v>EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 79-ISC-8
December 1979
Air
Industrial Surface
Coating
(Can Coating)
Emission Test Report
Metal Container
Corporation
Jacksonville, Florida
-------
INDUSTRIAL SURFACE COATING
(Can Coating)
EMISSION TEST REPORT
METAL CONTAINER CORPORATION
JACKSONVILLE, FLORIDA
Prepared For:
Mr. R.T. Harrison
Tech. Manager
ESED/EMB
Office of Air Quality Planning
and Standards
Environmental Protection Agency
Contract No. 68-02-2820
Work Assignment 14
Prepared By:
Samuel S. Cha
Task Manager
TRC Project 1198-E80
-------
TABLE OF CONTENTS
SECTION
1.0 INTRODUCTION
2.0 CONCLUSIONS
3.0 PROCESS DESCRIPTION
4.0 METHODOLOGY
4.1 Flow Rate Measurement
4.2 Volatile Organic Compound (VOC) Measurements -
TGNMO Manual Method .
4.3 Volatile Organic Compound (VOC) Measurements -
Flame lonization Analyzer (FIA)
4.4 • Material Balance
4.4.1 Can Counting Data -.'-..
4.4.2 Coating Usage
4.4.3 Coating Applied
5.0 RESULTS AND DISCUSSIONS
5.1 Flow Rate Measurement
5.2 VOC Measurements - TGNMO Manual Method
5.3 VOC Measurement - FIA Method
5.4 Coating Usage (Coating Sprayed) and Can Counting
5.5 Coating Applied
5.6 Material Balance
PAGE
1
2
4
6
6
7
7
9
9
10
11
11
11
14
14
17
17
APPENDICES
B
C
D
PROCEDURE FOR MEASUREMENT OF TOTAL GASEOUS NONMETHANE
ORGANIC COMPOUNDS - EPA REFERENCE METHOD 25
PROCEDURE FOR DIRECT MEASUREMENT OF TOTAL
GASEOUS ORGANIC COMPOUNDS USING A FLAME
IONIZATION ANALYZER
FLOW RATE MEASUREMENT RESULTS
VOC EMISSION MEASUREMENT RESULTS USING EPA
REFERENCE METHOD 25 (TGNMO)
VOC EMISSION MEASUREMENT RESULTS USING FLAME
IONIZATION ANALYZER
1. COATING USAGE (COATING SPRAYED) TEST RESULTS
2. CAN COUNT RESULTS
COATING APPLIED WEIGHING RESULTS
-------
LIST OP FIGURES
FIGURE
3-1
4-1
TABLE
5-1
5-2
5-3
5-4
5-5
PAGE
EMISSION CONTROL SYSTEM 5
INTEGRATED GAS SAMPLE APPARATUS . 8
LIST OF TABLES
PAGE
INTERIOR COATING OPERATION, TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA - OCTOBER, 1979 12
SUMMARY OF VOC EMISSION MEASUREMENT DATA - TGNMO
INTERIOR SPRAY COATING METHOD. TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA - OCTOBER, 1979 13
COMPARISON OF TGNMO vs TIA MEASUREMENT RESULTS
INTERIOR COATING OPERATION. TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA - OCTOBER, 1979 15
SUMMARY OF COATING USAGE DATA AND CAN COUNTING
DATA, INTERIOR SPRAY COATING. TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA - OCTOBER, 1979 16
SUMMARY OF VOC MATERIAL BALANCE (PER 1000 CANS)
INTERIOR SPRAY COATING. TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA - OCTOBER, 1979....... 18
-------
1.0 INTRODUCTION
As Work Assignment 14 of Contract 68-02-2820, the Office of Air Quality
Planning and Standards, Environmental Protection Agency, (EPA-OAQPS) has as-
signed TRC Environmental Consultants, Inc. to perform a Volatile Organic Com-
pound (VOC) material balance study of an interior spray coating operation in
a two-piece can manufacturing plant. Results from this study will provide
EPA-OAQPS with additional background information needed to develop New Source
Performance Standards (NSPS) for the two-piece can manufacturing industry.
A two-piece can interior spray coating operation at the Metal Container
Corporation at Jacksonville, Florida, manufacturing plant, was selected by
EPA-OAQPS as the site for the material balance study. Can manufacturing line
number 4 was selected as the process to be tested because of its accessibili-
ty. The tests were conducted during the week of October 16, 1979 by TRC
staff along with Mr. William King of Research Triangle Institute, also an EPA
.contractor, and EPA staff member Mr. R.T. Harrison.
A total of six material balance tests were performed. For each test,
total coating-solvent usage, coating applied per can, number of cans coated,
and volatile organic compound emissions from three exhausts were quantified.
Volatile organic compound material balances for the process were then de-
rived. This report summarizes TRC's approach, test results and conclusions.
-1-
-------
2.0 CONCLUSIONS
Test results of four of the six sets of data collected indicate a range
of from 58 to 97 weight percent of the total solvent used in the coating to
be collected from three principal locations (the oven, the elevator, and the
baghouse exhaust). Another source, the oven quench (cooling zone) exhaust,
was not quantified. Therefore, the range of emissions presented is conserva-
tive; since the oven quench exhaust is omitted. For two of the six sets of
data, VOC emissions from these three locations appeared to be greater than
the theoretical amount of solvent in the coating used. This was primarily a
result of the increased VOC concentration measured at the oven exhaust. No
explanation for this increase in concentration is apparent. The amount of
solvent in the coating is based on the manufacturer's specification that 80
weight percent of the coating is solvent.
The weight percentage of the total solvent used in the coating emitted as
VOC emission from the main oven exhaust (including the oven bypass exhaust)
ranged from 40 to 120 percent when expressed as propane. If Runs 5 and 6 are
omitted, then the range is 40 to 85 weight percent. Because these data are
derived from the TGNMO data which expresses VOC emissions as carbon, propane
was chosen as the surrogate compound to estimate the molecular weight to car-
bon ratio of the VOC being emitted.
The weight percent of the total solvent used in the coating emitted as
VOC emissions from the enclosed elevator exhaust ranged from 11 to 16 percent
for the six runs when expressed as propane.
The weight percent of the total solvents used in the coating emitted as
VOC emissions from the baghouse exhaust ranged from 16 to 30 percent for five
of the six runs, when expressed as propane. Data for Run 6 are not obtain-
ed. It is assumed that the baghouse was installed to remove entrained coating
-2-
-------
spray ,and that the spraying operation is a continuous operation, therefore,
the portion of the VOC emission from the baghouse exhaust due to the drying
of the coating in the baghouse is constant.
-3-
-------
3.0 PROCESS DESCRIPTION
The can manufacturing operations tested represent a segment of the overall
two-piece can manufacturing process. The overall process begins with aluminum
coil which is continuously fed to a cupper press that stamps a blank from the
coil and forms a shallow cup. Next, the cups are extruded to final size,
wall-ironed, washed and dried. A reverse roll coater then applies a base coat
to the can exterior which is cured at 190°C (370°F) in a gas-fired oven.
After the can cools, the exterior top coat is applied and cured in a similar
•' >i:.
manner. Finally, the can receives an 80% solvent, 20% solids (by weight)
interior spray coat which is cured in a flat bed, gas-fired oven. Emissions
from the base coat oven, top coat printer and oven, and the interior spray
coater and oven are ducted to a thermal oxidizer (incinerator) maintained at
620°C (1150°F). Figure 3-1 depicts the process ducting and emission
control system.
This study focused upon the emissions from the interior spray coating
operation which includes the can transfer conveyor and flat bed oven. The
overspray emissions from the interior spray coating are exhausted through a
baghouse to remove particulates. The coated cans are transferred from the
spray area to the flat bed oven through an enclosed elevator-conveyor system.
Following the oven, the cans enter a cooling zone which is supplied with
ambient air. The air is drawn from outside the building, passes through a
heat exchanger, and is exhausted back to the atmosphere. Emissions from the
elevator-conveyor enclosure, flat bed oven, and the baghouse discharge are
ducted to the thermal oxidizer.
-4-
-------
THERMAL OXIDIZER,
DAMPER CLOSED
n i •
r-\f
n
\
D
MASHER
n
fH
-.
1
N
Jp
PRTNTFR1 •;
CHAIN PIN OVEN
• -.
.
r^FI A
(J rLn
07
'
BY PASS STACK
NOT IN USE
DAMPER
CLOSED
BASE COATER'S
PIN OVEN
WASHER
PIN OVENS
BASE COATER
1
PRINTER
TOP VIEW
BED OVEN
ELEVATION
FLAT BED OVEN
V
INTERIOR COATER
BAGHOUSE
1 EXHAUST DUCT OF INTERIOR COATER BAGHOUSE 5 DOWN-STREAM OF THERMAL OXIDIZER
2 EXHAUST DUCT OF FLAT BED OVEN 6 OVEN BY-PASS STACK
3 EXHAUST DUCT AT ELEVATOR 7 OVEN COOLING ZONE AIR INTAKE
4 COMBINED DUCT OF 2 AND 3 AND QUENCH STACKS
FIGURE 3-1: EMISSION CONTROL SYSTEM
-------
4.0 METHODOLOGY
4.1 Flow Rate Measurement
Flow rate measurements were conducted at the following locations!
o Baghouse exhaust duct
o Flat bed oven exhaust duct
o Elevate conveyor enclosure duct (Elevator duct)
o Combined duct of the flat bed oven exhaust duct and the elevator duct.
o Oven by-pass stack (on the roof) °
o Cooling zone quenching stack (on the roof)
A standard pitot tube and an inclined manometer were used in accordance
with EPA Reference Methods 1 and 2 to determine flow rates. The exhaust air
temperatures were measured using a thermocouple and potentiometer.
4.2 Volatile Organic Compound (VOC) Measurements - TGNMO Manual Method
VOC were measured using EPA Method 25 - "Determination of Total Gaseous
Non-Methane Organic Compounds as Carbon-Manual Method" (see Appendix A) , also
known as the TGNMO Manual Method. Each sample was drawn from the stack
through a chilled condensate trap by means of an evacuated gas collection
tank. Total gaseous non-methane organics (TGNMO) were determined by combining
the analytical results obtained from independent analysis of the condensate
trap and evacuated tank fractions. After sampling was completed, the organic
materials collected in the condensate trap were oxidized to carbon dioxide and
quantitatively collected in an evacuated vessel; a portion of the carbon
dioxide was reduced to methane and measured by a flame ionization detector
(FID) . A portion of the sample collected in the gas sampling tank was
injected into a gas chromatographic (GC) column to achieve separation of the
non-methane organics from carbon monoxide, carbon dioxide and methane. The
non-methane organics were oxidized to carbon dioxide, reduced to methane and
measured by FID.
-6-
-------
TRC's subcontractor - Pollution Control Science, Inc., Miamisburg, Ohio,
performed the TGNMO method analysis of the collected samples.
4.3 Volatile Organic Compound (VOC) Measurements - Flame lonization Analyzer
(FIA)
VOC were measured using a modification of "Direct Measurement of Total
Gaseous Organic Compounds Using a Flame lonization Analyzer" (See Appendix
B) . This method is known as the FIA or FID method. The method specifies that
the sample be continuously drawn from the source through a heated sample line
and glass fiber filter to a flame ionization analyzer. Ions formed through
the combustion of a specific volatile organic compound in a H2-0_ flame
establish a current that is proportional to the mass flow rate of the volatile
organic at two polarized electrodes. The electrode current is measured on a
potentiometric recorder and compared with a calibration curve based on propane
(C3H8).
Due to the limited number of analyzers available, physical constraints of
sampling locations, and restrictive safety measures, an integrated bag sampl-
ing technique was used instead of continuous monitoring. Figure 4-1 shows the
sampling apparatus. Samples collected in the Tedlar bags were then measured
by the flame ionization analyzer. At sources where positive pressures were
observed, the Tedlar bags were filled without the sampler. In those cases,
needle valves were used for flow control.
4.4 Material Balance
A material balance was performed for comparion of the sampled VOC emission
data, the coating usage and the coating mass (weight) remaining on the can
after curing. Ideally, if all coating-solvent lost were captured and there-
fore sampled, the difference between the coating usage and the coating-solvent
—7—
-------
STACK WALL
FILTER
(GLASS WOOL)
PROBE
QUICK
CONNECTS
FEMALE
TEDLAR OR
ALUMINIZED
MYLAR BAG
TEFLON SAMPLE LINE-
VACUUM' LINE
NEEDLE VALVE
FLOW METER-
CHARCOAL TUBE
PUMP
RIGID LEAK-PROOF
CONTAINER
FIGURE 4-1: INTEGRATED-BAG SAMPLING TRAIN. (MENTION OF TRADE NAMES
OR SPECIFIC PRODUCTS DOES NOT CONSTITUTE ENDORSEMENT BY THE
ENVIRONMENTAL PROTECTION AGENCY.)
-8-
-------
recovered should be minimal. Mathematically, it could be expressed as:
coating usage = coating remaining on cans + sampled emission.
Realistically, there is always some coating-solvent that cannot be recovered.
The material balance data have been expressed in terms of Kg(lb)/1000 cans.
4.4.1 Can Counting Data
At the beginning and end of each test run, and also the beginning and end
of each test day, can counts were recorded from the plant's computerized pro-
duction monitoring system. The exact number of cans processed by line 4 was
recorded and a total plant production determined from this data and the rela-
tive operating conditions of the other three lines. The number of cans pro-
cessed on line 1 (which processes 16 ounce cans) was converted to an equiva-
lent number of 12 ounce cans by multiplying the recorded number of cans by the
ratio of the interior surface area of the two sizes of cans.
4.4.2 Coating Usage
Coating solvent usage was determined by measuring the liquid level in the
coating feed tank at the beginning and end of each test run. The level dif-
ferences were converted to volume used and the coating usage weight determined
by multiplying the volume by the coating density, 7.9 Ib/gal. In addition to
the coating usage per test run, daily coating usages were also determined to
enlarge the data base.
Because the interior spray coating operations for all four production
lines are fed from the same coating tank, the coating usages determined repre-
sent those for the entire plant. Line 4 coating usage was determined using a
factor based on that line's proportion of the total plant can production.
-9-
-------
4.4.3 Coating Applied
During each test run, four cans were numbered and an initial mass
determined by weighing. The cans were placed on the conveyor and passed
through the spray coater and oven. After exiting the oven, the cans were
retrieved and weighed. The difference between the initial and final masses
represents the coating applied to the can.
-10-
-------
5.0 RESULTS AND DISCUSSIONS
5.1 Flow Rate Measurement
Stack and duct flow rate measurement data are listed in Appendix C, and
summarized in Table 5-1.
The measured oven exhaust duct flow rate was significantly lower than the
design value. An investigation on October 18, 1979 revealed that the damper
at the oven by-pass stack was not completely closed. The Plant engineer was
notified but no immedidate adjustments were made. It should also be noticed
that the oven exhaust duct flow rate in Table 5-2 were calculated as the dif-
ference between the measured flow rate of combined ducts from oven and eleva-
tor, and the measured flow rate of elevator duct. Because of the proximity of
a duct elbow and exhaust fan, no accurate velocity traverse could be conducted.
5.2 VOC Measurements - TGNMO Manual Method
The results of the VOC measurement using the Total Gaseous Non-Methane
%
Organic Method are summarized in Table 5-2. The complete measurement results,
are presented in Appendix D. VOC emissions are expressed as equivalent carbon
and propane concentrations. The mass emission rates are calculated using pro-
pane as the surrogate compound to estimate the mass of the VOC emissions. For
each run emission rate at the oven by-pass stack is estimated using the air
flow rate measured on October 18, and the VOC concentration measured during
the test run at the oven exhaust. The total emission rate at the interior
spray coating area, is calculated by adding the emission rates from the bag-
house exhaust duct, the conveyor-elevator enclosure exhaust duct, oven exhaust
duct and oven by-pass stack, and expressing it on a 'per 1,000 can1 basis.
These total emission rates ranged from 0.247 kg (0.544 Ib) to 0.611 kg (1.351
Ib) and averaged 0.401 kg (0.885 Ib).
-11-
-------
TABLE 5-1
INTERIOR COATING OPERATION, TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA
OCTOBER, 1979
SOURCE
BAGHOUSE
DATE
10-16-79
10-17-79
10-18-79
FLOW RATE
(SCFM)
914
972
908
THERMO OXIDIZER
EXHAUST STACK
10-16-79
9479
OVEN EXHAUST DUCT
QUENCH STACK
BY PASS STACK
OVEN(ON ROOF)
ELEVATOR VENT
ELEVATOR AND
OVEN DUCT
10-16-79
10-18-79
10-18-79
10-18-79
10-18-79
10-16-79
10-17-79
10-17-79
10-16-79
10-17-79
10-18-79
532
1792
2005
2629
3170
200
183
190
824
758
654
-12-
-------
TABLE 5-2
SUMMARY OF VOC EMISSION MEASUREMENT DATA - TGNMO
INTERIOR SPRAY COATING METHOD, TWO-PIECE CAN PLANT
JACKSONVILLE. FLORIDA
OCTOBER, 1979
Sampling Period
Sample l.D.
Ho. of Can Coated Flow Ra*e
During the Period
Line 4 SCFM
M3/Min
Mass Emission Rate
VOC Concentration Measured Per Sampling Period
as Carbon as Propane as Propane
ppm
Mg/L
ppm
Mg/L
Kq
Ib
Mass Emission Rate
Per 1000 Cans
as Propane
Kg
Ib
JO/17/79
Run 1
Run 2
Run 3
1045 -1120 (35 min.)
Baghouse Duct
Elevator Duct
Oven Duct*3
Oven By-pass Stack* I
1351-1425 (34 min.)
Baghouse Duct
Elevator Duct
Oven Duct*3
Oven By-pass Stack*^
1602-1635 (33 min.)
Baghouse Duct
Elevator Duct
Oven Duct*3
Oven By-pass Stack* 1
21263
972
183
573
2,629
19826
972
190
583
2629
24683
972
190
583
2629
27.5
5.2
16.3
53.7
27.5
5.4
16.5
53.7
27.5
5.4
16.5
53.7
4515
11881
2450
2450
2741
8897
1844
1844
3053
12178
1682
1682
2.2555
5.933
1.223
1.2a3
1.368
4.444
0.921
0.921
1.524
6.080
0.840
0.840
1505
3960
817
817
914
2632
615
615
1018
4059
561
561
2.755
7.251
1.495
1.495
Sub total
1.672
5.432
1.126
1.126
Sub total
1.863
7.431
1.027
1.027
Sub total
2.652
1.314
0.852
3.893'
8.711
1.564
0.993
0.632
2.848
6.037
1.691
1.319
0.559
2.522
6.091
5.848
2.898
1.878
8.584
19.208
3.448
2.190
1.393
6.281
13.310
2.729
2.907
1.233
5.561
13.431
0.125
0.062
0.040
0.183
0.410
0.079
O.OSO
0.032
0.144
0.305
0.069
0.053
0.023
0.102
0.247
0.276
0.137
0.088
0.404
0.903
0.174
0.110
0.070
0.317
0.671
0.151
0.118
0.050
0.225
0.544
10/18/79
Run 4
1510-1545 (35 min)
Baghouse Duct
Elevator Duct
Oven Duct*3
Oven By-pass Stack
24225
908
166
488
3)70
25.7
4.7
13.8
53.7
4603
11981
1B41
1841
7.299
5.983
0.920
0.920
1534
3994
114
614
2.810
7.313
1.124
1.124
Sub total
2.527
1.202
0.543
3.529
7.801
5.572
2.651
1.198
7.782
17.201
0.104
0.050
• 0.022
0.145
0.322
0.230
0.109
(1.049
0.321
0.710
10/19/79
Run 5
Run 6
1010-1047 (37 min.)
Baghouse Dunt
Elevator Duct
Oven Duct*3
Oven By-pass Stack
1330-1405 (35 min.)
Baghouse Duct*^
Elevator Duct
Oven Duct*3
Oven By-pass Stack
22,520
908
166
488
3170
23320
108
166
488
1170
25.7
4.7
13.8
53.7
25.7
4.7
13.8
03.7
3551
13740
3734
3734
-
12185
4938
4398
1.773
6.861
1.86S
1.865
6.085
2.466
2. 466
1,184
4S60
1245
1245
4062
1646
1646
2.167
U.386
2.279
2.279
Sun total
7.417
3.014
3.014
Sub total
2.060
1.458
1.165
7.365
2.248
(2.060)
1.223
1.457
9.464
14.204
4.543
3.214
2.568
16.680
27.005
(4.543)
2.696
1.212
20.867
31.318
0.091
0.065
0.052
0.336
0.544
(0.091)
0.052
0.062
0.406
0.611
0.202
0.143
0.114
0.741
1.199
(0.202)
0.116
0.138
0.895
1.351
•1 10/18/79 flow rate data was used.
*2 Run 5 VOC concentration data was used.
*3 Flow ratp measurement obtained calculated as fhe difference hot-.ween the measured comhinc-J flow rato of the oven and elevator duct and the
-------
It should be noted that the VOC concentration measured at the oven exhaust
duct on October 19, 1979 during runs 5 and 6, show a large degree of inconsis-
tency with those measured on October 17 and October 18. No explanation could
be suggested. (See further discussion in Sections 5.3 and 5.6).
5.3 VOC Measurement - FIA Method
Results derived using the FIA method are summarized in Tables 5-3 and com-
plete results are presented in Appendix E. Samples taken at the conveyor-ele-
vator enclosure duct exceeded the instruments upper measurement limit for five
of the six runs.
Table 5-3 also compares TGNMO results versus FIA results. For samples
taken at the baghouse exhaust duct, results are in good agreement between the
two methods. For the oven samples, due to the condensation problem in the FIA
bag samples, TGNMO results are approximately 1.7 to 2.0 times higher than FIA
results. In Runs 5 & 6, this discrepancy is five times higher. This suggests
«
that some unknown factors may be involved in Runs 5 and 6.
5.4 Coating Usage (Coating Sprayed) and Can Counting
Coating usage was determined by measuring the levels in the coating feed
tank before and after each run, and also at the beginning and at the end of
each testing day. Results of the coating usage are summarized in Table 5-4.
Also included in Table 5-4 are the number of cans processed at the inter-
ior spray operation for the entire plant and for line 4, where the emission
tests were conducted, for each test run and for each testing day. Based on
that information, coating sprayed (coating usage) per 1000 cans was
-14-
-------
TABLE 5-3
COMPARISON OF TGNMO VS FIA MEASUREMENT RESULTS
INTERIOR COATING OPERATION, TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA, OCTOBER, 1979
SAMPLE 1
VOC MEASUREMENT
TGNMO
FIA
PPM as Propane
RATIO
TGNMO/FIA
Run 1 (21, 263)
Baghouse Duct
Elevator Duct
Oven Duct
Run 2 (19,826)
Baghouse Duct
Elevator Duct
Oven Duct
1505
3960
817
914
2632
615
1511
4500
470
093
3011
299
0.996
0.880
1.738
0.920
0.874
2.057
Run 3 (24,683)
Baghouse Duct
Elevator Duct
Oven Duct
Run 4 (24,225)
Baghouse Duct
Elevator Duct
Oven Duct
Run 5 (22,520)
Baghouse Duct
Elevator Duct
Oven Duct
1018
4059
561
1534
3994
614
1184
4560
1245
1251
3696
332
1412
3151
333
1060
3040
379
0.814
098
690
086
268
1.844
1.117
1.500
3.285
Run 6 (23,320)
Baghouse Duct
Elevator Duct
Oven Duct
4062
1646
1412
3151
333
1.289
4.943
-15-
-------
TABLE 5-4
SUMMARY OF COATING USAGE DATA AND CAN COUNTING DATA
INTERIOR SPRAY COATING, TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA
OCTOBER, 1979
CT>
I
Sampling Period
Start End
10/16/79 1330:00 1630:00
10/17/79 0817:00 1600:00
0817:00 1427:00
1042:15 1126:20
1349:00 1427:00
1559:00 1637:00
10/18/79 1410:00 1735:00
1508:15 1548:45
10/19/79 0845:00 1405:00
1008:45 1048:25
1327:00 1404:30
Time
Interval
(min)
180
463
370
44.1
38.2
37.7
205
40.5
320
39.7
37.5
Ho. of 12 Ounce
Equivalent Can Sprayed
From Can Counting (1)
total
4 lines
490531
056400
701516
95465
74008
53463
473697
114147
852435
108245
99602
Line
4
139531
281302
234154
26791
22275
28199
143997
28032
182731
24163
24986
% in.
Total
28%
33%
33%
28%
30%
53%
30%
25%
21%
22%
25%
Interior Coating
Density = 7.9
Total (Measured)
Gal Ib kg
65.10 514.
127.76 1009.
107.85 852.
14.09 111.
11.17 88.
(8.00) (63.
68.49 541.
15.05 118.
119.02 940.
15.55 122.
13.60 107.
3
3
0
3
2
2)
1
9
3
9
4
233.2
457.7
386.4
50.5
40.0
28.7
245.4
53.9
426.4
55.7
48.7
Sprayed
Ib/gal
Line 4 (Calcul.)
Gal Ig kg
18.23 143.9 65.3
42.16 333.1 151.1
35.59 281.2 127.5
3.95 31.2 14.1
3.35 26.5 12.0
4.24 33.5 15.2
20.55 162.3 73.6
3.76 29.7 13.5
24.99 197.5 89.6
3.42 27.0 12.3
3.40 26.9 12.2
AVE
Coating Weight
Sprayed Per
1000 Can
Ib kg
1.048
1.179
1.215
1.166
1.192
1.182
1.142
1.042
1.103
1.135
1.078
1.132
0.475
0.534
0.551
0.529
0.540
0.536
0.518
0.472
0.500
0.515
0.489
0.514
Remarks
Kun 1
1045-1120
(35 rain.)
Run 2
1351-1425
(34 min.)
Run 3
1602-1635
(33 min.)
Kun 4
1510-1545
(35 min.)
Kun 5
1010-1047
(37 min.)
Run 6
1330-1405
(35 min.)
Run 1 to
Run 6
(1) Lino one was producing 16 ounce cans; the equivalent number of 12 ounce cans produced on that line was caluclated by multiplying the
recorded number of 16 ounce cans by the ratio of the 16 ounce to 12 ounce interior can surface area (1.35).
-------
calculated and expressed in weight (Ib and kg) per 1000 cans coated. Results
calculated by using daily can counting and coating usage and by using interme-
diate can counting and coating usage (per each testing run) were consistent.
For the six test runs, the coating weight sprayed ranged from 1.042 Ibs (0.472
kg) to 1.192 Ibs (0.540 kg) and averaged 1.132 Ibs (0.514 kg). The complete
data are listed in Appendix F.
5.5 Coating Applied
For each of the six test runs, four cans were pre-weighed, passed through
the coating and drying operation, and re-weighed after exiting the oven. Re-
sults of these weighings are listed in Append.!x G and also included in Table
5-5. The weights were consistent for all six test runs, ranging from 0.082 kg
(0.181 Ibs) to 0.092 kg (0.204 Ibs) per 1000 cans. This represents approxi-
mately 17% of the coating usage weight which is in good agreement with the 20%
solids content of the coating.
5.6 Material Balance
Table 5-5 shows the material balance for the interior spray coating opera-
tion. Test Runs 5 and 6 are excluded from the average since the results from
these two runs indicate that the mass of VOC recovered, i.e. the sum of the
coating applied on the can and the total emission is more than the mass of
coating originally sprayed.
For the other four test runs, 62% of the coating applied was accounted for
as emissions at various locations (See Section 5.2 for distribution among
those locations) ; 17% of the coating was applied and remained on the cans.
This left 21% of the coating unaccounted for. It is believed that a major
portion of this 21% escaped as fugitive emissions at the coater, from the can
-17-
-------
TABLE 5-5
SUMMARY OF VOC MATERIAL BALANCE (PER 1000 CANS)
INTERIOR SPRAY COATING, TWO-PIECE CAN PLANT
JACKSONVILLE, FLORIDA
OCTOBER, 1979
Test Coating
(Run) Weight
No. Date Sprayed
kg Ib
1 10/17/79 0.529 1.166
100%
2 10/17/79 0.540 1.192
100%
3 10/17/79 0.536 1.182
100%
4 10/17/79 0.472 1.042
100%
5 10/17/79 0.515 1.135
100%
6 10/17/79 0.489 1.078
100%
AVE (1-4) 0.519 1.145
100%
\ Coating
Weight
Applied
kg Ib
0.090 0.198
17% .
0.082 0.181
15%
0.092 0.204
17%
0.086 0.189
18%
0.090 0.199
17%
0.089 0.197
18%
0.086 0.190
17%
Total
Sampled
Emissions
kg Ib
0.410 0.903
77%
0.307 0.671
57%
0.247 0.544
46%
0.322 0.710
68%
0.544 1.199
106%
0.611 1.351
125%
0.321 0.708
62%
Difference*
kg Ib
0.029
6%
. 0.151
28%
0.197
37%
0.064
14%
-0.119
-23%
-0.211
-43%
0.110
21%
0.063
0.333
0.434
0.141
-0.262
-0.465
0.243
*Difference = Coating Weight - Coating Weight Applied - Total Sampled Emissions
or = Coating Unaccounted For
-18-
-------
transfer conveyor-elevator enclosure, remained in the oven and condensed and/-
or was emitted from the quench (cooling zone) exhaust of the oven. in
addition, the method for calculating emissions from the sampling results may
also attribute some of the loss.
-19-
------- |