United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 80-WFB-6
May 1980
Air
Nonfossil Fueled Boilers

Emission Test Report
U.S. Sugar Company
Bryant,  Florida

-------
          NONFOSSIL FUELED BOILERS
            Emission Test Report
             U.S.  Sugar Company
               Bryant,  Florida
           Project No.:   80-WFB-6
                Prepared for

       Environmental Protection Agency
Office of Air Quality Planning and Standards
         Emission Measurement Branch
           Research Triangle Park
            North Carolina  27711
                     by

    James A. Peters and Charles F.  Duncan
 Contract 68-02-2818, Work Assignment No. 25
                  May 1980
        MONSANTO RESEARCH CORPORATION
              DAYTON LABORATORY
             1515 Nicholas Road
             Dayton, Ohio  45407

-------
                        TABLE OF CONTENTS

Figures	      iv
Tables 	       v

     1.  Introduction	       1
     2.  Summary of Results	       2
     3.  Process Description 	       8
     4.  Location of Sampling Points 	      12
     5.  Sampling and Analytical Procedures	      13

Appendices

     A.  Complete emission results 	      17
     B.  Field data sheets	      31
     C.  Analytical data sheets	      58
     D.  Boiler operating data	      75
     E.  Project participants	      87
                              111

-------
                             FIGURES
Number                                                      Page
  1       Schematic process flow diagram:   Bryant Mill,
            U.S. Sugar Corp	      8
  2       Plot plan:  Bryant Mill,  U.S.  Sugar Corp ...      9
  3       Spray impingement scrubber 	     10
  4       Diagram of Bryant Plant emission control
            equipment	     11
  5       Scrubber outlet/roof stack sampling location .     12
                                IV

-------
                             TABLES

Number                                                      Page

   1      Bryant Plant Sampling and Analysis Schedule. .       2

   2      Particulate Emission Data and Stack Gas Para-
            meters, U.S. Sugar-Bryant Mill, December 17-
            18, 1979 (English Units)	       4

   3      Particulate Emission Data and Stack Gas Para-
            meters, U.S. Sugar-Bryant Mill, December 17-
            18, 1979 (Metric Units)	       4

   4      Summary of Integrated Gas Analyses, U.S. Sugar-
            Bryant Mill, December 17-18, 1979	       5

   5      Summary of Andersen Particle Sizing Results,
            U.S. Sugar-Bryant Mill, December 17-18, 1979      5

   6      Summary of NO  Emissions, U.S. Sugar-Bryant
            Mill, December 17-18, 1979	       6
   7      Summary of Fuel Ultimate Analyses, U.S. Sugar-
            Bryant Mill, December 17-18, 1979	

   8      Summary of Boiler Operations During Testing,
            U.S. Sugar-Bryant Mill, December 17-18, 1979

-------
                            SECTION 1

                          INTRODUCTION


The Bryant Mill of U.S. Sugar Corporation in Bryant,  Florida was
emission tested by Monsanto Research Corporation (MRC) for the
U.S. Environmental Protection Agency (EPA) under Contract No.
68-02-2818, Work Assignment No. 25.  The objective of the sampling
program was to obtain emissions data from well-controlled sources
within the nonfossil fuel boilers category that could possibly be
used for the development of new source performance standards.

The field test work was monitored by Dan Bivins, Field Testing
Section, Emission Measurement Branch, EPA.  The sampling performed
by MRC was directed by Charles F. Duncan as team leader.  Gaseous
and particulate emissions were determined at the outlet of the
pollution control device serving Boiler #2.  A composite sample
of boiler feed was collected with each run so that a material
balance could be attempted.

The sampling at the Bryant Mill was conducted by MRC during
December 16-18, 1979.  The collection methods employed were EPA
Methods 1, 2, 3, 4, 5, 6, 7, and 9, with particulate sizing by
Andersen cascade impactor.

Quality assurance/quality control in the sampling area covered
such activities as instrument calibration, using standard or
approved sampling methods, chain-of-custody procedures, and pro-
tocols for the recording and calculation of data.  QA/QC in the
analysis area involved using only validated analysis methods,
periodic operator QC checking and training, sample QC by the use
of splits, reference standards, and spikes, and interlaboratory
audits.

-------
                             SECTION 2

                       SUMMARY OF RESULTS
Pollutants which were measured for this emission test were  partic-
ulate matter, particle  size,  C02,  CO,  S02 ,  NO ,  and plume opacity.
Table 1 presents the sampling and analysis  scnedule in condensed
form.

      TABLE  1.  BRYANT  PLANT  SAMPLING AND ANALYSIS SCHEDULE

Sampling
site
Scrubber
outlet
Scrubber
outlet
Scrubber
outlet
Scrubber
outlet
Scrubber
outlet
Total
number of
samples
3
3
3
3
3 runs,
4 samples
each
Minimum
Sampling sampling Initial analysis
Sample type
Particulate
matter
Particle-size
distribution
Integrated gas
analysis
S02
NO
X
method time Type
EPA 5 60 min
Andersen
EPA 3 C02, 02,
CO
EPA 6 , Same as
option 2 Method 5
EPA 7 15 min
intervals
Method


EPA 3


Scrubber     3      Opacity
  outlet

Scrubber  3  samples,  ASTM
  outlet  2 fuel
         analyses
         each
EPA 9
                     Ultimate   ASTM
                      analysis
                      and heat-
                      ing value
The-Bryant Mill  operates three waste-fired boilers  fed with
bagasse.  The  center boiler,  Boiler #2, was tested.   Boiler #2
utilizes dual  scrubbers in parallel for pollution abatement.  The
outlet stack is  located directly above the scrubbers.

-------
Three test runs were performed, each consisting of 96 minutes of
sampling time.  Forty-eight traverse points were used, six points
in each of the eight sampling ports.  The first run was completed
December 17.  During the run, the boiler operated normally, in the
range of 145,000 to 160,000 Ib/hr of steam, until more than half-
way through the test, when the bagasse feed was interrupted.  The
steam loading dropped to about 60,000 Ib/hr and oil began to be
burned.  The test was interrupted several minutes after the drop
in steam loading and was begun again after the bagasse feed rate
and the boiler operation returned to normal almost 2 hours later.
During the last several minutes of the test before the interrup-
tion, about 75 gal of oil was burned.  Bagasse alone was burned
the remainder of the run.

The remaining two runs of the test were completed on December 18.
Through both runs the boiler operated normally and bagasse alone
was burned.  The steam loading ranged from 125,000 to 165,000
Ib/hr, with an average of 151,000 Ib/hr, in Run 2 and from 130,000
to 170,000 Ib/hr, with an average of 144,000 Ib/hr, in the third
run.  Both runs were within the normal operating range.  During
the third run, soot blowing was performed.

Tables 2 and 3 contain the summarized particulate emission data
and stack gas parameters.  Moisture in the stack gas was unusually
high -- 32 percent H20.  Integrated gas analysis results for each
run are given in Table 4.

Table 5 contains a summary of the particle sizing results; each
Andersen cascade impactor run was made after completing a Method 5
run.  The #1 impactor test was discarded because the filter media
was soaked with water.  Due to the boiler #2 plume merging with the
other boilers' plumes, opacity readings were not able to be made.

Samples for SC>2 emissions were taken concurrently with particulate
emission runs by using the back half of the Method 5 train.  Due
to the very low sulfur content of the bagasse feed, emissions of
S02 were below the detection limit  (3.4 mg S02/m3) of Method 6,
and no data are presented.

Samples for NO  emissions were collected just after each particu-
late emission rest and are summarized in Table 6.

Composite fuel samples of bagasse were taken with each run from
the conveyor feeding the boiler, and ultimate analysis and fuel
values were determined.  A fuel oil sample from run #1 was also
collected and analyzed for fuel value.  Table 7 presents the fuel
analysis results.

A summary*of boiler operating conditions during testing is given
in Table 8.  Average steam temperatures and pressures were deter-
mined by averaging 15-min readings  in order to calculate steam
enthalpy-

-------
      TABLE 2.  PARTICULATE EMISSION DATA AND STACK GAS  PARAMETERS, U.S.
                SUGAR-BRYANT MILL,  DECEMBER 17-18, 1979  (ENGLISH UNITS)
Stack
Run Time, temperature, Flow, H20, Is
number Date min °F dscfm "/,
1 12/17/79
2 12/18/79
3 12/18/79
Average
96 161 58,515 31.3
96 164 58,720 33.1
96 162 58,825 31.7
96 162 58,687 32.0

okinetic.
% gr/dscf
105.7 0.1298
105.6 0.1001
101.6 0.1135a
0.1145
Emissions
Actual
Ib/hr Ib/mm Btu
65.1 0.3505
50.4 0.2547
57.2 0.3034
57.6 0.3029

Corrected to
12% CO,
gr/dscf
0.1442
0.1082
0.12053
0.1243
aRun #3 included a soot blow.
TABLE 3. PARTICULATE EMISSION DATA AND STACK GAS PARAMETERS,
U.S. SUGAR-BRYANT MILL, DECEMBER 17-18, 1979 (METRIC UNITS)
Run
number Date
1 12/17/79
2 12/18/79
3 12/18/79
Average
Stack
Time, temperature, Flow, H20,
min °C dncmpm %
96 72 1,657 31.3
96 73 1,663 33.1
96 72 1,666 31.7
96 72 1,662 32.0

Iscrkinetic,
% gr/dncm
105.7 0.2971
105.6 0.2292
101.6 0.2599a
0.2621
Emissions
Actual
kg/hr kg/GJ
29.5 0.1506
22.9 0.1097
26.0 0.1307
26.1 0.1303

Corrected to
12% C02
gr/dncm
0.3301
0.2478
0.27603
0.2846
aRun #3* included a soot blow.

-------
  TABLE 4.   SUMMARY OF INTEGRATED  GAS ANALYSES,  U.S.
              SUGAR-BRYANT MILL, DECEMBER 17-18, 1979

Run
number
1
2
3
Average
Date
12/17/79
12/18/79
12/18/79
C02,
%
10.8
11.1
11.3
11.1
CO,
%
0.0
0.0
0.0
0.0
02,
%
9.2
9.0
9.4
9.2
N2,
%
80.0
79.9
79.3
79.7
MW
Ib/lb mole
30.1
30.1
30.2
30.1

TABLE  5.  SUMMARY  OF ANDERSEN PARTICLE SIZING RESULTS,
           U.S. SUGAR-BRYANT MILL,  DECEMBER 17-18,  1979
                            Run No.  1
                            Discarded
                            Run No.  2
       Flow rate  = 0.927 acfm
       Isokinetic rate = 107.1%
          Stage
 Size range
             Percent in
             size range
         Cumulative %
         10.50
0 >10.50
1
2
3
4
5
6
7
Filter
6.50
4.30
2.95
1.88
0.94
0.58
0.39
0.0
- 10.50
- 6.50
- 4.30
- 2.95
-1.88
-0.94
-0.58
- 0.39
3.99
1.46
3.06
7.98
11.30
12.40
12.90
19.15
16.49
11.30
94.55
94.55
91.52
83.54
72.24
59.94
46.94
27.79
11.30
0
                            Run No.  3
       Flow rate  =  0.908 acfm
       Isokinetic rate = 105.5%
          Stage
 Size range
             Percent in
             size range
         Cumulative %
         10.60
   >10.60
6.60
4.40
3.00
1.90
0.96
0.59
0.40
      10.60
      6.60
      4.40
      3.00
      1.90
      0.96
      0.59
0.0 - 0.40
 6.56
 2.01
 4.28
 7.47
 8.66
 8.66
10.48
20.60
16.68
14.59
91.43
91.43
87.14
79.67
71.01
62.35
51.87
31.27
14.59
 0

-------
        TABLE  6.   SUMMARY OF  NOX EMISSIONS,  U.S.  SUGAR-
                   BRYANT MILL,  DECEMBER 17-18, 1979
Run
number
1-1
1-2
1-3
1-4
Average
2-1
2-2
2-3
2-4
Average
3-1
3-2
3-3
3-4
Average
Date 	 ppm
12/17/79 0.572
0.402
0.525
0.751
6.563
12/18/79 1.222
1.119
0.580
0.079
0.928
12/18/79 0.751
1.090
1.776
0.367
0.996
Ib/dscf
(x 10"6)
0.068
0.048
0.062
0.089
0.067
0.145
0.132
0.069
0.093
o.llo
0.089
0.129
0.210
0.043
6.118
lb/hra
0.237
0.167
0.218
0.312
0.234
0.510
0.467
0.242
0.329
6.387
0.314
0.455
0.742
0.153
6.416
Ib/mm Btu
(x 10-3)
1.276
0.699
1.174
1.680
1.257
2.577
2.360
1.223
1.663
1.956
1.665
2.413
3.936
0.811
2.206
gm/ncm
(x 10'2)
0.108
0.076
0.099
0.142
(5.106
0.232
0.212
0.110
0.150
6.176
0.142
0.207
0.337
0.069
6.189
Xg/hra
0.108
0.076
0.099
0.141
6.106
0.231
0.212
0.110
0.149
6.176
0.142
0.206
0.336
0.069
0.188
kg/GJ
(X 10"")
5.511
3.906
5.502
7.196
5.529
11.064
10.154
5.268
7.137
8.406
7.138
10.355
16.890
3.468
9.463
Based on corresponding Method 5 run for volumetric  flow rate.
       TABLE 7.   SUMMARY OF FUEL ULTIMATE ANALYSES,  U.S.
                  SUGAR-BRYANT  MILL, DECEMBER 17-18,  1979
Bagasse conveyor
feed to boiler
Run #1
as rcvd
dry basis
Run #2
as rcvd
dry basis
Run #3
as rcvd
dry basis
Carbon
%
19.99
46.59
18.38
46.38
19.57
46.31
Hydrogen
%
8.87
5.90
9.00
5.77
8.90
5.88
Nitrogen
%
0.15
0.36
0.15
0.39
0.17
0.40
Sulfur
%
0.01
0.01
0.01
0.01
0.01
0.01
Ash
%
0.47
1.09
1.13
2.85
0.67
1.58
Oxygen
%
70.51
46.05
71.33
44.60
70.68
45.82
Fuel value
(Btu/lb)
3,406
7,939
3,210
8,101
3,480
8,233
Fuel oil  (No.  6)
  from Run t3
TJltimate analysis on fuel oil not performed.
                                                                18,524

-------
          TABLE  6.   SUMMARY OF NOx  EMISSIONS,  U.S. SUGAR-
                     BRYANT  MILL, DECEMBER 17-18, 1979

Run
number Date
1-1 12/17/79
1-2
1-3
1-4
Average
2-1 12/18/79
2-2
2-3
2-4
Average
3-1 12/18/79
3-2
3-3
3-4
Average

ppm
57.2
40.2
52.5
75.1
56.3
122.2
111.9
58.0
7.9
92.8
75.1
109.0
177.6
36.7
99.6
~
Ib/dscf
(x 10-«)
6.76
4.76
6.21
8.88
6765"
14.46
13.25
6.86
9.35
10.98
8.88
12.90
21.02
4.34
11.79
— 	 —
. 	 •
lb/hra
23.7
16.7
21.8
31.2
2374
51.0
46.7
24.2
32.9
3T7T
31.4
45.5
74.2
15.3
4T76"
.
Ib/mm Btu
0.1276
0.0899
0.1174
0.1680
0.1257
0.2577
0.2360
0.1223
0.1663
0.1956
0.1665
0.2413
0.3936
0.0811
0.220^
_
qm/ncm
0.1083
0.0762
0.0994
0.1423
0.1066
0.2317
0.2122
0.1099
0.1498
0.1759
0.1423
0.2065
0.3366
0.0695
0.1887
—
kg/hra
10.77
7.58
9.89
14.15
10.60
23.11
21.17
10.97
14.94
17.55
14.22
20.65
33.65
6.94
18.87
kg/GJ
(x IP"3)
55.11
39.06
55.02
71.96
55.29
110.64
101.54
52.68
71.37
84.06
71.38
103.55
168.90
34.68
94.63
3Based on corresponding Method 5 run  for volumetric flow rate.
          TABLE 7.   SUMMARY OF  FUEL ULTIMATE ANALYSES, U.S.
                     SUGAR-BRYANT MILL, DECEMBER  17-18, 1979
Bagasse conveyor Carbon
feed to boiler %
Run #1
as rcvd
dry basis
Run «2
as rcvd
dry basis
Run «3
as rcvd
dry basis

19.
46.

18,
46.

19.
46.

.99
.59

.38
.38

,57
.31
Hydrogen
%

e
5

9
5

e
5

.87
.90

.00
.77

.90
.66
Nitrogen
%

0
0

0
0

0
0

.15
.36

.15
.39

.17
.40
Sulfur
%

0.01
0.01

0.01
0.01

0.01
0.01
Ash
%

0.47
1.09

1.13
2.85

0.67
1.58
Oxygen
%

70
46

71
44

70
45

.51
.05

.33
.60

.66
.62
Fuel value
(Btu/lb)

3,406
7,939

3,210
8,101

3,480
8,233
  Fuel oil (No. 6)
   from Run #3
                                                                18,524
  Ultimate analysis on fuel oil not performed.

-------
     TABLE 8.   SUMMARY OF BOILER OPERATIONS  DURING TESTING,
               U.S.  SUGAR-BRYANT MILL,  DECEMBER 17-18,  1979


Date
Time start
Time end
Total min
Steam meter end
Steam meter start
Total, tons steam
Total steam used, Ib
Average steam loading, Ib/hr
Average steam temperature, °F
Average steam pressure, psig
Steam heat output, mm Btu/hr
Steam heat output, GJ/hr
Bagasse feed rate, ton/hr
No. 6 fuel oil feed rate, gal

Run 1
12/17/79
5:35 8:02
4:22 7:18
73 44
530 682
445 629
85 53
276,000
142,000
575.6
202.2
185.71
195.94
41.0
75

Run 2
12/18/79
1:22
11:25
117
948
801
147
294,000
151,000
580.6
202.2
197.87
208.77
41.0
0

Run 3
12/18/79
7:20
5:25
115
392
254
138
276,000
144,000
577.8
196.7
188.54
198.93
41.0
0

aRough estimate based on steam loading provided by EPA contractor
 observing emission test.

The bagasse feed rate was not measured in the test because no
technique for doing so could be devised.  There did not appear to
be any reasonably accurate method of determining the amount of
bagasse carried by the conveyor, dropped off into each feed chute,
or moved by the feeder.  The mill personnel have never measured or
estimated the bagasse feed rate directly and could not offer any
suggestions for doing it.

The mill does not have any instrumentation that monitors the
scrubber.  The only surveillance of the scrubber consists of visu-
ally checking the water collected in the bottom of the unit to
ensure that it is at the proper level.  During the test, the water
level was correct, and the scrubber appeared to operate properly.

The test boiler operated normally during all test runs.  Even with
the drastic drop in steam production in run #1, the average steam
load in that run was high enough to present normal operation.  Thus,
the emission samples taken should be representative of normal
boiler operation.

-------
                            SECTION 3

                       PROCESS DESCRIPTION
Bagasse is a waste product of the sugar cane extraction process
and has a heating value of about 3,600 Btu/lb  (wet).   The bagasse
represents about 30% of the weight of raw sugar cane.   In order
for the sugar cane mill to avoid a large solid waste  disposal pro-
blem and to minimize power cost requirements,  bagasse is used as
the primary fuel for on-site process steam production.   No.  6 fuel
oil is used to start the boilers and maintain  steam production
when the bagasse feed fails or become too wet; the Bryant Mill has
more bagasse than they can utilize and uses as little fuel oil as
possible.  Figure 1 shows the schematic process flow  diagram for
the Bryant Mill.
                                               FINAL MOLASSES
   RAW CANE
                                              STEAM
           BUNKER CFUaOIL(b)
           Figure 1.  Schematic process  flow diagram:
                      Bryant Mill,  U.S.  Sugar Corp.
                                8

-------
At their mill in Bryant, U.S. Sugar Corp. operates three bagasse
boilers producing process steam  (see Figure 2).  All three identi-
cal boilers are Riley Stoker Corp. vibrating-grate spreader-stoker
boilers, designed for a continuous steam production rate of 120,000
Ib/hr and a two-hour peak rate of 132,000 Ib/hr.  However, Boiler
#2 normally operates continuously (with a full bagasse feed rate)
in the range of 130,000 to 160,000 Ib steam/hr.  The overall
thermal efficiency  for a typical bagasse boiler unit is 55% (rang-
ing from 50% to 70%).  In comparison with large fossil-fired steam
generators this efficiency range is rather low; however, since
bagasse is a plentiful by-product fuel, thermal efficiencies are
of secondary importance.
                     STACKS
  o
MOUSSES TANKS  J
                                            SUGAR WAREHOUSES
      Figure 2.  Plot plan:  Bryant Mill, U.S.  Sugar  Corp.

Normal utilization of bagasse  in boiler #2  is  approximately 64,000
to £2,000 Ib/hr.  Bagasse  alone is burned about 90  percent of the
time; fuel oil is burned with  a low bagasse feed rate or alone
during a disruption of bagasse feed.

-------
One chain conveyor feeds bagasse to all boilers,  with a small
amount of bagasse  falling from each link  into  each feed chute.
There are three  feed chutes to the test boiler.   A rotary feeder
in each chute  regulates the bagasse feed  rate  through the chute.

All three identical boilers are followed  by  spray impingement
scrubber units,  similar to the diagram in Figure 3,  however,
Boiler #2 has  a  pair of smaller scrubbers in parallel rather than
a single scrubber.  The scrubbers on Boiler  #2 are Joy Manufac-
turing Co.  Turbulaire scrubbers, Size 40,  Type D, installed in
1973.  A by-pass duct is provided for emergency use, such as for
cleaning due  to  plugging of the scrubber.  The scrubbers are placed
as close as possible to the boiler air heater  containing the water
tubes because  a  single I.D. fan pulls exhaust  gases through the
scrubbers rather than pushing, and internal  drag forces are minim-
ized to save  energy-  The scrubbers are then ducted to an exhaust
stack which terminates 65 ft above ground level.  A diagrammatic
sketch of the  control equipment and site  is  given in Figure 4.
                               CAS OUTLET
                 WATER SPRAYS
                  CAS INLET


                  GAS INLET
                  PASSAGE
                 ANNULAR
               IMPINGEMENT GAP
                                           - El Ml NATOR SUMP
RISER DUCT


GAS INLET
PASSAGE

PERIPHERAL NOZZLE

LI QUID INLET
                                    SAMPLE DISCHARGE
                                      DRAIN
              Figure 3.  Spray  impingement scrubber,
                                  10

-------
                                    ^ SAMPLING
                                    * PLATfORM
                                               ROOF
          PIPE RACK
                                                      GROUND
                                                      FLOOR
 Figure 4.   Diagram of Bryant Plant emission control equipment.

The mill  operates seasonally,  from November  through March; normal
operating time is 24 hr/day,  7 day/wk.
                                  11

-------
                             SECTION 4

                    LOCATION OF SAMPLING POINTS
At the scrubber  inlet  on  Boiler #2 no suitable locations existed
which could meet minimum  downstream stack diameter criteria.   The
scrubber outlet  was  sampled at the installed stack sampling  platform

A diagram of the Boiler #2  exhaust stack above the roof on the
scrubber outlet  is shown  in Figure 5.  A rectangular stack 40  in. x
82 in. extends about 10.9 ft above the roof line, and is situated
between the stacks of  Boilers #1 and #3.  Eight 3 in. diameter
ports are installed  56 in.  above the platform floor.  The sampling
platform has a guard railing on which is bolt-mounted a 15 in.  rail-
ing extension placed so that a probe can be supported on it.   The
top of the stack is  27 in.  above the port centerline; the down-
stream equivalent diameter  criteria are narrowly met.  The nearest
upstream disturbance is a reduction 2.8 diameters away.
       REMOVABLE
      RAILING EXTENSION
2T'
             \   ..
            T
            15"
            39"
                 56"
                                   3" I
                     oooooooo
                                         PLATFORM
                                              -82"-
                               EOUIVALfNTDIAM. = 2 ^-^ =53.8
                      SIDE VIEW
                                             FRONT VIEW
    Figure 5.   Scrubber outlet/roof stack sampling location.
                                 12

-------
                            SECTION 5

               SAMPLING AND ANALYTICAL PROCEDURES


The U.S. Sugar Corp., Bryant Mill was sampled for particulate
matter, particle size, S02,  NO ,  integrated gas analysis,  and
fuel analysis.

The following describes the methods used.

SAMPLING PROCEDURES

Particulate

Sampling for particulates was performed using the method outlined
in the Federal Register,  Method 5,  "Determination of Particulate
Emissions from Stationary Sources," modified so that the sample box
temperature was maintained at 325°F instead of 250°F.

Particle Size

Sampling for particle size was performed using an Andersen cascade
impactor with seven stages and a back-up filter.  Because of the
presence of entrained water or highly saturated stack gases, it
was decided to utilize an impactor preseparator to protect the
impactor substrates and jet stages from the effects of water.  This
was thought superior to a heated impactor because heating may
change stage collection efficiencies.

The sampling train used consisted of the following equipment listed
in order of the flow:  a 10 mm diameter probe tip; a curved (90°)
probe tip to Andersen head connector; cyclonic preseparator; stand-
ard Andersen heads; a 4 ft stainless steel probe; a Smith-Greenburg
impinger with water, then one charged with color indicating silica
gel; and an EPA-5 console equipped with a dry gas meter, digital
electronic thermometer and an inclined manometer.  Also, an S-type
pitot tube was connected to the probe so the stack velocity could
be continually monitored.

A total of three particle sizing runs were made.  Each run was
conducted for 13 to 20 min under isokinetic conditions.
                                13

-------
At the completion of each run, the moisture collected was measured
and the Andersen heads were opened and oven-dried for three hours.
After drying, each stage was weighed,  then the filter was removed
and the stage assemblies were cleaned, desiccated and reweighed to
provide partial tare weights.  The tare weights of the filters
were taken during the assembly of the heads (after desiccation for
24 hr).

All weight measurements were made with a Mettler analytical
balance.  The balance was calibrated daily and rezeroed before
each weight determination.  Calculations were performed using the
methods and tables provided in the Andersen manual.

Sulfur Dioxide

Sampling for S02 was performed using the alternate method outlined
in Section 2.1 of the Federal Register, Method 6, "Determination
of Sulfur Dioxide Emissions from Stationary Sources."  S02 was
determined simultaneously with particulate matter by replacing the
Method 5 water impinger system with a Method 8 isopropanol-filter-
peroxide system.

Nitrogen Oxides

Sampling for NO  was performed using the method outlined in the
Federal Registe?, Method 7, "Determination of Nitrogen Oxide Emis-
sions  from Stationary Sources."

Integrated Gas Analysis

Exhaust gas analysis was performed using the method outlined in
the Federal Register, Method 3, "Gas Analysis for Carbon Dioxide,
Oxygen, Excess Air, and Dry Molecular Weight."

Fuel

Three  samples of bagasse were grabbed from the boiler feed conveyor
during each Method 5 run, plus one sample of fuel oil during Run #1.

ANALYTICAL PROCEDURES

Particulate, S02, NO , Gas Analysis

All analytical procedures were performed using the methods described
in EPA Methods 3, 5, 6, and 7, previously mentioned in the Sampling
Procedures Section.

Fuel

Analysis of the bagasse feed was performed using ASTM D  3178  for
carbon and hydrogen, ASTM D 3176 for  oxygen, ASTM D 3179  for
                                 14

-------
nitrogen, ASTM D 3177 for sulfur, and ASTM D 3174 for ash.  Fuel
value was determined using ASTM D 2015.
Quality Assurance/Quality Control
Results of quality control tests are furnished with the analytical
data sheets provided in Appendix C.
                               15

-------
         APPENDICES
A.  Complete Emission Results
B.  Field Data Sheets
C.  Analytical Data Sheets
D.  Boiler Operating Data
E.  Project Participants
              16

-------