pouring station baghouse (because of the different A/C ratios assumed for
these systems), it is still more cost-effective to install a single control system at the lower A/C
ratio than to install two separate control systems. It is also likely that foundries that have to
control both scrap preheater and pouring-station emissions will install a single control system for
both of these emission sources to save on costs. However, based on the logic used in the control
cost model, separate control systems were designed for these emission sources if no additional
control was required for the EIFs.
6.4 MOLD- AND CORE-MAKING CONTROL SYSTEMS
Two emission reduction measures for mold- and core-making lines were considered. For
binder systems that employ a TEA gas catalyst, the emission reduction method is the installation
of an acid/wet (absorptive) scrubber. Four of the 96 foundries had uncontrolled TEA emissions
from their mold- and core-making TEA gas binder systems. For other binder systems, there may
be restrictions on the HAP content of the binder system components. Alternative binder systems
are available that can meet the HAP restrictions pertaining to methanol with no further costs
associated with adaptation, for example. However, requirements for the use of naphthalene-
depleted solvents (see Section 6.3.2) are expected to increase the operating costs of 61 of the 96
foundries.
6-13
-------
6.4.1 Acid/Wet Scrubber Control Costs
Costs associated with the installation and operation of two acid/wet scrubbers to control
emissions of TEA were calculated using the cost algorithms reported in the OAQPS Control Cost
Manual (EPA, 1996). Based on the TEA usage rates at the two foundries with uncontrolled TEA
mold- and core-making lines, two scrubbers were sized based on removing 25 tons/yr and
125 tons/yr, approximately a 25-percent excess capacity compared to current usage rates. From
the available source test data, TEA inlet (uncontrolled) concentrations ranged from 10 to
130 ppm. However, the systems with the highest flow rates also had the highest TEA
concentrations. Therefore, the smaller (actually a median size compared to the available test
data) scrubber that could remove 25 tons/yr of TEA was assumed to operate 4,000 hr/yr with an
inlet TEA concentration of 50 ppmv (median value from the test data). The larger scrubber,
capable of removing 125 tons/yr of TEA, was assumed to operate 4,000 hr/yr and at an inlet
TEA concentration of 100 ppmv.
The cost functions presented in the OAQPS Control Cost Manual are provided in third-
quarter 1991 dollars. These costs were scaled to 1998 dollars by using the Chemical
Engineering Plant Cost Index (using 361 for 1991 and 389.5 for 1998). The calculated control
costs for the two acid/wet scrubbers are shown in Table 6-3.
TABLE 6-3. SUMMARY OF CONTROL COSTS FOR ACID/WET
SCRUBBING SYSTEMS: 1998 $
Model scrubber
Scrubber 1 (25 tons/yr)
Scrubber 2 (125 tons/yr)
Total capital,
S103
$157
$309
Annual
capital*
$22.3
$44.1
Annual operating,
$103/yr
$31.6
$50.8
Total annual,
$103/yr
$53.9
$94.9
* Reflects capital recovery based on a 20-year life and a 7-percent interest rate.
6.4.2 Naphthalene-Depleted Solvent Pollution Prevention Costs
Naphthalene-containing solvents are used primarily in phenolic urethane binder systems.
The phenolic urethane binder system types were the ones most commonly used by the major
source foundries. These binder systems use a naphthalene-containing solvent that is
approximately 10 percent naphthalene. The naphthalene content of this solvent can be reduced
to approximately 3 percent by using a naphthalene-depleted solvent, effectively reducing the air
emissions of naphthalene from these mold- and core-making lines by approximately 70 percent.
6-14
-------
The naphthalene-depleted solvent costs 3 to 5 cents more per pound than the typical
naphthalene- containing solvent (in 1998 dollars). No additional or modified equipment is
needed to use the naphthalene-depleted solvent.
Twenty-two foundries were using the phenolic urethane binder system in at least one
mold- and core-making line in 1998. It was estimated that 2,773 tons/yr of naphthalene solvent
were being used by these 22 foundries. At 4 cents per pound, the estimated nationwide increase
in operating costs to use naphthalene-depleted solvent would be $222,000/yr.
6.5 MONITORING, REPORTING, AND RECORDKEEPING
Cost estimates are provided for continuous CO monitors for cupolas and continuous
VOC monitors for scrap preheaters. The costs of the continuous emission monitoring systems
(CEMSs) were estimated using EPA's CEMS Cost Model, Version 3.0 (in 1998 dollars).
Beyond the CEMS, other monitoring requirements considered for the control options are
continuous parameter monitoring requirements. For example, each baghouse is assumed to
install and operate a bag leak detection system. Foundries with TEA scrubber systems are
assumed to install and operate a pH-monitoring system, and gas and liquid flow rate monitors.
For venturi wet scrubbers used to meet a PM emission limit, the monitored parameters include
the pressure drop and gas and liquid flow rates. Finally, some recordkeeping and reporting costs
were estimated for developing a scrap selection and inspection plan, conducting performance
tests, and evaluating low-HAP-emitting binder formulations as required in the proposed rule.
These costs are described in the following sections. All capital costs for monitoring and
recordkeeping equipment were annualized using a capital recovery factor of 0.1424 based on
10-year equipment life and 7-percent interest rate.
6.5.1 Continuous CO Monitoring Systems
CEMS costs for an in situ continuous CO monitor placed after a control device at an
existing plant were estimated in 1998 dollars. The purchase cost of the continuous CO monitor
was estimated to be $55,000. Annual operating costs—which include operation and
maintenance of the monitor, annual relative accuracy test audit (RATA), periodic quality
assurance (QA) reviews, recordkeeping, and annual training and update—were estimated to be
$19,900/yr. Based on a capital recovery factor of 0.1424, the total annualized cost for operating
a continuous CO monitor was $27,700/yr. For the 63 cupolas operated at the 96 foundries, the
total annualized cost for CO monitors should be $1.75 million. However, because of an error in
6-15
-------
the spreadsheet calculation, certain cupolas were double counted so that the total annualized cost
for CO monitors used in the EIA was $2.08 million.
6.5.2 Continuous VOC Monitoring Systems
CEMS costs for an extractive continuous VOC monitor placed after a control device at an
existing plant were estimated (using the total hydrocarbon concentration [THC] monitoring costs
from EPA's CEMS Cost Model) in 1998 dollars. The purchase cost of the continuous VOC
monitor was estimated to be $73,700. Annual operating costs—which include operation and
maintenance of the monitor, annual RATA, periodic QA reviews, recordkeeping, and annual
training and update—were estimated to be $18,500/yr. Based on a capital recovery factor of
0.1424, the total annualized cost for operating a continuous VOC monitor was estimated to be
$29,000/yr. Of the 96 foundries anticipated to be major sources of HAP emissions, 33 were
operating scrap preheaters. If a foundry had multiple scrap preheaters, it was assumed the
exhausts from the scrap preheaters were combined and a single stack per foundry was monitored.
The total annualized cost for VOC monitors, therefore, was estimated to be $957,000.
6.5.3 Bag Leak Detection Systems
Each baghouse will need to be equipped with a bag leak detection system. These systems
will have an installed capital cost of $9,000 each, with an annual operating cost of $500/yr (EPA,
1998a). There are a total of 339 baghouses, either existing or required to be installed, at the 96
major source iron and steel foundries. Consequently, the total capital cost for bag leak detectors
was calculated as $3.05 million, with an annual operating cost of $170,000/yr.
6.5.4 Parameter Monitoring Systems
The costs for parameter monitoring systems were estimated using a generic parameter
monitoring system. Monitoring system costs were evaluated from control equipment supply
company catalogues for pH, pressure, temperature, and flow measurement systems and
associated electronic recording systems. These costs ranged from $1,500 to $3,000 per
monitoring and data recording system (in 1998 dollars). Therefore, the general cost of
equipment for any parameter monitoring system was estimated to be $2,500. It was estimated
that the installation, calibration, troubleshooting, training, and QA procedure development costs
for a new monitoring system would be $5,000, so that the total installed cost per new monitoring
system would be $7,500.
6-16
-------
The annual operating costs were estimated to be $2,000/yr. These costs are largely for
calibration and maintenance of the equipment, but also include summarizing and annual
reporting of the data.
6.5.4.1 Parameter Monitoring Systems for Venturi (PM) Wet Scrubbers. Existing
venturi wet scrubbing systems are expected to be able to meet PM emission limits from EIFs,
scrap preheaters, and pouring stations. If a venturi scrubbing system is employed, both the
pressure drop and scrubbing liquid flow rate must be monitored. Both of these monitoring
systems were assumed to be in place at each venturi scrubber control device. Annual operating
costs for both parameter monitoring systems were assumed to be $4,000/yr ($2,000 x 2). There
were 21 existing venturi wet scrubbing systems associated with EIFs, scrap preheaters, and
pouring stations at the major source foundries, so that the total annualized monitoring costs for
these systems would be $84,000.
6.5.4.2 Parameter Monitoring Systems for Acid/Wet Scrubbing Systems. The
acid/wet scrubbing systems used to control TEA emissions from mold- and core- making
operations are required to monitor flow rate and pH of the scrubbing liquor. All acid/wet
scrubbing systems were assumed to have flow monitors already in place, and it was assumed that
all systems would have to have a pH monitor installed. Consequently, the monitoring costs per
scrubbing system would be $7,500 capital costs and $4,000 annual operating and maintenance
costs. Forty-seven foundries were using TEA gas in their mold- and core-making lines.
6.5.5 Foundry Recordkeeping, Reporting, and Compliance Costs
Several work practice standards were also considered as possible control options. These
work practices generally require foundries to prepare plans and maintain records of certain
emission-reducing activities. It was assumed that all foundries manually ignited mold vents after
pouring if these vents did not auto-ignite. Therefore, no control costs were attributed to these
activities. Costs were estimated, however, for developing a scrap selection and inspection plan,
conducting performance tests, and evaluating low-HAP- emitting binder formulations. These
costs were calculated based on an estimate of the technical person-hours required to complete
each activity and the frequency of occurrence.
Labor rates and associated costs were based on Bureau of Labor Statistics (BLS) data.
Technical, management, and clerical average hourly rates for civilian workers were taken from
the March 2002 Employment Cost Trends (http://stats.bls.gov). Wages for civilian workers
6-17
-------
(white-collar occupations) were used as the basis for the labor rates, with a total compensation of
$28.49/hr for technical, $42.20/hr for managerial, and $18.41/hr for clerical. These rates
represent salaries plus fringe benefits and do not include the cost of overhead. An overhead rate
of 110 percent was used to account for these costs. The fully burdened wage rates used to
represent respondent labor costs were technical at $59.83, management at $88.62, and clerical at
$38.66.
The number of technical hours needed for each compliance activity was first estimated.
For each technical hour needed, 0.05 managerial hours and 0.10 clerical hours were also
assumed to be required. Consequently, the total labor cost, including technical, managerial, and
clerical labor, for a compliance activity would be $68.125 per technical hour expended.2 The
2002 labor costs were multiplied by 0.8837 (139.8/158.2) to estimate the compliance costs in
fourth-quarter 1998 dollars. Therefore, the compliance costs were calculated using $60.20 per
technical hour expended.
6.5.5.1 Performance Tests. A total of 70 technical hours was estimated for each
performance test. This included time to prepare a site-specific QA test plan, conduct the
performance test, and prepare the final source test report. The total number of stacks requiring
performance testing was estimated to be 431. Some stacks will require both a performance test
and a CEMS performance evaluation. It is assumed that both of these activities can be
performed at the same time without a significant increase in the technical hours needed for the
initial compliance demonstration. Additionally, annual performance evaluations of the CEMS
were included in the CEMS monitoring costs.
The performance tests are required once every 5 years. The compliance period is 3 years.
Therefore, the required performance tests can be fairly evenly distributed across the compliance
period. The average annual compliance cost associated with conducting performance tests at a
given foundry was calculated as the total costs for conducting all required performance tests at
that foundry and distributing those costs evenly over 5 years. Thus, 6,034 technical hr/yr were
estimated for conducting performance tests, for a total annual cost of $363,000/yr.
2 As stated, these labor rates are based on March 2002 statistics. According to BLS data, the Employment Cost
Index (ECI) for civilian workers in March 2002 was 158.2, while the ECI for civilian workers in December 1998
was 139.8.
6-18
-------
6.5.5.2 Scrap Selection and Inspection. Most of the 96 foundries had a scrap selection
and inspection program; however, it is anticipated that many foundries will have increased the
number of technical hours spent on scrap selection and inspection to comply with requirements
in the proposed MACT standard. It was assumed that the scrap inspection requirements would
increase a typical foundry's inspection process by 0.5 hr/day or 182 hr/yr (assuming 365
operating days/yr). A one-time scrap selection plan must be prepared and communicated within
the foundry. This activity was assumed to require an additional 10 hr/yr. Therefore, a total of
192 technical hr/yr per foundry were estimated. Of the 96 facilities considered to be major
sources of HAP emissions, two of these foundries each operate two adjacent plants. Although
they are considered a single facility under the CAA, these adjacent plants have separate scrap-
receiving areas. Thus, the scrap selection and inspection costs were estimated based on 98
foundry plants. The total nationwide costs for the scrap selection and inspection program were
estimated to be $1.13 million.
6.5.5.3 Low-HAP-Emitting Binder Evaluation. Under this control option, each
foundry is required to evaluate the potential to substitute currently used binder systems with
alternative binder formulations that have lower HAP emissions. This evaluation must be
performed once every 5 years so that newly developed low-HAP-emitting binder systems can be
periodically evaluated. Eighty technical hours were estimated for each evaluation. As
evaluations are required once every 5 years, the average annual technical hours required to
complete these evaluations would be 16 hr/yr per foundry. Again, the binder evaluations were
calculated based on 98 foundry plants so that the total nationwide costs for evaluating low-HAP-
emitting binder systems were estimated to be $94,400/yr.
6.5.5.4 Miscellaneous Recordkeeping Costs. Costs associated with maintaining
required records at the foundry level were estimated based on two filing cabinets (for capital
equipment costs of $400) and a pack of 10 writeable CDs (for annual operating costs of $32).
Again, these costs were estimated for 98 foundry plants.
6.6 TOTAL NATIONWIDE COSTS
The total nationwide costs for each of the major control or monitoring systems are
provided in Table 6-4. Table 6-4 also summarizes the estimated recordkeeping and reporting
costs. The total annual nationwide cost of the proposed MACT for iron and steel foundries is
projected to be $21.7 million.
6-19
-------
TABLE 6-4. NATIONWIDE COST ESTIMATES FOR IRON FOUNDRY MACT: 1998 $
Source
Baghouse replacement of cupola venturi scrubbers
Cupola afterburners
Baghouses on EIF and scrap preheaters
Baghouses on pouring stations
Acid/wet scrubber systems for TEA control
Use of naphthalene-depleted solvents
Total Emission Control Costs
Continuous CO monitoring system
Continuous VOC monitoring system
Bag leak detection systems
Venturi scrubber monitoring systems
Acid/wet scrubber parameter monitoring systems
Performance tests
Scrap selection and inspection
Binder system evaluation
Other recordkeeping costs
Total Monitoring, Recordkeeping and
Reporting Costs
Total Engineering Control Costs
Total
capital,
$1,000
108,754
555
13,190
7,770
933
0
131,202
4,125
2,432
3,051
0
353
0
0
0
39
10,000
141,202
Annual
capital*
10,266
79
1,245
733
133
0
12,457
587
346
434
0
50
0
0
0
6
176
6,802
Annual
operating,
$l,000/yr
(7,407)
228
5,550
4,747
165
427
3,710
1,493
611
170
84
188
363
1,133
94
3
4,139
7,849
Total j
annual,
$l,000/yr
2,859
307
6,795
5,480
298
427
16,167
2,080
957
604
84
238
363
1,133
94
9
5,563
21,730
Reflects capital recovery based on a 20-year life and 7-percent interest rate for baghouse emission controls and a 10-year life
and 7-percent interest rate for cupola afterburners, TEA scrubbers, and all monitoring equipment.
6.7 REFERENCES
Cole-Farmer, 1996. Products catalogue '97-'98. Cole Farmer Instrument Company, Vernon
Hills, IL.
Mitchell Instrument Co., 1999/2000. Product Catalog No. 399. San Marcos, California.
U.S. Environmental Protection Agency. 1991. Handbook: Control Technologies for Hazardous
Air Pollutants. Office of Research and Development, Washington, DC. EPA-625/6-
91/014, pp. 4-80 through 4-90.
U.S. Environmental Protection Agency. 1996. OAQPS Control Cost Manual. 5th ed. Office of
Air Quality Planning and Standards, Research Triangle Park, NC. EPA-453/B-96-01.
U.S. Environmental Protection Agency. 1998. Compilation of Information from Information
Collection Request Forms Submitted by Iron and Steel Foundries to the U.S. EPA Office
of Air Quality Planning and Standards. Office of Air quality Planning and Standards,
Research Triangle Park, NC.
6-20
-------
APPENDIX A
IRON AND STEEL FOUNDRIES REPORTING
IN THE 1998 EPA SURVEY
-------
APPENDIX A. IRON AND STEEL FOUNDRIES REPORTING
IN THE 1998 EPA SURVEY.
ID
AL-01
AL-02
AL-03
AL-04
AL-05
AL-06
AL-07
AL-08
AL-09
AL-10
AL-11
AL-12
AL-13
AL-14
AL-15
AL-16
AL-17
AL-18
AL-19
AL-20
AL-21
AL-22
AL-23
AL-24
AL-25
AL-26
AL-21
AL-28
AL-29
AL-30
AL-31
AL-32
AL-33
AL-34
AL-35
AL-36
AL-37
AL-38
AL-39
Name
U.S. Pipe & Foundry Co., Inc.
Southern Ductile Casting Co
Alabama Ductile Casting Co.
Talladega Foundry & Machine Co.
Mueller Co.
Glidewell Specialties Foundry Co
Barry Pattern & Foundry Co., Inc.
Vermont American, Auburn Division
Me Wane Cast Iron Pipe Co.
Southern Alloy Corporation
Griffin Wheel Company
Imperial Casting Co.
Robinson Foundry Inc.
Aliceville Cast Products
Southern Tool, Inc.
M&H Valve Company
Southern Ductile Casting Co.
Southern Ductile Casting Co.
Citation Foam Casting Co.
American Alloy Products, Inc.
Lawler Machine & Foundry Co.
Stockham Valves & Fittings, Inc.
U.S. Pipe & Foundry Company
U.S. Pipe & Foundry Co., Inc.
Jacobs Manufacturing, Inc.
ABC Rail Corp.
The Foundry of the Shoals, Inc.
Mobile Pulley & Machine Works, Inc.
Bama Foundry, Inc.
American Safety Tread Co.
Phenix Foundry
Talladega Castings & Machine Co.
Tommie Corporation
American Cast Iron Pipe Co.
Denver Thomas, Inc.
Union Foundry Co.
Opelika Foundry Co.
Brewton Iron Works, Inc.
Bells Novelty Casting Co.
City
Birmingham
Bessemer
Brewton
Talladega
Albertville
Calera
Birmingham
Auburn
Birmingham
Sylacauga
Bessemer
Florence
Alexander City
Aliceville
Oxford
Anniston
Centreville
Selma
Columbiana
Cullman
Birmingham
Birmingham
Bessemer
Anniston
Bridgeport
Calera
Florence
Mobile
Montgomery
Pelham
Phenix City
Talladega
Thorsby
Birmingham
Birmingham
Anniston
Opelika
Brewton
Anniston
State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
Metal p
Iron
189,287
26,258
89,560
4,235
40,780
8,537
1,217
0
109,164
88
0
4,042
17,514
5,404
0
20,000
2,421
7,260
20,849
10
3,750
35,993
183,957
19,837
600
0
13,244
1,930
241
875
1,326
20
4,630
386,965
4,609
32,059
8,225
825
650
oured, tons ]
Steel
0
0
0
0
0
824
0
897
0
1,225
87,075
0
0
0
328
0
0
0
0
389
0
0
0
0
0
134,883
0
6,111
0
0
0
1,320
0
0
9,314
0
0
0
0
per year
Total
189,287
26,258
89,560
4,235
40,780
9,361
1,217
897
109,164
1,313
87,075
4,042
17,514
5,404
328
20,000
2,421
7,260
20,849
399
3,750
35,993
183,957
19,837
600
134,883
13,244
8,041
241
875
1,326
1,340
4,630
386,965
13,923
32,059
8,225
825
650
A-l
-------
Metal poured, tons
ID
AR-01
AR-02
AR-03
AR-04
AZ-01
AZ-02
AZ-03
AZ-04
AZ-05
CA-01
CA-02
CA-03
CA-04
CA-05
CA-06
CA-07
CA-08
CA-09
CA-10
CA-11
CA-12
CA-13
CA-14
CA-15
CA-16
CA-17
CA-18
CA-19
CA-20
CA-21
CA-22
CA-23
CA-24
CA-25
CA-26
CA-27
CA-28
CA-29
CA-30
CA-31
CA-32
CA-33
CA-34
Name
Southern Cast Products, Inc.
Central Foundry and Stove Works
Bentonville Castings Co.
Nibco Inc.- Blytheville Division
American Aerospace Technical Castings
Ruger Investment Casting,
Arizona Castings Inc.
ME-West Castings, Inc.
Dolphin, Incorporated
Techni-Cast Corp.
Acme Castings, Inc.
Westelectric Castings, Inc.
Image Casting
Bell Foundry Company
Precision Cast Products, Inc.
Amcast Precision Products
Alhambra Foundry Company, Ltd.
K.P. Iron Foundry
Grass Valley Steelcast
Lodi Iron Works, Inc.
Lodi Iron Works, Inc.
Wyman-Gordon Company
Richmond Micro Metals
Macaulay Foundry, Inc.
Coastcast Corporation
California Electric Steel
Pomona Foundry Inc.
Commercial Castings
PAC Foundries
Waterman Foundry
Pacific Steel Casting Co.
Pacific Steel Casting Company
Pacific Steel Casting Company
Covert Iron Works
American Brass & Iron Foundry
U.S. Pipe & Foundry Co., Inc.
Soundcast Co.
Gregg Industries
West Coast Stainless Products
Pacific Alloy Castings, Inc.
Modern Pattern & Foundry Co
Dameron Alloy Foundries
Globe Iron Foundry, Inc.
City
Jonesboro
Clarksville
Bentonville
Blytheville
Phoenix
Phoenix
Tempe
Tempe
Phoenix
South Gate
Huntington Park
Commerce
Oxnard
South Gate
Oakland
Rancho
Cucamonga
Alhambra
Fresno
Grass Valley
Lodi
Gait
San Leandro
Richmond
Berkeley
Rancho Dominguez
Angels Camp
Pomona
Fontana
Port Hueneme
Exeter
Berkeley
Berkeley
Berkeley
Huntington Park
Oakland
Union City
Costa Mesa
El Monte
Vemon
South Gate
Vemon
Compton
Commerce
State
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
Iron
392
28
7,927
23,055
0
0
73
13
0
103
329
0
0
6,214
360
0
4,500
1,639
0
2,462
0
0
188
11,330
0
0
578
5,000
0
7,437
0
0
0
1,825
41,193
77,343
927
31,279
0
4,100
0
0
4,500
Steel
0
0
0
0
339
508
0
26,117
2,928
558
1,055
4,000
150
0
400
200
0
0
389
0
220
338
188
0
2,199
806
30
0
840
0
12,000
7,900
7,700
0
10,298
0
927
0
850
0
383
756
0
per year
Total
392
28
7,927
23,055
339
508
73
26,130
2,928
661
1,384
4,000
150
6,214
760
200
4,500
1,639
389
2,462
220
338
375
11,330
2,199
806
608
5,000
840
7,437
12,000
7,900
7,700
1,825
51,491
77,343
1,853
31,279
850
4,100
383
756
4,500
A-2
-------
Metal poured, tons
ID
CA-35
CO-01
CO-02
CO-03
CO-04
CO-05
CT-01
CT-02
CT-03
CT-04
CT-05
FL-01
FL-02
FL-03
FL-04
FL-05
FL-06
GA-01
GA-02
GA-03
GA-04
IA-01
IA-02
IA-03
IA-04
IA-05
IA-06
IA-07
IA-08
IA-09
IA-10
IA-11
IA-12
IA-13
IA-14
IA-15
IA-16
IA-17
IA-18
IA-19
IL-01
IL-02
IL-03
Name
Strategic Materials Corp.
Svedala Industries Pumps & Process
Fountain Foundry, Inc.
Western Foundries
Goltra Castings Co., Inc.
The Electron Corp.
The Noank Foundry
Wyman Gordon Investment Castings
The Taylor and Fenn Company
Philbrick Booth & Spencer, Inc.
American Industrail FerroCast
Miami Castings
Precision Castings Incorporated
Consolidated Castings, Inc.
Florida Cast Products
Maddox Foundry & Machine Works
U.S. Pipe & Foundry Co., Inc.
Wheland Foundry - Warrenton
West Point Foundry & Machine Co.
Grinnell Corporation
Goldens' Foundry & Machine Co.
Seabee Corp Steel Foundry
Quality Foundry Co.
Griffin Wheel Company
Bloomfield Foundry, Inc.
Griffin Pipe Products Co,
Seneca Foundry, Inc.
Sivyer Steel Corporation
Max-Cast
Clow Valve Company — Foundry
Alloys Foundry
Iron Foundry
Quinn Machine & Foundry Inc.
Progessive Foundry, Inc.
Crane Valves
Blackhawk Foundry & Machine
Eagle Iron Works
John Deere Foundry Waterloo
Russelloy Foundry, Inc.
The Dexter Company
Wagner Castings Company
Wagner Havana
Chicago Heights Gray Iron Foundry
City
South Gate
Colorado Springs
Pueblo
Longmont
Golden
Littleton
Noank
Groton
Windsor
Hartford
Pawcatuck
Miami
Boca Raton
Jacksonville
Tampa
Archer
Medley
Warrenton
West Point
Statesboro
Columbus
Hampton
Stockton
Keokuk
Bloomfield
Council Bluffs
Webster City
Bettendorf
Kalona
Oskaloosa
Cedar Falls
Cedar Falls
Boone
Perry
Washington
Davenport
Des Moines
Waterloo
Durant
Fairfield
Decatur
Havana
Chicago Heights
State
CA
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IL
IL
IL
Iron
53
1,639
2,668
969
0
25,359
5
0
2,475
0
0
0
0
10
0
713
58,912
32,000
1,169
113,000
10,610
94
493
0
5,225
152,890
5,174
0
3
25,050
0
10,450
7,570
9,833
24,146
25,238
2,500
251,535
2,100
69,307
200,000
60,000
646
Steel
1,017
250
232
340
578
0
0
1,075
275
1,127
75
259
41
0
216
320
0
0
0
0
0
2,246
0
142,084
0
0
0
31,313
0
0
1,664
0
0
0
0
0
0
0
0
0
0
0
0
per year
Total
1,070
1,889
2,900
1,309
578
25,359
5
1,075
2,750
1,127
75
259
41
10
216
1,033
58,912
32,000
1,169
113,000
10,610
2,340
493
142,084
5,225
152,890
5,174
31,313
3
25,050
1,664
10,450
7,570
9,833
24,146
25,238
2,500
251,535
2,100
69,307
200,000
60,000
646
A-3
-------
Metal poured, tons
ID
IL-04
IL-05
IL-06
IL-07
IL-08
IL-09
IL-10
IL-11
IL-12
IL-13
IL-14
IL-15
IL-16
IL-17
IL-18
IL-19
IL-20
IL-21
IN-01
IN-02
IN-03
IN-04
IN-05
IN-06
IN-07
IN-08
IN-09
IN-10
IN-11
IN- 12
IN-13
IN-14
IN-15
IN-16
IN-17
IN-18
IN-19
IN-20
IN-21
IN-22
IN-23
IN-24
IN-25
IN-26
IN-27
Name
American Steel Foundries
Caterpillar Inc., Mapleton Foundry
National Castings Inc.
National Castings Inc.
Cast-Rite Steel Castings Corp.
Excelsior Foundry Co.
Lemfco Inc.
Aurora Industries, Inc.
Penberthy, Inc
Rockbridge Castings Inc.
Prime Cast, Inc.
Fansteel/Escast, Inc.
Gunite Corporation
Castwell Products, Inc.
The Francis & Nygren Foundry Co
Sterling Steel Foundry Inc.
Marengo Foundry Corp.
Sarcol, Inc
Chrysler Indianapolis Foundry
ABC Rail Corp.
Weil-McLain
Shenango Industries, Inc.
Tate Model & Engineering
Harrison Steel Castings Co.
Electric Steel Castings Company
Urschel Laboratories, Inc.
Grede New Castle, Inc.
Richmond Casting Company
Indianapolis Casting Corporation
Auburn Foundry Plant #1
Aubum Foundry Plant #2
Bremen Castings, Inc.
Columbia City Engineering, Inc.
Decatur Casting Division
Elkhart Foundry & Machine Co.
Interstate Castings
Precision Propeller, Inc.
Dalton Corporation
Casteel Service, Inc.
Aero Metals, Inc.
Teledyne Casting Service
Atlas Foundry Company, Inc.
RMG Foundry
Kendon Corporation
North Manchester Foundry Inc.
City
Granite City
Mapleton
Chicago
Melrose Park
Chicago
Belleville
Galena
Montgomery
Prophetstown
Rockbridge
South Beloit
Addison
Rockford
Skokie
Chicago
Sauget
Marengo
Chicago
Indianapolis
Anderson
Michigan City
Terre Haute
Kokomo
Attica
Indianapolis
Valparaiso
New Castle
Richmond
Indianapolis
Aubum
Aubum
Bremen
Columbia City
Decatur
Elkhart
Indianapolis
Indianapolis
Kendallville
Kingsbury
LaPorte
LaPorte
Marion
Mishawaka
Muncie
North Manchester
State
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
Iron
0
175,000
0
0
0
3,911
5,105
67
228
17
15,968
0
224,250
55,051
4,052
0
3,959
0
189,150
0
39,150
1,233
4
2,717
0
0
70,448
9,853
192,000
239,536
149,022
30,840
361
14,000
7,100
3,500
0
101,411
94
31
44,484
18,832
44,797
153
13,280
Steel
58,096
0
23,475
19,407
630
0
0
36
283
0
0
2,014
0
0
0
1,234
208
46
0
4,600
0
1,176
0
65,211
5,530
41
0
0
0
0
0
0
0
0
0
0
200
0
690
2,561
0
0
0
0
0
per year
Total
58,096
175,000
23,475
19,407
630
3,911
5,105
103
512
17
15,968
2,014
224,250
55,051
4,052
1,234
4,167
46
189,150
4,600
39,150
2,409
4
67,928
5,530
41
70,448
9,853
192,000
239,536
149,022
30,840
361
14,000
7,100
3,500
200
101,411
784
2,592
44,484
18,832
44,797
153
13,280
A-4
-------
Metal poured, tons per year
ID
IN-28
IN-29
IN-30
IN-31
IN-32
IN-33
IN-34
IN-35
IN-36
IN-37
IN-38
KS-01
KS-02
KS-03
KS-04
KS-05
KS-06
KS-07
KY-01
KY-02
LA-01
LA-02
LA-03
LA-04
LA-05
LA-06
MA-01
MA-02
MA-03
MA-04
MA-05
MA-06
MA-07
MA-08
MA-09
MA-10
MD-01
MD-02
MD-03
ME-01
ME-02
MI-01
MI-02
Name
Plymouth Foundry, Inc.
Rochester Metal Products
Gartland Foundry Company, Inc.
MedCast
Sterling Casting Corporation
Intat Precision, Inc.
Waupaca Foundry, Inc., Plant 5
Golden Casting Corporation
Sibley Machine and Foundry
American Steel Foundries
OMCO Cast Metals
Farrar Corporation
ACME Foundry
Grede Foundries Inc.
Griffin Wheel Company
Metlcast Products, Inc.
Ferroloy Foundry, Inc.
Atchinson Casting Corp.
Grede Foundries Inc.
Carrollton Casting Center
Bogalusa Iron Works, Inc.
Nadler Incorporated
Vulcan Foundry
Hendrix Manufacturing Co.
HICA Steel Foundry & Upgrade Co
Pearce Foundry, Inc.
Rodney Hunt Company
Ware Foundry, Inc.
Harmony Pattern & Casting Co
Trident Alloys, Inc.
LeBaron Foundry Inc.
Charlette Bros Foundry, Inc.
KomTek
Wollaston Alloys, Inc
Jahn Foundry Corp.
Belcher Corporation
Kaydon Ring & Seal Co.
ABC Rail Corp.
Pangbom Corporation
Enterprise Foundry, Inc.
Etheridge Foundry & Machine Co.
Shellcast, Inc.
Village Castings Company
City
Plymouth
Rochester
Terre Haute
Warsaw
Buffion
Rushville
Tell City
Columbus
South Bend
East Chicago
Winchester
Norwich
Coffeyville
Wichita
Kansas City
Salina
Wichita
Atchinson
Cynthiana
Carrollton
Bogalusa
Plaquemine
Denham Springs
Mansfield
Shreveport
Prairieville
Orange
Ware
Swansea
Springfield
Brockton
Blackstone
Worcester
Braintree
Springfield
Easton
Baltimore
Baltimore
Hagerstown
Lewiston
Portland
Montague
Caseville
State
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KY
KY
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
ME
ME
MI
MI
Iron
866
35,687
11,266
0
8,800
75,900
148,212
33,239
9,826
0
21,000
9,600
44,533
34,327
0
4,673
3,400
0
44,681
41,340
1,040
119
18,581
0
0
1,373
3,250
168
907
0
15,253
480
0
0
12,300
9,348
1,135
10,240
1,127
2,141
561
0
116
Steel
0
0
0
195
0
0
0
0
0
25,061
0
0
0
0
86,348
0
0
46,158
0
0
0
145
0
3,566
1,480
343
0
0
0
303
0
0
54
1,627
0
0
23
0
1,076
0
0
95
0
Total
866
35,687
11,266
195
8,800
75,900
148,212
33,239
9,826
25,061
21,000
9,600
44,533
34,327
86,348
4,673
3,400
46,158
44,681
41,340
1,040
264
18,581
3,566
1,480
1,716
3,250
168
907
303
15,253
480
54
1,627
12,300
9,348
1,158
10,240
2,203
2,141
561
95
116
A-5
-------
Metal poured, tons per year
ID
MI-03
MI-04
MI-05
MI-06
MI-07
MI-08
MI-09
MI-10
MI-11
MI-12
Ml-13
MI-14
MI- 15
MI-16
MI-17
MI- 18
MI- 19
MI-20
MI-21
MI-22
MI-23
MI-24
MI-25
MI-26
MI-27
MI-28
MI-29
MI-30
MI-31
MI-32
MI-33
MI-34
MI-35
MI-36
MI-37
MN-01
MN-02
MN-03
MN-04
MN-05
MN-06
MN-07
MN-08
Name
Midland Iron Works, Inc.
Giddings & Lewis Casting Technology
Dock Foundry Company
Northland Castings Corp.
Grede Foundries Inc.
Astech, Inc.
Hayes Albion Corporation
Thunder Bay Manufacturing Corp
Omega Castings, Inc.
Bay Cast Incorporated
CMI- Cast Parts, Inc.
East Jordan Iron Works, Inc.
Hastings Manufacturing Company
Pioneer Foundary
Great Lakes Casting Corp.
Midwest Metallurgical Lab.
Eagle Alloy, Inc.
West Michigan Steel
New Haven Foundry
RLM Industries, Inc.
Huron Casting, Inc.
Process Prototype Inc.
Sparta Foundry
Sturgis Foundry Corporation
Grede-Vassar, Inc.
Briggs & Stratton
Bemier Cast Metals, Inc.
Kurdziel Industries
Steeltech Ltd.
Specialty Castings, Inc.
GM Powertrain-Saginaw Metal Casting
Operations
GM Powertrain - Saginaw Malleable Iron Plant
Burgess-Norton Mfg. Co. Plant 3
Temperform Corporation
CWC Castings-Textron Inc.
Municipal Castings, Inc.
Dezurik Foundry
M E International - Duluth
AEGoetze - Lake City
Invest Cast, Inc.
Minncast Inc.
Grede Foundries Inc.
Pier Foundry
City
Midland
Menominee
Three Rivers
Hart
Kingsford
Vassar
Albion
Alpena
Battle Creek
Bay City
Cadillac
East Jordan
Hastings
Jackson
Ludington
Marshall
Muskegon
Muskegon
New Haven
Oxford
Pigeon
Romulus
Sparta
Sturgis
Vassar
Revenna
Saginaw
Rothbury
Grand Rapids
Springport
Saginaw
Saginaw
Muskegon
Novi
Muskegon
Madison
Sartell
Duluth
Lake City
Minneapolis
Fridley
St. Cloud
St. Paul
State
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
Ml
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI .
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
Iron
216
11,921
767
1,600
89,480
644
171,980
13,355
0
814
108,246
235,455
1,949
2,770
59,707
3,321
0
0
57,190
0
0
125
34,680
19,595
61,964
55,000
84
113,216
0
1,494
448,682
315,276
4,538
11
52,704
5,840
12,133
16,620
12,600
0
0
38,000
5,180
Steel
0
0
0
0
0
1,196
0
0
420
7,322
0
0
0
0
0
369
11,667
10,048
0
850
29,639
0
0
0
0
0
0
0
1,200
0
0
0
0
1,076
0
0
0
30,867
0
883
2,340
0
0
Total
216
11,921
767
1,600
89,480
1,840
171,980
13,355
420
8,135
108,246
235,455
1,949
2,770
59,707
3,690
11,667
10,048
57,190
850
29,639
125
34,680
19,595
61,964
55,000
84
113,216
1,200
1,494
448,682
315,276
4,538
1,087
52,704
5,840
12,133
47,487
12,600
883
2,340
38,000
5,180
A-6
-------
Metal poured, tons per year
ID
MN-09
MN-10
MN-11
MN-12
MN-13
MN-14
MN-15
MO-01
MO-02
MO-03
MO-04
MO-05
MO-06
MO-07
MO-08
MO-09
MO-10
MO-11
MO- 12
MS-01
MS-02
MS-03
MS-04
MT-01
NC-01
NC-02
NC-03
NC-04
NC-05
NE-01
NE-02
NE-03
NE-04
NE-05
NE-06
NH-01
NH-02
NH-03
NH-04
NH-05
NJ-01
NJ-02
Name
Badger Foundry Company
Gorman Company
United Machine & Foundry
Northern Castings Company
Brom Machine & Foundry Co., Inc.
Waltek Inc.
The Dotson Company, Inc.
Gold Foundry & Machine Works
Milton Gold D/B/A Foundry & Machine Works
The Carondelet Corporation
Mans-Steel Foundary
Gardner Denver Industrial Machinery
Monett Steel Castings
Ralston Purina Comp. Pet Products Support
Center
Midwest Alloys Foundry, Inc.
Missouri Precision
St. Louis Precision Casting Co.
Didion & Sons Foundry Co.
Standard Electric Steel Corp.
Dews Foundry
Laurel Machine & Foundry
ESCO Corporation
Southern Cast Products
AFFCO, Inc.
Sanders Co., Inc.
Hallman Foundry, Inc.
Stovall Foundry, Inc.
Foundry Service Company
Charlotte Pipe & Foundry Co.
Sioux City Foundry Company
Spitz Foundry, Inc.
Omaha Steel Castings Co.
Deeter Foundry, Inc.
Paxton-Mitchell Company
Phoenix Casting & Machining
Joy Technologies Inc
Pine Tree Castings, Division of Sturm, Ruger &
Co.
Hitchiner Mfg. Co., Inc
Nashua Foundries, Inc.
K.W. Thompson Tool Co.
General Foundry
Bierman-Everett Foundry Co.
City
Winona
Winona
Winona
Hibbing
Winona
Anoka
Mankato
Independence
Independence
Pevely
Mansfield
LaGrange
Monett
St. Louis
O'Fallon
Joplin
St. Louis
St. Peters
Springfield
Haiti esburg
Sandersville
Newton
Meridian
Anaconda
Elizabeth City
Sanford
Gastonia
Biscoe
Charlotte
South Sioux City
Hastings
Omaha
Lincoln
Omaha
Juniata
Claremont
Newport
Milford
Nashua
Rochester
Flagtown
Irvington
State
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
Iron
17,625
4,102
1,267
30,150
457
0
20,821
0
0
0
0
19,243
0
0
0
0
250
14,604
0
2,105
1,000
0
328
1,161
66
2,812
1,886
33,759
157,075
7,724
292
0
23,397
6,073
150
0
0
0
1,600
0
1,000
700
Steel
0
0
0
0
0
500
0
166
191
1,054
6,071
0
729
50
353
4,668
500
0
414
0
0
22,190
6,232
806
0
0
0
0
0
0
0
12,608
0
0
0
4,660
3,454
6,109
0
171
0
0
Total
17,625
4,102
1,267
30,150
457
500
20,821
166
191
1,054
6,071
19,243
729
50
353
4,668
750
14,604
414
2,105
1,000
22,190
6,560
1,967
66
2,812
1,886
33,759
157,075
7,724
292
12,608
23,397
6,073
150
4,660
3,454
6,109
1,600
171
1,000
700
A-7
-------
ID
NJ-03
NJ-04
NJ-05
NV-01
NY-01
NY-02
NY-03
NY-04
NY-05
NY-06
NY-07
OH-01
OH-02
OH-03
OH-04
OH-05
OH-06
OH-07
OH-08
OH-09
OH- 10
OH- 11
OH-12
OH- 13
OH-14
OH- 15
OH-16
OH-17
OH-18
OH-19
OH-20
OH-22
OH-23
OH-24
OH-25
OH-26
OH-27
OH-28
OH-29
OH-30
OH-31
OH-32
OH-33
OH-34
OH-35
Name
U.S. Pipe & Foundry Co., Inc.
Atlantic States Cast Iron Pipe
Griffin Pipe Products Co.
Wyman Gordon Sierra Cast Division
Oneida Foundries, Inc.
Jamestown Iron Works Inc.
Kennedy Valve
Frazer & Jones Co.
Gray-Syracuse, Inc.
Wormuth Brothers Foundry, Inc.
Pohlman Foundry Co., Inc.
Griffin Wheel Company
Rimer Enterprises, Inc.
Columbia Foundry Company
MorCast Precision, Inc.
The Wagnerware Copr. - Sidney Division
Quaker City Castings, Inc.
T & B Foundry Company
SanCasT, Inc.
Foundry Division
Tri-Cast, Inc.
The Blanchester Foundry Co.
Clow Water Systems Co.
OSCO Industries, Inc.
Cast-Fab Technologies, Inc.
Babcock & Wilcox Company
The Pioneer City Casting Company
Plant #3 & #4 - Foundry
United Foundries - Canton Plant
Chris Erhart Foundry & Machine Co.
Minster Machine Company
Buckeye Steel Castings Co.
Funk FineCast Inc.
Elano Corp., Casting Division
Webster Manufacturing Co.
Concorde Castings, Inc.
Hamilton Foundry Division
Ironton Iron, Inc.
Ohio Foundry
The Sawbrook Steel Castings Co
Alloy Cast Steel Company
Miami-Cast Inc.
The Quality Castings Co.
Quincy Castings, Inc.
St. Marys Foundry, Inc.
City
Burlington
Phillipsburg
Florence
Carson City
Oneida
Falconer
Elmira
Solvay
Chittenango
Athens
Buffalo
Groveport
Waterville
Columbiana
Columbus
Sidney
Salem
Cleveland
Coshocton
Dayton
Akron
Blanchester
Coshocton
Portsmouth
Cincinnati
Barberton
Belpre
Blanchester
Canton
Cincinnati
Minster
Columbus
Columbus
Xenia
Tiffin
Eastlake
Harrison
Ironton
Lima
Cincinnati
Marion
Miamisburg
Orrville
Quincy
St. Marys
State
NJ
NJ
NJ
NV
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
Metal pi
Iron
121,229
112,129
100,000
0
430
500
33,190
30,000
0
5,600
3,676
0
0
1,623
0
1,088
9,550
6,352
10,933
1,407
1,827
4,560
125,536
74,555
30,216
0
769
1,100
25,671
1,684
8,310
0
0
0
3,600
0
29,500
105,270
29,700
0
0
1,350
55,513
11,710
11,584
Dured, tons
Steel
0
0
0
730
0
0
0
0
1,434
0
0
136,469
466
2,836
9
0
5,140
0
0
4,006
0
0
0
0
0
4,025
58
0
0
0
0
89,070
690
28
0
323
0
0
0
7,323
404
0
0
0
0
per year
Total
121,229
112,129
100,000
730
430
500
33,190
30,000
1,434
5,600
3,676
136,469
466
4,459
9
1,088
14,690
6,352
10,933
5,413
1,827
4,560
125,536
74,555
30,216
4,025
827
1,100
25,671
1,684
8,310
89,070
690
28
3,600
323
29,500
105,270
29,700
7,323
404
1,350
55,513
11,710
11,584
A-8
-------
ID
OH-36
OH-37
OH-38
OH-39
OH-40
OH-42
OH-43
OH-44
OH-45
OH-46
OH-47
OH-48
OH-49
OH-51
OH-52
OH-53
OH-54
OH-55
OH-56
OH-58
OH-59
OH-60
OH-61
OH-62
OH-63
OH-64
OH-65
OK-01
OK-02
OK-03
OK-04
OK-05
OK-06
OK-07
OK-08
OK-09
OK-10
OK- 11
OR-01
OR-02
OR-03
OR-04
OR-05
OR-06
OR-07
Name
Sandusky International, Inc.
Ohio Foundry, Inc.
Fisher Cast Steel Prod. Inc.
Bescast, Inc.
Xenia Foundry
Precision Metalsmiths, Inc.
OSCO Industries, Inc.
Kurdziel Iron of Wauseon
Electro-Alloys Corp.
Ford Motor Company
GM Powertrain - Defiance Plant
U.S. Casting Company
American Steel Foundries
The General Casting Co.
The General Casting Co.
The General Casting Co.
The General Casting Co.
The General Casting Co.
The General Casting Co.
Kenton Iron Products, Inc.
Eljer Plumbingware, Inc.
Commercial Casting Co.
The G & C Foundry Co.
The Knapp Foundry Company
OSCO Industries, Inc.
The O.S. Kelly Company
The Bimac Corporation
Big Four Alloy Castings
B&L Foundry, Inc.
American Alloy Division 2
American Foundry Group, Inc
American Alloy Division
Grede-Pryor Foundry Inc.
Flanagan Iron Works
The Electron Corporation
Central Machine & Tool Co.
Tonkawa Foundry, Inc.
Jencast,-Jensen International
ESCO Corp. - Main Plant
PED Manufacturing, Ltd.
Durametal Corporation
ESCO - Plant 3
Eagle Foundry Company
Varicast, Inc. (Portland Plant)
Wolf Steel Foundry, Inc.
City
Sandusky
Tallmadge
West Jefferson
Willoughby
Xenia
Euclid
New Boston
Wauseon
Elyria
Brookpark
Defiance
Canal Fulton
Alliance
West Liberty
Delaware
Cincinnati
Columbus
Delaware
Grafton
Kenton
Salem
New Philadelphia
Sandusky
Akron
Jackson
Springfield
Dayton
Tulsa
Tonkawa
Muskogee
Bixby
Bixby
Pryor
Tulsa
Blackwell
Enid
Tonkawa
So. Coffeyville
Portland
Oregon City
Tualatin
Portland
Eagle Creek
Portland
Hubbard
State
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
Metal pc
Iron
0
21
0
0
1,480
2
5,439
23,288
0
526,000
1,368,167
27
0
7,677
8,461
2,800
2,526
19,449
8,997
2,564
31,181
134
18,675
803
33,355
5,000
0
0
34
0
0
0
105,000
1,426
33,000
979
860
3,580
392
0
3,367
0
8,234
3,021
0
mred, tons
Steel
10,291
401
1,500
34
0
170
0
0
4,434
0
0
2,653
35,200
0
0
0
0
0
0
0
0
0
0
0
0
0
394
156
50
2,484
151
717
0
49
0
74
0
4,037
19,213
434
459
16,031
3,529
500
6,594
per year
Total
10,291
422
1,500
34
1,480
172
5,439
23,288
4,434
526,000
1,368,167
2,680
35,200
7,677
8,461
2,800
2,526
19,449
8,997
2,564
31,181
134
18,675
803
33,355
5,000
394
156
84
2,484
151
717
105,000
1,474
33,000
1,053
860
7,617
19,605
434
3,826
16,031
11,763
3,521
6,594
A-9
-------
Metal poured, tons
ID
OR-08
OR-09
PA-01
PA-02
PA-03
PA-04
PA-05
PA-06
PA-07
PA-08
PA-09
PA-10
PA- 11
PA-12
PA-13
PA- 14
PA-15
PA-16
PA-17
PA-18
PA-19
PA-20
PA-21
PA-22
PA-23
PA-24
PA-25
PA-26
PA-27
PA-28
PA-29
PA-30
PA-31
PA-32
PA-33
PA-34
PA-35
PA-36
PA-37
PA-38
PA-39
PA-40
PA-41
PA-42
PA-43
Name
PCC Standards, Inc.
Columbia Steel Casting Co
Brighton Electric Steel Casting
Ward Manufacturing, Inc.
Zum Cast Metals Operations
Goulds Pumps, Inc.
T.B. Wood's Sons Company
Tyler Pipe- Ransom Industries Inc.
Grinnell
Tom Ondrejko Co.
Empire Steel Castings, Inc.
EAFCO, Inc.
WHEMCO Midland Foundry
Centec Roll Corporation
Benton Foundry, Inc
United Foundry Company, Inc
Weatherly Casting & Machine Co.
Somerset Foundry & Machine
Wyano Foundry
Hazleton Pumps, Inc.
Delvest, Inc
Urick Foundry
Victaulic Company of America
Victaulic Company of America
Ephrata Manufacturing Co.
Pennsylvania Steel Foundry & Machine
CM1- Quaker Alloy, Inc.
Investment Casting Corp.
Damascus Steel Casting Co.
Muncy Foundry
McConway & Torley Corporation
Duraloy Technologies, Inc.
Saxonburg Foundry Co.
Hamburg Mfg., Inc.
Washington Mould Company
Donsco, Inc.
Mt. Joy Foundry
Wrightsville Foundry - Building #8
W.O.HickokMfg. Co.
McConway & Torley Corporation
The Frog, Switch & Manufacturing Co.
Hodge Foundry
Advanced Cast Products, Inc.
Frontier Foundry, Inc.
Penncast Corporation
City
Portland
Portland
Beaver Falls
Blossburg
Erie
Ashland
Chambersburg
Macungie
Columbia
Washington
Reading
Boyertown
Midland
Bethlehem
Benton
Johnstown
Weatherly
Somerset
Wyano
Hazelton
West Chester
Erie
Easton
Alburtis
Ephrata
Hamburg
Myerstown
Meadville
New Brighton
Muncy
Kutztown
Scottdale
Saxonburg
Hamburg
Washington
Wrightsville
Mt. Joy
Belleville
Harrisburg
Pittsburgh
Carlisle
Greenville
Meadville
Titusville
Marietta
State
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
Iron
0
2,938
0
122,211
16,847
1,747
31,166
37,274
126,997
57
63
28,541
0
9,309
18,248
80
2,824
2,321
1,384
601
0
33,026
49,244
39,687
3,444
0
0
0
0
2,814
0
0
660
5,300
1,165
49,331
18,404
18,421
328
0
0
18,000
25,000
0
0
Steel
4,920
29,710
1,344
0
0
2,413
0
0
0
13
4,105
0
32,279
249
0
80
0
0
0
324
216
0
0
0
0
8,147
5,187
6
1,500
3,582
28,017
3,756
0
0
0
0
0
0
0
20,041
25,145
0
0
34
3,341
per year
Total
4,920
32,648
1,344
122,211
16,847
4,160
31,166
37,274
126,997
69
4,168
28,541
32,279
9,558
18,248
160
2,824
2,321
1,384
925
216
33,026
49,244
39,687
3,444
8,147
5,187
6
1,500
6,396
28,017
3,756
660
5,300
1,165
49,331
18,404
18,421
328
20,041
25,145
18,000
25,000
34
3,341
A-10
-------
Metal poured, tons
ID
PA-44
PA-45
PA-46
PA-47
PA-48
PA-49
PA-50
PA-51
PA-52
PA-53
PA-54
RI-01
RI-02
RI-03
SC-01
SC-02
SC-03
SC-04
SC-05
SC-06
SC-07
SC-08
SD-01
TN-01
TN-02
TN-03
TN-04
TN-05
TN-06
TN-07
TN-08
TN-09
TN-10
TN-11
TN-12
TN-13
TN-14
TN-15
TN-16
TN-17
TX-01
TX-02
TX-03
TX-04
TX-05
Name
The General Casting Co.
Kennametal Castings
EMI Company
Spring City Foundry Company
CMI- Tech Cast, Inc.
Hempfield Foundry
Quality Investment Castings, Inc.
Monaca Plant
Nova Precision Casting Corp.
Hale Pump - Foundry Division
Harcast Co., Inc.
Seaboard Foundry Inc.
Cumberland Foundry Co., Inc.
Fairmount Foundry Inc.
Conbraco Industries, Inc.
Cast Products Co., Inc.
Carolina Casting Corporation
Pinebrook Foundry
Grede Foundries Inc. - Greenwood
Synehi Castings
US Filter/Wheelabrator Cast Products
Bahan Machine & Foundry
Mereen Johnson Machine Co.
Wheland Foundry - No. 2 Foundry
U.S. Pipe & Foundry Co., Inc.
Tennessee Investment Casting Co.
Lodge Manufacturing Co.
Camden Casting Center
Clinch River Casting, Inc.
Accu-Cast
Acheson Foundry & Machine
Mueller Company
Wheland Foundry - Middle Street
Wheland Foundry - #1 Foundry
Cleveland Foundry & Mfg. Co, Inc.
John Bouchard & Sons Foundry
American Magotteaux Corporation
Eureka Foundry
Clarksville Foundry, Inc.
Vestal Manufacturing Co.
Manufactured Alloys, Inc.
National Foundry & Mfg. Inc.
Texaloy Foundry
Dai-Air Investment Castings, Inc.
Consolidated Castings Corp.
City
Shippensburg
Bedford
Erie
Spring City
Myerstown
Greensburg
Blandon
Monaca
Auburn
Conshohocken
Chester
Johnston
Cumberland
Woonsocket
Conway
Westminster
Hardeeville
Great Falls
Greenwood
Greenwood
Walterboro
Taylors
Webster
Chattanooga
Chattanooga
Bristol
South Pittsburg
Camden
Caryville
Chattanooga
Chattanooga
Chattanooga
Chattanooga
Chattanooga
Cleveland
Nashville
Pulaski
Chattanooga
Clarksville
Sweetwater
Luling
Crane
Floresville
Point
Hutchins
State
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
RI
RI
RI
SC
SC
SC
SC
SC
SC
SC
SC
SD
TN
TO
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TO
TN
TX
TX
TX
TX
TX
Iron
9,659
0
69,800
1,042
0
2,289
3
76
0
700
0
1,280
925
850
0
100
66
25
85,400
750
975
1,200
3,975
200,000
49,656
8
18,637
50,800
850
0
540
50,000
262,800
164,820
844
4,403
57,323
5,670
817
14,104
0
0
1,345
0
0
Steel
0
358
0
0
2,244
0
281
1,824
200
0
87
0
0
0
4,172
0
265
0
0
0
525
0
0
0
0
400
0
0
0
600
60
0
0
0
0
0
16,378
0
12
0
6
350
0
180
1,640
per year
Total
9,659
358
69,800
' 1,042
2,244
2,289
284
1,900
200
700
87
1,280
925
850
4,172
100
332
25
85,400
750
1,500
1,200
3,975
200,000
49,656
408
18,637
50,800
850
600
600
50,000
262,800
164,820
844
4,403
73,701
5,670
829
14,104
6
350
1,345
180
1,640
A-ll
-------
Metal poured, tons per year
ID
TX-06
TX-07
TX-08
TX-09
TX-10
TX-11
TX-12
TX-13
TX-14
TX-15
TX-16
TX-17
TX-18
TX-19
TX-20
TX-21
TX-22
TX-23
TX-24
TX-25
TX-26
TX-27
TX-28
TX-29
TX-30
TX-31
TX-32
TX-33
UT-01
UT-02
UT-03
UT-04
UT-05
VA-01
VA-02
VA-03
VA-04
VA-05
VA-06
VA-07
VA-08
VA-09
VT-OI
Name
American Spincast, Inc.
Alamo Iron Works Foundry
Texcast, Inc
Oil City Iron Works, Inc.
Greens Bayou Foundry, Inc.
Western Iron Works, Inc.
Taylor Foundry Company
Mabry Foundry Inc. of Beaumont
Texas Foundries
Goulds Pumps - Turbine Division
Delta Centrifugal Corporation
Victoria Precision Alloys, Inc.
Lufkin Industries, Inc
Hensley Industries, Inc.
Harrisburg Woolley Tool Co.
Penatek Industries, Inc.
Sure Cast, Inc.
Centrifugal Castings, Inc.
Gulf Star Foundry
A A Foundries, Inc.
Texas Precision Metalcraft, Inc.
Tyler Pipe Company
Southwest Steel Casting Company
Gainesville Foundry, Inc.
Texas Steel Company
Martin Foundry- Martin Sprocket & Gear
Smith Steel Casting Co., Inc.
Henderson Manufacturing Co., Inc.
Star Foundry & Machine
Maca Supply Company
Tony Metals Corp.
Pacific States Cast Iron Pipe
GSC Foundries, Inc., Steel Division
O.K. Foundary Co., Inc.
Emporia Foundry, Inc.
Walker Machine & Foundry Corp
mtermet Corporation
Internet Corporation
Graham- White Manufacturing Co
Newport News Shipbuilding and Dry
Intermet Corporation
Griffin Pipe Products Co.
Vermont Castings - Foundry Division
City
Belton
San Antonio
Houston
Corsicana
Houston
San Angelo
Wichita Falls
Beaumont
Lufikin
Slaton
Temple
Victoria
Lufkin
Dallas, Dallas
County
Odessa
Odessa
Bumet
Temple
Corpus Christi
San Antonio
Sugar Land
Tyler
Longview
Gainesville
Fort Worth
Dallas
Marshall
Pittsburg
Salt Lake City
Springville
Spanish Fork
Provo
Ogden
Richmond
Emporia
Roanoke
Radford
Radford
Salem
Newport News
Lynchburg
Lynchburg
Randolph
State
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
Iron
0
2,125
0
14,640
500
10,000
8,690
18,000
123,286
4,866
0
140
75,000
0
2,539
2,040
6
0
12
193
0
219,308
0
5,038
0
8,338
0
0
234
0
20
103,440
0
1,289
14,508
7,929
98,600
135,600
2,921
0
148,213
137,600
20,704
Steel
5,390
0
111
0
0
0
0
0
0
0
2,303
220
0
34,895
0
1,005
676
1,975
0
1
250
0
17,170
0
33,341
0
1,170
2,100
1,248
558
33
0
270
0
0
0
0
0
0
517
0
0
0
Total
5,390
2,125
111
14,640
500
10,000
8,690
18,000
123,286
4,866
2,303
360
75,000
34,895
2,539
3,045
682
1,975
12
194
250
219,308
17,170
5,038
33,341
8,338
1,170
2,100
1,482
558
53
103,440
270
1,289
14,508
7,929
98,600
135,600
2,921
517
148,213
137,600
20,704
A-12
-------
Metal poured, tons
ID
VT-02
VT-03
WA-01
WA-02
WA-03
WA-04
WA-05
WA-06
WA-07
WA-08
WA-09
WA-10
WA-11
WA-12
WA-13
WA-14
WI-01
WI-02
WI-03
WI-04
WI-05
WI-06
WI-07
WI-08
Wl-09
WI-10
WI-11
WI-12
WI-13
WI-14
WI-15
WI-16
WI-17
WI-18
WI-19
WI-20
WI-21
WI-22
WI-23
WI-24
WI-25
WI-26
WI-27
WI-28
WI-29
Name
Brown Foundry
Vestshell Vermont, Inc.
Vancouver Foundry Co (DBA Varicast)
Mackenzie Specialty Castings
Western Steel Casting Co.
Northwest Castings
D & L Foundry, Inc.
N.E.W. Castings Inc.
Dyko Foundry
Sather Mfg. Co.
Meltec-Division of Young Corp
Spokane Steel Foundry
Spokane Precision Castings
Quali-Cast Foundry, Inc.
Atlas Foundry & Machine Co
SeaCast/ Eagle, Inc.
Waupaca Foundry, Inc. - Plant 1
Tomahawk Foundry Inc.
Cast Tools, Inc.
Shelmet Precision Casting Co.
Pelton Casteel, Inc.
Austin Gray Iron Foundry
Northern Precision Casting Co., Inc.
Belgium Foundry
Washbum Iron Works, Inc.
Willman Industries, Inc.
Modem Plate Co. Inc. of WI
Racine Steel Castings
Roloff Manufacturing Corporation
Baker Manufacturing Co.
Kirsch Foundry Inc.
Prime Cast, Inc.
Castalloy Corporation
Iroquois Foundry Co.
Vilter Manufacturing Corporation
Bay Engineered Castings
Investment Casting Plant
Torrance Casting, Inc.
Aelco Foundries
Briggs & Stratton
The Falk Corporation
Grede Foundries Inc.
Kramer International, Inc.
Brillion Iron Works
Maynard Steel Casting Co.
City
Swanton
St. Albans
Vancouver
Arlington
Seattle
Seattle
Moses Lake
Spokane
Spokane
Everett
Seattle
Spokane
Spokane
Chehalis
Tacoma
Marysville
Waupaca
Rice Lake
Racine
Wild Rose
Milwaukee
Sheboygan
Lake Geneva
Belgium
Washburn
Cedar Grove
Racine
Racine
Kaukauna
Evansville
Beaver Dam
Beloit
Waukesha
Browntown
Milwaukee
DePere
Fond du Lac
La Crosse
Milwaukee
West Allis
Milwaukee
Milwaukee
Milwaukee
Brillion
Milwaukee
State
VT
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
Iron
1,780
0
0
436
90
0
12,288
809
938
2,643
0
7,032
0
0
0
0
206,916
740
0
0
0
3,504
0
3,889
3,001
10,469
39
0
4,842
8,597
9,898
6,299
0
20,439
820
12,949
0
5,380
0
101,000
0
2,009
218
251,430
0
Steel
0
163
5,137
498
2,910
2,643
0
0
902
0
3,000
3,014
332
1,954
10,632
832
0
0
154
276
15,168
0
1,600
0
0
0
0
13,732
0
0
0
2,870
3,753
0
100
0
1,310
0
4,322
0
23,266
42,113
372
0
23,560
per year
Total
1,780
163
5,137
934
3,000
2,643
12,288
809
1,840
2,643
3,000
10,046
332
1,954
10,632
832
206,916
740
154
276
15,168
3,504
1,600
3,889
3,001
10,469
39
13,732
4,842
8,597
9,898
9,169
3,753
20,439
920
12,949
1,310
5,380
4,322
101,000
23,266
44,122
590
251,430
23,560
A-13
-------
ID
WI-30
WI-31
WI-32
WI-33
WI-34
WI-35
WI-36
WI-37
WI-38
WI-39
WI-40
WI-41
WI-42
WI-43
WI-44
Wl-45
WI-46
WI-47
WI-48
WI-49
WI-50
WI-52
WI-53
WI-54
WI-55
WI-56
WV-01
WV-02
WV-03
Name
Milwaukee Malleable & Grey Iron Works
OMC Milwaukee
Stainless Foundry & Engineering
Badger Iron Works, Inc.
Northern Stainless Steel Corp.
Grede Foundries, Inc.
Richland Center Foundry Co.
Wisconsin Investcast
Navistar International, Transportation Co
Wisconsin Centrifugal
J&L Fiber Services
Waunakee Alloy Casting Corp.
Waupaca Foundry Plant 2/3
Waupaca Foundry, Inc. - Plant 4
Liberty Foundry
Northern Steel Castings, Inc.
Precision Metalsmiths, Inc.
Berlin Foundry Corporation
Spuncast Inc.
Neenah Foundry Company Plant #2
Neenah Foundry Company Plant #3
Waukesha Foundry Inc.
Mid-City Foundry Co., Inc.
Mid-City Foundry Co., Inc.
Winsert, Inc.
Craft Cast
Kelly Foundry & Machine Co., Inc.
Centre Foundry & Machine Co
Sturm Inc.
City
Milwaukee
Milwaukee
Milwaukee
Menomonie
Pewaukee
Reedsburg
Richland Center
Watertown
Waukesha
Waukesha
Waukesha
Waunakee
Waupaca
Marinette
Wauwatosa
Wisconsin Rapids
Markesan
Berlin
Watertown
Neenah
Neenah
Waukesha
Milwaukee
Grafton
Marinette
Jackson
Elkins
Wheeling
Barboursville
TOTALS
State
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
wv
wv
wv
Metal p<
Iron
3,663
0
6
3,963
0
273,987
19,529
12
73,811
0
3,669
0
615,353
325,133
32,978
0
221
43,939
51
137,586
118,343
0
6,348
2,258
0
0
1,632
20,364
0
15,595,229
»ured, tons
Steel
0
425
5,846
0
490
0
0
500
0
12,494
5,503
508
0
0
0
4,100
883
0
5,048
0
0
807
0
0
509
345
0
0
4,000
1,904,131
per year
Total
3,663
425
5,852
3,963
490
273,987
19,529
512
73,811
12,494
9,172
508
615,353
325,133
32,978
4,100
1,104
43,939
5,099
137,586
118,343
807
6,348
2,258
509
345
1,632
20,364
4,000
17,499,360
A-14
-------
APPENDIX B
ESTIMATED HAP EMISSIONS
FROM MOLD AND CORE MAKING OPERATIONS
-------
B.I ESTIMATED HAP EMISSIONS FROM MOLD AND CORE MAKING
OPERATIONS
The EPA made emission estimates for mold- and core-making operations by determining
the HAP content of the binder systems and estimating the fractions of HAP that were evaporated
during mixing and curing. Actual chemical usage for foundries plus information on Material
Safety Data Sheets was used to determine HAP content whenever this information was available.
When no data were available, usage was estimated from the amount of sand reported used for
each system, the proportion of chemical to sand (reported or estimated), and typical values for
HAP constituents of the binder components. Table B-l gives HAP content in the components,
and Table B-2 gives typical proportions of the constituents; information in these tables was
furnished by the chemical supplier industry. Table B-3 shows the overall HAP content of the
systems based on information from Tables B-l and B-2 and also shows typical proportions of
chemicals to sand, which were determined from information reported by foundries.
Because no actual emission data were available, estimates of the fraction of HAP emitted
were made using information provided by the AFS, which consists of values for the fractions of
HAP that react during the mold- and core-making process, evaporate during the process, and
remain in the mold or core after it is cured but before it is exposed to molten metal. These values
are given in Table B-4.
In determining emission factors, some of the values reported in Table B-4 were modified
based either on more definitive information or on the premise that a more conservative estimate
was appropriate. First, the fraction of phenol reacted in the phenolic Novolac system was
assumed to be 35, not 95, percent based on information from an industry supplier. Second, the
values given for nonreacting constituents (naphthalene, cumene, xylene, and biphenyl) of the
phenolic urethane no-bake and cold-box systems were recalculated. The values given by the
AFS were weight loss data determined from a study made by supplier companies for the Ohio
Cast Metals Association and the AFS (RMT, 1998). The values assume that the fraction of
weight lost by all chemicals is the same. In fact, the formaldehyde, phenol, and MDI in these
systems react almost completely and cannot evaporate. The proportions of the nonreacting
constituents that evaporate must therefore be greater that the overall weight loss. Taking this
fact into account and using the composition information in Tables B-l and B-2, the EPA
estimates that approximately 9 percent instead of 3.25 percent of the nonreacting chemicals in
B-l
-------
the phenolic urethane cold-box system and 16 percent instead of 5.85 percent of those chemicals
in the phenolic urethane no-bake system are emitted.
Finally, because the basis for estimates for the other systems is not given, EPA made the
conservative assumption that all of the chemicals that do not react are emitted, which includes
the fractions reported emitted plus those reported remaining in the mold or core. Taking into
account the above considerations, Table B-5 shows the fractions of chemicals that the EPA
assumes are emitted.
To summarize all of the above information, Table B-6 is a list of emission factors derived
from the information presented in Tables B-l through B-5. These factors were used to estimate
average or nominal mold- and core-making emissions from foundries.
The EPA also made worst-case emission estimates based on the highest reported HAP
content of binder systems and the highest reported chemical-to-sand ratios, in order to identify
foundries that are obviously not major sources of HAP based on mold- and core-making
operations alone. Table B-7 gives estimates of HAP emissions per 50,000 tons of sand
processed; this level of operation results in HAP emissions on the order of 10 and 25 tons based
on individual and total HAP, respectively, for several systems. This basis of estimation also
allows estimates to be made easily based on tons of metal poured, assuming that the sand-to-
metal metal ratio is known. Table B-8 gives data needed to develop Table B-7, and tables B-9
and B-10 show intermediate determinations in terms of emissions per 100 pounds of chemicals
and tons of sand, respectively.
Tables B-l through B-10 use nomenclature for binder systems that is used by the industry
as reported in an AFS guidance document for estimating emissions from these systems. The
nomenclature used in EPA's 1998 industry survey is somewhat different. To avoid confusion,
Table B-l 1, which gives a comparison between the two systems, is included in this discussion.
B-2
-------
TABLE B-l. HAP CONTENT OF SAND BINDER SYSTEM COMPONENTS1>2
Binder system
Acrylic/epoxy/SO2
Furan hot box
Furan nobake
Furan/SO2
Furan warm box
Phenolic baking
Phenolic ester
nobake and cold box
Phenolic hot box
Phenolic nobake
Phenolic Novolac
liquid
Phenolic Novolac
flake
Component
Resin
Resin
Resin
Catalyst
Resin
Oxidizer
Resin
Catalyst
Resin
Resin
Resin
Resin
Catalyst
Resin
Resin
HAP present
Cumene
Formaldehyde
Phenol
Formaldehyde
Methanol
Methanol
Formaldehyde
Methanol
Dimethyl phthalate
Methyl ethyl ketone
Formaldehyde
Methanol
Phenol
Formaldehyde
Phenol
Formaldehyde
Phenol
Formaldehyde
Phenol
Formaldehyde
Methanol
Methanol
Phenol
Formaldehyde
Methanol
Phenol
Formaldehyde
Methanol
l|
Amount of HAP in component, percent
Range
5. minimum3
2. -5.
0. -4.
0. -1.
2. -4.
20. - 30.
1.-4.
1.-3.
40. - 50.
0. 2.
O.-l.
45. -55.
3. -14.
0. -2.
2. -8.
0. -2.
2. -8.
1.-4.
8. - 14.
0. -2.
2. -4.
20. - 30.
1.-4.
0. -3.
0. -15.
1.5-8.04
0. -0.24
0. 10.44
Typical |
5.
3.
0.
0.1
3.
27.
2.
2.
45.
2.
0.5
50.
8.
1.
4.
0.5
5.
2.
12.
0.5
3.
27.
2.
0.5
5.
5.54
O.4
O.4
B-3
-------
TABLE B-l. HAP CONTENT OF SAND BINDER SYSTEM COMPONENTS*'2
Binder system
Phenolic urethane
nobake and cold box
Urea formaldehyde
Component
Resin
Coreactant
Resin
HAP present
Phenol
Formaldehyde
Naphthalene
Cumene
Xylene
Naphthalene
Cumene
Xylene
Formaldehyde
Amount of HAP in component, percent
Range
3. -8.
O.-l.
0. -2.
0. 2.
O.-l.
0.-3.
0. 1.
O.-l.
1.-4.
Typical
6.
0.1
1.
0.5
0.1
1.
0.1
0.1
1.
Source: Stone, 1999, and Jonathan A. Stone, Delta Resins and Refractories, Delta-HA (private communication to J. H.
Maysilles, U.S. EPA., November 15,1999) except where noted.
Only HAP that could be emitted because of incomplete reaction or nonreaction are listed.
Information supplied by Joe Fox, Ashland Chemical, Inc. Private communication to J. H. Maysilles, U.S. EPA, August 16,
2000.
Information is based on Material Safety Data Sheets from foundries that use this system.
B-4
-------
TABLE B-2. PROPORTION OF COMPONENTS IN BINDER SYSTEMS'
Binder system
Furan nobake
Furan/SO2
Furan warmbox
Phenolic nobake
Phenolic urethane
nobake and coldbox
Component
Resin
Catalyst
Resin
Oxidizer
Resin
Catalyst
Resin
Catalyst
Resin
Coreactant
Proportion in system, percent
Range
55-80
20-45
50-55
45-50
75-85
15-25
55-75
25-45
50-60
40-50
Typical
70
30
55
45
80
20
68
32
55
45
1 Information supplied by Jonathan A. Stone, Delta Resins and Refractories, Delta-HA.. Private communication-to J. H.
Maysilles, U.S. EPA.
B-5
-------
TABLE B-3. AVERAGE CHEMICALS/SAND RATIOS AND TYPICAL CHEMICAL CONTENT IN BINDER SYSTEMS
Binder system
Acrylic/Epoxy/SO,
Furan hot box
Furan nobake
Furan/SO,
Furan warm box
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake
Phenolic-Novolac flake
Phenolic-Novolac liquid
Phenolic urethane nobake
Phenolic urethane cold box -
resin plus coreactant
Phenolic urethane cold box -
gas
Urea formaldehyde
Pounds
chemical
per ton of
sand '
34.
40.
24.
30.
32.
30. 3
•> i
jj.
32.
30.-'
27.
50.
50.
25.
30.
*t
:>.
30.3
Chemical content of system, percent.2
Phenol
0.
8.
4.
4.
5.
8.2
5.5
2
-> *>
j.j
3.3
Formal-
dehyde
•}
_3 _
0.07
1.1
0.4
1.
0.5
0.5
T
0.34
0.
0.5
0.055
0.055
1.
Naph
thalene
1.
1.
Cumene
5.
0.32
0.32
Xylene
0.1
0.1
Biphenyl
o.4
Methanol
10.2
1.1
10.
o.4
10.7
0.
5.
Glycol
ethers
O.4
Dimethyl
phthalate
20.2
Methyl
ethyl
ketone
0.9
TEA !
100. 5
Source: EPA, 1998.
MD1 is not included because essentially all MD1 is reacted and therefore not available for evaporation.
Values are based on nominal values for other systems; not enough information is available for these systems to establish values.
Not mentioned in the source of information for table C-l; assumed to be zero.
Assumes that the catalyst gas is triethylamme; dimethylethylamine, which is not a HAP, is used in some operations
B-6
-------
TABLE B-4. HAP EMITED FROM CHEMICAL BINDER SYSTEMS
USED FOR SAND CORES AND MOLDS (AFS, 1998)
Binder system
Furan hotbox
Furan nobake
Furan/SO2
Furan warmbox
Phenolic baking
Phenolic hotbox
Urea formaldehyde
Phenolic Novolac
liquid (Shell)
Phenolic Novolac
flake (Shell)
Phenolic nobake
(Acid catalyzed)
HAP and component in
which it is used
Formaldehyde, resin
Phenol, resin
Formaldehyde, resin
Methanol, resin
Methanol, catalyst
Formaldehyde, resin
Methanol, resin
Dimethyl phthalate, oxidizer
Methyl ethyl ketone,
oxidizer
Formaldehyde, resin
Methanol, catalyst
Phenol, Part I
Formaldehyde, Part I
Formaldehyde, resin
Phenol, resin
Formaldehyde, Part I
Phenol, Part I
Formaldehyde, Part I
Methanol, Part I
Phenol, resin
Phenol, resin
Formaldehyde, resin
Methanol, resin
Methanol, acid
Percent
reacted
95
98+
98
0
0
98
0
0
0
95
0
95
95
95
95
98
95
95
0
95
98
98
0
0
Percent emitted during
core and mold making '
5
0
2
50
50
2
50
50
50
5
100
0
5
5
0
2
0
5
100
0
0
2
50
50
Percent
remaining in
mold or core
0
2
0
50
50
0
50
50
50
0
0
5
0
0
5
0
5
0
0
5
2
0
50
50
B-7
-------
TABLE B-4. HAP EMITED FROM CHEMICAL BINDER SYSTEMS
USED FOR SAND CORES AND MOLDS (AFS, 1998)
Binder system
Phenolic urethane
nobake
Phenolic urethane
coldbox
Alkyd oil
Phenolic ester
nobake
HAP and component in
which it is used
Formaldehyde, Part I
Phenol, Part I
Xylene, Part I
Cumene, Part I
Naphthalene, Part I
Methylene phenylene
isocyanate, Part II
Xylene, Part II (catalyst)
Cumene, Part II (catalyst)
Naphthalene, Part II
Formaldehyde, Part I
Phenol, Part I
Xylene, Part I
Cumene, Part I
Naphthalene , Part I
Methylene phenylene
isocyanate, Part II
Naphthalene, Part II
Xylene, Part II
Biphenyl, Part II
Lead, resin
Cobalt, resin
Methylene phenylene
isocyanate, coreactant
Formaldehyde, resin
Phenol, resin
Percent
reacted
98
98
0
0
0
99.99
0
0
0
98
98
0
0
0
99.99
0
0
0
0
0
99.99
98
98
Percent emitted during
core and mold making '
2
0
5.85
5.85
5.85
0
5.85
5.85
5.85
2
0
3.25
3.25
3.25
0
3.25
3.25
3.25
0
0
<0.01
2
0
Percent
remaining in
mold or core
0
2
94.15
94.15
94.15
0.01
94.15
94.15
94.15
0
2
96.75
96.75
96.75
0.01
96.75
96.75
96.75
100
100
0.01
0
2
B-8
-------
TABLE B-4. HAP EMITED FROM CHEMICAL BINDER SYSTEMS
USED FOR SAND CORES AND MOLDS (AFS, 1998)
Binder system
Phenolic ester cold
box
Acrylic/epoxy/SO2
HAP and component in
which it is used
Formaldehyde, resin
Phenol, resin
Methanol, coreactant
Cumene, Part I
Percent
reacted
98
98
0
0
Percent emitted during
core and mold making '
2
0
50
1.5
Percent
remaining in
mold or core
0
2
50
98.5
Percent emitted up to the time that metal is poured.
B-9
-------
TABLE B-5. FACTORS USED IN MOLD AND CORE MAKING EMISSIONS ESTIMATES;
PERCENTAGES OF CHEMICALS EMITTED ' 2
Binder system
Acrylic/Epoxy/SO2
Furan hot box
Furan nobake
Furan/SO2
Furan warm box
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake
Phenolic-Novolac flake or liquid
Phenolic urethane nobake
Phenolic urethane cold box - resin plus coreactant
Urea formaldehyde
Percent of compound emitted
Phenol
2
5
2
2
5
2
65 3
2
")
Formaldehyde
5
2
2
5
5
2
2
5
2
5
2
~>
2
Naphthalene
16<
94
Cumene
100
16"
94
Xylene
16J
94
Source: AFS, 1998, except where noted otherwise.
100 percent of all other compounds listed in table C-4 is assumed to be emitted.
Industry supplier estimate: Kenneth C. Pyles, Acme Resin Corp. Private communication to J. H. Maysilles, U.S. EPA, February 14, 2000.
4 EPA estimates.
B-10
-------
TABLE B-6. EMISSION FACTORS BASED ON TYPICAL CHEMICAL CONTENT OF SYSTEM
Binder system
Acrylic/Epoxy/SO,
Furan hot box
Furan nobake
Furan/SO,
Furan warm box
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake
Phenolic-Novolac flake
Phenolic-Novolac liquid
Phenolic urethane nobake
Phenolic urethane cold box -
resin plus coreactant
Phenolic urethane cold box -
gas
Urea formaldehyde
Pounds emitted per 100 Ib. of binder chemicals.
Phenol
0.
0.4
0.08
0.08
0.25
0.164
3.6
1.3
0.066
0.066
Formal-
dehyde
0.15
0.0014
0.022
0.02
0.05
0.01
0.01
0.1
0.0068
0.
0.025
0.0011
0.0011
0.02
Naphthalene
0.16
0.09
Cumene
5.0
0.051
0.029
Xylene
0.016
0.009
Biphenyl
0.
Methanol
10.2
1.1
10.
-
0.
10.7
0.
5.
Glycol
ethers
0.
Dimethyl
phthalate
20.2
Methyl
ethyl
ketone
0.9
Triethyl
amine
100.
,,
B-ll
-------
TABLE B-7. ESTIMATES FOR EMISSIONS BASED ON HIGH ORGANIC COMPOUND CONTENT
OF BINDER COMPONENTS, HIGH BINDER/SAND RATIOS
Binder system
Acrylic/Epoxy/SO;,
Furan hot box
Furan nobake - with methyl alcohol
Furan nobake - no methyl alcohol
Furan/SO2
Furan warm box - with methyl alcohol
Furan warm box - no methyl alcohol
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake - with methyl alcohol
Phenolic nobake - no methyl alcohol
Phenolic-Novolac flake - with methyl alcohol
Phenolic-Novolac flake - no methyl alcohol
Phenolic-Novolac liquid
Phenolic urethane nobake
Phenolic urethane cold box
Urea formaldehyde
Tons emitted per 50,000 tons of sand processed. ;
Phenol
0.65
0.65
10.5
2.75
2.25
6.0
1.55
1.55
155.
155.
78.
1.70
1.45
Formal-
dehyde
4.5
0.16
0.16
0.30
0.40
0.40
1.50
0.70
0.55
3.0
0.22
0.22
0.3
0.3
4.5
0.21
0.18
1.2
Naph-
thalene
7.0
3.25
Cumene
150.
4.5
2.15
Xylene
2.75
1.35
Methyl
alcohol
160.
0.
12.5
145.
0.
160.
0.
312.
0.
450.
Dimethyl
phthalate
190.
Methyl
ethyl
ketone
7.5
Triethyl
amine
3.0'
Total
HAP
150.
4.5
161.
0.81
210.
145.
0.40
12.0
3.45
2.80
9.0
162.
1.77
467.
155.
533.
16.2
11.4
1.2
Assumes 99 percent control of gas
B-12
-------
TABLE B-8. HIGH CHEMICAL/SAND RATIOS AND HIGH CHEMICAL CONTENT OF BINDER SYSTEMS
Binder system
Acrylic/Epoxy/SO2
Furan hot box
Furan nobake
Furan/SO2
Furan warm box
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake
Phenolic-Novolac flake
Phenolic-Novolac liquid
Phenolic urethane nobake
Phenolic urethane cold box -
resin plus coreactant
Phenolic urethane cold box -
gas
Urea formaldehyde
Pounds
chemical
per ton
of sand
60.
70.
40.
30.
42.
60.'
70.
56.
60.'
40.
120.
120.
70.
60.
12.
60.'
Chemical content of system, percent.
Phenol
3.2
14.
8.
8.
8.
7.7
8.0
4.0
4.8
4.8
Formal-
dehyde
5.
0.8
i ~i
0.75
2
2.
2
4.
1.1
0.2
3.0
0.6
0.6
4.
Naphthalene
2.4
2.4
Cumene
5.
1.6
1.6
Xylene
1.
1.
Biphenyl
Methyl
alcohol
15.7
1.65
13.8
15.7
10.4
15.
Glycol
ethers
Dimethyl
phthalate
25.
Methyl
ethyl
ketone
1.0
TEA
100.
Values are based on nominal values for other systems; not enough information is available for these systems to establish values.
B-13
-------
TABLE B-9. EMISSION FACTORS BASED ON HIGH CHEMICAL CONTENT OF SYSTEM
Binder system
Acrylic/Epoxy/SO,
Furan hot box
Furan nobake
Furan/SO,
Furan warm box
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake
Phenolic-Novolac flake
Phenolic-Novolac liquid
Phenolic urethane nobake
Phenolic urethane cold box -
resin plus coreactant
Phenolic urethane cold box -
gas
Urea formaldehyde
Pounds emitted per 100 Ib. of binder chemicals.
Phenol
0.064
0.7
0.16
0.16
0.4
0.54
5.2
2.6
0.096
0.096
Formal-
dehyde
0.25
0.016
0.044
0.038
0.1
0.04
0.04
0.2
0.022
0.01
0.15
0.012
0.012
0.08
Naphthalene
0.38
0.22
Cumene
5.0
0.26
0.14
Xylene
0.16
0.09
Biphenyl
Methyl
alcohol
15.7
1.65
13.8
15.7
10.4
15.
Glycol
ethers
Dimethyl
phthalate
25.
Methyl
ethyl
ketone
1.0
Triethyl
amine
100.
B-14
-------
TABLE B-10. EMISSION FACTORS BASED ON HIGH CHEMICAL/SAND RATIOS AND
CHEMICAL CONTENT IN SYSTEM
Binder system
Acrylic/Epoxy/SO:
Furan hot box
Furan nobake
Furan/SO:
Furan warm box
Phenolic baking
Phenolic ester nobake
Phenolic ester cold box
Phenolic hot box
Phenolic nobake
Phenolic-Novolac flake
Phenolic-Novolac liquid
Phenolic urethane nobake
Phenolic urethane cold box -
resin plus coreactant
Phenolic urethane cold box -
gas
Urea formaldehyde
Pounds emitted per ton of sand processed.
Phenol
0.026
0.42
0.11
0.090
0.24
0.062
6.2
3.12
0.067
0.058
Formal-
dehyde
0.18
0.0064
0.013
0.016
0.06
0.028
0.022
0.12
0.0088
0.012
0.18
0.0084
0.0072
0.048
Naphthalene
0.27
0.13
•
Cumene
3.0
0.18
0.086
Xylene
0.11
0.054
Biphenyl
Methyl
alcohol
'
6.3
0.50
5.8
6.3
12.5
18.
Glycol
ethers
Dimethyl
phthalate
7.5
Methyl
ethyl
ketone
0.30
Triethyl
amine
0.12'
Assumes 99 percent control.
B-15
-------
TABLE B-ll. COMPARISON OF NOMENCLATURE
FOR CHEMICAL BINDER SYSTEM
Identification used in MACT Standards
Development Information Request '
Identification used in AFS-CISA Form R
reporting guidance document2
Thermosetting systems
Shell
Hot box
Warm box
Core oil
Phenolic Novolac flake
Phenolic Novolac liquid
Furan hotbox
Phenolic hotbox
Furan warmbox
(Not identified)
Self-setting systems
Furan
Phenolic acid cured
Phenolic ester cured
Alkyd urethane
Phenolic urethane
Furan nobake
Phenolic nobake - acid catalyzed
Phenolic ester nobake
Alkyd oil
Phenolic urethane nobake
Gas-cured systems
Free radical-SO2
Epoxy-SO2
Furan-SO2
Phenolic urethane-amine
Ester cured phenolic
(Not identified)
Acrylic/Epoxy/SO2
Furan/SO2
Phenolic urethane coldbox
Phenolic ester coldbox
1 Questionnaire submitted by EPA to iron foundries in February 1998.
2 Form R Reporting of Binder Chemicals Used in Foundries, Second Edition, American Foundrymen's Society, Inc. and Casting
Industry Suppliers Association, 1998.
B-16
-------
B.2 REFERENCES
American Foundry Society (AFS), 1998. Form R Reporting of Binder Chemicals Used in
Foundries. American Foundry Society, Des Plaines, IL, and Casting Industry Suppliers
Association, Worthington, OH.
Residuals Management Technology, Inc. (RMT), April 1998. Technical and Economic
Feasibility Study for Control ofVOCsfrom Phenolic Urethane Cold Box and No Bake
Core- and Mold-Making Operations in Foundries. Prepared for the Ohio Cast Metals
Association and American Foundrymen's Society, Inc.
Stone, Jonathan A., 1999. Delta Resins and Refractories; Delta-HA. Memorandum to James H.
Maysilles, U.S. EPA, September 16,1999.
U.S. Environmental Protection Agency (EPA), 1998. Compilation of information from
questionnaire forms submitted by iron and steel foundries to the U.S. EPA Office of Air
Quality Planning and Standards. Office of Air Quality Planning and Standards,
Research Triangle Park, NC.
B-17
-------
APPENDIX C
DEVELOPMENT OF EMISSION FACTORS
FOR FOUNDRY PROCESSES
-------
C.I INTRODUCTION
This appendix details the evaluation of emissions test data and other information related
to the development of emission factors used in estimating baseline emissions and emissions
reductions effected by the regulatory alternatives.
The hierarchy used in evaluating emissions data assigned top priority to recent emissions
source test reports that directly measured HAP emissions using EPA standard sampling and
analytical methods or similar methods with the appropriate quality assurance indicators. The
next priority was given to source test data reported in the literature or provided in response to
EPA's MACT Standards Development Information Request (MSDIR) that measure only
indicators of HAP emissions (i.e., particulate matter [PM]), and to HAP emissions test data that
are more than 10 years old, or do not completely document the specific methods used and/or the
results of quality assurance test samples. The last priority was given to baghouse catch data or
other information that is an indirect assessment of emissions.
For the most part, this appendix presents average emissions data for each source test. A
complete compendium of emissions data for melting furnaces and pouring/cooling/shakeout
lines, including the results of individual runs, is given in Appendices D through G of this
document.
C.I SUMMARY OF RECENT FOUNDRY HAP EMISSION SOURCE TESTS
Four recent source tests directly measured HAP emissions from foundry operations and
have complete documentation. In April and May of 1997, the stack test team from the Casting
Emission Reduction Program (CERP) performed emission source tests at two green sand iron
foundries in Mexico. The resulting data are summarized in a single test report. In September of
1997, the EPA performed emission source tests at two green sand iron foundries in the United
States. Brief summaries of these source tests and the resulting emissions data are presented in
the following sections.
C.I.I CERP Source Tests
The CERP source tests were conducted at two automotive iron foundries operated in
Mexico. The studies primarily investigated HAP emissions from pouring, cooling, and shakeout
(PCS), although some emissions data were collected for induction furnace melting and core
making. The CERP source test reported recommended emission factors for PCS based in terms
of pounds of analyte per ton metal poured, and it primarily considered emissions from the
C-l
-------
casting of engine blocks. An independent review of the data, including emissions data for
casting bearing caps and manifolds. The emission factors developed from EPA's analysis were
essentially identical to those given in the CHRP source test report. The recommended emission
factors for PCS operations are provided in Table C-l. Note that the metal HAP emissions are
less than the principal organic HAP emissions by a factor of 25 or more.
TABLE C-l. HAP EMISSION FACTORS FOR PCS OPERATIONS; CERP STUDY
HAP Compound
Emission factor (Ib HAP per ton of metal poured) !
Pouring
Cooling
Shakeout
Total PCS
Organic HAP
Acetaldehyde
Benzene
Cresols (total)
Ethylbenzene
Formaldehyde
POM (total)
[Naphthalene]
Propanal
Styrene
Toluene
Xylenes (total)
2.94 x 1Q-4
2.19 x 10"'
1.65 x lO'6
1.01 x JO'4
1.38 x 10-4
3.56 x 1C'4
1.81 x 1Q-4
ND1
5.31 x IQ-5
1.05 x 10'3
6.12 x IQ-4
3.20 x 10°
3.49 x 1Q-2
9.27 x lO'4
1.87 x 10 "3
1.73x 10 -3
4.64x 10 -3
2.41 x 10 -3
3.71 x IQ-5
4.35 x lO'4
1.89 x ID'2
1.14x 1Q-2
5.78 x 1Q-2
2.68 x IQ-2
1.46 x 10 -2
2.81 x 10'3
2.57x IQ-2
2.21 x IQ-2
8.37 x IQ-3
5.70 x 10 -3
4.81 x IQ-3
2.21 x 1Q-2
1.78 x lO'2
6.13 x 1Q-2
6.39 x lO'2
1.54 x 1Q-2
4.88 x IQ-3
2.76 x 1Q-2
2.71 x 1Q-2 |
1.10 x 1Q-2
5.74 x 1Q-3
5.30x IQ-3
4.21 x 1Q-2
2.99 x IQ-2
Metal HAP
Cadmium
Chromium
Lead
Manganese
4.55 x ID'6
4.85 x 10'5
1.79 x IQ-4
8.37 x IQ-4
2.03 x IQ-5
2.31 x IQ-4
2.22 x lO'4
5.21 x 1Q-4
1.67 x 10'5
1.71 x 1Q-4
7.29 x 10'5
3.39x IQ-4
4.16 x IQ-5
4.51 x 1C'4 1
4.74 x 1C'4
1.70 x 10 -3
ND - Not detected.
C-2
-------
Table C-2 summarizes the emissions data for the induction melting furnace.
TABLE C-2. HAP EMISSION FACTORS FOR EIF; CERP STUDY
HAP compound
Cadmium
Chromium
Lead
Manganese
Nickel
Emission factor,
Ib/ton metal melted
0.000102
0.000074
0.00558
0.014
0.000897
The measured emissions from core making were small, generally several orders of
magnitude less than PCS emissions when accounting for the amount of core sand used per ton of
metal poured. However, the CERP report classified the capture efficiency of the exhaust system
for the core-making machines as "very poor." Additionally, emissions from core storage were
not measured, and these emissions may significantly contribute to the overall emissions of the
core-making operations, based on the total VOC emission measurements conducted in the RMT
study. Upon request from EPA, CERP provided data regarding the relative HAP composition
VOC emissions from the core storage area. These data, which are provided in Table C-3,
indicate that approximately 25 percent of the VOC emissions from core storage are HAP.
TABLE C-3. HAP CONTENT OF CORE ROOM STORAGE EXHAUST VENT
Parameter
Total VOC (ng) '
Total HAP (ng) l
% HAP
Test 1
16,309
4,644
28
Test 2
21,036
5,060
24
Test3
43,538
10,262
23
ng - nanograms.
C-3
-------
C.1.2 Waupaca-Tell City Foundry Source Test
The Waupaca foundry in Tell City, Indiana, is a completely new grey iron foundry that
started operation in February 1997. The foundry casts a diverse group of products, including
brake drums, shoes, rotors, calipers, and other parts. EPA measured emissions from the cupola
were measured by EPA during a source test in September 1997. During the time of the test, the
plant operated one large water-cooled cupola that melts at a rate of approximately 60 tons per hr
(tph), with a blast rate of 10,000 to 15,000 standard cubic feet per minute (scfm). Figure D-l is a
simplified schematic of the cupola gas handling system and emission control equipment.
The plant has four lines for pouring, cooling, and shakeout. Silica sand, bentonite, and
seacoal constitute the molding sand, which is recycled about 50 times prior to disposal in a
monofill. Resins and a catalyst are used to produce furan warm-box cores. Some of the
company's cast products use cores; others do not. During the source test, cores were not being
used on any of the lines. The properties of the molding sand measured during the test day are
given below.
Sand Property Value
Moisture (%) 3.5
Clay (%) 8.7
Loss on ignition (%, at 1800°F) 7.8
Volatile content (%, at 900°F) 4.0
A bonding agent was added to the sand in the amount of 38.1 pounds of bond per ton of sand
mulled. The bonding agent is a dry mixture of coal, brittle asphalt, cellulose, bentonite, starch,
and cereal. The material safety data sheet for the product indicates no volatile components and
no hazardous ingredients other than coal dust and crystalline quartz.
Tables C-4 and C-5 summarize the emission test results for the cupola melting furnace.
PCS emissions at Waupaca were measured only using Fourier transform infrared (FTIR)
analysis. These results were deemed to be unreliable based on a comparison with a gas
chromatography (GC) analysis performed during the EPA's second source test and therefore are
not included in this data summary.
C-4
-------
11 pOMCM* *W BCW
003
Figure C-1. Simplified Shematic of Cupola Gas Handling System at Waupaca-Tell City Foundry.
C-5
-------
TABLE C-4. SUMMARY OF CUPOLA PM AND METAL HAP EMISSIONS FROM
WAUPACA-TELL CITY FOUNDRY
Parameter
Air flow
rate
Particulate
m H ttpt*
Specific
HAP
metals
Total HAP
metals
Melt rate
Metals as
percent of
inlet PM
PM
emission
factor
Inlet (dscfm)
Outlet (dscfm)
gr/dscf, inlet
gr/dscf, outlet
Ib/hr, inlet
Ib/hr, outlet
Collection efficiency, %
Mn Ib/hr, inlet
Mn Ib/hr, outlet
Collection efficiency, %
Pb Ib/hr, inlet
Pb Ib/hr, outlet
Collection efficiency, %
Ib/hr, inlet
Ib/hr, outlet
Collection efficiency, %
tons/hr
Mn
Pb
Total metal HAP
Ib/ton metal, inlet
Ib/ton metal, outlet
Run 1
26,800
32,100
0.62
0.0026
143
0.71
99.5
3.67
0.0020
99.95
3.51
0.0096
99.73
7.36
0.027
99.64
41.4
2.57
2.45
5.15
3.45
0.017
Run 2
38,200
49,700
1.44
(')
472
(')
99.997
9.16
0.0029
99.97
8.75
0.0028
99.97
18.28
0.016
99.91
48.7
1.94
1.85
3.87
9.69
(')
Run 3
38,500
48,500
1.07
0.0011
354
0.45
99.87
7.15
0.0054
99.93
6.37
0.0056
99.91
13.82
0.019
99.87
45.7
2.02
1.80
3.90
7.75
0.0098
Average
34,500
43,400
1.04
0.0019'
323
0.58'
99.88 1
6.66
0.0034
99.95
6.21
0.0060
99.89
13.15
0.020
99.84
45.3
2.15
2.03
4.30
6.96
0.013'
' There appears to be an error in the filter weight measurement during run 2, which resulted in a negative value for
the PM filter catch and an extremely low reported PM
PM emission rate was therefore calculated using data
emission
from runs
rate (filter catch
1 and 3 only.
plus rinse catch).
The average
C-6
-------
TABLE C-5. SEMIVOLATILE HAP and PCDD/F' EMISSIONS FROM CUPOLA
BAGHOUSE OUTLET; WAUPACA-TELL CITY SOURCE TEST
Parameter
Runl
Run 2 Run 3
Average
Flow rate, dscfin
Acetophenone
bis(2-ethylhexyl)pthalate
Naphthalene
Phenol
2-4-6 Trichlorophenol
33,800
1.79
0.32
1.25
1.29
0.86
47,200 49,200
Concentration (ppb) 2
0.65 1.09
0.09 0.26
0.21 0.87
0.19 0.47
0.06 0.64
43,400
1.18
0.22
0.78
0.65
0.52
Concentration (ng/dscm, as measured)
Total D/F TEQ 3
1.07
2.60 6.36
2.52
Emission rate (Ib/hr)
Acetophenone
bis(2-ethylhexyl)pthalate
Naphthalene
Phenol
246 Trichlorophenol
Total D/F TEQ 3
Total semivolatile HAP
1.13 x 10'3
6.48 x 1Q-4
8.43 x 10 -4
6.39 x lO'4
8.88 x 1C'4
65.4
5.78 x 10 "4 1.01 x 10 -3
2.48 x 10 -4 7.62 x 10 A
2.00 x 1Q-4 8.50 x IQ-4
1.30 x 10 "4 3.42 x 10 -4
8.84 x lO'5 9.68 x lO'4
Emission rate (|o,g/hr) 4
217. 531.
9.07 x 10 -4
5.52 x IQ-4
6.31 x lO'4
3.70 x 10 -4
6.48 x 10 -4
202.
3.11 x 10-3lb/hr
1 PCDD/F Polychlorinated dibenzo-/?-dioxins/polychlorinated dibenzofurans.
2 ppb - parts per billion.
3 D/F TEQ = Dioxin/Furan Toxicity Equivalence to 2,3,7,8-TCDD.
4 |ig - micrograms.
C-7
-------
C.1.3 GM-Saginaw Metal Castings Operations (SMCO) Foundry Source Test
The GM Powertrain Group, part of the General Motors Corporation, operates a foundry
in Saginaw, Michigan, named Saginaw Metal Casting Operations (SMCO), which casts grey
iron and aluminum. This foundry was constructed by GM in 1918 and is currently operating
three cupolas ("B", "C", and "D") and two green sand lines for casting iron along with an electric
melting furnace and one casting line for aluminum to produce engine blocks for use in GM
automobiles. Cupola B was chosen by the EPA for testing, primarily because it had more
modern and complete controls and instrumentation. Cupola B has a diameter of 114 inches and
melts at a rate of about 55 tons per hour (tph) with a blast rate of 21,000 to 23,000 cfm, which
makes it among the larger cupolas in use in the United States. The blast is enriched with oxygen
at a rate of about 4 percent. Figure C-2 is a simplified schematic of the cupola gas handling
system and emission control equipment. Table C-6 summarizes the emission test results for the
cupola melting furnace.
The two iron pouring lines are labeled lines 3 and 4. Line 4 was selected for emissions
testing because it is newer and the layout is more amendable to sampling. The line has a
capacity of 270 molds per hour with two engine blocks per mold. Each horizontal mold contains
3,300 Ibs of green sand (lake sand, sea coal, and bentonite). The typical properties that are
measured and the range during the test days are given below.
Sand Property Range
Moisture (%) 2.8 to 3.3
Clay (%) 6.8 to 7.4
Comparability (%) 3.6 to 4.8
Green strength 164 to 221
Permeability 114 to 130
Loss on ignition (%) 3.8 to 5.0
During the test days, both 4- and 6-cyUnder engine blocks were poured on line 4. The
pouring weight of iron for the 4-cylinder block is 202.8 Ibs to produce a casting of 116.2 Ibs.
For the 6-cylinder block, the pouring weight is 250.4 Ibs and the casting weight is 149.2 Ibs. The
cores used in the molds include both hot-box and cold-box binder systems with phenol-
formaldehyde constituents. .
C-8
-------
Off gas
Wet cap
Recuperator
Charge
Ol ID/~, A
CUPOLA
Heated
blast air
Solids removed
Quencher
Solids
removed
Recuperator
I Blast air
Cooled gas
Scrubber
water
Venturi
scrubber
chamber
Cupola
b|0wer
Exhaust stack
Cleaned gas
I
T Scrubber water to
wastewater treatment
\
O
Exhaust fan
Figure C-2. SCHEMATIC OF THE CUPOLA GAS HANDLING SYSTEM AT GM-SMCO FOUNDRY
C-9
-------
TABLE C-6. SUMMARY OF CUPOLA
FROM GM-
PM, METAL HAP, AND PCDD/F EMISSIONS
SMCO FOUNDRY
Parameter
Particulate
matter
Specific HAP
metals
Total HAP
metals
D/F TEQ
Scrubber AP
Hot blast
temperature
Charging
rate
Metals as
percent of
inlet PM
PM emission
factor
gr/dscf, inlet
gr/dscf, outlet
Ib/hr, inlet
Ib/hr, outlet
Efficiency, %,
Mn Ib/hr, inlet
Mn Ib/hr, outlet
Pb Ib/hr, inlet
Pb Ib/hr, outlet
Total HAP Ib/hr, inlet
Total HAP Ib/hr, outlet
Efficiency for metal HAP, %
ng/dscm, outlet
u.g/hr, outlet
Inches of water column
op
tons/hr
Mn
Pb
Total metal HAP
Ib/ton iron poured, inlet
Ib/ton iron poured, outlet
Run 1
0.27
0.015
139
8.08
94.2
8.2
0.70
1.3
0.090
9.6
0.80
91.7
0.26
22.6
33
488
38.8
5.8
0.9
6.9
3.58
0.21
Run 2
0.34
0.016
169
7.61
95.5
9.2
0.52
0.84
0.076
10.1
0.60
94.1
0.36
33.2
35
464
41.7
5.4
0.5
6.0
4.05
0.18
Run 3
0.39
0.0035
186
1.71
99.1
7.6
0.10
1.7
0.030
9.4
0.14
98.5
0.11
9.8
38
366
45.4
4.1
0.9
5.0
4.10
0.038
Run 4
0.46
0.0035
215
1.71
99.2
8.1
0.070
1.8
0.015
10.0
0.09
99.1
0.10
9.6
42
364
43.5 1
3.8
0.8
4.6
4.94
0.039
Average
0.36
0.0095
177
4.8
97.3
8.3
0.35
1.4
0.053
9.8
0.41
95.8
0.21
18.8
37
421 !
42.4
4.7
0.8
5.6
4.29
0.12
For the 4-cylinder engine, 15.7 Ibs of hot-box cores and 74.5 Ibs of cold-box cores are
used for a total core weight of 90.2 Ibs per block or 180.4 Ibs per mold. For the 6-cylinder
engine, 17.9 Ibs of hot-box cores and 95.2 Ibs of cold-box cores are used for a total core weight
of 113 Ibs per block or 226 Ibs per mold. The PCS emissions were measured by EPA using both
C-10
-------
FTIR and gas chromatography/mass spectroscopy (GC/MS) techniques. Additionally, GM
employed its own GC sampling and analysis techniques to independently measure the PCS
emissions. Table D-7 summarizes the PCS emission test results from the three different
methods. Agreement between the EPA and GM measurements was not exact but nevertheless
was much better than agreement between either set of measurements and the FTIR results.
TABLE C-7. SUMMARY OF PCS EMISSION TESTING AT GM-SCMO
Compound
Toluene
Hexane
Benzene
Naphthalene
Formaldehyde
Ethyl benzene
m/p-Xy\ene
o-Xylene
Styrene
Total
Pouring emissions, Ib/hr
EPA
FTIR
ND
0.98
ND
ND
ND
ND
ND
ND
ND
0.98
EPA
GC/MS
0.043
ND
0.18
ND
ND
ND
ND
ND
ND
0.22
GM
GC
0.046
0.013
0.23
0.019
0.015
0.0031
0.014
0.0035
0.0055
0.35
Cooling emissions, Ib/hr
EPA
FTIR
22.2
12.3
ND
ND
ND
ND
ND
ND
ND
34.5
EPA
GC/MS
1.12
ND
3.52
ND
ND
0.055
0.28
0.08
0.024
5.08
GM
GC
1.43
0.11
6.16
0.47
0.15
0.092
0.84
0.13
0.063
9.45
Shakeout emissions, Ib/hr
EPA
FTIR
15.6
8.8
ND
ND
ND
ND
ND
ND
ND
24.4
EPA
GC/MS
2
ND
3.3
ND
ND
1.4
0.59
0.18
0.097
7.57
GM
GC
0.51
0.026
0.95
0.73
0.11
0.062
0.26
0.094
0.044
2.79
C.1.4 1995 Wisconsin Study
Residuals Management Technology, Inc., performed source emission testing for the
Wisconsin Cast Metals Association to characterize benzene and formaldehyde emissions from
PCS operations (RMT, 1995). Process emissions from pouring, cooling, or shakeout were tested
at eight different foundries, although none of the foundries were tested for all three processes
(i.e., PCS operations combined). Five of the foundries tested used green sand molds, two of the
foundries used no-bake molds, and one foundry used a shell process. From the data provided,
most of these foundries would be classified as iron foundries (type of metal poured was specified
for five foundries, all of which poured some type of iron). The direct emission measurement
data were reported along with projected actual emissions based on estimated capture efficiency
of each process at the given foundry. The resulting emissions data are provided in Table C-8.
C-ll
-------
TABLE C-8. HAP EMISSION FACTORS FOR PCS OPERATIONS;
WISCONSIN STUDY'
HAP
Benzene
Benzene
Benzene
Formaldehyde
Formaldehyde
Formaldehyde
Sand system
Green Sand
Nobake
Shell
Green Sand
Nobake
Shell
Emission factor, Ib HAP/ton metal poured
Pouring
5.70 x 10 "3
l.lOx 10':
Cooling
4.50 x 10°
3.20 x 10 -
1.30x 10': 2
3.20 x 10°
5.90 x lO'3
1.40x 10 '3
3.10 x 10-1
7.00 x 1Q-4 2
Shakeout
8.30 x 10 °
5.30 x JO'3
Not Meas. 2
3.90 x 10'3
8.00 x 10 '4
Not Meas. 2
Total PCS
5.90 x JO'2
4.80 x 10'2
(1.30 x 10'2) 2
8.50 x IQ-3
9.80 x 10 -3 i
(7.00 x 10'4) 2
1 Source: RMT, 1995.
2 Pouring and cooling line for the shell system was combined in single exhaust vent. No shakeout emissions were
measured for this system.
C.1.5 Summary and Comparison of Recent Source Test Results
Table C-9 provides a summary of the melting furnace emission test results for the studies
described in this section. Table C-10 summarizes the PCS emission factors for these studies.
TABLE C-9. COMPARISON OF EMISSION SOURCE TEST RESULTS FOR
CUPOLA MELTING
Parameter
Waupaca-
Tell City
GM-
Saginaw
Reported values
Cupola melting furnace
Controlled PM emission factor, Ib/ton metal
Uncontrolled PM emission factor, Ib/ton metal
Controlled total metal HAP emission factor, Ib/ton
metal
Total metal HAP cone., % of PM
Mn cone, in collected PM, % of PM
Pb cone, in collected PM, % of PM
Dioxin/Furan TEQ emission factor, ug/ton metal
0.013
7.0
0.00046
4.3
2.2
2.0
4.5
0.12
4.3
0.0097
5.6
4.7
0.8
0.44
0.09- 1.46 (WS)'
0.03-0.08 (FF)1
4.8- 66. 2-3
0.005 -0.0245
2.5 - 15. 6
1.- 10.6
1.4 -5.6
0.1 -2.74-5
'From test data provided in response to detailed ICR; see Table D-13. WS - wet scrubber; FF - fabric filter.
2 Source: Kearney, 1971.
'Source: Wallace, 1981.
4Source: EMCON, 1990. Source test at U.S. Pipe and Foundry Company, Inc., Union City, CA, 1990. Reported
in response to detailed ICR (CA-27). Metals emissions reported for one test run on 10/23/1990 were excluded
because they were an order of magnitude greater than those reported for the other two runs.
5 Source: Ecoserve, 1992.
6Source: Davis, 1975.
C-12
-------
TABLE C-10. COMPARISON OF EMISSION SOURCE TEST RESULTS FOR PCS
Source/HAP
Emission factor, Ib/ton metal poured
CERP
GM-
Saginaw '
WI Study
Green sand2
WI Study
Nobake2
Pouring \
Benzene
Toluene
Formaldehyde
Total organic HAP
0.00219
0.00105
0.000138
0.0050
0.0054
0.0011
0.000354
0.0082
0.0057
0.0032
0.011
0.0059
Cooling
Benzene
Toluene
Formaldehyde
Total organic HAP
0.0349
0.0189
0.00173
0.078
0.14
0.033
0.0035
0.22
0.045
0.0014
0.032
0.0031
Shakeout
Benzene
Toluene
Formaldehyde
Total organic HAP
0.0349
0.0189
0.00173
0.20
0.022
0.012
0.0026
0.065
0.0083
0.0039
0.0053
0.0008
PCS combined
Benzene
Toluene
Formaldehyde
Total organic HAP
0.0268
0.0221
0.0257
0.283
0.167
0.046
0.064
0.294
0.059
0.0085
0.048
0.0098
1 Values reported by GM using their GC method.
2 Source: RMT, 1995.
C-13
-------
C.2 SUMMARY OF REPORTED FOUNDRY EMISSIONS TEST DATA
This section summarizes the recent emissions test data reported in the literature or in
response to the MSDIR that is generally lacking complete documentation or that focused
primarily on PM emissions data or other surrogates for HAP emissions rather than directly
measuring HAP emissions. The primary data reported in this section include a compilation of
1990 to 1992 test data accumulated by the Wisconsin Department of Natural Resources (DNR),
the Ohio Cast Metals Association study of VOC emissions from mold/core making, and source
test data reported in response to the MSDIR.
C.2.1 Wisconsin Department of Natural Resources Compilation
This reference is simply a compilation of source test data results with no description of
the methods used and little information about the foundries. Like a subsequent 1995 Wisconsin
study, the data focuses on benzene and formaldehyde emissions from PCS operations. Results
for individual runs were provided, but no information was provided regarding the sampling and
analytical protocols, nor was information provided regarding quality assurance procedures.
Emissions from six foundries were tested; process data were available for five of these
foundries. A summary of the average emission factors calculated for each pouring line tested is
given in Table C-l 1.
C.2.2 Ohio Cast Metal Association Study
The Ohio Cast Metals Association study (RMT, 1998) tested phenolic urethane cold-box
and phenolic urethane no-bake binder systems from three different vendors. VOC emissions
were determined by weight loss measurements by each of the three vendors for each of the three
vendor products for each binder system over a 12-hour storage period. The average VOC
emissions from each binder system were reported per mass of binder added and per ton of sand
used. The average VOC emission factors for the two binder types are summarized in Table C-
12.
C-14
-------
TABLE C-ll. HAP EMISSION FACTORS FOR PCS OPERATIONS; WISCONSIN
DNR STUDY1.
HAP
Benzene
Formaldehyde
Foundry/Line
Foundry 1, 12 tph
Foundry 1, 20 tph
Foundry 1, 4 tph
Foundry 2, oil core
Foundry 2,
GS isocure
Foundry 3
Foundry 4
Foundry 5
Average for
GS foundries
Foundry 1, 12 tph
Foundry 1 , 20 tph
Foundry 2, oil core
Foundry 2,
GS isocure
Foundry 4
Foundry 5
Average for
GS foundries
Emission factor
(Ib HAP/ton metal poured)
Pouring
1.56x 10 -2
1.05 x 1Q-4
Cooling
5.24 x 10 -2
1.84x 10 "2
1.03 x 10 -2
Shakeout
5.04 x 10 -2
4.88 x 10'2
2.81 x IQ-2
7.41 x IQ-2
2.59 x 10'2
3.59 x 1C'2
4.59 x IQ-2
7.90 x 10 "3
7.50 x 10^
2.87 x 1Q-4
2.70 x 10 "2
2.67 x 10-3
1.68 x IQ-3
6.62 x IQ-3
3.35 x lO'2
5.36 x 10-3
7.18 x 1Q-5
2.80 x 10 -3
5.84 x 10 -3
5.83 x 10'3
1.83 x 10'3 4
2.18 x 10'3
2.51 x IQ-3
3.64 x lO'3
Total PCS
1.18x 10'1
6.72 x 10 "2 2
7.42 x 10 -2 3
3.59 x 10'2
5.25 x 10 -2
6.84 x 10 -2
8.78 x IQ-3
3.59 x IQ-4 3
5.84 x 10 "3
8.34 x 10 -3
7.65 x IQ-3
1 Source: WI-DNR, 1992 .
2 Includes only measurements of cooling and shakeout emissions.
3 Includes only measurements of pouring and shakeout emissions.
4 Calculated as average total PCS - average cooling - average shakeout.
TABLE C-12. VOC EMISSIONS FROM 12-HOUR STORAGE OF MOLDS
Parameter
Binder emitted, %
VOC emission factor, Ib VOC/ton sand
Phenolic urethane
cold box
3.26
0.65
Phenolic urethane
nobake
5.74
1.17
C-15
-------
C.2.3 Source Test Data Provided in Response to the MSDIR
Most of the emissions source test data submitted in response to the MSDIR pertained to
PM emission measurements. The one notable exception is the GM data, which is included in the
summary of the EPA's source test at the GM foundry (see Section C.I.3). Table C-13
summarizes the average PM emission factors calculated from the data obtained in response to the
detailed ICR. The data are organized by the emission source and control device used. When
results of individual runs were provided, Table C-13 indicates the number of individual runs that
were made and used in calculating the average. When only summary results were provided,
these data are identified as "Avg as reported" in the "Basis of reported values" column.
TABLE C-13 . SUMMARY OF PM EMISSIONS DATA SUBMITTED IN RESPONSE
TO THE MSDIR
Facility
ID
Emission source
Control device
Production
rate, tph
Basis of
reported values
Emission
rate, Ib/hr
Emission
factor, Ib/ton
Foundry
type
Cupolas controlled with wet scrubbers , '; :! i : -:r :/»:?> ;:••
181
181
181
31
77
105
107
140
143
157
202
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
AB'/WS
AB/WS
AB/WS
WS
WS
WS
WS
WS
WS
WS
WS
16.6
17.8
20.7
40.2
29.5
36.4
5.7
43
57
8
11/11/93 Avg 3 runs
6/1 3/95 Avg 3 runs
6/1 2/96 Avg 3 runs
Avg of 3 runs
Avg as reported
Avg as reported
Avg as reported
Avg as reported
Avg of 4 runs
Avg of 3 runs
Avg of 3 runs
10.9
14.7
18.3
12.32
16.53
3.27
8.31
16.20
28.66
5.72
0.654
0.829
0.894
0.306
0.240
0.560
0.090
1.460
0.265
0.451
0.715
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
^:"3V Cupolas iMi^Hetf •:}•/':>•.•. •• " ;"i>?V::;' ''^^^,^K^:''\': -'''\i W-
505
36
111
Cupola
Cupola
Cupola
FF
FF
FF
20
22
10
Avg of 3 runs
Avg of 3 runs
Avg as reported
1.63
1.69
0.30
0.082
0.077
0.030
Iron
Iron
Iron
Electric arc funraces-ffi^'confeoWi^fabrteflWSi-;:'VK"v: -o:'^%: • :" /:i:X'; • •• W:;|iT:,> -•;.•' r'- -^^^l/':'-,- - •
9
41
814
771
EAF
EAF
EAF
EAF
FF
FF
FF
FF
14.6
3.51
22.9
12.7
Avg of 3 runs
Avg of 3 runs
Avg of 3 runs
Avg of 3 runs
1.41
0.13
4.71
6.27
0.096
0.037
0.206
0.558
Steel
Steel
Steel
Steel
EAFs uncontrolled • , . ••;.''"'-•-' •'•'•..!'••.'"•'•" ,-.•"-•''.'•• • • - .'
17
EAF
FF (inlet?)
20.6
Avg of 3 runs
493
23.9
Iron
' , Electric induction furnaces (EIF) with PM control
201
52
200
200
200
EIF
EIF
EIF
EIF/Inoculation
EIF + pouring/cooling
Cyclone
FF
FF
FF
WS
4.435
4.65
3
2.75
4.38
Avg of 3 runs
Avg as reported
Avg as reported
Avg as reported
Avg as reported
2.47
0.37
0.4
0.22
2.92
0.558
0.080
0.133
0.080
0.667
Iron
Iron
Iron
Iron
Iron
.••"'I. Induction furnace* uncontrolled : ' •'" : .-.-•
201
31
EIF
EIF
Before APCD
None
4.435
6
Avg of 3 runs
Avg of 2 runs
39.7
2.62
8.940
0.437
Iron
Iron
C-16
-------
TABLE C-13 . SUMMARY OF PM EMISSIONS DATA SUBMITTED IN RESPONSE
TO THE MSDIR
Facility Emission source
ID
Control device Production
rate, tph
Basis of
reported values
Emission
rate, Ib/hr
Emission
factor, Ib/ton
Foundry
type
Miscellaneous processes
31
181
181
198
201
201
200
200
31
181
Mold cooling
Pouring/cooling
Pouring/cooling
Pouring/cooling/shakeout
Pouring/cooling/shakeout
Cooling/shakcout
Shakeout
Grinding/cutoff
Shot blast
Cleaning room
None
"Dust colllector"
"Dust colllector"
FF
None
WS
WS
FF
FF
Cartridge
4.62
17.88
20.22
2.4
9.4
13.91
4.97
1.9
20.3
17.88
Avg of 2 runs
Avg of 3 runs '96
Avg of 3 runs '98
Avg as reported
Avg of 2 runs
Avg of 2 runs
Avg as reported
Avg as reported
Avg of 3 runs
Avg of 3 runs
1.51
1.52
2.85
1.49
0.80
1.38
1.18
0.27
1.33
0.70
0.326
0.085
0.141
0.621
0.085
0.099
0.237
0.142
0.066
0.039
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
AB - Afterburner.
C.3 SUMMARY OF HISTORIC AND ADDITIONAL DATA REGARDING
FOUNDRY EMISSIONS
This section summarizes the emissions test data performed over 10 years ago and
additional data reported in response to the MSDIR regarding baghouse catch data. It includes a
summary of the Scott, Bates, and James studies, which are well documented but were performed
more than 20 years ago. Binder formulations have changed dramatically over the past 20 years,
and the binder materials used in these studies probably are not representative of the binder
formulations used by industry. Historical data on PM emissions are also summarized as well as
estimates of uncontrolled PM emission factors based on baghouse catch data reported in
response to the MSDIR.
Table C-14 provides a listing of the emissions data for pouring and cooling processes
from the studies of Scott, Bates, and James and associated bench-scale emission studies. The
primary criticism of these studies is that they were performed approximately 20 years ago and,
according to representatives of the American Foundry Society, binder formulations have
changed significantly (with reduced HAP content) since these studies were done. It is apparent
from the median and average emission factor values that there are some high emission factors
that appear to skew the average emission factors toward the high end. Consequently, the median
emission factors from these studies appear to be the most appropriate for comparison with more
recent studies.
C-17
-------
Historical data on PM emissions, primarily from melting operations, is summarized in
Table C-15. Table C-16 provides a summary of emission factors for uncontrolled melting
furnace and metal treatment operations estimated based on baghouse catch data. Table C-17
provides a similar summary of emission factors for finishing operations.
TABLE C-14. EMISSION FACTOR SUMMARY FROM BENCH SCALE STUDIES
HAP/Binder system
Benzene
Pouring/cooling - green sand
Pouring/cooling - dry sand
Pouring/cooling - silicate ester
Pouring/cooling - core oil
Pouring/cooling - alkyd isocyanate
Pouring/cooling - phenolic urethanc
Pouring/cooling - phenolic no-bake
Pouring/cooling - low N: furan-H,PO4
Pounng/cooling - med N, furan-TSA
Pouring/cooling - furan hot box
Pouring/cooling - phenolic hot box
Pouring/cooling - shell (phenolic)
Cooling - shell (phenolic)
Cooling - N; furan-TSA
Cooling - phenolic urethane
Cooling - alkyd isocyanate
Metal pour
rate, tph
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.017
0.030
0.170
0.170
0.170
Sand use, tph Emission rate. Emission
Ib/hr factor, Ib/ton
0.110
0.106
0.126
0.152
0.115
0.106
0.123
0.121
0.115
0.112
0.117
0.015
0.023
0.374
0.374
0.255
3.84E-03
3.84E-03
1.98E-03
6.48E-03
1.93E-02
1.61E-02
3.04E-02
2.12E-03
1 .48E-02
2.25E-03
4.37E-03
7.54E-03
3.05E-03
9.99E-03
6.73E-03
7.92E-03
Median emission factor:
Average emission factor:
Formaldehyde
Pounng/cooling - green sand
Pouring/cooling - dry sand
Pouring/cooling - core oil
Pouring/cooling - alkyd isocyanate
Pouring/cooling - phenolic urethanc
Pouring/cooling - phenolic no-bake
Pouring/cooling - low N-, furan-H1PO4
Pouring/cooling - med N, furan-TSA
Pouring/cooling - furan hot box
Pouring/cooling - phenolic hot box
Pounng/cooling - shell (phenolic)
Pooling - shell (phenolic)
Cooling - N, furan-TSA
Cooling - phenolic urethane
Hydrogen cyanide
Pouring/cooling - green sand
Pouring/cooling - dry sand
Pouring/cooling - silicate ester
Pouring/cooling - core oil
Pouring/cooling - alkyd isocyanate
Pouring/cooling - phenolic urethane
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.017
0.030
0.170
0.170
0.044
0.044
0.044
0.044
0.044
0.044
0.110
0.106
0.152
0.115
0.106
0.123
0.121
0.115
0.112
0.117
0.015
0.023
0.374
0.374
Median
Average
0.110
0.106
0.126
0.152
0.115
0.106
2.65E-05
6.61E-05
2.65E-04
3.84E-04
6.61E-05
2.65E-05
8.73E-04
2.12E-04
3.97E-05
2.65E-05
3.97E-05
2.44E-05
8.98E-03
1.30E-04
Emission Factor:
Emission Factor:
5.22E-03
2.65E-04
2.51E-04
2.38E-04
6.35E-04
3.17E-03
0.0870
0.0870
0.0450
0.1470
0.4380
0.3660
0.6900
0.0480
0.3360
0.0510
0.0990
0.4560
0.1020
0.0588
0.0396
0.0466
0.0930
0.1940
0.0006
0.0015
0.0060
0.0087
0.0015
0.0006
0.0198
0.0048
0.0009
0.0006
0.0024
0.0008
0.0528
0.0008
0.0015
0.0076
0. 1 1 90
0.0060
0.0057
0.0054
0.0144
0.0720
Metal type Reference'
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Steel (1025)
Steel (1025)
Steel (1025)
Steel (1025)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Steel (1025)
Steel (1025)
Steel (1025)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
A
A
A
A
A
A
A
A
A
A
A
B
B
B
A
A
A
A
A
A
C-18
-------
TABLE C-14. EMISSION FACTOR SUMMARY FROM BENCH SCALE STUDIES
HAP/Binder system
Pouring/cooling - phenolic no-bake
Pouring/cooling - low N, furan-H,P04
Pouring/cooling - med N, furan-TSA
Pouring/cooling - furan hot box
Pouring/cooling - phenolic hot box
Pouring/cooling - shell (phenolic)
Hydrogen cyanide (continued)
Cooling - shell (phenolic)
Cooling - N, furan-TSA
Cooling - phenolic urethane
Cooling - alkyd isocyanate
Cooling - alkyd isocyanate
Cooling - phenolic urethane
Cooling - shell
Cooling - phenolic no-bake
Cooling - furan N, free
Cooling - furan no-bake (med N;?)
Metal pour
rate, tph
0.044
0.044
0.044
0.044
0.044
0.017
0.030
0.170
0.170
0.170
0.170
0.170
0.170
0.170
0.170
0.170
Sand use, tph Emission rale,
Emission
Metal type Reference '
Ib/hr factor, Ib/ton
0.123
0.121
0.115
0.112
0.117
0.015
0.023
0.374
0.374
0.255
0.425
0.425
0.425
0.425
0.425
0.425
Median
7.94E-05
1.20E-03
1.98E-03
1.46E-02
5.16E-03
1.19E-02
2.30E-03
2.30E-03
9.37E-04
6.00E-04
3.27E-03
3.24E-03
1.19E-02
7.94E-05
1.06E-04
1.98E-03
emission factor:
Average emission factor:
Phenol
Pouring/cooling - green sand
Pouring/cooling - dry sand
Pouring/cooling - silicate ester
Pouring/cooling - core oil
Pouring/cooling - alkyd isocyanate
Pouring/cooling - phenolic urethane
Pounng/cooling - phenolic no-bake
Pouring/cooling - low N, furan-H,PO4
Pouring/cooling - med N, furan-TSA
Pounng/cooling - furan hot box
Pouring/cooling - phenolic hot box
Pouring/cooling - shell (phenolic)
Cooling - shell (phenolic)
Cooling - phenolic urethane
Cooling - alkyd isocyanate
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.017
0.030
0.170
0.170
0.110
0.106
0.126
0.152
0.115
0.106
0.123
0.121
0.115
0.112
0.117
0.015
0.023
0.374
0.255
Median
8.07E-04
5.03E-04
3.70E-04
1.59E-04
3.97E-04
1.18E-02
2.65E-03
5.29E-05
3.17E-04
7.94E-05
4.89E-04
1.46E-03
3.26E-03
4.08E-04
2.55E-04
emission factor:
Average emission factor:
POM (benzene soluble participates)
Cooling - N, furan-TSA
Cooling - phenolic hot-box
Cooling - phenolic no-bake
Cooling - phenolic urethane
Cooling - phenolic urethane
Cooling - alkyd isocyanate
Cooling - shell (phenolic)
Cooling - alkyd isocyanate
Cooling - silicate ester
Cooling - core oil
Cooling - dry sand
Cooling - furan hot-box
Cooling - furan med. N2
Cooling - furan N, free
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.080
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
0.120
1.68E-05
1 .65E-04
3.53E-05
3.53E-05
1.54E-05
2.36E-04
2.36E-04
2.87E-05
2.87E-05
7.83E-04
8.93E-04
3.99E-04
2.76E-04
0.0018
0.0273
0.0450
0.3300
0. 1 1 70
0.7200
0.0766
0.0135
0.0055
0.0035
0.0192
0.0191
0.0700
0.0005
0.0006
0.0117
0.0167
0.0765
0.0183
0.0114
0.0084
0.0036
0.0,090
0.2670
0.0600
0.0012
0.0072
0.0018
0.0111
0.0880
0.1090
0.0024
0.0015
0.0090
0.0400
0.0002
0.0021
0.0004
0.0004
0.0002
0.0030
0.0030
0.0004
0.0004
0.0098
0.0112
0.0050
0.1750
0.0034
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Steel (1025)
Steel (1025)
Steel (1025)
Steel (1025)
Iron
Iron
Iron
Iron
Iron
Iron
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Steel (1025)
Steel (1025)
Steel (1025)
Steel
Iron (grey)
Iron (grey)
Steel
Iron (gray)
Steel
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
A
A
A
A
A
A
B
B
B
B
C
C
C
C
C
C
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
D
E
E
D
E
D
E
E
E
E
E
E
E
E
C-19
-------
TABLE C-14. EMISSION FACTOR SUMMARY FROM BENCH SCALE STUDIES
HAP/Binder system
Cooling - green sand
Toluene
Pouring/cooling - green sand
Pouring/cooling - dry sand
Pouring/cooling - silicate ester
Pouring/cooling - core oil
Pouring/cooling - alkyd isocyanatc
Pouring/cooling - phenolic urcthane
Pouring/cooling - phenolic no-bake
Pounng/cooling - low N: furan-H,POj
Pouring/cooling - med N: furan-TSA
Pouring/cooling - furan hot box
Pouring/cooling - phenolic hot box
Pouring/cooling - shell (phenolic)
Cooling - shell (phenolic)
Cooling - N, furan-TSA
Cooling - phenolic urethane
Cooling - alkyd isocyanate
Xylenes
Pouring/cooling - green sand
Pouring/cooling - dry sand
Pouring/cooling - silicate ester
Pouring/cooling - core oil
Pouring/cooling - alkyd isocyanatc
Pouring/cooling - phenolic urethane
Pouring/cooling - phenolic no-bake
Pouring/cooling - low N; furan-H,PO4
Pouring/cooling - med N, furan-TSA
Pounng/cooling - furan hot box
Pouring/cooling - phenolic hot box
Pouring/cooling - shell (phenolic)
Cooling - shell (phenolic)
Cooling - N, furan-TSA
Cooling - phenolic urethane
Cooling - alkyd isocyanate
Metal pour Sand use, tph Emission rate.
rate, tph
0.080
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.017
0.030
0.170
0.170
0.170
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.044
0.017
0.030
0.170
0.170
0.170
Emission
Metal type Reference '
In/hr factor, Ib/ton
0.120 2.25E-03
Median emission factor:
Average emission factor:
0.110 3.97E-04
0.106 1.06E-03
0.126 3.97E-04
0.152 1.32E-03
0.115 5.56E-03
0.106 2.51E-03
0.123 1.72E-03
0.121 3.97E-04
0.115 2.88E-02
0.112 1.19E-04
0.117 7.94E-04
0.015 3.17E-03
0.023 9.52E-04
0.374 3.84E-02
0.374 2.10E-03
0.255 3.46E-03
Median emission factor:
Average emission factor:
0.110 1.32E-05
0.106 7.94E-04
0.126 2.65E-04
0.152 1.46E-03
0.115 2.30E-02
0.106 1.72E-03
0.123 3.57E-04
0.121 9.66E-03
0.115 9.26E-04
0.112 1.59E-04
0.117 6.61E-04
0.015 7.94E-04
0.023 4.02E-04
0.374 5.17E-04
0.374 3.22E-03
0.255 4.60E-03
Median emission factor:
Average emission factor:
0.0281
0.0030
0.0162
0.0090
0.0240
0.0090
0.0300
0.1260
0.0570
0.0390
0.0090
0.6540
0.0027
0.0180
0.1920
0.0317
0.2260
0.0124
0.0203
0.0270
0.0912
0.0003
0.0180
0.0060
0.0330
0.5220
0.0390
0.0081
0.2190
0.0210
0.0036
0.0150
0.0480
0.0134
0.0030
0.0189
0.0270
0.0185
0.0622
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Steel (1025)
Steel (1025)
Steel (1025)
Steel (1025)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Iron (grey)
Steel (1025)
Steel (1025)
Steel (1025)
Steel (1025)
E
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
Key to references: A - Scott, 1977; B - Scott, 1976; C - Emory, 1978; D - Scott, 1978; E - Southern Research Institute, 1979.
C-20
-------
TABLE C-15. SUMMARY OF HISTORIC DATA FOR PM EMISSION FACTORS
Reference
Metal rate, PM emission PM emission
tons/hr rate, Ib/hr factor, Ib/ton
Foundry type
Comments
Cuoola - uncontrolled
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
B
25
45
35
55
17
4
10
22
20
26
50
60
38
18
11
13
34
42.5
35
15
8
45
16
50
EAF - controlled
29.7
31.8
3.9
4.3
7
16
15
16
7.3
15
5.5
with FF
3.86
4.27
2.71
1.46
2.43
2.43
3.78
1.65
0.38
2.56
0.64
7.50
9.60
11.4
12.1
15.1
17.4
18.3
19.5
19.9
20.4
20.6
20.8
22.9
36.0
37.6
40.4
40.5
44.7
45.7
46.6
48.5
50.0
53.4
66.3
0.13
0.13
0.70
0.34
0.35
0.15
0.25
0.10
0.05
0.17
0.12
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron (grey)
Iron (grey)
Iron
Iron (grey)
Steel
Steel
Steel
Steel
Steel
Melt rate in Vol 3, Em. Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit Vl-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit Vl-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit Vl-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit Vl-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit Vl-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit Vl-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Melt rate in Vol 3, Em.Factor in Vol. 2 Exhibit VI-9
Average total catch - Facility A
Average total catch - Facility B
Average probe and filter catch - Facility C
Average total catch - Facility D
Average probe and filter catch - Facility E
Average probe and filter catch - Facility F
Average of four runs - Facility G
Average of two runs - Facility H
Average of three runs - Facility I
One 4-hour test - Facility J
Average of 3 runs - Facility K
EAF - uncontrolled
B
B
A
A
A
A
A
A
A
29.7
4.3
13
13
5
8
1.7
6.25
5
195
69.8
6.57
16.2
12.0
6.00
20.0
18.3
10.0
4.00
40.0
Iron
Iron (grey)
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Average total catch - Facility A
Average total catch - Facility D
Melt rate = charge/cycle time; Em. Fact, from Vol. 2 Exhibit VI- 16
Melt rate = charge/cycle time; Em. Fact, from Vol. 2 Exhibit VI- 16
Melt rate = charge/cycle time; Em. Fact, from Vol. 2 Exhibit VI- 16
Melt rate = charge/cycle time; Em. Fact, from Vol. 2 Exhibit VI-16
Melt rate = charge/cycle time; Em. Fact, from Vol. 2 Exhibit VI- 16
Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI- 1 6
C-21
-------
TABLE C-15. SUMMARY OF HISTORIC DATA FOR PM EMISSION FACTORS
Reference '
Metal rate
tons/hr
, PM emission PM emission
rate, Ib/hr factor, Ib/ton
Foundry type Comments
EAF - uncontrolled (continued)
A
A
A
A
A
A
A
A
A
A
A
A
C
C
C
C
C
C
C
C
C
C
C
C
C
C
1.7
1
1.5
1.5
2.6
3
6
5
3.4
2
8
11
Induction
3.9
3.6
3.6
0.65
0.38
2
0.5
4
12.7
10.7
13.4
5.30
15.3
12.8
6.10
29.4
12.7
11.0
7.50
15.0
furnace - uncontrolled
1.50
1.10
0.75
0.35
0.57
0.55
0.34
0.57
0.66
0.26
0.31
0.77
1.30
3.30
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron Melt rate = charge/cycle time; Em.Fact. from Vol. 2 Exhibit VI-16
Iron (nodular) Cold scrap
Iron (malleable) Cold scrap
Iron (malleable) Hot scrap
Iron (malleable) Includes charging, melting, superheating, pouring
Iron (malleable) Includes charging, melting, superheating, pouring
Iron (malleable) Includes charging, melting, superheating, pouring
Iron (ductile) Includes charging, melting, superheating, pouring
Iron (ductile) Includes charging, melting, superheating, pouring
Iron (ductile) 2-ton furnace
Iron (graphite) 2-ton furnace
Iron (graphite) 2-ton furnace
Iron (ductile)
Steel
Iron 30% oily borings
1 Key to references: A - Kearney, 1971; B - EPA, 1980; C - Shaw, 1982.
C-22
-------
TABLE C-16. UNCONTROLLED EMISSION FACTORS FOR MELTING FURNACES AND METAL
TREATMENT BASED ON BAGHOUSE CATCH DATA REPORTED IN MSDIR
Facility
ID
495
465
519
670
497
778
81
36
231
296
809
808
286
573
605
554
367
76
358
516
9
496
16
765
138
814
481
546
512
76
641
534
780
561
818
177
666
560
157
818
808
649
802
Source ?M collected,
Ib
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
Cupola
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EAF
EIF
EIF
EIF
EIF
EIF
EIF
EIF
EIF
Metal treatment
Metal treatment
Ductile treatment
Ductile treatment
Ductile treatment
70,740
340
548,000
200
82,000
3,000,000
5,258,000
140,000
600
34,000
800
25,000
1,000
9,000
3,600
3,600
20,000
24,000
88,000
2,760,000
506,200
209,860
1,800
726,000
23,360
1,557,740
136.5
1,080
120
216,000
20
18
24.4
4,080
370,000
100
54,600
400
360
22
2,000
2,000
5,000
Metal _, .. ..
, Collection
processed, . , ,
. period, hours
ton*?
8,690
39
44,532
14
5,668
150,000
235,455
6,000
25
1,300
28.9
820
30
268
100
80
312
7,321
26,137
630,720
87,075
29,980
239.8
86,348
2,474
136,469
10.5
60
5.5
7,321
60
20
23.2
2750
184,000
35
16,906
100
500
12
820
700
1,000
1459
4.5
2670
4
750
4000
4461
400
1
80
1
12
8
24
9
11
20
1820
4712
8760
6072
24
5816
2891
5568
8
10
1
1820
24
56
1
1000
8760
24
172
9
1
12
10
500
Emission
factor,
Ih/ton
8.14
8.72
12.31
14.29
14.47
20.00
22.33
23.33
24.00
26.15
27.68
30.49
33.33
33.58
36.00
45.00
64.10
3.28
3.37
4.38
5.81
7.00
7.51
8.41
9.44
11.41
13.00
18.00
21.82
29.50
0.33
0.90
1.05
1.48
2.01
2.86
3.23
4.00
0.72
1.83
2.44
2.86
5.00
Metal type
Iron
Gray and ductile iron
Gray iron
Iron
Gray iron
Ductile iron
Gray and ductile iron
Iron
Gray iron
Iron
Ductile iron pipe
Ductile iron pipe
Gray iron
Iron
Cast iron
Iron
Ductile iron
Steel
Mn steel
Iron
Steel
Low alloy, stainless, Mn steel.
Steel
Steel
Steel
Steel
Low carbon and stainless steel
Gray and ductile iron
Iron
Steel
Gray and ductile iron
Steel and stainless steel
Gray and ductile iron
Iron and steel
Ductile iron
Gray iron
Gray and ductile iron
Steel and stainless steel
Ductile iron
Ductile iron
Ductile iron
Ductile iron
Gray and ductile iron
C-23
-------
TABLE C-17. SUMMARY OF EMISSION FACTORS FOR UNCONTROLLED FINISHING OPERATIONS BASED
ON BAGHOUSE CATCH DATA REPORTED IN MSDIR
Facility
ID
802
453
433
433
358
406
609
462
149
2
433
433
542
332
16
516
262
268
466
519
465
587
512
609
95
385
358
358
519
230
72
16
384
389
599
225
207
159
619
619
619
127
Operation(s), as reported
Journal grinder
Shot blast, finishing
Arc wash riser contacts
Riser removal
Scrap preparation
Electric grinding
Grinding, grit blasting
Grinding
Dry grinding and welding
Shot blasting
Shot blasting
Shot blasting
Shot blast
Casting cleaning
No. 5 shot blast
Mechanical finishing
Shotblast
Shot blast
Cleaning
Shot blasting
Shotblasting
Blast Cleaning
Iron grinding
Grit blasting
Casting Cleaning
Cutoff grinding
Cutoff
Cutoff
Grinding
Grinding
Shotblasting
No. 7 shot blast
Shotblast machines
Grinding
Finishing
Blasting and Grinding
Shotblast
Cutoff
Jet blast
Sand blasting
Sand blasting
Shot blasting
PM
collected,
Ib
2,000
0.3
500
500
11,400
25
2,500
1,500
12,660
8
500
2,000
400
100
3,000
469,100
250
20
10
112,000
100
16,200
200
1,000
20,000
2,000
30,000
30,000
170,000
200
94,000
24,000
76
14,000
480
6,000
240
2,000
44,000
44,000
44,000
60
Metal
processed,
tons
9,500
1.25
1,200
1,200
26,000
50
4,205
1476
11,250
6
320
1,200
205
50
1400
1,881
100
8
4
44,532
39
6,000
70
350
6,885
600
8,500
8500
44,532
45
21,000
5,152
16.3
3,000
100
1,215
48
394
8,191
8,191
8,191
11
Collection
period,
hours
4.2
1
320
320
4,712
8
2,000
2,838
4,500
8
160
320
2,000
80
24
8,760
150
8
8
2,670
8
1,774
24
450
1,997
400
5,814
5,814
2,670
40
5,000
24
16
6,000
16
150
24
2,080
3,840
3,840
3,840
8
Emission
factor,
Ib/ton
0.21
0.24
0.42
0.42
0.44
0.50
0.59
1.02
1.13
1.33
1.56
1.67
1.95
2.00
2.14
2.49
2.50
2.50
2.50
2.52
2.56
2.70
2.86
2.86
2.90
3.33
3.53
3.53
3.82
4.44
4.48
4.66
4.66
4.67
4.80
4.94
5.00
5.08
5.37
5.37
5.37
5.45
Metal type
Iron, steel
Stainless steel
Steel
Steel
Mn steel
Gray Iron Castings
Iron
Ni-hard iron
Carbon, low alloy steel
Carbon steel
Steel
Steel
Ductile & Gray Iron
Iron and steel
Steel
Iron
Iron
Iron
Cast iron & ductile iron
Gray Iron Castings
Gray iron and ductile iron
Iron and steel
Iron
Iron
Steel
Iron and steel
Mn steel
Mn steel
Gray Iron Castings
Unspecified
Iron
Steel
Steel, iron
Steel
Steel
Grey, ductile, & malleable iron
Gray and ductile iron
Steel & stainless steel investment
castings.
Steel
Steel
Steel
Iron
C-24
-------
TABLE C-17. SUMMARY OF EMISSION FACTORS FOR UNCONTROLLED FINISHING OPERATIONS BASED
ON BAGHOUSE CATCH DATA REPORTED IN MSDIR
Facility
ID
227
349
14
581
780
358
608
541
147
323
138
276
72
358
358
519
567
433
112
376
537
762
810
286
384
818
818
811
818
214
255
255
184
529
19
413
413
760
16
159
516
811
811
811
818
140
Operation(s), as reported
Cutoff
Electric Air Arc
Grinding
Air arc cutoff
Grinding/shotblast
Cutoff
Shot blasting
Shot blast
Shotblast Castings
Blast Cleaning
Cutoff, grinding etc.
Cleaning
Shotblasting
Cutoff
Cutoff
Shot blasting
Shotblast
Sand blast castings
Shot blasting, grinding
Shotblast (steel shot)
Cut-off, shot blast, grinding
Cleaning & finishing
Shotblast
Finishing
Cutoff, grinding, chopping
Grinding
Grinding
Grinding
Grinding
Grinding
Table blast
Shot blast, grinding
Blasting - shot
Grinding
Grit blasting
Shotblast
Cutoff grinding
Shot blast
No. 6 shot blast
Grinding
Mechanical finishing
Grinding
Grinding
Grinding
Grinding
Grinding
PM
collected,
Ib
158
2,000
8,000
50
62.6
40,000
22,500
500
40
40
40,000
100
176,000
134,000
140,200
410,000
3,600
6,000
3,000
10,000
500
1,000
2,000
220
10
1,216,000
1,438,000
38
13
80
800
1,000
82
100
286,000
3,360
3,360
100
6,000
4,000
2,996,000
25
25
25
50
52,000
Metal
processed,
tons
28.9
340
1,300
8.1
10.1
6,000
3,250
72
5
5
4,948
12
21,000
15,586
15,586
44,532
388.8
640
300
1,000
50
100
192
20
0.85
96,360
113,880
3
1
6
60
75
6.03
7
18,739
220
220
6.5
384
256
187,753
1.5
1.5
1.5
3
3,098
Collection
period,
hours
64
960
2,000
160
1
5,814
4,500
8
1
1
2,891
8
5,000
5800
4,845
2,670
1,920
320
20
400
24
20
16
8
16
8,760
8,760
1
1
16
80
120
24
120
4,000
20
20
40
24
2,080
8,760
1
1
1
1
2,080
Emission
factor,
Ib/ton
5.47
5.88
6.15
6.17
6.20
6.67
6.92
6.94
8.00
8.00
8.08
8.33
8.38
8.60
9.00
9.21
9.26
9.38
10.0
10.0
10.0
10.0
10.4
11.0
11.8
12.6
12.6
12.7
13.0
13.3
13.3
13.3
13.6
14.3
15.3
15.3
15.3
15.4
15.6
15.6
16.0
16.7
16.7
16.7
16.7
16.8
Metal type
Steel
Iron and steel
Iron
Stainless steel and superalloy
Gray iron and ductile iron
Mn steel
Grey, ductile and ni-resist iron
Ductile & Gray Iron
Iron
Iron and Steel
Steel
Iron
Iron
Mn steel
Mn steel
Gray Iron Castings
Mild and low alloy steel
Steel
Gray Iron
Steel
Steel
Gray & ductile iron
Ductile iron
Gray iron
Steel, iron
Ductile iron
Ductile iron
Ductile iron
Ductile iron
Steel investment castings
Steel
Steel
Ductile, gray, steel, niresist
Steel
Steel
Ductile and Gray Iron
Ductile and Gray Iron
Steel
Steel
Steel & stainless steel investment castings.
Iron
Ductile iron
Ductile iron
Ductile iron
Ductile iron
Gray iron castings
C-25
-------
TABLE C-17. SUMMARY OF EMISSION FACTORS FOR UNCONTROLLED FINISHING OPERATIONS BASED
ON BAGHOUSE CATCH DATA REPORTED IN MSDIR
Facility
ID
388
818
818
811
811
811
811
388
255
768
86
86
166
481
522
537
619
365
216
72
76
780
812
385
363
282
493
493
611
23
500
306
599
365
14
774
808
363
363
760
492
818
213
818
Operations), as reported
Shot blasting
Shot blasting
Rotary drum cleaning
Shot blasting
Grinding
Shot blasting
Shot blasting
Grinding
Shotblast & grind
Blast cleaning/grinding
Shot blast, chip & spool
Shot blast, chip & spool
Casting removal
Shot blasting
Shot blast, grinding
Sand blasting
Cut-off
Grinding
Manual & auto grinding
Grinding
Shot blast
Shotblast
Shot blast
Grinding
Cutoff saw
Cleaning, grinding
Cleaning
Cleaning
Finishing
Shot blast
Mechanical Finishing
Grinding/shotblasting
Finishing
Tumblast
Shot blasting
Cleaning & Finishing
Cutoff, grinder
Grinding
Shot blast
Grinding
Cleaning
Grinding, shot blast
Blasting - shot
Shot blast
PM
collected,
Ib
280
1,064,000
1,290,000
85
51
43
43
240
1,200
12,000
760
760
3,400
200
300
200
70,000
72,200
16,000
460,000
4,080
79.8
1,500
46,420
12
3,040
250
250
500
15,200
142,000
6,000
260
90,200
800
4,000
2,000
500
500
100
4,000
1,266,000
13,960
70
Metal
processed,
tons
16.68
63,072
76,212
5
3
2.5
2.5
13.9
65
630
39
39
172
10
15
10
3,454
3,444
750
21,000
181
3.46
64.4
1,950
0.5
123.2
10
10
20
599
5,545
234
10
3,444
30
144
72
17.5
17.5
3.25
127
39,420
425
2.1
Collection
period,
hours
8
8,760
8,760
1
1
1
1
8
120
24
24
24
4,000
9
10
24
3,840
1,353
4,000
5,000
120
1
18
5,856
40
40
14
7
40
2,652
5,297
18
16
1,722
30
24
12
40
40
80
504
8,760
434
1
Emission
factor,
Ib/ton
16.8
16.9
16.9
17.0
17.0
17.2
17.2
17.3
18.5
19.0
19.5
19.5
19.8
20.0
20.0
20.0
20.3
21.0
21.3
21.9
22.5
23.1
23.3
23.8
24.0
24.7
25.0
25.0
25.0
25.4
25.6
25.6
26.0
26.2
26.7
27.8
27.8
28.6
28.6
30.8
31.5
32.1
32.8
33.3
Metal type
Mn steel, stainless steel, carbon steel, and
nickel alloy (heat resistant)
Ductile iron
Ductile iron
Ductile iron
Ductile iron
Ductile iron
Ductile iron
Mn steel, stainless steel, carbon steel, and
nickel alloy (heat resistant)
Steel
Iron
Gray iron
Gray iron
Steel
Low carbon and stainless steel
Ductile iron
Steel
Steel
Iron
Iron and steel
Iron
Steel
Gray iron and ductile iron
Ductile iron
Iron and steel
Ductile and malleable iron
Gray iron
Stainless steel centrifugal castings.
Stainless steel centrifugal castings.
Gray & Ductile Iron
Ductile iron, stainless steel, carbon steel,
bronze, nickel alloys
Iron castings
Iron
Steel
Iron
Iron
Gray and Ductile Iron
Ductile iron pipe
Ductile and malleable iron
Ductile and malleable iron
Steel
Stainless and low alloy carbon steels.
Ductile iron
Iron
Ductile iron
C-26
-------
TABLE C-17. SUMMARY OF EMISSION FACTORS FOR UNCONTROLLED FINISHING OPERATIONS BASED
ON BAGHOUSE CATCH DATA REPORTED IN MSDIR
Facility
ID
93
93
557
818
100
516
533
166
140
533
533
533
282
666
636
257
645
541
262
512
471
581
1
169
294
581
762
7
60
523
669
23
232
169
282
110
309
682
666
214
666
486
570
216
1
270
Operation(s), as reported
Grinding
Shot blasting
Mechanical finishing cutoff
Shot blast
Shotblast
Mechanical finishing
High speed grinding
Gate grinding/finishing
Shot blasting
Pull through shot blast
Tumblast, grinding
Tumblast
Cleaning, grinding
Shotblasting
Cut-off, shot blast
Shotblast
Shot blasting
Grinding
Grinding
Rotoblast Shotblasting
Abrasive cleaning, blasting
Abrasive cutoff
Cleaning/finishing
Casting finishing
Finishing
Gate grinding
Cut-off, shot blast
Grinding/cleaning/deburing
Cleaning and finishing
Finishing
Grinding, shot blast
Cut-off, grinding
2013 DISA Shotblast
Casting cleaning
Cleaning
Shot blasting
Tumblast
Shot blast
Shotblast
Sand blast
Shotblast
Shot blast
Clean castings
Casting removal
Cleaning/finishing
Grinding
PM
collected,
Ib
100
200
200
590,000
6,000
3,064,000
300
13,000
122,000
200
320
320
3040
153,000
3,000
1,800
6,000
500
13,500
3,200
220
2,200
1,500
5,000
200
100
1,000
544
1,500
1,000
1,000
33,400
10,000
6,000
4,040
4,000
1,000
3,000
224,000
400
347,220
3,400
80,000
53,000
3000
480,000
Metal
processed,
tons
3
6
6
17,520
168
84,998
8
334
3,098
5
8
8
74
3,711
72
43.1
140
11.5
300
70
4.8
44.1
30
100
4
2
20
10
27.5
18
18
599
175
100
65.6
64
16
48
3,549
6
5,149
50
1,134
750
42
6,600
Collection
period,
hours
8
5
24
8,760
12
8,760
10
4,000
2,080
10
10
10
40
1
8
80
80
8
150
24
8
160
40
24
175
40
10
40
8
12
24
2,652
22
24
40
40
16
80
16
0
10
4,032
4,000
40
1,200
Emission
factor,
Ib/ton
33.3
33.3
33.3
33.7
35.7
36.0
37.5
38.9
39.4
40.0
40.0
40.0
41.1
41.2
41.7
41.8
42.9
43.5
45.0
45.7
45.8
49.9
50.0
50.0
50.0
50.0
50.0
54.4
54.5
55.6
55.6
55.7
57.1
60.0
61.6
62.5
62.5
62.5
63.1
66.7
67.4
68.0
70.5
70.7
71.4
72.7
Metal type
Iron
Iron
Iron
Ductile iron
Gray iron
Iron
Steel
Steel
Gray iron castings
Steel
Steel
Steel
Gray iron
Gray & Ductile Iron
Iron and steel
Gray and ductile iron
Iron
Ductile & Gray Iron
Iron
Iron
Steel
Stainless steel and superalloy
Gray and ductile iron
Ductile Iron
Steel
Stainless steel and superalloy
Gray & ductile iron
Iron
Gray and ductile iron, alloy steel
Grey and ductile iron
Iron, Steel, Alloy steel, Stainless steel
ductile iron, stainless steel, carbon steel,
bronze, nickel alloys
Ductile iron
Ductile Iron
Gray iron
Gray and ductile iron
Iron and steel
Iron
Gray & Ductile Iron
Steel investment castings
Gray & Ductile Iron
Iron
Iron
Iron and steel
Gray and ductile iron
Gray iron
C-27
-------
TABLE C-17. SUMMARY OF EMISSION FACTORS FOR UNCONTROLLED FINISHING OPERATIONS BASED
ON BAGHOUSE CATCH DATA REPORTED IN MSDIR
Facility
ID
507
552
86
516
389
33
534
560
499
581
231
525
172
508
782
44
194
244
782
63
60
184
216
241
37
60
610
37
166
581
110
282
782
123
782
214
16
Operation(s), as reported
Cutoff, Grind, Shotblast
Shotblast
Shotblast, stand grind, chip &
spool
Cleaning room
Shot blast
Cleaning/grinding
Shot blasting
Cut off saw and stand
grinding
Mechanical Finishing
Grit blasting
Hydro shotblast
Shot blast, grinding
Blasting
Finishing
Casting cleaning
Cutoff, blasting, grinding
Hand grinding
Surface Grinders
Casting cleaning
Cutoff, shot blast, grinding
Heat treating, cleaning
Shotblast
Casting cleaning
Tumbleblast
Grinding and finishing
Casting cleaning
Grit blasting
Casting cleaning
Casting cleaning
Sand/grit blasting
Grinding
Cleaning
Casting cleaning
Heat treating, grinding
Casting cleaning
Sand blasting
Tumble blast
PM
collected,
Ib
2,000
6,000
4,000
16,150,000
1,060,000
620
374
1,320
56000
100
10,000
4,000
3,000
1,000
2,000
22,800
54
6,000
3,000
3,200
1,000
400
100,000
180
1,000
1,500
2,000
1,500
76,000
1,520
8,000
5,060
5,000
20,000
3,000
3,120
6,000
Metal
processed,
tons
27
80
52
207,292
12,884
7.5
4.46
14
588
1.02
100
40
28
9
18
198
0.47
50
25
25
7.5
3
750
1.3
6.6
8
10
7
334
6.1
32
19.6
15
53
6
6
8
Collection
period,
hours
40
10
24
8,760
6,000
40
40
30
1,920
40
20
90
8
1,000
8
3,810
40
160
8
24
8
24
4,000
8
8
8
100
8
4,000
80
40
40
8
40
8
16
24
Emission
factor,
Ib/ton
74.1
75.0
76.9
77.9
82.3
82.7
83.9
94.3
95.2
98.0
100
100
107
111
111
115
115
120
120
128
133
133
133
138
152
188
200
214
228
249
250
258
333
377
500
520
750
Metal type
Steel
Steel
Gray iron
Iron
Steel
Gray Iron
Steel
Steel & stainless steel
Stainless steel and steel castings
Stainless steel and superalloy
Grey iron
Gray and ductile iron
White iron, gray iron, Ni hard & ductile/iror
Iron
Gray iron
Steel
Steel-cobalt-inconel
Ductile iron, carbon & low alloy steel,
stainless & high alloy steel
Gray iron
Gray & ductile castings
Gray and ductile iron, alloy steel
Ductile, gray, steel, niresist
Iron and steel
Iron
Gray iron
Gray and ductile iron, alloy steel
Steel
Gray iron
Steel
Stainless steel and superalloy
Gray and ductile iron
Gray iron
Gray iron
Steel and stainless steel
Gray iron
Steel investment castings
Steel
C-28
-------
C.4 REFERENCES
Casting Emission Reduction Program (CERP). Foundry Process Emission Factors: Baseline
Emissions from Automotive Foundries in Mexico. November 1998.
Davis, J. A., E. E. Fletcher, R. L. Wenk, and A. R. Elsea. Screening Study on Cupolas and
Electric Furnace in Gray Iron Foundries. Final Report. Battelle - Columbus
Laboratories. August 15, 1975.
Ecoserve, Inc. Determination ofAB 2588 Emissions From the Gray Iron Foundry Cupola
Baghouse at American Brass and Iron, Oakland, CA. (November 1991). April 1, 1992.
EMCON Associates. Compliance Testing to Quantify Emissions at U. S. Pipe and Foundry
Company, Union City, CA. (October 1990.) December 1990.
Emory, M.B., P.A. Goodman, R.H. James, and W.D. Scott. July 1978. Nitrogen-Containing
Compound Emissions. American Industrial Hygiene Association Journal, pp. 527 533.
A. T. Kearney and Company. System Analysis of Emissions and Emissions Control in the Iron
Foundry Industry. Three volumes. NTIS publication PB 198 348, PB 198 349, and PB
198350. February 1971.
Residuals Management Technology, Inc. (RMT), June 1995. Wisconsin Cast Metal Association
Group Source Emission Testing. Test Report Prepared for Wisconsin Cast Metal
Association.
Residuals Management Technology, Inc., 1998. Technical and Economic Feasibility Study for
Control ofVOCs from Phenolic Urethane Cold Box and No Bake Core- and Mold-
Making Operations in Foundries. Prepared for the Ohio Cast Metals Association and
American Foundrymen's Society, Inc. April 1998.
Southern Research Institute. Binder Decomposition During Pouring and Solidification of
Foundry Castings, Part II: Paniculate Emissions from Foundry Molds. AFS
International Cast Metals Journal, pp. 14-15. June 1979.
Scott, W. D., and C. E. Bates, 1976. Binder Decomposition During Solidification and Cooling of
Steel Castings. A Report to the Steel Founders' Society of America. Southern Research
Institute. October 1976.
Scott, W. D., C. E. Bates, and R. H. James, 1977. Chemical Emissions From Foundry Molds.
AFS (American Foundry Society) Transactions; 77-98. Southern Research Institute.
Scott, W. D., C. E. Bates, and R. H. James, 1978. Polynuclear Aromatic Hydrocarbons in Mold
Decomposition Effluent. Report to the Steel Founders' Society of America. Southern
Research Institute. March 1978.
Shaw, F. M. Induction Furnace Emissions. International Cast Metals Journal. British Cast Iron
Research Association. June 1982.
C-29
-------
U.S. Environmental Protection Agency (EPA), 1980. Electric Arc Furnaces in Ferrous
Foundries - Background Information for Proposed Standards. Office of Air Quality
Planning and Standards, Research Triangle Park, NC. EPA-450/3-80-020a. May 1980.
U.S. Environmental Protection Agency, 1999a. Iron and Steel Foundries Manual Emissions
Testing of Cupola Baghouse at Waupaca Foundry in Tell City, Indiana. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. EPA-454/R-99-017A and
EPA-454/R-99-017B. June 1999.
U.S. Environmental Protection Agency, 1999b. Iron and Steel Foundries Manual Emissions
Testing of Cupola Wet Scrubber at General Motors Corp., Saginaw, Michigan. Office of
Air Quality Planning and Standards, Research Triangle Park, NC. EPA-454/R-99-025A
and EPA-454/R-99-025B. July 1999.
Wallace, D., P. Quarles, P. Kielty, and A. Trenholm. January 1981. Summary of Factors
Affecting Compliance by Ferrous Foundries. EPA-340/1 -80-020. Midwest Research
Institute.
Wisconsin Department of Natural Resources (WI-DNR). June 1992. Hazardous Organic
Emissions from Iron Foundry Operations. Letter and Attachments from Susan Lindem,
(WI-DNR) to G. Gschwandtner, E.H. Pechan and Associates.
C-30
-------
APPENDIX D
SOURCE TEST PARTICIPATE MATTER DATA
FOR CUPOLA BAGHOUSES
-------
D.I SOURCE TEST PARTICIPATE MATTER DATA FOR CUPOLA BAGHOUSES
This appendix presents the individual sampling run data for the source tests available to
characterize the control performance for baghouses applied to cupolas (Chapter 4). Summary
test data are given in Table D-l along with information on melting rates and capacities and a
description of the control systems and the processes they serve.
The data in Table D-l represent a range of cupola sizes and types of baghouses. The
design melting rates range from 3.5 to 80 tons per hour, and ventilation rates range from 30,000
to 195,000 actual cubic feet per minute. The cupolas include both recuperative and non-
recuperative, and both above and below charge take off. The baghouses include both negative
and positive pressure operating modes and employ both shaker and pulse jet cleaning systems.
Some were installed about 30 years ago, and some are relatively new (rebuilt). The design air-
to-cloth ratios cover a range of 1.68 to 5.1 feet per minute. No information is available on the
ages of the bags in service when the tests were conducted.
The reported results were checked to ensure the weights of PM from the filter and the
probe catch were above detection limits. When the reported catch was less than 3 rag, a
detection limit value of 3 mg and the sample volume were used to estimate the detection limit in
gr/dscf. Values calculated in this manner are reported as "less than" (<).
D-l
-------
TABLE D-l. PM SOURCE TEST RESULTS FOR BAGHOUSES SERVING CUPOLAS
Foundry WI-35 (tested March 1998)
Run
1
2
3
Avg
PM
(gr/dscf)
O.0006
<0.0006
O.0006
<0.0006
PM
(Ib/hr)
O.4
O.4
O.4
O.4
Flow
(dscfm)
75,974
75,412
74,847
75,411
Flow
(acfm)
107,297
107,145
105,854
106,765
Temp
(°F)
271
273
274
273
Air: cloth ratio
(ft/min)
1.7
1.7
1.7
1.7
Melt rate
(tph)
Cupola information Baghouse information
45 tph capacity, Installed 1998, negative
afterburner, pressure, pulse jet, horizontally-
recuperative, above supported bags, 10.8 oz Nomex
charge takeoff fabric, aircloth = 2.4 ft/min,
design for 280°F and 148,000
acfm
Foundry WI-35 (tested November 1998)
Run
1
2
3
Avg
PM
(gr/dscf)
<0.0007
<0.0008
0.0008
0.0008
PM
(Ib/hr)
O.4
O.4
O.4
O.4
Flow
(dscfm)
59,651
56,350
57,002
57,668
Flow
(acfm)
86,905
81,221
82,220
83,449
Temp
(°F)
279
270
271
273
Aircloth ratio
(ft/min)
1.4
1.3
1.3
1.3
Melt rate
(tph)
40
40
42.5
41
Foundry WI-35 (tested May 2000)
Run
1
2
3
Avg
PM
(gr/dscf)
O.0007
O.0007
O.0007
O.0007
PM
(Ib/hr)
O.4
O.4
O.4
O.4
Flow
(dscfm)
61,074
60,856
61,132
61,021
Flow
(acfm)
88,945
88,346
88,483
88,591
Temp
(°F)
271
269
267
269
Aircloth ratio
(ft/min)
1.4
1.4
1.4
1.4
Melt rate
(tph)
D-2
-------
Foundry IN-01 (tested March 2000)
Run
1
2
3
Avg
PM
(gr/dscf)
0.00086
0.00079
0.00069
0.00078
PM
(Ib/hr)
0.43
0.42
0.39
0.41
Flow
(dscftn)
58,178
61,481
65,454
61,704
Flow
(acfm)
81,782
87,303
95,494
88,193
Temp
(°F)
259
270
293
274
Air: cloth ratio
(ft/min)
Melt rate
(tph)
69.5
61.8
68.6
66.6
Cupola information
75 tph capacity,
afterburner, below
charge takeoff
Baghouse information
New baghouse, pulse jet,
horizontally-supported bags
Foundry MI-26 (tested December 1995)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0012
0.0023
0.0017
0.0017
PM
(Ib/hr)
0.22
0.40
0.29
0.30
Flow
(dscfrn)
20,987
20,987
21,029
21,001
Flow
(acfrn)
Temp
(°F)
Aircloth ratio
(ft/min)
Melt rate
(tph)
10
Cupola information
15 tph capacity,
afterburner, above
charge takeoff
Baghouse information
Installed 1995, positive pressure,
shaker, fiberglass fabric,
aircloth = 0.75 ft/min, design
for 500°F and 25,700 acfm
Foundry NC-05 (tested February 2000)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0019
0.0027
0.0019
0.0022
PM
(Ib/hr)
1.15
1.69
1.14
1.33
Flow
(dscfrn)
65,932
64,883
64,879
65,231
Flow
(acfrn)
102,298
105,026
102,995
103,440
Temp
(°F)
288
292
296
292
Air: cloth ratio
(ft/min)
2.3
2.3
2.3
2.3
Melt rate
(tph)
62.9
59.8
65.3
62.7
Cupola information
70 tph capacity,
afterburner, above
charge takeoff
Baghouse information
New baghouse, negative
pressure, pulse jet, air:cloth =
1.76 ft/min, design for 350°F
and 79,000 acfm
D-3
-------
Foundry NJ-3 (tested August 1991)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0048
0.0055
0.0026
0.0043
PM
(Ib/hr)
12.7
11.2
3.5
8.9
Flow
(dscfm)
306,488
238,254
159,297
234,680
Flow
(acfm)
390,656
305,489
211,491
304,017
Temp
(°F)
213
217
241
224
Aircloth ratio
(ft/min)
3.5
2.7
1.9
2.7
Melt rate
(tph)
87
67
88
81
Cupola information
2 cupolas with 64 tph
capacity (only one
operates at a time),
afterburner,
recuperative, below
charge takeoff
Baghouse information
Installed 1974, positive pressure,
shaker, fiberglass fabric,
aircloth = 1 .75 ft/min, design
for 500°F and 195,000 acfm,
controls melting
Foundry NJ-3 (tested September 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0012
0.0023
0.0014
0.0016
PM
(Ib/hr)
3.06
1.89
2.99
2.6
Flow
(dscfm)
219,000
220,100
240,200
226,433
Flow
(acfm)
263,000
282,000
316,000
287,000
Temp
(°F)
175
216
235
209
Air: cloth ratio
(ft/min)
2.4
1.9
2.8
2
Melt rate
(tph)
80
90
75
82
Foundry IN-34 (tested September 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0026
O.0003
0.0011
O.0013
PM
(Ib/hr)
0.71
O.14
0.46
<0.5
Flow
(dscfm)
32,100
49,700
48,500
40,300
Flow
(acfm)
45,000
69,600
68,200
56,600
Temp
(°F)
231
253
254
243
Aircloth ratio
(ft/min)
1.2
1.8
1.8
1.5
Melt rate
(tph)
53
41
47
50
Cupola information
80 tph capacity,
afterburner,
recuperative, below
charge takeoff
Baghouse information
Installed 1997, negative
pressure, pulse jet, Nomex,
air: cloth =1.8 ft/min, design for
320°F and 70,000 acfm, controls
melting and charging
D-4
-------
Foundry VA-8 (tested January 1998)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0039
0.0028
0.0026
0.0031
PM
(Ib/hr)
1.64
1.14
1.08
1.29
Flow
(dscfrn)
48,697
47,588
48,934
48,407
Flow
(acfrn)
70,363
69,934
72,472
70,923
Temp
(°F)
278
281
283
281
Airxloth ratio
(ft/min)
2.6
2.6
2.7
2.6
Melt rate
(tph)
49
51
53
51
Cupola information
2 cupolas with 65 tph
capacity (only one
operates at a time),
afterburner,
recuperative, below
charge takeoff
Baghouse information
Installed 1997, negative
pressure, pulse jet, Nomex,
air: cloth = 3.74 ft/min, design
for 375°F and 100,000 acfm,
controls melting and charging
Foundry FL-6 (tested February 1998)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0028
0.0031
0.0051
0.0037
PM
(Ib/hr)
0.52
0.67
1.11
0.77
Flow
(dscfm)
21,976
25,178
25,288
24,147
Flow
(acfrn)
35,420
42,114
41,495
39,676
Temp
(°F)
246
266
272
261
Air: cloth ratio
(ft/min)
0.9
0.7
0.7
0.8
Melt rate
(tph)
17.7
19.8
25.1
20.9
Cupola information
22 tph capacity,
afterburner,
recuperative, above
charge takeoff
Baghouse information
Installed 1998, negative
pressure, reverse air, fiberglass
fabric, airxloth = 1.68 ft/min,
design for 460°F and 65,000
acfm, controls melting and
charging
Foundry IA-19 (tested February 1998)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0026
0.0015
0.0022
0.0021
PM
(Ib/hr)
0.92
0.58
0.90
0.80
Flow
(dscfrn)
41,861
46,281
46,811
44,984
Flow
(acfm)
58,271
63,363
64,433
62,022
Temp
(°F)
245
233
238
239
Air: cloth ratio
(ft/min)
4.2
4.6
4.7
4.5
Melt rate
(tph)
13.5
13.5
13.5
13.5
Cupola information
20 tph capacity,
afterburner,
recuperative, below
charge takeoff
Baghouse information
Installed 1992, negative
pressure, pulse jet, Nomex felt
fabric, air: cloth = 5.1 ft/min,
design for 450°F and 70,000
acfm, controls melting,
charging, tapping
D-5
-------
Foundry IN-35 (tested November 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0044
0.0043
0.0043
0.0043
PM
(Ib/hr)
1.71
1.68
1.66
1.69
Flow
(dscfm)
45,055
44,780
44,773
44,869
Flow
(acfm)
66,407
66,018
66,532
66,319
Temp
(°F)
213
215
212
213
Aircloth ratio
(ft/min)
4.1
4.1
4.1
4.1
Melt rate
(tph)
Cupola information
22 tph capacity,
afterburner,
nonrecuperative,
above charge takeoff
Baghouse information
Installed 1997, positive pressure,
pulse jet, Tuflex fabric, aincloth
= 4.65 ft/min, design for 400°F
and 75,000 acfm, controls
melting
Foundry SD-1 (tested March 1995)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0058
0.0035
0.0047
0.0046
PM
(Ib/hr)
0.72
0.48
0.62
0.61
Flow
(dscfm)
14,580
16,008
15,336
15,308
Flow
(acfm)
20,403
21,992
21,567
21,321
Temp
(°F)
227
216
231
225
Aincloth ratio
(ft/min)
2.7
2.9
2.9
2.8
Melt rate
(tph)
4.3
4.3
6.4
5.0
Cupola information
3.5 tph capacity, no
afterburner,
nonnonrecuperative,
above charge takeoff
Baghouse information
Installed 1994, negative
pressure, pulse jet, 16 oz Nomex '
fabric, aircloth = 3.96 ft/min,
design for 400°F and 30,000
acfm, controls melting and
charging
Foundry WI-49/50 (tested September 1995)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0044
0.0047
0.0060
0.0050
PM
(Ib/hr)
1.2
1.2
1.5
1.3
Flow
(dscfm)
30,852
30,826
29,750
30,476
Flow
(acfm)
59,684
59,347
60,281
59,771
Temp
(°F)
338
332
339
336
Air: cloth ratio
(ft/min)
3.0
3.0
3.0
3.0
Melt rate
(tph)
29.7
28.4
24.4
27.5
Cupola information
2 cupolas, 30 tph
capacity, afterburner,
recuperative, above
charge takeoff
Baghouse information
Installed 1994, negative
pressure, pulse jet, woven
fiberglas fabric, air: cloth = 2.4
to 3.7 ft/min, design for 450°F
and 50,000 to 70,000 acfm,
controls melting
D-6
-------
APPENDIX E
SOURCE TEST PARTICIPATE MATTER DATA
FOR ELECTRIC INDUCTION FURNACE FILTERS
-------
E.I INTRODUCTION
This appendix presents the individual sampling run data for the source tests available to
characterize the control performance for fabric and cartridge filters applied to EIF (Chapter 4).
Summary test data are given in Table E-l along with information on furnace melting rates and
capacities and a description of the filters and the processes they serve.
The data in Table E-l represent a range of furnace sizes and types of filters. The design
furnace melting rates range from 0.8 to 15 tons per hour, and ventilation rates range from 6,500
to 225,000 acfm. All of the foundries produce iron in the furnaces tested. The filters include
both negative and positive pressure operating modes and employ both shaker and pulse jet
cleaning systems. Some were installed about 20 to 25 years ago, and some are relatively new
(rebuilt). The design air-to-cloth ratios cover a range of 1.7 to 11.8 ft/min. No information is
available on the ages of the bags in service when the tests were conducted.
The reported results were checked to ensure the weights of PM from the filter and the
probe catch were above detection limits. When the reported catch was less than 3 mg, a
detection limit value of 3 mg and the sample volume were used to estimate the detection limit in
gr/dscf. Values calculated in this manner are reported as "less than" (<).
E-l
-------
TABLE E-l. PM TEST RESULTS FOR FILTERS SERVING EIF AND SCRAP PREHEATERS
Foundry MI-04 (tested August 1994 )
Run
1
2
3
Avg
PM*
(gr/dscf)
O.0006
O.0006
0.0006
O.0006
PM*
(Ib/hr)
O.027
O.027
<0.027
<0.027
Flow
(dscfm)
Flow
(acfm)
Temp
(°F)
Air-cloth ratio
(ft/min)
Melt rate
(tph)
4.1
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester
Design gas flow rate: 50,000 acfm
Design operating temperature: 80°F
Design air-to-cloth ratio: 6 ft/min
Serves 3 EIF, 1.5 tons/hr design melt rate for each
* The results were reported as O.0002 gr/dscf and were adjusted to O.0006 gr.dscf based on the best estimate of the detection limit.
Foundry CA-01 (tested March 1996)
Run
1
PM
(gr/dscf)
O.0002
PM
(Ib/hr)
<0.05
Flow
(dscfm)
41,000
Flow
(acfm)
43,110
Temp
(°F)
90
Air-cloth ratio
(ft/min)
2.56
Melt rate
(tph)
1.3
Baghouse design and service data
Positive pressure, shaker cleaning; in series with 2 prefilters
and a HEPA filter
Fabric: polyester
Design gas flow rate: 49,600 acfm
Design operating temperature: 81°F
Design air-to-cloth ratio: 2.95 ft/min
Serves 8 EIF, (0.5 to 1.75 tons/hr design melt rate), 4
casting stations, 4 mold spray/coating stations, 1 Hawley
system
E-2
-------
Foundry IN-13 (tested October 1996 )
Run
1
2
3
Avg
PM
(gr/dscf)
O.0006
O.0006
<0.0006
O.0006
PM
(Ib/hr)
<0.34
<0.34
<0.34
<0.34
Flow
(dscfin)
66,943
66,453
67,590
66,995
Flow
(acfm)
71,590
72,190
73,100
72,290
Temp
(°F)
95
102
100
99
Air-cloth ratio
(ft/min)
2.91
2.94
2.97
2.94
Melt rate
(tph)
33.8
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester
Design gas flow rate: 72,500 acfm
Design operating temperature: 150°F
Design air-to-cloth ratio: 2.95 ft/min
Installed 1995
Serves 3 EIF, 10.7 tons/hr design melt rate for each;
controls charging, melting, holding furnaces, ladle
metallurgy
Foundry WI-43 (tested November 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
<0.0010
O.0011
<0.0011
<0.0011
PM
(Ib/hr)
<0.6
<0.6
<0.6
<0.6
Flow
(dscfm)
60,236
59,491
58,117
59,281
Flow
(acfm)
66,964
66,543
65,870
66,459
Temp
(°F)
111
115
122
116
Air-cloth ratio
(ft/min)
4.0
3.9
3.9
3.9
Melt rate
(tph)
112
114
137
121
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester
Design gas flow rate: 1 10,000 acfm
Design operating temperature: 100°F
Design air-to-cloth ratio: 6.5 ft/min
Installed 1995
Serves 10 EIF, 1 1 tons/hr design melt rate each; controls
charging, melting, magnesium treatment
E-3
-------
Foundry WI-43: scrap preheater only (tested November 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
O.0007
<0.0007
O.0007
<0.0007
PM
(Ib/hr)
<0.4
<0.4
<0.4
<0.4
Flow
(dscfin)
71,594
72,303
73,230
72,376
Flow
(acfin)
88,045
88,649
87,282
87,992
Temp
(°F)
169
167
149
162
Air-cloth ratio
(ft/min)
7.8
7.9
7.7
7.8
Preheat
rate (tph)
56
69
58
61
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: fiberglass
Design gas flow rate: 80,000 acfin
Design operating temperature: 3 10°F
Design air-to-cloth ratio: 7.1 ft/min
Installed 1995
Serves 3 scrap preheaters, 33 tons/hr design rate each
Foundry MN-7 (tested August 1996)
Run
1
2
3
Avg
PM
(gr/dscf)
<0.0010
<0.0013
0.0014
<0.0012
PM
(Ib/hr)
<1.0
<1.2
1.3
<1.2
Flow
(dscfin)
110,900
111,900
109,600
110,800
Flow
(acfm)
118,500
120,600
118,800
119,300
Temp
(°F)
99
103
107
103
Air-cloth ratio
(ft/min)
3.9
3.9
3.9
3.9
Melt rate
(tph)
7.55
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester (Dacron) felt (16 oz) singed finish
Design gas flow rate: 1 19,300 acfin
Design operating temperature: 103°F
Design air-to-cloth ratio: 3.9 ft/min
Installed 1991; Serves one EIF, 15.2 tons/hr design melt
rate; controls charging, melting, tapping, holding furnaces,
ladle metallurgy, pouring/cooling
E-4
-------
Foundry WI-47 (tests of 3 systems)
Run
Avg
Avg
Avg
PM
(gr/dscf)
0.0011
0.0006
0.0052
PM
(Ib/hr)
0.4
0.22
2.92
Flow
(dscfin)
44,052
46,032
65,132
Flow
(acfin)
Temp
(°F)
Air-cloth ratio
(ft/min)
Melt rate
(tph)
3.0
2.8
4.4
Design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester
Design gas flow rate: 50,000 acfin
Design air-to-cloth ratio: 7 ft/min
Installed 1991
Serves preheater and one EIF, 3.5 tons/hr design melt rate;
controls charging, melting
Negative pressure, pulse jet cartridge cleaning
Fabric: cartridge collector
Design gas flow rate: 40,000 acfin
Design air-to-cloth ratio: 1.3 ft/min
Installed 1991
Serves two EIFs, 5 tons/hr design melt rate for each;
controls charging, melting; also controls inoculation and
cast cooling
Venturi scrubber with <13 in water pressure drop; 73,500
acfin
Serves two EIF for melting (5 tph each); also pouring and
cooling
E-5
-------
Foundry IN-24 (tested December 1996)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0017
0.0014
0.0026
0.0019
PM
(Ib/hr)
0.34
0.28
0.50
0.37
Flow
(dscfin)
23,050
23,171
22,909
23,043
Flow
(acfin)
23,111
23,074
22,842
23,009
Temp
(°F)
62
59
60
61
Air-cloth ratio
(ft/min)
1.55
1.55
1.53
1.55
Melt rate
(tph)
4.4
Cartridge filter design and service data
Negative pressure, pulse jet cartridge cleaning
Fabric: cellulose cartridge
Design gas flow rate: 25,000 acfin
Design operating temperature: 180°F
Design air-to-cloth ratio: 1.68 ft/min
Installed 1996
Serves two EIF, 4.5 tons/hi design melt rate controls
charging, melting, tapping
Foundry CA-09 (tested October 1987)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0015
0.0023
0.003
0.0023
PM
(Ib/hr)
0.076
0.113
0.145
0.11
Flow
(dscfin)
5,906
5,727
5,630
5,754
Flow
(acfrn)
6,503
6,427
6,426
6,452
Temp
(°F)
102
113
121
112
Air-cloth ratio
(ft/min)
1.4
1.3
1.3
1.3
Melt rate
(tph)
0.8
Baghouse design and service data
Negative pressure, shaker cleaning
Fabric: polyester
Design gas flow rate: 9,600 acfin
Design operating temperature: 130°F
Design air-to-cloth ratio: 2 ft/min
Installed 1997
Serves three EIFs, two at 0.8 tph and one at 1.5 tph design
melt rate each; controls melting, charging, preheater, and
sand reclaimer
E-6
-------
Foundry MN-12 (tested March 1995 and May 1996)
Run
1
2
3
4
5
6
Avg
1
2
3
4
5
6
Avg
PM
(gr/dscf)
0.0034
0.0014
0.0024
0.0022
0.0026
0.0012
0.0022
0.0009
0.0016
0.0028
0.0005
0.0006
0.0019
0.0014
PM
(Ib/hr)
0.38
0.14
0.21
0.24
0.31
0.14
0.47*
0.11
0.19
0.35
0.06
0.07
0.22
0.33*
Flow
(dscfhi)
13,200
11,700
10,300
12,700
13,700
13,800
25,100*
14,700
14,000
14,400
13,800
14,200
13,500
28,200*
Flow
(acfm)
13,500
12,200
11,000
13,100
14,100
14,200
26,000*
15,600
14,900
15,500
14,700
14,700
14,200
29,900*
Temp
(°F)
86
90
78
86
82
84
84
105
104
111
105
89
95
102
Air-cloth ratio
(ft/min)
2.54
2.29
2.07
2.46
2.65
2.67
2.45
2.93
2.80
2.91
2.76
2.76
2.67
2.80
Melt rate
(tph)
5.8
6.0
6.3
5.8
6.4
6.4
6.1
5.2
5.3
5.3
5.1
5.3
5.3
5.2
Baghouse design and service data
Positive pressure, shaker cleaning
Fabric: felt
Design gas flow rate: 29,800 acfm
Design operating temperature: 100°F
Design air-to-cloth ratio: 2.8 ft/min
Installed 1980
Serves two EIF, 4.7 tons/hr design melt rate each; controls
charging, melting, tapping, ladle metallurgy; two stacks on
baghouse
* The baghouse has two stacks; Runs 1-3 are for one stack and Runs 4-6 are for the other stack.
E-7
-------
Foundry PA-06 (tested July 1995; one of two baghouses in parallel)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0022
0.00124
0.00064
0.0014
PM
(Ib/hr)
0.71
0.39
0.2
0.43
Flow
(dscfhi)
37,936
36,578
36,267
36,927
Flow
(acfrn)
41,151
40,150
39,414
40,238
Temp
(°F)
106
108
104
106
Air-cloth ratio
(ft/min)
Melt rate
(tph)
8.0
Baghouse design and service data
Negative pressure, reverse pulse cleaning (two baghouses
in parallel)
Fabric: polyester
Design gas flow rate: 95,094acfm for two baghouses
Design operating temperature: 120°F
Design air-to-cloth ratio: 4.38 ft/min
Installed 1996
Serves one EIF at 10 tons/hr design melt rate each; also
controls inoculation and carbon/silicon adjustment
Foundry PA-06 (tested July 1995; one of two stacks; doubled flow and emission rate to estimate for both stacks)
Run
1
2
3
Avg
PM
(gr/dscf)
0.00225
0.00116
0.00117
0.0015
PM
(Ib/hr)
1.32
0.68
0.68
0.89
Flow
(dscfin)
68,464
68,402
68,094
68,320
Flow
(acfin)
75,040
75,204
74,434
74,893
Temp
(°F)
97
95
93
95
Air-cloth ratio
(ft/min)
Melt rate
(tph)
8.0
Baghouse design and service data
Negative pressure, reverse pulse cleaning (two baghouses
in parallel)
Fabric: polyester
Design gas flow rate: 95,094acfm for two baghouses
Design operating temperature: 120°F
Design air-to-cloth ratio: 4.57 ft/min
Installed 1996
Serves one EIF atlO tons/hr design melt rate each; also
controls inoculation and carbon/silicon adjustment
E-8
-------
Foundry OH-43 (tested October 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0038
0.0013
0.0018
0.0023
PM
(Ib/hr)
2.25
0.81
1.09
1.38
Flow
(dscfin)
69,695
71,174
71,568
70,812
Flow
(acfm)
74,979
76,590
78,190
76,586
Temp
(°F)
83
83
93
86
Air-cloth ratio
(ft/min)
6.04
6.17
6.30
6.34
Melt rate
(tph)2
9.4
5.9
12.2
9.2
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester
Design gas flow rate: 65,000 acfm
Design operating temperature: 90-1 10°F
Design air-to-cloth ratio: 5.24 ft/min
Installed 1996
Serves two EIF, 15 tons/hr design melt rate each; controls
melting, grinding, shot blasting, pouring
2 Tons per hour transferred; both furnaces were operating, but there was only one charge during the test. Test includes both melting and holding.
Foundry TX-11 (tested October 1993)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0030
0.0021
0.0020
0.0024
PM
(Ib/hr)
2.29
1.74
1.71
1.91
Flow
(dscfin)
81,362
77,351
76,379
78,364
Flow
(acfm)
93,159
90,950
90,057
91,389
Temp
(°F)
95
111
112
106
Air-cloth ratio
(ft/min)
3.11
3.03
3.00
3.05
Melt rate
(tph)
3.85
Baghouse design and service data
Negative pressure, shaker cleaning
Fabric: Nomex
Design gas flow rate: 90,000 acfm
Design operating temperature: 100°F
Design air-to-cloth ratio: 3 ft/min
Installed 1977
Serves one EIF, 3.75 tons/hr design melt rate; controls
charging, melting, tapping, ladle metallurgy
E-9
-------
Foundry MI-28 (tested March 1996)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0031
0.0028
0.0027
0.0029
PM
(Ib/hr)
1.03
0.94
0.96
1.03
Flow
(dscfin)
38,480
39,512
41,190
39,728
Flow
(acfin)
Temp
(°F)
Air-cloth ratio
(ft/min)
2.10
2.20
2.30
2.20
Melt rate
(tph)
5.20
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: Polyester
Design gas flow rate: 70,000 acfin
Design operating temperature: 135°F
Design air-to-cloth ratio: 3.9 ft/min
Installed 1995
Serves 3 EIFs, 9 tons/hr design melt rate and 2 scrap
preheaters; controls charging, melting, tapping
Foundry IN-11 (tested September 1990)
Run
1
2
3
Avg
1
2
3
Avg
PM
(gr/dscf)
0.0032
0.0050
0.0026
0.0036
0.0019
0.0037
0.0017
0.0024
PM
(Ib/hr)
1.435
2.217
1.140
1.597
1.456
2.827
1.303
1.862
Flow
(dscfin)
52,383
52,200
52,100
52,228
89,280
88,683
89,633
89,199
Flow
(acfin)
61,842
62,017
61,534
61,798
103,143
102,427
104,083
103,218
Temp
(°F)
143
143
142
143
135
136
139
137
Air-cloth ratio
(ft/min)
2.14
2.15
2.13
2.14
3.57
3.54
3.60
3.57
Melt rate
(tph)
Unknown
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester (Dacron)
Design gas flow rate: 100,000 acfin
Design operating temperature: unknown
Design air-to-cloth ratio: 3.46 ft/min
Installed 1990
Two identical baghouses serving three EEF each, 10 tons/hr
design melt rate each; controls preheater, charging, melting,
tapping
E-10
-------
Foundry IN-29 (tested February 1997)
Run
1
2
3
Avg
PM
(gr/dscf)
0.0025
0.0017
0.0076
0.0039
PM
(Ib/hr)
0.85
0.59
2.56
1.33
Flow
(dscfin)
40,367
39,694
39,033
39,698
Flow
(acfm)
42,354
41,609
41,037
41,667
Temp
(°F)
86
85
86
86
Air-cloth ratio
(ft/min)
12.5
12.3
12.1
12.3
Melt rate
(tph)
24
20
23
23
Baghouse design and service data
Positive pressure, pulse jet cleaning •
Fabric: polyester felt
Design gas flow rate: 40,000 acfm
Design operating temperature: 175°F
Design air-to-cloth ratio: 1 1 .8 ft/min
Installed 1996
Serves two EIF, 10.5 tons/hr design melt rate; controls
preheating, melting
Foundry IN-12 (tested March 1990)
Run
1
2
Avg
PM
(gr/dscf)
0.0056
0.0068
0.0062
PM
(Ib/hr)
2.38
2.86
2.62
Flow
(dscfin)
49,122
49,247
49,185
Flow
(acfm)
51,817
51,865
51,841
Temp
(°F)
99
99
99
Air-cloth ratio
(ft/min)
Melt rate
(tph)
15
Baghouse design and service data
Uncontrolled induction furnaces (3 at 5 tph)
Foundry PA-46 (tested October 1995)
Run
1
2
3
Avg
PM
(gr/dscf)
0.008
0.009
0.008
0.008
PM
(Ib/hr)
10.76
11.25
10.55
10.85
Flow
(dscfhi)
155,000
150,000
155,000
153,000
Flow
(acfm)
Temp
(°F)
Air-cloth ratio
(ft/min)
Melt rate
(tph)
15
Baghouse design and service data
Negative pressure, pulse jet cleaning
Fabric: polyester
Design gas flow rate: 225,000 acfm
Design operating temperature: 100°F
Design air-to-cloth ratio: 6.8 ft/min
Installed 1995
Serves five EIF, 3.3, 3.3, 4.1, 6.8, and 12.7 tons/hr design
melt rate; controls charging, melting, tapping
E-ll
-------
APPENDIX F
SOURCE TEST PARTICULATE MATTER DATA
FOR ELECTRIC ARC FURNACE BAGHOUSES
-------
F.I INTRODUCTION
This appendix presents the individual sampling run data for the source tests available to
characterize the control performance for baghouses applied to EAF (Chapter 4). Summary test
data are given in Table F-l along with information on furnace melting rates and capacities and a
description of the control systems and the processes they serve.
The data in Table F-l represent a range of furnace sizes and types of baghouses. The
design furnace melting rates range from 2.5 to 15 tons per hour, and ventilation rates range from
31,000 to 225,000 acfm. The baghouses include both negative and positive pressure operating
modes and employ both shaker and pulse jet cleaning systems. Some were installed about 30
years ago, and some are relatively new (rebuilt). The design air-to-cloth ratios cover a range of
2.3 to 5.7 ft/min. No information is available on the ages of the bags in service when the tests
were conducted.
The reported results were checked to ensure the weights of PM from the filter and the
probe catch were above detection limits. When the reported catch was less than 3 mg, a
detection limit value of 3 mg and the sample volume were used to estimate the detection limit in
gr/dscf. Values calculated in this manner are reported as "less than" (<).
F-l
-------
TABLE F-l. PM TEST RESULTS FOR BAGHOUSES SERVING EAF
Foundry IN-7 (tested December 1997)
Run
1
2
3
Average
PM
loading
(gr/dscf)
0.0006
0.0004
0.0005
0.0005
PM mass
flow rate
(Ib/hr)
0.15
0.11
0.13
0.13
Flow rate
(dscfin)
29,200
32,100
30,300
30,500
Flow rate
(acfm)
Temp,
(°F)
Air-to-cloth
ratio
(ft/min)
Melt rate
(tph)
3.7
3.7
3.1
3.5
Baghouse design and service data
Negative pressure; shaker cleaning.
Fabric: Dacron/cotton.
Design gas flow rate: 3 1,200 acfm.
Design operating temperature: 100 °F.
Design air-to-cloth ratio: 2.51 ft/min.
Rebuilt 1985.
Serves one EAF, 3.6 tons/hr design melt rate.
Foundry IA-09 (tested August 1996)
Run
1
2
3
Average
PM
loading
(gr/dscf)
0.00083
0.00063
0.00041
0.00062
PM mass
flow rate
(Ib/hr)
0.62
0.47
0.29
0.46
Flow rate
(dscfm)
85,099
85,200
79,414
83,238
Flow rate
(acfhi)
87,520
87,030
81,406
85,319
Temp,
(°F)
127
129
126
127
Air-to-cloth
ratio
(ft/min)
2.4
2.4
2.3
2.4
Melt rate
(tph)
5.65
Baghouse design and service data
Positive pressure; shaker cleaning.
Fabric: 10.5 oz. polyester.
Design gas flow rate: 85,000 acfm
Design operating temperature: 90 °F.
Design air-to-cloth ratio: 2.36 ft/min.
Installed 1974.
Serves two EAFs, 6.0 tons/hr design melt rate
each, one holding furnace with 61 tons capacity,
and one holding furnace with 40 tons capacity.
F-2
-------
Foundry IA-09 (tested July 2002)
Run
1
2
3
Average
PM
loading
(gr/dscf)
0.0007
0.0007
0.0006
0.00067
PM mass
flow rate
(Ib/hr)
0.51
0.50
0.42
0.48
Flow rate
(dscfin)
85,927
83,992
80,727
83,549
Flow rate
(acfm)
93,624
89,854
86,978
90,152
Temp,
(°F)
127
117
121
122
Air-to-cloth
ratio
(ft/min)
2.6
2.5
2.4
2.5
Melt rate
(tph)
5.65
Baghouse design and service data
Positive pressure; shaker cleaning.
Fabric : 1 0 . 5 oz. polyester.
Design gas flow rate: 85,000 acfm
Design operating temperature: 90 °F.
Design air-to-cloth ratio: 2.36 ft/min.
Installed 1974.
Serves two EAFs, 6.0 tons/hr design melt rate
each, one holding furnace with 61 tons capacity,
anH nnp VmlHina fiirnapp with 40 tnn<3 ranar.itv
Foundry IA-09 (tested May 1995)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0013
0.001
0.00072
0.0010
PM mass
flow rate,
Ib/hr
1.0
0.63
0.50
0.71
Flow rate,
dscfin
87,520
87,030
81,406
85,319
Flow rate,
acftn
Temp,
op
Air-to-cloth
ratio,
ft/min
Melt rate
(tph)
—
—
—
5.65 total
Baghouse design and service data
Positive pressure; shaker cleaning.
Fabric: 10.5 oz. polyester.
Design gas flow rate: 85,000 acfin
Design operating temperature: 90 °F.
Design air-to-cloth ratio: 2.36 ft/min.
Installed 1974.
Serves two EAFs, 6.0 tons/hr design melt rate
each, one holding furnace with 61 tons capacity,
and one holding furnace with 40 tons capacity.
F-3
-------
Foundry TX-19 (January 1995)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0030
O.0013
O.0013
O.002
PM mass
flow rate,
Ib/hr
1.18
<0.5
<0.5
<0.7
Flow rate,
dscfm
46,100
47,700
46,700
46,800
Flow rate,
acfin
51,000
52,500
51,600
51,700
Temp,
op
114
114
118
115
Air-to-cloth
ratio,
ft/min
2.34
2.41
2.37
2.37
Melt rate
(tph)
Baghouse design and service data
Negative pressure; shaker cleaning.
Fabric: 10.5 oz. seamless polyester.
Design gas flow rate: 50,000 acfm.
Design operating temperature: 250 °F.
Design air-to-cloth ratio: 2.30 ft/min.
Serves two EAFs, 5 tons/hr design melt rate
each.
Foundry AL-11 (tested September 1995)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0019
0.0017
0.0009
0.0015
PM mass
flow rate,
Ib/hr
1.77
1.58
0.87
1.41
Flow rate,
dscfm
109,000
108,000
113,000
110,000
Flow rate,
acfm
122,000
123,000
127,000
124,000
Temp,
OF
121
130
126
126
Air-to-cloth
ratio,
ft/min
3.05
3.08
3.18
3.10
Melt rate
(tph)
9.1,9.4
9.4, 9.5
9.1,9.5
9.2, 9.5
Baghouse design and service data
Negative pressure; pulse jet cleaning.
Fabric: 18 oz. polyester dual density felt.
Design gas flow rate: 140,000 acfin.
Design operating temperature: 200 °F.
Design air-to-cloth ratio: 3. 50 ft/min.
Rebuilt 1995.
Serves two EAFS, 9.25 tons/hr design melt rate
each.
F-4
-------
Foundry MN-3 (tested May 1993)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0021
0.0019
0.0019
0.0020
PM mass
flow rate,
Ib/hr
2.64
2.29
2.45
2.46
Flow rate,
dscfrn
146,200
142,200
151,000
146,500
Flow rate,
acfm
155,600
150,000
157,100
154,200
Temp,
oF
84
85
85
85
Air-to-cloth
ratio,
ft/min
2.27
2.19
2.30
2.25
Melt rate
(tph)
4.8, 3.9
4.8, 4.4
6.3, 4.4
5.3,4.2
Baghouse design and service data
Negative pressure; shaker cleaning.
Fabric: polyester.
Design gas flow rate: 1 80,000 acfin
Design operating temperature: 100°F
Design air-to-cloth ratio: 2.4 ft/min
Installed 1980.
Serves two EAFs, 4.3 tons/hr design melt rate
each.
Foundry MI-09 (tested October 1996)
Run
1
2
3
4
Average
PM
loading
(gr/dscf)
0.0044
0.0030
0.0017
0.0015
0.0027
PM mass
flow rate
(Ib/hr)
1.03
0.69
0.39
0.35
0.62
Flow rate
(dscfm)
26,702
26,365
26,716
26,544
26,582
Flow rate
(acfm)
31,467
31,868
31,447
31,654
31,609
Temp,
(°F)
144
159
143
151
149
Air-to-cloth
ratio
(ft/min)
Melt rate
(tph)
12
Baghouse design and service data
Positive pressure; shaker cleaning.
Fabric: Polyester.
Design gas flow rate: 200,000 acfm.
Design operating temperature: 170°F.
Design air-to-cloth ratio: 2.33 ft/min.
Built 1987.
Serves three EAF, 15 tons/hr design melt rate.
F-5
-------
Foundry OH-1 (tested March 1994)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0025
0.0030
0.0025
0.0027
PM mass
flow rate,
Ib/hr
4.45
5.26
4.42
4.71
Flow rate,
dscfm
208,000
205,000
206,000
206,000
Flow rate,
acfm
234,000
230,000
230,000
231,000
Temp,
op
96
103
102
100
Air-to-cloth
ratio,
ft/min
Melt rate
(tph)
Baghouse design and service data
Design gas flow rate: 225,000 acfm.
Design operating temperature: 150 °F.
Design air-to-cloth ratio:
Serves three EAFs, 13 tons/hr design melt rate
each.
Foundry OH-1 (tested May 1997)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0063
0.0076
0.0059
0.0066
PM mass
flow rate,
Ib/hr
Flow rate,
dscfm
Flow rate,
acfm
Temp,
op
Air-to-cloth
ratio,
ft/min
Melt rate
(tph)
Baghouse design and service data
F-6
-------
Foundry WI-45 (tested September 1990)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0033
0.0025
0.0035
0.0031
PM mass
flow rate,
Ib/hr
1.97
1.45
1.77
1.73
Flow rate,
dscfm
33,550
33,800
33,667
33,700
Flow rate,
acfm
Temp,
°F
Air-to-cloth
ratio,
ft/min
Melt rate
(tph)
2.07
2.16
2.46
2.23
Baghouse design and service data
Positive pressure; shaker cleaning.
Fabric: polyester/cotton.
Design gas flow rate: 35,000
Design operating temperature: 125 °F.
Design air-to-cloth ratio: 5.7 ft/min.
Installed 1979.
Serves one EAF, 2.5 tons/hr design melt rate
and sand mulling.
Foundry IA-17 (tested January 1995)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0069
0.0029
0.0035
0.0044
PM mass
flow rate,
Ib/hr
5.35
2.55
2.68
3.53
Flow rate,
dscfm
82,000
92,100
85,200
86,400
Flow rate,
acfm
Temp,
op
Air-to-cloth
ratio,
ft/min
Melt rate
(tph)
8.3,11.6
9.6, 14.1
Baghouse design and service data
Negative pressure; shaker cleaning.
Fabric: woven Dacron.
Design gas flow rate: 120,383.
Design operating temperature: 1 82 °F.
Design air-to-cloth ratio: 2.59 ft/min.
Installed 1972.
Serves two EAFs, 12 tons/hr design melt rate
each.
F-7
-------
Foundry PA-11 (tested November 1994)
Run
1
2
3
Average
PM
loading,
gr/dscf
0.0058
0.0080
0.0103
0.0080
PM mass
flow rate,
Ib/hr
4.9
6.3
7.6
6.3
Flow rate,
dscfin
99,000
92,000
86,000
92,000
Flow rate,
acftn
Temp,
OF
Air-to-cloth
ratio,
ft/min
Melt rate
(tph)
15.1
15.1
8.2
12.8
Baghouse design and service data
Negative pressure; shaker cleaning.
Fabric: polyester.
Design gas flow rate: 120,000.
Design operating temperature: 130 °F.
Design air-to-cloth ratio: 3.2 ft/min.
Installed 1977.
Serves one EAF, 15 tons/hr design melt rate.
F-8
-------
APPENDIX G
SOURCE TEST PARTICULATE MATTER DATA
FOR POURING, COOLING AND SHAKEOUT
-------
G.I INTRODUCTION
This appendix presents the individual sampling run data for the source tests available to
characterize the control performance for baghouses and wet scrubbers applied to pouring,
cooling, and shakeout at iron foundries (Chapter 4). Summary test data are given in Table G-l
along with descriptions of the control systems and the processes they serve. The test data and
control device information were compiled from EPA's comprehensive 1998 survey of the
industry. Some respondents provided information on each test run, and other respondents only
provided the average for the test. Individual run information is presented when available.
There are many common features of the baghouses listed in the table. Almost all are
negative pressure baghouses with pulse jet cleaning, and most of the air-to-cloth ratios are in the
range of 5 to 7.4 ft/min. The table includes one shaker baghouse and four cartridge filters. The
design flow rates for baghouses range 10,000 to 375,000 acfm. It is common for a single control
system to serve multiple operations, such as different combinations of pouring, cooling,
shakeout, and sand handling.
The wet scrubbers are low pressure drop devices with a range of 3.2 to 13.5 inches of
water. The types include venruri, cyclonic, and centrifugal scrubbers. The design flow rates for
the scrubbers range from 32,000 to 104,000 acfm, and liquid-to-gas ratios range from 2 to 8
gallons per 1,000 actual cubic feet. The scrubbers are applied to various combinations of
pouring, cooling, and shakeout, and one also serves as the control device for an induction
furnace.
G-l
-------
TABLE G-l. PM SOURCE TEST RESULTS FOR BAGHOUSES AND SCRUBBERS SERVING
POURING, COOLING, AND SHAKEOUT
1 Foundry
ro
IN- 13
WI-43
WI-43
WI-43
IN-13
WI-43
WI-43
WI-43
WI-42
WI-42
WI-01
WI-43
I A- 17
WI-01
MN-12
Date
Oct-96
Oct-93
Oct-93
Oct-93
Oct-96
Oct-93
Oct-93
Aug-95
Jan-00
Jan-00
Jun-94
Aug-95
Feb-96
Jun-94
Mar-95
Run no.
1
2
3
Average
Average
Average
Average
1
2
3
Average
Average
Average
Average
Average
Average
Average
Average
1
2
3
Average
Average
1
2
3
Average
PM
(Ib/hr)
0.51
0.28
0.23
0.34
0.56
0.24
0.44
0.57
0.35
0.17
0.36
0.39
0.42
0.87
1.10
0.80
1.89
0.67
0.40
0.42
0.50
0.50
0.10
0.20
0.16
0.15
PM
(gr/dscf)
0.00043
0.00024
0.00020
0.00029
0.0005
0.0005
0.0005
0.00085
0.00054
0.00026
0.00055
0.0008
0.0009
0.0009
0.001
0.001
0.001
0.0012
0.0018
0.0011
0.0011
0.0013
0.0015
0.0009
0.0020
0.0016
0.0015
Flow
(dscfm)
138,000
136,000
134,000
137,000
135,000
55,000
94,000
78,000
76,000
76,000
76,000
56,000
55,000
.
102,000
128,000
93,000
184,000
43,000
42,000
45,000
43,000
39,000
12,400
11,600
11,700
11,900
Design intormation tor fabric filters
Flow
facfm^
150,000
143,000
60,000
101,000
85,500
60,000
60,000
60,000
150,000
150,000
198,000
180,000
50,000
51,000
12,400
Air: cloth
(It/mint
4.5
5.6
7.1
7.1
4.4
7.1
7.1
7.1
6.6
6.5
NR
7.1
1.4
(cartridge)
NR
9.4
Material
polyester
polyester felt
polyester felt
polyester felt
polyester
polyester felt
polyester felt
polyester felt
polyester felt
polyester felt
NR
polyester felt
cellulose
NR
polyester
Cleaning
tvpe
pulse jet
pulse jet
pulse jet
pulse iet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
NR
pulse jet
pulse jet
NR
shaker
Operations served
shakeout Lines 1 & 2; return sand s
cooling and grinding (C22)
shakeout and grinding (C33)
cooling and sand handling (C3 1)
shakeout Lines 3 & 4
cooling, shakeout, grinding (C32)
cooling, shakeout, grinding (C34)
cooling, shakeout, grinding (C32)
2 pouring/cooling lines; 2 cast
cooling lines
shakeout
5 shakeout/cast cooling lines
cooling, shakeout, grinding (C35)
shakeout and sand transfer with
Line 802
pouring/cooling Lines 1 & 5; sand
mullor
pouring
G-2
-------
irtoundry Date
,„
SC-07
IN-29
OH-48 Nov-96
WI-43 Oct-93
WI-15
MN-12 Mar-95
OH-13 Feb-96
TX-19 Jan-95
I A- 17 Feb-96
WI-01 Jun-94
Run no.
1
2
3
Average
Average
1
2
3
Average
Average
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
Average
PM
(lb/hr)
0.62
0.86
1.08
0.85
0.75
0.30
0.06
0.16
0.17
1.66
1.49
0.25
0.19
0.24
0.23
1.76
1.22
1.58
1.52
1.1
0.29
0.54
0.64
4.43
3.79
2.60
3.61
1.18
I'M
(gr/dscf)
0.0013
0.0018
0.0024
0.0019
0.0019
0.0034
0.0007
0.0017
0.0019
0.0022
0.0024
0.0029
0.0023
0.0029
0.0027
0.0033
0.0023
0.0030
0.0028
0.005
0.0015
0.002
0.003
0.0038
0.0033
0.0023
0.0031
0.0032
Flow
(dscfm)
55,000
56,000
52,000
54,000
46,000
10,306
10,655
10,802
10,588
89,000
73,000
10,100
9,800
9,800
9,900
63,000
63,000
63,000
63,000
21,259
22,481
21,002
21,581
136,000
134,000
132,000
134,000
43,000
Design information tor fabric filters
Flow
facfm^
60,000
51,000
10,000
96,000
75,000
10,000
65,000
30,000
140,000
198,000
Air: cloth
(ft/mm)
NR
(cartridge)
5.5
9.1
8.9
7
5.8
6.5
6.5
7.4
NR
Material
NR
polyester felt
polyester
acrylic
coated
polyester
polyester
felt
polyester
polyester
singed
polyester
NR
Cleaning
tvpe
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
NR
Operations served
pouring/cooling
shakeout
shakeout
cooling and shakeout (C19)
pouring, cooling, shakeout
cooling
pouring, cooling, shakeout,
miscellaneous (DS23)
shakeout
shakeout and sand transfer
Line 803
pouring/cooling Lines 2 & 4
handling
with
; sand
G-3
-------
foundry Date
ID
SC-07
IA-17 Nov-97
OH-43 Oct-97
IN-11
IA-17 Feb-96
IA-17 Feb-96
WI-01 Jun-94
MI-04
OH-13 Mar-98
TX-11
Run no.
1
2
3
Average
1
2
3
Average
1
2
3
Average
Average
1
2
3
Average
1
4
5
Average
Average
Average
1
2
3
Average
1
2
3
Average
I'M
(lb/hr)
0.70
0.67
0.70
0,69
15.17
11.60
9.11
11.96
1.79
0.91
2.06
1.59
4.91
3.65
3.21
3.92
2.96
7.03
7.48
5.82
1.46
1.30
1.05
0.94
6.56
2.85
1.83
0.69
0.95
1.16
PM
(gr/dscf)
0.0035
0.0037
0.0041
0.0038
0.0047
0.0038
0.0029
0.0038
0.0045
0.0023
0.0049
0.0039
0.0041
0.0054
0.0040
0.0036
0.0043
0.0022
0.0052
0.0056
0.0043
0.0044
0.0047
0.0020
0.0018
0.0122
0.0053
0.0089
0.0033
0.0047
0.0056
Flow
(dscfm)
23,000
21,000
20,000
21,000
377,000
356,000
366,000
367,000
46,000
46,000
49,000
47,000
.
106,000
106,000
104,000
106,000
157,000
158,000
156,000
157,000
39,000
32,000
63,000
63,000
63,000
63,000
24,000
24,000
24,000
24,000
Design information
Flow
facfro^
20,000
375,000
50,000
174,000
110,000
110,000
101,000
75,000
65,000
170,300
Aincloth
(ft/rnin)
NR
(cartridge)
5
5.2
3.6
5.9
5.9
NR
5.1
6.5
2
lor tabric
Material
NR
polyester
polyester
polyester
polyester
polyester
NR
polyester
polyester
polyester
filters
Cleaning
tvpe
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
NR
pulse jet
pulse jet
shaker
Operations served
shakeout
shakeout and sand transfer with
Line 801
cooling; bond and sand storage
shakeout
shakeout and sand transfer with
Line 802
shakeout and sand transfer with
Line 802
pouring/cooling Lines 2 & 4; sand
handling
shakeout
pouring, cooling, shakeout;
miscellaneous (DS23)
pouring, cooling, shakeout
G-4
-------
1 .foundry Date
ID
IA-17 Feb-96
OH-43 Oct-97
IN- 11
AZ-04
OH-22 May-93
Run no.
1
2
3
Average
1
2
3
Average
Average
Average
1
2
3
Average
FM
(Ib/hr)
2.02
1.92
1.25
1.73
2.50
3.21
2.57
2.76
6.3
11.2
24.6
15.7
17.2
I'M
(gr/dscf)
0.0074
0.0073
0.0048
0.0065
0.0063
0.0075
0.0059
0.0066
0.0076
0.0109
0.019
0.042
0.027
0.029
Flow
(dscfm)
32,000
31,000
30,000
31,000
46,000
50,000
51,000
49,000
.
67,911
67,863
68,314
68,029
Design
Flow
facfm^
35,600
50,000
180,000
45,000
80,000
information lor tabric titters
Air:cloth
(ft/min)
7.4
5.2
7.4
6.1
6.4
Material
singed
polyester
polyester
polyester
polyester
polyproplene
Cleaning
tvpe
pulse jet
pulse jet
pulse jet
pulse jet
pulse jet
Operations sei
•yed
shakeout and sand transfer with
Line 803
shakeout and sand
cooling
shakeout
shakeout
shakeout
G-5
-------
(foundry Date
ID
TN-09 Mar-86
TN-09 Sep-90
TN-09 Mar-87
TN-09 Mar-87
WI-28 Oct-90
WI-47
TN-09 Mar-86
OH-22 May-93
WI-47
OH-22 Oct-94
Run no.
1
2
Average
Average
1
2
3
Average
1
2
3
Average
1
2
Average
Average
1
2
3
Average
1
2
3
Average
Average
1
2
3
Average
FM
(Ib/hr)
0.49
0.50
0.50
1.06
0.66
1.12
0.63
0.80
1.40
2.00
2.20
1.90
1.50
1.25
1.38
2.92
2.90
0.80
2.30
2.00
7.3
3.3
3.5
4.7
1.18
8.52
14.2
8.73
10.5
PM
(gr/dscf)
0.0011
0.0011
0.0011
0.0023
0.0021
0.0033
0.0019
0.0024
0.0032
0.0046
0.0050
0.0043
0.0051
0.0043
0.0047
0.0052
0.0091
0.0021
0.0052
0.0055
0.0085
0.0039
0.0042
0.0055
0.0064
0.01
0.016
0.01
0.012
Flow
(dscfm)
39,000
34,000
36,000
54,000
51,000
51,000
51,000
52,000
34,000
34,000
34,000
65,000
37,000
44,000
52,000
42,000
99,871
98,072
98,931
98,958
21,000
102,769
101,986
106,224
103,660
Design
Flow
(acfm)
75,000
75,000
45,000
54,000
60,000
73,500
75,000
99,000
32,000
104,000
information
Type
cyclonic
cyclonic
centrifugal
centrifugal
venturi
venturi
cyclonic
venturi
venturi
venturi
for wet scrubbers
Ap
(in water)
13.5
13.5
6.0
5.8
6
13
13.5
3.5
13
3.2
L:G
(1,000
jsal/acft
8
8
2.5
2.5
5
2
8
NR
2
NR
Operations served
shakeout
shakeout
shakeout
pouring and cooling
cooling, shakeout
induction furnace, pouring, cooling
shakeout
shakeout
shakeout
shakeout
NR = not reported
G-6
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-453/R-02-013
2.
4. TITLE AND SUBTITLE
National Emission Standards for Hazardous Air Pollutants
(NESHAP) for Iron and Steel Foundries - Background
Information for Proposed Standards
7. AUTHORS)
Jeff Coburn, RTI and Kevin Cavender, EPA
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 2771 1
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 2771 1
3. RECIPIENTS ACCESSION NO.
5. REPORT DATE
December 2002
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT
NO.
10. PROGRAM ELEMENT NO.
1 1 . CONTRACT/GRANT NO.
68-D6-0014
13. TYPE OF REPORT AND PERIOD
COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report provides the background information for the proposed NESHAP to control metallic
organic emissions of hazardous air pollutants (HAP) from iron and steel foundries. The emissic
control techniques, estimates of emissions, control costs, and environmental impacts are present
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b. IDENTIFIERS/OPEN ENDED TERMS c. COSATI
Field/Group
emission controls
environmental impacts
estimates of air emissions
18. DISTRIBUTION STATEMENT
Release Unlimited
Air Pollution Control
Iron and Steel Foundries
Hazardous Air Pollutants
1 9. SECURITY CLASS (Report) 2 1 . NO. OF PAGES
Unclassified
20. SECURITY CLASS (Page) 22. PRICE
Unclassified
md
n
^d.
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |