Report No. 77-NMM-5
O I Mil ^H PFIZER, INC.
VICTORVILLE, CALIFORNIA
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Emission Measurement Branch
Research Triangle Park. North Carolina
-------
EMISSION SOURCE TEST FROM A BAGHOUSE
SERVING A TALC GRINDING MILL
AT
PFIZER, INC.
Victorville, California
Task Order #12
Contract No. 68-02-1405
July, 1977
Robert J. Bryan, Director of Field Services
Robert L. Norton, Project Manager
Pacific Environmental Services, INC.
CORPORATE AND ENGINEERING 1930 14th Street Santa Monica, California 90404 Telephone (213) 393-9449
MIDWEST OPERATIONS Suite 228N 2625 Butterfield Road Oak Brook, Illinois 60521 Telephone (312) 325-5586
-------
TABLE OF CONTENTS
Section Page
I. INTRODUCTION 1-1
II. SUMMARY OF RESULTS II-l
III. PROCESS DESCRIPTION AND OPERATION III-l
IV. SAMPLING PORT LOCATION IV-1
V. SAMPLING PROCEDURES V-l
APPENDIX A. CALCULATIONS A-l
APPENDIX B. SOURCE TEST DATA SHEETS B-l
APPENDIX C. VISIBLE EMISSIONS DATA SHEETS C-l
APPENDIX D. CALIBRATION DATES D-l
APPENDIX E. SAMPLE IDENTIFICATION LOG E-l
APPENDIX F. EPA METHOD 17 F-l
LIST OF TABLES
Table Page
II-l DATA SUMMARY II-2
II-2 DATA SUMMARY II-5
II-3 PARTICLE SIZE DATA II-8
LIST OF FIGURES
Figure Page
IV-1 EXISTING STACK IV-2
IV-2 TEMPORARY STACK EXTENSION IV-3
IV-3 OUTLET SAMPLING POINTS IV-4
IV-4 INLET SAMPLING PORTS IV-4
IV-5 INLET SAMPLING POINTS IV-5
V-l OPACITY VERSUS TIME a) Outlet Test #1 b) Outlet
Test #2 c) Outlet Test #3 d) Outlet Test #4) V-4
-------
I. INTRODUCTION
Source tests were performed on the talc grinding facility
operated by Pfizer, Inc. located in Victorville, California.
Sampling took place on June 20, 21, 22, 1977 by Pacific Environ-
mental Services, Inc. (PES) for particulates using EPA Method
17 (see Appendix F), particle size, and opacity (EPA Method 9).
The source test team consisted of Robert Norton, Bob Bakshi, Keith
Duval and Ron Holliday all of PES. PES personnel were accompanied
on site by EPA observers Bob Martin and Jim Eddinger.
The facility tested consisted of a pebble grinding mill
exhausting into a pulse-jet baghouse. Samples were taken both at
the inlet and outlet of the baghouse for particulate grain loading,
at the inlet for particle size data. Opacity observations were
performed at the outlet. A more detailed description of the process
can be found in Section III and of the sampling procedures in
Section V.
All field data sheets and sample calculations can be found
attached to this report.
-------
II. SUMMARY OF RESULTS
Tabular data on results from all tests can be found in this
section, (Tables II-l and II-2). The average baghouse outlet grain
loading found for the four test runs was 0.0285 grains/standard
cubic foot. The baghouse inlet grain loading was found to be 6.18
grains/standard cubic foot. This yields an average baghouse
efficiency of 99.54%.
Tabular results for the particle size breakdown can be found
in Table I1-3.
II-l
-------
Table II-l
DATA SUMMARY
Run Number
Test Date
Sampling Time, 24-hour clock
D Sampling Nozzle Diameter, in.
Tt Net Time of Test, Min.
P. Barometric Pressure, in. Hg Absolute
P Average Orifice Pressure Drop, in. H^O
V Volume of Dry Gas Sampled at Meter
m Conditions, DCF
T Average Gas Meter Temperature, F
V Volume of Dry Gas Sampled at Standard
mstd Conditions*, DSCF
V Total H20 Collected in Impingers and
w Silica Gel, ml
w Volume of Water Vapor Collected at
9 Standard Conditions, SCF
%M % Moisture in Stack Gas, by Volume
Md Mole Fraction of Dry Gas
% C02 Volume % Dry
% 02 Volume % Dry
% CO Volume % Dry
% N2 Volume % Dry
% EA Percent Excess Air
MWd Molecular Weight of Stack Gas, Dry Basis
Outlet #1
6-20-77
1300
0.25
120
27.26
1.92
87.946
134.9
71.50
17.9
0.84
1.17
0.9883
0
18.75
0
81.25
-
28.75
Outlet #2
6-21-77
0800
0.25
(0.1875)
120
27.32
1.87
85.776
117.0
72.04
26.3
1.24
1.69
0.9831
0
19.5
0
80.5
-
28.78
Outlet #3
6-21-77
1200 i
0.25
(0.1875) :
120
27.30
1.60
80.824
140.0
65.17
19.1
0.90
1.36
0.9864
0
19.5
0
80.5
-
28.78
aStandard Conditions - 20°C, 760 mmHg.
II-2
-------
Table II-l (continued)
DATA SUMMARY
Run Number
MW Molecular Weight of Stack Gas,
Wet Basis
Cp Pitot Tube Coefficient
T Average Stack Temperature °F
Np Net Sampling Points
P . Static Pressure of Stack Gas, in. Hg
P Stack Gas Pressure, in. Hg Absolute
V Stack Gas Velocity at Stack
Conditions, fps
Ar Stack Area, ft2
s
Q. Dry Stack Gas Volumetric Flow Rate
at Standard Conditions
DSCFM
Q Stack Gas Volumetric Flow Rate at
a Stack Conditions, ACFM
% I Percent Isokinetic
% 0 Percent Opacity
Particulates - probe, and filter catch
mg
gr/DSCF
gr/ACF
Ib/hr
Outlet
#1
28.62
0.87
143.0
24
0
• 27.26
45.64
1.54
3325.1
4216.8
81
0.21
100.6
0.022
0.018
0.62
Outlet
#2
28.60
0.87
135.3
24
0
27.32
49.42
1.54
3636.5
4566.7
89*
0.75
112.1
0.024
0.020
0.75
Outlet
#3
28.64
0.87
152.3
24
0
27.30
44.56
1.54
3195.6
4112.8
91*
2.01
115.2
0.027
0.022
0.75
*These samples required use of two different nozzle sizes because of the
wide range of velocities encountered in the duct.
II-3
-------
Table II-l (continued)
DATA SUMMARY
Run Number
Date
Volume of Gas Sampled - Nm ^a'
Average Stack Temperature - C
Stack Volumetric Flow Rate - Nm3/min^b^
Stack Volumetric Flow Rate - m /min^
Particulates - probe, cyclone, and
filter catch
mg
mg/Nm
3
mg/m
kg/hr
Outlet #1
6-20-77
2.02
61.7
94.17
119.42
100.6
49.8
40.5
0.28
Outlet #2
6-21-77
2.04
57.4
102.99
129.33
112.1
55.0
46.2
0.34
Outlet #3
6-21-77
1.85
66.8
90.50
116.47
115.2
62.3
50.2
0.34
aDry normal cubic meter at 20 °C, 760mm Hg.
Dry normal cubic meters per minute at 20 °C, 760mm Hg.
cActual cubic meters per minute
II-4
-------
Table II-2
DATA SUMMARY
Run Number
Test Date
Sampling Time, 24-hour clock
D,, Sampling Nozzle Diameter, in.
n
T. Net Time of Test, Min.
P. Barometric Pressure, in. Hg
Absolute
P Average Orifice Pressure Drop,
m in. H20
V Volume of Dry Gas Sampled at Meter
m Conditions, DCF
T Average Gas Meter Temperature, °F
V Volume of Dry Gas Sampled at
std Standard Conditions, DSCF
V Total H20 Collected in Impingers
w and Silica Gel, ml
w Volume of Water Vapor Collected at
gas Standard Conditions, SCF
%M %Moisture in Stack Gas, by Volume
M. Mole Fraction of Dry Gas
%C02 Volume % Dry
%02 Volume % Dry
%CO Volume % Dry
%N2 Volume % Dry
%EA Percent Excess Air
MW. Molecular Weight of Stack Gas,
Dry Basis
Outlet #4
6-22-77
0750
0.25
(0.1875)
120
27.37
1.93
89". 188
119.3
74.76
26.2
1.24
1.63
0.9837
0
19.5
0
80.5
-
28.78
Inlet #1
6-22-77
1130
0.1875
24
27.37
1.07
14.194
120.2
11.85
3.4
0.16
1.34
0.9866
0
19.5
0
80.5
-
28.78
Particle Sizing
Inlet #2
6-22-77
1230
0.1875
30
27.37
1.6
20.317
128.4
16.75
6.7
0.32
1.85
0.9815
0
19.5
0
80.5
-
28.78
II-5
-------
Table II-2 (continued}
DATA SUMMARY
Run Number
MW Molecular Weight of Stack
Gas, Wet Basis
Cp Pi tot Tube Coefficient
T Average Stack Temperature °F
N Net Sampling Points
P t Static Pressure of Stack
51 Gas, in. Hg
P Stack Gas Pressure, in. Hg
Absolute
V Stack Gas Velocity at Stack
Conditions, fpm
2
A Stack Area, ft
s
Q. Dry Stack Gas Volumetric Flow
Rate at Standard
Conditions ,c DSCFM
Q Stack Gas Volumetric Flow
a Rate at Stack Condi-
tions, ACFM
% I Percent Isokinetic
% 0 Percent Opacity
Particulates - probe
and filter catch
mg
gr/DSCF
gr/ACF
Ib/hr
Outlet #4
28.60
0.87
136.8
24
0
27.37
49.56
1.54
3646.2
4579.4
95*
1.28
199.5
0.041
0.034
1.29
Inlet #1
28.64
0.87
165.7
12
-0.29
27.08
58.43
1.23
3250.3
4312.3
97
-
4742.2
6.18
5,16
172.06
Particle Sizing
Inlet #2
28.58
0.87
164.7
1
-0.29
27.08
70.07
1.23
3883.5
5171.3
92
-
N.A.
*This sample was collected using two different nozzle sizes because of the wide
range of velocities encountered in the duct.
II-6
-------
Table II-2 (continued)
DATA SUMMARY
Run Number
Date
Volume of Gas Sampled - Mm ^
Average Stack Temperature - °C
Stack Volumetric Flow Rate -
Nm3/min(b)
Stack Volumetric Flow Rate -
m3/min(c)
Particulates - probe
and filter catch
mg
mg/Nm
kg/hr
kg/Mton
Particulates - total catch
mg
mg/Nm
3
mg/m
kg/hr
kg/Mton
Percent impinger catch
Outlet #4
6-22-77
2.53
58.2
103.26
129.69
199.5
78.9
66.1
0.59
Inlet #1
6-22-77
0.40
74.3
92.05
122.12
4742.2
11,855.5
9897.7
78.05
Inlet #2
6-22-77
0.58
73.7
109.98
146.45
aDry normal cubic meter at 20°C, 760 mm Hg.
Dry normal cubic meter per minute at 20°C, 760mm Hg.
cActual cubic meters per minute
II-7
-------
Table 111-3. PARTICLE SIZE DATA
Particle Size
Range
1-5
5-10
10-15
15-20
20-25
25-30
30-35
35-40
40-45
45-50.
50-55
55-60
60-65
65-70
70-75
Total Count
Sampl
Particle
Count
2925
1855
1024
759
476
334
210
139
87
67
34
21
12
9
7
7959
e #1
% of
Total
36.8
23.3
12.9
9.5
6.0
4.2
2.6
1.7
1.1
0.8
0.4
0.3
0.2
0.1
0.1
100
Sample
Particle
Count
3181
1898
998
895
497
366
229
142
104
94
45
36
15
9
11
8520
n
% of
Total
37.3
22.3
11.7
10.5
5.8
4.3
2.7
1.7
1.2
1.1
0.5
0.4
0.2
0.1
0.1
100
Average
Particle
Count
3053
1876
1011
827
486
350
220
140
96
80
40
28
14
9
9
8239
% of
Total
37.1
22.8
12.3
10.0
5.9
4.2
2.7
1.7
1.2
1.0
0.5
0.3
0.2
0.1
0.1
100
00
-------
III. PROCESS DESCRIPTION AND OPERATION
A. GENERAL
The talc ore is trucked to the processing plant from both
underground mines and surface quarries. At the plant, the raw
ore is first reduced in size by a primary crusher. A secondary
crusher is used to further reduce the ore. After crushing, the
ore is conveyed to one of three grinding lines for product sizing.
B. PEBBLE MILL
The pebble mill is a cylindrical, horizontal, slow-speed
rotating drum containing a mass of pebbles as the grinding media.
The baghouse controlling particulate emissions from the
pebble mill was manufactured by Mikro Pul and contains 96 bags.
The bags are cleaned by a pulse air system every fifteen seconds.
New bags were installed two weeks prior to testing. The baghouse
is designed for an air-to-cloth ratio of 7:1. The dust collected
by the baghouse is recycled back to the process.
C. PROCESS OPERATION
During the testing, the operation of the process was normal
according to the operators. Since only a few process meters or
instruments were available to monitor, discussions with plant per-
sonnel were used to determine if the process was operating normally
during the testing. In addition, a pressure drop meter was installed
across the baghouse. The pressure drop across the baghouse ranged
from 2 inches of water (outlet run No. 1) to 9 inches of water
(outlet run No. 3). During the testing, three different product
grades were processed by the pebble mill.
III-l
-------
IV. SAMPLING PORT LOCATION
At the outlet the Pebble mill baghouse, sampling could
not be conducted in the existing stack. The stack consisted of a
four foot rectangular extension attached directly to the blower
with a weather cover attached (see Figure IV-1). A flow control
damper was located at the exit point of the stack. Because of
these existing sampling conditions, a temporary stack extension was
constructed and attached at 90° to the existing stack (see Figure
IV-2). A flow diverter was installed to make a more gentler flow
bend and reduce the problem of particle impingement on the stack
walls. Sampling was conducted in the rectangular stack approxi-
mately four stack diameters downstream from the flow diverter and one
stack diameter upstream from the stack exit. Using Figure 1-1 from
Method 1 (Federal Register, Tuesday June 8, 1976) and the curve for
small ducts (stack equivalent diameter 15 inches), twenty-four
sampling points were used. The sampling points used are shown in
Figure IV-3.
Sampling ports were installed in the baghouse inlet duct in
a straight section of ducting (15 inches in diameter) located inside
the mill building. This location was selected because of the length
of straight ducting between the disturbances. The location of the
sampling ports are shown in Figure IV-4. Twelve sampling points
(six on a diameter) were used as suggested by the E'PA Project Officer,
Mr. Bob Martin. By using twelve sampling points and sampling two
minutes at each point, the sampling could include a complete tra-
verse of the duct before the pressure drop, due to the high grain
loading, adversely effected the operation of the test. The sampling
points used are shown in Figure IV-5. The particle sizing sample
was taken for thirty minutes on,the inlet at one sampling point[9(16)],
IV-1
-------
Weather Con
Existing
Stack
BloweH
Motor
Blower
Flow Damper
Baghouse
Figure IV-1. EXISTING STACK
IV-2
-------
Temporary
Stack
Extension
Sampling Port
Existing
Stack
Permanent Hand Rail
(PES Installed)
Figure IV-2. TEMPORARY STACK EXTENSION
IV-3
-------
19
18 '
7
6
20
17
8
.5
13 1/2 in ^
21
16
9
4
22
15
10
3
23
14
11
2
24
13
12
1
16 3/8
1
a) Runs Outlet #1, 2, 4.
1
7 '
18
19
2
8
17
20
3
9
16
21
4
10
15
22
5
11
14
23
6
12
13
24
(b) Run Outlet #3
Figure IV-3. OUTLET SAMPLING POINTS
IV-4
-------
Sampling Ports
From Mill
Figure IV-4. INLET SAMPLING PORTS
To Baghouse
Existing
Scaffolding
15 in.
Figure IV-5. INLET SAMPLING POINTS
IV-5
-------
V. SAMPLING PROCEDURES
A. SUMMARY
Particulate samples were obtained using EPA Method 17 (Federal
Register, September 14, 1976) which included an tn-stack filter.
Fiber glass thimbles (19 x 90 mm) were used for both inlet and out-
let mass loadings and an alundum thimble was used for the particle
sizing sample. Opacity observations were made using EPA Method 9
(Federal Register, December 23, 1971).
B. EQUIPMENT
Flow control sampling equipment used for the Pfizer, Inc.
source test was manufactured by Microchemical Specialties Company
(Misco) and consisted of a Model 7200 CM control module (modified
with magnehelic gages instead of manometers). Stack temperatures
were measured using an iron-constantan thermocouple and a portable
potentiometer manufactured by Thermo Electric ("Minimite" Model
31101). Flue gas analyses were performed using a Fyrite test kit
manufactured by Bacharach Instruments (#10-5,000).
Alundum thimbles of 0.1 micron porosity were used for the
particle sizing sample (Western Precipitation RA-84). The fiber-
glass thimbles used for the mass loading samples were obtained from
BGI, Inc. (Schleicher & Schuell, BGI No. 603GV/19 x 90).
Isokinetic conditions were maintained using a stainless steel
nomograph from Nutech Corporation (Model 211-2).
C. SAMPLING PROCEDURES
The PES test crew arrived on site on Monday, June 20, 1977 and
installed all temporary stack extensions, supports etc. PES also
made arrangements to have a welder on site to install sampling ports
in the inlet ductwork and to assemble and install a safety handrail
V-l
-------
atop the baghouse where most work was to take place. Sampling
at the outlet using EPA Method 17 began in the afternoon of June
20, 1977 when Outlet run #1 consisting of sampling twenty-four points
at a rate of six minutes per point (total sample time = two hours)
was conducted.
Outlet runs #2 and #3 were performed on Tuesday, June 21, 1977
and differed from Outlet #1 in that two nozzles had to be used.
Both a 0.25 inch diameter and 0.1875 inch diameter nozzle were used
because the installed flow diverter resulted in an uneven flow
pattern in the duct. The 0.25 inch diameter nozzle was used for
three traverses across the duct while the 0.1875 inch diamter nozzle
was used for the one traverse nearest the top of the duct where the
velocity was highest (see data sheets Appendix B). Leak checks
were performed during each test as described in EPA Method 17 which
included leak checking the in-stack filter system in the stack after
it had reached a temperature equilibrium.
Outlet test #4, inlet test #1 and inlet test #2 (particle
sizing sample) were all performed on Wednesday, June 22, 1977. The
outlet sample was fun for two hours as all previous tests and used
two sampling nozzles as in outlet tests #2 and #3. Inlet test #1
was run for twenty-four minutes (two minutes per point) because it
was feared the high grain loading would result in a rapid increase
in pressure drop which would exceed the capacity of the sampling
apparatus. The particle sizing sample was taken for thirty minutes
at a point selected which was approximately the average flow encoun-
tered in the previous inlet test.
Sample recovery procedures were performed on site in a
laboratory located in the mill building. The filter thimbles were
removed from the thimble holder and carefully placed in a sample
container. The nozzle and all parts of the thimble holder, located
before the filter, were washed with acetone and the washings placed
in a teflon-capped glass sample bottle. Both the acetone washings
V-2
-------
and thimble sample containers were labeled using EPA labels and
logged on an EPA Sample Identification Log.
The moisture collected in the impingers was measured and
recorded and the water was then discarded. The silica gel moisture
trap was also weighed on site and the silica gel was retained for
regeneration.
Field data sheets for all sampling runs can be found in
Appendix B.
D. OPACITY OBSERVATIONS
Opacity observations were made during each particulate test
at the baghouse outlet by two members of the test crew. Each crew
member took part in the opacity observations and all members were
certified as visible emissions evaluators by the California Air
Resources Board. Dates of the last certification can be found in
Appendix D.
Opacity observations were performed simultaneously, by two
crew members during the two hour test with opacity observations
made every fifteen seconds. The average opacity for each test run
was as follows:
Test #1 - 0.21
Test #2 - 0.75
Test #3 - 2.01
Test #4 - 1.28
Figures V-l illustrates the average observed opacity versus time
for the four outlet particulate tests.
E. LABORATORY ANALYSIS
Analyses for mass loading of the particulate matter were
performed in the PES laboratory. The acetone washings were trans-
ferred to a tared drying dish and allowed to evaporate at room
V-3
-------
20 -
15
10
5
20
15
10
5
20
15
10
5
\
1 2
Time, Mrs.
(a)
20
15
10
5
1 2
Time, Hrs. .
(c)
1 2
Time, hrs,
(b)
^— —
1 2
Time, Hrs.
(d)
Figure V-l. OPACITY VERSUS TIME [a) Outlet Test #1 b) Outlet Test #2 c) Outlet Test #3
d) Outlet Test #4]
-------
temperature in a laboratory hood. The sample was then desiccated
for twenty-four hours, weighed, and the weight increase noted.
The fiber glass filter thimbles were placed in the desiccator,
dried for twenty-four hours, weighed and the particulate weight
noted. The particulate found on the filter was added to that from
the acetone wash to yield the total particulate catch. All labora-
tory weighings were performed on a Torbal analytical balance to the
nearest 0.1 mg.
The particle sizing sample was obtained in an alundum thimble
and brought back to the PES laboratory for sample preparation. The
thimble was washed thoroughly with isopropyl alcohol to remove the
particulate matter and to place the particles into solution. This
solution was then sent to Spectrex Corporation where the particle
size range determinations were performed using a scanning laser
beam particle counter (Spectrex Prototron Particle Counter Model
ILI 1000). The number of particles in each of fifteen five micron
ranges (from 1 to 75 microns) were obtained as the results.
V-5
------- |