FTIR and Emissions Test at an Iron Foundry Waupaca Foundry, Inc., Plant No. 5, Tel City, Indiana

r, EPA
            United States
            Environmental Protection
            Agency
               Office of Air Quality
               Planning and Standards
               Research Triangle Park, NC 27711
EPA-454/R-99-035
September 1999
            Air
FTIR EMISSIONS TEST AT AN IRON
FOUNDRY
            Waupaca Foundry, Inc.
            Plant No. 5, Tel City, Indiana

-------
FTIR EMISSIONS TEST AT AN IRON FOUNDRY
                                   Waupaca Foundry, Inc.
                             Plant No. 5, Tell City, Indiana
                                            Prepared for

                 Office of Air Quality Planning and Standards
                Emissions, Monitoring and Analysis Division
                    Emission Measurement Center (MD-19)
               Research Triangle Park, North Carolina 27711
                                     Mr. Michael Ciolek
                               Work Assignment Manager
                           EPA Contract No. 68-D-98-027
                                  Work Assignment 2-13
                         MRI Project No. 104951-1-013-04
                                        September, 1999

-------
                  GENERAL DISCLAIMER
 This document may have problems that one or more of the following disclaimer
                        statements refer to:
This document has been reproduced from the best copy furnished by the
sponsoring agency. It is being released in the interest of making
available as much information as possible.

This document may contain data which exceeds the sheet parameters.  It
was furnished in this condition by the sponsoring agency and is the best
copy available.

This document may contain tone-on-tone or color graphs, charts and/or
pictures which have been reproduced in black and white.

The document is paginated as submitted by the original source.

Portions of this document are not fully legible due to the historical nature
of some of the material. However, it is the best reproduction available
from the original submission.

-------
PREFACE

This draft report was prepared by Midwest Research Institute (MRI) for the U. S.
Environmental Protection Agency (EPA) under EPA Contract No. 68-D-98-027, Work
Assignment No. 2-13. Mr. Michael Ciolek is the EPA Work Assignment Manager (WAM).
Dr. Thomas Geyer is the MRI Work Assignment Leader (WAL). The field test was performed
under EPA Contract No. 68-D2-0165, Work Assignment No. 4-25 and a draft report was
submitted under EPA Contract No. 68-W6-0048, Work Assignment No. 2-08. Mr. Michael
Ciolek was the EPA WAM for the Emission Measurement Center (EMC) under Work
Assignment 4-25 and Mr. Michael Toney was the WAM under Work Assignment No. 2-08.
Mr. John Hosenfeld was the MRI WAL under Work Assignment 2-08 and Dr. Thomas Geyer
was the MRI task leader for Work Assignment 2-08, task 08.

This report presents the procedures, schedule, and test results for an emissions test
performed at Waupaca Foundry in Tel City, Indiana. The emissions test used Fourier transform
infrared (FTIR) sampling procedures to measure hazardous air pollutants (HAP's) and other
pollutants.

This report consists of one volume (354 pages) with seven sections and four appendices.

Midwest Research Institute
John Hosenfeld
Program Manager
Approved:
Jeff Shular
Director, Environmental Engineering Division

September 30,1999
in

-------
                           TABLE OF CONTENTS
1,0 INTRODUCTION	1-1
      1.1 BACKGROUND 	1-1
      1.2 PROJECT SUMMARY 	1-1
      1,3 PROJECT PERSONNEL	1-5

2.0 PROCESS AND CONTROL EQUIPMENT OPERATION 	2-1
      2.1 INTRODUCTION	2-1
      2.2 PROCESS DESCRIPTION  	2-1
           2.2.1 Iron Melting in Cupolas  	2-1
           2.2.2 Pouring, Cooling and Shakeout	,	2-4
      2.3 SUMMARY OF PROCESS OPERATING DATA COLLECTED DURING
           SOURCE TEST	2-5
           2.3.1 Process Operating Data for Cupola Melting Operations	2-5
           2.3.2 Process Operating Data for Pouring, Cooling and Shakeout
                 Operations	2-11

3.0 TEST LOCATIONS AND GAS COMPOSITION	,	 3-1
      3.1 BAGHOUSE OUTLET - STACK	3-1
      3.2 BAGHOUSE INLET DUCT	3-1
      3.3 MOLD COOLING LINE 	3-1
      3.4 MOLD SHAKE-OUT HOUSING  	3-5
      3.5 VOLUMETRIC FLOW 	3-5

4.0 RESULTS	4-1
      4.1 TEST SCHEDULE	4-1
      4.2 FIELD TEST PROBLEMS AND CHANGES	4-1
      4.3 FTIR RESULTS	4-2
           4.3.1 Mold Cooling Line and Shake-out Housing 	4-2
           4.3.2 Baghouse Inlet and Outlet 	4-2
      4.4 ANALYTE SPIKE RESULTS	4-2

5.0 TEST PROCEDURES	5-1
      5.1 SAMPLING SYSTEM DESCRIPTION	5-1
           5.1.1 Sample System Components	5-1
           5.1.2 Sample Gas Stream Flow	5-3
      5.2 FTIR SAMPLING PROCEDURES	5-3
           5.2.1 Batch Samples	.5-4
           5.2.2 Continuous Sampling	.,	5-4

-------
                     TABLE OF CONTENTS (CONTINUED)
      5.3 ANALYTE SPIKING	5-5
           5.3.1 Analyte Spiking Procedures	5-5
           5.3.2 Analysis of Spiked Results 	5-6
                 5.3.2.1 Determination of Formaldehyde Standard  	5-6
                 5.3.2.2 Determination of Concentrations in Spike Mixtures 	5-6
                 5.3.2.3 Determination of Percent Recovery 	5-7
      5.4 ANALYTICAL PROCEDURES 	5-9
           5.4.1 Computer Program Input 	5-10
           5.4.2 EPA Reference Spectra 	5-11
      5.5 FTIR SYSTEM	5-13

6.0 SUMMARY OF QA/QC PROCEDURES	6-1
      6.1 SAMPLING AND TEST CONDITIONS	6-1
      6.2 FTIR SPECTRA	6-2

7.0 REFERENCES	7-1
APPENDIX A - VOLUMETRIC FLOW DATA
      A-l. MOLD COOLING AND SHAKE-OUT HOUSING LINE FLOW DATA
      A-2. BAGHOUSE FLOW DATA

APPENDIX B - FTIR DATA
      B-l. FTIR RESULTS TABLES
           TABLE B-l.  FTIR RESULTS FROM THE MOLD COOLING LINE
           TABLE B-2.  FTIR RESULTS FROM THE SHAKE-OUT HOUSING LINE
           TABLE B-3.  FTIR RESULTS AT THE BAGHOUSE INLET
           TABLE B-4.  FTIR RESULTS AT THE BAGHOUSE OUTLET
      B-2. FTIR FIELD DATA RECORDS
      B-3. FTIR FLOW AND TEMPERATURE READINGS

APPENDIX C - CALIBRATION GAS CERTIFICATES

APPENDIX D - TEST METHODS
      D-l. EPA METHOD 320
      D-2. EPA FTIR PROTOCOL
                                   VI

-------
                    TABLE OF CONTENTS (CONTINUED)
                                                                  Page

                            LIST OF FIGURES

Figure 2-1. Simplified schematic of cupola gas handling system	,	2-2
Figure 2-2. Hourly melting rates during cupola testing 	2-7
Figure 2-3. Blast air flow rate through cupola	2-10
Figure 3-1. Schematic of cupola gas handling system, sampling points A and B  	3-2
Figure 3-2. Schematic of baghouse inlet and baghouse outlet. Sampling points
         A and B, respectively	3-3
Figure 3-3. Schematic of mold cooling and mold shake-out gas handling system;
         sampling points C and D  	3-4
Figure 4-1. Example spectra of spiked and unspiked baghouse outlet samples	 4-10
Figure 5-1. Sampling system schematic	5-2
                            LIST OF TABLES

TABLE 1-1. SUMMARY OF FTM RESULTS FROM THE MOLD COOLING
          AND SHAKE-OUT HOUSING DUCTS	1-3
TABLE 1-2. SUMMARY OF FTIR RESULTS (ppm) AT THE CUPOLA
          BAGHOUSE INLET AND OUTLET	1-4
TABLE 1-3. PROJECT PERSONNEL 	1-5
TABLE 2-1. TYPICAL CUPOLA CHARGE MATERIALS	2-3
TABLE 2-2. SUMMARY OF CUPOLA CHARGING DURING THE TEST DAYS	2-6
TABLE 2-3. PROCESS DATA DURING THE DAYS OF CUPOLA TESTING 	2-8
TABLE 2-4. PERIODS WHEN CUPOLA WAS "ON RELIEF' DURING TESTING ..... 2-9
TABLE 2-5. PROCESS IRON CHEMISTRY AT CUPOLA	2-11
TABLE 2-6. TYPICAL RESULTS FROM GREEN SAND ANALYSIS	2-12
TABLE 3-1. FLOW DATA AT WAPAUCA MOLD COOLING AND
          SHAKE-OUT HOUSING PROCESSES 	3-5
TABLE 3-2. CUPOLA BAGHOUSE INLET AND OUTLET GAS COMPOSITION
          AND FLOW SUMMARIES	3-6
TABLE 4-1. TEST SCHEDULE AT WAUPACA FOUNDRY	4-1
TABLE 4-2. FORMALDEHYDE SPIKE RESULTS FROM THE MOLD COOLING
          PROCESS	,	4-5
TABLE 4-3. FORMALDEHYDE SPIKE RESULTS FROM THE WAUPACA
          SHAKE-OUT HOUSING PROCESS	4-5
TABLE 4-4. TOLUENE SPIKE RESULTS FROM THE MOLD COOLING PROCESS  ... 4-6
TABLE 4-5. TOLUENE SPIKE RESULTS FROM THE SHAKE-OUT HOUSING
          PROCESS 	4-6
TABLE 4-6. SUMMARY OF FORMALDEHYDE SPIKE RESULTS AT THE CUPOLA
          BAGHOUSE INLET	4-7
                                  Vll

-------
                  TABLE OF CONTENTS (CONTINUED)
                                                             Page

TABLE 4-7. SUMMARY OF FORMALDEHYDE SPIKE RESULTS AT THE CUPOLA
         BAGHOUSE OUTLET	4-7
TABLE 4-8. SUMMARY OF TOLUENE SPIKE RESULTS AT THE CUPOLA
         BAGHOUSE INLET 	4-8
TABLE 4-9. SUMMARY OF TOLUENE SPIKE RESULTS AT THE CUPOLA
         BAGHOUSE OUTLET 	4-8
TABLE 4-10. COMPARISON OF FTIR SPECTRA OF SAMPLES FROM TOLUENE
         (60 ppm) CYLINDER TO EPA TOLUENE REFERENCE SPECTRA	4-9
TABLE 5-1. DETERMINATION OF FORMALDEHYDE STANDARD
         CONCENTRATION	5-6
TABLE 5-2. MEASURED ANALYTE CONCENTRATIONS AND MIXING FLOW
         RATES FOR THE SPIKE MIXTURES	5-8
TABLE 5-3. PROGRAM INPUT FOR ANALYSIS OF MOLD COOLING
         AND SHAKE-OUT HOUSING SAMPLE SPECTRA 	5-11
TABLE 5-4. PROGRAM INPUT FOR ANALYSIS OF BAGHOUSE
         INLET AND OUTLET SAMPLE SPECTRA  	5-12
TABLE 5-5. PROGRAM INPUT FOR ANALYSIS OF CTS SPECTRA
         AND PATH LENGTH DETERMINATION 	5-13
TABLE 5-6. RESULTS OF PATH LENGTH DETERMINATION	5-13
                               Vlll

-------
1.0 INTRODUCTION
1.1 BACKGROUND
The Emission Measurement Center (EMC) of the U. S. EPA received a request from the
Metals Group of the Emission Standards Division (ESD) and Source Characterization Group of
the Emission Monitoring and Analysis Division (EMAD), both in the Office of Air Quality
Planning and Standards (OAQPS), U, S. EPA, to perform emissions testing at iron foundries,
specifically on cupola emission control devices, as well as pouring, cooling, and shake-out
operations. The test program was performed in September, 1997 under Work Assignment 4-25,
under EPA Contract No. 68-D2-0165. This draft report was prepared under Work
Assignment 2-08, under Contract No. 68-W6-0048.
1.2 PROJECT SUMMARY
The cupola melting process is used to melt iron for casting into automotive and machine
parts. It is potentially a significant source of HAP emissions, including metal and organic
compounds. Emissions from the mold pouring, cooling, and shake-out are also potential sources
of HAP emissions.
The principal emission point at a cupola furnace is the exhaust from the furnace itself.
Emission controls for the Waupaca Plant No. 5 include a movable cap on the cupola, that seals
the charge, coarse grain separator, afterburner, drop out chamber, heat exchangers (recuperators),
dry calcium hydroxide injection system, pulse-jet baghouse, and stack. Cupola emissions testing
was conducted at the stack (outlet) and an inlet location to the baghouse to determine the
measurable emissions released during the melting process. Testing was also conducted at the
cooling line and shake-out housing ducts to determine the measurable emissions released during
the cooling and shake-out of the castings. Pouring operations had no emission capture or control
system; thus, no testing was conducted at the mold pouring location.
Three test runs were conducted at the cupola inlet and outlet locations over a three day
period simultaneously with manual method testing conducted by Pacific Environmental Services
(PES). One test run over a 4 hour (hr) period was conducted by FTIR only at both the cooling
line and shake-out housing locations. A summary of the FTIR results at the cooling and
shake-out housing locations is presented in Table 1-1. Emissions from the mold cooling and
shakeout housing included CO, methane, and ethylene. The emissions also contained a mixture
of heavier aliphatic hydrocarbon compounds. In the draft report the mixture of heavier

1-1

-------
hydrocarbons was represented by "hexane" because hexane and isooctane are the only aliphatic
hydrocarbons in the EPA library of HAP reference spectra. There are many hydrocarbon
compounds that are structurally similar to hexane and also have similar spectral features. After
the draft report was submitted, EPA directed MRI to measure quantitative spectra of some
additional hydrocarbon compounds. MRI selected candidate compounds that, based on their
infrared spectra in the region of the analyses, near 2900 cm"1, were likely to be components of
the sample mixture. MRI obtained commercially-prepared cylinder standards of butane, n-
heptane, pentane, 1-pentene, 2-methyl-l-pentene, 2-methyl-2butene, 2-methyl-2-pentene, and 3-
methylpentane. MRI then measured FTIR reference spectra of these compounds in the
laboratory. MRI also measured new high-temperature spectra of the HAPs hexane and isooctane.
Documentation of the new reference spectra and a brief description of the laboratory procedures
is presented in Appendix B.
The new spectra were used in revised analyses that gave the results presented in Table 1-1
and in Tables B-l and B-2. The new spectra made it possible to better represent the sample
mixture spectrum. Consequently hexane was not detected in mold cooling emissions, and was
only detected in one sample in the shakeout housing emissions. The reported hexane
concentrations are lower in the revised results because the spectrum of the sample hydrocarbon
mixture, which was represented by "hexane" in the draft results, is better represented by some of
the new spectra of other non-HAP hydrocarbons. In particular, 3-methylpentane and 1-pentene
were detected in cooling and shakeout housing process emissions. Butane and 2-methyl-2-
butene were also measured at the shakeout housing. The revised results give a more accurate
representation of the process emissions, but it's possible that other hydrocarbon compounds
could be measured in the emissions if their reference spectra were available.
The FTIR results from the cupola baghouse inlet and outlet locations are presented in
Table 1-2. Toluene was included in the analysis because this compound was spiked at the inlet
and outlet. Additional description of the results is in Section 4.
EPA Method 320 uses an extractive sampling procedure. A probe, pump, and heated line
are used to transport samples from the test port to a gas manifold in a trailer that contains the
FTIR equipment. Infrared spectra of a series of samples are recorded. Quantitative analysis of
the spectra was performed after the FTIR data collection was completed. All spectral data and
1-2

-------
results were saved on computer media for review after the test is completed.  A compact disk
containing all of the FTIR data was provided with the draft report.
      TABLE 1-1.  SUMMARY OF FTIR RESULTS FROM THE MOLD COOLING
                       AND SHAKE-OUT HOUSING DUCTS
Compound || Cooling Uncertainty || Shakeout Uncertainty
Toluene ppma
Ib/hr
kg/hr
Hexane ppm
Ib/hr
• kg/hr
Ethylene ppm
Ib/hr
kg/hr
Methane ppm
Ib/hr
kg/hr
Carbon Monoxide ppm
Ib/hr
kg/hr
Formaldehyde ppm
Ib/hr
kg/hr
3-Methylpentane ppm
Ib/hr
kg/hr
Butane ppm
Ib/hr
kg/hr
1-Pentene ppm
Ib/hr
kg/hr
2-Methyl-2butene ppm
Ib/hr
kg/hr
17.5 3.9
5.01
2.27
ND 30.6
13.3 0.8
1.158
0.525
178.5 2.6
8.859
4.017
402.3 28.1
34.95
15.85
ND 2.97
5.42 1.68
1.45
0.656
ND 34.90
17.9 3.89
3.88
1.76
ND 8.81
0.81 3.5
0.33
0.15
0.16 17.3
0.060
0.027
3.4 0.8
0.42
0.19
26.0 1.6
1.82
0.826
106.7 ' 19.1
13.06
5.92
ND 1.74
3.35 1.05
1.263
0.573
3.21 5.83
0.816
0.370
0.92 8.82
0.28
0.13
7.02 1.53
2.19
0.991
aAverage ppm concentration for the Run.
                                      1-3

-------
          TABLE 1-2.  SUMMARY OF FTIR RESULTS (ppm) AT THE CUPOLA BAGHOUSE INLET AND OUTLET3
Compound
HC1


Toluene


Methane


Formaldehyde

ppmc
Ib/hr
kg/hr
ppm
Ib/hr
kg/hr
ppm
Ib/hr
kg/hr
ppm
Ib/hr
kg/hr
9/8/97(12:02-17:07)
Inlet
33.5
5.2
2.4
ND


5.2
0.6
0.3
0.3
0.04
0.02
Uncb
3.3


2.9


1.2


1.7

Outlet
23.3
4.7
0.4
ND


4.7
0.7
0.1
ND

Unc
3.3


2.9


1.2


1.6

9/9/97(7:49-14:19)
Inlet
27.7
6.2
2.8
ND


5.3
0.6
0.3
ND

Unc
3.6


3.2


1.3


1.8

Outlet
16.3
4.6
0.6
0.4
0.3
0.1
4.9
1.1
0.1
ND

Unc
3.4


3.0


1.2


1.7

9/10/97(7:53- 14:19)
Inlet
29.7
6.7
3.0
ND


4.8
0.6
0.3
ND

Unc
3.3


2.9


1.2


1.7

Outlet
22.6
6.4
0.8
ND


4.7
1.0
0.1
ND

Unc
3.1


2.8


1.1


1.6

a PES did not complete a run on 9/8, but completed a manual run on 9/9 and two manual runs on 9/10. The PES flow data from 9/9 were used to calculate mass
  emission rates for the MRI runs on 9/8 and 9/9. The PES flow data from their first manual run on 9/10 were used to calculate emission rates for the MRI run
  on 9/10.
  Estimated uncertainty in ppm in the reported concentration.
c Average ppm concentration for the Run.

-------
1.3  PROJECT PERSONNEL
      The EPA test program was administered by the EMC. The Test Request was initiated by
the Metals Group of the ESD and the Source Characterization Group of the EMAD, both in
OAQPS. Some key project personnel are listed in Table 1-3.
                       TABLE 1-3.  PROJECT PERSONNEL
Organization and Title
Waupaca Foundry, Inc.
P.O. Box 249
311 S. Tower Road
Waupaca, WI 54981
Waupaca Foundry, Inc.
P.O. Box 189
9856 State Highway 66
Tell City, IN 47586
U. S. EPA, EMC
Work Assignment Manager
Work Assignment 4-25
U. S. EPA, EMC
Work Assignment Manager
Work Assignment 2-08
MRI
Work Assignment Leader
Work Assignment 4-25
Work Assignment 2-13
MRI
Work Assignment Leader
Work Assignment 2-08
Name
JeffLoeffler
Keith Tremblay
Michael K. Ciolek
Michael L. Toney
Thomas J. Geyer
John Hosenfeld
Phone Number
(715) 258-6629
(812)547-0700
(919)541-4921
(919)541-5247
(919)851-8181
Ext 3 120
(816)753-7600
Ext 1336
                                      1-5

-------
2.0 PROCESS AND CONTROL EQUIPMENT OPERATION
The material in Section 2 was prepared by Research Triangle Institute and provided to
MRI by the EMC. It was included in the report without MRI review.
2.1 INTRODUCTION
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. The plant operates one large cupola that
melts at a rate of about 60 tons/hr (tph), and operates four pouring lines. This section of the test
report provides a description of the cupola operation for iron melting, and the casting operation,
including pouring, cooling, and shake out.
2.2 PROCESS DESCRIPTION
2.2.1 Iron Melting in Cupolas
The Waupaca foundry in Tell City operates a large, water-cooled cupola that melts at a
rate of approximately 60 tph, with a blast rate of 10,000 to 15,000 standard cubic feet per
minute (scfm), which makes it a large cupola by U.S. industry standards. Figure 2-1 is a
simplified schematic of the cupola gas handling system and emission control equipment.
The cupola is charged with metal scrap, re-melt, coke, and limestone at the top of the
cupola, using one of two automated skip buckets. The level of metal within the cupola is
monitored, and the charge material in the skip bucket is dumped into the cupola when the level of
charge falls below a set level. The seal from the charge material and a draft on the cupola
prevent gases from escaping. If, for any reason, the charge material cannot be added to the
cupola within 5 minutes (min) of the level falling below the set point, the cupola will
automatically go "off blast" until the appropriate charge level in the cupola can be achieved.
The blast air is preheated to about 1,000°F in the blast air recuperator and is introduced
into the bottom of the cupola through 8 tuyeres. The blast is also enriched with oxygen under
certain melting conditions. The off gas from the cupola is removed at 250-300°F. The off-take
duct is lined with refractory material and leads to a coarse grain separator where heavy particles
are removed. The separator is cooled with non-contact water.
2-1

-------
     Charge
                              (3) Oil heat exchangers
                                                                       Exhaust Stack
                 Combustion
                 Chamber
   CUPOLA
                Solids Removed
                                Combustion Air
                                                                         Exhaust fan
               Figure 2-1. Simplified schematic of cupola gas handling system.
       After removal of the heavy particles, the gas enters a large combustion chamber where
combustion air is introduced and the CO is burned. Two burners are used when necessary to
maintain the combustion temperature. The gas leaves the combustion chamber at approximately
1650°F, and enters a dropout chamber where additional heavy particles are removed. The hot
gas then passes through an air-to-air heat exchanger (blast air recuperator), followed by a series
of three oil heat exchangers that are used to cool the air. These oil heat exchangers are not
currently used for heat recovery. (Modifications may be made in the future to recover and use
the  heat, such as for heating the building.)
       The gas from the heat exchangers is injected with a dry mixture (mostly calcium
carbonate and magnesium oxide) in a venturi mixer that increases the gas velocity and suspends
the  injected particles. Dry injection is used to improve pollutant removal in the baghouse.

                                          2-2

-------
During the source test, the dry powder injection was turned off to afford better characterization of
uncontrolled emissions and baghouse efficiency for HAP constituents. The gas is then sent to a
10 module, negative pressure baghouse that uses a high temperature fabric designed to withstand
temperatures of up to 320°F. The temperature of the gas at this point is typically 280 to 290°F.
A fan pulls the gas through the system and discharges the cleaned gases through a stack.
The pressure drop across the baghouse is monitored, and when the pressure drop
increases to 6 inches (in.) of water, individual bag house compartments are cycled off-line, the
bags are cleaned with a pulses of air, then the compartments are brought back on-line. The
baghouse uses plenum pulsing. During testing, each compartment was off line for approximately
8 min for cleaning, with cleaning pulses occurring approximately every 30 seconds (sec) during
this interval.
The plant routinely monitors several parameters associated with the cupola, including
blast air and oxygen rate, and afterburner air addition rates, as well as temperature at various
points in the process. The combined air flow rate through the recuperator and the baghouse
system is not directly monitored, but can be estimated from the blast air, oxygen and afterburner
air addition rates. During testing, the combined flow rate of offgas was also measured by the test
crew at the final stack sampling location. The plant also records the amount of each type of
material added to the cupola by the automatic skip buckets for each charge load. The
composition of a typical charge is given in Table 2-1 and contains approximately 4 tons of iron.
The iron includes remelt from the foundry, steel scrap, and pig iron.
TABLE 2-1. TYPICAL CUPOLA CHARGE MATERIALS
Material
Remelt from foundry
Steel scrap
Pig ironb
Silicon bricks
Blend bricks (Si, Mn, Cr)
Silicon carbide
Coke
Limestone
Typical range (lbs/charge)a
3,500 to 4,500
3,200 to 4,000
600 to 1,400
70 to 105
260 to 300
2 10 to 250
500 to 900
280 to 300
f* Typical range observed during the test days.
Remelt, steel scrap, and gray iron bricks are the sources of iron and total 8,700 Ibs (4.35 tons) per charge.
2-3

-------
2-2.2 Pouring, Cooling and Shakeout
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 warmbox cores. Some of the company's cast
products use cores, and others do not. During the source test, cores were not being used on any
of the lines. The lines are all similar except that Line 4, which is designed to handle larger
molds, has an automatic pouring station and the other lines do not. Line 4 is typically used for
casting the larger size parts.
Pouring emissions are not captured at any of the four pouring stations. Cooling emissions
are captured by hoods that cover the entire cooling line prior to shakeout. The shakeout
operation is totally enclosed and evacuated to capture the emissions. After shakeout, the parts
are transferred to a casting cooling house where they are placed on a metal "tree." The parts then
proceed to a "spinner house" and are shot blasted to remove residual sand. The spinner house is
also evacuated to the duct that removes emissions from shakeout. The captured emissions from
shakeout and cooling are sent to a baghouse for gas cleaning. There are three baghouse systems;
each system predominantly receives emissions vented from a single line, but a few of the vents
from a given line are routed to another line's baghouse system. Consequently, controlled
emissions represent contributions from multiple lines and multiple processes.
The ductwork for the cooling lines are interconnected with either other cooling lines or
shakeout enclosure ductwork. Therefore, it is impossible to get a representative sample for
cooling emissions that could be attributed to an entire cooling line. The least amount of
interconnection was on Line 4, so it was selected for emissions testing. The first third of the
cooling section of Line 4 was ducted to a single vent that had a long, straight vertical section
before connecting with other ductwork. Ports were installed in this straight section of the vent,
so that the uncontrolled emissions from the first third (approximately 20-25 min) of the cooling
line could be measured. The shakeout enclosure ductwork has a short vertical rise, then elbows
to a horizontal section where it is tied to the vent from the spinner housing and the last hood from
the cooling line. Ports were installed in the short vertical duct from the shakeout enclosure
approximately one foot (ft) above the roof of the enclosure prior to the point where the ducts
from the spinner house and the end of the cooling line join the shakeout enclosure duct. This
point represents uncontrolled emissions from the shakeout operations.

2-4

-------
2.3 SUMMARY OF PROCESS OPERATING DATA COLLECTED DURING SOURCE TEST
2.3.1  Process Operating Data for Cupola Melting Operations
       Testing of the melting operations was conducted over a three-day period. Single test runs
were performed on September 8th and 9th, and two runs were performed on September 10th.
During testing, process information was collected from the operating room's computer control
panel.  Process information collected included cupola charging data, process chemistry, gas flow
rates, temperatures, baghouse pressure drop, and cupola stack opacity.
       Table 2-2 and Figure 2-2 present metal charging rates for the cupola during the three days
of testing. Table 2-2 and Figure 2-2 show that the average metal production rate for
September 8th was higher than on the 9  or 10 .  Table 2-2 and Figure 2-2 also show that hourly
production rates varied significantly within a given day.
                                         2-5

-------
TABLE 2-2. SUMMARY OF CUPOLA CHARGING DURING THE TEST DAYS
9/08/97
Time period




11:02-11:57
12:00-12:54
13:00-13:58
14:03-14:55
15:00-15:57




Average rate
(tons/hr)
Tons
charged




45
56
59
51
55




53.2
9/09/97
Time period

8:01-8:56
9:01-9:56
10:02-10:46
11:03-11:50
12:00-12:51
13:00-13:56
14:06-14:55





Average rate (tons/hr)
Tons
charged

47
52
42
35
26
58
30





41.4
9/10/97
Time period
7:01-7:56
8:03-8:54
9:01-9:59
10:04-10:55
11:01-11:57
, 12:03-12:59
13:03-13:41
14:02-14:58
15:01-15:28
16:13-16:56
17:00-17:56
18:01-18:57
19:02-19:58
Average rate
(tons/hr)
Tons
charged
48
35
49
47
54
59
40
' 35
32
45
51
56
55
46.6
                            2-6

-------
to
• 8-Sep H9-Sep DIG-Sep
i\j -
KC\ -
"^
o
i- en
| °°
|
« 40
2
O)
~ 30
0)
E
•c 20
3
O
10
0 -
)

--J
8
oo
o
o

H —

CD
8
cb
CD
o

H —

=
-
CO
8
o
8

=
o
8
6
o

-tJ
mm
8
NJ
-------
Table 2-3 presents average gas flow rates and temperatures for several locations in the
cupola flue gas system on September 8th, 9th, and 10th. As can be seen from Table 2-3, the
average blast rate was lowest on September 9th, and the cupola typically operated with oxygen
addition on that day. On September 8th and 10th, the average blast rate was higher, and oxygen
addition was not used. The average baghouse pressure drop was lower on September 9th than on
September 8th or 10th, (2.8 in. of water versus 4.4 in. of water), and the average opacity was
higher on September 9th than September 10th (5.7 percent versus 2.8 percent).

TABLE 2-3. PROCESS DATA DURING THE DAYS OF CUPOLA TESTING
Process Parameter
Average Value on Testing Date
September 8
September 9
September 10
Cupola process air flow information
Blast rate (scfm)
Oxygen addition (on/off)
Temperature in (F)
Temperature out (F)
14,794
off
1,156
330
9,131
on
1,067
232
13,665
off
1,101
297
Afterburner air flow information
Primary air (scfm)
Secondary air (scfm)
Cooling air (scfm)
Temperature out (F)
5,271
5,893
2,794
1,717
2,764
2,823
2,300
1,639
4,969
5,476
2,792
1,668
Baghouse information
Temperature in (F)
Pressure drop (inches H2O)
Opacity (%)
297
4.4
Not Recorded
281
2.8
5.7t
299
4.4
2.81"
^Average of opacity readouts recorded every 15 minutes; opacity readouts are 6-minute averages from KVB EPA-2
stack mounted opacity monitor.
Continuous records of blast air flow rates are presented in Figure 2-3. Blast air rates were
significantly reduced (i.e., the cupola was placed "on relief) for varying lengths of time on
September 9th and 10th. The specific times when the cupola was on relief are listed in Table 2-4.
2-8
-------
TABLE 2-4, PERIODS WHEN CUPOLA WAS "ON RELIEF" DURING TESTING
September 8
Continuous blast.

September 9
10:54-11:09

September 10
13:51 - 14:04
14:11-14:28
15:27 - 16:03
16:06 - 16:08
Table 2-5 presents average process iron chemistry values (from cupola) for September 9th
and 10 . Although process chemistry values were not recorded on September 8 , average trace
metal impurity levels were typical on all three days.
2-9
-------
18000
16000
8-September
- - - 9-September
10-September
2000
6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 0:00
Clock time (hours:minutes)

FIGURE 3-3. BLAST AIR FLOW RATE THROUGH CUPOLA.
Figure 2-3. Blast air flow rate through cupola.
-------
TABLE 2-5. PROCESS IRON CHEMISTRY AT CUPOLAa
Element
ElCe
(Elemental Carbon Equivalents)
CalcC
(Calculated Carbon)
Si
Mn
P
S
Ni
Mo
Cr
Cu
Al
Ti
Sn
Mg
V
Pb
Concentration (%)
September 9
3,665
2.470
1.050
0.350
0.055
0.070
0.045
0.105
0.205
0.114
0.008
0.009
0.005
0.006
0.006
0.001
September 10
4.045
3.450
4.490
0.595
0.042
0.090
0.090
0.025
0.240
0.230
0.009
0.011
0.009
0.001
0.011
0.002
a The process chemistry values reported here are considered typical; the process chemistry values were not
specifically recorded on September 8, but were also considered typical.
2.3.2 Process Operating Data for Pouring, Cooling and Shakeout Operations
Emissions from cooling (first third section) and shakeout were measured on
September 5,1997. Line 4 employs an automated molding machine. Testing was conducted
only when the entire mold line was filled with recently poured molds. During the day of the
source test of the cooling and shakeout operations, Line 4 was used to cast brake drums. Each
mold produced two brake drums, and used 189 pounds (Ib) of poured metal. For the test day,
249 molds per operating hour were produced. Each mold contains 1393 Ib of green sand (lake
sand, sea coal, and bentonite), so the molds had a sand to metal ratio of 7.35:1.
The properties of the molding sand measured during the test day are given in Table 2-6.
A bonding agent was added to the sand in the amount of 38.1 Ib of bond per ton of sand mulled.
2-11
-------
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.
TABLE 2-6. TYPICAL RESULTS FROM GREEN SAND ANALYSIS
Property
Moisture (%)
Clay (%)
Loss on ignition (%, at 1800T)
Volatile content (%, at 900 °F)
Value
3.5
8.7
7.8
4.0
2-12
-------
3.0 TEST LOCATIONS AND GAS COMPOSITION
Figure 3-1 is a schematic showing an overview of the cupola gas handling system that
presents the locations of both cupola test points. The baghouse inlet (location "A") and the outlet
stack (location "B") were sampled concurrently.
Figure 3-2 is a schematic showing a closer view of the cupola test locations. Location
"A" was at the duct leading to the baghouse, and location "B" was at the exhaust stack following
the baghouse.
Figure 3-3 is a schematic with a view of test locations at the mold cooling and mold
shake-out housing. The sample location of the mold cooling line was the collector duct for the
first several cooling line vents. The shake-out housing was sampled at the duct drawing from the
enclosure.
3,1 BAGHOUSE OUTLET - STACK
The test ports on the stack are located at about 70 ft above ground level. Access to the
stack ports is at roof level, which can be reached by a ladder on the baghouse. Test ports on the
7 ft 9 in.-diameter stack allowed for concurrent FTIR and manual sampling.
3.2 BAGHOUSE INLET DUCT
The test ports on the inlet duct are located at roof level about 70 ft above the ground.
Ports allowed simultaneous testing by both FTTR and manual methods. The baghouse inlet and
outlet were sampled by FTIR concurrently using a dual line extractive sampling system.
3.3 MOLD COOLING LINE
Two ports were utilized on the vertical duct that collects the emissions from the first
seven take-off vents over the cooling line conveyer following the pouring station. That
vertically-oriented duct is inside the main facility. Sampling was conducted at a height of
approximately 30 ft above the facility floor to obtain volumetric flow, diluent, moisture, and
FTIR data across the diameter of the duct interior. A dual-line system was used to conduct
concurrent testing at the mold cooling line and shake-out housing.
3-1
-------
Charge Movable
I cap
Oil heat exchangers
Cupola
Exhaust
stack
Air-Air heat exchanger
Combustion
chamber
Heated
blast air
m Dry
injection
(Baghouse
inlet)
Blast air
Baghouse
Solids
removed
Cupola blower