United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 81-PLY-4
January 1982
Air
Plywood/Veneer
Emission Test Report
Georgia-Pacific
Springfield Plant
Springfield, Oregon
-------
PLYWOOD/VENEER
EMISSION TEST REPORT
GEORGIA-PACIFIC PLYWOOD PLANT
SPRINGFIELD, OREGON
JUNE 1981
Environmental
Consultants, Inc.
EMB Report 81-PLY-4
ESED Project 80/02
EPA Contract No. 68-02-3543
Work Assignment No. 1
TRC Project No. 1460-E80-51
Prepared for:
C.E. Riley
Task Manager
Prepared by:
Peter W. Kalika, P.E.
Program Manager
Eugene A. Brackbill, P.E.
Work Assignment Manager
John H. Powell
Project Scientist
Eric A. Pearson
Project Scientist
January 1982
800 Connecticut Blvd.
East Hartford, CT 06108
(203) 289-8631
-------
This report has been reviewed by the Emission Standards and Engineering
Division, Office of Air Quality Planning and Standards, Office of Air,
Noise and Radiation, Environmental Protection Agency, and approved for
publication. Mention of company or product names does not constitute
endorsement by EPA. Copies are available free of charge to Federal
employees, current contractors and grantees, and nonprofit organizations -
as supplies permit - from the Library Services Office, MD-35, Environ-
mental Protection Agency, Research Triangle Park, NC 27711.
Order: EMB Report 81-PLY-4
-------
PREFACE
The work described herein was conducted by personnel from TRC - Environ-
mental Consultants, Inc., Research Triangle Institute (RTI), Del Green Asso-
ciates (DGA); CH~MHill, Engineers, Planners, Economists and Scientists;
Pollution Control Science, Inc., (PCS); Georgia-Pacific (G-P) in Springfield,
Oregon; the National Council of the Paper Industry for Air and Stream Improve-
ment, Inc. (NCASI); and the United States Environmental Protection Agency
(EPA) Emission Measurement Branch (EMB).
The scope of work was issued under EPA Contract 68-02-3543, Work Assign-
ment 1. The work was performed under the supervision of Eugene A. Brackbill,
P.E., TRC work assignment manager, and John H. Powell, TRC field crew chief.
Robert L. Chessin of RTI monitored process operations and was assisted by
Paul Willhite of DGA. RTI was responsible for preparing Section 3 and Appen-
dix I of this report, both of which deal with process descriptions and opera-
tions. Mark S. Boedigheimer supervised Method 5X analyses performed by
CHJIHill. David Robinson supervised Method 25 analysis performed by PCS.
Victor Dallons supervised NCASI sampling and analysis activities as well as
providing helpful suggestions and comments in support of the test program.
Mitch Steffensen and Pete Fetter of Georgia-Pacific, provided invaluable
assistance and guidance to TRC, EPA and RTI in the performance of the test
program. Clyde E. Riley, Office of Air Quality Planning and Standards
(OAQPS), Emission Measurement Branch, EPA, served as task manager and was
responsible for coordinating the test program.
Edwin J. Vincent, OAQPS, Chemical and Petroleum Branch, EPA, served as
project lead engineer. He was also responsible for coordinating and directing
the process operations monitoring.
-11-
-------
TABLE OF CONTENTS
SECTION PAGE
PREFACE ii
1.0 INTRODUCTION 1-1
1.1 Background 1-1
1.2 Summary of Process and Emissions 1-2
1.3 Applicability of EPA Reference Test Methods 1-3
1.3.1 EPA Method 5X (Provisional) 1-3
1.3.2 EPA Method 25 1-5
1.3.3 Comparability of Test Methods 1-6
1.4 Measurement Program Summary 1-7
1.4.1 Scrubber Inlet 1-7
1.4.2 Scrubber Outlet 1-7
1.4.3 Georgia-Pacific Scrubber System 1-8
1.4.4 Fugitive Emissions 1-8
1.4.5 Ambient Air Measurements 1-8
1.4.6 Clean-Up Evaluations 1-9
1.5 Report Sections 1-9
2.0 SUMMARY AND DISCUSSION OF RESULTS 2-1
2.1 Background and Definitions 2-1
2.1.1 Particulate Emissions 2-1
2.1.2 Condensible Emissions 2-2
2.1.3 Noncondensible Emissions 2-2
2.1.4 Total Organic Emissions 2-2
2.2 Method 5X - Particulate/Condensible
Organics Emission Tests 2-3
2.2.1 Scrubber Inlet 2-6
2.2.2 Scrubber Outlet 2-8
2.3 Method 25 - Total Organic Tests 2-13
2.3.1 Scrubber Inlet 2-16
2.3.2 Scrubber Outlet 2-16
2.4 Visible Emissions 2-21
2.5 Scrubber Operational Summary 2-21
2.6 Summary of Fugitive Emissions 2-23
2.7 Ambient Air Measurements 2-23
2.8 Clean-Up Evaluation 2-27
2.9 Possible Test Interferences 2-27
3.0 PROCESS DESCRIPTION AND OPERATIONS 3-1
3.1 Process Equipment 3-1
3.2 Emission Control Equipment 3-1
3.3 Production and Control Equipment Monitoring 3-2
3.4 Process Operating Conditions During Test Program. . . 3-2
4.0 DESCRIPTION OF SAMPLING LOCATIONS 4-1
4.1 Scrubber Inlet 4-1
4.2 Scrubber Outlet 4-1
4.3 Scrubber Operational Measurement Locations 4-5
4.4 Fugitive Emissions 4-5
-111-
-------
TABLE OF CONTENTS (Continued)
SECTION PAGE
5.0 SAMPLING AND ANALYTICAL METHODS 5-1
5.1 EPA Reference Methods 5-1
5.2 Preliminary Measurements 5-2
5.3 Measurements for Particulate, Condensible and
Noncondensible Emissions 5-2
5.3.1 EPA Method 5X (Provisional) - Particulate and
Condensible Organic Compounds 5-2
5.3.2 EPA Reference Method 25 - Condensible and
Noncondensible Organic Compounds 5-10
5.4 Preliminary Moisture Determination 5-22
5.5 Preliminary Velocity Determination 5-22
5.6 Visible Emissions 5-23
5.7 Pressure Drop Measurements 5-23
5.8 Scrubber Solution Samples 5-23
5.9 Fugitive Emissions 5-24
5.10 Ambient Temperature and Relative Humidity 5-25
6.0 QUALITY ASSURANCE 6-1
6.1 Method 5X 6-1
6.2 Method 25 6-3
6.3 Method 9 6-3
-IV-
-------
LIST OF FIGURES
SECTION PAGE
1-1 Veneer Dryer Exhaust and Scrubber System 1-4
4-1 Veneer Dryer Exhaust and Scrubber System
Sampling Locations 4-2
4-2 Scrubber Inlet Sampling Port Configuration and
Traverse Point Locations 4-3
4-3 Scrubber Outlet Sampling Port Configuration and
Traverse Point Locations 4-4
5-1 Modified EPA Particulate and Condensible Organics
Sampling Train 5-4
5-2 Method 25 Sampling Train 5-11
5-5 Method 25 Flow Control Assembly Adjustment 5-13
5-6 TRC Nonmethane Organic Analyzer 5-18
5-7 TRC Condensate Recovery and Conditioning Apparatus . . 5-21
LIST OF TABLES
2-la Summary of Method 5X Particulate and Condensible
Organic Collection Efficiency for Georgia-Pacific
Scrubber System (English Units) 2-4
2-lb Summary of Method 5X Particulate and Condensible
Organic Collection Efficiency for Georgia-Pacific
Scrubber System (Metric Units) 2-5
2-2 Summary of Method 5X Particulate and Condensible
Organic Measurements for Gases Entering the
Georgia-Pacific Scrubber System 2-7
2-3 Summary of Method 5X Particulate and Condensible
Organic Measurements for Gases Exiting the
Georgia-Pacific Scrubber System 2-9
2-4 Comparison of Particulate and Condensible Measured
Emissions with Calculated Emissions 2-10
2-5a Summary of Method 25 Total Organic Collection
Efficiency for Georgia-Pacific Scrubber System
(English Units) 2-14
2-5b Summary of Method 25 Total Organic Collection
Efficiency for Georgia-Pacific Scrubber System
(Metric Units) 2-15
-v-
-------
LIST OF TABLES (Continued)
SECTION PAGE
2-6 Summary Method 25 Individual Total Organic
Measurements for Gases Entering the Georgia-Pacific
Scrubber System 2-17
2-7 Summary Method 25 Individual Total Organic
Measurements for Gases Exiting the Georgia-Pacific
Scrubber System 2-18
2-8 Summary of Method 25 Individual Total Organic Trap,
Tank Measurements for Gases Entering the
Georgia-Pacific Scrubber System 2-19
2-9 Summary of Method 25 Individual Total Organic Trap,
Tank Measurements for Gases Exiting the
Georgia-Pacific Scrubber System 2-20
2-10 Georgia-Pacific Scrubber System Operational
Data Summary 2-22
2-lla Fugitive Emission Summary 2-24
2-llb Fugitive Emission Summary 2-25
2-llc Fugitive Emission Summary 2-26
2-12 Method 5X Clean-up Evaluation Results 2-28
-vi-
-------
1.0 INTRODUCTION
1.1 Background
Section 111 of the Clean Air Act of 1970 charges the administrator of the
United States Environmental Protection Agency with the responsibility of
establishing Federal Standards of Performance for New Stationary Sources
(SPNSS) that may significantly contribute to air pollution. When promulgated,
these standards of performance for new stationary sources are to reflect the
degree of emission limitation achievable through application of the best dem-
onstrated emission control technology. Emission data collected from con-
trolled sources in the plywood industry will provide a portion of the database
used by EPA to develop SPNSS.
EPA's Office of Air Quality Planning and Standards selected the Georgia-
Pacific (G-P) plywood plant in Springfield, Oregon, as a site for an emission
test program because it is considered to employ process and emission techno-
logy representative of modern plywood manufacturing plants.
The test program was designed to determine the emission rate of particu-
late matter and condensible and noncondensible organic material emitted from
the veneer drying operation. A second objective was to measure the collection
efficiency of the Georgia-Pacific scrubber system for condensible and non-
condensible organic emissions.
TRC - Environmental Consultants, Inc. was retained by the EPA Emissions
Measurement Branch (EMB) to perform emission measurements at the G-P plywood
plant in Springfield, Oregon. Testing was performed during the week of June
8, 1981 on the veneer dryer emissions and their pollution control, a G-P
scrubber system. This report has been prepared in accordance with EPA Con-
tract No. 68-02-3543 under the provisions of Work Assignment No. 1.
1-1
-------
The Research Triangle Institute (RTI), the New Source Standard (NSS) con-
tractor/ was responsible for coordinating the overall test program with G-p
personnel and for assuring that process and control equipment operating con-
ditions were suitable for testing. All process data were monitored and
recorded by RTI. Fugitive emissions from the veneer dryers/ ambient air tem-
perature and relative humidity were monitored and recorded by RTI.
Additional testing for total organic compounds was performed by the
National Council of the Paper Industry for Air and Stream Improvement, Inc,
(NCASI) simultaneously with the TRC test program. This testing was performed
at the request of the American Plywood Association (APA) for research purposes
and to provide an additional measure of quality assurance.
1.2 Summary of Process and Emissions
The G-P Springfield plant is a combination veneer and lay-up facility, and
is considered to employ process and emission control technology representative
of modern plywood manufacturing plants. The plywood production rate for the
drying operation is approximately 800,000 square feet (3/8-inch basis) per
24-hour day.
The veneer drying operation begins after the veneer has been peeled from
the log at the lathe operation and is transferred to the drying operation.
Here, the veneer is continuously hand-fed onto the dryer feed conveyor and
into the dryer. The purpose of the operation is to thermally drive the mois-
ture out of the veneer in preparation for the lay-up and laminating operations
which follow. During the drying operaion, organic compounds are driven out of
the veneer. These organic compounds are the emissions of interest.
1-2
-------
The G-P Springfield plant has four veneer dryers. Each is a steam heated,
multideck unit, with the number of drying zones varying. Each dryer has two
exhausts from the heated zones, except dryer 4, which has three exhausts. The
nine exhausts are ducted to a common manifold which carries the dryer emis-
sions to the G-P scrubber system. A schematic drawing of the veneer dryer
exhaust system is presented in Figure 1-1.
1.3 Applicability of EPA Reference Test Methods
EPA is required to publish a national reference test method for each reg-
ulated source category and pollutant for which a New Source Performance
Standard (NSPS) is established. Reference test methods are usually specified
by a State regulatory agency during the State Implementation Planning process
and may be different from national reference test methods.
The purpose of establishing a national reference test method is to ensure
that emission data collected from a specific source is representative of that
source and comparable to data collected at other designated sources. The pri-
mary purpose of this test program was to collect emission data using standard-
ized test methods which allow the data to be evaluated to develop a national
SPNSS. Two different test methods were selected by EPA to measure emissions
from plywood veneer drying operations. These methods are briefly described in
the following subsections and are described in detail in Section 5.
1.3.1 EPA Method 5X (Provisional)
Provisional Method 5X is similar to the Oregon Department of Environmental
Quality (ODEQ) Method 7 used to measure condensible organic emissions. EPA
Method 5X measures particulate matter and condensible organic matter. "Par-
ticulate matter" is defined as any finely divided solid or liquid material,
1-3
-------
Spray-JJ
Nozzles
Dryer 4
Dryer 2
Dryer 1
Figure 1-1. Veneer Dryer Exhaust and Scrubber System
Georgia-Pacific Plywood Plant
Springfield, Oregon
1-4
-------
other than uncombined water, that condenses at or above the filtration temper-
ature range of 350 +25 F (177 +_14 C) , and is collected by the probe and
filter (front half of the sampling train). "Condensible organic matter" is
defined as any material remaining after extraction, filtration and ambient
evaporation of the ether-chloroform extract of the impinger portion of the
sampling train. Particulate matter and condensible organic matter are quanti-
fied gravimetrically and results are expressed as the mass of collected mater-
ial.
The purpose of the 350 F filtration temperature is to precondition the
Method 25 slipstream sample being withdrawn from the Method 5X sample stream.
This temperature was selected on the basis of average veneer dryer operating
temperatures throughout the industry. This temperature condition excludes
from the Method 25 samples only matter than can condense at or above 350 F.
It does not affect Method 5X results because the remaining sample is caught in
the condenser portion of the train.
1.3.2 EPA Method 25
EPA Reference Method 25, as promulgated in the October 3, 1980 Federal
Register (volume 145, no. 194, 65959 ff.), applies to the measurement of
organic compounds as total gaseous nonmethane organics (TGNMO). Emissions are
expressed as equivalent carbon (C,) mass. Method 25 sample fractions are
separated by a gas chromatographic column, oxidized to carbon dioxide (C0_),
and reduced to methane (CH.) prior to analysis by flame ionization detector
(FID). Since all the sample organic compounds are reduced to CH , the pro-
blems associated with the variable FID response characteristic for different
organic compound structures is eliminated. This allows comparison of emission
data on a uniform C basis. Method 25 is discussed in greater detail in
Section 5 of this report.
1-5
-------
Major procedural modifications made to Method 25 were required to measure
accurately emissions from plywood veneer drying facilities. These modifica-
tions are discussed in Section 5. An additional condensate trap immersed in
water ice was placed in the sampling train ahead of the standard dry ice im-
mersed condensate trap. The purpose of the additional trap is to condense
moisture that would freeze in the dry ice immersed trap and cause a premature
sample flow stoppage. In this manner gas stream moisture content, which may
range from 30 to 60 percent by volume, may effectively be reduced to 3 percent
or less before entering the dry ice immersed trap.
The use of the Method 5X sampling train as a sample preconditioner also
represents a major modification. In addition to the 350 F sample stream
temperature, isokinetic sample extraction from the source using Method 5X was
also deemed necessary to obtain a representative Method 25 sample. This is
particularly the case when moisture-saturated gas streams, such as those
following wet scrubbing devices, are being sampled. Entrained water droplets
may contain organic materials that would not be collected using the normal
Method 25 constant sampling rate procedure.
1.3.3 Comparability of Test Methods
Methods 5X and 25 are not related and measured results can not be compared
under any circumstances. Condensation temperatures differ by more than
100 F between the two methods, and consequently different condensible com-
pounds are collected by each method. In addition, it has been demonstrated
that Method 5X has limited collection capabilities for organic compounds with
high-vapor pressures. A loss of organic material is experienced even during
normal Method 5X sample recovery and analysis operations.
1-6
-------
1.4 Measurement Program Summary
The measurement program was conducted at the G-P Springfield facility dur-
ing the week of June 8, 1981. The emission tests were designed to measure the
veneer dryer organic emissions and to determine the collection efficiency of
the G-P scrubber system for those emissions. Tests were performed at the
veneer dryer exhaust duct (scrubber inlet) and at the outlet of the scrubber
system.
All emission testing was performed by TRC and NCASI personnel. RTI per-
sonnel monitored process operating conditions, while DGA personnel monitored
fugitive emissions/ ambient temperature and relative humidity. Scrubber oper-
ational data and solution samples were collected by TRC personnel.
1.4.1 Scrubber Inlet
Preliminary Measurements
Preliminary testing was performed on June 8 to determine volumetric flow
rate and stack gas moisture content.
Method 5X - Particulate and Condensible Organics Tests
Three Method 5X tests were performed, one each on June 9, 10, and 11, con-
currently with tests performed at the scrubber outlet.
Method 25 - Total Organic Tests
Eighteen Method 25 samples were taken at this location concurrently with
the Method 5X tests performed. Six Method 25 samples were taken concur-
rently with each Method 5X test.
1.4.2 Scrubber Outlet
Preliminary Measurements
Preliminary tests were performed on June 8 to determine volumetric flow
rate and stack gas moisture content.
Method 5X - Particulate and Condensible Organics Tests
Three Method 5X tests were performed at this location, one each on June 9,
10, and 11 concurrently with tests performed at the scrubber inlet.
1-7
-------
Method 25 - Total Organic Tests
Eighteen Method 25 samples were taken at this location concurrently with
the Method 5X samples (six per test run), and simultaneously with Method
25 samples taken at the scrubber inlet.
Method 9 - Visible Emissions
Scrubber outlet visible emissions were not monitored as planned because of
overcast sky background conditions. The scrubber outlet plume was bluish-
white and was therefore indistinguishable from the overcast sky. Overcast
skies were present on June 8, 9 and 10. During the last test day, June
11 / the sky began to clear and only scattered clouds were present in the
afternoon. However, the final test sequence was nearly completed by this
time. Consequently, no visible emission observations were recorded.
Although the scrubber outlet stack had an attached steam plume, the Method
9 observations were not cancelled because of this condition. The method
provides for attached steam plumes by requiring that observations be made
at the point where the condensed water vapor is no longer visible.
1.4.3 Georgia-Pacific Scrubber System
Static pressure upstream and downstream of the system induced draft fan
was measured with U-tube water manometers and recorded at 30-minute inter-
vals. These measurements were used to calculate pressure drop (AP) across
the scrubber system.
Scrubber solution samples were taken every 30 minutes during the scrubber
outlet Method 5X test period. One-hundred-mi samples were collected from the
scrubber recirculation tank every 30 minutes during each test. The individual
samples collected during each test were composited for analysis.
1.4.4 Fugitive Emissions
Fugitive emissions from the veneer dryers were monitored by DGA during
each Method 5X test.
1.4.5 Ambient Air Measurements
Ambient air temperature and relative humidity were monitored and recorded
by DGA at the beginning and end of each Method 5X test.
1-8
-------
1.4.6 Clean-Up Evaluations
Prior to any emission testing, two Method 5X sampling trains were prepared
and charged, ready to perform a test. The unexposed trains were then cleaned
according to the method and samples recovered. The samples were analyzed to
establish background and/or contamination levels from the .sample collection
equipment.
1.5 Report Sections
The remaining sections of this report present the Summary and Discussion
of Results (Section 2), Process Description and Operations (Section 3),
Description of Sampling Locations (Section 4), Sampling and Analytical Pro-
cedures (Section 5), and Quality Assurance (Section 6). Descriptions of
methods and procedures, field and laboratory data, and calculations are pre-
sented in various appendices as noted in the Table of Contents.
1-9
-------
2.0 SUMMARY AND DISCUSSION OF RESULTS
A summary of all emission measurements and collected data is presented in
this section. Section 2.1 provides a brief background discussion and defini-
tions of measured parameters. Section 2.2 presents Method 5X particulate/
condensible organics results with a complete breakdown and discussion of par-
ameters at both sampling sites. Method 25 total organic emission results are
described in detail in Section 2.3, which includes a discussion of emissions
at both sampling sites as well as a breakdown of major analytical data.
Section 2.4 discusses visible emissions observations. A summary of scrubber
operational data is presented in Section 2.5. Fugitive emissions are dis-
cussed in Section 2.6. A summary of ambient air measurements is presented in
Section 2.7. A full discussion of the Method 5X clean-up evaluation and
results may be found in Section 2.8.
2.1 Background and Definitions
The test program was designed to measure particulate matter/ condensible
and noncondensible organic material emitted from veneer dryers, and to deter-
mine the collection efficiency of the G-P scrubber system as a control for
those emissions.
2.1.1 Particulate Emissions
Particulate emissions are defined as any finely divided solid or liquid
o
matter, other than uncombined water, that condenses at or above 350 +25 F
(177 _+14°C) and is collected in the probe and filter (front half) of the
Method 5X sampling train.
2-1
-------
2.1.2 Condensible Emissions
Condensible emissions are defined differently in Methods 5X and 25.
Although called by the same name, these two sample fractions differ signifi-
cantly in content and composition and may not under any circumstances be com-
pared.
Method 5X condensibles are collected in glass impingers containing deion-
ized distilled water and immersed in a water ice bath, and on a back-up filter
following those impingers. Any material remaining after extraction, filtra-
tion and ambient evaporation of the impinger solution, plus any material
collected on the desiccated back-up filter, is defined as Condensible organic
matter. Quantification of this matter is done gravimetrically.
Method 25 condensibles are collected in two stainless-steel traps, one
immersed in water ice followed by another packed in dry ice. Material
collected in the traps is oxidized, reduced and analyzed by flame ionization.
Results are expressed as a concentration of carbon (C,).
2.1.3 Noncondensible Emissions
Noncondensible emissions are measured by Method 25 only and are those that
pass through both ice traps to the evacuated sample tank at the end of the
Method 25 train. These samples are oxidized, reduced and analyzed by FID.
Results are expressed as concentrations of carbon (C,).
2.1.4 Total Organic Emissions
Total organic emissions are those collected by the complete Method 25 sam-
pling train drawing a preconditioned sample slipstream from a Method 5X
train. These emissions include Condensible and noncondensible emissions as
defined above.
2-2
-------
2.2 Method 5X - Particulate/Condensible Organics Emission Tests
A summary of Method 5X data collected at the scrubber inlet and outlet is
presented in Tables 2-la (English units) and 2-lb (metric units) . These
tables include relevant emission data: stack gas temperature, moisture con-
tent and volumetric flow rate; veneer dryer production rate; and a summary of
the total measured particulate/condensible emissions by concentration, mass
emission rate, and emission rate per unit production.
Emission data are presented for the three test series. Testing was per-
formed concurrently at the scrubber inlet and outlet. Emissions at the scrub-
ber inlet averaged 18.3 Ibs/hr (8.29 kg/hr) or 0.53 lbs/1000 ft2 veneer on a
3/8-inch basis (2.56 kg/1000 m2 on 9.5 mm basis) for the three tests.
Emissions from the scrubber outlet averaged 14.9 Ibs/hr (6.77 kg/hr) or 0.43
lbs/1000 ft2 veneer (2.10 kg/1000 m2) for the three tests. The concentra-
tions of the emissions from the two sources, however, differed markedly. The
average scrubber inlet concentration was 0.164 gr/DSCF (0.376 g/NM3), while
the scrubber outlet averaged only 0.103 gr/DSCF (0.236 g/NM3) for the three
tests.
The removal efficiency of the G-P scrubber system for particulate/
condensible organics averaged 16.4 percent for the three tests. Efficiencies
ranged from 29 percent during test 1 to 6 percent during test 3.
Detailed summaries of this test data are presented in Sections 2.2.1 and
2.2.2 and in Appendix A. Sample equations and calculations are presented in
Appendix B. Field data sheets appear in Appendix C. Sampling logs and sum-
maries are shown in Appendix D. Calibration data for the Method 5X sampling
train are found in Appendix F. Laboratory analysis data are presented in
Appendix G.
2-3
-------
TABLE 2-la (English Units)
GUMMARY OF METHOD SX PARTICULATE AND CONUENS1BLE ORGANIC
COLLECTION EFFICIENCY FOR UEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood Plant, Springfield, Oregon
Run Number
Date
Emission Description!
Volume Gas Sampled (DSCF)a
Stack Gas Flow Rate (DSCFM)b
Stack Temperature (°F)
Percent Moisture of Volume0
Percent isoklnesis
M Scrubber Pressure Drop (inches lljO)
"•*• Production Rate (1000 ft«/hr)d
Particulate-Condensible Results
Total Catch
gr/DSCF
Ibs/hour
lbs/1000 ft1
Scrubber System Collection Efficiency
Run 1
6-9-81
Uncontrolled Controlled
45.3 38.3
12400 15200
303 166
31.9 38.3
07.4)e
116.5 120.4
14.2
35.7
0.198 0.115
23. Of 16. 3£
0.644 0.456
(») 29.1
Run 2
6-10-B1
Uncontrolled Controlled
42.1 37.3
12000 17500
314 166
34.0 37.5
(37.4)e
111.4 102.1
15.0
34.4
0.169 0.107
18. 5£ 15.9
0.538 0.462
14.1
Run 3
6-11-81
Uncontrolled Controlled
42.4 37.4
12600 16100
309 168
31.8 39.6
(39.3)e
99.2 111.3
13.9
34.1
0.124 O.OB6
13.4 12. 6£
0.393 0.370
5.97
Average
Uncontrolled Controlled
43.3 37.7
12300 16300
309 167
32.6 3U.5
(38.0)a
109.0 111.3
14.4
J4.7
U.164 0.10J
18.3 14.9
0.525 0.429
16.4
a Standard conditions! 29.92 inches Hg at 6B°F.
b Outlet volumetric flows (DSCFM) suspected to be slightly biased high due to possible wind Interference.
0 Outlet moisture results calculated with psychometric equation. (See Section 5.)
d 1000 square feet veneer per hour on 3/8 inch basis) includes trim factor; does not account for redry material.
e Theoretical moisture content at saturation ( )) all results based on this.
f Results are average of concentration and area ratio methods due to high isokinetic values. (Bee Table 2-4.)
-------
TABLE 2-lb (Metric Units)
SUMMARY OP METHOD 5X PARTICULATE AND CONDENSIBLE ORGANIC
COLLECTION EFFICIENCY FOR GEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood Plant, Springfield, Oregon
ro
I
ui
Run Number Run 1
Date 6-9-81
Emission Description! Uncontrolled Controlled
Volume Gas Sampled (NM')a 1.28 1.08
Stack Gas Flow Rate (NM»/Min)b 351 430
Stack Temperature (°C) 151 74.4
Percent Moisture of Volume0 31.9 38.3
(37.4)e
Percent Isokinesis 116.5 120.4
Scrubber Pressure Drop (mm HjO) 55.9
Production Rate (1000 m'/hr)d 3.32
Particulate-Condenslble Results
Total Catch
g/NM» 0.454 0.263
kg/hour 10. 4£ 7.39£
kg/1000 ft» 3.14 2.23
Scrubber System Collection Efficiency % 28.9
Run 2 Run 3
6-10-81 6-11-81 Average
Uncontrolled Controlled Uncontrolled Controlled Uncontrolled controlled
1.19 1.06 1.20 1.06 1.23 1.07
340 495 357 456 348 461
157 74.4 154 75.6 154 75.0
34.0 37.5 31.8 39.6 32.6 38. 5
(37.4)e (39.3)e (3B.O)8
111.4 102.1 99.2 111.3 109.0 111.3
59.1 54.7 56.7
3.20 3.17 3.22
0.387 0.245 0.284 0.197 0.376 0.236
8.39£ 7.21 6.08 5.72£ 8.29 6.77
2.62 2.26 1.92 1.80 2.56 2.10
14.1 5.92 16.3
a Standard conditional 760 nun llg at 20°C.
b Outlet volumetric flows (NM'/min) suspected to be slightly biased high due to possible wind interference.
c Outlet moisture calculated with psychometric equation. (See Section 5.)
d 1000 square meters veneer per hour on 9.5 millimeter basisi included trim factor) does not account for redry material.
e Theoretical moisture content at saturation ( )j all results based on this.
£ Results are average of concentration and area ratio methods due to high isoklnetic values. (See Table 2-4.)
-------
2.2.1 Scrubber Inlet
A summary of Method 5X data collected at the scrubber inlet is presented
in Table 2-2. Data presented include sample volume; stack gas flow rate, tem-
perature, and moisture content; isokinesis for each test; veneer production
rate; front half (particulate) and total (particulate/condensible) emissions.
Tests 1, 2 and 3 were performed at the scrubber inlet on June 9, 10, and
11, respectively. Measured particulate emissions ranged from 0.77 to 2.93
Ibs/hr (0.02 to 0.08 lbs/1000 ft2 veneer), averaging 1.83 Ibs/hr (0.05
lbs/1000 ft2 veneer). Total particulate/condensible emissions ranged from
13.4 to 23.0 Ibs/hr (0.39 to 0.64 lbs/1000 ft2 veneer) for an average of
18.3 Ibs/hr (0.53 lbs/1000 ft* veneer). Particulate matter accounted for
approximately 10 percent of the total sample weight while the remaining 90
percent of the catch was condensible organics.
Measured particulate grain loadings averaged 0.016 gr/DSCF for tests 1, 2
and 3; ranging from 0.007 to 0.025 gr/DSCF. Total particulate/condensible
grain loadings ranged from 0.124 to 0.198 gr/DSCF, for a three-test average of
0.164 gr/DSCF. The bulk of the total emission concentration was accounted for
by condensible organics (90 percent).
The average stack gas temperature was 309 F with an average moisture
content of 32.6 percent. Moisture content varied from 31.8 percent to 34.0
percent over the three tests. The average stack gas flow rate was 12,300
DSCFM and did not vary significantly among the three tests.
Isokinesis averaged 109 percent for the three tests performed. Isokinesis
for test 1 was 116.5 percent due to a nomograph calculation error, while test
2 was high at 111.4 percent due to a higher than expected gas stream moisture
content. Isokinesis was acceptable for test 3 at 99.2 percent. Leak checks
were performed at the conclusion of each test and leak rates were acceptable
at less than 0.02 cfm.
2-6
-------
TABLE 2-2
SUMMARY OF METHOD SX PARTICIPATE AND CONDENSIBLE ORGANIC MEASUREMENTS
FOR GASES ENTERING THE GEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood Plant, Springfield, Oregon
to
-4
Run Number
Date
Volume Gas Sampled (DSCF)a
Stack Gas Flow Rate (DSCFM)
Stack Temperature (°F)
Percent Moisture by Volume
Percent Isokinesis
Production Rate (1000 ft'/hr)b
Particulate-Condensible Results
Front Half Catch (Probe and Filter)
milligrams
gr/DCSF
1 be/hour
1000 ft*
Total Catch
milligrams
gr/DSCF
Iba/hour
lbs/1000 ft1
Percent Condensible Emissions
Run 1
6-9-B1
45.3
12400
303
31.9
116.5
35.7
74.2
0.0253
2.93C
0.082
582. 3
0.198
23.0°
0.644
87.3
Run 2
6-10-81
42.1
12000
314
34.0
111.4
34.4
44.2
0.0162
1.67°
0.049
459.6
0.169
18.5°
0.538
90.4
Run 3
6-11-81
42.4
12600
309
31.8
99.2
34.1
19.7
0.0072
0.77
0.023
340.6
0.124
13.4
0.393
94.2
Averaqe
43.3
12300
309
32.6
109.0
34.7
46.0
0.0162
1.71
0.049
460.8
0.164
18.3
0.525
90.0
a Standard Conditionsi 29.92 inches llg at 68°F.
b 1000 square feet veneer per hour on 3/8 inch baslsi Includes trim factor) does
not account for redry material.
c Results are average of concentration and area ratio methods due to high
isokinetic values. (See Table 2-4.)
-------
The mass emission rates for tests 1 and 2 were recalculated using the area
ratio method because of the unacceptable anisokinetic conditions. The results
are presented in Table 2-4 and are only slightly higher than those obtained
from the concentration method, which is the normal approach. This result is
probably due to the small percentage of particulate matter in the gas stream
which would escape collection by the sampling nozzle under superisokinetic
sampling conditions. An explanation of the area ratio method for calculating
mass emission rates is presented in Section 5.3.1.4 of this report. Mass
emission rates presented in Tables 2-1 and 2-2 represent the average of the
two calculation methods for tests 1 and 2.
2.2.2 Scrubber Outlet
A summary of Method 5X data collected at the scrubber outlet is presented
in Table 2-3. Data presented include sample volume; stack gas flow rate, tem-
perature, and moisture content; isokinesis for each test; veneer production
rate; front half (particulate) and total (particulate/condensible) emissions.
Three emission tests were performed at the scrubber outlet. Testing was
performed concurrently with tests at the scrubber inlet on June 9, 10 and 11.
Measured particulate emissions for tests 1, 2 and 3 ranged from 2.59 (0.08
lbs/100 ft2) to 3.70 Ibs/hr (0.11 lbs/1000 ftM, averaging 3.20 Ibs/hr
(0.09 lbs/1000 ft2 veneer). Total measured particulate/condensible
emissions ranged from 12.6 Ibs/hr (0.37 lbs/1000 ft2) for test 3 to 16.3
Ibs/hr (0.46 lbs/1000 ft2) for test 1. The average total emission rate was
14.9 Ibs/hr (0.43 lbs/1000 ft2 veneer). Particulate material collected
during these three tests accounted for approximately 21 percent of the total
emissions on the average, while the remaining 79 percent was condensible
organics.
2-8
-------
to
I
TABLE 2-3
SUMMARY OF METHOD 5X PARTICULATE AND CONDENSIBLE ORGANIC MEASUREMENTS
FOR GASES EXITING THE GEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood
Run Number
Date
Volume Gas Sampled (DSCP)a
Stack Gas Flow Rate (DSCFH)b
Stack Temperature (°F)
Percent Moisture by Volume0
Percent Isoklnesia
Production Rate (1000 ft»/ht)d
Particulate-Condensible Results
Front Half Catch (Probe and Filter)
milligrams
gr/DCSF
Ibs/hour
lbs/1000 ft1
Total Catch
milligrams
gr/DSCF
Ibs/hour
lbs/1000 ft1
Percent Condenslble Emissions
Plant, Springfield, Oregon
Run 1 Run 2 Run 3
6-9-81 6-10-81 6-11-81
38.3
15200
166
38.3
(37.4)e
120.4
35.7
56.9
0.023
3.30f
0.092
286.9
0.115
16.3*
0.457
80.2
37.3
17500
166
37.5
(37.4)a
102.1
34.4
41.9
0.017
2.59
0.075
257.6
0.107
15.9
0.462
83.7
37.4
16100
168
39.6
(39.3)e
111.3
34.1
61.6
0.025
3.70f
0.109
208.8
O.OB6
12. 6f
0.370
70.5
Average
37.7
16300
167
38.5
(38.0)e
111.3
34.7
53.5
0.022
3.20
.092
251.1
0.103
14.9
0.430
78.7
a Standard Conditionsi 29.92 inches Hg at 68°F.
b Outlet volumetric flows (DSCFM) suspected to be slightly biased high due to
possible wind interference.
c Actual measured moisture adjusted to saturated conditions using psychometric
equation. (See Section 5.)
d 1000 ft' per veneer hour on 3/8 in. basis; includes trim factor; does not
account tor redry material.
0 Theoretical moisture content at saturation ( ), all results based on this.
f Results are average of concentration and area ratio methods due to high
isokinetlc values. (See Table 2-4.)
-------
TABLE 2-4
COMPARISON OF PARTICULATE AND CONDENSIBLE MEASURED EMISSIONS
WITH CALCULATED EMISSIONS
(Concentration Method vs. Area Ratio Method)
Georgia-Pacific Plywood Plant, Springfield, Oregon
Emission Rate (Ibs/hr)
Sample No.
Sample Fraction
Concentration
Method
Area-Ratio
Method
Average
5X-l-Inlet
(%I* = 116.5)
5X-2-Inlet
(%I* = 111.4)
5X-l-Outlet
(%I* = 120.4)
5X-3-Outlet
(%I* = 111.3)
Front half
Back half
Total
Front half
Back half
Total
Front half
Back half
Total
Front half
Back half
Total
2.69
18.4
21.1
1.67
15.7
17.4
2.98
12.0
15.0
3.51
8.39
11.9
3.16
21.6
24.8
1.88
17.7
19.6
3.59
14.5
18.1
3.89
9.30
13.2
2.93
20.0
23.0
1.78
16.7
18.5
3.30
13.3
16.3
3.70
8.85
12.6
* isokinesis
2-10
-------
Particulate grain loadings measured at the scrubber outlet averaged 0.022
gr/DSCF for these tests, ranging from 0.017 gr/DSCP to 0.025 gr/DSCF. Total
grain loadings (particulate/condensible) ranged from 0.086 to 0.115 gr/DSCF,
averaging 0.103 gr/DSCF for the three tests.
The average stack gas temperature measured during the three tests was
167 F. The measured moisture content of the gas stream averaged 38.5 per-
cent for the three tests with slight variation.
The moisture content of the gas stream as measured during each test
exceeded saturation at the measured temperature. This phenomenon was not
surprising since entrained water droplets were observed in the gas stream by
TKC and EPA personnel. In accordance with EPA Method 4, the gas stream
moisture content was recalculated assuming saturation of the gas stream at the
average stack gas temperature for each test. An explanation of this procedure
is presented in Section 5.3.1.4. Recalculated gas stream moisture contents
ranged from 37.4 percent for tests 1 and 2 to 39.3 percent for test 3, averag-
ing 38.0 percent. These saturation moisture values were carried through the
remainder of the emission calculations.
Measured stack gas flow rates ranged from 15,200 DSCFM to 17,500 DSCFM,
averaging 16,300 DSCFM. The average outlet stack gas flow rate was measured
to be approximately 25 percent greater than that measured entering the
scrubber system. One reason for the difference in the flow rate measurements
may be leaks within the system (around spray nozzles, in the cyclone ductwork,
and before the fan) which would draw ambient air into the system. Another
reason may be the inherent inaccuracy of EPA Method 2 at stack gas velocities
of approximately 500 feet per minute (fpm) . A third possible reason is wind
interference. Since sampling was performed only 9 feet from the top of a 9
foot i.d. stack, even a small gust of wind created turbulence within the
2-11
-------
duct. Momentary negative flow rates were indeed observed during the test pro-
gram. These reasons may account for the differences between the measured
inlet and outlet flow rates as well as the wide variation in measured flow
rates from test to test.
Isokinesis was acceptable only for test 2 at 102 percent. Isokinesis was
unacceptable for tests 1 and 3 at 120.4 percent and 111.3 percent, respective-
ly. The average isokinesis for the three tests performed was 111.3 percent.
Varying stack gas moisture content was a major factor affecting the unaccept-
able isokinetic conditions. The preliminary determination indicated 15.8 per-
cent moisture/ but measured moistures were more than twice this during the
subsequent tests. It was later discovered that different grades of veneer
with varying moisture contents were dried throughout the test program, as
shown in Table 3-1.
The mass emission rates for tests 1 and 3 were recalculated using the area
ratio method because of unacceptable superisokinetic conditions. The results
are presented in Table 2-4 and are only slightly higher than those obtained
from the concentration method, which is the normal approach. This fact is
probably due to the small percentage of particulate matter in the gas stream
which would escape collection by the sampling nozzle under anisokinetic sampl-
ing conditions. An explanation of the area ratio method for calculating mass
emission rates is presented in Section 5.3.1.4 of this report. Mass emission
rates presented in Tables 2-1 and 2-3 are the average of the two calculation
methods for tests 1 and 3. Leak checks were performed at the conclusion of
each test and leak rates were acceptable at less than 0.02 cfm. Some diffi-
culty was encountered maintaining probe and filter outlet temperatures at 350
o
+25 F during these tests. Further discussion is presented in Section
5.3.1.1.
2-12
-------
2.3 Method 25 - Total Organic Tests
A summary of the Method 25 total organic data (condensible and nonconden-
sible) collected at the scrubber inlet and outlet is presented in Tables 2-5a
(English units), 2-5b (metric units). These tables include TRC, PCS, and
NCASI average emission data: stack gas flow rate, moisture content and tem-
perature; veneer drying production rate, and a summary of the total organic
emissions by concentration, mass emission rate, and emission rate per unit
production. All emissions are expressed as carbon (C,). NCASI calculates
the emission rate as Ibs/hr equivalent methane (CHJ instead of carbon
(C ). Their data in the tables have been converted to Ibs/hr C to
present the data on a consistent basis, conforming with Method 25.
Emission data are presented for the three test series. Testing was per-
formed simultaneously at the scrubber inlet and outlet on June 9, 10 and 11.
Total organic emissions entering the scrubber system ranged from 23.8 Ibs/hr
(10.6 kg/hr) or 0.67 lbs/1000 ft2 veneer (3.20 kg/1000 m2) to 35.8 Ibs/hr
(16.2 kg/hr) or 1.05 lbs/1000 ft2 (5.10 kg/1000 ra2), averaging 30.2 Ibs/hr
(13.6 kg/hr) or 0.87 lbs/1000 ft2 veneer (4.22 kg/1000 m2). Emissions
exiting the scrubber system ranged from 30.9 Ibs/hr (13.8 kg/hr) to 43.6
Ibs/hr (19.6 kg/hr) or 1.22 lbs/1000 ftz (5.87 kg/1000 m2) for an average
emission rate of 38.8 Ibs/hr (17.5 kg/hr) or 1.12 lbs/1000 ft2 veneer (5.39
kg/1000 m2). The collection efficiency of the system was measured to be
less than zero for tests 1 and 2 and 13.7 percent for test 3. The average
collection efficiency of the scrubber system was less than zero.
Detailed summaries of these test data are presented in Sections 2.3.1 and
2.3.2, and in Appendix A. Sample equations and calculations are presented in
Appendix B. Field data sheets appear in Appendix C. Sampling logs and sum-
maries are shown in Appendix D. Laboratory analysis data are presented in
Appendix G.
2-13
-------
TABLE 2-5a (English Units)
SUMMARY OF METHOD 25 TOTAL ORGANIC COLLECTION
EFFICIENCY FOR GEORGIA-PACIFIC SCRUBBER SYSTEM
Geocgia-Pacific Plywood Plant, Springfield, Oregon
Run Number
Date
Emission Description!
Stack Gas Flow Rate (DSCFM)3' b
Stack Temperature (°F)
Percent Moisture by Volume0
Scrubber Pressure Drop (inches H2O)
Production Rate (1000 ft'/hr)a
to
1
£ Total Organic Results8
parts/million, Cj
gr/DSCF, Cj
Ibs/hour, Cj
lbs/1000 ft1 , Cj
System Collection Efficiency (%)
Run 1 Run 2 Run 3
6-9-81 6-10-81 6-11-81
Uncontrolled Controlled Uncontrolled Controlled Uncontrolled Controlled
12,400 15,200 12,000 17,500 12,600 16,100
303 166 314 166 309 168
31.9 37.4 34.0 37.4 31.8 39.3
14.2 15.0 13.9
35.7 34.4 34.1
1,027 1,536 1,389 1,286 1,524 1,030
0.22 0.33 0.30 0.28 0.33 0.22
23.6 43.6 31.1 42.0 35.8 30.9
0.67 1.22 0.90 1.22 1.05 0.91
<0 <0 13.7
Average
Uncontrolled
12,300
304
32.6
14.
34.
1,313
0.21)
30.2
0.87
<0
Controlled
16,300
167
38.0
4
7
1,284
0.28
38.8
1.12
a Standard conditions! 29.92 inches Hg at 68°F.
b Outlet volumetric flows (DSCFM) suspected to be slightly biased high due to possible wind interference.
c Actual measured moisture adjusted to saturated conditions using psychometric equation (controlled emissions only).
d 1000 square feet veneer per hour on 3/8 inch basis) includes trim factor) does not account for redry material.
e Emission results calculated and reported as Cj. Does not include front half results from Method 5X collector, and cannot be compared
with Method 5X mass determinations.
-------
TABLE 2-5b (Metric Units)
SUMMARY OF METHOD 25 TOTAL ORGANIC COLLECTION
EFFICIENCY FOR GEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood Plant, Springfield, Oregon
Run Number
Date
Run 1
6-9-81
Run 2
6-10-81
Run 3
6-11-81
Average
Emission Description:
Stack Gas Flow Rate (NM'/Hln)3' b
Stack Temperature (°C)
Percent Moisture by Volume0
Scrubber Pressure Drop (nun IIjO)
Production Rate (1000 meters'/hr)d
Total Organic Results8
parts/million, C^
g/USCP, Cj
kg/hour, Cj
kg/1000 m> , Cj
System Collection Efficiency (%)
Uncontrolled Controlled Uncontrolled Controlled Uncontrolled Controlled Uncontrolled Controlled
351.2
150.6
31.9
430.2
74.4
37.4«
360.7
3.32
1,027 1,536
0.50 0.76
10.6 19.5
3.20 5.87
<0
339.8
156.7
34.0
495.6
74.4
37.4«
381.0
3.20
1,389 1,286
0.69 0.64
14.0 19.1
4.37 5.95
<0
356.8
153.9
31.8
456.0
75.6
39.3£
353.1
3.17
1,524 1,030
0.76 0.50
16.2 14.0
5.10 4.42
13.6
348.3 461.6
154 74.8
32.6 JB.UC
365.8
3.22
1,313
0.62
13.6
4.22
1,284
0.63
17.5
5.41
<0
a Standard conditions: 760 mm Hg at 20%.
b Outlet volumetric flows (NM'/min) suspected to be slightly biased high due to possible wind interference.
0 Actual measured moisture adjusted to saturated conditions using psychometric equation (controlled emissions only).
d 1000 square meters veneer per hour on 9.5-mm basis) includes trim factor) does not account for redry material.
8 Emission results calculated and reported as Cj. Does not include front half results from Method 5X collector, and cannot be compared
with Method 5X mass determinations.
f Theoretical moisture content at saturation.
-------
2.3.1 Scrubber Inlet
A summary of Method 25 condensible and noncondensible organics data
collected at the scrubber inlet is presented in Tables 2-6 and 2-8. Table 2-6
shows relevant emission data and presents total organic emissions calculated
by TRC, PCS, and NCASI as concentration, mass emission rate, and emission rate
per unit production. Table 2-8 presents a breakdown of the total organic
emissions into condensible and noncondensible organics as analyzed by the
three laboratories. In addition, individual sample train analyses results are
shown. The relative standard deviation between the paired sample trains is
also presented.
Emissions of carbon (C.) from the scrubber as analyzed by TRC, PCS and
NCASI showed good overall correlation. The precision of the test data between
paired samples (relative standard deviation-RSD) was excellent overall,
averaging 19.8 percent RSD for the three laboratories involved.
2.3.2 Scrubber Outlet
A summary of Method 25 condensible and noncondensible organics data
collected at the scrubber outlet is presented in Tables 2-7 and 2-9. Table
2-7 shows relevant emission data and presents total organic emissions calcula-
ted by TRC, PCS and NCASI as concentration, mass emission rate, and emission
rate per unit production. Table 2-9 presents a breakdown of the total organic
emissions into condensible and noncondensible organics as analyzed by the
three laboratories. In addition, individual sample train analyses results are
shown. The relative standard deviation between paired sample trains is also
presented.
Emissions of carbon (C-) from the scrubber as measured by TRC and NCASI
showed good correlation. The average emissions calculated by TRC were
slightly greater than those calculated by NCASI and PCS. There is no apparent
2-16
-------
TABLE 2-6
SUMMARY METHOD 25 INDIVIDUAL TOTAL ORGANIC MEASUREMENTS
FOR GASES ENTERING THE GEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood Plant Springfield, Oregon
to
I
Run Number
Date
Stack Gas Flow Rate (DSCFM)a
Stack Temperature (°F)
Percent Moisture by Volume
Production Rate (1000 ft»/hr)b
Laboratory Performing Analysis
Total Organic Results*-
parts/million, Cj
g/DSCF, Cj
Ibs/hour, Cj
lbs/1000 ft»r, Cj
Run 1
6/9/81
12,400
303
31.9
35.7
TRC PCS NCASI
1016 655 1210
0.22 0.19 0.26
23.7 19. B 28.1
0.66 0.55 0.78
Run 2
6/10/81
12,000
314
34.0
34.4
TRC PCS NCASI
1482 1392 1295
0.32 0.30 0.28
33.3 31.2 29.1
0.97 0.91 0.84
Run 3
6/11/81
12,600
309
31.8
34.1
TRC PCS NCASI
2137 1101 1334
0.47 0.24 0.29
50.4 25.9 31.4
1.48 0.76 0.92
Average
12,300
J09
32.6
34.7
TRC PCS
1545 1116
0.34 0.24
35.8 25.6
1.03 0.74
NCASI
1280
0.28
29.5
0.85
a Standard Conditions: 29.92 inches llg at 68°F, NCASI uses O°C.
b 1000 square feet veneer per hour on 3/B inch basis; includes trim factor) does not account for redry material.
c Emission reaults calculated and reported as Cj . Does not include front half results from Method 5X collection, cannot be
compared with Method 5X mass determination.
-------
N)
I
TABLE 2-7
SUMMARY OF METHOD 25 INDIVIDUAL TOTAL ORGANIC MEASUREMENTS
FOR GASES EXITING THE GEORGIA-PACIFIC SCRUBBER SYSTEM
Run Number
Date
Stack Gas Flow Rate (DSCFM)a> b
Stack Temperature °F
Percent Moisture by Volume0
Production Rate (1000 ft«/hr)d
Laboratory Performing Analysis
Total Organic Results0
parts/million, Cj
gr/DSCF, G!
Ibs/hour, Cj
lbs/1000 ft« , Cj
Georgia-Pacific
Run 1
10/9/81
15,200
166
37.4
35.7
TRC PCS NCASI
1762 1538 1306
0.38 0.33 0.28
50.1 43.7 37.2
1.40 1.22 1.04
Plywood Plant, Springfield, Oregon
Run 2 Run 3
10/10/81 10/11/81
17,500 16,100
166 168
37.4 39.3
34.4 34.1
TRC PCS NCASI TRC PCS NCASI
1694 941.6 1221 1264 903 923
0.37 0.20 0.27 0.28 0.20 0.20
55.5 30.8 40.0 38.1 27.2 27.8
1.61 0.90 1.16 1.12 0.80 0.82
Average
16,300
167
38.0
34.7
TRC PCS NCASI
1573 1127
0.34 0.25
48.0 34.2
1.38 0.99
1150
0.25
35.1
1.01
a Standard Conditional 29.92 Inches Hg at 6B°F, NCASI uses O°C.
b Outlet flows suspected to be slightly biased high due to wind Interference.
c Theoretical moisture content at saturation. (See Section 5.)
d 1000 square feet veneer per hour on 3/8 inch basis; Includes trim factort does not account for redry material
8 Emission results calculated and reported as Cj. Does not include front half results from Method 5X collector, and cannot be compared with
Method 5X mass determinations.
-------
TABLE 2-8
SUMMARY OF METHOD 25 INDIVIDUAL TOTAL ORGANIC TRAP, TANK MEASUREMENTS
FOR GASES ENTERING THE GEORGIA-PACIFIC SCRUBBER SYSTEM
Georgia-Pacific Plywood Plant, Springfield, Oregon
Condensible Catch
Run 1
June 9
Run 1
June 10
Run 3
June 11
Overall Average
Stack Gas
Flow Rate Sample
DSCFM3 I.D. No.
12,400 1-A
1-B
1-C
1-D
9-1
9-2
Average
12,000 2-A
2-B
2-C
2-D
10-1
10-2
Average
12,600 3-A
3-B
3-C
3-D
11-1
11-2
Average
12,300
Lab
TRC
TRC
PCS
PCS
NCASI
NCASI
TRC
TRC
PCS
PCS
NCASI
NCASI
TRC
TRC
PCS
PCS
NCASI
NCASI
HjO
Ice Trap
(ppm)
477
245
509.7
720.1
N.A.b
N.A.b
488
483
489
539
549.7
N.A.b
N.A.b
515
545
940
469.0
576.7
N.A.b
N.A.b
639
547
C02
Ice Trap
(ppm)
300
601
267.7
164.1
1231.4
1045.6
601.6
163
212
352.2
367.9
1531.1
863.2
5B2
469
433
550. 8
315.9
1143.1
1155.2
678
621
NonCondensible Catch
(ppm)
409
n.d.
25.0
22.5
60.9
82.3
100
804
814
433
541.8
63.3
132.0
465
1102
786
159.2
102.6
161.1
209.4
420
328
Total
Catch
(ppm)
1186
846
802.4
906.7
1292
1128
1027
1450
1515
1324
1459
1594
995
1389
2116
2159
1206
995.2
1304
1365
1524
1314
Pair
Average
(ppra)
1016
855
1210
1027
1482
1392
1295
1389
2137
1101
1334
1524
1314
Emission
Rate
(Ibs/hr)
23.7
19.8
28.1
23.8
33.3
31.2
29.1
31.1
50.4
25.9
31.4
35.8
30.2
Relation
Standard
Deviation
4.22
11.6
4.0
6.6
32.2
14.6
3.06
16.6
70.3
7.4
31.2
36.3
19.8
a Standard Conditionsi 29.92 inches Hg at 60°F; NCASI uses 0°C.
b NCASI does not use an 1)2° ice trap in their train.
-------
TABLE 2-9
SUMMARY OF METHOD 25 INDIVIDUAL TOTAL ORGANIC TRAP, TANK MEASUREMENTS
FOR GASES EXITING THE GEORGIA-PACIFIC SCRUBBER SYSTEM
NJ
I
N)
o
Georgia-Pacific Plywood Plant, Springfield, Oregon
Condensible Catch
Stack Gas
Flow Rate Sample
DSCFM3 I.D. No.
Run 1
June 9 15,200
Run 2 17,500
June 10
Run 3 16,000
June 11
/
Overall Average 16,300
1-A
1-B
1-C
1-D
9-1
9-2
Average
2-A
2-B
2-C
2-D
10-1
10-2
Average
3-C
3-D
3-A
3-B
11-1
11-2
Average
11 20
Ice Trap
Lab (ppm)
TRC
TRC
PCS
PCS
NCASI
NCASI
TRC
TRC
PCS
PCS
NCASI
NCASI
TRC
TRC
PCS
PCS
NCASI
NCASI
815
758
1180.2
905.0
N/Ab
N/Ab
915
465
792
579.9
571.0
N/Ab
N/Ab
606
831
441
534.6
414.3
N/Ab
N/Ab
555
692
C02 Total
Ice Trap Non-Condensible Catch Catch
(ppm) (ppm) (ppm)
278
740
405.6
116.6
1242.9
1276.3
676.6
332
303
117.5
102.4
1270.2
1039.9
527
196
193
176.8
641.1
949.4
743.4
483
562
546
388
323.3
145.5
49.2
44.9
249
687
810
316.8
177.5
23.6
108.5
354
443
424
0
39.0
68.6
85.2
177
260
1639
1886
1909
1167
1292
1321
1536
1484
1905
1032
. 850.9
1294
1148
1286
1470
1058
711.4
1094
1018
828.6
1030
1284
Pair
Average
(ppm)
1762
1538
1307
1536
1694
941.6
1221
1286
1264
903
923
1030
1284
Emission
Rate
(Ibs/hr)
50.1
43.7
37.2
43.6
55.5
30.8
40.0
42.0
38.1
27.2
27.8
30.9
38.8
Relation
Standard
Deviation
10
2.9
63
25.3
5.7
7.3
11. a
8.3
4.3
3.3
b.9
4.8
12.8
a Standard condltionsi 29.92 inches llg at 68°F| NCASI uses 0°C.
b NCASI does not use an II20 ice trap in their train.
-------
explanation for this difference. The precision of the test data between
paired samples (relative standard deviation) was excellent overall, averaging
12.8 percent RSD for the three laboratories involved.
2.4 Visible Emissions
Visible emissions observations were not conducted during this sampling
program as planned. Overcast skies prevented an accurate determination of the
scrubber outlet plume opacity. Further details of the decision to abort this
phase of the test program are presented in Sections 1.4 and 5.7.
2.5 Scrubber Operational Summary
A summary of operational parameters of the G-P scrubber system during the
test program is presented in Table 2-10. Pressure drop measurements (AP)
across the scrubber system are presented as well as scrubber solution analysis
data.
Scrubber solution samples were taken from the recirculating tank every 30
minutes and then composited into one sample per test. Sample analyses for
total organic carbon (TOC) ranged from 3,010 mg/1 for test 1 to 2,860 mg/1 for
test 3, averaging 2,956 mg/1 for the three tests.
Pressure drop (AP) measurements across the scrubber were made at 30
minute intervals during the test program. The AP gradually increased during
tests 1 and 2, and averaged 14.2 and 15.0 inches water, respectively. During
test 3, however, the AP peaked shortly after the start of testing and then
gradually declined for the rest of the test, averaging 13.9 inches water. The
three test average AP was 14.4 inches water.
2-21
-------
TABLE 2-10
GEORGIA-PACIFIC SCRUBBER SYSTEM OPERATIONAL DATA SUMMARY
Georgia-Pacific Plywood Plant, Springfield, Oregon
June 1981
Pressure Drop
Measurements
Run Number Date Time
1 June 9 1400
1430
1500
1530
1600
1630
1700
1730
Average
2 June 10 1300
1400
1430
1500
1530
1600
1630
Average
3 June 11 1145
1215
1245
1315
1345
1415
1445
Average
AP
(in. H90)
12.5
13.0
14.0
14.8
14.5
14.7
15.1
15.2
14.2
14.7
14.8
15.0
14.8
15.2
15.4
15.3
15.0
13.8 -
14.5
14.0
13.9
13.8
14.2
12.9
13.9
Scrubber Solution
Volume Collected TOC
(ml) (mg/1)
800 3010
. 700 3000
700 2860
2-22
-------
2.6 Summary of Fugitive Emissions (Provided by RTI)
The temperature and pressure changes that a veneer dryer is subjected to
make it very difficult for a dryer to be completely air tight. Door seals and
skins, green and dry end baffles, and abort stacks will with time all even-
tually develop leaks. Door seals and dryer skins most readily develop them.
At Springfield all the dryers had fugitive emissions. The three older dryers
leaked more from around the elephant ears than from individual door seals.
The jet dryer also experienced door leaks with quality checks showing varying
amounts from one day to another. It was impossible to estimate volume of fug-
itive gases from any fugitive source.
At the abort stacks there were also fugitive emissions. Very little was
seen from the jet dryer abort stacks, while opacities (unofficially) up to 30
percent were seen from the other three dryer abort stacks. Cooling section
air volumes are large. No bluish haze was seen coming from cooling section
exhausts from the three longitudinal dryers. However, the jet dryer cooling
stacks showed some bluish opacity.
All of the fugitive emissions evaluations were purely qualitative and
visual. Tables 2-lla, 2-llb and 2-llc contain fugitive emission data.
2.7 Ambient Air Measurements
A summary of ambient temperature and relative humidity measurements by RTI
and DGA is presented along with process information in Table 3-1. Ambient
temperatures ranged from 54 to 75 F, while relative humidity ranged from
36 percent to 76 percent during the test program.
2-23
-------
TABLE 2-1la
FUGITIVE EMISSION SUMMARY
Georgia-Pacific Plywood Plant, Springfield, Oregon
(Provided by RTI)
June 9, 1981
to
I
CO
Abort
Dryer I/Time Green End
1 2ilO
2:45
3>40
4:50
Si 30
2 It 50
2i45
3:45
4:50
5:30
3 2:00
2:45
3:45
4:45
5:30
4 1:45
2:45
' 3:40
4:40
5:27
Building
Pans
H
11
H
H
H
II
II
H
N
N
N
N
N
1
L
25%
30%
M
H
M
II
II
L
10%
10%
30%
30%
10%
20%
N
N
N
N
N
2
Stacks
Dry End
L
M 30%
II 20%
II 30%
H 30%
L
L
M
M 10%
M 10%
II 20%
M 30%
M 30%
L
H 30%
N
N
N
N
N
3
Cooling
N
N
H
N
N
N
N
N
H
N
N
N
H
N
N
5%
10%
10%
10%
15%
4
Overall1
L
H
M
H
II
M
M
M
M
M
H
II
H
M
11
5
100
200
200
250
300
150
200
200
150
150
250
250
250
ISO
300
L -
L -
L -
VL
L
6
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
(Opacity)
2:00
5%
5%
-
5%
-
not
Time
2:10
3:25
4:15
5:05
5:40
2:05
3:25
4:15
5:05
5:40
2:05
3:15
4:13
5:00
5:40
2:00
3:18
4:12
5:00
5:40
N »
T •
VL •
L •
observed M -
2:45
3:45
4:45
5:30
5%
5%
5%
5t
10%
10%
5%
5%
-
-
-
-
5%
5%
5%
10%
-
-
-
5%
«
•
••
M
II »
VII "
Door
Seals
2M
2L, 2T
4M
1L
3L
1L
H
N
N
N
N
N
N
N
N
1M, 1L
1L
1H, 1L
111, 1L
2M
None
Trace
Inside
Elephant Above
Ears Dryers
1M
1M, 1L
1M, 211
4M, III
2M
1L
N
1M
1L
1L
N
N
N
N
N
NA
NA
NA
NA
NA
KEY
M
M
H
II
M
L
L
M
M
H
H
H
11
H
H
L
L
L
L
M
1 cfm estimates
are gross
Very Light estimates
Light
Moderate % opacity
Heavy
on blue
Very Heavy plumes
-------
TABLE 2-llb
FUGITIVE EMISSION SUMMARY
Georgia-Pacific Plywood Plant, Springfield, Oregon
(Provided by RTI)
June 10, 1981)
NJ
Ul
Inside Appearance
Dryer
1
2
3
'i
4
I/Time
12:20
1:15
2:20
2:45
3:45
4:15
4:45
12:17
1:15
2:15
2:50
3:45
4:15
4:45
12:15
1:15
2:15
2:50
3:50
4:15
4:40
12:13
1:12
2:15
2:48
3:50
4:15
4:40
Roof
Vents
12:20
1:15
2:15
2:45
3:45
4:15
4:40
Abort
Green End
S 10%
11
H
M 10
M 15
L 10
L 10
11
II
H
L 10%
L
L 5t
L 10
M 20t
L
M 10%
11 20%
M 10
H 20
II 20
N
N
H
N
N
N
N
907 908
5 5
5
5 5
5 5
5 5
5 5
5
Stacks
Dry End
M 20
M
II 20
H 20
11 30
H 30
M 20
M
S
M 101
M 10%
M 10
M 10
M 10
M 20%
L
M 10%
M 20
11 20
II 20
H 20
N
N
N
N
N
N
N
909
N
-
-
-
-
-
-
Door Elephant
Cooling
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
10-15
5
10%
10-15
N
5-10
5
910
5
-
5
5
5
5
-
Overall1
M 150 cfm
M 150 cfm
H 250 cfm
M 150 cfm
M 200 Cfm
11 250 cfm
M 200 cfm
M 300 cfm
M 150 cfm
M 200 cfm
L 150 cfm
L 150 cfm
L 100 cfm
L 150 cfm
M 200 cfm
50 cfm
M 100 cfm
II 250 cfm
H 250 cfm
11 250 cfm
II 250 cfm
Had at Dry
End
H
H
II
N
It
H
911 912
5 5
5
-
5
-
-
5
Time
12:10
1:25
2:07
2:40
3:40
4:05
4:38
12:08
1:25
2:06
2:40
3:40
4:05
4:35
12:05
1:20
2:05
2:40
3:35
4:05
4:33
12:03
1:20
2:03
2:38
3:35
4:00
4:33
N -
T «
VL -
L •
M *
II "
VII "
Seals
1 VL
2 VL
2 VL
1 S, 1 L
2 L
2 L
2 L
2 M
2 VL
N
N
N
N
N
N
N
N
N
N
N
N
1 L, 1 M
2 M
1 L
1 L
1 L
1 M
1 M, TR
KEY
None
Trace
Very Light
Light
Moderate
Heavy
Very Heavy
Ears Top
N L
N L
N L
1 L M
1 L M
1 L L
1 L M
IS L
IS M
1 M M
1 L II
1 L M
1 L L
1 L L
N M
N M
N II
N II
N H
N M
N H
NA M
NA M
NA L
NA L
NA L
NA M
NA L
cfm estimates
are gross
estimates
t opacity
on blue
plumes
Comments
M S mid fans
Mod. around
mid fans
11 at dry end
None at Green
Large leakage
around
exhaust dry
end stk.
Cooling stks.
from this
dryer and
roof vents
are the major
roof-ton
emitters.
-------
TABLE 2-llc
FUGITIVE EMISSION SUMMARY
Georgia-Pacific Plywood Plant, Springfield, Oregon
(Provided by RTI)
June 11, 1981)
to
I
cn
Abort Stacks
Dryer/Time
1
2
3
4
10:50
Ilt46
ill 20
12:55
1(06
2:28
11:10
11:45
12:20
12:55
1:46
2:28
11:10
11:45
12120
12:55
1:48
2:27
11:10
11:45
12:17
12t55
1:48
2:26
Roof
Vents
11:10
11:45
12:19
12:55
1:48
2:27
Green End
II
II
M 5
M 5
M 5
M 10
II
H
II
II
II
M 10
H 30
II
H
II 10
II 10
M 20
N
N
N
N
N
N
907 908
5
5
5
10 5
5 10
10 5
Dry End
H
II 20
H 10
H 10
H 30
H 20
M
H 10
M 10
L
M 10
M 10
1! 20
H 20
H 20
H 20
II 20
II 20
N
N
N
N
N
N
909
-
-
5
-
-
-
Cool ing
N
N
N
N
N
N
N
N
H
N
N
N
N
N
N
N
N
N
lot
5-10
0-5
5-10
5
5-10
910
5
-
-
5
5
5
Ov«rall'
300 H
300 H
250 H
250 II
300 II
250 H
350 H
300 H
250 11
150 M
300 H
200 M
350 H
350 11
350 11
350 11
300 H
250 H
911 912
5
5
-
-
-
5
Time
10:42
11:32
12:08
12:45
1:26
2:17
2:47
10:39
11:30
12:07
12:45
1:26
2:16
2:47
10>38
11:30
12:06
12:43
1:27
2:15
2:48
10:37
11:28
12:06
12:43
1:27
2:14
2:47
N
T
VL
L
H
H
VH
Inside
Appearance
Door Elephant
Seals
3 L
3 L
2 L
2 VL
2 VL
2 L
1 L
N
N
N
N
N
N
N
N
N
N
N
N
N
N
2 VL, 3 T
2 VL
N
N
N
N
N
KEX
• None
» Trace
« Very Light
» Light
» Moderate
- Heavy
» Very Heavy
Ears
1 L
1 L
1 M
1 M
1 L
1 L
1 L
1 L
1 L
1 L
1 L
1 L
1 L
1 L
N
N
N
N
N
N
N
NA
NA
NA
NA
NA
NA
NA
Cfm
Top
L
M
M
L
M
11
M
L
L
M
L
L
M
M
H
H
11
H
M
H
L
L
VL
VL
VL
VL
VL
estimates
Comments
M e top
center Cans
M 9 center
fans at top
Stack leaks
k middle fans
hlgn
emissions
are gross
estimates
% »
opacity
on blue
plumes
-------
2.8 Clean-Dp Evaluation
Results of the clean-up evaluations performed on both Method 5X sampling
trains are presented in Table 2-12. Clean-up evaluation rationale and pro-
cedures are presented in Section 1.4.6 and Section 6.1. Front half total
residue collected was 55.5 mg and 12.1 rag for the inlet and outlet sampling
trains, respectively. Back half total residue collected was 39.8 mg and 164.2
mg, respectively. Total residue collected during the clean-up evaluation was
96.3 mg and 176.3 mg for the inlet and outlet trains, respectively.
The high blank value of the inlet probe wash is probably due to the fact
that the probe was not acid washed before the evaluation. The high value
detected as impinger residue was probably due to a large amount of chromium
residue (2.50 mg/1) remaining from the pretest chromic acid wash of the glass-
ware. Further analysis for chromium was performed on the actual test impinger
solutions with only a trace amount being detected in test 1 samples. There-
fore, it is believed no test sample interference resulted from chromium con-
tamination. Chromium analytical data are presented in Appendix G.
2.9 Possible Test Interferences
A possibility exists that components and reagents used in the Method 5X
sampling train may cause interferences with the Method 25 samples drawn from
the Method 5X train. At the time of this report, a study is being performed
by TRC to quantify the possible interfering effects of acetone, silicone
vacuum grease, and silicone rubber sealant (RTV) on the Method 25 procedures.
The scrubber outlet Method 5X filtration temperature could not be main-
o
tamed in the planned 350 H^25 F range due to insufficient heater capacity.
Temperatures ranged from 310° to 340°F. Although not a factor in the
2-27
-------
TABLE 2-12
METHOD 5X CLEAN-UP EVALUATION RESULTS, JUNE 8, 1981
Georgia-Pacific Plywood Plant, Springfield, Oregon
Train
Component
Sample Fraction
Residue Weight (g)
Inlet Outlet
Front Half
Probe Wash (DD H20)
Probe Wash (acetone)
Front Filter
Front Half Total
0.0337*
0.0448
NAt
0.0785
0.0116
0.0113
NAt
0.0229
Back Half
Impinger Water
Organic Extraction
Evaporation
Acetone Rinse
Back-up Filter
Back Half Total
Total Sample
0.0014
0.0230
0.0151
NAt
0.0395
0.1180
0.0019
0.1360**
0.0260
NAt
0.1639
0.1868
* Probe not acid washed prior to test program.
** Upon further analysis, it was discovered that Sample 5X-0-0-4 contained
2.50 mg/liter chromium, indicating residue remaining from the chromic acid
cleaning solution used in the pretest preparation of the glassware.
t Filters not inserted into trains.
2-28
-------
Method 5X sample collection, this may have resulted in a slight low bias for
the Method 25 samples. Organic materials which would have passed through the
o
filter at 350 +25 F may instead have condensed and been collected on the
lower temperature filter.
2-29
-------
3.0 PROCESS DESCRIPTION AND OPERATIONS (Provided by RTI)
This section describes the plywood manufacturing process, specifically the
veneer drying process and its emission control/ a G-P scrubber system. Pro-
duction monitoring as well as process operational conditions during the test
program are also discussed.
3.1 Process Equipment
The veneer drying operation begins after the veneer has been peeled from
the log at the lathe operation. The veneer then proceeds to the drying opera-
tion. Here, the veneer is continuously hand-fed onto the dryer feed conveyor
and into the dryer. The purpose of the operation is to thermally drive the
moisture out of the veneer in preparation for the layup and laminating opera-
tions which follow. During the drying operation, organic compounds are also
driven out of the veneer.
The G-P Springfield plant has four veneer dryers. Each is a steam heated,
multideck unit, with the number of drying zones varying between dryers.
Dryers 1, 2 and 3 are longitudinal dryers, with 22, 18 and 18 zones, respect-
ively. Dryer 4, a new unit, is a 22 zone jet dryer. Each dryer has two
exhausts from the heated zones, except dryer 4 which has three exhausts. Atop
each exhaust is an abort damper for emergency use only. These are a source of
fugitive emissions. The eight exhausts are ducted to a common manifold which
carries the exhaust to the Georgia-Pacific scrubber system.
3.2 Emission Control Equipment
The Georgia-Pacific scrubber system shown in Figure 1-1 includes a wet
spray zone, six wet cyclones, a packed tower, and a mesh pad entrainment
separator. As dryer exhaust gases pass through the 35.25-inch inside diameter
3-1
-------
duct, six nozzles inject water countercurrently into the gas stream to
saturate and cool the gas stream, thereby condensing the organics. Solids are
separated and agglomerated droplets are collected in the six wet cyclones
which follow. The remaining moisture laden gases are drawn through an induced
draft fan and forced through a packed tower and a mesh pad to rid the effluent
of aerosols. All water within the scrubber system is recirculated. The gas
stream, at approximately 165 F and 35 percent moisture by volume, then
discharges to the atmosphere through a 9-foot i.d. stack.
3.3 Production and Control Equipment Monitoring
All production monitoring data collected by RTI and DGA is presented in
Table 3-1. Scrubber operational data, collected by TRC, is presented in
Table 2-10.
3.4 Process Operating Conditions During Test Program
The operation of each dryer is set according to the size, thickness, and
kind of wood being dried. The operation of the three longitudinal dryers does
not frequently vary, but it appears that the jet dryer makes more frequent
changes. During the testing hours of the first day (June 9) there was a
change of production, despite efforts on Georgia-Pacific's part to keep the
dryer operation steady state. Dryer 4 changed from drying 1/6-inch sapwood to
1/6-inch heartwood. This was not considered to be a major change worthy of
cancelling the test run.
It is normal for small plugups in the feeding and outloading mechanisms to
occur and this did happen during the tests. On the third test day (June 11)
there was more of this type upset than usual on Dryer 4, but these were con-
sidered minor and insufficient cause for cancelling a test.
3-2
-------
TABLE 3-1
SUMMARY OF OPERATING CONDITIONS (Provided by RTI)
Georgia-Pacific Plywood Plant, Springfield, Oregon
June 8
June 9
June 10
June 11
I Production
(ft2, per hour
on 3/8-in. basis)
S apwood
Heartwood
Total
19,935
12,090
31,424
21,604
14,143
35,747
21,509
12,863
34,372
19,847
14,221
34,068
II Redry Rate (%)
9.7
9.0
9.6
11.4
III Steam Use (Ibs
per hour)
No
Evaluation
51,430
50,467
51,450
IV Temperatures
No
Evaluation
steady
325-375°F
steady
325-375°F
steady
325-375°F
V Fugitives
1. abort stacks
2. door leaks
3. above dryers
4. cooling stacks
No
Evaluation
150-300 CFM*
nos. 1&4
nos. 1 & 3 had
blue haze
150-300 CFM*
150-300 CFM*
nos. 1,2 & 4 less small leaks
than on June 9
noticeable
above all
dryers
VI Weather No cloudy,
Evaluation showers,
mid 60s
53-74%
rel.
humidity
cloudy, small
showers,
60°-75°
36-70%
rel.
humidity
morning fog,
sunny ,
54°-66°
51-76%
rel.
humidity
*Per dryer, except no. 4
3-3
-------
Steam usage, dryer temperatures, and drying times were maintained evenly
throughout the three days of tests. Process operating conditions are
summarized in Table 3-1.
Production figures provided are not the actual square footage of green
veneer dried in the steam-heated dryers but rather a figure that accounts for
trim and shrinkage. A full green veneer sheet is approximately 54 inches by
101 inches and will eventually be trimmed to 48 inches by 96 inches following
shrinkage in the dryer. The amount of shrinkage depends on the original
moisture level. As is the case with all western softwoods, Douglas fir
sapwood will shrink more than heartwood. An expected shrinkage loss is 5
percent to 7 percent. The production figures reported are, therefore, approx-
imately 85 percent of the actual throughput of the dryers. All veneer has
been converted to a 3/8-inch basis.
3-4
-------
4.0 DESCRIPTION OF THE SAMPLING LOCATIONS
This section presents a description of each sampling location and a sum-
mary of the work performed at each site. Figure 4-1 presents a schematic
layout of the veneer dryer exhaust system and identifies all sampling loca-
tions.
4.1 Scrubber Inlet
The inlet to the scrubber system was sampled employing EPA Methods 1, 2,
4, 5X and 25 in the 35.25-inch inside diameter insulated duct at sampling
ports 45 above the horizontal duct axis. These ports were located 30 feet
downstream (>8 diameters) and 24 feet upstream (>2 diameters) from the
nearest respective flow disturbances. In accordance with EPA Method 1, sam-
pling was performed at 12 traverse points. Sampling port and traverse point
locations are presented in Figure 4-2. Duct static pressure was also
measured at this location.
Method 5X tests performed at this location were 60 minutes in duration as
were the Method 25 tests performed simultaneously. A total of three Method
5X and 18 Method 25 tests was performed at this location.
4.2 Scrubber Outlet
Sampling ports were located 9 feet upstream (1 diameter) from the top of
the 9-foot i.d. stack and approximately 6 feet (2/3 diameter) downstream from
the mesh pad entrainment separator. In accordance with EPA Method 1,
sampling was performed at 48 traverse points. Sampling port and traverse
point locations are presented in Figure 4-3.
4-1
-------
Scrubber
Outlet _
Sampling
Location
Scrubber
Solution
Sampling
Location
Spray- -J
Nozzles
Inlet
Location
PRESSURE MEASUREMENT LOCATIONS
Figure 4-1. Veneer Dryer Exhaust and Scrubber System Sampling Locations
Georgia-Pacific Plywood Plant
Springfield, Oregon
4-2
-------
351/4"I.D.
To Scrubber
Traverse Point
Number
1
2
3
4
5
6
Traverse Point Location
From Inside Duct Wall
(Inches)
1.6
5.1
10.4
24.8
30.1
33.7
Figure 4-2. Scrubber Inlet Sampling Port Configuration and Traverse
Point Locations, Georgia-Pacific Plywood Plant
Springfield, Oregon
4-3
-------
108"
From
Scrubber
T
108"
108"I.D.
Traverse
Point
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Traverse Point Location
From Inside Duct Wall
1.2
3.5
5.9
8.5
11.3
14.3
17.4
21.0
24.8
29.4
34.9
43.0
65.0
73.1
78.6
83.2
87.0
90.6
93.7
96.7
99.5
102.1
104.5
106.8
Figure 4-3. Scrubber Outlet Sampling Port Configuration and Traverse
Point Locations Georgia-Pacific Plywood Plant,
Springfield Oregon
4-4
-------
Method 5X tests performed at this location were 96 minutes in duration.
Method 25 sampling began 18 minutes into each Method 5X test and ran for 60
minutes, leaving 18 minutes of Method 5X testing after the completion of the
Method 25 tests. A total of three Method 5X and 18 Method 25 tests were per-
formed at this location.
4.3 Scrubber Operational Measurement Locations
Scrubber solution samples were taken from the scrubber recirculation tank
during each test. Pressure drop across the scrubber system was calculated
from static pressures in the duct measured at the scrubber inlet and at two
pressure taps each located 2 feet from the scrubber fan inlet and outlet.
These sampling locations are shown in Figure 4-1.
4.4 Fugitive Emissions
Fugitive emissions were observed by RTI and DGA around the veneer dryers
and their abort dampers.
4-5
-------
5.0 SAMPLING AND ANALYTICAL METHODS
This section presents descriptions of sampling and analysis procedures
employed during the emission testing conducted at the Georgia-Pacific plywood
facility in Springfield, Oregon during the week of June 8, 1981. EPA Methods
1, 2, 4, 5X*, 9, 22 and 25 were used to measure emissions at the veneer dryer
exhaust and from the scrubber outlet. These methods are presented in greater
detail in Appendix G.
5.1 EPA Reference Methods Used in This Program
The following EPA Reference Methods were used for the testing at the G-P
plywood plant. These methods** were taken from CFR 40, July 1, 1980, part
60, "Standards of Performance for New Stationary Sources," Appendix A, pp.
183 ff.; and Federal Register, volume 45, no. 194, Friday, October 3, 1980,
pp. 65959 ff.
Method 1 - Sample and Velocity Traverses for Stationary Sources
This method specifies the number and location of sampling points within a
duct, taking into account duct size and shape and local flow disturbances.
Method 2 - Determination of Stack Gas Velocity and Volumetric Flow Rate
This method specifies the measurement of gas velocity and flow rate using
an S-type pitot tube, manometer, and temperature sensor. The physical
dimensions of the pitot tube and its spatial relationship to the temper-
ature sensor and a sampling probe are also specified.
Method 4 - Determination of Moisture Content in Stack Gases
This method specifies the procedures by which the water vapor content of
a gas stream be can determined.
* Method 5X will be assigned a reference letter designation when the NSS
regulation is proposed in the Federal Register. This method was derived
from EPA Method 5 and ODEQ Method 7.
** With the exception of the Provisional Method 5X, which has yet to be
proposed.
5-1
-------
Method 5X - Determination of Particulate and Condensible Organic
(Provisional) Emissions from Stationary Sources in the Plywood Industry
This method, based upon EPA Method 5 and ODEQ Method 7, describes pro-
cedures for measuring emissions in the context of the following defini-
tions. Particulate matter is material which condenses at or above fil-
tration temperature and is collected by the front half of the sampling
train. Condensible organic matter is that material which remains after
extraction, filtration, and evaporation of the impinger portion of the
train.
Method 9 - Visual Determination of the Opacity of Emissions From Sta-
tionary Sources
This method specifies the procedures by which opacity of emissions are
measured.
Method 22 - Visual Determination of Fugitive Emissions from Material
Processing Sources
This method specifies the procedures for visual determination of the pre-
sence and total time of occurence of fugitive process emissions.
Method 25 - Determination of Total Gaseous Nonmethane Organic Emissions
as Carbon
This method describes procedures for the sampling and analysis of gaseous
nonmethane organic emissions. An emission sample is drawn through a con-
densate trap and into an evacuated tank. Trap and tank contents are oxi-
dized to carbon dioxide, reduced to methane, and analyzed by a flame ion-
ization detector.
5.2 Preliminary Measurements
Before the start of emission sampling, each location was tested according
to EPA Methods 1, 2 and 4 to determine the preliminary stack gas velocity and
moisture content within the ducts.
5.3 Measurements for Particulate, Condensible and Noncondensible Emissions
5.3.1 EPA Method 5X (Provisional) - Particulate and Condensible Organic
Compounds
This section presents a summary of procedures followed by TRC during par-
ticulate and Condensible organic sample collection, recovery and preparation,
analysis, and data reduction. Deviations from the specified method are
5-2
-------
explained in this section. Further details of this method are presented in
Appendix G.
5.3.1.1 Method 5X - Sample Collection
The sampling train was a modified EPA Method 5X train as shown in
Figure 5-1. This train was designed and built by TRC. A slipstream was
drawn from behind the heated Method 5X filter to quadruplicate TRC and dupli-
cate NCASI Method 25 sampling trains. Vacuum grease was used in the assembly
of the Method 5X train ahead of the Teflon sample line-impinger train connec-
tion for test 1. This may have caused contamination of the total organic
compound samples for test 1. No vacuum grease was used at those locations
during tests 2 and 3. A minimum amount of grease was used in the impinger
train. Leak checks were performed on the complete sampling train (modified
5X train attached to the six Method 25 trains) before and after each test.
Field data were recorded on standard EPA Method 5 data sheets which are pre-
sented in Appendix C.
The Method 5X sampling train is essentially the same as that described by
EPA Method 5 with the following modifications. A flexible Teflon sample line
was used to connect the outlet of the 4-1/2 inch glass-fiber Gelman Spectro-
grade no. 64948 filter to the impinger train. Since the filter was at a
higher elevation than the impinger train, condensation in the sample line ran
into the first impinger and not back into the filter. The Method 5X impinger
train consisted of four impingers and a 2-1/2 inch glass-fiber filter. The
first impinger was a modified Greenburg-Smith (impingement plate removed)
charged with 100 ml of deionized distilled (D.D.) water. The second impinger
was a regular Greenburg-Smith unit also charged with 100 ml D.D. water. The
third was another modified Greenburg-Smith and was empty. The fourth was
5-3
-------
if,
j
STACK WALL-
17
13
LEGEND
1 - NOZZLE
2 - PROBE
3 - FILTER HOLDER
4 - HEATED FILTER BOX
5 - IMPINGER ICE BATH
6 - UMBILICAL CORD
7 - VACUUM GAUGE
8 - MAIN VALVE TO PUMP
9 - PUMP
10 - BYPASS VALVE
11 - DRY GAS METER
12 - ORIFICE AND MANOMETER
13 - PITOT TUBE AND MANOMETER
14 - THERMOCOUPLE READOUT
15 - FLEXIBLE TEFLON SAMPLE LINE
16 - BACK-UP FILTER HOLDER
17 - THERMOCOUPLES
I I I II I
SLIPSTREAM" TO
METHOD 25 TRAINS
12
Figure 5-1. Modified EPA participate and condensible organics sampling train,
(August 18, 1977 Federal Register)
-------
also a modified Greenburg-Smith type and was charged with 200 grams of silica
gel. A 2-1/2 inch glass-fiber filter (similar to the 4-1/2 inch filter) was
inserted between the third and fourth impinger to collect any organic
material condensed but not collected in the impingers.
Prior to initial field use, all glassware was washed with a chromic acid
solution and rinsed with D.D. water and acetone according to Method 5X.
Sampling train operations were identical to those of EPA Method 5, with
several exceptions. In order to prevent condensation of organic materials in
the probe and on the 4-1/2 inch glass-fiber filter, the stainless steel probe
and the filter were heated to 350° +25°F. Thermocouples were inserted
into the probe and the filter outlet gas stream to ensure that proper tem-
peratures were maintained. These temperatures were noted on the field data
sheet during routine data recording intervals.
During sampling at the scrubber outlet it was sometimes impossible to
maintain probe and filter outlet temperatures in the range of 350 +25 F
because of insufficient heater capacity. Probe and filter outlet tempera-
tures ranged between 310 to 340 F at the scrubber outlet throughout the
test program and averaged about 320 F. Filter box temperatures were main-
tained at 350 ^25 F with no problems.
Impinger outlet temperatures were monitored and maintained below 68 F
throughout the test program. These temperatures were not, however, recorded
on the scrubber outlet field data sheets.
Velocity pressure at the scrubber outlet was extremely low as measured by
the S-type pitot tube, hovering near the lower detection limit of EPA Method
2. In addition, wind gusting across the top of the stack sometimes caused
turbulence within the 9-foot i.d. duct, creating the illusion of negative
flows. If this phenomenon was more than momentary, sampling was halted until
flows again appeared positive, and then restarted.
5-5
-------
A 2 1/2-inch filter was used in the front half of the inlet sampling
train for tests 1 and 2, and the outlet sampling train for test 2. This was
done because the 4 1/2-inch filter assembly would not pass the leak check.
During outlet test 3, the filter assembly was broken in the process of
changing sampling ports. The filter was recovered and replaced with another
filter assembly to complete the test.
5.3.1.2 Method 5X - Sample Recovery and Preparation
Sample recovery was performed in an improvised laboratory on site.
Because this area had a clean, wind-free environment and was well lighted, it
was suited for sample recovery and preparation for shipment.
Sample recovery was performed in accordance with EPA Methods 5 and 5X as
presented in Appendix G. At the conclusion of each test run, separate sample
fractions were collected from each Method 5X sampling train by a three-person
clean-up crew. The liquid samples were placed in glass sample jars with
Teflon-lined lids, and the filters were placed in inert petri dishes and
sealed. The sample fractions collected were as follows:
Container 1 - 4-1/2 inch glass-fiber filter (2-1/2 inch filters were used
during test 1-1, 1-2, and 0-2).
Container 2 - D.D. H2O wash of nozzle, probe and front half of the 4-1/2
inch filter holder.
Container 3 - Acetone wash of nozzle, probe and front half of the 4-1/2 inch
filter holder.
Container 4 - Exposed impinger solution from impingers 1, 2 and 3 and D.D.
H^ wash of impingers, connectors, Teflon sample line, back
half of 4-1/2 inch filter holder and front half of 2-1/2 inch
filter holder.
Container 5 - Acetone wash of first three impingers, connectors, Teflon sam-
ple line, back half of 4-1/2 inch filter holder, and front half
of 2-1/2 inch filter holder.
Container 6 - 2-1/2 inch glass-fiber filter.
5-6
-------
The probe and nozzle were brushed and rinsed three times with D.D. H-O,
which was deposited in container 2. The front half of the 4-1/2 inch filter
holder was also rinsed with D.D. H2O, which was deposited in container 2.
The probe, nozzle and front half of the 4-1/2 inch filter holder were brushed
and rinsed with acetone in the same manner and deposited in container 3.
The Teflon sample line was drained into the impinger train. The Teflon
sample line was not brushed because the particulate catch in the sample line
is generally considered to be insignificant. Impinger contents were weighed
to determine moisture catch and deposited in container 4. The Teflon sample
line, impingers, connectors and the back half of the 4-1/2 inch filter holder
were rinsed three times with D.D. HO into container 4, and then rinsed
three times with acetone into container 5.
Prior to tests 2 and 3 both the probe and Teflon sample line were washed
with D.D. H_0 after the acetone wash to remove any acetone residue which
might have contaminated the EPA Method 25 samples. These washes were dis-
carded and the components allowed to dry at ambient conditions before being
reassembled. A possibility of Method 25 acetone contamination exists for
test 1.
Both filters were removed from their holders and deposited into their
respective petri dishes, containers 1 and 6. Filter residue on the filter
holders was scraped and deposited into the same acetone rinse containers as
the front halves of their respective filter holders. The stainless steel and
glass filter frits used in the filter holders were not rinsed during sample
recovery, because any organic material collected on the frits is generally
considered to be insignificant. Glass and/or metal particles could become
detached and contaminate sample fractions.
Silica gel samples were weighed immediately at the conclusion of each
test and the weights recorded by the clean-up crew. All Method 5X samples
5-7
-------
were packed in shock-proof containers and driven to the CH_MHill laboratory
in Corvalis, Oregon for analysis at the conclusion of the test program.
5.3.1.3 Method 5X - Sample Analysis
With the exception of the silica gel samples, all sample fractions were
analyzed by CHJlHill. CHJIHill was chosen to perform the analytical
*L ft
phase of the Method 5X sampling program because of their extensive experience
with Oregon DEQ Method 7, from which EPA Method 5X was derived. All analyses
were performed in accordance with EPA Method 5X and as approved by EPA/EMB.
The sample fractions were analyzed as follows:
Container 1 - (4-1/2 inch glass-fiber filter) - desiccate and weigh after 24
hours.
Container 2 - (D.D. 1^0 probe rinse) - evaporate, desiccate and weigh after
24 hours.
Container 3 - (acetone probe rinse) - evaporate, desiccate and weigh after
24 hours.
Container 4 - (impinger water solution and D.D. ^0 rinse) - extract,
desiccate and weigh.
Container 5 - (impinger acetone wash) - evaporate, desiccate and weigh.
Container 6 - (2-1/2 inch glass-fiber filter) - desiccate and weigh after 24
hours.
Silica gel samples were weighed on site with a triple-beam balance at the
conclusion of each test by the Method 5X sample recovery crew. The weight
gain of the silica gel was determined to the nearest 0.5 gram and recorded.
All analytical data were recorded on the data sheets as presented in
Appendix H. Sample residue remaining after analysis was~ retained for 90 days
after the end of the field program after which they were discarded according
to EPA instructions.
5-8
-------
5.3.1.4 Method 5X Data Reduction
All Method 5X data reduction was performed in a manner identical to pro-
cedures described by EPA Method 5. (See Appendices B and G.) The only
variation from these calculations was as follows. Because of the
unacceptable super- isokinetic sampling conditions during tests 1 and 2 at
the scrubber inlet and tests 1 and 3 at the scrubber outlet, the particulate
mass emission rate (MER) for these runs were calculated by two methods: the
concentration method (by which calculations are normally done) and the area
ratio method.* With the former method, the concentration of particulate
matter entering the nozzle is calculated and then multiplied by the
volumetric flow rate to obtain the mass emission rate:
(m/V) x Q = MER (Ibs/hr) (Eq. 5-1)
where m = amount of particulate sampled (Ibs)
V = volume of sampled gas (DSCF)
Q = volumetric flow rate (DSCF/hr)
If the nozzle sampling velocity is greater than the stack gas velocity
(superisokinetic sampling conditions), then the calculated mass flow rate
will be less than the true MER. This is because the heavier particles will
leave their streamlines (gas streamlines diverted into the nozzle) and will
not enter the nozzle, as they would under isokinetic conditions. Since the
volume of gas sampled is greater than what would be sampled under isokinetic
conditions, the concentration (m/V) will be less than that under isokinetic
conditions.
With the area ratio method, the mass of particulate matter collected is
divided by the sampling time and then multiplied by the ratio of the stack
area to the nozzle area to obtain the mass emission rate:
* Brenchley, D.F., C.D. Turley and R.F. Yarmac. Industrial Source Sampling.
Ann Arbor Science Publishers, Inc., 1973, p. 173 ff.
5-9
-------
(m/t) x (A /A ) = HER (Ibs/hr) (Eq. 5-2)
s n
where m = amount of particulate sample (Ibs)
t = sampling time (hrs)
As = area of stack (ft2.)
An = area of nozzle (ft"2.)
Again, if the nozzle sampling velocity is greater than the stack gas
velocity, then the MER calculated by this method will be somewhat greater
than the true MER. The lighter particles follow the diverted streamlines
into the nozzle; the amount of particulate matter sampled in time (t) is
therefore assumed to be greater than what should be sampled. The volume of
sampled gas is not a factor in this calculation. The average of the two
calculated MERs was used as an estimate of the true MER.
Gas stream moisture content measured at the scrubber outlet exceeded the
saturation values for the duct temperatures measured. Therefore moisture
values were recalculated using the following psychometric equation:
HO =
B
where Py = vapor pressure of air at a given temperature
PB = barometric pressure
5.3.2 EPA Reference Method 25 - Condensible and Noncondensible Organic
Compounds
This section presents a summary of procedures followed by TRC during con-
densible and noncondensible organic sampling equipment preparation, sample
collection, field sample recovery, and sample analysis. The TRC Method 25
sampling train is shown in Figure 5-2. Deviations from the method are also
5-10
-------
SWAGELOK
CONNECTORS
H2°-l£
ICE W**
CONDENSATE TRAPS
VACUUM
GAUGE
FLOW
RATE
CONTROLLER
ON/OFF FLOW
VALVE
OUICK r±|
CONNECT '1
EVACUATED
SAMPLE
TANK
Figure 5-2. Method 25 Sampling Train
-------
explained in this section. Further details of Method 25 are presented in
Appendix G. NCASI Method 25 procedures are also presented in Appendix G.
5.3.2.1 Method 25 - Sampling Equipment Preparation
This procedure is based on and supplements EPA Method 25 / "Determination
of Total Gaseous Nonmethane Organic Emissions as Carbon."*
Condensate Trap
After being checked for any sign of physical damage, each trap was inter-
connected to a hydrocarbon (HC)-free air cylinder, flowmeters and C0_ moni-
tor (nondispersive infrared detector (NDIR)) and inserted in the furnace as
shown in Figure 5-3. The trap was then purged with the HC-free air at a 100
ml/min flow rate with the furnace operating at a temperature of 600 C. A
propane torch was used to heat those portions of the trap and probe assembly
that extend outside the furnace. The purge was performed until the C02
monitor indicated a concentration of 10 ppra or less.
Sample Tank
Each sample tank was connected to a cylinder of HC-free air, a vacuum
pump, and a mercury manometer as shown in Figure 5-4. The tank was evacuated
to 29 inches Hg vacuum after which the three-way valve was switched and the
tank pressurized to 10 inches Hg with HC-free air. This cycle was repeated
three times. After the third pressurization, the tank was connected to the
TGNMO analyzer and a sample analysis was performed. If a nonmethane organic
concentration greater than 10 ppm was measured, the tank was again subjected
* Federal Register, volume 45, no. 194, October 3, 1980, pp. 65959-73.
5-12
-------
® FLOW
T METER
HC __
FREE I
""
Figure 5-3
t
3-WAY
/— v VALVE
T
HC
FREE
AIR
co2
\^\ ANALYZER "*"
FLOW
\ METER
-* ^ TUflP
iw '»"'
|
*— J !?5? ^- rURNACE
Mm%
. Method 25 Trap Preparation
nilTTK
TANK
v^x
Figure 5-4. Method 25 Tank Purging and Evacuation
ADJUSTMENTVALVE
FLOWMETER
ON/OFF
VALVE
FLQW CONTROL
ASSEMBLY
PRESSURE GAUGE
TRAP
QUICK CONNECT
TANK
Figure 5-5. Method 25 Flow Control Assembly Adjustment
5-13
-------
to the evacuation-pressurization analysis procedure until accepted. Each
tank was then evacuated and pressurized with dry nitrogen for shipment to the
field.
Flow Control Assembly
The sampling train was assembled as shown in Figure 5-5 and leak
checked. The probe end cap was removed and the probe connected to a flow
meter as shown. The sample flow shut-off valve was opened and the flow con-
trol valve adjusted to achieve a flow rate of 50 +5 ml/minute. The flow con-
trol adjustment screw was sealed after the flow rate was achieved. The flow
control valve number and calibration data were recorded on forms presented in
Appendix E.
5.3.2.2 Method 25 - Sample Collection
The sampling train was a modified EPA Method 25 apparatus. The modifi-
cation consists of placing an additional condensibles trap, immersed in a
water ice bath, ahead of the trap immersed in dry (C0_) ice. (See
Figure 5-2.) The additional trap is intended to remove the high moisture
content associated with the process emission streams and prevent freezeup in
the dry ice trap which leads to premature sample flow stoppage.*
The sample tanks were shipped to the site slightly pressurized with dry
nitrogen. Immediately prior to each test, tanks were evacuated. The tank
vacuum, ambient temperature and barometric pressure were recorded on the
field sampling data sheet. (See Appendix C.)
* "Method Development for the Plywood/Plywood Veneer Industry," EPA Contract
68-02-3543, Work Assignment 1. TRC - Environmental Consultants, Inc.,
August 1981.
5-14
-------
With the flow shut-off valve in the closed position, the train was
checked again after a minimum period of 10 minutes. If the indicated vacuum
had not changed, the portion of the sampling train behind the shut-off valve
did not leak and was considered acceptable. Assuring that the probe tip was
tightly capped, the front part of the sampling train was leak checked by
opening the flow shut-off valve. After a short period to allow pressure
stabilization (not more than 2 minutes), the indicated gauge vacuum was
formed. The tank vacuum as indicated by the vacuum gauge was recorded and
assembled as shown in Figure 5-2. The pretest leak check was then per-
formed. After a minimum period of 10 minutes, the indicated vacuum was again
noted. The leak check was considered acceptable if no visible change in
vacuum occurred. The pretest leak rate (inches Hg/10 minutes) was recorded
if observed. At the completion of the leak checks, the sample flow shut-off
valve was closed.
After the leak check had been performed, the sample tank number and trap
numbers for each sampling train were recorded on the field data sheet with
the respective test run number and sampling site. Four TRC and two NCASI
sampling trains were connected to each Method 5X sampling train at the
insulated outlet of their respective hotbox filters. Immediately prior to
sampling, the gauge vacuum and clock time were noted. The flow shut-off
valve was opened and sampling begun. TRC gauge vacuum readings were recorded
every 5 minutes during the sampling period. At the end of the sampling
period, the flow shut-off valve was closed, the time and final gauge vacuum
recorded. After the Method 5X sampling was completed, the Method 25 probe
lines were disconnected from the Method 5X interface and tightly capped.
A post-test leak check was performed prior to disassembly of the sampling
train. After assuring that the probe had been tightly capped, the flow shut-
5-15
-------
off valve was opened and the gauge vacuum monitored for a minimum of 10
minutes. The leak check was acceptable if no visible change in tank vacuum
occurred. The post-test leak rate (inches Hg/10 minutes) was recorded if
observed. At the completion of the leak check, the flow shut-off valve was
closed.
5.3.2.3 Method 25 - Field Sample Recovery
After the post-test leak check was completed, the TRC sampling train com-
ponents were disconnected. Both ends of each condensibles trap were tightly
sealed. The traps were then packed in dry ice for sample preservation and
shipment to the laboratory.
5.3.2.4 Method 25 - Sample Analysis
Two inlet sampling trains and two outlet sampling trains from each test
were analyzed by TRC. The other two inlet sampling trains and two outlet
sampling trains from each test were analyzed by Pollution Control Science,
Inc., (PCS). NCASI analyzed the samples collected with their equipment.
Additional analyses of the PCS analyzed samples were performed at the TRC
laboratory. The purpose of this approach was to identify the cause of the
poor paired sample data precision obtained during the method development pro-
gram.
The analyses were performed in general accordance with the method as pub-
lished (Appendix F). Prior to the nonmethane organic analysis of the tank
samples, a preliminary analysis was performed with an FID to determine the
relative sample concentrations. The purpose of this analysis was to provide
additional information for the resolution of any poor precision. If the pre-
5-16
-------
liminary tank sample analyses indicated relatively equal tank concentrations
for paired samples/ the nonmethane organic analyses should also produce
relatively equal values. These data are presented in Appendix F.
Preliminary Sample Tank Analysis
The preliminary sample tank analysis was performed using the nonmethane
organic analyzer. However, the separation column, oxidation catalyst, and
reduction catalyst were bypassed. The sample tank was pressurized to approx-
imately 100 mmHg gauge and then connected to the analyzer. The sample loop
was purged an
-------
Ul
I
CD
I
SEPARATION
COLUMN
NONMETHANE
ORGANIC
(BACKFLUSH)
CO
C02
1 CH4
COLUMN
BACKFLUSH VALV
OXIDATION
CATALYST
HEATED CHAMBER
VALVE
AIR
SAMPLE
INJECT
VALVE
REDUCTION
CATALYST
HEATED CHAMBER I
|
DATA
RECORDER
SAMPLE
TANK
CALIBRATION
CYLINDERS
MOLECULAR
SIEVE
FLOW
METER
CARRIER
GAS
He
Fiaure 5-6. TRC Nonmetharie Organic Analyzer
-------
A six-port valve (Carle Model 5521) was substituted for the two four-port
valves in the oxidation catalyst flow scheme. One four-port valve was used
instead of two four-port valves in the reduction catalyst flow scheme. In
effect the latter valving modification precluded hydrogen venting within the
laboratory.
The exit line from the oxidation furnace to the six-port valve was heat
traced to avoid condensation. Additionally, all four switching valves incor-
porated in the analyzer were enclosed in a heated, insulated compartment
thermostatically controlled to maintain a constant 100°C temperature.
The separation column used was prepared by Supelco, Inc. It was a 4-1/2
foot long, 1/8-inch diameter stainless steel tube with two packed sections.
The injection side section was 3 feet long and contains 10 percent OV-101
(liquid methyl silicone) on 80/100 mesh Supelcoport. The following section
was 1-1/2 feet long packed with 60/80 mesh Poropak Q.
The reduction catalyst was a Byron Instruments unit with integral
heater. This was mounted within the Varian gas chromatograph oven to ensure
constant temperature operation.
Although not clearly shown in Figure 5-6, a single combustion air source
services both the oxidation catalyst and the flame ionization detector.
Individual metering valves are used after the flow splitter to regulate the
supply to each device.
The condensate recovery and conditioning apparatus equipment was assem-
bled by TRC as shown in Figure 5-7 and was essentially the same as the con-
figuration detailed by the method. The NDIR incorporated was an Anarad AR
400, with a range of 0 to 10,000 ppm C02>
The TRC arrangement did not incorporate the vacuum pump in a direct link
with other equipment. Instead it was located remotely. This was done to
avoid contamination by the oil mist vented from the vacuum pump.
5-19
-------
A tube furnace was used for volatilization of the condensate trap
sample. This provides more even, high temperature heating of the trap. A
propane torch was used to heat those parts of the trap, including the probe,
which remain outside the furnace during the sample recovery procedure.
Valves A, B, C and D in Figure 5-7 and their connecting tubing were enclosed
in a thermostatically controlled oven maintained at 180°C to prevent con-
densation. An oxygen rich carrier gas passed through the condensate trap
during heating and oxidized the organic compounds to C0_ and water vapor.
The flow exited the trap, passed through a water trap and NDIR, and entered
the intermediate collection vessel.
Analyzer Operating Conditions;
Gas Regulator Pressure (psig) Flow Rate (cc/min)
Helium 42 25
Air 45 30 FID
50 Oxidation Catalyst
Hydrogen 20 30
Separation column normal temperature - 0°C
Separation column backflush temperature - 100°C
Oxidation catalyst temperature - 750°C
Reduction catalyst temperature - 100°C (32 VAC)
Condensate Recovery Conditions;
Gas Regulator Pressure (psig) Flow Rate (cc/min)
Oxygen 10 150
Air 15 50
Oxidation catalyst temperature - 850°C
Details of the NCASI analyzer and procedures are presented in Appendix H.
Nonmethane Organic Analysis Procedure
The analysis was performed in accordance with the published procedure.
(See Appendix H.) However, the condensate trap carbon dioxide purge (Section
A.3.2 of the published procedure) was modified. After briefly purging the
trap according to the procedure, the valves were switched so that the trap
5-20
-------
L±3
FLOW
METERS
FLOW
xCONTROL\
V. VALVES^
HEATED
SWITCHING
VALVES
CONNECTORS
CATALYST
HEATED
CHAMBER
I
I
i ' I ' '
Llt-i
I SAMPLE
CONDENSATE
| TRAP
I
OXIDATION
CATALYST
I
HEATED |
_CHAMBER,_ J
VENT
1 1 \~* J
NDIR
ANALYZER
_,
VALVE A
QUICK A
CONNECT[Q|
HoO
TRAP
MERCURY INTERMEDIATE
MANOMETER
Figure 5-7. TRC Condensate Recovery and Conditioning Apparatus
5-21
-------
was bypassed. After the trap had been bypassed, the carrier gas flow con-
tinued through the system and into the tank for approximately 5 minutes. It
was then vented to the atmosphere through the valve located downstream of the
NDIR. (See Figure 5-7.) This time period was sufficient to purge the inter-
connecting tubing and NDIR cell volume. Prior to resuming flow through the
condensate trap, the valve was switched to introduce again the flow into the
sample tank. The trap was removed from the dry ice bath and allowed to warm
to room temperature (determined by touch) . The trap was placed back into the
dry ice bath and the valves switched to resume carrier gas flow through the
trap after frosting appeared on external trap surfaces. The procedure was
then completed as described. This modification to the procedure was intended
to assure the removal of any CO which may be trapped within the ice
crystals present in the trap.*
5.4 Preliminary Moisture Determination
Preliminary moisture tests were performed at the scrubber inlet and
outlet prior to emission testing. Testing was performed in accordance with
EPA Method 4. Data were recorded on field moisture determination forms as
presented in Appendix C.
5.5 Preliminary velocity Determination
Preliminary velocity measurements were made at the scrubber inlet and
outlet prior to emission testing. EPA Methods I and 2 were followed in
measuring the velocity of the gas stream. Data were recorded on the field
data sheets (Traverse point Location for Circular Ducts and Preliminary
Velocity Traverse, Appendix C).
* "investigation of Carbon Dioxide Interference with Method 25." EPA Contract
68-02-2814, Work Assignment 41. Midwest Research Institute, April 15,
1981, p. 7.
5-22
-------
5.6 Visible Emissions
Scrubber outlet visible emissions were not monitored as planned because
of overcast sky background conditions. The controlled emissions were bluish-
white which made it impossible to distinguish the emissions from the overcast
sky. Overcast skies were present on June 8, 9 and 10. During the last test
day, June 11, the sky began to clear and only scattered clouds were present
in the afternoon. However, the final test sequence was nearly completed by
this time. Consequently, no visible emission observations were recorded.
Although the scrubber outlet stack had an attached steam plume, the
Method 9 observations were not cancelled because of this condition. The
method provides for attached steam plumes by requiring that observations be
made at the point where the condensed water vapor is no longer visible.
5.7 Pressure Drop Measurements
The pressure drop across the scrubber system was measured in order to
determine if the unit was operating at design conditions and to provide a
means of correlating a scrubber operating parameter with collection effi-
ciency. O-tube water manometers were used to measure static pressure at
three points at 30-minute intervals during the test period. Measurements
were made at the inlet sampling location, just before, and just after the
induced draft fan. Pressure drop across the scrubber system was then cal-
culated from the static pressures.
5.8 Scrubber Solution Samples
Scrubber solution samples were taken from the holding tank of the recir-
culating system concurrently with the particulate/condensible organics test-
ing performed at the scrubber outlet. These samples were taken to determine
5-23
-------
the relationship between the total organic carbon content of the scrubber
solution and the actual measured emissions.
A 100-ral sample was taken approximately every 30 minutes while Method 5X
was being performed at the outlet. These samples were taken by dipping a 100
ml graduated cylinder into the holding tank. Sample numbers and collection
times were recorded on the Scrubber Solution Sample Collection form. The
100-ral aliquots collected during a test were composited into one sample jar
for that test.
The composite samples were placed in shock-proof containers and trans-
ported to CH2MHill at the conclusion of the test program. The scrubber
solution samples were analyzed for total organic carbon (TOC) following Stan-
dard Method 415.1. All analyses were performed within two weeks of sample
collection.
5.9 Fugitive Emissions
The purpose of these measurements was to visually determine the frequency
of occurrence of emissions that are not emitted directly from the process
stack or duct. These are generally referred to as fugitive emissions and
include such emissions as those: (1) escaping capture by process equipment
exhaust hoods; (2) emitted during material transfer; (3) emitted from build-
ings housing material processing or handling equipment; (4) emitted directly
from process equipment.
EPA Method 22 modified guidelines, as presented in Appendix F, were used
to determine fugitive emissions from the veneer dryer doors and abort
stacks. The method does not require that the opacity of emissions be deter-
mined. Instead, the amount of time that any visible emissions occur during
the observation period is measured.
5-24
-------
Fugitive emissions from the veneer dryers were monitored by RTI and DGA.
Observations were recorded periodically throughout the test program.
5.10 Ambient Temperature and Relative Humidity
Outdoor ambient air temperature and relative humidity were measured with
a psychrometer at the beginning and end of each test period. Measurements
were made by RTI and their subcontractor to determine if a relationship
exists between ambient temperature and relative humidity, and the emissions
from the veneer dryers. Data was recorded on a form presented in Appendix I.
5-25
-------
6.0 QUALITY ASSURANCE
The TRC quality assurance program is designed to ensure that emission
measurement work is performed by qualified people using proper equipment
following written procedures in order to provide accurate, defensible data.
This program is based upon the EPA Quality Assurance Handbook for Air Pollu-
tion Measurement Systems, Volume III (EPA-600/4-7-027b).
At the beginning of each day, a meeting was held to orient personnel to
the activities scheduled for that day and to discuss results from the pre-
vious day, and to determine if any special considerations were appropriate
for the day's work.
6.1 Method 5X
TRC's measurement devices, pitot tubes, dry gas meters, thermocouples,
probes and nozzles are uniquely identified and calibrated with documented
procedures and acceptance criteria before and after each field effort.
Records of all calibration data are maintained in TRC files. Samples of
these calibration forms are presented in Appendix F.
All Method 5X sampling shall be 100 +10 percent isokinetic. Probe and
hotbox temperatures were maintained at 350° +25°F. Deviations from these
criteria were reported to the EPA/EMB task manager to decide whether a test
run should be repeated or continued.
A single clean-up evaluation test was performed on each initial set
(collector train) of glassware prior to collecting field samples. The eval-
uation tests (Method 5X) were performed in the field clean-up laboratory and
were observed by the EPA task manager. Necessary changes or modifications to
the clean-up procedures were specified by the EPA task manager prior to
collecting field samples. The sets of glassware, including the probes,
6-1
-------
were prepared and precleaned before conducting the clean-up evaluation
tests. The impingers were precharged as specified in the actual test pro-
gram. Afterward, the sample collectors, including probes, were cleaned and
the blank samples recovered and analyzed as specified in the actual test pro-
gram. Results are presented in Section 2 of this report.
In summary, the evaluation tests were designed to precondition the sample
collectors, to establish blank background values, and to educate the clean-up
personnel in specific sample recovery procedures.
Acetone was provided by CH_MHill in glass-lined containers. Both the
acetone and D.D. water were analyzed by CH MHill prior to field use. Resi-
due data from this preliminary analysis was evaluated by the EPA/EMB task
manager with respect to the suitability for use during the test program.
These data are presented in Appendix H. In addition, three blank samples of
D.D. water, acetone, and both 2-1/2 inch and 4-1/2 inch filters were
collected for background analysis. All clean-up evaluation and blank samples
were analyzed in conjunction with the actual test samples.
All sample recovery was performed by a three-person clean-up crew.
Appropriate sample recovery data were recorded on the sample identification
log, sample handling log, chain-of -custody form, and analytical data forms as
presented in Appendix D.
Recovered samples were secured in padlocked, shock-proof, steel con-
tainers for storage and shipment for analysis.
All preparation and analysis of Method 5X samples were performed by
111, which has extensive experience with Oregon DEQ Method 7, from
which Method 5X derives. CH MHill adhered to the standards of quality
assurance set forth in the Quality Assurance Handbook for Air Pollution
6-2
-------
Measurement Systems, Volume III (EPA-600/4-7-027b) and the Handbook for
Analytical Quality Control in Water and Wastewater Laboratories (EPA-600/4-
79-019, March 1979).
6.2 Method 25
Method 25 traps were burned out according to the method prior to testing
and spot-checked for contamination. All Method 25 tanks were flushed with
nitrogen and checked for contamination prior to field use.
Six sampling trains were used to provide a check on data precision. Two
trains were analyzed by TRC; two by PCS, and NCASI analyzed the remaining two
trains. All tanks and traps have permanently engraved identification numbers.
Analyzers were calibrated over the specified ranges using certified cali-
bration gases. Certification forms are provided in Appendix F.
6.3 Method 9
The TRC observer had been certified within the past 6 months to perform
visible emission evaluations.
6-3
------- |