ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
EPA-330/2-75-005
Boiler Stack Emission Monitoring
Kekaha Sugar Company
Kekaha, Kauai, Hawaii
NATIONAL ENFORCEMENT INVESTIGATIONS CENTER
DENVER.COLORADO
1
JULY 1975
I®;
-------
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
BOILER STACK EMISSION MONITORING
KEKAHA SUGAR COMPANY
KEKAHA, KAUAI, HAWAII
NATIONAL ENFORCEMENT INVESTIGATIONS CENTER
DENVER, COLORADO
JULY 1975
-------
CONTENTS
INTRODUCTION 1
SUMMARY AND CONCLUSIONS 2
POWER BOILER OPERATIONS AND EMISSION CONTROL EQUIPMENT 3
Power Boi 1 ers 3
Air Pollution Control Equipment 4
DISCUSSION OF STUDY 5
Stack Monitoring 5
Visible Emission Evaluation 10
REFERENCES - 11
APPENDICES
A Processing of Sugar Cane
B Stack Monitoring Procedures
C Analytical Procedures for Stack Monitoring Samples
D Boiler Operating Data, Kekaha Sugar Company
E Visible Emissions Evaluations
Kekaha Sugar Company, May 28-29, 1975
ii
-------
TABLES
I Summary of Stack Monitoring Data 8
Kekaha Sugar Company, May 28-29, 1975
II Summary of Boiler Operating Data 9
Kekaha Sugar Company, May 28-29, 1975
FIGURE
1 Kekaha Boiler Stack 6
iii
-------
INTRODUCTION
The Kekaha Sugar Company Mill has a rated capacity of 2,270 kkg*
(2,500 tons)/day net cane** and generally operates 24 hours per day
6 days/week.
The processing of cane to raw sugar and molasses generally follows
the flow chart contained in Appendix A. As is typical at most sugar
mills in the Hawaiian Islands, the bagasse*** is used as the primary
fuel for the boilers. The bagasse available for fuel is generally about
30 percent of the net cane processed.
At the request of Region IX, EPA, the National Enforcement In-
vestigation Center (NEIC) conducted stack monitoring tests and visible
emissions observations on May 28 and 29 to determine if applicable
state air pollution regulations were being met. The allowable par-
ticulate emission rate is 0.4 kg/100 kg (0.4 lb/100 lb of bagasse
burned,. For visible emissions from existing sources, the regulations
state
"No person shall cause or permit the emission of visible air
pollutants of a shade or density equal to or darker than that
designated as No. 2 on the Ringelmann Chart or 40 percent opa-
city, except" that "A person may discharge into the atmosphere
from any single source of emission, for a period or periods ag-
gregating not more than 3 minutes in any 60 minutes, air pol-
lutants of a shade or density not darker than No. 3 on the
Ringelmann Chart or 60 percent opacity when building a new fire
or when breakdown of equipment occurs.'-/
Information on process and boiler operations and air pollution con-
trol practices were obtained from Mr. John Robinson, Factory Manager,
and Mr. Robert Inouye, Plant Engineer. Mr. Harold Younguist, Pollution
Control Engineer of the Hawaii Department of Health observed the tests.
* Metric tons.
** Net cane is a calculated value and is essentially equal to the gross
cane brought into the plant minus the soil, leaves, rock and other
trash.
*** Bagasse is the solid material remaining after the milling process
has romoved the juice from the sugar cane. It serves as an excel-
lent fuel source with a heating value of 2,580 gram-cal/gm
(4,650 BTU/lb).
-------
-2-
SUMMARY AND CONCLUSIONS
1. Stack monitoring was conducted May 28 and 29 to determine par-
ticulate emissions. The average particulate emission was 0.22 kg/100 kg
(0.22 lb/100 lb) of bagasse burned which was substantially less than
that allowed i.e. 0.4 kg/100 kg (0.4 lb/100 lb) of bagasse burned.
2. Visible emission observations were conducted as specified in
Method 9 of the Standards of Performance for New Stationary Sources.
Forty observations, at 15 second intervals were made during each stack
particulate emission test. At the time of the observations the plant
was operating normally. The average plume opacities for the three
observation periods ranged from 23-30 percent. These results show
the stack plume meets the applicable state regulations which require
that visible emissions from existing sources not equal or exceed 40
percent opacity.
3. Air pollution control equipment at this mill consists of a
multiple cyclone unit (Western Precipitation Corp. Multiclone Unit),
having a rated capacity of 4,200 actual m3/min (148,000 acfm) and
which is designed to operate at 90 percent efficiency. Stack gas
volumes were about 65-75 percent of the rated capacity. The results
indicate this unit was operating effectively.
-------
-3-
POWER BOILF.R OPERATIONS AND EMISSION CONTROL EQUIPMENT
Both steam and electricity are used by the Kekaha Sugar Company to
supply power for cane processing at the mill. High pressure steam pro-
duced in a bagasse-fueled boiler is used to generate electric power and
to drive the majority of the mechanical equipment in the mill. Low
pressure steam exhausted from the electric turbines and mill equip-
ment is used in the sugar juice evaporation and concentration processes.
The total electrical power requirement of the mill is about 3,500 kw.
Approximately 1,500 kw is used for the irrigation pumps which supply
water to the plantation fields and 2,000 kw is used for in-house mill re-
quirements. The majority of the electric power is supplied by three
steam powered turbine generator units having total capacity of 6,500 kw.
One 2,500 kw condensing or extracting (1 kg/cm2 exhaust turbine is op-
erated at full capacity. The second 2,500 kg turbine is a full condens-
ing unit which is normally operated at approximately 500 kw. The third
unit is a 1,500 kw condensing or extracting (1 kg/cm2 exhaust) unit which
is a stand-by unit operated only when the mill needs to burn excess bag-
asse or when it requires additional 1 kg/cm2 (15 psi) extracted steam.
Supplemental electrical power is obtained from two company-owned
hydro-generation facilities located in the Waimea Canyon above the mill.
The smaller unit, located approximately 5 km (3 miles) away from the
mill, is rated at 500 kw although it is normally operated at 275 kw.
The larger unit, located approximately 11 kw (7 miles) up the Canyon
from the plant, has a rated capacity of 1,000 kw and is normally op-
erated at 750 kw.
The electrical system is tied into the Kauai Electric Company's
distribution system for the Island. The mill can purchase additional
electrical power from this system on an as-needed basis. Likewise,
the mill can supply power to the Kauai Electric system during peak
demand periods. During 1974, the mill purchased 700,000 kw hrs of
power and supplied 10,400,000 kw hours to the system.
POWER BOILERS
Two small boiler units and the main boiler are available at the
mill for steam production. The two small (Combustion-Engineering Inc.)
steam boiler units are available for standby steam production during
periods when the main power boiler is not operational. One unit,
rated at 800 Hp and 23 kg/cm2 (325 psi), can be fueled with both
bagasse and/or Bunker C fuel oil. The other unit, rated at 600 Hp
and 22 kg/cm2 (317 psi), is fueled solely with fuel oil. Neither of
these units has been operated in the past two years.
-------
-4-
The main power boiler at this mill is a Combustion Engineering,
Inc. Model V.U. 40S steam generator. It was installed new in 1954 and
was one of the first boilers in the Hawaiian Islands designed for
bagasse as a primary fuel, with Bunker C fuel oil as a supplemental
fuel. This boiler is a dual drum, water tube and water wall unit de-
signed for 86,000 kg (190,000 lb)/hour of steam at a maximum delivery
pressure of 33 kg/cm2 (475 psi). With the current mill operating modes
the boiler is normally operated at 29-29.5 kg/cm2 (410-420 psi).
Bagasse from the last pressing mill of the sugar cane processing
train [Appendix A, Figure A-1] is transported via conveyor belt into
four feed chutes at the power boiler - keeping these chutes filled at
all times. Any excess bagasse is diverted to a bagasse storage area.
During the periods when bagasse production from milling processes is
not sufficient to supply the boiler fuel demand, bagasse is withdrawn
from storage and feeds the four chutes.
The bagasse enters the furnace via four spreader stokers, one lo-
cated at the bottom of each bagasse chute. The stokers propel the
bagasse into the furnace by rotating flipper arms. Approximately 50
percent of the bagasse burns in suspension, the remaining 50 percent
burns on the furnace grate. The bagasse feed rate into the furnace is
determined by the rotating flipper arm speed. Conversely, the speed of
the arms is controlled by steam demand; i.e., as the steam demand in-
creases, the amount of bagasse fed into the boiler increases.
Ashes are removed from the furnace by a traveling grate system.
The grate moves from back to front below the furnace and dumps ashes
into a hopper located in front. The ashes are then sluiced by water
into the plant wastewater channel.
A manually operated atomized oil burner is located at each corner
of the furnace. The majority of the fuel oil is used at the beginning
and end of the processing season. Fuel oil is also used whenever the
steam demand exceeds that which is produced by burning bagasse only.
These situations occur when the bagasse is extremely dirty or when the
bagasse feed systems are affected by mechanical problems.
AIR POLLUTION CONTROL EQUIPMENT
Particulate emissions from the boiler originate from unburned
bagasse and ashes which are entrained in the combustion gases. Com-
bustion air is supplied to the furnace by a forced draft (FD) fan which
draws in ambient air and forces it through an air preheater unit. The
FD fan is rated at 1,960 m3/min (69,400 CFM) at 100°F. From the air
preheater unit the warm air is introduced into the furnace beneath the
ash grate. This air supports the combustion of the bagasse both on the
grate and in suspension. The combustion gases are then drawn by an in-
duced draft (ID) fan [rated capacity of 4,200 m3/min (149,000 CFM) at
-------
-5-
380°F] up through the steam superheater tubes, the boiler tubes, and
exit at the upper rear portion of the boiler. These exhaust gases
pass through a cyclone dust collector, the air preheater unit, the ID
fan and are exhausted to the atmosphere through a 3 m (10 ft) diameter
24.3 m (80 ft) high steel stack.
The cyclone type dust collector is a Western Precipitation Corp.
multiclone unit, Type 24 V.D.A., Model P-47930A, size 36-4. The unit
is rated at 4,200 actual nvfym'in (148,600 ACFM) of exhaust gas and has
a design collection efficiency of 90%. Ashes collected in this multi-
clone unit fall into four hoppers. Motorized rotary valves located at
the bottom of these hoppers continually discharge the collected ash
into a convion open tank. From here the ashes are sluiced to the waste-
water discharge channel. The rotary valves have mechanical seals which
form airlocks to maintain the vacuum in the multiclone hoppers.
The company has installed a Bailey smoke density meter in the exhaust
gas ductwork between the dust collector and the air preheater unit.
This unit transmits a signal to a recorder located in the boiler
control room. Smoke opacity is continuously recorded here in Ringelmann
units.
DISCUSSION OF STUDY
STACK MONITORING
The required three stack tests were conducted on May 28 and 29,
1975. The stack monitoring procedures used including equipment, its
calibration, testing, and sample handling are discussed in Appendix B.
The boiler stack at this mill is 3.05 m (10 ft) in diameter. The
two sample ports are located about 12.2 m (40 ft) above the exahust
gases inlet and approximately 7.6 m (25 ft) from the stack top [Figure 1].
A 305 m (10 ft) probe was used. Sampling was conducted in accordance
with Federal Regulations which dictate the number of sampling points
necessary based on port location (Method 1)1/. Since the ports were
located 4 diameters downstream from a disturbance, i.e., the exhaust
inlet, and about 2.5 diameters upstream from a disturbance, i.e., the
stack exit, 36 sampling points, 18 on each diameter were required.
During a test each point was sampled three minutes in order that
1.7-2.1 (60-75 ft^) of air could be sampled. Due to the light par-
ticulate load, only one filter was required per test. Isokineticity
ranged from 102-108 percent*.
* Isokinetic sampling can be defined as sampling in a manner so that
the sampling velocity (Vn) is equal to the stack velocity (Vs). If
Vn < Vs i.e. percent less than 100, sampling is under isokinetic; if
Vn > Vs sampling is over isokinetic. For a valid test sampling must
be in the range 90 to 110 percent.
-------
SAAiPLING PORTS
GAS FLOW
FROM t D FAN
^ oo
-------
-7-
2/
Samples were recovered as specified for Method 5- [Appendix B].
The samples were shipped to the NEIC laboratory in Denver for analyses
[Appendix C]. The NEIC Chain of Custody procedures were followed.
During the stack monitoring tests NEIC personnel observed boiler
operating procedures in the control room. Pertinent operating readings
were obtained every ten minutes from the boiler instrumentation panels.
These data included the steam production rate, bagasse production rate,
superheater steam pressure and temperature, feed water pressure and
temperature, and feed water flow rate [Appendix D]. The average steam
production rates during monitoring ranged from about 77-78.5 percent of
the boiler capacity of 86,260 kg (190,000 lb)/hr.
The steam production rates observed during any one monitoring test
differed by as much as 9,100 kg (20,000 lb)/hr. The changes generally
resulted when the cane line stopped and bagasse had to be taken from
storage to meet the fuel demand. The slight delay was reflected in a
steam production drop,
The Company does not actually measure the amount of bagasse fed
to the boiler unit. Over the years, through interpretation of empirical
data, Company personnel have determined that approximately 2.3 kg (2.3
lbs) of steam are produced in the boiler for each kilogram (pound) of
bagasse burned.
NEIC personnel requested that the company certify certain produc-
tion and boiler operating parameters during the test periods, including
bagasse usage rate [Appendix D]. The particulate emission i.e. kg/100
kg (lb/100 lbs) of bagasse were calculated using the bagasse figures
provided [Table II].
To ascertain the particulate emission, the amount of material
collected on the filter and in the cyclone along with any material con-
tained in the sampling probe and nozzle, i.e., the "front half" of
the sampling train were measured [Table I]. The results show that the
average particulate emission was 0.22 kg/100 kg (0.22 lb/100 lb) of
the bagasse burned which was substantially less than that allowed i.e.
0.4 kg/100 kg (0.4 lb/100 lb) of bagasse burned.
The "back-half" collection represents the material collected in
the impinger case and is that material existing in a gaseous state at
120°C (248°F) but condensing at lower temperatures. This material is
not presently considered in determining particulate emissions. It is
presented herein for information purposes only [Table I].
The complete field data and analytical results are on file at NEIC
and are available upon request.
-------
-8-
Table I
Summary of Stack Monitoring Data
Kekaha Sugar Company
May 28-29, 1975
Volume Metered
Average Stack
Temperature °R
Molecular weight
Percent Moisture
Average Stack gas
Velocity
m/sec.
ft/sec.
Stack gas volume
m3/min.
(ftvmin.)
Percent Isokinetic
Particulate Collec-
tion -J
Front half-gm
Back half-gm
Total-gm
Bagasse Fired
kg/hr
(lb/hr)
Emission Rate
Front half collec-
tion.
kg/100 kg bagasse
(lb/100 lb bagasse)
Total Collection
kg/100 kg bagasse
(lb/100 lb bagasse)
Test 1
1791
(63.25)
805
27.2
24.9
7.32
(24.03)
3207
(113,240)
104
1.33
0.13
1.46
28,400
(62,800)
0.25
(0.25)
0.27
(0.27)
Test 2
1696
(59.88)
804
27.7
20.6
6.70
(21.97)
2932
(103,530)
102
1.03
0.17
1.20
29,000
(64,000)
0.19
(0.19)
0.22
(0.22)
Test 3
1564
(55.24)
812
26.9
25.4
6.26
(20.54)
2741
(96,790)
108
1.27
0.12
1.39
29,200
(64,600)
0.22
(0.22)
0.24
(0.24)
1/
2/
SCF - Standard Cubic Feet
Front half collection - Particulates contained on the filter, in the
cyclone, and in the sampling probe and nozzle.
Back half collection - The material contained in the Impinger case
which is in a gaseous form at 120°C (248°F) but condenses at lower
temperatures.
The back half material is not at present considered as part of the stack
emission for new stationary sources.
-------
TABLE II
SUMMARY of boiler operating data
KEKAHA SUGAR COMPANY
MAY 28-29, 1975
Test
No.
Date and
Time
Steam
Production
kq (lb)/hr
Steam
Production
kq (lb)/hr
Bagasse
Fuel Used
Metric Tons (Tons)/hr.
Moisture
Content %
kg(lb) Steam
/kq(lb) baqasse
1
May 28, 1975
66,700
65,800
28.4
46.5
2.30
(1500 to 1710)
(147,000)
(145,000)
(31.4)
2
May 29, 1975
67,100
67,100
29.0
46.5
2.31
(0810 to 1110)
(148,000)
(148,000)
(32.0)
3
May 29, 1975
67,600
67,600
29.2
46.5
2.29
(400 to 1610)
(149,000)
(149,000)
(32.3)
I
vo
I
-------
-10-
VISIBLE EMISSION EVALUATION
Visual determinations of the boiler stack plume opacity were con-
ducted as specified in the Federal Regulations (Method 1).— Forty ob-
servations at 15 second intervals were made for a ten minute period dur-
ing each particulate sampling test. At the observation times, the plant
was operating normally. The average opacities during tests 1, 2, and 3
were respectively 27, 30, and 23 percent. The range of opacity readings
were 15-45 percent, 20-45 percent, and 15-35 percent [Appendix E], The
average opacities as well as most individual operations were less than
40% opacity.
-------
REFERENCES
Public Health Regulations, Air Pollution Control, Department of
Health, State of Hawaii, Chapter 43, March 21, 1972.
Standards of Performance for New Stationary Sources, Environmental
Protection Agency, Federal Register, Vol. 36, No. 247, Part II,
Appendix Test Methods, December 23, 1971.
-------
APPENDIX A
PROCESSING OF SUGAR.CANEl/
The processing of sugar cane to raw sugar and molasses generally
follows the steps discussed below [Figure Al].
Cane Cleaning
The replacement of hand cutting by mechanical harvesting has resulted
in an increase of mud and dirt content. This has necessitated cane
washing and it is the practice normally followed at mills in the Hawaiian
Islands. The cane is usually washed with warm barometric condenser waters.
Mi 11i nq
Following cleaning, the cane proceeds into the milling process of
which the purpose is to extract the juice from the cane stock. This
extraction is accomplished with revolving cane knives, shredders,
crushers, and roll mills. The knives cut the cane into chip's in preparation
for grinding and to provide a more even feed rate to the mills. Shredders
further prepare the cane. These two operations increase mill capacity.
The crushers extract 40-70% of the juice. After the juice is extracted,
the remaining material (bagasse) representing about 30% by weight of the
cane entering the system, is usually used as feed for the boiler system
with the excess being hauled to landfill or, in some cases, discharged
in the wastewater.
Clarification
The juice from the milling operation contains impurities such as
fine particles of bagasses, guns, and waxes. Screening will remove the
courser particles which are returned to the mills. The majority of the
T7 Development Document for Interim Final Effluent Limitations Guidelines
and Proposed New Source Performance Standards for the Raw Cane Sugar
Processing Segment of the Sugar Processing Point Source Category,
Effluent Guidelines Division, Office of Water and Hazardous Material,
U. S. Environmental Protection Agency, Washington, D.C. 20460, pp 32-46.
-------
-2-
remaining impurities are removed by clarification.
In the clarification process lime, heat and a small amount of phos-
phate are used to aid in removing the remaining impurities. During the
process a flocculent precipitate is formed and most of the suspended
solids remaining in the juice are occluded and settle out with the
precipitate. The precipitate is separated from the juice by settling
and decantation in continuous clarifiers.
Filtration
The clarification process separates the juice into two portions:
(1) the clarified juice and (2) the precipitated sludge or muds. The
first portion represents 80-90% of the juice which is usually taken directly
to the evaporator system. The second portion is generally thickened by
rotary vacuum filters. The filtered juice is cycled back through the
clarification system.
The filter cake produced (20-75 kg/kkg or 40-150 lbs/ton of cane
ground) has a moisture content of 70-80 percent. It is general practice
to collect the cake in storage bins for subsequent removal to the cane fields.
In most cases, the filter cake is not discharged.
Evaporation
The clarified juice is about 85 percent water and 15 percent solids.
To obtain crystallization, enough water must be removed to produce a
60 percent solids syrup. This concentration process is usually accomplished
using multiple-effect evaporators in the interest of better fuel economy.
-------
-3-
Crystallization
The concentrated syrup from the evaporators is placed in single effect,
batch type evaporators called "vacuum pans". The pans are operated
using exhaust steam or vapor from the first stage evaporators discussed
above. Seed crystals are introduced into the vacuum pan at the beginning
of the operation. The pan must be maintained in a narrow range of sugar
concentration and temperature so that the seed crystals grow.
After sugar crystals form in the pan", the mixture of crystals and
syrup (massecuite) is agitated gently in a mixer. After mixing, the
crystals are separated from the syrup in a high speed centrifuge with
the raw sugar going to storage.
The vacuum pans generally operate in series with each pan
crystallizing a different grade of massaceuite. The last pan yields
a low sugar and final (blackstrap) molasses which is usually used for
animal feeds. The low grade sugar off the final pan is subsequently
melted into syrup and mixed with the syrup from the evaporators.
-------
Boiler Feod Water
c i;: i wu. f
s -Jl.'.M I
Y
Dilution of Molasses
Imbibition Washing
D iscnorgo
Electricity
7"
Steam
-w—c
/t-~;\
IT
[ CU'QSSC
7"
Condenser
Water From
Uarometr Ic
Leys
Turbogenerators
Steam Turbines
Cane
Leveler
-a» Bearing Ccoling
Water--Ceo led
ana Recycled or
D ischarccd
Mechanical Mill Drive
Feed Tab^
Cane j'- Wash
U U U
/l\ /l\ /l\
—Carrier
Imbibition Water
Cane Wash Water
to Discharge
or Recycle
Juice To
Clarification
(Sheet 2)
TYPICAL SUGAR FACTORY WITH'CANE WASH
FIGURE A-l
Sheet 1 of 3
-------
TYPICAL EVAPORATION ¦SYSTEM
FIGURE A-l '
Sheet 2 cf 3
-------
SY1 :RC
CVAPORA'I ION
(Sheet 2)
SYKUP
TANKS
A
Water'
•x <->
•i j
cu
r.
o
s-
ra
CO
1 I
Condenser Water to Cane Wash, Other Uies, or IMschnnje
f\y
Steam
¦,3— Seed Sugar
./
Mixer
"A" Molasses
MOLASSES
TANKS
Water.
I i
Steam
Seed Sugar
<
Mixer
^— "B" Molasses
Centrifuge
Condensate to
Condensate Tank
("Sheet 1)
r
Commercial
Raw Sugar
C
VACUUM
PAN
/
0 Q" 0 Crystallizers
Centrifuge
Mixers
l Centrifuge
:j. Final MoT ass
Seed Sugar
Sheet 3 of 3
-------
TM
APPENDIX B
STACK MONITORING PROCEDURES
Equipment
Particulate samples were collected using a Scientific Glass
sampling train equipped and operated as required by "Method 5 - Determi-
nation of Particulate Emission from Stationary Sources"!/. The method
requires that the stack gas be sampled between 90 and 110 percent
isokinetically; i.e., the kinetic energy of the stack gas must equal the
kinetic energy of the gas entering the sample nozzle within + 10 percent.
Since the kinetic energy of a gas is dependent only upon the mass and
velocity of the gas stream and the mass of the gas entering the nozzle is
equal to that in the stack at any given time, isokineticity can be satis-
fied by adjusting the gas entry velocity into the nozzle to that of the
stack velocity. The two velocities are related by an equation using
pressure drop across a S-type pitot tube (stack velocity) and a calibrated
orifice (nozzle velocity) located just prior to where the sampled gas exits
from the train. Rapid adjustments to the isokinetic sampling rate are made
possible by the EPA - Method 5 sampling train shown below and a aomograph
which rapidly solves the pressure drop equation.
Inl irluCn I r Hui
HfATEO ANEA HLHRMOLDfR
IMrMMtRb ICE OATH y
uy r.v.s vaive
Tv
THEMWETLra'
VACUUM
GAUGE
^ MAIN OaLVS
DRY TEST MEI EH AIR-TIGHT
puwi*
1/ Standards of Performance for New Stationary Sources, Environmental
Protection Agency, Federal Register, Vol. 36, No. 247, Part II,
Appendix-Test Methods, Method 5, December 23, 1971.
-------
-2-
The probe is moved to the location to be sampled and the velocity
pressure in the stack (pitot tube pressure differential, aP) is read
on an inclined monometer. Using the nomograph, the operator solves
the equation and manually adjusts the sampling rate to obtain the desired
pressure drop across the calibrated discharge orifice.
The stack gases are pulled through the nozzle and probe to a cyclone
which collects large particulates and then through a glass fiber filter
which collects the remaining particulates. The train to this point is
heated to 120°C (248°F) to eliminate condensation.
The gases then pass through four impinger tubes which are partially
imnersed in ice water to maintain temperatures around 21°C (70°F).
The first two impinger tubes are prefilled with 100 ml of distilled water.
The third is empty and 200 gms of silica gel is added to the fourth. A
Greenburg-Smith impinger is used in the second tube; modified (3.8 cm
(1.5 inch) ID opening). Greenburg-Smith impingers are used in the first,
third and fourth tubes. The first,three tubes collect water vapor and
other condensable material. The air is dried by the silica gel in the
fourth tube. As shown in the preceding figure, the heated box and impinger
case are at the stack.
The clean, dry gases then flow through the umbilical cord to the
leakless vacuum pump located at the control console. The pump operates
at a constant rate; valves adjust the amount of pump discharge air
recycled to the inlet of the pump and the amount drawn through the
nozzle to the pump. Since the pump is leakless, the air which is pulled
through the nozzle is equal to the amount which discharges from the
-------
-3-
pump under positive pressure to the 4.96 cmh (175 cfh) Rockwell dry
gas meter. Air discharges from the meter through the calibrated orifice
to atmosphere.
During the test, temperatures were measured at six points throughout
the train with chrome!-alumel thermo-couples and displayed on a digital
temperature indicator (DTI) at the control console. The locations
monitored for temperature are the stack gas, the probe, the filter oven,
the impinger exit, the inlet to the gas meter and the exit from the gas
meter.
Molecular weight of the stack gas was determined using an Orsat
analyzer to analyze a composite sample taken with a hand pump during the
test period. The Orsat analyzer contains three reagents which selectively
absorb CO2, O2 and CO. The molecular weight was then calculated by
assuming that the remaining gas was nitrogen.
Static pressure of the stack was determinedby facing the openings
of the S-type pi tot tube parallel to the gas stream. The pressure differ-
ential (compared to atmosphere) was read on the inclined monometer at the
control console. Barometric pressure was read from a Lloyds barometer.
The percent moisture of the stack gas was estimated for the first test.
The calculated moisture content from each test was used for the following
test.
Calibration
The DTI was calibrated on site with a Mini Mite ™ pyrometer.
Nozzle diameter was measured with calipers to an accuracy of + 0.025 mm
(+ 0.001 inch).
-------
-4-
Napp Inc. manufactured the S-type pi tot tube used for stack velocity
pressure measurement. A record of the Napp Inc. calibration of the
pitot tube is on file at NEIC. The gas meter was calibrated by the
Public Service Company of Colorado and verified by NEIC personnel at the
TM
calibration laboratory against an American wet test meter to be accurate
within + ]%. The meter was again verified to be accurate by the Gasco
Company in Honolulu, Hawaii, just prior to use. The discharge orifice
was also calibrated against the wet test meter at the NEIC calibration
laboratory and was found to have a reference pressure drop of 1.72 at
a flow rate of 0.021 S cum/min (0.75 scfm).
Test Procedures
The Kekaha stack is 3.05 m (10 ft) in diameter. The sample ports
are at 90° to each other and located approximately 4 diameters downstream
from the stack gases inlet and two diameters down from the top of the stack,
thus 36 sampling points, 18 on each diameter are required by Method 1-1A
The NEIC team used a 3.05 m (10 ft) probe. The points were located at
the centroids of thirty six equal area portions of the total stack
cross section.
The static pressure was determined and a velocity traverse was
made. These data and estimates of percent moisture and molecular weight
were required to select a nozzle diameter. After the nozzle was selected,
the nomograph was set arid a leak check of the sampling train was performed.
EPA Method 5 allows no more than 566 cc (0.02 cubic feet) leakage per
minute. After satisfying the leakage requirements, the probe was inserted
to the first sampling point and the test begun.
\J Op. Cit. Method 1.
-------
-5-
The stack gas was sampled for three minutes at each point for a total
of 108 minutes sampling per test. This amount of time was adequate to
obtain the desired 1.7 - 2.1 (60 - 75 ft^) of air. The 3 minute
sampling time per point was a deviation from that prescribed for Method 5
in the December 23, 1971 Federal Register which calls for a minimum of
5 minutes per sampling point. However, revisions to Method 5 are being
promulgated changing the minimum sampling time to two minutes. Current
practice is to sample a sufficient time (but not less than two minutes
at each point) to obtain the aforementioned air volume. Sampling for 5
minutes at a large number of points yields a larger than desirable air
volume.
Sample Recovery
At the end of the sampling period, the sample was recovered as
specified in Method 5^. The probe was removed from the stack and
disconnected from the train. All open connections were capped immediately
to prevent gaining or losing particulate. The probe and nozzle were
rinsed with acetone and brushed. The rinse was collected in a clean glass
jar for later analysis.
The glass-ware portion of the train was taken to the plant chemistry
laboratory for cleanup. The filter v/as placed in a glass petri-dish
and sealed with aluminum foil. All glass-ware between the filter and
the probe was rinsed with acetone. All of the acetone rinse from this
portion of the train was placed in the same jar as the probe rinse.
T7 Ibid
-------
-6-
Water from the impingers was measured and the silica gel weighed
to determine percent moisture in the stack gas. The water was placed in
a jar for subsequent ether-chloroform extraction. All back-half glass-ware
was rinsed with acetone which was collected in a third jar. A sample of
the distilled water and acetone used was also taken to determine what
weight was added to the samples from the clean-up liquids.
All samples were shipped to the NEIC laboratory for analysis.
Samples were handled under standard NEIC Chain of Custody procedures.
-------
APPENDIX C
ANALYTICAL PROCEDURES FOR
STACK MONITORING SAMPLES
Filters
Gel mans glass filters were preweighed directly on the pan of a Mettler
analytical balance after 24 hours desiccation. They were then placed in
petri dishes prewashed and acetone rinsed. Finally, the petri dishes
were completely wrapped in foil to exclude dust.
Upon receiving the filters back in the lab they were placed in pre-
weighed 100 ml beakers. The aluminum foil used to seal the filters
was found to be free of particulates collected during the sampling phase
and, therefore, discarded. One unused filter was used as a control.
Beakers were weighed until constant weight was achieved. As the material
on the filters rapidly picked up weight upon removal from the desiccator
only about three filters could be weighed at a time.
Chloroform - Ethyl Ether Extracts
The materials used here were a 25 ml graduate, 2000 ml separatory funnels,
pre-weighed 100 ml beakers, two square yards of chiffon nylon cloth, and
Burdick and Jackson brand solvents.
Each sample was serially extracted with three 25 ml portions of CCI3 then
three 25 portions of ethyl ether. The combined extract, in a 100 ml beaker,
was placed in a hood and the air intake covered with a nylon chiffon
cloth to prevent dust and particulates from entering the sample. After
air drying, the sample was desiccated and weighed until constant weight
was achieved.
Water Portions
The water portions were measured before extraction in a large graduate
cylinder. After extraction they were taken down in 800 ml beakers and
transferred to 100 ml pre-weighed beakers and taken to dryness. The take
down was done on a concentric ring type water bath at approximately 95°C.
Measurements were recorded when constant weight was attained.
Acetone Rinses
These were first taken down in the quart sample containers inside of a
hood, the intake of which was covered with nylon cloth to restrict dust
from entering. When the samples were below 100 ml, they were transferred
to pre-weighed 100 ml beakers and taken down to dryness. The entire take
down procedure was accomplished at around 70°F and a blank was run along
with the samples. After no acetone smell could be detected, the samples
were placed in a desiccator and weighed 24 hours later. In some cases,
constant weight was not achieved until several days later.
Richard C. Ross
-------
Hawaii Air Sampling - Summary of Results***
Fi1ters
Sample No. Particulate Wt. (g)
4201-052§X#2) .3326
4201-0526(#2) .3968 ***4201-Honokaa
4202-0528(#l) .4698 Sugar Mill
4201-0525(#1) .3692 4202-Kekaha
4201-0525(^1) .3166 Sugar Mill
4201-0526(#3) .2811
4201-0526(^3) .3580
4202-0529(#2) .3444
4202-0529(#2) .4265
Blank -.0002
CClq-Ethyl Ether Extract Residues Residue Weight (g) Original Water Volume
4202-0529(#2) .0081 505 ml
4202-0529(#1) .0194 620
4202-0529(#3) .0043 560
4202-0528(^1) .0006 265
4201-0526(^2) (Blank) -.0006 275
4202-0529(#3) (Blank) -.0007 435
4201 -0525(//I) .0670 705
4201-0526(#2) .0575 865
4201-0526(#3) .0553 775
Water Residues - Post Extraction Weight (g)
4202-0529(#2T .1407
4202-0529(^1) .0446
4202-0529(^3) .1032
4202-0528(# 1) .0375
4201-0526(#2) (Blank) .0039
4202-0529(^3) (Blank) -.0003
4201-0525(^1) .4907
4201-0526(#2) .4056
4201-0526(#3) .4019
Acetone Residues Weight After 24 hrs (g) Constant Weight (g)
4201 -0526 (?2jO> 2.2073 " 1.6843
4201-0526(#2)B.H.** .0320 .0320
4202-0529(^3) Blank .0026 ..0026
4201-0525(^1) Blank ..0055 .0055
4201-0525(#1)F.H. 1.8652 1.8528
4201-0525(^1)B. H. .0242 .0242
4202-0529(#3)F.H. .8474 .8434
4202-0529(#3)B.H- .0129 .0129
4202-0529(#2)F.H. .6884 .6866
4202-0529(#2)B.H. .0285 .0285
4202-0528(#l)F.H. .8641 .8614
4202-0528(irl )B.H. ..0383 .0383
4201-0526(#3)F.H. 2.5003 1.5995
4201-0526(//3)B.H. ,.0522 .0522
*F.H. Front Half Collection **Back Half Collection
-------
APPENDIX D
BOILER OPERATING DATA
KEKAHA SUGAR COMPANY
-------
J/, /,., / . C
Jtt^o c
s^rLJc .- r/.„ A s"
»2SL _jfo/£v L. IW. ^,y_ iW re,.p 4-"^ T""P °X
>— - P-" —
500 /SC0CO 75^> ^C- 1^600 5CO /&% "f/t> 1&$~
(^ZS~)
f52.c'>o 75Z. 4»o it%coc> t?Co /£>%¦>
'62o ]^)LCO ?5Z t'df-ltcfr <&£ 2lo <$% 4^ dco 7±) ~Z_- /*£ <4-ciS 2Jd &'£> <^(b 4-jO ^>3%
J^cjcc'o 752. «£c£, /¦ic^cco 43 2Z5* fSb ¦£fSS' ^zlYo
/ 6-5-0 /^e«' 75Z ^>c, J^CC'O ^:° f£;>
600 ISCjCoo Is z. 41^ J-^SyOto 4^s 2^^ 7 2^ -f/o -feT
'kic J£c ocx> 7 /£%> ^o ^'5"
C-UO
C5"C> /4Zj&co 7oZ ^.,c Ifycoe 2z*f <9% *fto <$£$' 4rS%
y
fe£ i^.tco 752. 4/6 I5fyoo -495- 22^ 2% 4C>^ 3T/ ^
l^L£)
I-7/0 140, ceo 7SC
yC/^i £rJj/V-tL^ c'-tu C^-
C/^-Z y^-
t^*c c<- ckju^^d^u^ ¦~+'t''<.(:c(
(jr*L- C^l^I (^£rf*- ¦qJcu
-------
rv"-CA COMPUTATION SHEET
project /fe /fc< A 1 ^c.'>*? < C
<->830
ISSjcoo
}5~Uj <• 00
!5 2.) cud
7SZ
7SS
755
mz
(%3)
mc
Wzsj
Vv$
1 SOjOoc
/ ¥5 00c'
/
V?2
HC1S
ZZ5
2-2.S
0
0
8
¥7$
t Bo
y?S
5 $
5- O
^3
0 HC
/ 0 ts
/o
¥oc
¥7c
*?
C °3o
)55, *<*-
755
c^'
^7. ace
J
¥?s
2.2.8
/£
^(J c
5-. 1.
v YVc
)¥j, OcO
755
*¦ 0 *
c//o
ftyS'jooo
Z?o
5
*//¦> i
—f -7 -*
> i
/ ^/C'l
ill c :
?'.-> <2 I
r> !
t-l A r \
Z/?C
C¦OS lift - 1 7
-------
"l3"
'K*.crc ^
fl^L !L<TV|0^ ^ ^
/4co. ,'44.JCC 7S-2_ (37a0<, 49s- z?g g% «^o -^£r'<^..s7a
4"10 /4^6oft 75 z. 4^ \ilboo--S/fa
i H3^ !'£H/iioo 052. "T^S" SlfP^
i±Sd i^cco 75Z. /J7^o -?%r ;?36 7% ^SS S,5%
Ct'H) &
6'co (a f&P ^ 5"%
''5s-C J'SZjbc-c-- )s~L 4/z -^76"" 2z^- Jt7a '1-os 'iSB 4>
4"3o /jft'cvo 7£Z- (-4c& l^tco ^s~ -Z2.X IUOc i^7 Z 3.5%
(qj-?) '
"9+0 /3Z.CCO 75 Z- /*£A>r> -2 56 4no ^5" O" .5
(_<\-z^) ^ -
. £5£ /55/*> 75 Z- -tfc ^'ooo -4^5" 236 i"Z%> 4io
Is^cco %-Z ^ /6^0:0 4?^ -210 /^°/6 -fts ^.£C/?
ikIO .M^±_'l5Z -K.H /%>cc -rts /4-% -f/S" ^r/S
-------
ENVIRONMENTAL PROTECTION ^GlNCV
orncc of enforcement
NATIONAL FIFLD INVESTIGATIONS CENTER-DENVE
BUILDING 53 BOX 75227, DENVER FEDERAL CENTER
DENVER, COLORADO 80225
On the basis of my experience with the operation of this plant,
I submit that the following figures are the best determinations of the
pertinent operational parameter? for this facility during the period
3-00 P.M. to 5 15 P.M. f May 28 f 1975.
Average bngas.se usage
Averaf; gar.se moisture content
Fuel oil usage (average)
Steam production (average)
Juice production (average)
Boiler efficjency
Tons Net Cane Per Hour
CERTIFIED BY;
31.40 tons/hr
46.5
None
145,000
gal/'hr
lbs/hr
240,000 lbs/hr
68.0
113
7o
Tons
f FACTORY SUPERINTENDENT
i£nature Title Date
KEKAHA SUGAR COMPANY, LIMITED
May 30, 1975
Plant
As sanitary engineer for the Environmental Protection Agency, I agree
that the above figures arc Lhe best determinations of the pertinent opera-
tional parameters for thi? facility during the period 3:00 P.M. to
S: IS P.M. __lLay 2L6 » 1975-
11 Tc'i --u/C^ i *
7/22/75
Title ( ''J Date
Chief, Field Operations Branch 7/22/75
Signa Lure ' - • — ™
Title
Date
-------
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
NATIONAL FIELD INVESTIGATIONS CENTER-DENVER
BUILDING 53, BOX 25227. DENVER FEDERAL CENTER
DENVER, COLORADO 80225
TEST NO. 2
On the basis of my experience with the operation of this plant,
I submit that the following figures are the best determinations of the
pertinent operational parameters for this facility during the period
8 35 A.M. to 11.15 A.M. May 29 1975.
Average bagasse usage
32.04
tons/hr
Avera^'v bagasse moisture content
46.5
7
/O
Fuel oil usage (average)
None
gal/hr
Steam production (average)
148,000
lbs/hr
Juice production (average)
234,000
lbs/hr
Boiler efficiency-
68.0
%
Tons Net Cane Per Hour
111
Tons
factory SUPERINTENDENT
Title
May 30, 1975
Date
KChAHA SUGAR COMPANY, LIMITED
Plant
As sanitary engineer for the Environmental Protection Agency, I agree
that the above figures are the best determinations of the pertinent opera-
tional parameters for this facility during the period 8:15 A.M. to
11:15 A.M. , May 29 , 1975.
Signature
Si gnature
7/22/75
Title ( (/ Date
Chief, Field Operations Branch 7/22/75
Tit]e Date
-------
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
NATIONAL FIELD INVESTIGATIONS CENTER-DENVER
BUILDING 53, BOX 25227. DENVER. FEDERAL CENTER
DENVER, COLORADO 80225
TEST NO. 3
On the basis of my experience with the operation of this plant,
I submit Lhat the following figures are the best determinations of the
"pertinent operational parameters for this facility during the period
2:00 P.M. to 4 15 P.M. , May 29 , 1975.
Average bagasse usage
32.26
tons/h
Average baga^oe moist ^rc content
46.5
%
Fuel oil usage (average)
None
gal^hr
Steam production (average)
149,000
lbs/hr
Juice production (average)
234,000
lbs/hr
Boiler efficiency
68.0
%
Tons Net Cane Per Hour
111
Tons
CERTIFIED BY:
factory SUPERINTENDENT
Title
May 30, 1975
Date
KN\AHA SUGAR COMPANY, LIM1TI-U
Plant
As sanitary engineer for the Environr.ier.tal Protection Agency, I agree
that the above figures arc the best determinations of the pertinent opera-
tional p.irameteis for this facility during the period 2:00 P.M to
4:15 P.M. May 29 , 1975.
TJL
Signature
*
'1 's)%< {Tf'i-y y>v-,-c.^ 7/22/75
TiLlo
Date
Signature
Chief, FLe]d Operations Branch 7/22/75
TiL]e Date
-------
APPENDIX E
VISIBLE EMISSIONS EVALUATION
KEKAHA SUGAR COMPANY
MAY 28-29, 1975
-------
^/Ac/T KUM it..
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
l-'VTIONAL FIELD INVESTIGATIONS CENTER-DENVER
tUILDING 53, BOX 25227, DENVER FEDERAL CENTER
DENVER, COLORADO 80225
Date /n
Address , J^u>
34
5
i^o
35
Fuel ,/ < Cv-"
6
5^
•>
36
Observation'^be^an ^^"Ended ^ ^
Remarks:,
/?, A^/'/cJ {"
S-/fS/ ^
i 4'/ Z5 £>/jlcrC.^.
*•
7
'1c
«pr
>-
37
8
d\j>:
33
9
/_r
:.7
-------
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
NATIONAL FIELD INVESTIGATIONS CENTER-DENVER
BUILDING 53, BOX 25227, DENVER FEDERAI CENTER
DENVER, COLORADO 80225
Dale (5~ - 2 y ~ 9 Location « ,
Observer EU „ cJ NamC— f f'"7; ^
Address UtUrjJir? A/r)
Observation Points',-? ^
0
15
30
45
0
15
30
45
/sir ^ st#<£
0
2T
XO
2f
30
Stack - Distajice From 3cr,J]ieight
I
) 1
'U'
yj
J*
31
V ^ !n '' 'l i-i
Wind - Speed r- - Direction
2
n
/c
~?V
32
Type of Installation /,^
3
3T
jy
l'5~
33
I
4
;?r
-3*
25"
34
5
fr
.?r
35
Fuel
G
4o
XT
3
-------
ENVIRONMENTAL PROTECTION AGENCY
OFFICE Oh ENFORCEMENT
NATIONAL FJElD INVESTIGATIONS CENTER-DENVER
BUILDING 53, BOX 25227, DfcNVEft FEDERAL CENTER
DENVER, COLORADO 80225
Pate fy)±t h 'i /^ Location ^
_, y' / I / Name y/jT/a-( JjTy -r a ^ Co
Observer -j -j-~
Address X-, , Jr^y.. , . .
Observe tioa Point S £*J- c?7"7 .'fA
0
15
30
45
0
15
30
45
0
IS
as
3 &
30
Stack - Distance From/:^ Height
1
Je
25
7S
/$
' '
31
Wind - Speed £~~ Direction
2
y-'
Pc
-
.}
13 2
Type of Installation
3
/$"
3S
,1°
Jo
33
4
J£
f~S
/O
34
Fuel
5
A
26
3*>
35
35
i U. ./<.yi£.
6
2$
po
PS
36
Remarks /j /i // /
£< /r /, /y?M* *
£ /'r /
^ /- •/; >*u' a dih€~jst/'' f'C f'
Ac
/-} y~ ^ i-4
7
Po
A
36
37
8
<8
29
".O
Mana[
Addru
cr Mr. John Robinson
¦v, Kckaha Sugar Company
Kckaha, Kauai,~Tfaw'ai i"
------- |