a WJEQ-TR-1993-0012
JSLED/% EPA/600/R-93/077
AD-A279 762
G
ii
EVALUATION OF INNOVATIVE PAINTING PROCESSES
R
hJI K. Pandalal and G. Pandalal
s
T ENVIRONICS DIRECTORATE
JL 130 Barries Drive, Suite 2
R Tyndail AFB FL 32403-5323
O
N Pandalal Coalings Company
837 Sixth Avenue
Brackenridge PA 15014
US EPA/AEERL
Organlca Control Division MS-61
L Reaearch Triangle Park NC 27*711
A
Q April 1993
o
Final Technical Report for Period October 1990 • October 1991
A
T
O
R
Y
k
AIR FORCE MATERIEL COMMAND
TYNDALL AIR FORCE BASE, FLORIDA 32403-S32|
-------
NOTICES
This report was prepared as an Account of work sponsored by an
agency of the United States Government. Neither the United States
Government nor any agency thereof, nor any employees, nor any of their
contractors, subcontractors, or their employees, make any warranty,
expressed or implied, or assume any legal liability or responsibility for
the accuracy, completeness, or usefulness of any privately owned rights.
Reference herein to any specific commercial product, process, or service
by trade name, trademark, manufacturer, or otherwise, does not
necessarily constitute or imply its endorsement, recommendation, or
favoring by the United States Government or any agency, contractor, or
subcontractor thereof. The views and opinions of the authors expressed
herein do not necessarily state or reflect those of the United States
Government or any agency, contractor, or subcontractor thereof.
When Government drawings, specifications, or other data are used for
any purpose other than in connection with a definitely Government-related
procurement, the United States Government incurs no responsibility or any
obligation whatsoever. The fact that the Government may have formulated
or in any way supplied the said drawings, specifications, or other data
is not to be regarded by implication, or otherwise in any manner
construed, as licensing the holder or any ovher person or corporation! or
as conveying any rights or permission to manufacture, use, or sell any
patented invention that may in any way be related thereto.
This report has been reviewed by the Public Affairs Office (PA} and
is releasable to the National Technical Information Service (NTXS). At
NTXS, it will be available to the general public, including foreign
nationals.
This report has been reviewed and is approved for publication.
Veil J Lamb, Col, USAF, BSC
Director, Environ!cs Directorate
EDWARD N. COPPOLA, Major, USAF
Chief, Environmental Compliance Division
-------
1 REPORT DOCUMENTATION PAGE
form Apr>',*? (fit* Mu«m
1 1. SUmtMfNTARY NOTIS
|«» DISTRIBUTION / AVAILABILITY STATEMENT
J Publication Approved
I Distribution Unlimited
1?b DISTRIBUTION CODI
h
| *' ,RA<- (M*w>mum oowords) -phis report details an evaluation of two novel
1 spray painting techniques that may decrease volatile organic compound
I(VOC) emissions associated with application of surface coatings. One
|technique uses supercritical carbon dioxide as a solvent and a
Ipropellant to deliver and disperse the coating. The second system,
J called the Ultra-*Low Volume (ULV) process utilizes pressurized
I nitrogen to deliver a constant-flow, very soft spray through the gun
Jnozzle. The C02 system encountered problems in delivering urethane
J topcoats and was eliminated from field testing. Data were gathered on
JVOC emissions, paint consumption, coat thickness, and workplace
I exposure during application of an epoxy primer and a silica-filled
jchemical resistant (CARC) topcoat to a fleet of trucks at Warner
IRobins AFB GA. The ULV process was compared to a conventional air
latomizing gun system. Results revealed n 50-percent decrease in VOC
jemi ssione, and a 30-percent decrease in paint consumption at no
[increase in exposure when using the ULV system.
I 14. SUBJICT TtRMS
I (I!) spray painting, transfer efficiency, constant-pressure
i airless spray gun, air pollution, VOCs, pollution prevention
15, NUMBER Of PAG(S
48
16, PRICi CODI
1 <7. SlCURITY CLASSIFICATION
1 or REPORT
1 Unclassified
11. SECURITY CLASSIFICATION
Of THIS CAGE
Unclassif ied
19. SCCURUY CLASSIFICATION
Of ABSTRACT
Unclassi f ied
70. IIMUATION or ABSTRACT
None
NSN 75*0-01.J80 S500 St*r»d*fd ?Otm 296 (H#v 189)
I'ifv-H »«#M %~* ?If t»
MS *0?
-------
PREFACE
This report was prepared by Pandalai Coating Company, 837
Sixth Avenue, Brackenridge PA 15014, under contract #68D00088 for
the U.S. Environmental Protection Agency (EPA) Air and Energy
Engineering Research Laboratory (AEERL), Research Triangle Park
NC 27711, and the Civil Engineering Laboratory (CEL), Air Force
Engineering Support Agency, 139 Barnes Drive Tyndall AFB FL
32403-5319.
This technical report summarizes work done between 15 October
1990 and 15 October 1991. The EPA program monitors were Charles
H. Darvin and Bobby E. Daniel? the CEL project officer was Dr.
Joseph D. Wander.
Generous cooperation by BASF personnel in Atlanta GA, Air
Force personnel at Wright Laboratories and Warner Robins Air
Logistics Center, David F. Pulley of the Naval Air Development
Center (NADC)in Warmister PA, and Lyndon S. Cox an EPA Senior
Environmental Employee is gratefully acknowledged.
kacoaston Tor /
"iris c-p.Asi m
OTIC "/B
IW.".'- .*"•«*
jn_, t ¦. vlon-
~
~
; tl%i
»" k "5r»i/»-r
.Slav.
(*'
m«c-1- \ u.>.
iii
(The reverse of this page is blank)
-------
EXECUTIVE SUMMARY
OBJECTIVE* The objective of the project was to perform a
practical evaluation of two innovative spray painting
technologies that can decrease volatile organic compound (VOC)
emissions associated with application of surface coatings.
BACKGROUNDt Solvents added to decrease the viscosity of coatings
to levels compatible with application methods are a major source
of volatile organic compounds (VOCs). These VOCs are subject to
regulation under environmental codes, including Title 3 of the
Clean Air Act Amendments. While end-of-pipe technology exists to
capture or destroy these volatiles from the spray booth emissions
air stream, it is expensive to maintain and operate. A desirable
alternative, when practical, is to lower the solvent content
(i.e., raise the solids content) of the coating as applied.
Practicality depends on the availability of a delivery system
that can apply a high-solids coating formulation at the viscosity
supplied. This approach, defined as pollution prevention, lowers
the amount of solvent available for emission to the atmosphere.
This study initially addressed two approaches to spraying high-
viscosity coatings. Unicarb technology uses supercritical carbon
dioxide, a relatively nontoxic, nonpolluting material, as both a
solvent to adjust viscosity and as the propellant to deliver and
disperse the paint. The embodiment of this technology at the
time of this study proved incompatible with curient two-component
urethane topcoat formulations, and thus was not evaluated in the
field. However the principle remains valid, and reexamination of
Unicarb process as a means of applying urethane topcoats can be
reexamined when the process and coating are compatible.
The second approach is a low-tech device employing a portable
pressure cell comprising two connected chambers, one pressurized
with nitrogen gas, and the second isolated from the first by a
floating piston that transmits the gas pressure to a charge of
paint. Flow of paint from the second chamber through a noz2le
results in a constant-flow, very soft delivery of a paint spray.
This technology, now called the Ultra-Low Vclume fUlAH spray
process, is available from commercial outlets under license from
Air Compliance Technologies.
SCOPE? During the course of this study, the ULV gun was qualified
to apply M^.L-C-83286 and MIL-C-85285 urethane topcoats, and was
subsequently used in the field to apply MIL-P-23377 epoxy primer
and a silica filled chemical agent resistant (CARC) topcoat, MIL-
C-53039. Data were gathered on VOC emissions, paint consumption,
coat thickness, and workplace exposure during application of
prime and top coats to two equivalent (same number of each size)
sets of trucks. One set was painted with the ULV gun, and the
other with a conventional air atomizing gun, by painting shop
staff at Warnec Robins AFB GA.
v
-------
METHODOLOGY8 Standard procedures were maintained during tests.
Painters wore supplied-air respirators, and airflow in the booths
was verified by the Warner Robins bioenvironmental engineering
(BEE) survey. BEE personnel also monitored representative
painting sessions. A Zahn viscometer cup was used to measure
viscosity at intervals corresponding roughly to refilling. Film
thickness, measured using a thickness gauge (Inspector, Elcometer
Instruments), is reported as an average of 10 determinations for
each truck. Paint consumption was measured by weighing the paint
gun systems full and after delivery, on a scale accurate to ±0.02
pounds. Consumption was the net weight loss. VOC concentrations
were measured with a flame ionization detector (Rosemount Model
400A Hydrocarbon Analyzer). Stack samples were delivered to the
FID by ADI 01320T dual-head, Tef lonR-diaphragm pumps. Continuous
VOC mear 'rements were recorded on strip charts. Tabular VOC data
were collected at 15-second intervals, integrated over time by
application of the trapezoid rule, and compared to the
mechanically integrated strip chart data to confirm calibration.
TEST DESCRIPTION: The test consisted of measuring paint
consumption, VOC emissions, and paint coat thicknesses during
painting of each of 14 trucks. However, experimental
irregulrities caused elimination of the data from the first
weekend, leaving only two 2.5-ton and two 5-ton trucks as the
sample population for each treatment group. Personal exposure
sampling was conducted by the BEE group during one of the tests
performed with each of the guns. The painters and shop
supervisor were surveyed for their impressions of the two guns.
RESULTS: After normalization to correct for coat thickness, paint
consumption of the CARC topcoat applied with the ULV gun averaged
20 percent less than with the conventional unit. The ULV gun
lowered VOC emissions by nearly 50 percent, a result of greater
transfer efficiency (noted above) and the capability of the ULV
gun to deliver undiluted, high-viscosity paint. A qualitative
decrease in density of overspray was seen during the study.
Personal exposure data are complicated by the occurrence of a
solvent spill during the exposure test for the ULV gun, but still
show that the ULV gun creates no increase in exposure compared to
the conventional gun. Impressions of the ULV gun were uniformly
favorable, both about its handling and about the coat delivered.
CONCLUSIONS: The ULV gun delivered satisfactory prime and top
coats. Transfer efficiency was significantly better than for the
conventional gun. Both of the latter observations have potential
to lower the rate of VOC emission associated with spray painting.
RECOMMENDATION: The ULV gun should be examined further, both to
determine suitability for specific applications to ground
equipment, aircraft, parts, etc., and to establish optimal
viscosities for application to minimize VOC emissions consistent
with satisfactory coat thickness and characteristics.
vi
-------
TABLE OF CONTENTS
Section Title Page
I INTRODUCTION...... 1
A. OBJECTIVE ..... 1
B. BACKGROUND. 1
1. UNICARB Process 1
2. Air Compliance Technologies'
ULV Process 2
C. SCOPE/APPROACH. 2
II LABORATORY EXPERIMENTAL PHASE 4
A. LABORATORY TESTING OF THE ULV PROCESS 4
III FIELD TESTING. 6
A. EXPERIMENT. 6
1. Viscosity Measurements................ 6
2. Thickness Measurements. 7
3. Paint Consumption 7
4. VOC Measurements...................... 7
B. ANALYTICAL METHODS 7
C. RESULTS 9
IV DATA ANALYSIS 13
A. ULV DATA. ............ 18
V PROCESS-DEPENDENT PAINT SPECIFICATIONS.. 20
VI MATERIALS SAVINGS 21
4
VII WORKER SAFETY AND HEALTH EFFECTS............. 23
VIII QUALITY CONTROL. 24
A. DATA QUALITY INDICATORS , 24
IX CONCLUSIONS AND RECOMMENDATIONS.............. 26
APPENDIX
A STRIP CHART RECORDER OUTPUT. 27
B RAW DATA. 31
vii
-------
LIST OF FIGURES
Figure Page
1 Schematic Diagram Of VOC Sampling.... 8
2 Graphical Illustration of Trapezoid Rule....... 9
3 Emissions Comparison Of ULV and conventional spray
Prime Coat, Night Shift, 20 September 1991 10
4 Prime Coat VOC Emissions vs Time for ULV Spray
Day Shift, 13 September 1991 14
5 Prime Coat VOC Emissions vs Time for Conventional
Spray, Day Shift, 20 September 1991 15
6 Top Coat VOC Emissions vs Time for Conventional
Spray, Day Shift, 20 September 1991 16
7 Top Coat VOC Emissions vs Time for Conventional
Spray, Night Shift, 20 September 1991 17
8 One Gun vs Two Guns—Conventional Spray Gun,
Primer. 18
-------
LIST OF TABLES
TABLE PAG!
1 AIR FLOW RATES FOR EACH BOOTH 6
2 AMOUNT OF PAINT USED AND DRY FILM THICKNESS
FOR EACH TRUCK PAINTED. . 11
3 REFILLING CHART FOR CUP GUN.. 12
4 TEST DATA FOR ALL PRIMER AND TOPCOAT RUNS.. 13
5 REFILLING CHART FOR DAY SHIFT PRIMING ON
20 SEPTEMBER 1991 15
6 REFILLING CHA3T FOR NIGHT SHIFT PRIMING ON
20 SEPTEMBER 1991 16
7 PROPERTIES OF MATfRIAL APPLIED BY ULV AND
CONVENTIONAL SYSTEMS 20
i NORMALIZED MATERIAL SAVINGS FOR 5-TON
TRUCKS - 21
9 NORMALIZED MATERIAL SAVINGS FOR 2.5-TON
TRUCKS 21
10 RESULTS OF WRALC AIR SAMPLING 23
11 SPAN GAS ID ZERO GAS CHECK AT END OF RUN.. 25
ix
(The reverse of this page is blank)
-------
Section I
INTRODUCTION
The content of this report is the result of a study carried
out by Pandalai Coatings Company (PCC) to evaluate two innovative
painting processes. The study was carried out under contract
#68000088 for the Air Force Civil Engineering and Support Agency
and the Air and Energy Engineering Research Laboratory (AEERL) of
the Environmental Protection Agency in RTP, North Carolina. The
two innovative painting processes included were the VniCarb
process, and Air Compliance Technologies' Ultra-Low Volume (ULV)
spray process. A brief description of the fundamental principles
involved in the two processes is given in the following sections.
A. OBJECTIVE
The objective of this study was to evaluate and define the
painting efficiency and benefits of two selected painting
systems. The study was conducted through laboratory evaluations
and operational evaluations at a painting facility. Each system
was required to be able to apply the complete range of USAF
coatings applicable with conventional spray technology. The
systems were also required to utilize typical USAF coatings
including high solids coatings.
B. BACKGROUND
1. UNICARB Process
In the UniCarb process, carbon dioxide (C02) is used undar
supercritical conditions to replace part of the coating «*olve*.rt
formulation that is used to dilute the coating to the required
spray viscosity. The amount of carbon dioxide that can be aculsd
depends on the resin system and the type and amount of solvents
that are present in the resin system. The UniCarb system has
several advantages;
a. Pollutant emissions are significantly reduced since the
coating solvent loading is reduced, therefore health hazards from
potentially toxic solvents are reduced.
b. The possible reduction of VOC emissions from the spray
painting operations may permit compliance with the Clean Air Act.
c. Waste by-product carbon dioxide can be used.
d. The cost of bulk carbon dioxide is 50 to 60 percent
less than the cost of solvents used in modern coating systems.
e. System can reduce or eliminate the need to use
1
-------
viscosity modifying diluent organic solvents or thinners.
Laboretory experiments on the ability of the technology to
apply USAF specification coatings were conducted at the Union
Carbide West Charleston WV research facility. Since the coating
would be diluted with an alien compound (C02) and heated,
compatioility, application, and pot life of the two-component
system were evaluated. USAF specifications require the pot life
for mixed two-component urethane coatings to be at least four
hours. In addition, the coating quality must meet the
requirements for gloss, orange peel and other surface and
functional properties.
It was found that the UNICARB system is not capable of
utilizing two-component systems without premixing. Thus the two-
component urethane used for the evaluation was mixed in a
separate vessel during the laboratory evaluations and combined
with carbon dioxide withir the system. Since the present UNICARB
system heats the premixed coating for use with the high pressure
carbon dioxide, premature setting of the coating occurs. Based
on this limitation the system could not meet the pot life
specificetion for the mixed coating. This one limitation
eliminated the UNICARB system from further consideration during
the evaluation program. Thus this technology, used under its
current operating conditions, was considered incompatible with
present USAF two-component topcoe ts.
2. Air Compliance Technologies' ULV Process
In the ULV process, compressed nitrogen gas acts upon the
back of a floating piston to force paint through a spray gun and
the nozzle. Depending upon the viscosity and solids content of
the paint and the size of the tip opening, the pressure of the
nitrogen gas in the cylinder is adjusted between 700 to 950 psig.
Spray gun tip openings range from 0.012 to 0.020 inches. As in
conventional airless spray painting operations, the paint is
driven from a container by a displacement pump and compressed at
the top of the piston atop the nitrogen cylinder. When the spray
gun is triggered, paint is pushed through the nozzle opening at a
constant flow rate. Unlike compressor operated systems, which
surge at each stroke of the pump, this constant flow rate permits
deposition of the coating at an even, uniform thickness.
Theoretically this eliminates peaks and valleys, which
necessitate multiple passes over the surface. The uniformity of
thickness allows the usage of less paint, thus increasing the
painting efficiency in area covered per unit of paint used.
C. SCOPE/APPROACH
An examination of the Unicarb and ULV systems indicates that
the two technologies are at different stages of development. The
ULV system appears to be a simple system that is suitable for
2
-------
operations in small job shops as well as in a large painting
operations. The Unicarb system is geared toward larger painting
operations utilizing single component coatings. The Unicarb
system could not be used during the field testing phase using any
of the aerospace two-component coating systems, due to problems
of paint clogging. However, since the ULV system did not exhibit
these problems, it advanced to on-site testing and evaluations at
an operating USAF facility.
The system evaluation objectives were accomplished through a
series of tests, beginning with the laboratory testing of the
Unicarb and ULV systems. Upon successful completion of the
laboratory evaluation, the ULV system was evaluated at an
operating facility. Factors evaluated during laboratory and
fiald testing included surface quality, pot life, relative
emissions generated during painting (compared to conventional
pointing system emissions), volume of paint used per unit area
covered, amount of solvent contained in the applied paint, and
ease of use of the system.
- Since the Unicarb system was eliminated from further
evaluation after the initial laboratory studies were completed,
additional discussions of that system are omitted from this
report.
3
-------
Section II
LABORATORY EXPERIMENTAL PHASE
The paint system used during initial laboratory evaluations
included a MIL-P-23377 primer and a MIL-C-83286 topcoat.
Specially formulated high-viscosity variants of the above coating
were used for the laboratory evaluation phase. This low-VOC
topcoat system nominally conforming to MIL-C-83286 was specially
formulated by and procured from Crown Metro Aerospace Coatings
Company in Greenville, South Carolina. Two colors, camouflage
grey and white, were shipped to Atlanta, Georgiii, for initial
laboratory evaluation. Based on data obtained from Crown Metro,
the volume of solids was 50 percent and the VOC content of both
the grey and the rfhite was less than 3.5 pounds per ga3Ion of
coating, the maximum amount of VOC permissible for compliance in
many states. These data were verified in PCC's laboratories.
Km LABORATORY TESTING OF THE OL? PROCESS
Approximately 2 gallons of both white and grey camouflage
colors were used for preliminary laboratory testing. Test panels
were treated and primed by the Air Force Materials Laboratory,
Wright-Patterson AFB OH, and supplied to the test laboratory for
t.opcoating. The final preparation of test panels, with the
application of the topcoat, was carried out at the BASF research
facility near Atlanta GA.
Since the ULV technology is based on the application of
nitrogen pressure to force the high-viscosity coating through the
nozzle, the atomization will depend on the temperature of the
coating at the time of application, rheological properties of the
coating, and nozzle opening. These settings were determined and
panels prepared for quality assurance laboratory evaluations at
Wright-Patterson AFB.
Work with the specifically modified high-solids coating
presented a number of problems during the laboratory phase of the
project. One problem encountered was slow drying of the modified
coating. Work carried out at PCC laboratories and at the Air
Force Materials lab confirmed that the problem was due to
insufficient catalyst to promote rapid and complete drying. It
was therefore decided to utilize the standard compliant two-
component urethane coating system MIL-C-85285 for all the further
evaluations.
This coating system and the ULV process had never been
utilized before at Warner Robins Air Logistics Center. Therefore,
the Air Force corrosion group and the F-15 engineering group
required that test panels be prepared and tested for compliance
with the substitute coating system before permission for the
4
-------
field test program could be granted. Test panels of 2024-T0 and
2024-T3 aluminum were anodized, chromate conversion-coated and
primed with MIL-P-23377 epoxy primer. These panels were allowed
to dry for 3 to 4 hours and sent to the ULV corporation for
application of the MIL-C-852S5 top coat. They were then sent to
the Naval Air Development Center (NADC) laboratories for
compliance testing. Quality approval of the test panels was
granted by the NADC laboratory. Approval for field testing of
the ULV system was granted by the site engineering authority
based on the NADC approval of the test panels.
Although it was originally planned to paint F-15 aircraft
parts, military vehicles belonging to the Air Logistics group
were substituted for the field testing. This change ensured a
steady supply of similar items to be painted over a predictable
length of time. The change also necessitated a change in paint.
Initially it was scheduled that a two-component polyurethane
coating, MIL-C-46168, for which the voc content is 4.65 pounds
per gallon would be used for the testing. However, due to
changes in military vehicles, a single-component moisture-curing
polyurethane, MIL-C-53039, for which the VOC content is 3.50
pounds per gallon was used as the topcoating system.
The test program was scheduled to be carried out during three
successive weekends in September 1991. The parameters to be
monitored included paint consumption, total hydrocarbon emission,
paint dry film thickness, and worker comparison and opinions, for
both the ULV and conventional air and airless spray painting
processes.
5
-------
Section IZI
FIELD TESTING
JU EXPERIMENT
The site selected for the field test program was Warner
Robins Air Logistics Center, where F-15, C-130, and C-141
military aircraft are repaired and maintained. The component
painting shop located in building 180 was selected as the
location for the painting evaluations.
An experimental plan was devised to test the ULV system
against the conventional pressure pot spray systems by comparing
the following parameters and variables.
- Exhaust VOC concentrations
- Paint consumption per unit area
- Surface area covered
- Spray tip size
- Spray pressure
- Solvent content of coating system
- Viscosity of paint
- Film thickness
- Total estimated VOC emissions
The evaluations were run on three consecutive weekends, 13
September 1991 to 27 September 1991. Equal numbers of both 2.5-
and 5-ton trucks were painted. During the first weekend, six
trucks were monitored; three 2.5-ton and three 5-ton vehicles.
During the next two weekends a total of eight trucks were
monitored—four of each size.
Such baseline data as exhaust fan airflow rate and paint
composition had to be obtained from the paint shop records. Fan
flow rates (Table 1) for each booth were measured in June of 1990
by the Warner Robins Bioenvironmental Engineering Group. The
coating content was obtained from the Material Safety Data Sheet
(HSDS).
1. Viscosity Measurements
The viscosity of the paint was measured by using a Zahn
viscometer cup. Viscosity measurements were generally taken each
time the large pressure pot or container for the ULV system was
TABLE 1 AIR FLOW RATES FOR EACH BOOTH
Equipment
Paint Spray Booth # 73 3
Paint Spray Booth # 734
Measured CFM
39,029
66,243
6
-------
refilled. This ensured that the paint viscosity was
approximately constant throughout the test,
2. Thickness Measurements
The dry film thickness of the paint was measured using an
Inspector" thickness gauge by Elcometer Instruments. Ten dry film
thickness measurements were taken at different locations on each
truck and averaged.
3. Paint consumption
To measure the weight of the paint used, a method by which
the spray gun, hose, and paint container were weighed together
was devised. An Ohio Valley Industrial scale accurate to 0.02
pounds gave weights before and after each painting cycle. These
measurements permitted a determination of paint consumption, and
indirectly were used to calculate total VOC emissions.
4. VOC Measurements
A Rosemount model 400A flame ionization detection
Hydrocarbon Analyzer was used to measure VOC concentration in the
exhaust duct. Outputs from the machine are in the form of a
digital readout and a variable DC output suitable for a standard
strip chart recorder. From the strip chart recorder output, an
integration over time was used to calculate the total emissions
during the duration of the experiment.
Booths numbered 733 and 734 located in building 180 were
used for the painting evaluations, and their exhausts were
sampled for total VOC emissions. Emission samples from the
booths were drawn in parallel by an ADI 01320T dual-head Teflon*
diaphragm pump. This arrangement allowed valves to be put on
both suction ends of the pump, which created two separate
sampling lines—one for each booth. Both booth #733 and #734
have two separate fans and stacks, and sampling took place just
beyond each exhaust fan. Since each spray booth had two stacks
both stacks were sampled simultaneously or the sample combined
from both to obtain a representative result for booth emissions.
Figure 1 is a schematic diagram of the sampling arrangement.
B. Analytical Methods
VOC emissions data in parts per million (ppm) were obtained
in two forms, strip chart recordings, and total VOC emissions
taken every 15 seconds and saved in spreadsheet form. Two
methods were subsequently used to analyze these two data sets. A
Planix digital rolling planimeter was used to calculate the area
in square inches under the curve drawn as the strip chart
recorder output. This area was then converted into the
appropriate units of ppm-^ec. The spreadsheet of output
7
-------
Figure 1 Schematic Diagram of VOC Sampling
measurements, taken at equal time intervals, conveniently lends
itself to a simple mathematical manipulation via the trapezoidal
rule equation:
AREA -
-------
the digital data, use of the trapezoid rule provides an estimate
of the area under a series of sensor rsadings{i.e.,n readings
create n-1 trapezoids), The sum of the area of these trapezoids
is approximately the area under the curve. The quality of the
approximation improves as the number of intervals increases, as
the typical period of accumulation is about an hour and the data
were sampled at 15-second intervals, area so calculated are a
reasonable approximation of the true area, h graphical
illustration of this method is shown in figure 2.
Figure 2 Graphical Illustration of Trapezoid Rule
C. RESULTS
Measurements of emissions during testing are given in Table 2
on page 9. It gives the amount of paint used for each priming and
topcoat operation# and the final dry film thickness—where it was
measured—on each truck. Measurements of emissions are also
provided for analysis. These graphs are the basis for the
conclusions and recommendations of the report. Figure 3 overlays
displays of the amount of VOC measured versus time during
painting of nominally identical 2.5-ton trucks with the
conventional and UbV guns. From this figure, it can be seen that
the overall emissions rates from the ULV system are only two-
thirds that for the conventional spray gun. This run provides a
good comparison between the two spray systems for many reasons.
First, operator bias is eliminated because the same operator used
9
-------
both systems. Second, the duration of priming of the trucks is
approximately the same, so background emissions will not be a
factor. Finally, both systems were used on "identical" trucks in
the same booth, bo the effect of these variables can be
considered equivalent for this comparison.
In Figure 3, it is easy to misinterpret the conventional
spray readings as lower than actual. This is because, in the
conventional process of priming, the two-component primer,
MIL-P-2 3 377, was first thinned and then sprayed using one-quart,
cup guns. Valleys in the VOc emission curve indicate the time
during which the cup guns were refilled when using the
conventional system. Table 3 shows the refilling schedule for
300-
s
fi-
fe
•
200 -
M
C
©
•
I 1
I
E
w
100-
1
>
\
0
0
-r-
10
-~r~
20
—r*
30
Conventional Spray
ACT Spray
T™
40
50
60
Time (mill)
Figure 3 Emissions Comparison of ULV and Conventional Spray
Prime Coat, Night Shift, 20 September 1991
the cup guns used in this particular run. Comparison of the
chart and the graph verifies the correspondence between the
valleys in the pattern of VOC emissions from the conventional
system and the times when the gun was refilled. No such valleys
appear in the process line for the ULV spray, on which refilling
was not necessary because the ULV system comes equipped with a
four-gallon storage unit for spraying. This volume was usually
sufficient tc complete painting of one unit.
10
-------
TABLE 2 Amount of Paint Used and Dry Film Thickness for Each
Truck Painted
DATE
BOOTH
NO.
TRUCK
ID#
TRUCK
SIZE
METHOD
COAT
PAINT
USED
FILM
THICKNESS
TONS
LBS
MIL
9/13
2
87K1703
2.5
CONV.
PRIMER
24.4
*
9/13
2
87K2085
2.5
CONV,
PRIMER
19.89
*
9/13
3
85K479
5
ULV
PRIMER
9.57
*
9/13
3
A -JV-J
O *3 aS J
5
ULV
PRIMER
6.23
*
9/13
2
87K1703
2.5
ULV
TOP
88.12
10.5-12
9/13
2
87K2085
2.5
ULV
TOP
*
10.5-12
9/13
3
85K479
5
ULV
TOP
107.5
10- 12
9/13
3
83K3
5
ULV
TOP
*
10- 12
9/14
2
87K2347
2.5
CONV.
PRIMER
12.3
*
9/14
3
83K1
5
CONV.
PRIMER
17.3
*
I 9/14
2
87K2347
2.5
CONV.
TOP
44.2
6- 7.5
| 9/14
3
83K1
5
CONV.
TOP
52.9
6- 7
19/20
2
87K1991
2.5
CONV.
PRIMER
6,26
*
9/20
2
87K2133
2.5
CONV.
"DT> TMIfD
6.2
#
9/20
3
87K1695
5
CONV.
PRIMER
16.16
*
9/20
3
87K2086
5
CONV.
PRIER
10.92
*
9/20
2
87K1991
2.5
CONV.
TOP
31.9
5-7
9/20
2
87K2133
2.5
CONV.
TOP
50.1
6-7.5
9/20
3
87K1695
5
CONV.
TOP
60.4
7-8
9/20
3
87K2086
5
CONV.
TOP
53.7
6-7.5
9/27
3
86K9107
5
ULV
PRIMER
9.9
*
9/27
3
87K3323
5
ULV
PRIMER
*
*
9/27
2
86K10086
2.5
ULV
PRIMER
5.9
*
9/27
2
87K10091
2.5
ULV
PRIMER
*
*
9/27
3
86K09107
5
ULV
TOP
95.3
8-10
9/27
3
87K03323
5
ULV
TOP
*
*
9/27
2
85K10086
2.5
ULV
TOP
83.4
6.5-8
9/27
2
87K10091
2.5
ULV
TOP
*
h
* no data available
NOTE: Due to inconsistencies in the spray process, clogging, spillage,
and operator error, data from the first weekend of testing were not used
for purposes of analysis, although they are presented here.
11
-------
4
9
12
19
21
23
27
31
34
39
45
TABLE 3 REFILLING CHART FOR CUP GUN
Cup Weight
Cup Weight
Total
Before (lbs)
After(lbs)
Weight(lbs)
5.60
3.26
2.34
5.72
3.46
2.26
5.72
3.30
2.42
5 - 64
3.28
2.36
5.52
3.30
2.22
5.54
3.26
2.28
5.52
3.24
2.28
5 • 54
3.16
2,38
5.42
3.68
1.74
5.42
3.20
2.22
5.44
3.18
2.26
5.50
3 .18
2 .32
12
-------
Section IV
DATA ANALYSIS
Copies of the strip chart recorder data are in Appendix A,
and the raw numerical data appear in spreadsheet form in Appendix
B« Also included on the spreadsheet are calculations of painted
area and paint used. This will be addressed later in this
section. Table 4 lists the conditions for each individual prime
and top coat run. Individual graphs and accompanying tables
follow in this report. When analyzing the data, several factors
that affect emissions must be considered:
- Operator error
- Paint shop procedures
- Spills/leaks of paint and/or thinner
- Ambient ventilation conditions
- Direction of spray by operator
Table 4 TEST DATA FOR ALL PRIMER AND TOPCOAT RUNS
Thinner
Painting To
Truck
Booth
Time Spray
Pressure Spray Tip
Coating
Coat ing
Viscosity
Date
IDf-Siz*
dumber Minutes Equipment
PS I
Size bib
Type
Ratio
seconds
9/13/91
85K479-5T0N
3
15
ULV
775
12
Primer
1/4.6
IS
*»
83K3-5T0N
3
20
(*
600
12
Primer
1/3.0
21
I*
83K3-5T0N
3
300?
UL V
800
12-20
Top Coat
*
•
•»
85K479-5T0N
3
300?
•I
800
12-20 Top Coat
•
•
•l
87K1703-2.5T0N
2
61
ULV
100
30
Primer
1/11.8
23
«
87K2Q85-2.5T0N
2
44
li
100
30
Primer
1/9.7
23
«
87K1991-2.5T0N
2
13
II
100
30
Primer
1/3.0
23
9/20/91
87K2233-2.5T0N
2
13
com
100
30
Primer
1/2.9
23
•1
87K2086-2.5T0N
3
45
i*
100
30
Primer
1/7.4
21
II
87K1695-5TQN
2
40
ii
100
30
Primer
1/4.9
24
M
87K1991-2.5T0N
2
158
CONY
100
30
Top Cost
1/4.0
29
(1
87K2133-2.5T0N
3
150
«i
100
30
Top Coat
1/4,0
30
87K1695-5T0N
3
150
..
100
30
Top Coat
1.5/4
21
II
87K2086-5T0N
3
125
i»
100
30
Top Coat
1.3/4
24
9/27/91
09107-5TON
3
45
ULV
600
10
Primer
1/4.5
18
II
03323-5T0N
3
45
It
600
10
Primer
1/4.5
18
"
10666-2.STON
2
45
H
1000
10
Primer
1/2.7
18
II
10091-2.5T0*
2
45
"
1000
10
Primer
1/2.7
18
"
09107-5TON
3
65
«
1000
16,14
Top Coat
1/14.4
34
««
03323-5T0N
3
85
It
1000
16,14
Top Coat
1/14.4
34
~t
10866-2.5T0N
2
120
•1
800
14,14
Top Coat
1/12.6
34
u
10091-2.5T0N
2
120
M
800
14,14
Top Coat
1/12.6
34
* No data available
13
-------
Rome tint* from the following analysis should not be included
in the ovor.iI 1 averaging because they involve one or more of the
above factors, and do therefore not represent norma I painting
operations. These are Identified for each instance. Figure 4
may be interpreted as follows: for the interval from
approximately 12 to 15 minutes into the operation emissions drop
drastically. This corresponds to the period from completion of
spraying one vehicle until starting work on the next. During
this changeover period, the paint container and gun were weighed
to determine the amount of paint used for priming a single
vehicle. The total time used to complete priming of both trucks,
25 minutes, is less than the total time recorded in Table 3 for
the Zirst two entries, because the entry in the table includes
spraying time plus time required to clean the spray guns and the
Kpra-Pnc UlV system.
Prime Coai VOC Emissions *f Tlmr for lll/V Spra '
S
Bu
6.
VI
e
0
1
E
UJ
Time (ittln)
Figure 4 Prime Coat VOC Emissions vs Time For UI.V Spray,
Day shift, 1 J September 1991
The weight of primer given in Table 5 is the total weight of
the cup and paint removed from the conventional spray apparatus,
ny subtracting the before and after weights, an accurate value
for total weight change is obtained. As in the preceding
example, each valley in the emissions graph, Figure 5,
corresponds to a time when the one-quart cup gun used for priming
14
-------
Prime Coat VOC Emissions vs Time for Conventional Spray
300
Time (mln)
Figure 5 Prime Coat VOC Emissions vs Time for Conventional Spray
Day Shift, 20 September 1991
needed to be refilled. Also, only one gun was running during the
priming operation. The first operator finished at 1:03 p.m. and
the second operator imir, .diately began priming the second truck in
the booth. Observations made on this run included comments that
Table 5 - REFILLING CHART FOR DAY SHIFT PRIMING ON
20 SEPTEMBER 1991
Primer Weight Primer Weight Total Primer
Time (min)
4
7
12
Before (lbs) After (lbs)
5.32 3.14
5.36 3.14
5.50 3.b4
Weight (lbs)
2.18
2 .22
1.86
Total weight (6.26)
(End Priming of Truck #8" K1991 and Begin priming of Truck
#87K213 3)
15 5.32 3.46 1.H6
19 5.32 3.2 2.12
24 5.50 3.64 1.86
Total Weight (5.84)
Total used (both Trucks) 12.1
the prime coat varies in thickness. The foreman pointed out that
15
-------
this is a common occurrence as the prime coat is necessary only
as an agent for the topcoat to adhere to the surface. Figure 5
and Table 5 represent day shift priming on 20 September 1991 of
5-ton trucks.
300
200
.2
J
E
w
130
Time (m!n)
Figure 6 Top Coat VOC Emissions vs Time For Conventional Spray
Day Shift, 20 September 1991
rifibl A |E DCrTTTTMf! 17AD CUTt?T DDTMTM^ AM OA CPDTPMQFD
a IPX Q O MCtJr JL jLj JL IN vj X * Ui» fi JL «3 ri X O it X X lr Jr> X l«JL n J w £ U v Hi * 1 JUiil
1991
Cup Weight Cup Weight
Primer Total
Time (min)
Before(lbs)
After(lbs)
Weight (lbs)
4
5.60
3.26
2.34
9
5.72
3 . 46
2.26
13
5.72
3. 30
2.42
21
5.64
3.28
2.36
23
5.52
3.30
2.22
27
5.54
3. 16
2.38
31
5.54
3.26
2.28
34
5.42
3.68
1.74
38
5.52
3.24
2.28
45
5.42
3.20
2.12
51
5.44
3.18
2.26
56
5.50
3.18
2.32
Total Primer used (26.98)
16
-------
Several features of Figure 6 are worthy of comment. During
the first 15 minutes, the emissions are approximately half of the
emissions during the rest of the operation because only one gun
was operating at that time. The two valleys at 75-80 minutes and
95-100 minutes correspond to times when the painters were
refilling their respective pressure pots. The emission peak
value at 120 minutes occurred when the first painter cleaned his
spray gun with solvent by spraying solvent into the booth.
400
£
Cu
B-
m
m
E
W
200
Tiwe Mil in)
Figure 7 Top Coat VQC Emissions vs Time for Conventional Spray,
Night Shift, 20 September 1991
Two anomalies in Figure 7 qualify for comment. The spike
near 55 minutes, which rises to over 300 ppm, occurred when
painting resumed before the booth exhausts were turned on. This
caused the concentration of paint and paint solvents to build up,
After raridly rising, the concentration of paint and paint
sol\ * also declined as rapidly when the exhaust fans were
'• : on. Second, the second painter used considerably more
iit than the first because the second painter coated one truck
bu : also did extensive touch-up work on both trucks in the booth.
The difference in emissions between one and two guns can be
measured by overlaying data from the first 15 minutes in Figure 7
17
-------
with a 15-minute segment later in that session. This is shown in
Figure 8, which shows a noticeable difference between one- and
two-gun application. Although the difference is not double, as
might be expected intuitively, area calculations indicate the
following:
One gun
Two guns
Area under Curve (total Emissions)
1191.56 ppm-minutes
1798.54 ppm-minutes
The refilling chart for this sequence is given as Table 6.
Figure 8 shows approximately a 50-percent increase in emissions
between one and two guns. Data of this type are not available
for the ULV spray system because, when the ULV system was
operated for priming, only one gun was used. However, for
topcoats, two guns were used. No comparison of emissions between
the top coat and the primer for the purpose of comparing one- and
two-gun emissions for ULV would be relevant.
300
0-
W5
e
o
'tn
IsJ
200
One Uuil
Tw o < J tins
100 -
Time (itiin)
Figure 8 One Gun vs Two Guns —Conventional Spray Gun, Primer
A. ULV DATA
When examining the data for the ULV spray system, one must
keep in mind several, differences between the spray methods.
During this testing, only one ULV unit was used. For this
18
-------
reason, only one weight is recorded for every two trucks either
primed or painted. Finally, each time the ULV system was used,
it had a different operator, so operator inconsistency is
potentially a significant factor in this testing. One episode
included both spillage and clogging of the guns, which caused
both a time delay and an increase in VOC emissions. Operator
proficiency with the ULV equipment can only improve with
experience and familiarity.
Human factors data, based upon comments of operators at the
time that they were using the equipment follow;
1. The ULV system was easy to handle.
2. There was better visibility of the progress of spraying,
because there was less overspray.
3. The ULV system took less time to complete the same job.
19
-------
Section V
PROCESS-DEPENDENT PAINT SPECIFICATIONS
The tests indicate that the viscosity and solids content is
dependent on which painting process is used. To use the minimum
amount of solvent or a high-solids coating, the airless system is
preferable to the air-assisted painting process. As shown in
Table 1, this study showed that the ULV system permitted the use
of material at 40 seconds viscosity whereas the air assisted
process tolerated only material to 24—second viscosity.
TABLE 7 PROPERTIES OF MATERIAL APPLIED BY ULV AND
CONVENTIONAL SYSTEMS
Process
Conventional
ULV
Zahn Cup #2
Viscosity (Sec)
18-24
40
% Solids
30
50
20
-------
section VI
MATERIAL SAVINGS
Paint consumption for each process is given in the results
table (Table 3). Although the weight difference is less than 10
percent, this figure should not be viewed alone. The dry film
thickness of the coating must also be taken into account. It is
reasonable to extrapolate the paint used for a particular
thickness as a linear function of the dry film thickness. In
this analysis, since the ULV system's dry film thickness exceeded
the conventional in each case, the amount of paint used for the
conventional system was normalized to give the same dry film
thickness. Based on this, Table 8 shows that the weight of the
coating on two trucks painted with the conventional system would
be 144.97 pounds whereas the ULV system used only 95.3 pounds—
which gives a material savings of approximately 34 percent for an
equal coating thickness.
Table 8 NORMALIZED MATERIAL SAVINGS FOR 5-TON TRUCKS
Truck ID Method Actual Kass (lbs) Normalized Mass (lbs)
87K1695 Conventional 60.4 80.53
87K2086 Conventional 53.7 64.44
86K09107 ULV
47K03323 ULV
95.3
95.3
In the case of 2.5-ton trucks, using the same assumptions as
for 5-ton trucks, the normalized material savings; is shown in
Table 9. In this example, the normalized sum of the paint used
on the two trucks with the conventional system is 92.4 pounds
whereas the ULV system used only 83.4 pounds, which gives a
material savings of approximately 10 percent. This savings was
lowered by an incident of spillage of paint and by episodes of
clogging problems with the ULV system. These may be avoided when
operators become more familiar with the system.
T*bl« 9 NORMALIZED MATERIAL SAVINGS FOR 2.5-TON TRUCKS
Truck ZD Method Actual Kass (lbs) Normalized Mass (lbs) )
87K1991 Conventional
87K2133 Conventional
86K10866 ULV
87K10091 ULV
31.9
50.1
83.4
38.6
53.8
92 ¦ 4
21
-------
Averaging the two values for material savings gives a minimum
average of 22 percent. An equivalent statement is that the
topcoat consumption is at most 80 percent of the usage by the
conventional system. Sources at Mobay Corporation report that
approximately 5 million dollars are spent annually on the MIL-C-
85285 coating. A 20-percent savings in coating cost for the two
aircraft topcoats represents a savings of one million dollars
annually. These are rough estimates, and a detailed cost study
is warranted to develop actual costs and estimates of savings.
22
-------
Section VIZ
WORKER SAFETY AND HEALTH EFFECTS
Limited health effects data were taken. Air sampling in the
painters' breathing zones was conducted for both conventional and
ULV spraying operations. The samples were used to determine
airborne concentrations of various contaminants. Samples were
analyzed for methyl isoamyl ketone, butyl acetate, hexamethylene
diisocyanate (HMDI), and total chromium.
Results of the sampling were not consistent with the stack
measurements of hydrocarbons, which had previously shown much
lower concentrations during the ULV operation than during use of
the conventional method. The data in Table 10 are suspect
because a spill of paint thinner occurred just outside the booth
at the beginning of the ULV operation during these evaluations.
This spill is suspected to have influenced the results obtained
for the or^nic vapors. This conclusion is supported by the
results obtained for total chromium (a particulate in the primer)
emissions when sprayed. This was detected at 0.014 milligrams of
chromium per cubic meter of air (mg/m3) during the ULV operation
and 0.038 mg/m3 during the conventional operation (see table 10)
It is reasonable to expect that if the observed elevation of VOC
concentrations during use of the ULV system were produced by
spraying primer there would be a commensurate increase in
particulates. This was not the case.
Table 10 RESULTS
OF WRALC
AIR SAMPLING
ULV
Conventional
Contaminant
Results
fma/m3V
Results fmcr/m3'
Total Chrome
0.014
0.038
HMDI
0.0034
None Detected
Total Chrome
Blank
Blank
HMDI
Blank
Blank
Butyl Acetate
11.3
12.5
Methyl Isoamyl Ketone
201
161
Butyl Acetate
38.3
1.2
Methyl Isoamyl Ketone
455
52
Butyl Acetate
Blank
Blank
Methyl Isoamyl Ketone
Blank
Bl^nk
Because it appears that the measurements obtained for
the organic vapors are not typical of normal ULV operation,
they cannot reliably be used in comparison with those
obtained during the conventional method as a means of
evaluating worker exposure.
23
-------
Section VIII.
QUALITY CONTROL
In any experimental data acquisition program, the
quality of the data obtained roust be verified for accuracy
and bias. Factors involved in quality control include the
number of trucks painted, the operator, such external
conditions as temperature, humidity, ambient wind velocity,
paint thinner spills, direction of spray, and thickness of
coating. The quality of the experimental results improves
with the number of objects painted and with increased
operator familiarity with the ULV spray system.
JU DATA QUALITY INDICATORS
1. Precision:
The Rosemount 4G0A FID instrument is a very
sensitive instrument with a response time of less than
0.6 seconds for 90 percent full-scale output. According to
manufacturers specifications, the precision of the
instrument is ±1 percent of full scale.
2. Bias:
The operating concentration range is a function of
the full scale output to which the machine is calibrated.
For example, if 1000 ppm methane span gas is used to
calibrate at the 0 to 5 VDC output, a reading of 0.5 volts
will indicate 100 ppm of total HC. According to
manufacturers specifications, the bias of the instrument is
±1 percent of full scale over a 24 hour period.
3. Completeness:
The output of the blower (CFM, Table 1) through the
stack is constant. As the total HC measurement can be read
almost instantly by the instrument, there is no limitation
on acquiring all necessary data upon which to base
recommendations. Thus, the completeness is 100 percent.
4. Representativeness:
Zero gas and methane span gas are homogeneous
mixtures, given that they are properly stored after
procurement. They are expected to be representative gases
that will have reliable response in the FID detector.
5. Sampling Procedure:
The total hydrocarbon measurement is performed in
situ. A pump attached to the instrument continuously pulls
the air under test through the FID. The instrument is
initially calibrated and standardized prior to any
measurements. A schematic of the sampling set up is
provided in Figure 1.
24
-------
6. Analytical Procedures:
The instrument was calibrated to display
concentration in volts and percent full scale. Both
dovmscale and upscale calibration points on the recorder and
digital readout were first set using the following method.
The zero standard gas (nitrogen) flows through the sample
port while the zero point was set using the zero adjustment
inside the instrument panel. The upscale point adjustment
was made by using a standard span gas of known voc
concentration. For this experiment, 500 ppm methane in
nitrogen gas was used.
Because the instrument used for VOC measurement is
adjusted for zero with gas containing no VOCs, and the span
gas used contained 500 pm methane, it was workable to check
the zero reading and span calibration before and after each
measurement run. This check gave instrument responses that
fell within five percent of the expected value for every
run. Examples of some of these readings are listed in Table
11 below. All of this information is in the raw data in
spreadsheet form in Appendix B.
Table 11 SPAN GAS AND ZERO GAS CHECK AT END OF RUN
Run
Zero Gas (ppm)
Span Gas
1
1.3
485
2
1.6
506
3
0.7
493
25
-------
Section XI
CONCLUSIONS AND RECOKMENDATION8
Data Bust be gathered on a much larger scale during field
tests before information on transfer efficiency of the ULV
system can be obtained on a strong statistical basis.
However, this preliminary study does show that the process
appears to offer advantages of decreased paint consumption,
reduced overspray, and possibly lower risks to human health.
Although this is implied by the results, more data are
required before statistical inference can achieve a high
confidence level.
The evaluations indicate that a significant reduction in
total VOC emissions is possible with the ULV system. The
measurements collected indicate a 50-percent reduction in
average VOC concentration. This can be confirmed by a
corresponding average of 20-percent decrease in paint
consumption. The product evaluation indicates that the ULV
system is capable of producing comparable coating thickness
with less paint, producing a lower concentration of VOC
emissions, and being at least as safe as a conventional
spray system in terms of health effects for the operators of
the system.
The operators also found the ULV system easy to use.
There were unquantified labor savings based on the
operator's observation of a reduction in the time needed to
paint objects.
The average value for material savings appears to be
about 20 percent. It is estimated that approximately 5
million dollars are spent annually on the MIL-C-85285
coating. A 20-percent material savings in this coating cost
alone is equivalent to a savings of one million dollars.
These 20-percent savings is only approximate, and a detailed
study is necessary to develop actual cost and realistic
savings.
These preliminary findings indicate that a larger
definitive operational test is needed to obtain a firmer
statistical basis for the efficiency of the ULV painting
system.
2£
-------
APPENDIX A - STRIP CHART RECORDER GRAPHS
27
-------
<09»Hdd) DOA
(09«Hdd) DOA
28
-------
.
* • t
1 ,
. , -
;
¦
.
.i ;i!
: j ; | j • '
i! Hi
* 1 . » »
• * t *
H| i;
a
• i 4
h
' | H '
0.: :"3
.f*» l|<(
o;;i ;l'2
ii**
Pi!:
ill
i *«
liii
i,i- ii!!
liii ii
iill IMi
ill
Hi
ijii
1!
liii 1!
II! i
ii
\im i1
-------
a
•H
¦
O
X
t
(09»Wdd) DOA
(09»Wdd) DOA
30
-------
APPENDIX B - RAW DATA
31
-------
Time
(min)
ii
0.25
(IJ
075
I
1.25
1.5
1.75
*
2.25
13
2.75
3
3.25
5.5
575
4
4.25
4.5
475
5
5b 25
55
5.75
6
6.25
fc.5
ft 75
7
7.25
7.5
7.75
11
*25
R5
8 75
9
9.25
9.5
9.75
10
10.25
I0.5
10.75
i:
11.25
I »-5
11.73
12
12.25
12.5
12.75
IS
13.25
13.5
13.75
14
TOPCOAT • BOOTH #3 TWO PRESSURE POTS
Trapezoid VOC
Area ppm
214
*4 2
VOC Calculations
21 9
65 7
Area • Topcoat
23,7
71.1
9395475
23.2
69.6
21.5
IMS
207
62.1
2»2
61 >6
23 1
693
19.3
57.9
17.7
53 1
2ft 2
786
2ft 6
79 8
26 2
78.6
25.2
756
236
70 R
23.2
6«fi
21
63
18 8
yi4
25 n
774
24 3
729
31 9
95 7
25.1
753
66
22*4
67.2
21 4
64 2
253
759
2(17
62 1
22 ft
684
273
HIV
36 9
HI! 7
28
84
21.9
65 7
23 8
71 4
26 2
78 6
31 2
93 6
27 1
HI 9
24 7
74 1
24 5
" 73 5
28 4
85 2
2ft
78
27 2
Hi 6
3
-------
MS 5
|6| 4
133 8
226 8
215?
IK4 S
211 3
1«#
1563
129 9
H67
153 7
2(1* 7
194 1
142 H
125 4
114
icti n
l!W I
H14 7
87 6
101.1
148 5
203 4
192,3
216 6
174.1
168 6
2fM
1893
225 9
205 2
21.1 3
194 4
2193
1W>9
1953
l«6 2
213
177.1
155 4
156 6
1893
159 9
I6II8
1512
172.8
174 9
144 y
1542
1497
1842
139 5
1393
180.9
2U9 4
1959
174.9
200 t
160 8
iys
151.2
149 «
125 4
153 3
132 3
54 7
Ml |
4S> |
54 2
39 4
7K7
76 4
MJ
53 I
42 4
39 5
38 2
44 5
45 2
35 9
S3 I
A3 I
59
55 t
69 5
65.fi
Ml.
63 9
655
72 7
WV3
898
73.7
684
65 8
76 I
56
34 3
32 ft
41.5
39 3
44 I
59
74
63 M
$K 2
72 1
72 9
978
124 5
1179
132 3
177
222
19 M
|74*
216 3
22R.1
185 4
153 3
185 I
139 5
183 9
158 7
174 3
151.2
lf<*5
1578
1692
171) 4
172 8
1566
2184
2178
223 3
1968
1368
1923
44 25
44 5
44 75
45
45 25
45 5
45 75
4*
4ft. 2 3
46 5
4ft 75
47
47 25
47,5
47.73
4H
4K25
4M 5
48 75
4V
4W 25
495
4975
5"
5S.* 25
31), 5
50 75
31
31 25
51.3
31 75
52
32.35
52.5
52.75
53
53 25
535
53 75
54
34 25
54 5
54 75
55
5525
55.3
5575
56
56.25
56.5
56 75
57
57.25
57.5
5X75
58
58 23
58 5
5875
59
59 23
59 5
59 75
6ft
6n25
48 5
51 8
44 ft
756
71 V
61 5
TO 4
Ht>
52 I
43 3
3K9
51 9
669
W?
47 6
41 R
3 *
34 6
34 7
34 9
2U 2
337
49 5
67 K
64 1
72.2
58 I
56 2
fi8
63 I
753
m4
71 I
frl 8
73 I
m.%
65 I
65 «
71
59 1
51 8
52 2
63 1
53 3
53 6
50 4
576
58 3
483
51 4
499
61 4
46 5
53 1
6*1 3
69 8
65 3
58.3
66.7
53 6
65
50 4
497
41 8
51 1
44 I
33
-------
Ml. 75
53 6
|M"8
75 5
39
11*.
61
52 7
15* 1
75 75
33 1
9U ;
M 25
4* 5
145 5
76
29 1
87.;-
M 5
52 1
IV- J
V, 25
25 4
7fi:
61 75
51 4
IM 2
76 5
26 6
79 »
62
4*5
145 5
76 75
24
7;
62.25
61.5
184 5
77
21.2
6J t
62 5
53 5
l«i 5
77.25
18 6
55 >
62 75
51 3
1519
775
197
59 1
63
447
134 I
77 75
32.6
971
61.25
39 9
119 7
78
369
I Ml "
63 5
49.2
147 6
78 25
4(18
122 '
(S3,75
52.ft
157 8
78 5
47 6
N2I
64
51 V
155 7
7H 75
44 5
13.1 J
ru 25
61 2
18.1 6
79
393
H7i
64 5
58 3
174 9
79 25
55 1
165 *
4«
67.25
47 4
m:
82
65 6
l*Kr*»
67 5
52 3
156 9
82 25
frf»4
ii*»
67 75
55 A
Ifrf.H
82 5
f«R 4
2**5;
68
75 2
225 6
82 75
778
233.'
68 25
66 6
l'»R
83
773
11| «
68 5
52 A
157 8
83 25
68 3
205;
r# 75
49 5
1485
R1 5
67 1
iiii :*
69
67 9
203 7
83 75
484
145:
69 25
52 8
1584
84
489
ija:
m 5
59 R
179 4
8.1 25
65 I
195*
<,9 75
56 I
13
85 75
66 7
2'xi 1
71 25
62 4
|H7 2
86
727
218 I
7| 5
62 1
|K6 3
86 25
61 4
mt:
71,75
51 5
154 5
86 5
483
14< i.
72
594
178 2
86 75
493
147«
7225
61.3
18.1 9
87
563
168«.
72 5
58 1
174 3
R7.25
60 5
181 f
72.75
648
194 4
87 5
593
177«
73
54 6
163 8
87.75
708
212.»
7325
73 5
594
58.6
178 2
1758
*8
88.25
66 7 •
66,5
21*1 I
199 f
73.75
63 8
191 4
8H5
632
lR9r
74
52.3
156 9
88 75
587
176 1
74 25
50.7
152 1
R9
456
1361
74 5
55 3
1659
R9 25
38 5
115?
74 75
51 7
155.1
895
43 2
1291
75
46.3
I3R9
89 75
493
147 S
75.25
45 3
1359
""MS
669
2<«>"
34
-------
Oil 25
filfi
l"OR
|i»5
524
157 2
VU.S
m i
2«»7 3
H'5 25
4A3
13*9
VII75
5? 7
1A| I
105.5
43.2
12V ft
Ml
4M jt
lii y
H»5 75
4| 8
1254
Ml .25
¦li!
I2«»
I0A
35 4
Him 2
VI 5
34 3
|02 «
It* 25
2«4
*H 2
VI.75
31 A
94,H
W6.5
29,3
87 9
31.H
95.4
11*75
31 4
V4 2
92 25
2 A,V
WI7
107
34 5
103 5
«2 5
2f».3
7* V
l« 17.25
OA V
I4i>7
92.75
2^. J
7*9
107.5
41 *
1254
V.I
22.7
*«R 1
IU7.75
4| «
125 7
V3.25
2A H
W>4
|l*
41 2
129 A
V.I 5
23.4
70.2
11*25
3R 2
114 A
V3.75
2(1.4
6i,:
lis* 5
40*
122 4
U4
37 A
112 r
H«R 75
45 5
l.i'« 5
V4 25
44 A
131 H
lie;
44 5
133 5
94 5
41 7
125.1
1W25
44 V
134 7
<14 75
42.5
127 5
H»» 5
4ft I
14-1 1
V5
4(1
!2<«
H"J 75
51 *
155 4
VS. 2$
A 2 7
IRK 1
1 in
5«7
|7o |
**5 5
45
13*i
11(125
52 9
15«7
VS. 75
77 2
ill A
1 Jii.5
4| V
125 7
VA
fil 5
1*4.5
lit) 75
57 5
172 5
Vr» 25
72 K
2IH4
HI
5A4
Oi9 2
ua 5
«5 A
|<*. R
III 25
51 *
155 1
VA 75
to. 5
211 5
1115
45 I
1.15 3
V7
«»v
IR2 7
111.75
M |
|«2.3
V7 25
5*5
175 5
111
71 5
214 5
975
53 2
15U A
112 25
65.1
1959
97.75
55 4
If* 2
112 5
M
1«2
W
fW 2
|77 A
112 75
5« 7
I7A 1
VH.25
59
177
113
5-1 1
1*2.3
UH 5
57
171
113 25
52.2
l 5
AI.2
IRlfi
115 25
423
I2A9
UK 1.75
fil V
IRS 7
115.5
«)V
1*2 7
lot
51 V
155 7
115 75
A2.2
IHAfi
101.25
41*. 7
HA I
lift
524
157 2
im 5
5R 5
175 5
IIA25
51.7
155 1
mi.75
SB.2
174 A
1 |A 5
5.1.3
15° *J
Ml2
fil.6
IK/1 R
IIA 75
60. V
|R2 1
102.25
57.2
171 fi
117
48.2
"|44 A
102.5
566
IA9K
117.25
484
145 2
10275
4R*
1458
117 5
43B
131 J
HO
47.R
1434
117.75
42.3
12*9
103.25
32.R
I5R4
11*
<03 .
120.9
ll«5
53 4
160.2
11*25
3*3
114V
103.75
547
Jf*U
MRS
453
1359
m
59.*
170 4
11R.75
WR
|A4 4
1114.25
61.3
1*3 V
'.IV
4R.5
145 5
104.5
64
IV2
1 19.25
446
133 R
1(14 75
60.9
1H2.7
119.5
4T,4
139 2
35
-------
»
119.75
40.3
120
44.6
120.25
44.3
120.5
60.9
121
58.1
121.25
61.6
121.5
46.9
121.75
33.3
122
39.4
122.25
39.8
122.5
40.5
122.75
48.8
123
48.9
123.25
44.1
123.5
48.8
123.75
46.1
124
42.8
124.25
36
124 .5
39
124.75
46.4
125
29.8
125.25
30.7
125.5
29.6
125.75
28.3
126
25.2
126.25
24.3
Average**
49.5
120.9
133,
15* . y
293.7
174.3
184.8
140.7
117.9
118.2
119.4
121.5
146.4
146.7
132.3
146.4
138.3
128.4
108
117
139.2
89.4
?2.1
68.8
84.3
75.6
72.9
148.5
Check span gases- 67.51
Check zero gases- 1.6%
4
m
36
------- |