&EPA
United States
Environmental Protection
Agency
Region I
Office of
Research and Development
J.F. Kennedy Federal Building
Boston, MA 02203
March 1979
Evaluation of
Wood-fired Boilers
and Wide-Bodied
Cyclones in the
State of Vermont
-------
EVALUATION OF WOOD-FIRED
BOILERS AND WIDE-BODIED CYCLONES
IN THE STATE OF VERMONT
Prepared By:
Cedric R. Sanborn
FOR:
THE ENVIRONMENTAL PROTECTION AGENCY
REGION I
BOSTON, MASSACHUSETTS 02203
AND
THE AIR POLLUTION CONTROL SECTION
AGENCY OF ENVIRONMENTAL CONSERVATION
DIVISION OF ENVIRONMENTAL ENGINEERING
MONTPELIER, VERMONT 05602
MARCH 1, 1979
-------
ABSTRACT
A two part emissions testing program for the evaluation of wood-fired boilers
and wide-bodied cyclones was conducted by the Vermont Agency of Environmental
Conservation during the summer of 1977« The first part consisted of testing for
participate, gaseous, and organic matter from small (less than 25 x 10° BTU input)
industrial wood-fired boilers. Tests were conducted on 10 boilers for total
particulates, sulfur dioxide (S02), nitrogen oxides (NOX), and particle sizing of
the exhaust gases. The second part of the program consisted of testing particulate
emissions from wide-bodied cyclones which are used by the wood products industry
to collect and/or transport wood "wastes"-
The primary purpose of the testing program was to qualify and quantify
particulate and gaseous emissions from wood-fired boilers and to acquire a workable
knowledge of the combustion characteristics of wood-fired boilers. The results
of the testing program were used to develop specific regulations for emissions
from wood-fired boilers and wide-bodied cyclones.
The average SC^ emission rate was less than the minimum detectable limit of
3.*» mg/DSCM (2.12 x 10~'Ib/DSCF). NOX emissions averaged k.97 mg/OSCM (3.1 x 10~7
Ib.DSCF), with a high concentration of 30.*4*» mg/DSCM (1.9 x 10~6 Ib/OSCF). The
low sulfur dioxide (S02) and nitrogen oxides (NOX) emissions were most likely due
to both the low sulfur and nitrogen content of the fuel and low firebox temperatures.
Results of particle sizing indicated that up to 40% of the particles emitted
by an uncontrolled wood-fired boiler and up to 80% of the particles emitted by a
controlled boiler are one (1) micron or less in diameter. There may be a potential
health problem associated with wood-fired boilers, since a high percentage of the
particles emitted are in the respirable range (0.1 - 1.0 microns).
The particulate emissions ranged from 0.073 - 1.142 g/OSCM (0.032 - 0.499
gr/OSCF) <£» 12% C02 (excluding boiler H), with an average rate of 0.684 g/OSCM
(0.299 gr/OSCF).
-------
-2-
The results of comparison testing between EPA Method 5 and the high-volume
test method indicates that statistically there is no significant difference in the
overall means between the two methods. However, it was found on a case by case
basis there may be a substantial difference (up to 53%)•
Results of the wide-bodied cyclone tests indicate that if the amount of
sanderdust introduced to the unit is kept to a minimum, the unit will be able to
meet a participate emission standard of 0.137 g/OSCM (0.06 gr/OSCF). The average
emission rate was found to be 0.114 g/OSCM (0.05 gr/OSCF) with a low rate of .009
g/OSCM (0.004 gr/OSCF). Based on field observations, if the opacity of the cyclone
exceeds 20%, the unit is probably incapable of meeting the 0.137 g/OSCM (0.06 gr/
OSCF) standard.
-------
FIGURES I""1
TABLES iv
NOMENCLATURE v
ACKNOWLEDGEMENT v1
INTRODUCTION 1
CONCLUSIONS 2
Part I - WOOD-FIRED BOILERS 5
SECTION
2. BACKGROUND 6
3. TEST METHODS 7
4. TYPES OF BOILERS 17
5. FUEL 24
6. CONTROLS 27
7.0 EMISSIONS 28
7.10 GENERAL DISCUSSION 28
7.20 CALCULATION TECHNIQUES 34
7.21 lb/106 BTU INPUT 34
7.22 gr/DSCF @ 1235 C02 35
7.23 EMISSION RATES 42
7.30 COMPARISON OF HIGH VOLUME & METHOD 5 TESTS 44
7.40 PARTICLE SIZING 58
8. GASEOUS EMISSIONS 62
9. ASH ANALYSIS 66
Part II - WIDE-BODIED CYCLONES 71
INTRODUCTION 72
1. TEST METHODS 73
-------
n
Page
2. DISCUSSION 74
3. PARTICLE SIZING 79
4. WOOD FUEL SIEVE ANALYSIS 84
REFERENCES 86
APPENDICES
A. REFERENCE METHODS 1-5 A - 1
B. HIGH VOLUME TEST METHODS B - 1
C. REFERENCE METHOD 6 C-l
D. REFERENCE METHOD 7 D-l
E. COMPLETE RESULTS OF METHOD 5 SAMPLING E - 1
F. STATE OF VERMONT WOOD BOILER REGULATIONS F - 1
G. STATE OF VERMONT - CYCLONE REGULATIONS G - 1
-------
111
FIGURES
Number Page
1 Participate Sample Train 10
2 Schematic Diagram of Typical High-Volume Train 11
3 SOg Sample Train 12
4 NOX Sample Train 13
5 A & B Schematic Andersen Sample Head 14-15
6 Collection Substrate for Andersen Sampler 16
7 Dutch Oven and HRT Boiler 21
8 Small Spreader - Stoker Boiler 22
9 Pneumatic Wood Feeder For Pneumatic Stoker 23
10 Regression Analysis - Method 5 vs. High Volume (Vermont Data) 56
11 Regression Analysis - Method 5 vs. High Volume (Vermont-Boubel Data) . 57
12 Particle Size Distribution - Wood-Fired Boilers 61
13a. Particle Size Distribution - Wide-Bodied Cyclone 82
13b. Particle Size Distribution - Wide-Bodied Cyclone 83
-------
Number
iv
TABLES
Page
1 Boiler Capacities 19
2 Type of Fuel & Firing 20
3 Chemical Analysis of Wood Boiler Feed 26
4 Excess Air Vs. Color of Filter 30
5 Color of High-Volume Filters 31
6 Color of Impinger Water 32
7 "F" Factors 37
8 Emission Rate (EC), Wood Boilers - Calculated 38
9 Emission Rate (Ep), Wood Boilers - "F" Factor . 39
10 Comparison of Emission Rates+, EC & Ep, Wood Boilers 40
11 Overall Comparison, EC, Ep 41
12 Comparison of Particulate Emission Standards 43
13 Grain Loadings Corrected to 12% C02 - Method 5 46
14 Grain Loadings Corrected to 12% COg - High-Volume 48
15 Comparison of Grain Loadings Corrected to 12% COg, Method 5 Vs.
High-Volume 50
16 Overall Comparison of Method 5 and High-Volume, Corrected to 12% COg 51
17 Impactor Data - Boilers 59
18 Results of NOX Sampling 63
19 Vermont Wood Ash Analysis by AA 68
20 Vermont Wood Ash Analysis by XRF 69
21 Ash Sample Location 21
22 Material Handled by Cyclones During Test 76
23 Cyclone Emission Data 77
24 Impactor Data - Cyclones 80
25 Sieves Analysis 85
-------
NOMENCLATURE
d - Difference of population mean
dF - Degree of freedom
DSCF - Dry Standard Cubic Feet
DSCM - Dry Standard Cubic Meter
g - Gram
gr - Grain
Ha - Research Hypothesis
H0 - Null Hypothesis
H.P. - Horse Power
H-V - High-Volume Test Method
M-5 - EPA Reference Method 5
mg - Milligram
N - Population (sample) Size
NOX - Oxides of Nitrogen
R2 _ Coefficient Correlation
S - Standard Deviation
S02 - Sulfur Dioxide
Sd - Standard Deviation of the Differences
Sx - Standard Error of the Mean
T - Test statistic
T,^- T Distribution
u - Mean Grain Loading
x~ - Population Mean
XRF - X-ray Fluorescence
- Probability (Type I)
- Probability (Type II)
-------
ACKNOWLEDGEMENT
Chief Technician for the Vermont Field Testing was Lawrence McGlll.
Assistance was also provided by George Apgar, Lena Blaise, Richard Cambio,
Richard Couture, James Feeley, Harold Garabedian, Christian Jones and Paul
Wishinski. Sample preparation and laboratory analysis were conducted by the
Vermont Industrial Hygiene Laboratory under the supervisionof Benjamine Levadie.
Analysis were performed by Michael Blanchet, Cindy Parks, and Debbie Voland.
Graphical assistance was provided by Gary Durkee.
The x-ray fluorescence analysis was done by Dr. Thomas Spittler, United
States Environmental Protection Agency, Lexington, Massachusetts.
Fuel analysis were made by Schwartzkopf Microanalytical Laboratory, Woodside,
New York.
-------
INTRODUCTION
A two part emissions testing program for the evaluation of wood-fired boilers
and wide-bodied cyclones was conducted by the Vermont Agency of Environmental
Conservation and the GCA/Technology Division during the summer of 1977. The first
part consisted of testing for particulate, gaseous, and organic matter from small
(less than 25 x 106 BTU input) industrial wood-fired boilers. The State of Vermont
conducted tests on 10 boilers for total particulates, sulfur dioxide (S02), nitrogen
oxides (NOX), and particle sizing of the exhaust gases. The 6CA tests were for
evaluation of the organic compounds of the flue gas. GCA tested five of the above
mentioned boilers. The second part of the program which was done by the State of
Vermont consisted of testing particulate emissions from wide-bodied cyclones which
are used by the wood products industry to collect and/or transport wood "wastes".
The primary purpose of the testing program was to qualify and quantify particulate
and gaseous emissions from wood-fired boilers and to acquire a workable knowledge of the
combustion characteristics of wood-fired boilers. As more and more facilities turn to
the burning of wood as a primary fuel, the need for specific emission regulations for
wood-fired boilers becomes apparent. The results of the program were used to develop
these regulations for Vermont.
Additional boiler tests were made by the State of Vermont during the summer of 1978.
Some of the data generated from these tests have been included in this report.
In order to maintain the confidentially of the results, each facility has been coded
with a letter (A,B,C, etc.). The boilers in the original study are coded as A, B, C, D,
E, F, 6, H, I, and J, with boilers N and 0 added in 1978. The cyclones are designated
as B, D, E, F, I, J, K, L, and M. Each cyclone at a facility was also assigned a number
(1, 2, 3 etc.).
-------
-2-
SECTION 1
CONCLUSIONS
Both sulfur dioxide (S02) and nitrogen oxides (NOX) emissions are relatively low.
This is due to both the low sulfur and nitrogen content of the fuel and a low firebox
temperature. The average S02 emission rate was less than the minimum detectable limit
of 3.4 mg/DSCM (2.12 x TO'7 Ib/DSCF). NOX emissions averaged 4.97 mg/DSCM (3.1 x 10~7
Ib/DSCF), with a high concentration of 30.44 mg/DSCM (1.9 x 1Q-6 Ib/DSCF) and a low
concentration of 1.59 mg/DSCM (9.9 x 10-8 Ib/DSCF).
Results of particle sizing indicated that up to 40% of the particles emitted by
an uncontrolled wood-fired boiler are one (1) micron or less in diameter. The single
unit (N) that was equipped with a multiclone collector had a particulate size distribution
showing 80% of the particles to be less than one (1) micron. There may be a potential
health problem associated with wood-fired boilers, since a high percentage of the particles
emitted are in the respirable range (0.1 - 1.0 microns).
The particulate emissions ranged from 0.073-3.549 g/DSCM (0.032 - 1.551 gr/DSCF) @
12% C02. However boiler (H), with the high emission rate of 3.549 g/DSCM (1.551 gr/DSCF)
@ 12% C02 is not considered to be representative of normal boiler operation. The second
highest emission rate was 1.142 g/DSCM (0.499 gr/DSCF) @ 12% C02- The average particulate
emission rate for boilers (excluding H and A) was 0.684 g/DSCM (0.299 gr/DSCF) @ 12% C02.
It is interesting to note that the four boilers which are hand fired had the lowest
particulate emission rates.
It was found that the majority of the boilers operated with 250-400 percent excess
combustion air. This leads to excess particulate (fly ash and unburned carbon) carryover,
as well as reduced combustion and boiler efficiency. Both the placement and amount of
combustion air is critical to proper operation of wood-fired boilers. Since wood is
approximately 80% volatiles, the majority of the required air is needed to provide a
secondary combustion zone for the volatiles which are driven off as the wood first
starts to burn. Only a minimum amount of air is needed to maintain combustion of the
-------
-3-
fixed carbon. Not only is the fly ash carried over, but it is also possible to carry
out unburned carbon particles which are lifted from the bed, as well as the fines in the
fuel which may not even reach the primary combustion zone. Two of the boilers, E and F,
which had emission rates in excess of the Vermont standard of 1.03 g/DSCM (0.45 gr/DSCF)
@ 12% COg, have had their air systems modified to reduce the amount of combustion air.
The two units were both retested (1978) after modification, and in both cases the
particulate emission rate was reduced by more than 50%. Not only were the particulate
emissions reduced but the overall combustion efficiency and firebox temperature increased,
while the fuel usage was reduced. Therefore, in most cases for boilers similar to the ones
that were tested, a reduction of excess air and proper placement of the air, will reduce
emissions to meet standards. It will not be necessary in most cases to install add on
control equipment.
The results of the comparison testing between EPA Method 5 and the high-volumetest
method indicates that there is no significant difference in the overall means between the
two methods. However, it was found on a case by case basis there may be a substantial
difference (up to 53%). The most likely reason for the difference is the overall test
times. A Method 5 test is run for a minimum of one hour, while a high volume test is run
for twenty (20) minutes or less. Because of its shorter test period, the high volume
method is more apt to reflect any short term cycle in boiler operation. Disparity between
the two methods can be reduced if five or more high volume tests are run per set. The
high volume sampler is the preferred test method since it is simple and therefore easier
and less expensive to use. Since it is statistically comparable to a Method 5 test, it
can be used as a screening test to determine if a Method 5 test would be required or
perhaps as a compliance test.
Results of the wide-bodied cyclone tests indicate that if the amount of sanderdust
introduced to the unit is kept to a minimum, the unit will be able to meet a particulate
emission standard of 0.137 g/DSCM (0.06 gr/DSCF). The average emission rate was found to
be 0.114 g/DSCM (0.05 gr/DSCF) with a low rate of .009 g/DSCM (0.004 gr/DSCF). A cyclone
that handles hardwood waste has the potential for higher emissions, since hardwood
-------
-4-
particles have a lower resin content than softwood particles and therefore do not have a
tendency to agglomerate. A unit that handles large amounts of sander or planer dust will
probably not be able to meet a 0.137 g/DSCM (0.06 gr/DSCF) standard. Based on field
observations, if the opacity of the cyclone exceeds 20%, the unit is probably incapable
of meeting the 0.137 g/DSCM (0.06 gr/DSCF) standard.
-------
-5-
PART 1
WOOD-FIRED BOILERS
-------
-6-
SECTION 2
BACKGROUND
There are approximately 45 Industrial wood-fired boilers in Vermont at the present
time. The largest of these is rated at 30.2 million K cal/hr. (120 x 106 BTU/hr),
although the average size is 1.6 million K cal/hr (6.5 x 10^ BTU/hr). The largest boiler
tested during the original program was rated at 5.5 million K cal/hr (22 x 106 BTU/hr.).
A wide range of emission rates will be found due to variations in boiler design, feed
system, type of wood fuel, and most importantly the degree of operator control. Several
of the boilers have been modified at least once and may bear little resemblance to the
original unit. Less than 10% of the boilers are equipped with emission control equipment.
The collectors used for the older boilers (pre 1975) are instack fly ash collectors, with
the newer units equipped with multi-clones. The maximum steam pressure that the pre-1975
boilers operate at is 150 psi.
Extensive testing has been done on large, high pressure (200-600 psi) boilers in
the Northwest to determine particulate and gaseous emissions. These units burn Western
woods (Douglas Fir, Redwood, etc.), with the majority of them burning a bark and wood
mixture, with a high percentage of bark. The boilers in Vermont primarily burn Eastern
kiln-dried hardwoods, with little or no bark content. Thus,the results of the Western
stddies may not be applicable to boilers in the Northeast, Vermont in particular.
-------
-7-
SECTION 3
TEST METHODS
3.10 Participates
Testing for particulates was performed using two different test methods: a) EPA
Method 5, which is the standard test method, and b) a high-volume test method, which is
being considered as an alternate and comparative test method. The original scope of
work called for simultaneous testing with the two methods, however due to the small size
of the stacks (0.81 - 1.22 meters diameter) it was felt that the high volume method would
cause undue interference with the Method 5 test, therefore the two methods were not run
simultaneously.
3.11 Method 5
The Method 5 tests were performed in accordance with 40 CFR 60, Appendix A, Reference
Methods 1-5, as specified prior to the August 18, 1977 revisions. Due to time
restrictions, the sampling time was reduced from five minutes to three minutes per point
for those locations where the number of required sampling points exceeded 24.
For a Method 5 test, particulate matter is withdrawn isokinetically from the
stack and passed through a heated box containing a fiberglass filter. The gas is then
cooled in an impinger box to 70°F or less, before it enters a dry gas meter which measures
the total gas flow in dry cubic feet. The filter is removed and placed in a sample
dish. The nozzle, probe, and any glassware proceeding the filter are washed with acetone
and the wash placed in a sample bottle. The total amount of moisture collected in the
impingers is measured. In determining the separate particulate weight gains of the filter
and acetone wash, each must be dried, desiccated and weighed. See Appendix A for a copy
of the reference methods. See Figure 1 for schematic of test train.
3.12 High-Volume
Testing for particulates was also performed using a manual Rader high-volume sampler.
Testing was performed in accordance with methods (Modified slightly for boiler application)
outlined in the Oregon Air Pollution Control Regulations (See Appendix B). The Oregon test
-------
-8-
was used since at the time of the test program no formal ASTM or EPA test method had been
developed for this test method.
For the high volume tests, from three to six sample points were used on each stack
depending on the size of the stack and accessibility. In all cases the sample was drawn
isokinetically for a total test time of 15-18 minutes. The test time was dependant upon
the temperature of the stack, amount of particulate collected, and the moisture content
of the exhaust gas. Unlike a Method 5 test train, the high volume train does not have a
heated probe, glassware, or a dry gas meter. The flow through the system is controlled by
a butterfly valve, while the total flow is determined using the recorded orifice pressure.
After the test, the filter is removed and placed in a sample envelope. The nozzle, probe
and filter housing are washed with acetone and the wash placed in a sample bottle. See
Figure 2 for schematic of high volume test train.
3.20 Gaseous
3.21 Sulfur Dioxide:
Three tests for sulfur dioxide (SOg) were performed on each boiler in accordance with
EPA Reference Method 6 (See Appendix C). A gas sample was extracted from a single sampling
point in the stack at a rate proportional to the stack gas velocity. The gas sample passes
through a midget bubbler containing fifteen (15) milliliters (ml) of 80 percent isopro-
panol, which retains any acid mist and sulfur trioxide. The 502 in tne 9as stream is
captured in a three (3) percent hydrogen peroxide solution in two midget impingers. A
fourth and final midget impinger is left dry.
After the sample has been taken, the probe is removed from the stack and the sample
train is purged with ambient air fifteen (15) minutes. After purging, the contents of the
midget bubbler is discarded, and the contents of the three midget impingers placed in a
sample bottle. The midget impinger and the connecting glassware are washed with distilled
water and the wash added to the sample bottle. See Figure 3 for schematic of SOX sample
train.
-------
-9-
3.22 Oxides of Nitrogen (NOX):
Testing for NOX was performed on each of the boilers, following test procedures
outlined in EPA Reference Method 7. A grab sample is collected in an evacuated flask
containing a dilute sulfuric acid - hydrogen perioxide absorbing solution. The nitrogen
oxides, except nitrous oxide, are measured colorimetrically using the phenoldisulfuric
acid (PSD) procedure. See Appendix D for complete testing and analytical procedures
and Figure 4 for schematic.
3.30 Particle Sizing
In-stack particle sizing of the particulates was performed on each of the boilers,
using an Anderson Mark III eight stage cascade impactor. The sample was withdrawn
isokinetically from a single point in the stack. A single point and constant orifice
pressure drop was used in order to maintain a constant velocity through the sizer. If
the velocity is changed then the size of the particles collected on each plate would
also change. Slotted fiberglass filters were used as the collection media (the backup
filter is not slotted). Boilers C, D and F were tested without a backup filter. This
will bias the results, since the particles in the<0.6 micron range were not collected.
See Figure 5 a, b, and c for schematic of sampler and collection substrate.
-------
HEATED AREA
THERMOMETER
CHECK
VALVE
REVERSE-TYPE
PITOT TUBE
VACUUM
LINE
o
i
Figure 1. Particulate sampling train.
-------
FIGURE 2
SCHEMATIC DIAGRAM
TYPICAL HIGH VOLUME PARTICULATE SAMPLING TRAIN
^K_
V5
S" type.
COMPONENTS:
1. Attached pilot tube - "P" type or
2. Nozzle
3. Probe
4. Differential pressure gauge or manometer
5. Filter holder
6. Calibrated orifice
7. Differential pressure gauge or manometer
8. Thermometer or thermocouple
9. Control valve or damper
10. Optional flexible coupling
11. High volume blower
-------
-12-
FIGURE 3
S02 SAMPLING TRAIN
GLASS LINED
PROBE
TYPE S
PITOT TUBE
STACK WALL
SILICA GEL DRYING TUBE
MIDGET BUBBLER MIDGET IMPINGERS
GLASS WOOL
PITOT MANOMETER
ICE BATH
THERMOMETER
NEEDLE VALVE
PUMP
DRY GAS METER
ROTAMETER
-------
PROBE
\
/
FILTER
-13-
FLASK VALVE
\
FLASK
FLASK SHIELD
SQUEEZE BULB
PUMP VALVE
PUMP
THERMOMETER
FIGURE 4
NOX SAMPLING TRAIN
-------
TOP VIEW
7.0"
ANDERSEN 2000 INC.
50-800 SERIES
STACK SAMPLER
(O.D. 2.875")
3/8 O.D. PITOT TUBE
7 TO 8
GOOSE NECK NOZZLE
2-7/8
FIGURE 5A Sampler Adaptation to "EPA type" Pitobe?
-------
-15-
UOUSING
PLATE
HOLDER
CONE
FIGURE 5B
ANDERSON SAMPLE HEAD9
-------
-16-
FIGURE 6
COLLECTION SUBSTRATE PLATE 9
-------
SECTION 4
TYPES OF BOILERS
The two main boiler configurations that are used in Vermont are a firetube/Dutch
oven and a watertube/pneumatic stoker.
The firetube boilers operate at a relatively low pressure (less than 150 psi) and
generally involve pile burning of the fuel in the Dutch oven. The fuel, either fed by
hand or gravity, forms a pile on the grates. The height of the pile is maintained by
visual inspection and the feed rate is determined by the plant's steam demand. Underfire
air passes through the grates either by forced draft or natural draft through the ash
doors. Overfire air is either drawn in through the oven doors and/or through air jets
in the bridge wall.
The majority of the firetube/Dutch oven boilers were manufactured prior to 1930.
On some units, attempts have been made to reduce the size of the pile by feeding the fuel
penumatically. However, this method does not seem to adapt itself readily to Dutch ovens
and higher participate emissions occur (Plant C).
The Dutch oven acts as a primary combustion chamber, burning the fixed carbon in the
fuel while it is in the pile. The volatiles that are driven off are combusted in the
secondary combustion zone, provided adequate over-fire air is supplied. Generally the
over-fire air is injected through jets in the bridge wall itself.
One interesting note is that almost no soot blowing is done on firetube boilers. The
tubes are generally cleaned once a year, during the July shutdown. What effect this has
on boiler efficiency has yet to be determined.
The watertube boilers generally utilize a pneumatic stoker to distribute the fuel
over the fixed grates. The fuel is blown in pneumatically and forms a thin la^er over the
grates. Many of the smaller particles will burn while still in suspension. Underfire air
comes through the grates, which helps promote primary combustion and keeps the ash and fuel
from plugging the grates. Overfire air is added above the grates to form the secondary
-------
-18-
combustion zone for the combustion of the volatiles. The rate of feed is generally
controlled by a screw conveyor whose speed is controlled by the steam demand from the
plant. Once the speed of the screw conveyor has been set to correspond with steam demand,
a watertube/spreader stoker setup (unlike a firetube/Dutch oven) may be almost totally
independent, needing little operator control. However this type of operation may actually
lead to higher participate emission rates, since the automatic controls cannot make adjust-
ments to compensate for variations within the fuel.
The boilers that were tested ranged from a rated capacity of 150 HP to 425 HP. The
State of Vermont Air Pollution Control Regulations define one (1) boiler horsepower (H.P.)
as a unit that is equal to ten (10) square feet of boiler heating surface. It is common
for a watertube boiler to operate at a rate that is 200% over the rated horsepower, while
firetube boilers may operate at 150% of rated horsepower. Of ten boilers, five utilize
pneumatic firing and five are manually fired. The type of fuel ranged from kiln dried
sawdust and shavings to hogged fuel to large chunks and scraps. The boiler at Plant D
burns hogged kiln dried maple which is wetted with water prior to firing. This technique
increases retention time and minimizes carry over. A complete description of each boiler,
and the feed system used is contained in Tables 1 & 2.
-------
-19-
TABLE 1 - BOILER CAPACITIES
APPROXIMATE*
PLANT
A
B
C
D
E
F
G
H
I
J
NO,
H.P. RATING i«0. BOILERS AT PLANT TESTED
425
150
167
150
250
200
257
300
150
150
2
2
2
2
1
1
1
1
1
1
1
1
2*
1
1
1
1
1
1
1
MANUFACTURER FEED RATE
OF BOILER - Tn/Hr
Riley
D. M. Dillon
D. M. Dillon
D. M. Dillon
Keeler
D. M. Dillon
D. M. Dillon
Erie City
Dillon
D. M. Dillon
1.1
0.40
0.40
0.36
0.50
0.44
0.62
1.29
0.33
0.40
*Test done in common stack with both units in operation
+Feed rate during test - since tests were conducted in summer, this rate does not
represent the maximum possible feed rate.
-------
-20-
TABLE 2 - TYPE OF FUEL & FIRING
PLANT
A
B
C
D
E
F
G
H
I
TYPE OF FUEL
Maple, Birch & Pine
(Kiln Dried)
Ash
Mixture Kiln
Dried & Green
Mixture of Hard &
So ftwoods
Maple - Most Kiln
Dried
Oak - Maple
& Pine (Kiln Dried)
Maple
(Kiln Dried)
Mixture of Hard &
Softwoods
Mixture of Hard &
Softwoods
Mixture of Maple
& Pine - Mostly
Kiln Dried
Mixture of Hard &
Softwoods
CONDITION OF FUEL
METHOD OF FIRING
Hogged, shavings, sawdust Pneumatic stoker
Shavings, sawdust,
occasional hand firing
Sanderdust, sawdust-some
hogged fuel
Gravity feed
to pile on grates
Pneumatic feed to
pile on grates
Hogged fuel, shavings, saw- Hand fired
dust-material wetted before
firing-occasionally wet
sawdust & bark used
Hogged, shavings, sawdust, Pneumatic stoker
sanderdust
Hogged, shavings, sawdust, Pneumatic stoker
Edgings-Plywood scrap
Hogged-Plywood scrap,
cores, bark, veneer scraps
Shavings, sawdust, hogged
fuel, occasional hand
firing
Cores, plywood trimmings,
sanderdust, sawdust, veneer
scraps
Hand fired
Pneumatic stoker
Gravity feed from
collector to pile
on grates
Hand fired &
gravity feed from
col1ectors
-------
-21-
D. M. DILLON STEAM BOILER WORKS INC., FITCHBURG, MASS
Setting of Horizontal Return Tubular Boiler
Showing Extension Front or Dutch Oven
This form of setting is used in sawmills and other plants where sawdust, shavings
and slabs constitute the bulk of fuel. Sawdust and shavings are fed through the open-
ings on top of furnace, while the regular fire doors are used for coal and large slabs. On
account of the large grate which is installed with this form of setting, the full rating of
the boiler is obtained when green fuel is used.
FIGURE 7
-------
-22-
FIGURE 8
STEAM OUT £
STiCK
*.R HEATER "CULT1JL|
COLLECTOR
Small spreader-stoker furnace.
-------
-23-
FIGURE 9
PNEUMATIC WOOD FEEDER FOR PNEUMATIC STOKER
DEFLECTOR PLATE
PNEUMATIC STOKER
-------
-24-
SECTION 5
FUEL
Most of the wood fuel that is burned is generated as waste within the plant, and
has a moisture content that varies from 6% (kiln dried) to approximately 60% (green).
The furniture manufacturers use primarily kiln dried fuel consisting of sawdust, shavings,
hogged fuel, and limited quantities of sanderdust. The use of sanderdust is limited, due
to the fact that it can be highly explosive. The species, size, and moisture content of
the fuel varies, depending on the type of wood processed at a given facility, the type of
operation, the availability of the wood waste and the boiler feed systems. (See Table 2)
A spreader stoker requires that the larger pieces be hogged, while a Dutch oven can be
hand fired with chunks and large pieces.
One of the main advantages of wood is that its composition (based on an ultimate
analysis) is relatively constant from species to species, which is beneficial to good
combustion control. Its main disadvantage is that the fuel mixture as it is fed to
the boiler is not uniform. This lack of uniformity may cause a temporary or periodic
"smokey" condition which requires an operator to correct. Wood has low sulfur (less
than 0.05%) and ash (less than 5%) content.
I
The average $£, N2 and ash content (by weight) was 0.02 percent, 0.18, and 0.72
respectively, excluding plants G, H, and J. These three have been excluded since the
fuel used at each plant contains large quantities of plywood scraps. It was felt that
the glue in the plywood adversely affected the results of the analysis. For example
the fuel used at Plant G, which burns exclusively plywood scraps, had an ash content
of eight (8) percent, while the average for raw wood was 0.72 percent. Due to its
overall low $2, N2 and ash content, wood is desirable as fuel. Wood fuel contains 70-
80 percent volatiles with the remainder of it being ash and fixed carbon. Due to the
high volatile content, wood has burning characteristics similar to those of a gaseous
fuel, rather than a solid fuel such as coal. Because the volatiles are driven off as
the wood starts to burn, a secondary combustion zone must be provided above the bed in
order to utilize these gases.
-------
-25-
The BTU value for hardwoods is less than 8,500 BTU per dry pound of wood. Bark
and softwoods have a slightly higher BTU value per pound since they contain more resins
than the hardwood. However, particulate emissions from bark are higher since they
usually contain large amounts of dirt, sand, etc. Bark also has a higher sulfur
content (up to 0.1%). Dry, resin-free wood has a high heat value (hhv) of 8,300 BTU/lb.,
while the hhv of resin alone is 16,900 BTU/lb.3 Thus a high resin content will increase
the overall hhv of the fuel.
Less than 10% of the wood-fired steam generators in Vermont utilize bark, keeping
potential emissions to a minimum. There is only one boiler that uses fly ash re-
injection which provides an opportunity for additional combustion of any unburned carbon
particles. However, re-injection may also lead to higher particulate emission rates.
-------
-26-
TABLE 3 - CHEMICAL ANALYSIS OF WOOD BOILER FEED
FACILITY H
A 6
B 6
C 6
D 6
E 6
F 6
G 5
H • 5
I 6
J 6
N 6
0 6
.19
.10
.12
.11
.24
.26
.93
.36
.30
.14
.34
.44
C
48.57
47.78
48.45
47.36
47.57
47.73
46.30
45.11
49.51
47.16
48.52
47.51
S
0.005
0.025
0.019
0.007
0.004
0.018
0.057
0.032
0.014
0.032
0.021
0.027
N
0.086
0.10
0.88
0.074
0.065
0.083
2.00
0.34
0.094
1.77
0.10
0.15
0
41
38
38
41
41
41
39
39
39
39
42
41
ASH
.70
.74
.47
.36
.54
.14
.44
.44
.69
.21
.46
.84
0.
0.
0.
1.
0.
0.
8.
7.
2.
2.
0.
1.
0.
23
85
32
12
33
44
25
71
80
08
79
,17
,72
1.43
HEATING VALUE,
BTU/DRY LB.
8,360
8,440
8,392
8,430
8,220
8,387
8,400
8,056
8,856
8,300
8,264
8,241
PERCENT
MOISTURE
5.
23.
10.
8.
5.
5.
11.
27.
5
4
8
3
9
3
4
8
5.8
10.5
50.3
46.4
-------
-27-
SECTION 6
CONTROLS
The primary control device presently used on most boilers is an optical opacity
monitor. When the opacity reaches a preset point, usually 20% opacity, the operator
is alerted and can then make the necessary changes to combustion conditions to reduce
the visible emissions. In some cases an additional overfire air fan also comes on
automatically when the preset point is reached.
An add-on control device that is used most often for pre-1975 boilers is an instack
fly ash collector. Three of these are in use at this time.
The boiler at Plant E is the only unit equipped with an instack collector that was
tested. The ash that is collected by this system is not reinjected to the boiler. The
newer boilers are all equipped with multiclone type collectors.
While both the instack collector and the multiclone effectively capture the larger
particles (greater than 10 microns), they do nothing to reduce visible emissions caused by
smoke. Only an operator can make the necessary adjustments to the boiler feed and air
systems to reduce visible emissions, making him the most important control factor. Also
an automatic feed system cannot compensate for changes in the fuel mixture which may
significantly alter the combustion characteristics causing a temporary smoking condition.
-------
-28-
SECTION 7
EMISSIONS
7.10 General Discussion
As discussed previously, there are two main types of boiler configurations in
Vermont. Each unit has its own distinct and unique set up for providing underfire air,
overfire air, and damper control. The primary responsibility of the fireman has not
been to provide maximum combustion efficiency, but to operate the boiler in a manner that
provides the required steam load and at the same time keeps visible emissions to a
minimum. Since the majority of the plants have an excess of wood waste, which is used
for fuel, maintaining a high combustion efficiency has been of little concern. However,
with the cost of fuel oil increasing, and excess wood "waste" becoming a saleable item,
most facilities are now trying to increase combustion efficiency.
The amount of particulates emitted is dependent upon the percent excess air, fly
ash carryover, condition of the fuel, and the combustion efficiency of the system. The
amount of excess air required to reduce carryover should be limited to 125% for an under-
fired system and 50% for a primarily overfired system. As the underfire air reaches
125% or more excess airj the rate of carryover increases. On the other hand, when the
!
percent excess air provided thru overfire air exceeds 50%, no significant change in
particulate emission has been observed.4 The matter that is carried out by the gas is
a combination of fly ash and unburned carbon.
In a system that is equipped with both overfire and underfire air, it is essential
that less than 20% of the total air be underfire air. Recent studies of Junge^ have
shown that for the best combustion efficiency with minimum particulate emissions, the
percent excess air should be held between 100-130%, with an overfire air to underfire
air ratio of 9 to 1.
This helps to reduce carryover, and allows the fuel to remain in the primary
combustion zone longer. The high percentage of over-fire air provides the air necessary
to promote combustion of the volatiles in the secondary combustion zone.
-------
-29-
Average excess air observed during the Vermont study was approximately 300%.
High excess air rates indicate that gas velocities through the boiler are too high,
thus reducing retention time and increasing carryover of fly ash and unburned carbon
particles. Boiler efficiency is further reduced because excess air must be raised from
ambient temperature to boiler temperature, an increase that could be as much as 1800-
2000°F. The placement of the combustion air is just as important as the percent excess air.
In order to burn the pile of wood in Dutch ovens, larger amounts of underfire air
are needed to keep the pile from settling on the grate and smothering the fire. In the
cause of a spreader stoker, less underfire air will be needed, since the fuel forms only
a thin bed that is readily lifted by the incoming air.
Different types of particulate carryover are apparent from a visual inspection of
the test filters. In cases where high rates of underfire were used the material on
the filter is black, indicating partially burned solid carbon which was lifted from the
bed. This condition was more prevalent on the Dutch oven systems. For systems using a
spreader stoker and large amounts of excess underfire air, the filters are brownish in
color, indicating that the fines were being carried out without any combustion. This
condition also exists for boilers that have the overfire air jets located immediately
above the grates. The jets must be far enough above the grates so that the burning bed
is not disrupted. The overfire air should also be introduced such that the fines (from
a pneumatic stoker) are not carried out without ever entering the combustion zone. Where
the excess air was kept to a minimum (100-200%), material on the filter appeared light
grey, indicative of only fly ash carryover.
A further indication of the combustion efficiency is the color of the impinger water.
Dark brown discoloration indicated high levels of volatiles which were not subjected to a
proper secondary combustion zone.
Even though the percent excess air was found to be high, the stack gas velocities
were found to be fairly low (less than fifteen (15) feet per second). Therefore the
velocities through the boiler itself should also be relatively low, minimizing carryover.
-------
-30-
TABLE 4 - EXCESS AIR VS. COLOR OF FILTER
B
H
E*
TEST
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
% EXCESS AIR
447
1121
649
317
328
327
232
298
261
219
269
249
685
532
385
416
374
505
159
131
112
132
201
184
370
311
399
215
174
177
294
279
294
COLOR OFFILTER
Brown
Grey
Brown
Black
Lt. Brown
Lt. Brown
Black
Black
Black
Grey
Black
Lt. Grey
Black
Dk. Brown
Dk. Grey
Black w/wood
Dk. Brown
Dk. Brown
Lt. Grey
Lt. Grey
Lt. Grey
Dk. Grey
Dk. Grey
Dk. Grey
Black
Dk. Brown (2)
Black w/wood
Grey
Lt. Grey
Lt. Grey
Grey
Dk. Grey
Grey
-------
-31-
TABLE 5 : COLOR OF HIGH VOLUME FILTERS
TEST COLOR
1 Grey
2 Brown
3 Brown
B 1 Black
2 Grey
3 Dark Grey
C 1 Black
2 Black
3 Black
D 1 Grey
2 Dark Grey
3 Black
E ' 1 Grey with Wood Particles
2 Dark Grey with' Wood Particles
3 Dark Grey with Wood Particles
F 1 Brown
2 Brown with Fly Ash
3 Light Grey
4 Dark Brown
G 1 Light Grey
2 Light Grey
3 Light Grey
H 1 Grey
2 Grey
3 Grey
I 1 Dark Grey
2 Black with Wood Particles
3 Black with Wood Particles
4 Grey with Wood Particles
-------
-32-
TABLE 6 : COLOR OF IMPINGER WATER
PLANT
A
B
C
D
E
F
G
H
I
J
E*
TEST
1-3
1-3
1
2-3
1-3
1-3
1-3
1-3
1-2
3
1
2
3
1-3
1-3
COLOR
Clear
Clear
Slightly Discolored
Clear
Clear
Slightly Discolored
Brown
Clear
Clear
Slightly Discolored
Brown
Dark Brown
Light Brown
Slightly Discolored
Clear
*1978 Retest
-------
-33-
The results of instack particle sizing show that the majority of the narticles
emitted are less than five (5) microns in size. As much as half of the particles were
one (1) micron or less. The carryover generally seemed to be limited to the smaller
size particles, another indication of low velocities through the system, even though
the excess air rate was high. There is also a possibility that the fly ash particles
break up relatively easy, which would increase the percent of fines emitted. The rate
of break up would be increased significantly with high rates of excess air.
Another factor to be considered was the variations in the feed rate due to steam
demands. If the speed of a screw conveyor is dependent upon steam pressure, a sharp
increase in demand may cause temporary smothering of the fire and increase in emissions.
Units that are fired independent of actual steam demand, such as those that are hand
fired, may also undergo temporary smothering if the fuel is improperly charged.
The estimated particulate emission rate for a wood boiler burning bark-free fuel
(5-50% moisture is 2.3 - 6.80 kg (5 - 15 Ib) per ton of fuel on an as-fired basis. In
cases where the fuel is kiln dried, the estimated emission rate should be closer to
2.3 kg (5 Ib) per ton of fuel.1 In most cases the emission rates from the boilers tested
fell within the 2.3 - 6.8 kg (5 - 15 Ib) per ton range. The main exception to this was
boiler H which had severe combustion problems, as well as improperly placed overfire air
jets.
-------
-34-
7.20 Calculation Techniques
7.21 Q.b/IO6 BTU Input)
The particulate emission rate (lb/106 BTU input) was calculated using both the
pollutant mass rate (PMR) and the "F" factor technique.
The problem associated with using the PMR approach was that, in most cases for small
industrial boilers, the BTU input to the boiler was not known, nor was the steam output
known. Thus several assumptions had to be made in order to calculate the BTU input,
these assumptions included boiler capacity, percent of boiler capacity in use at time
of test, and overall boiler efficiency. The amount of fuel fed to the boiler may be
measured directly or estimated. Depending on the feed system in use, the estimate may
be based on production figures from the plant or by weighing an average charge. If the
amount of steam produced can be obtained from a chart or intergrator reading, it must be
assumed that the chart or integrator is calibrated. If the output is known, a boiler
efficiency must be assumed in order to calculate the BTU input. In any event the BTU
value of the fuel must be determined, either by estimate or analysis. However, even
with all the assumptions, a workable estimate of the total BTU input can be made.
An alternate method of calculating the emission rate in pounds per million (lb/10 )
BTU input is the "F" factor approach. This approach gives an emission rate (E) using
the following equation:
20.9
E - CF ..........
20.9 - %Qz (7-1)
where C is the concentration of the particulates in pounds per standard cubic foot of
flue gas (Ib/DSCF), 02 is the percent oxygen of the flue gas and F is the ratio of the
volume of dry flue gases generated to the gross calorific value (GCV) of the fuel combusted
(DSCF/106 BTU).
The value for C is obtained using:
N
C = 2.205 x 10'6 --—- (7.2)
mstd
where Mn is the particulate collected in mg, and Vm is the volume of gas sampled
at standard condition (DSCF).
-------
-35-
The value of F is determined using an ultimate analysis of the fuel and the
equation:
106 3.64% H + 1.53* + 0.14% H - 0.46%0
F = (7.3)
GCV
where GCV is the gross calorific value of the fuel.
Equation 7-1 utilizes the particulate concentration determined at the stack (C),
the percent Og at the test location, and a BTU analysis of the fuel. All of these values
are readily available and the calculations fairly straight forward. In the event that
an ultimate analysis is not available, a standard "F" factor of 9223 has been developed.
A list of the calculated "F" factor for each test is contained in Table 7. The calculated
factors are based on an utlimate analysis (Table 3) and equation 7-3. The critical
factor in determining the emission rate using the "F" factor is the percent 62- Tables
8 and 9 contain the calculated emission rates using both the "F" factor and the PMR
techniques.
A comparison of the results obtained using the "F" factor and PMR approach is
contained in Tables 10-11. There was no difference in the two averages, however differences
of up to 50% were observed (Plant D - Table 10). Test 1-2 has been excluded since it was
felt, based on Tests 1-1 and 1-3, not to be a representative test. Plant A has been
excluded from the comparison analysis between EC and Ep, since the gas sample
was. not valid. This will bias the results obtained using the "F" factor approach,
since the percent 02 is a critical part of the equation. At this point it was felt that
the "F" factor technique was the more appropriate method and would yield a number that
was more representative of the actual emission rate. This was because no assumptions had
to be made as was necessary for the PMR approach.
7.22 gr/DSCF @ 12% Kb
An alternate approach to quantifying the particulate emission rate is grain loading.
In this case the emission rate was given as grains per dry standard cubic foot of exhaust
gas (gr/DSCF). Since the amount of excess air was not the same for all boilers, then a
baseline had to be made so that all grain loadings were compared equally. Therefore all
-------
-36-
measured grain loadings were corrected to 12% C(>2. This correction was made using the
following equation:
Cc = 12 Cm
~
(7.4)
Where:
Cc = corrected grain loading (gr/DSCF) at 12% C02
Cm = measured gain loading (gr/DSCF) at stack conditions
% C02 = percent C02 measured at the sampling point.
The concentration in the stack, Cm, was determined as follows:
Cm = 0.0154 (MN) (7.5)
VMSTD
Where:
MN = total particulate catch in milligrams
VMSTD = volume of gas sampled (FT3) at standard conditions (68°F and 29.92 in. H2)
on a dry basis.
For the grain loading technique, no assumptions regarding the boiler operation in
BTU input had to be made when determining the emission rate.
-------
-37-
TABLE 7 - "F" FACTORS
PLANT
I W*ll 1
A
B
C
D
E
F
G
H
I
J
N
0
*
DSCH/J
1.045
1.033
1.056
1.009
1.045
1.041
0.998
0.983
1.021
1.041
1.059
1.049
1.038
DSCF/106 BTU
9291.5
9184.3
9395.1
8978.6
9294.2
9170.2
8880.3
8745.3
9083.3
9245.1
9415.2
9334.2
9233
*EPA-40 CFR 60.45 (F) (4) (v)
-------
-38-
TABLE 8 - EMISSION RATE (Ec), WOOD BOILERS - CALCULATED
PLANT x, lb/106 BTU
B
C
D
E
F
G
H
I
I*
J
0.258
0.716
0.134
0.836
1.035
0.348
2.724
1.533
0.791*
1.194
0.008
0.337
0,059
0.064
0.193
0.026
0.573
1.287
0.742*
0.256
A —
0.005
0.194
0.034
0.037
0.111
0.015
0.331
0,742
0.524*
0.148
*Excltiding Test No. 2
-------
-39-
TABLE 9 - EMISSION RATE (EF), WOOD BOILERS - "F" FACTOR
PLANT x. lb/106 BTU S - S=
B
C
D
E
F
G
H
I
I*
J
0.167
0.755
0.068
1.058
0.923
0.238
2.981
l.«84
0.996*
0.702
0.006
0.319
0.024
0.331
0.304
0.026
0.610
1.193
0.047*
0.004
i*'
0.003
0.184
0.014
0.191
0.175
0.015
0.352
0.844
0.027*
0.002
*Excluding Test No. 2
-------
-40-
TABLE 10 - COMPARISON OF EMISSION RATES*, Ec AND Ep, WOOD BOILERS
PLANT
B
C
D
E
F
G
H
I
I*
0
Er, lb/106 BTU
0.258
0.716
0.134
0.836
1.035
0.348
2.724
1.533
0.791*
1.194
EF, lb/106 BTU
0.167
0.755
0.068
1.058
0.923
0.238
2.981
1.684
0.996*
0.702
DIFFERENCE, d
0.091
-0.039
0.066
-0.222
0.112
o.no
-0.257
-0.151
-.205*
0.492
0.957 AVE.
0.957 AVE.
0.000 AVE.
*Excludes Test No. 1-2
+Excludes Plant A
-------
-41-
TABLE H - OVERALL COMPARISON* Ec & EF
Method x S S;
EF 0.957 0.861 0.272
Ec 0.957 0.757 0.239
*Excludes Plant A
-------
-42-
7.23 Emission Rates
For discussion purposes, Plant A is excluded wherever the results are corrected
to 12% C02- Plant H is used only for determining the high average for participate emissions
and for comparison analysis between Method 5 and the high volume method, since this boiler
is not felt to be representative of normal or proper operation.
The average emission rate, using the "F" factor approach and excluding Plants A
and H is 329 nanogram/J (0.76 lb/106 BTU) or 2.43 Kg/hr (5.36 Ib/hr).
Based on the Method 5 tests the average particulate emission rate is 0.684 g/DSCM
(0.299 gr/DSCF) at 12% C02 which is well below the Vermont standard of 1.029 g/DSCM
(0.45 gr/DSCF) @ 12% C02. This corresponds approximately to 5.44 Kg/hr (12 Ib) per ton
of fuel.
The lowest paniculate emission rate was 0.073 g/DSCM (0.032 gr/DSCF) @ 12% C02
and 1.055 g/DSCM (0.461 gr/DSCF) at 12% C02 respectively, both had their air feed systems
modified in early 1978. These boilers were retested in August 1978, and the emission
raes were then found to be 0.398 g/DSCM (0.174 gr/DSCF) at 12% C02 and 0.599 g/DSCM
(0.262 gr/DSCF) at 12% C02. This represents a 35% reduction in emissions for boiler E
and a 43% reduction in emissions for boiler F. These reductions are due to reducing and
reapportioning the flow of combustion air. Therefore it is possible to substantially
reduce emissions by utilizing proper combustion air placement and quantities. For
boilers of the type tested, add-on control equipment may not be necessary.
-------
-43-
TABLE 12 - COMPARISON OF PARTICULATE EMISSION STANDARDS
EXISTING, lb/106 BTU
Plant Actual
A 0.382
B 0.258
C 0.716
D 0.134
E 0.836
F 1.035
G 0.348
H 2.724
I 0.791
J 1.194
Ex 0.395
FX 0.586
+6as Analysis not valid, emission rate (gr/DSCF) not known.
*Proposed regulations adopted, August 1978.
XRetested 1978
' BTU
Allowable
0.34
0.50
0.50
0.50
0.32
0.50
0.50
0.26
' 0.50
0.50
PROPOSED, GRA
Actual
+
0.081
0.380
0.032
0.493
0.461
0.119
1.551
0.499
0.330
0.32 0.174
JNS/DSCF (12%
Allowable
0.450
0.450
0.450
0.450
0.450
0.450
0.450
0.450
0.450
0.450
0.450
0.50
0.261
0.450
-------
-44-
7.30 Comparison of High Volume and Method 5 Tests
7.31 Emission Rates
Comparison testing was originally performed on nine of the boilers, although the
tests were not always run simultaneously due to limitations of the locations and/or stack
size. Boiler J was not tested with the high volume sampler due to the high (greater than
500°F) stack temperature, which adversely affects the operation of the sampler. From
Table 15 it can be seen that the overall grain loadings corrected to 12% COg determined
using Method 5 average 6.6 percent higher than the loadings determined using the high-
volume sampler. For purposes of this comparison, the results of Plants A & H were
included since only a comparison of the results obtained using each test method is being
made. However the results of Plant A are not representative of actual emissions when
corrected to 12% C02 due to an invalid gas sample. The average grain loading for Method
5 and the high-volume method were 1.007 g/DSCM (0.440 gr/DSCF) at 12% C02 and 0.940 g/DSCM
(0.411 gr/DSCF) at 12% COg respectively. In both cases the standard deviation (S) and
the standard error of the mean (Sx) were essentially the same. A comparison of the two
methods is contained in Tables 15 and 16. Although the overall difference is only 6.6
percent, it should be noted that differences of up to 53% were observed on a test by test
basis. In order for the shprt-term high-volume sampler to produce results similar to
those of Method 5, more than three runs will be needed per test site. A set of at least
five runs would help to minimize the bias obtained using the high-volume sampler. The
bias is due to the fact that high volume tests may reflect short-term cycles in the boiler
operation, since the test period is 20 minutes or less. A Method 5 test will have less
bias since the test must be conducted over a period of at least one hour. It may also
be advantageous to increase the high-volume test time.
The high-volume sampling was performed using a manual Rader high-volume sampler.
The major drawback of using this system to test boilers is that control of flow through
the sampler is extremely difficult when stack temperature exceeds 35QQF. Under such
high temperatures, the sampler butterfly valve is affected. Therefore, 1t is possible
that the total volume reported may not be equal to the actual volume sampled because the
-------
,45-
flow must be adjusted constantly. This could account for, in part, the overall 6.6
difference in grain loading as shown in Table 15.
An automatic high volume sampler is available which, among other things, records
the total amount of air sampled. This would remove most of the doubt about the total
volume of air used for the calculations. It would be worthwhile to do some comparison
testing between the automatic and manual samplers in order to determine if there is a
significant difference in results.
-------
-46-
PLANT
B
H
TEST
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
Average
1
2
3
Average
2.2
1.8
1.4
4.6
4.2
4.6
5.0
4.2
5.2
6.2
5.8
5.4
2.2
3.2
4.2
3.4
4.0
2.8
6.8
8.2
9.0
7.4
5.8
6.6
3.8
4.2
3.6
6.0
7.0
6.6
TABLE 13 GRAIN LOADINGS CORRECTED TO
12% C02 - METHOD 5
RECORDED
LOADING (gr/DSCF)
0.083
0.093
0.087
0.087
0.030
0.031
0.029
0.030
0.237
0.083
.145
.155
0.
0.
0.014
0.022
0.011
0.016
0.129
0.113
0.123
0.122
0.136
0.094
0.145
0.125
0.078
0.071
0.087
0.076
0.901
0.654
0.012
0.856
0.163
0.552
0.145
0.154*
0.287
0.164
0.191
0.185
0.180
CORRECTED LOADING (gr/DSCF)
0.453
0.620
0,746
0.606
0.078
0.089
0.076
0.081
0.569
0.237
0.335
0.380
0.027
0.046
0.024
0.032
0.704
0.424
0.351
0.493
0.480
0.282
0.621
0.461
0.138
0.104
0.116
0.119
1.461
1.353
1.840
1.551
0.515
1.577
0.483
0.499*
0.858
0.328
0.327
0.336
0.330
*Excludes Test No. 2
-------
Table 13 (continued) ~47~
RECORDED
PLANT TEST % CO; LOADING (gr/DSCF) CORRECTED LOADING (gr/DSCF)
E+ 1 5.2 0.071 0.165
2 5.4 0.098 0.217
3 5.2 0.061 0.141
Average 0.174
+1978 Retest
-------
-48-
TABLE 14 GRAIN LOADINGS CORRECTED
TO 12% C02 - HIGH-VOLUME
TEST
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
Average
1
2
3
4
Average
1
2
3
Average
1
2
3
Average
1
2
3
4
Average
^^•Mi^^^
2.2
1.8
1.4
4.6
4.2
4.6
5.0
4.2
5.2
6.2
5.8
5.4
2.2
3.2
4.2
3.4
3.4
4.0
2.8
6.8
8.2
9.0
7.4
5.8
6.6
3.8
3.8
4.2
3.6
RECORDED
LOADING (gr/DSCF)
0.128
0.070
0.061
0.086
0.048
0.037
0.013
0.033
0.097
0.090
0.086
0.091
0.027
0.018
0.026
0.024
0.094
0.084
0.089
0.089
0.156
0.110
0.073
0.210
0.137
CORRECTED
LOADING (gr/DSCF)
0.
0.
.068
.056
0.060
0.061
0.535
1.167
0.809
0.837
0.093
0.190
.251
.086
0.
0.
0.698
0.4767
0.523
0.563
0.125
0.106
0.034
0.088
0.233
0.257
0.198
0.229
0.052
0.037
0.058
0.049
0.513
0.315
0.254
0.361
0.551
0.388
0.219
0.900
0.515
0.120
0.082
0.080
0.094
868
414
471
0.155
1.584
0.294
0.600
0.717
0.287
0.474
-------
"1978 Retest
.49.
PLANT
TEST
1
2
3
4
5
Average
RECORDED
LOADING (gr/DSCF)
0.069
0.054
0.071
0.069
0.062
CORRECTED
LOADING (gr/DSCF)
0.159
0.124
0.154
0.162
0.151
0.150
-------
-50-
TABLE 15 - COMPARISON OF GRAIN LOADINGS
CORRECTED TO 12% C02. METHOD 5
VS. HIGH-VOLUME
PLANT
A
B
C
D
E
F
G
H
I*
E+
*Excludfng Test No.
+Re tested 1978
X
METHOD 5
0.606
0.081
0.380
0.032
0.493
0.461
0.119
1.551
0.499*
0.174
0.440 Ave.
2
X
HIGH- VOLUME
0.563
0.088
0.229
0.049
0.361
0.515
0.094
1.584
0.474
0.150
0.411 Ave.
x %
0.043
-0.007
0.151
-0.017
0.132
-0.054
0.025
•0.033
0.025
0.024
0.029 Ave.
DIFFERENCE
7.1
-8.6
39.7
-53.1
26.8
411.7
21.0
-2.1
5.0
14.0
6.6 Ave
-------
-51-
TABLE 16 - OVERALL COMPARISON OF METHOD 5
AND HIGH-VOLUME, CORRECTED TO 12% C02
METHOD x, gr/DSCF S_Ave. S?
High Volume 0.411 0.454 0.144
Method 5 0.440 0.440 0.139
-------
-52-
7.32 Statistical Analysis
In order to determine if there is a significant difference between the particulate
emission rate determined using Method 5 and the emission rate determined using the Rader
high volume method, a statistical analysis of the overall results of the two test methods
was made. In the event there was a significant difference between the two methods, it
was hoped that a correlation between the Method 5 tests and the high volume test could be
made by using linear regression techniques. A paired sample analysis of the Vermont
data was made to determine if the results obtained using the two methods were statistically
equivalent. Additionally, data from the Boubel Study5 was added to the Vermont data
for purposes of performing a linear regression analysis. The combined data base should
generate a line that better represents the differences, if any, in the two test methods.
STATISTICAL ANALYSIS OF METHOD 5 VS. HIGH VOLUME,
PER DRY STANDARD CUBIC FOOT CORRECTED TO 12%
Vermont Data
PLANT
A
B
C
D
E
F
G
H
I
E*
M-5
0.606
0.081
0.380
0.032
0.493
0.461
0.119
551
0.499
0.175
1
H-V
0.563
0.088
0.229
0.049
0.361
0.515
0.094
1.584
0.474
0.150
CONCENTRATIONS GIVEN IN GRAINS
DIFFERENCE
d
0.043
-0.007
0.151
-0.017
0.132
-0.054
0.025
-0.033
0.025
0.025
*1978 Retest
PAIRED SAMPLE ANALYSIS
N = 10 dF - 9 I = 0.029 Sj = 0.066 °^= 0.05
Uj = mean grain loading (gr/DSCF) of Method 5
U£ = mean grain loading (gr/DSCF) of high volume method
H0: UT = U2
Ha: UT / U2
REJECT H0 IF /T/ > T
-------
-53-
T = d = 0.029 = 1.389 (7.6)
0.066/nO
Tot/i from T Tables = 2'262
Since 2.262> 1.389 Accept Hull Hypothesis (H0)
95% Confidence Interval For Difference in Means (Ui - U2 = 0)
d - d" = T/ Sd
(7.7)
d = 0.029 +. 2.262 (0.066)
d = 0.029 ±0.047
B Analysis of Null Hypothesis (M
C = 0.05 Two-sided T Test
d - /do - da/ - XO-0.047/
Sd 0.066 (7.8)
d = 0.71
From OC Curves
B = 0.30
Linear Regression (Vermont only)
Y = §o + BT X! (7.9)
From HP-25 Program
B0 = -0.038
§! = 1.021
Y = -0.038 + 1.021 X
Where:
X = Grain Loading Determined Using Method 5
Y = Grain Loading Determined Using High-Volume
R2 = 0.979
Linear Regression (Vermont & Boubel )
Additional data points (Boubel)
-------
-54-
Modified M-5 High-Volume Difference
0.138 0.204 -0.066
0.263 0.223 0.040
0.186 0.265 -0.079
0.234 0.240 -0.006
0.116 0.106 0.010
0.086 0.113 0.027
0.254 0.246 0.008
Y =
From HP-25 Program
Y = -0.007 + 0.990 X
R2 = 0.971
From the paired sample analysis of the Vermont data, using a significance level of
<=< = 0.05, it can be seen that the critical T value ("£j of 2.262 is greater than the test
statistic (T) of 1.389. Since the value of T does not exceed the T^value, the null
hypothesis (H0), that the values of the two mean grain loadings Ui and Ug are not
significantly different (i.e., U-j = Ug) can be accepted.
However there is still some risk or probability that the null hypothesis (H0) is
incorrectly accepted using the above technique. This is known as the B risk. From the
results of the B test above, it can be seen that there is a 30% probability of incorrectly
!
accepting the null hypothesis (H0) that there is no significant difference in the means.
Therefore, there is a possibility that 30% of the time the two means (Ui and Ug) may be
significantly different even though the T test indicates no difference. This difference
can be seen from the results of Table 15, where the following differences in the means
are observed; Plant C-40%, Plant E-27%, and Plant D-53%.
A linear regression was done in order to correlate the values obtained using the high
volume method and Method 5. For the first regression analysis, the ten data points from
the Vermont Study only were used (See FigurelO). The calculated equation for the line
(y = 0.038 + 1.021x) shows almost a one to one relationship between the two test methods.
The R2 value of 0.979 reflects minimal scattering of the data and indicates a good fit
of the line to the data points.
-------
-55-
For the second linear regression, the results of the Boubel5 study were added to
the Vermont data. The calculated equation of the line (y = 0.007 + 0.990'x) changed
slightly due to the overall effect of the additional data points. (See Figure 11).
Again, a relatively high R2 value (0.971) was obtained.
It can be shown statistically that at a significance level of 0.05 there is no
significant difference between the results obtained by the two test methods. It would
appear in the case of wood-fired boilers that the shorter and easier high volume test
method could be used in place of the more complex Method 5. However, it can be seen
from the results of the Vermont study that the actual differences may be as much as 50%
on a case by case basis. This should be kept in mind if the high volume test is sub-
stituted for the Method 5 test.
One possible way to minimize the difference would be to conduct a series of at
least five (5) high volume tests per test set rather than the customary three. Because
the high volume test is generally only 15-18 minutes, it is possible to bias the test
due to fluctuations in the boiler load and feed rate which may cause a corresponding
change in the emission rate. By performing more than three tests, the overall test time
will be increased and the bias minimized. Currently, more comparison testing is being
conducted by the State of Vermont to substantiate the hypothesis that the two test methods
are not significantly different. In any event, the high volume test method can be used
as a relatively inexpensive screening method to see if the more complex Method 5 test is
required. With a sufficient number of tests (5 or more) per test location, the results
obtained using the high volume test method should be equivalent to those obtained using
the Method 5 tests.
-------
(VERMONT DATA)
FIGURE 10
-------
1M LOADING,GR/OSCF.a»il2Vo .ddij, METHOD 5
(VERMONT-BOUBEL DATA)
FIGURE II
-------
-58-
7.40 Particle Sizing.
The results of the instack particle sizing is shown in Figure 12 and in Table 17.
Boilers C, D, and F were tested without a backup filter. It should also be noted that
boiler N was equipped with a multi-clone collector which significantly reduces the
amount of large particles (> 5u) that are emitted. This effect can be seen from Figure
12. Boiler E was equipped with an instack fly ash collector. However the collector
does not seem to substantially alter the particle size distribution, when the results
are compared to the size distribution of the other boilers. All of the remaining boilers
did not have particulate control devices.
There was the possibility that some of the larger particles would break up when
either entering the impactor or moving through the collection plates. This would
indicate that the percent of small particles (less than 5 microns) actually emitted
was less than indicated by the impactor. In order to verify the cascade impacted results,
an alternate method of sizing should be performed. Assuming the results were unbiased,
for an uncontrolled small industrial wood-fired boiler, it would appear that 40% (by
weight) of the particles emitted were less than one (1) micron in size, while only 20%
(by weight) were larger thjan 10 microns. For a boiler controlled with a multiclone,
the size distribution will be similar to that of boiler N.
-------
-59-
TABLE 17 - IMPACTOR DATA - BOILERS
BOILER
C+
D+
E
F+
G
H
I
J
*Back Up
TEST
1
2
3
1
2
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Filter Not Used
,£1.0 u
21.9
37.0
24.6
0.0
6.4
56.7
49.6
38.3
9.5
48.8
38.4
76
68
56.5
4.4
3.6
1.9
34.4
30.0
^~W.Q
50.5
58
52.5
PERCENT (BY WT)
4C5.0 u 4J110.0 u
45.9 74.77
65.6 86.1
57.0 84.2
5.9 42.0
24.5 71.4
67.5 79.5
69.4 82.6
57.7 72.5
20.7 36.6
61.6 71.8
58.1 69.9
92 96
87.8 92.6
87.4 94.3
10.6 21.3
9.2 25.5
9.9 25.4
38.3 45.5
33.0 39.6
47.1 57.5
58.3 68.7
62.5 68.6
68.4 77.4
-------
Table 17 (continued) "6C~
PERCENT (BY WT)
BOILER TEST ^ 1.0 u <5.0 u <10.0 u
N*
*Equipped With
G+
1
2
Multi-Clone
1
2
85
81
63.0
57.5
98.5
93.3
i
80.0
76.0
99
99
87.9
83.0
1978 Retest
-------
IS
18
i 8pi iy« gsra
rrrmt—HT—H-rttlrH-rH—H—r—I—H-HiiJ r4;
•H44^iHr|i-'-tr~H-HHi^
±lxii114-iii:4:i.:iirxi^J:. :.:riJ_:
^ffifr+ffl ^t-r-M-t-I-Tt-1 T-,-»: '
^:;;u:plT^:|liiIr:;':::'::;HTl!lr
j. [lilmliffltffil; bJ±H3];tG ":J±
. _I_J_H_ ' I -LiJ-JJ _1 L.._l...J—!_4_.. I-.1. J J-t.-U^' f
g_jJ_mi±j^j.TT i _.'_j-I--l,'.u^.'
!-f-|Tl_LJL|~L1 U, :.J\r • V- ;-: J . -' i..-: — '.. -+ Ij-j-.iJ-. i_!-!.,;...;I, L'.j-i-fa-|.-|_!
Sffllffl:*: tfiftl-u Hlflfi
••1
11 i ''' i j
1T ht-irr-f- "*
iS!MBlffiip«ffi£i
::v:::-,: •! .;::;--:-5ft3
11 Ar
•igft ip4|j p jig
Lilii:
i±ffi(:jffll{ffi'j.V.-\V-::
»i3SS™
:
. •
'1 -
te
FIGURE 12
-------
-62-
SECTION 8
GASEOUS EMISSIONS
NOy
The samples from boilers J and D were contaminated in the laboratory and samples
A and C were stored improperly making these four sample sets invalid. In most cases
there was less than one microgram of NOX detected in the sample. The maximum calculated
emission rate was 0.32 kg/hr (0.718 Ib/hr). The average emissions rate was 4.97 mg/DSCM
with a high concentration of 30.44 mg/DSCM and a low of 1.59 mg/DSCM. The NOX emission
rate was determined using the concentration determined from the laboratory analysis in
conjunction with the stack gas flow rate.
An elemental analysis was performed for each wood fuel sample (Table 3). The
nitrogen present ranged from 0.065-2.00 percent by weight. Thus, the amount of nitrogen
available for NOX formation from the fuel itself was minimal. Using a Leeds & Northrup
optical pyrometer, the fire box temperatures ranged from 1800-2200°F, with the average
temperature being 1900°F. A low fire box temperature minimizes NOX formation from the
free nitrogen in the combustion air. As a contrast, the fire box temperature in boiler H
was 2200°F, (the highest value recorded) which corresponded with the higher NOX emission
rate (0.32 Kg/hr). Due to overall low fire box temperatures and low nitrogen content of
the fuel, relatively low emission rates of NOX were expected.
SO*
A laboratory analysis of the sample for the sulfur dioxide fraction was done by the
barium-thorin titration method. The analysis showed the S02 content in all of the samples
to be less than the minimum detectable limit of 3.4 mg/DSCM (Method 6). It can be seen
from Table 3 that the average sulfur content of the wood fuel is less than 0.022 percent
by weight, with a maximum sulfur content of 0.057 percent. .Because of the small
quantities of sulfur present in the fuel, a low quantity of S02 was emitted. Using the
minimum detectable limit of 3.4 mg/DSCM and an average stack gas flow rate of 9905m?/hr, the
amount of SOg emitted would be less than 33 g/hr (0.1 Ib/hr).
-------
-63-
TABLE 18- RESULTS OF NOX SAMPLING
PLANT: F
TF - 66°F
T! - 250°F
Flask No.
102
103
107
108
109
m
1 1 1
121
me
IUD
*NOX present
PLANT: B
Tp - 68°F
T - 205°F
Flask No.
102
103
104
105
107
108
109
111
121
4P, in. H?0
+1.2
-38.1
-18.6
-26.8
-37.2
untn
VU1U
-25.8
unm
VU1U
unin
vuiu
is less than value
AP, in. H00
-40.2
-15.0
+1.2
+0.4
-13.7
-13.4
-12.2
-0.2
-39.3
Pp - (30.09 - A?) 1n Hg
P! - 29.97 in. Hg
Vol.
Flask, ml No.., ug
2032 2.5
2034 .dl.O
2031 <1.0
2014 41.0
2051 4 1 .0
2034 2.5
given
PF - (29.40 -4P) 1n. Hg
Pj - 30.21 in. Hg
Vol.
Flask, ml Nou, ug
2032 <1.0
2034 41 .0
2041 41.0
2049 <.1.0
2031 41.0
2014 <1.0
2051 <1.Q
2031 <1.0
2034 <1 .0
August 18,
Concentration
Ib/DSCF
3.4 x 10'7
1.4*
1.4*
1.4*
1.4*
3.4
August 31,
Concentration
Ib/DSCF
2.2 x 10'7
2.2
2.2
2.2
2.2
2.4
2.2
2.2
2.2
T977
Emission
Rate, Lb/hr
0.141
0.058*
0.058*
0.058*
0.058*
0.141
1977
Emission
Rate, Lb/hr
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
*NOX present is less than value given
-------
Table 18 (continued)
-64-
PLANT: I
Tp - 70°F
Tj - 360° F
Flask No.
102
103
104
105
107
108
109
111
121
*NO present
PLANT: E
Tp - 68° F
Tj - 3300 F
Flask No.
102
103
104
105
107
108
109
111
121
4P, in. H?0
-29.8
-28.5
-31.2
-34.3
-35.3
-33.7
-32.9
-10.1
-33.1
is less than the
AP, in. HZ0
-25.9
-19.2
-17.5
-20.9
-16.5
-19.4
-16.7
-18.6
-19.4
Pp - (29.55 -^P) in. Hg
P! - 30.08 in. Hg
Vol.
Flask, ml NO.,, ug
2032 <1.0
2034 <1.0
2041 < 1 .0
2049 <1.0
2031 <1.0
2014 <1.0
• 2051 <1 .0
2031 <1.0
2034 <1 .0
value given
PF - (29.45 -aP) in. Hg
Pj - 29.70 in. Hg
Vol.
Flask, ml NO.,, ug
2032 5
2032 <1
2041 5
2049 2.5
2031 10
2014 <1
2051 <1
2031 <1
2034 <1
September
Concentration
Ib/DSCF
1.15 x 10"7
1.17
1.17
1.16
1.17
1.18
1.16
1.17
1.17
September
Concentration
Ib/DSCF
4.9 x 10"7
9.8 x 10'8*
4.9 x 10"7
2.4 x 10'7
9.8 x 10"7
9.9 x 10~8*
9.7 x 10"8*
9.8 x 10'8*
9.8 x 10-8*
7, 1977
Emission
Rate, Lb/hr
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0.023
30, 1977
Emission
Bate, Lb/hr
0.170
0.034*
0.170
0.083
0.339
0.034*
0.034*
0.034*
0.034*
*NOX present is less than the value given
-------
Table 18 (continued)
-65-
PLANT: 6
TF - 75°F
Tj - 400°F
Flask No.
102
103
108
109
*NOX present
PLANT: H
TF - 75°F
Tj - 240° F
Flask No.
104
105
107
111
121
140
AP, in. HoO
-10.6
+1.6
-5.7
+6.5
is less than the
AP, in. H?0
-6.9
+6.7
+1.5
+1.5
+9.2
-2.0
Pp - (29.35 -2M>) in. Hg
Pj - 29.31 in. Hg
Vol.
Flask, ml N0«, ug
2032 8
2034 5
2014 3
2051 <1
value given
Pp - (29.35 -^P) in. Hg
P! - 29.98
Vol.
Flask, ml NO*, ug
2041 7.5
2049 13
2031 <1
2031 3
2034 11.5
2026 8
October 4,
Concentration
Ib/DSCF
6.7 x 10'7
4.2 x 10'7
2.5 x 10"7
8.2 x 10'8*
October 5,
Concentration
Ib/DSCF
1.1 x 10~6
1.9 x 10"6
1.5 x 10~7*
4.4 x ID'7
1.7 x lO'7
1.2 x 10'6
1977
Emission
Rate, Lb/hr
0.218
0.137
0.082
0.027*
1977
Emission
Rate, Lb/hr
0.416
0.718
0.057*
0.166
0.064
0.454
*NOX present is less than the value given
-------
-66-
SECTION 9
ASH SAMPLES
Samples of wood ash were collected from the fire box of each of the boilers that
were sampled. Some additional samples were taken in the breeching area, and at the base
of the stack where possible. A preliminary analysis of thirteen ash samples was made
using both atomic adsorption (AA) and x-ray/fluorescence (XRF) techniques. Fourteen
elements were analyzed for using the AA, with the major emphasis on heavy metals, while
eleven elements were analyzed for using the XRF. The results of these analyses are
contained in Tables 19 and 20.
Concentrations obtained using XRF may not be representative of actual concentrations
due to the lack of sensitivity of the instrument and the fact that no sample preparation
is required prior to analysis. Additionally the print out from the unit doesn't lend
itself easily to quantification. In most cases, the concentrations determined using XRF
were consistently lower than the concentrations found using AA techniques.
All of the samples which were analyzed on the AA were run in duplicate in order
to verify the results. Four of the samples required additional ashing due to a
high percentage of unburneti material. These samples were analyzed both before and after
ashing.
As can be seen from Table 19, the concentration of a given element varied consid-
erably from sample to sample. This variability could result from:
1. The species of wood;
2. The soil characteristics where the tree was cut; or
3. Process or boiler contamination.
From the results of the AA analysis, it can be seen that wood ash contains high
concentrations of aluminum, calcium, iron, potassium, and magnesium. However the ash
also contained appreciable amounts (greater than 10 ppm) of cobalt, copper, chromium,
nickel and lead.
The fourteen elements that were found using the AA represents 40-60 percent of the
total ash sample. It is felt that large portions of the remaining ash consists of
-------
-67-
siliconcompounds. There are also small quantities of heavy metals (such as
Vanadium, Titanium, Strontium, etc.) present.
Presently the wood ash is either dumped in a landfill type environment or used on
a limited scale as fertilizer for agricultural purposes. At the existing level of use,
wood ash disposal is not felt to be a problem. However, the ramifications of increased
use of wood and the disposal problems that go with it are unknown and should be fully
investigated before long term financial conmittments to burn wood on a large scale are
made by industry, municipalities and the general public.
-------
TABLE 19 - VERMONT WOOD ASH ANALYSIS BY AA TECHNIQUES
Sample I.D.
176-11179
177-11204
177-11204*
178-11206
179-11207
180-11176
181-11175
182-11121
182-11121*
183-11122
184-11123
184-11123*
185-11124
186-11125
187-11126
187-11126*
188-11127
AL
44.16*
7.11
64.63
14.02
8.9
21.07
16.47
0.154
2.56
21.08
4.34
13.61
11.86
13.54
6.61
15.63
17.46
Ca
131.95
19.62
178.4
428.9
231.0
101.81
425.6
12.96
214.0
241.45
38.23
119.80
151.4
441.5
60.52
142.3
99.6
Cd
0.002
0.0006
0.005
0.0023
0.0033
0.164
0.0075
0.004
0.0059
0.003
0.0021
0.0067
0.078
0.0024
0.0011
0.0025
0.0019
Co
0.613
0.0030
0.0268
0.0292
0.0256
0.0232
0.0398
0.0027
0.0439
0.0868
0.0092
0.0287
0.0621
0.0173
0.0040
0.0304
0.0170
CR
0.1288
0.0069
0.0624
0.0409
0.0170
0.0592
0.1131
0.0110
0.1815
0.0572
0.0237
0.0742
0.0354
0.0357
0.2165
0.5089
0.0278
Cu
0.1602
0.0245
0.2224
0.0690
0.5440
0.5075
0.1072
0.0114
0.1885
0.5309
0.0267
0.0837
0.0339
0.1775
0.0800
0.1881
0.1647
ELEMENTS
Fe
39.51
0.3715
3.3737
12.26
5.272
27.42
22.01
3.8284
63.22
113.14
4.3517
13.64
122.95
17.63
41.44
97.43
16.86
K
21.265
3.907
35.50
15.99
58.51
23.03
16.00
1.852
30.57
25.06
1.834
5.75
31.13
11.60
17.897
42.07
4.304
Hg
187.405
5.922
53.77
206.15
318.70
223.5
391.25
3.258
53.79
253.9
5.934
18.60
22.86
16.91
12.39
29.14
7.40
Mn
3.703
0.808
7.351
8.856
0.500
7.794
20.809
0.474
7.822
16.114
1.163
2.644
17.615
5.735
1.782
4.190
2.426
Na
17.53
0.691
6.281
2.209
1.674
7.595
4.324
1.703
28.13
9.829
2.793
8.75
12.112
11.625
10.980
25.81
0.0365
Ni
0.059
0.004
0.0365
0.079
0.038
0.111
0.094
0.0059
0.0978
0.102
0.0228
0.0540
0.0553
0.148
0.0649
0.1525
0.037
Pb
0.0762
0.0132
0.1204
0.0432
0.0380
0.3164
0.0817
0.0053
0.0874
0.1187
0.0228
0.0716
0.2143
0.0862
0.0139
0.0326
0.0299
Zn
0.0177
0.1608
0.1139
2,8860
8.0866
0.4166
0.0053
0.0896
0.3159
0.0112
0.0352
2.3894
0.1173
0.0033
0.0078
0.'2S66
CO
*Sample Completely Ashed
+A11 concentrations are given 1n mg/g (parts per thousand)
-------
-69-
20 - Vermont Wood Ash Analysis by XRF*
11204
11122
11123
11124
11125
11126
11127
11175
11176
11179
11121
11206
11207
Ti
<100
200
200
1,050
< 50
200
750
400
850
700
<50
100
200
V
<30
66
66
130
< 30
<30
100
130
130
100
<30
<30
66
Cr
<20
<20
<20
22
<20
44
66
<20
44
66
^20
<20
66
Mn
760
2,180
790
3,740
940
1,500
550
2,960
1,600
780
600
650
62
Fe
130
12,500
1,500
3,700
1,040
2,300
4,700
3,000
7,000
5,600
450
400
530
Rb
31
33
22
55
44
51
35
44
100
80
<7
15
120
Ni
<4
36
<4
<4
<4
13
<4
9
22
7
<4
<4
<4
Cu
<6
90
13
30
20
30
10
23
60
66
16
6
96
Zn
50
20
37
210
16
37
50
50
1,010
92
27
7
370
Pb
< 9
13
9
94
27
<9
18
22
76
18
<9
<9
9
Sr
54
10
88
360
560
590
150
540
230
370
110
220
450
1. Samples analyzed directly on XRF.
2. Quantitation performed by standard addition technique. Cr, Cu, Pb and Rb added
t!T> caim-il a 1 1 tf\A
to sample 11204.
3. Minimum detection limit expressed as
-------
-70-
TABLE 21
ASH SAMPLE LOCATIONS
SAMPLE I.D.
11204
11122
11123
11124
11125
11126
11127
11175
11176
11179
11121
11206
11207
PLANT
I
E
H
C
J
G
H
E
A
F
C
D
B
LOCATION
Fi rebox
Fi rebox
Boiler house roof
Breeching
Firebox
Fi rebox
Base of stack
Fly ash collector
Fi rebox
Fi rebox
Fi rebox
Fi rebox
Firebox
-------
-71-
PART II
WIDE-BODIED CYCLONES
-------
-72-
INTRODUCTION
The primary purpose of testing the cyclones was to develop participate emission
standards for wide-bodied cyclones. The emission standard is based on the high volume
method. Additionally for ease of computation and applicability, the emission standard
was to bein grains per dry standard cubic foot of exhaust gas. In order to determine
overall particulate size collection efficiency under various load conditions particle
sizing was conducted at the cyclone outlet and a sample of the material entering such
unit was collected for a sieve analysis.
-------
-73-
SECTION 1
TEST METHODS
Particulates
Testing for participates was performed using a manual Rader high-volume sampler.
Testing was performed in accordance with methods outlined in the Oregon Air Pollution
Control Regulations (See Appendix B). The Oregon test was used since at the time of the
test no formal ASTM or EPA test method had been developed for this test method. The test
procedure involves "mapping out" the face (outlet) of the cyclone to determine the flow
pattern. The point with the highest flow rate and at least five (5) representative points
were selected for testing. Each point was sampled isokinetically for three minutes, for
a total test time of at least 18 minutes. After the testing was completed the filter was
removed and placed in an envelope. The nozzle probe and filter housing were cleaned with
acetone and the acetone wash placed in a sample jar.
Particle Sizing
Particle sizing was performed at the outlet of the cyclone. An Anderson Mark III
eight stage cascade impactor was used. The sample is drawn isokinetically from a
single sample point maintaining a constant orifice pressure drop in order to keep the
velocity through the sizer constant.
-------
-74-
SECTION 2
DISCUSSION
A total of twenty-four (24) different cyclones were tested, with two of the cyclones
(1-5 and L-l) tested under two different load conditions for a total of twenty-six
sets of tests. Cyclone 1-5 which was tested once when handling only softwood wastes
and again with primarily hardwood wastes, is listed as Cyclone I-5A while handling hard-
woods and 1-5 for softwood wastes. Cyclone L-l was tested once when handling only planer
dust and shavings and again when handling only hogged material. The cyclone is denoted
as L-1P and L-1H respectively. From Table 22 it can be seen that a wide variety of wood
species and types of wood waste (hogged, sanderdust, shavings, etc.) were handled by the
cyclones that were tested.
The average particulate emission rate for the cyclones was 0.114 grains per dry
standard cubic meter (g/DSCM) [0.05 grains per dry standard cubic foot (gr/DSCF)], with
a high emission rate of 0.448 g/DSCM (0.196 gr/DSCF) (I-5A) and a low emission rate of
.009 g/DSCM (0.004 gr/DSCF) (M-3). (See Table 23) The testing showed that five of the
units had an emission rate of 0.137 g/DSCM (0.06 gr/DSCF), the.1Imitation presently.set by
the State of Vermont. Cyclone No. I-5A had an average emission rate of 0.448 g/DSCM
(0.196 gr/DSCF) when carrying mixed kiln dried hardwood waste and an average emission rate
of 0.119 g/DSCM (0.052 gr/DSCF) when handling softwood waste. This was probably-"caused by
the fact that hardwood dust (sander, planer, etc) is generally smaller in size then the
dust from softwood. Also since the resin content of the hardwood wastes is low, these
particles do not readily agglomerate as do the softwood particles. In the case of
cyclone 1-5, only seven (7) percent of the softwood particles emitted are 10 microns or
smaller in size, while 41 percent of the hardwood particles were smaller than 10 microns
in size. Therefore it could be expected that wide-bodied cyclones handling hardwood
waste would have a higher emission rate and would emit smaller size particles than a
similar unit handling only softwood waste.
-------
-75-
Sanderdust, due to its small size (2-80vO is not effectively collected by a wide-
bodied cyclone. Cyclone J-2 handles sanderdust exclusively and had an emission rate of
0.371 g/DSCM (0.162 gr/DSCF) with 68.5 percent of the particles less than 10 microns
in size and 26 percent of the particlesless than 5.0 microns in size. Cyclone E-6
was also fed a large amount of sanderdust, however there were also planer shavings
and hogged material mixed in with the dust which help reduce the number of sanderdust
particles emitted. In this case only 40 percent of the particles are less than 10 microns
in size.
The average emission rate of those collectors which were meeting the particulate
emission standard of 0.137 g/DSCM (0.06 gr/DSCF) was 0.071 g/DSCM (0.031 gr/DSCF). In
most cases if there is little or no sanderdust introduced to the collector, a wide-
bodied cyclone will be able to operate within standards. A summary of the test results
is contained in Table 23.
-------
-76-
TABLE 22- MATERIAL HANDLED BY CYCLONES DURING TESTS
CYCLONE ID. TYPE OF WOOD
E -1 & 6 Oak - Maple
E -4 & 5 Oak - Maple
F-l Maple
F-2 Maple
D-1 Maple
D-2 • Maple
D-4 Maple
B-l Ash
B-2 Ash
M-l & 2 Oak, Birch, White Ash
M-3 Birch, Oak, White Ash
I 1 Pine
I 2 Pine
I 3 Pine
1-5 & 7 Pine
I-5-A Maple
L IP Assorted Hardwoods
L 1-H Assorted Hardwoods
J 1 Assorted Hardwoods & Softwoods
J 2 Assorted Hardwoods & Softwoods
CONDITION OF WOOD
Hogged, sawdust, planer
shavings (Kiln dried)
Sanderdust, shavings, sawdust
(Kiln dried)
Sanderdust, sawdust (Kiln dried)
Sawdust, shavings, planer
shavings, sanderdust, hogged
(Kiln dried)
Shavings (Kiln dried)
Sawdust, planer shavings
(Kiln dried)
Hogged (Kiln dried)
Shavings, sawdust (Kiln dried)
Shavings, sawdust (Kiln dried & Green)
Hogged, sawdust
Chips, sawdust
Planer shavings, sawdust (Kiln dried)
Sawdust, planer shavings, sanderdust
(Kiln dried)
Hogged (Kiln dried)
Hogged, sawdust, sanderdust,
planer shavings (Kiln dried)
Hogged, sawdust, sanderdust,
planer shavings (Kiln dried)
Planer shavings (Kiln dried)
Hogged (Kiln dried)
Sawdust, shavings
Sanderdust
-------
-77-
TABLE 23- CYCLONE EMISSION DATA
Cyclone I.D.
E - 1
E - 4
E - 5
E - 6
F- 1
F- 2
D - 1
D - 2
D - 4
B - 1
B - 2
M - 1
M - 2
Test No.
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
4
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
gtVDSCF
0.047
0.049
0.048
0.038
0.035
0.035
0.075
0.209
0.150
0.112
0.076
0.073
0.040
0.042
0.074
0.066
0.027
0.042
0.021
0.009
0.009
0.008
0.018
0.029
0.019
0.015
0.020
0.023
0.020
0.020
0.010
0.019
0.026
0.027
0.008
0.187
0.102
0.027
0.025
0.009
Emission Hate Average
Rate, QP./DSCF
0.048
0.036
0.145
0.087
0.055
0.030
0.009
0.022
0.019
0.017
0.024
0.099
0.020
-------
Table 23- Continued
Cyclone I.D.
-78-
Test No.
Emission Rate
ar/DSCF
Average
Rate, ftr./DSCF
M - 3
I - 1
I - 2
I - 3
I - 5A
I - 5
I - 7
L -IP
L 1 H
J - 1
J - 2
K
I -10
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
1
2
3
1
2
3
1
2
3
1
2
3
0.003
0.007
0.003
0.018
0.020
0.021
0.027
0.007
0.009
0.010
0.014
0.008
0.275
0.178
0.134
0.041
0.068
0.046
0.039
0.039
0.040
0.028
0.070
0.079
0.020
0.030
0.022
0.019
0.013
0.209
0.050
0.228
0.024
0.022
0.033
0.085
0.090
0.094
0.004
0.020
0.014
0.011
0.196
0.052
0.039
0.059
0.025
0.018
0.162
0.026
0.090
-------
-79-
SECTION 3
PARTICLE SIZING
Particle sizing was performed on nine different cyclones, with Cyclones 1-5 and
L-l, each tested twice under different conditions for a total of eleven sets of sizing
data. The size distribution for the cyclones is presented in Figures ISA and 13B. It
can be seen that in the majority of cases less than 35% of the particles (by weight)
are smaller than five microns, while 60% of the particles (by weight) are greater than
ten microns.
The original scope of work called for sizing to be done in each inlet to the
collectors. Due to the high concentrations and generally large size of the material enter-
ing the collector, it was not possible to conduct in line tests. However, samples of
wood waste entering the collectors was taken for most of the units that were tested.
A size distribution by weight of the material entering the cyclones is presented in Table
24.
-------
CYCLONE l.D. TEST <1.0 u
-80-
TABLE 24- IMPACTOR DATA - CYCLONES
PERCENT (BY WT)
F - 1
•-I - 5 *
I-- 5A +
I - 7
J -.2
E - 1
E -5
E -6
K
l\
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
0.0
0.0
0.0
0.0
0.8
4.5
0.0
1.0
i-.o
2.9
5.8
2.0
0.0
0.0
1.1
0.2
0.2
0.2
0.9
0.1
0.0
1.6
0.0
0.3
0.2
0.0
0.0
3.3
4.7
7.8
47.7
0.9
5.3
35.7
25.9
23.6
7.7
16.3
5.5
12.1
24.0
27.2
25.0
31.2
44.7
10.4
7.4
6.1
11.7
5.7
7.5
13.0
4.5
0.0
x-iu.v u
73.6
67.8
65.0
12.9
92.2
94.7
64.3
74.1
39.0
92.3
75.0
94.5
40.6
28.4
25.4
62.2
59.0
35.3
72.0
78.6
76.1
66.9
51.0
63.3
59.8
52.4
87.1
-------
-81-
TABLE24- IMP ACTOR DATA - CYCLONES (CONTINUED)
PERCENT (BY WT)
CYCLONE I.D. TEST <1.0 u <5.0 u >10.0 u
L
L
- P
- H
1
2
1
1.8
2.6
14.4
42.7
23.6
42.1
12.2
37.8
29.8
* - Handling Softwood
+ - Handling Hardwood
-------
99.a9
99.9 99.8
95 90
80 70 60 50 96 98 99
WIDE
:::: ; : I BMifl ^ i
FIGURE I3A
-------
99
OC.d 9J.8
Sb
\
' "' '
, • rnzJ
.••••: -::±m=
I l~l * I
, . . , .-,—r— p-L-^-J-f—j--,
v
r ^~r ; ; ; ; ' ; i :
i'i'i~r ' ' ' <-~ n''~rrn".-r'TTi~!~n"rf~l — r-r-H
•;' : ~^i i i i i"^1 r^rn ' 1 i , '~i i i : i
FIGURE I3B
-------
-84-
SECTION 4
. WOOD FUEL SIEVE ANALYSIS
An analysis was done on samples of wood fuels taken from various wood-working
plants in Vermont. This analysis consisted of quartering samples taken from boiler
feeds, cyclones, and the like, and running a known amount (about an ounce) of this
material through six sieves on an automatic sieve shaker for 5 minutes. The sieves
were then dismantled, and respective weights of collected materials were noted.
Results are exhibited on a percent of total sample basis. Samples were weighed
to the nearest l/10th of a gram. A small amount of material was lost in transfer.
-------
TABLE 2 5 . SIEVE ANALYSIS -85-
SIEVE SIZES
Sample
Description
E Boiler
E - #1 Cyclone
E - #3 Cyclone
E - #4 Cyclone
E #5 Cyclone
C Main Boiler Feed
C Hammer Mill
I — Boiler Feed
I - Cyclone #1
I - Cyclone #2
Rip Saw
#5 and #7 Cyclones
#10 Cyclone
B Boiler Feed
J Boiler Feed - Scrap
J - Cyclone #1
K Cyclone #1
L - Cyclone #1-H
L Cyclone fl-P
D - Boiler Fuel & Cyclone
D Cyclone #1
D Cyclone #2
M Boiler Feed
M Shaker #3
M Cyclone #4
F — Boiler Feed Cyclone
A - Boiler Feed
#10
0.0787"
51.6%
14.8%
73.7%
2.3%
53.8%
37.8%
92.5%
28.4%
11.3%
86.0%
32.3%
36.1%
21.1%
66.3%
(100%) All
0.2%
41.8%
98.7%
49.1%
#4 95.7%
58.4%
25.6%
59.0%
96.9%
6.0%
82.5%
76.4%
#18
0.0394"
20.8%
19.7%
14.6%
10.1%
25.0%
16.6%
7.3%
25.3%
17.1%
7.3%
41.0%
24.2%
75.9%
22.7%
Pieces
3.6%
28.2%
1.0%
23.9%
3.8%
31.1%
24.4%
20.1%
2.0%
23.2%
10.4%
19.4%
#40
0.0165"
18.7%
32.5%
9.9%
33.0%
18.2%
11.4%
0.2%
30.4%
29.6%
4.1%
20.3%
19.8%
2.5%
9.3%
Larger Than 1st
31.3%
21.6%
0.2%
15.7%
0.4%
8.8%
34.3%
13.3%
0.4%
46.1%
1.9%
3.0%
#80
0.0070"
6.6%
20.9%
1.4%
26.7%
1.8%
8.4%
0.2%
9.6%
20.3%
1.1%
5.5%
9.8%
0.2%
1.0%
Sieve
34.9%
5.5%
0.1%
7.6%
0.2%
0.5%
12.4%
5.2%
0.2%
14.1%
1.1%
0.9%
#140
0.004"
1.5%
9.1%
0.5%
15.6%
0.2%
7.8%
0.2%
3.4%
14.0%
0.4%
1.4%
5.2%
0.2%
0.3%
17.9%
1.1%
0.1%
5.2%
0.2%
0.3%
2.0%
1.6%
0.2%
4.0%
1.5%
0.2%
#400
0.0015"
0.8%
6.3%
0.5%
10.4%
0.2%
11.2%*
0.2%
2.9%
8.4%
0.2%
0.5%
3.5%
0.1%
0.3%
9.8%
0.7%
0.1%
3.4%
0.2%
0.3%
0.5%
0.5%
0.2%
2.6%
1.9%
0.2%
*6.5% less than last sieve
-------
-86-
REFERENCES
(1) Anonymous, Compilation of Air Pollutant Emission Factors. 2nd edition,
United States Environmental Protection Agency.
(2) Anonymous, Standards of Performance for New Stationary Sources. United
States Environmental Protection Agency 40 CFR 60. Appendix A as amended.
(3) Baumeister, Theodore, Mark'sStandard Handbook for Mechanical Engineers. 7th
edition. McGraw-Hill Book Co. Chapter 17, pp. 19-19.
(4) Junge, David C., Investigation of the Rate of Combustion of Wood Residue Fuel.
United States Environmental Protection Agency Contract No. EY-76-C-06-2227,
TPR No. 1, September 1977.
(5) Morford, Jerry M., The Comparison of a High-Volume Sampling Method with EPA Method
5 for Particulate Emission from Mood-Fired Boilers. Oregon State University,
1975.
(6) Brown, O.D., "Energy Generation From Wood Waste", National District Heating
Association, French Lick, Indiana, June 1973.
(7) Boubel, R.W., Control of Particulate Emissions From Wood-Fired Boilers. United
States Environmental Protection Agency Contract No. 68-01-3150.
(8) Supplied by Rader Companies, Inc., Portland Oregon.
(9) Supplied by Andersen 2000 Inc., Atlanta, Georgia.
-------
A-l
APPENDIX A
REFERENCE METHODS 1 - 5
-------
APP.A
Title 40—Protection of Environment
(1) Method 5 for conccitU'uUou of par-
.llcululu mailer tuul as.suclnU.'(l molsluro
Content;
(2> Method l (or sample- nnd velocity
traverses;
(3) Method 2 for velocity and volu-
metric flow rate; and
(4) Method 3 for gas analysis.
(b) I''or Method 5, the sampling time
tor each run shall be at least four hours.
When a single EAF is sampled, the sam-
pling tune lor each run shall also In-
clude an integral number of heats.
Shorter sampling tunes, when necessi-
tated by process variables or other fac-
tors, may. be approved by the Admin-
istrator. The minimum sample volume
shall be 4.5 dacm (160 dscf).
(c) For the purpose of this subpart,
the owner or operator shall conduct tho
demonstration of compliance with 00.-
272(a)(3) and furnish tho Adminis-
trator a written report of the results of
the test.
(d) During any performance test re-
quired under i C0.8 of tills part, no case-
ous diluents may be added to tho
effluent cos stream after the fabric In
any pressurized fabric filler collector,
unless the amount of dilution is sepa-
rately determined and considered in tho
determination of emissions.
(e) When more than one control de-
' vice serves the EAF(s) being tested, the
concentration of particulate matter shall
bo determined using the following
' equation:
Align:
.
by method 6.
" W-tol»l nunibor of control doviett
i-ii- '. totted. • • . .
0..-volun>olrlc Bow rote ot tlio effluent
eat stream In dsmi/hr (dscl/hr) M
•!•••• aotcrmlncil by motlicd 2.
, or (QO.-veluo o[ Uio applicable pararooUr for
, " '; each control device twled./
• ' (f) Any control device subject to tho
' provisions of this subpart shall be de-
signed and constructed to allow meas-
urement of emissions using applicable
• test methods and procedures. • •
(B) Where emissions from any EAP(s>
. are combined with emissions from faclll-
. tics not subject to tho provisions of this
subpart but controlled by a common cap-
* tore system and control device, tho owner
N-/
• : C.-concoulrallon of nartlculnlo irmllcr
or operator inny use ixny of the follow-
ing' proecilwos during a iiorforniiuiec
test:
(l) Dasa compliance on control of tho
combined emissions.
(2) Utilize a method acceptable to
the Administrator which compensates
for the emissions from the facilities not
subject to the provisions of this subpart.
(3) Any combination of the criteria
of paragraphs (g)(l) and (2> of tills
section.
(h) Where emissions from any EAF(s)
are combined with emissions from facili-
ties not subject to tho provisions of
this subpart, the owner or operator may
use any of the following procedures for
demonstrating compliance with 1 60.272
(a) (3) :
(1) Base compliance on control of tho
combined emissions.
(2) Shut down operation of facilities
not subject to the provisions of this
subpart.
(3) Any combination of tho criteria
of paragraphs (h) (l) and (h) (2) of this
. section.
ArrlNDUC A — RjUTJlEHOt MCTUODS
urrnoD i — BAUPLK AND VELOCITY TRAvniSKg.
FOR BTATJONAflY SOURCES
1. Principle and XppKcabltlly.
1.1 Principle. A sampling site and tho
number of traverse points are selected to aid
In tho extraction of a representative sample.
14 Applicability. Tlili method should
bo applied only when specified by the, test
procedures for determining compliance with
" the How Source Performance Standards. TJta-
1 leu otherwise specified, this method la not
• Intended, to Apply to gas streams other than
those emitted directly to the atmosphere
' without further processing,
9. Procedure. . , .
. 9.1 Selection of a Biunpllng slto and mini-
mum number of traverso points.
9.1.1 Select a sampling site that Is at leaot
eight stuck or duct diameters downstream
' and two diameters upstream from any .flow
> disturbance cueh as a bond, expansion, con.
': traction, or visible flame. For rectangular
••cross section, determine on equivalent dlam-
-•eter from the following equation: ••••••
... ..-. . i<- <•. i:-. •• • equation 1-1
••8.1.9 When the'1 above1' sampling slt»
• criteria eon be mot, tho minimum number
.,ot traverse points. Is twelve (U). ..
... 9.1.3 Some sampling situations render tho
...abovo .sampling slto criteria Impractical. '
.When this Is the' case, choose a convenient
'•' sampling location and use Figure 1-1 to de-
Chapter I—Environmental Protection Agency
App. A
tormina tho minimum number of ttiwurno
pulnla. Under uo ooudltloiu cliould a iuu»-
piinu point bo selected wltliln 1 inch or the
stuck Wftll. To obtain tbo number ot travoroo
pulnto lor steaks or duclo with a dlnmotur
IMS than 9 foot, multiply the number of
polnU obtained from Figure 1-1 by 0.07.
9.1.4 To uso Figure 1-1 ant measure the
distance from the chosen sampling location
to the noarest upstream and downstream dis-
turbances. Determine the corresponding
number ot traverse points for each alotanc*
from Figure 1-1. Select llio hlgliur of Uio
two numUora of iruvcno pulnM, or a yrculct
vuluo, ouoh Unit (or clrcutnr ulncltu lliu IHIUI-
bar lo a multiple of 4, and for rccUnnuiar
stacks the number followo the criteria of
seollon 3.9a.
2 J Oross-sectlonol layout and location ol
traveno points.
3.3.1 For circular stacks locate tho tra-
vorao points on at least two diameters ac-
cording to Figure 1-9 and Table 1-1. The
travorso axes shall divide tho stack cross
section Into equal parts.
NUMBin OF DUCT DIAMETERS UPSTREAM-
(DISTANCE A|
FROM POINT OF ANY TYPC OF
DISTURBANCE IOENO. EXPANSION. CONTRACTION. ETC.)
NUMBED OF OUCIDIAMEURS OOWNSTRCAM'
(DISTANCE 0)
Figure 1-1. Minimum number ol Iravorso points.
-------
Figure 1-2. Cross section of circular stack divided inlo 12 equal
areas, showing location of traverse points at centroid of each area.
o ^
.. _- -._
o
0
1
1
• 1 *
1
r t
1
0 | 0
J
II
1
p r
1
0 | 0
1
o
-- - --—
o
o
Figure 1-3. Cross section of rectangular stack divided into 12 equal
areas, with traverse points at centroid of each area.
00
Table 1-1. Location of traverse points 1n circular stacks
(Percent or stack diameter from Inside wall to traverse point)
Traverse
point
number
on »
d1imot«r
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Number of traverse points on a dlamettr
2
14.6
85.4
1
4
6.7
25.0
75.0
93.1
6
4.4
14.7
29.S
70.5
85.3
95.6
8
3.3
10.5
19.4
32.3
67.7
80.6
89.5
96.7
,
,
10
2.5
8.2
14.6
22.6
34.2
65.8
77.4
85.4
91.8
97.5
' ' '!'
12
2.1
6.7
11.8
17.7
25.0
35.5
64.5
75.0
82.3
88.2
M.3
97.9
14
1.8
5.7
9.9
14.6
20.1
26.9
36.6
63.4
73.1
79.9
85.4
90.1
94.3
98.2
V
16
1.6
4.9
8.5
12.5
16.9
22.0
28.3
37.5
62.5
71.7
78.0
83.1
87.5
91.5
95.1
98.4
18
1.4
4.4
7.5
10.9
14.6
18.8
23.6
29.6
38.2
61.8
70.4
76.4
81.2
85.4
89.1
92.5
95.6
98.6
20
1.3
3.9
6.7
9.7
12.9
16.5
20.4
25.0
30.6
38.8
61.2
69.4
75.0
79.6
83.5
87.1
90.3
93.3
96.1
S8.7
22
1.1
3.5
6«,0
8.7
11.6
14.6
18.0
21.8
26.1
31.5
39.3
60.7
68.5
73.9
78.2
82.0
8S.4
80.4
91.3
94.0
96.5
98.9
.24
1.1
.3.2
5.5
7.9
10.5
13.2
16.1
\9.4
23.0
27.2
32.3
39.8
60.2
67.7
72.8
77.0
80.6
03.9
86.8
8». 5
92.1
94.5
96.8
98.9
For rectangular stacks divide the
cross section Into M many equal rectangular
areas as traverse points, such that the ratio
of the length to the width of the elemental
areas U between ona ana two. Locate the
traverse points at the eentrold of each equal
area according to Figure 1-3.
9. Re/arenca.
Determining Dust Concentration In a Gas
Stream, ASMS Performance Tost Oodo #27,
New York, N.Y., 1857.
Devorkln, Howard, et al.. Air Pollution
Source Testing Manual. Air Pollution Control
District. IXM Angeles. Calif. November 1803.
Methods for Determination of Velocity.
Volume, Dust and Wat Content of Oases,
Western Precipitation Division of Joy Manu-
Co.. Los Angeles. Calif. Bulletin
WP-oO. 1000.
Standard Method tor Sampling stacks for
Partloulato Matter, to: 1071 Book of ASTM
Btoadards, Port 23. Philadelphia, Pa, 1071.
ASTM Designation D-2928-71.
UCTHOD i—DEXBamtrATiON or STACK OAS
vzLocrrr AND vottnonsio now BATI (TTPC
B VlfUV TUUE) . .
1. Principle on* applicability..
1.1 ?r'roipio. stack gas velocity Is deter-
mined r, -«i the got density and from mou-
uremont or Ui« velocity bead uilng a Type 3
(Btausoholbe or rovers* typo) pilot tube.
12 Applicability. This method should bo
applied: only when specified by the teat pro-
07
-------
App. A
Title 40—Protection of Environment
ctxiitrca for determining compliance with the
Now Source Performance Standards.
3. Apparatus.
2.1 Pltot tubo—Typo S (Figure 2-1), or
equivalent, with a coefficient within ±0%
ovor the working range.
2.3 Differential pressure gauge—Inclined
manometer, or equivalent, to measure veloc-
ity head to within 10% of the minimum
value.
9.3 Temperature gauga—Thermocouple or'
equivalent attached to the pltot tube to
measure (tack temperature to within i.s% or
the mt"l"w? absolute stock temperature.
2.4 Pressure gauge—Mercury-filled U-tubo
manometer, or equivalent, to measure stack
preosuro to within 0.1 in. Hg.
3.5 Bnromatct'—To measure atmospheric
pressure to within 0.1 In. Jig.
2.0 Oils analyzer—To nmilyxe ens composi-
tion for determining molecular weight.
3.7 Fltot tube—Stnudard typo, to call-
brato Typo S pltot tube,
3. Procedure.
3.1 Set up the apparatus as shown in Pig.
ure 3-1. Make euro oil connections are tight
and leak tree. Measure the velocity bead and
temperature at the traverse points epoeiaod
by Method I.
3.3 Measure the static pressure la tht
stack.
3.3 Determine the stack gas molecular
weight by gas analysis and appropriate cal-
culations as indicated in Method 3,
. PIPE COUPLIIMC
TUBING ADAPTER
•••-.:••. :-.!!•: .-.'• •: • ;
'. ». . \.a ».:•: :.•• . ..-4
'Figure 2-1. Pltot tube-manometer assembly,u:
68
Chaptor I—Environmental Protection Agoncy
App. A
I. calibration.
4.1 To calibrate tho pltot tube, measure
the velocity hend at come point In a flowing
fu ttre&m wltli both a Type 8 pltot tube and
t ttaudard typo pltot tube with known co-
.Mclent, Calibration ahould bo done In the
laboratory ana tho velocity of the flowing gas
itraun should be varied over the normal
verting range, It Is recommended that the
calibration bo repeated after uso at each field
ilte. •
4.2 Calculate the pltot tube coefficient
uilog equation 3-1.
f —
CB'»
2-1
where:
C»1(,,=Pltot tube coefficient of Type S
. . pltot tube.
CF,,,=Pltot tube coefficient of standard
type pltot tube (if unknown, use
0.00).
aptua Velocity head measured by stand-
ard typo pltot tube.
ipuiiss Velocity head measured by Type S
pltot tube.
44 Compare the coefficients of the Type 8
pltot tube determined first with on* leg and
then the other pointed downstream. Use the
pltot tube only if the two eoefflelents differ by
no more than 0.01. :
E. Calculation*.
Use equation 3-3 to calculate the etaek gas '
ttlooltj.
i "' Equation 2-2
wharsl
(V0.,..-8uek tas'veleeity, toit pcr'aaeond (tpjO.
\M,
Wbwtbenunlti
tube eoeRleltat, dtiBMSIentasa,
absolute staok las Uaperaturt,
„!!.
(VAli)>ri.»Ar vrlnrliy lioail ol snick cm, InellM
HiOtwiiHB.22).
1',-ALjviliili'Sim'k fiu iirrnuro, liu-lin UK.
M.~.\lolmil:ir wi-lr.lit ol Huck |!U (wet buU)
\\i.ll\i.-mu\f.
.
Mj-Dry moloculur wcljjhl of Hack cui (Irom
Hrthod 3).
D»«Pro|iorUon by voliunt ol water vapor la
(bo gas ilroaio (from Method 4).
Figure 2-3 shows a sample recording sheet
'for veioelty traverse data. Vse the averages
in the last two columns of Figure 9-3 to de-
termine the average stack go* veioelty from
Equation 3-3.
Uso Equation 3-3 to calculate the stack
gas volumetric flow rate.
Equation 2 '
whore:
Q.-Votumolrto flow rate, dry bask, itondard eondl-
lloiu, ll.Vlir.
A - Cross-vxtilonal arcu of ilnck. ft.*
T,j-Al»nlulo temperature at ilnndard eondltlani
at itandard tondltloni, WM
InehH
prcaiurt
lie.
6. Rtjerenett.
Mark, L. Bv Mechanical EnRinssrs' Hand-
book. Moaraw-HUl Book Co.. Inc. New York,
H.V., 1881.
• Perry. J. H.. Chemical Engineers' Hand-
book. Mearaw-IUU Book Co.. Inc., How York,
H.Y., 1000.
Shlgohara, n. T., W. V. Todd. and W. a
Smith. SlgnlOcanee of Errors in Stack sam>
i pllng Measurements. Paper presented at the
; Annual Meeting of the Air Pollution Control
: Association. St. Louis, lio, June 14-19. 1970.
Standard Method for Sampling Stacks for
PorticuUte Matter, la: 1971 Book of ASTM
< Standards, Part 33. Philadelphia, Pa, 1971,
• ASTM Designation D-2038-71.
Vennard, 3, SL, Elementary Fluid Mftehan-
Ics, John Wiley ft Sons, Inc.. Mew York, K.Y,
WT.
l.'Ji V .-
>
09
-------
App.
PL*
DAI
RUN
STA
BAR
STA1
OPEfl
A Tillo 40— Protection of Environment
NT
FE
NO.
CK DIAMETER, in..
DMETRIC PRESSURE, in. Ho.
1C PRESSURE IN STACK (Pg), In. Hfl.
tATORS
mu^
SI
Traversa point
number
.
"•
1.. • i .
•
Velocity head,
in. H2O
t
AVERAGE:
vfip-
•
.. * "
THEMATIC OF STACK
CROSS SECTION
Slock Temperature
Flguro 2-2. Velocity traverse data.
i-
Chapter I—Environmental Protection Agency
App. A
urriioo s—OAS AMACTBIB FOB CAIIDON DIOXIDX,
rZCCSS AW, AND D8Y KOLEOVtJUl WHOHT
1. Principle ant applicability.
1.1 Prtnolplo. An Uueemtod or grab gas
sample IB extracted tram ft sampling point
»nd analysed toe Us components wing on
Qnftt vuuner, "
la Applicability. ThU method should be
applied only when apaalfled by tha teat pro-
cedures for determining eompllanee with the
New Bourea Ferform*no« Staadsids. Tha test
procedure will Indicate whether • grab sam-
ple or on Integrated sample to to be used.
9. Apparatia.
9.1 Grab sample (Vlgora 8-1).
9.1.1 Probo—Stainless steel or Pyres'
glass, equipped with a filter to remove partle-
ulate matter.
3.1.3 Pump—One-way squeeze bulb, or
equivalent, to transport gas sample to
analyser.
9.3 integrated (ampto (Figun 3-9),
/
2.9.1 Probe—Stainless steel or Pyrex1
glnss, equipped wltli a litter to remove par-
tloulnto maucr.
22.2 Air-cooled condenser or equivalent—
To remove any excess moisture.
MS Needle valve—To adjust now rate.
2.2,4 Pump—Leak-tree, diaphragm type,
or equivalent, to pull gas.
2.2.0 Rate meter—To menaure B now
range from 0 to 0.036 elm.
3.9.8 Flexible bag—Tedlar,1 or equivalent
with a capacity of a to 9 ou. ft. Leak teat the
bag In the laboratory beloro using.
2.3.7 Pilot tube—Type S. or equivalent
Attached to tho probo no thnt tlio sampling
now rate con be regulated proportional to
the stock gas velocity when velocity is vary-
ing with Umo or a sample traverse is
conducted.
2.3 Anafysti.
2.3.1 Great analyzer, or equivalent.
1 Trade name. ,
FLEXIBLE TUBING
TO ANALYZER
LTEHIG
FILTER (GLASS WOOL)
ft
SQUEEZE BUtB
. Rgura 3-1. Grab-sampling train.
RATE I
•.; AIR-COOLED CONDENSER ..
PROBE •"•''••
FILTER (GLASS WQOL) (,
•;., i..' iKi.lVi* • i liim • .:.> •; i,'
QUICK DISCONNECT
Flguro 3-2, Integrated gas - sampling train.
-------
App, A
Tillu -10—I'rototllon of Envlronmonl
8. Procedure.
3.1 Grab sampling.
3.1.1 Sot up tbo equipment na oliown In
Flguro 3-1, waking sure nil connections an
look-free. Plnco tho proba la tlia alack nt a
sampling point and ptirca llio wimpllng lino,
3.1.9 Draw enntplo into tho aualyzcr.
3J Integrated campling.
3.2.1 Evacuato the flexible bac. Set up the
equipment OB shown In Figure 3-3 with the
bag disconnected. Place the probe In the
•tack and purge tho sampling line. Connect
tho bag. making euro that all connections are
tight and that them are no leaks.
3.2.3 Sample at a rato proportional to the
stack velocity.
3.3 Analysis.
3.3.1 Determine the CO,, O,, and CO con-
centrations as soon ne possible. Make as many
passes as are necessary to give constant read-
ings. IT more than ton passed aro nccesoary,
replace the absorbing solution.
332 For grab sampling, repeat the sam-
pling and analysts until three consecutive
samples vary no more than 0.5 percent by
volume (or each component being analyzed.
3.3.3 For integrated sampling, repeat the
analysis of tho sample until throo contioeu*
tlvo analyses vary no more than 0.3 percent
by volume for each component being
analyzed.
4. Calculations.
4.1 carbon dioxide. Average the three eon-
eemtlvo runs and report the result to tho
nearest 0.1% CO,.
4.3 Excess air. trie Equation 3-1 to calcu-
late axceaa air, and average tho rum. Report
the result to tho nearest 0.1% exceed air.
7o t A-
(%0»)-0.5(%CO)
».264(% N,)-(% Oi)+0.5(%C05
X100
equation 3-1
vhere:
%EA
,
Pereent excess air. .
HO,= Percent oxygen by volume, dry bosla,
XH,=Percent nitrogen by volume, dry
bail*.
%CO=Perc«nt eorbon monoxide by vol-
ume, dry basis.
0.304 =RatIo em oxygen to nitrogen In air
by volume.
4.3 Dry molecular weight. Use Equation
-2 to calculate dry molecular weight and
veraga the run*. Report the result to the
le.ircst tenth.
+o.38<%N,4-%co)
! ' equation 8-3
/hero: ;
ftMDry molecular weight, ibyib-mott.
%CO«— Percent carbon dioxide by volume,
dry basis.
«O«*Porcont oxygen by voluat, dry
basts,
%N*«Percant nitrogen by volume, dry
baits.
0.44=Molooulnr wolnht of cnrbon dloxldo
divided by 100.
0.33=MoIcculnr weight of oxygen divided
by 100.
0.20=Molocular weight of nitrogen and
CO divided by 100.
6. Re/trenoet.
Altsbuller, A. PH et all., Storage of Oases
and Vapors lu Plastic Bags, Int. J. Air ft
Water Pollution, 0:75-01, 1SC3.
Conner, William D., and J. S. Nader. Air
Sampling with Plastic Bugs, Journal of tho
American Industrial Hygiene Association.
25:201-207, May-Juno 1004.
Devorkiu, Howard, et nl., Air Pollution
Source Testing Manual, Air Pollution Con-
trol District, Los Angeles, Calif., November
1803.
UBTHOD 4—DXTEBMtNATIOK Or UOZ5TUUI
IN STACK CASES
1. Principle and applicability.
1.1 Principle. Molsturo is removed from
the gas stream, condensed, and determined
volumetrlcally.
1.2 Applicability. This method is appli-
cable for tho determination of malsturo In
otack gas only whon specified by test pro-
cedures for determining compliance with Now
Source Performance Standardn. This method
does not apply whon liquid droplets are pres-
ent In tho gas stream' and tho moisture is
subsoquontiy used in tho determination of
stnck gas molecular weight.
Other methods such as drying tubes, wot
bulb-dry bulb techniques, and volumotrlo
condensation techniques may b« used.
3. Apparatus.
3.1 Probe—Stainless steel or Pyrex1 glass
sufficiently heated to prevent condensation
' and equipped with a filter to remove portlou-
lata matter.
3.3 Implngcrs—Two midget tmplngors.
«aoh with 30 ml. capacity, or equivalent.
3.3 Ice both container—To condons*
moisture In Impingors.
2.4 Silica gel tubo (optional)—To protsot
pump and dry gaa meter.
34 Needle valve—To regulate gat flow
rate.
3.0 Pump—Leak-free, diaphragm typo, of
equivalent, to pull gas through train.
3.7 Dry gas motor—To measure to within
1% of the total sample volume.
X8 Botameter—To measure a flow rang*
from 0 to 04 c.lm. •
2.0 Graduated cylinder—25 ml.
3.10 Bolometer—Sufficient to read to
V..' 'within 0.1 Inch Bg.
* 3.1i Pttot tube—Type 8, or equivalent,
attached to probo BO that tbo lampling flow
•If liquid droplet* are present la the gal
itroaro, assume tne itnmm to bo laturatod,
determine the average stock goa temporature
by travenlng according to Method 1, and
aw a psychrometrto chart to obtain an ap-
proximation of tne moisture percentage.
. *Trnde name,
Cliaplor |—Ciwlroiununleil I'rolocllon Aguncy
rnto can bo rogxilatod proportional to tho
iitnck e" velocity when velocity Is varying
with tuno or a samplo travcroo is conducted.
3. Procedure.
3.1 Place exactly 5 ml. distilled water In
each tmplngor. Assemble tbo apparatus with-
out the probo as shown In Figure 4-1. Leak
check by plugf'lng tho Inlet to the flrst Un-
plugor and drawing a vacuum. Insure that
flow through the dry gas motor Is lean than
1 % of the sampling rato.
3.2 Connect tbo probo and sample ai a
App. A
constant rule of 0.076 o.f.m. or at a rato pro-
porlloniil to tho atnck EOS velocity. Conllnuo
uxmpllni; until tho dry gas motor rosters 1
cublo fuot or until vielble HnulU tlroplolu aro
carried over from the flrst Implncer to the
second, necord tompcrnturo, prcsouro, and
dry gas meter roadlncs as required by Flgur*
4-2.
33 After collcctlnR the sample, measure
the volume Increase to the nearest 0.0 ml.
4. Calculation*.
4.1 Voi umo of water vapor collected.
V...
Vvo=Volume of water vapor collected
(standard conditions), cu. ft.
Vi=Final volume of Implnger contents.
ml.
Vi=Initlal volume of Implngor con-
tents, ml.
R=Ideal gas constant, 31.83 inches
He—cu. f t./lb. molo-'B.
.,-, ,
nil. cquiitlon 4-1
pnto=Deuslty of water, 1 g./ml.
T. u= Absolute temperature at standard
conditions, 530* R.
p,i4=Absoluto prcssuro at standard con-
ditions, 29.02 inches Bg.
MDK>= Molecular weight of water, 18 Ib./
Ib.-mole.
FILTER '(GLASS WOOL)
HOTAMETER
CTt
DRY GAS METER
' ICE BATH
Flguro 4-1. Molsluro-sampling train.
-------
APP.A
LOCATION,
TEST
DATE
OPERATOR
Till* 40—Protection of Environment
COMMENTS
BAROMETRIC PRESSURE.
CLOCK TIME
GAS VOLUME THROUGH
METED. |Vm).
fl3
ROTAMETER SETTING
f|3/imn
METER TEMPERATURE.
•f
Figure 4-8. Field moisture dolermlnnllon.
T» ss Absolute temperature at meter (*P-f
400), *R.
3 Moisture content.
17.71
equation 4-2
where!
V» oDry cat volume through meter tt
standard conditions, ecu It.
Vm =Dry gas volume measured by meter,
on. ft.
?. oBarometrte pressure at the dry gas
meter, inches Hg.
P,u=Pre«suro at standard conditions, 3943
B..-7
whcrei
Bwo=
Bra
Absolute temperature at standard
conditions, 630* B. •• • • •
-+(0.026)
>•
equation 4-3
.•Proportion by volume of water vapor
in the gas stream, dlmenslonless.
sVolumo of water vapor collected
(standard conditions), ou. ft.
sDry gas volume through meter
(standard conditions), ou. ft,
{Approximate volumetric proportion
of water vapor in the gas stream
• leaving the implngers, 0.025.
Chapter I—Environmental Protection Agency
APR. A
5. Reference!,
Mr Pollution Engineering Manual, Diuilcl-
son, J. A. (cd.), U.S. DREW. PHS. National
Coater for Air Pollution Control, Olneinnatl,
Ohio. PUS Publication No. B9B-AP-40, 1007.
Oovorkln, Howard, et al., Air Pollution
Source Testing'Manual, Air Pollution Con-
trol District, Los Angeles. Calif., November
1803.
Methods for Determination of Velocity,
Volume, Dust end Mist Content of Oases,
Western Precipitation Division of Joy Manu.
footurlue Co., Los Angeles. Calif., Bulletin
VfP-CO. 1068.
METHOD G—DsmiMWATioir OP PAIITICUUIT*
EMISSIONS Faoic STATIONARY SOURCES
1. Principle and tppUoabfltty.
1.1 Principle. Partloulate matter is with-
drawn laoklnetloally from the source and Ita
weight Is determined gravlmetrloally after re-
moval of uncomblned water.
1.2 Applicability. This method Is applica-
ble for the determination of partlculeto emis-
sions from stationary sources only when
specified by the test procedures for determin-
ing compliance with New source perform-
ance Standards.
3. Apparatus.
2,1 Sampling train. The design specifica-
tions of tho partloulate sampling train used
by EPA (Figure 6-1) ore described In APTD-
OS81. Commercial models of thti train aro
available.
3.1.1 Nozzlo—stainless Bteei (310) with
sharp, tapered lending edge.
3,1.3 Probe—Py rex > glass with a heating
system capable of maintaining it minimum
gas temperature of 250* P. at the exit end
during sampling to prevent condensation
from occurring. When length limitations
(greater than about 8 ft.) are encountered at
temperatures less than 000* F.. Inooloy 825 >,
or equivalent, rimy bo used. Probes for sam-
pling gas streams at temperatures In excess
of 800* 7. must have been approved by the
Administrator.
9.1.3 Pltot tube—Typo 8. or equivalent.
attached to probe to monitor stock gas
velocity.
> Trade name.
3.1.4 Filter Holder—Pyrox1 glau with
henttng system capable of maintaining mini-
mum temperature of 225* V.
3,1.6 Iroplngars / Condenser—Four Impln-
gers connected In series with glass ball Joint
fittings. Tho first, third, and fourth Impln-
gers ore of tho areonburg-Smlth design,
modified by replacing the tip with a '/,-lnch
ID gloss tube extending to one-half Inch
from tho bottom of tho flask. Tito second Im-
plnger Is of tho Greenburg-Smlth dcslcn
with tho standard Up. A condonscr may be
used in place of the Impingors provided that
tho moisture content of tho staclc gtis can
atlll bo determined.
3.1.0 Molcrlnc system—Vacuum gauge,
leak-free pump, thermometers capable of
measuring temperature to within 6' F., dry
gas meter with 2% accuracy, and related
equipment, or equivalent, as required to
maintain an Icoklnctlo sampling rate and to
determine sample volume.
2.1.7 Barometer—To measure atmospheric
pressure to ±0.1 Inches Bg.
23 Sample recovery. '
23.1 Probe brush—At least as long as
probe.
233 Glass wash bottles—Two.
2.2.3 Glass sample storage containers.
23A Graduated cylinder—260 ml. 3>
3,3 Analysis. '
3.8.1 Gloss weighing dishes.
3.3.3 Desiccator.
3.3.3 Analytical balance—To measure to
±0.1 rag.
2.3.4 Trip balance—300 g. capacity, to
measure to ±0.06 g.
3. Reagent*.
3.1 Sampling.
3.1.1 Filters—Glass fiber, MSA 1108 BH«.
or equivalent, numbered (or identification
and prewelghed.
,3.1.2 Silica gel—Indicating type, 8-18
meih. dried at 173' C. (350* P.) for 3 hours
3,1.3 Water.
3,1.4 Crushed Ice.
3,2 Sample recovery.
3.2.1 Acetone—Reagent grade.
3.3 Analysis.
3.3.1 Water.
-------
TOOBE
REVERSE-TYPE
PITOT TUBE
Tillo 40—Protoellon of Environment
IWPIHCF.R TRAIN OPTIONAL. HAY nil REPLACED
BY AN EQUIVALENT CONDENSER
HEATED AREA BLTER HOLDER / THERMOMETER CHECK
\
^VACUUM
LINE
PIT01
ORIFH
IMPINGBB ICE BATH
BV-PASS.VALVE
VACUUM
GAUGE
'ALVE
THERKOKETEI
DRY TEST METER
Figure 5-1. Partlculate-samptlng train.
3.3.9 Deslceant—Dtterito,* indicating.
4. Procedure*
4.1 Sampling
4.1.1 Alter selecting the campling site tad
10 minimum number of sampling points,
itermlne the stack preuura, temperature.
oisturo, and range of velocity head. •
4.13 Preparation of collection train.
elgh to tbe nearest gram approximately 900
of elite* gel. lAbel a filter ot proper dimm-
er, desiccate1 for ot least 24 noun and
tlgh to tho nearest 0.5 mg. In a room where
e relative humidity Is leu than 60%. Place
0 ml. of water In each of the first two
iplngen. leare the third Implnger empty,
'Trade name. . •
• Dry using Drierlte' at 70' F.±10* P.
and place approximately 300 g. of prowelghed
alllea gel In the fourth Impluger. Set up the
train without the probe as In Figure 9-1.
Leak check tbe campling train at the sam-
pling alto by plugging up the Inlet to the fil-
ter holder and pulling a 16 In. Kg vacuum. A
leakage rate not In excess of 0.09 e.fja. at a
vacuum of 16 In. Hg la acceptable. Attach
the probe and adjust the heater to provide a
gas temperature ot about 3SO* P". at the probe
outlet. Turn -on the Alter beating system.
Place crushed lea around the Implngers. Add
more Ice during the run to keep the temper*
aturo of the gases leaving tbe lost implnger
as low as possible and preferably at 70* P«
or less. Temperatures above 70' V. may result
In damage to the dry gas meter'from either
moisture condensation or excessive heat.
Chapter I—-Environmental Protection Agency
App. A
4.1.3 Partlculnto train operation. For cacti
run, record tbo data required on tho cxiunplu
ulicot shown ID Figure 6-3. Take readings at
euch sampling point, at least every 5 minutes,
and when significant changes In stack con-
ditions necessitate additional adjustments
In flow rate. To begin sampling, position tho
nozzle at tho first traverse point with tho
tip pointing dlrcotly Into tbo gns stream.
Immediately start the pump and adjust tho
flow to Isoklnotlc conditions. Sample for at
5 minutes at each traverse point; sam-
pllni! tlinn nuir.t lie tho omnc for each point.
MuliitMii looklnctlc oampuni; throughout the
sampling period. Nomographo are available
which old In tbo rapid adjustment of the
sampling rate without other computations,
AJfTD-0570 details the procedure for using
those nomographs. Turn off the pump at the
conclusion of each run and record tbe Qnal
readings. Remove the probn nncl nozzle from
tho stock and handle In accordance with tho
Bamplo recovery process described In section
4.2.
HAHI ,
lOtAUM .. ., ....
OHIUIC
OIK
MM NO.
tMKI
U.IIIK
•
01 "°1 . . I.
IM.
Kill. II.
e>Ariai_ _ .,
1UVI«1 raiNI
htwalt
IOI>1
lAWtMO
IM
1,1. ~i
1VIHCI
>UIK
Mnw
If ,1. ta. Hi
tuci
imtunM
IV. •«
vuocinr
MAD
Ufjl.
mimic
DDimmiAi
ACMhf
our icc
Witt
1* HI.
I*. MjO
GAiiuni
voiuut
IVM It*
AUVINt ( 1
MMUffll
MSIMCOH
mm* K>
notttiMi
MOlliCOIA
noniuA
OUfAUftl HWIUttM
AtOnCAlUHt
Mir
Umml.'f
•»*.
««.
OUIIII
II«MI.'I
An-
•-1-111111™-™-*
UWIIIOI
tnmuiu*.
•'f
rimuiuu
Of CAl
LUVJK
CMIBI'W
iMitrmcu
•t
4.9 Sample recovery. Exercise enro In mov-
ing the collection train Irom the test site to
the sample recovery area to mitiimtM the
loss of- collected sample or the gain of
extraneous paniculate matter. Set aside a
i portion of tho acetone used In the sample
recovery as a blank for analysis. Measure tho
volume of water from the first three 1m-
plngors, then discard. Place the samples in
• containers as follows: • •• • .••
Container Wo. I. Remove the filter from
'•'• Its bolder, place IB thli container, and seal.
Container Wo. 3. Place loose partloulate
matter and acetone washings from all
sample-exposed suifnces prior to the filter
In this container and seal. Use a razor blade,
brush, or rubber policeman to lose adhering
particles.
Container Wo. 3. Transfer the silica gel
' from the fourth Implnger to tho original con-
tainer and aoat XJso a rubber policeman as
an aid In removing silica go! from the
Implnger.
-------
APP.A
Tltlo 40—Proloetlon of Environment
4,3 Annlyila. Record Uio data required on
the oxninpla sheet shown In Plguro 8-8.
Handle cnob sample cuiilulnar iia fniiuwa:
Container Wo. t, Transfer the Oltor and
tny loose partloulBto mallur from the aamplo
container to a tnrod glow weighing dUtb,
desloeate. and dry to a constant weight. Bo-
port muiu to the nearest OJ mg.
Container Wo. 2. Transfer the acetone
washing* to a tared beaker and evaporate to
'dryaeas at ambient temperature and prei-
nire. Desiccate and dry to a constant weight.
Beport results to the nearest 0.6 mg.
Container Wo. J. Weigh the spent alllca gel
tnd report to the nearest gram.
8. Calibration.
Use methods and equipment which hare
been approved by. the Administrator to
calibrate the orifice meter, pltot tube, dry
gas meter, and probe heater. Recalibrate
alter each test series.
fl. Calculation!.
6.1 Average dry gas meter temperature
and average orifice pressure drop. Bee date
sheet (figure 6-3).
03 Dry gas volume. Correct the sample
volume measured by the dry gas meter to
standard conditions (TO* P., 2953 inches Bg)
by using Equation 5-1.
equation 6-1
where:
V»|U~ Volume of gas sample through the
dry gas meter (standard condi-
tions) , cu. ft.
V_— Volume of gas sample through the
dry gas meter (meter condi-
tions) , cu. ft.
T.,.— Absolute temperature at standard
conditions, 030* R.
*„•• Average dry gas meter temperature,
Ft,,** Barometric pressure at the orifice
meter, Inches Hg.
4H«- Average pressure drop across the
orifice meter, Inches H,O.
ISA** Specific gravity of mercury.
P..(— Absolute pressure at statulard con-
ditions, 39.02 Inches Bg.
03 Volume of water vapor.
"»•
conditions) *
cu.lt.
• Vi,—Total volume of liquid collected In
Implngera and silica gel (sea fig-
ure B-S), ml.
pi^>-* Density of water, 1 g,/ml.
Miv-Moleoular weight of water, 18 In./
tb.-mole.
a—Ideal gas constant, 91.83 Inches
Bg—cu. f Wlb.-molB-'B.
T.u»Absolute temperature 'at standard
TP'I'tl""*, 630* R.
P.u— Absolute pressure at standard con-
ditions, MM inches Hg.
0.4 Moisture content.
Bw»™i
aquation 5-3
11.. •Proportion by volumtoi water vapor In tbogu
in, dlmwulonlMO.
, .
trt-Votnme ol water In Uio KOI sample (ttandud
oandllloni), cu. It.
Vl*iU«Voliiinool ens umplo through the dqr gat miter
(flandiudcaiidlUoin), cu. ft.
9& Total particulate weight. Determine
Uio VotoS particular catch from the eum of
the weights oa the analysis data sheet
(Figure 6-3).
0.0 Concentration.
0.0.1 Concentration In gr./s.o.f.
equation 6-4
wb«rt:
O'l-Conetntratlon ol partleulaU mnttw in MMk
B**j ST./IA!., dry boili.
M.-TOU! amount ol ptrtloalato matter eellocted,
tag.
Chapter I—Environmental Protection Agoncy
PLANT
DATE
App. A
RUN NO..
CONTAINER
NUMBER
WEIGHT OF PARTICULATE COLLECTED,
mg
FINAL WEIGHT
TARE WEIGHT
WEIGHT GAIN
vo
FINAL
INITIAL
LIQUID COLLECTED
TOTAL VOLUME COLLECTED
VOLUME OF LIQUID
WATER COLLECTED
IMPINGER
VOLUME/
ml
SILICA GEL
WEIGHT.
9
9*j ml
CONVERT WEIGHT OF WATER TO VOLUME DY DIVIDING TOTAL WEIGHT
INCREASE BY DENSITY OF WATER. (1 g. ml):
VOLUME
Figure 5 -3. Analytical data.
70
-------
APP- A Tlrlo 40—Protection of Environment
8.8.3 Concentration In Ibycu. ft.
V-..4
'e,-Cone«ntf»tlon of partleolstt mittv la itart
equation 6-5
M,-Total amount e' pirtlcutato matter eollfoted
V,lU-Volume of cu tainpte thrmiitlt dry cu motet
(ttsndard oondllloiii), cu. ft.
8.7 Isokinetlo variation.
•him:
I-
-Perctutoflsoklnetlc umpllDt.
•Total-velum* of liquid collec
nnd Jlllcn ccl (Sco Flu. »-3). ml.
m,n»UeiultyoCwutor, iR./ml.
collected In Imptafdi
R-Id«JraeoiWiint. SI.8J Inehei Il|-ca. Mb.
ir«,o-MoleouIariri>lFlit of water, 13 Ib Jlb.-moU.
V.-Volumocf fas •ampin Uiroupli Hie dry tai motor
(inrtarcvmhlliinid.cu. ft.
T«-Absolui« IIVCRWO dry ga niflur tampentur*
Piu-niiroinrtrlt prcsiuro ill sampling site. tithes
ATI -Arwico Pressure drop tcroit tho orlO« (u*
FlR.»-S),lnchwIIi6.
T,-Alj»lute «Ycrm:e Mack cm tempo-Blurt (set
«-Tolal'v\nii!llnr tlm», inhi.
V."Slsc!c f,a velocity rnlculoted by Method J,
Kfliiniionj-2.ll.ftw.
P.- Absolute jtnrk irni pn-jjurn. Inrhrs )lg.
A.-Cron-irctlannl urr.i of noult, si. ft.
0.8 Acceptable results. The following
range sets the limit on acceptable Isoklnetle
sampling results:
If 80% «g 110%. the results are acceptable;
otherwise, reject the result* ana repeat
tho ten.
7. nefcrenee.
Addendum to Specifications for Incinerator
Testing at Federal Faculties. PHS, NCAFC.
Deo. 0. 1007.
Martin, Robert M., Construction Details of
Zsoklnetle Source Sampling Equipment, En-
vironmental Protection Agency, APTD-5081.
Rom. Jerome J., Maintenance, Calibration
and Operation of Isoklnetle Source Sam-
pling Equipment, Environmental Protection
Agency. APTD-067B.
Smith. W..8.. n. T. ShlBehora, and W. R
Todd. A Method of interpreting Stack Sam-
pling Data. Paper presented at the OSd An-
nual Meeting of tho Air Pollution Control
Association. St. Louis, Mon June li-lB, 1070.
Smith, W. 3.. ot at., Stack das sampling
Improved and Simplified with New Equip.
tnant, APOA paper No. 07-110, 1007.
Specifications for Incinerator Testing at
federal Facilities, PUS, HCAPA, 1007.
cquation 8-0
HZTHOD S—DETKSMINATIOM OF BOLruK DIOXIDE
nCBSIONS FROM STATIONARY 8ODBCES
1. Principle and applicability.
1.1 Principle. A gas sample Is extracted
from the sampling point In the stack. The
add mist. Including sulfur trtojdde. Is sepa-
rated from the sulfur dioxide. The sulfur
dioxide fraction is measured by tho barium*
tbortn titratlon method.
1.3 Applicability. This method Is Appli-
cable for the determination of sulfur dioxide
emissions from stationary sources only when
specified by the tost procedures for determin-
ing compliance with New Source Performance
Standards.
3. Apparatus.
3.1 Sampling. See Figure 0-1.
3.1.1 Probo—Pyrex' gloss, approximately
8 to 0 nun. ID, with a heating system to
prevent condensation and a altering medium
to remove paniculate matter including sul-
furlo acid mist.
3.1.3 Midget bubbler—One, with glass
wool packed In top to prevent sulfurlo aoid
mist carryover.
3.1.3 Glass wool.
9.1.4 Uidget Iraplngors—Three.
9.1.8 Drying tube—Packed with e to 10
mesh indicating-typo silica gol, or equivalent,
to dry the sample.
3.1.0 Valve—Noodle valve, or equivalent,
to adjust flow rato.
3.1.7 Pump—leak-free, vacuum type.
9.1.8 Rate meter—Rotamoter or equiva-
lent, to measure a 0-10 s.o.f.h, flow range.
34.0 Dry gas meter—Sufficiently accurate
to measure the sample volume within 1%.
9.1.10 Pltot tube—Type S, or equivalent,
necessary only if a sample traverse is re-
quired, or if stock gas velocity varies with
33 Sample recovery.
3 J.I GUasa wash bottles—Two.
3.13 Polyethylene storage bottles—To
•ton implngor samples.
3* Analysis.
' Trade names.
i
OA
Choptor I—Environmental Protection Agency App. A
SILICA GEL DRYING TUBE
STACK WALL
PROBE (END PACKED
WITH QUARTZ OR
PYREX WOOL1 "• ^ MIDGET BUBBLER MIDGET IIWPINGEAS
. J GLASS WOOL
TYPE S WOT
THERMOMETER
PUMP
DRV GAS METER ROTAMETER
Figure 0-1. SOj sampling train.
3.8.1 Pipettes—Transfer type. 6 ml. and
10 ml. sizes (0.1 mL divisions) and 26 mL
tlze (0.3 ml. divisions).
3.3.3 Volumetric flasks—EO ml., 100 mL.
and 1.000 ml.
2.3.3 Burettes—5 mL and 60 ml.
3.3.4 Erlenmeyer flask—135 ml.
3. Reagents.
S.I Sampling,
8.1.1 Water—Delonlxed, distilled.
$.1.9 Isopropnnol, 80%—Mix 80 ml. of Iso-
propanol with 20 ml. of cttittlled water.
3.1.3 Hydrogen peroxide, 9%—dilute 100
ml. of 30% hydrogen peroxide to 1 liter with
distilled wntcr. Prepare fresh dally,
8 J Sample recovery.
8.3.1 Wntcr—Delonlxed, distilled.
3.3.3 Isopropanol, 80%.
3.3 Analysis.
3.3.1 Water—Delonteed, distilled.
3.3.3 Isopropanol.
34.3 Thorln Indicator—l-(o-arsonophen-
ylnzo)-3-naphthol-8,e-dlsulTonlo acid, dlso-
dlum salt (or equivalent). Dissolve 0.20 g. in
100 ml distilled water.
8.3.* Barium pnrohlorata (0.01 H)—Dis-
solve l.OS g. of bnrtum perohlorate
|Ba(01O,),«3HtOJ In 200 ml. distilled water
and dilute to 1 liter with Isopropanol. Stand-
ardize with sulfurlo odd. Barium chloride
uny bo used. .
14.6 SuUurio nold standard (0.01 W)—
Purchttie or standardise to ±0.0003 N
affMnst 0.01N NaOH which bos prevloxisly
been stnndordleed against potassium add
phthalats (primary standard grade),
4. Procedure,
4.1 Suuipllng.
4.1.1 Preparation of collection train. Pour -
16 ml. of 80% laopropanol Into the midget
bubbler and 16 ml. of 3% hydrogen peroxide
Into each of tho first two midget Implneers
Lcavo tbo final mtdgat Unplnger dry. Assem-
ble tho train OB shown In Ficure 0-1. Leak
chock tho sampling train at the sampling
site by plugging tho probe Inlet and pulling
a 10 inches Be vacuum. A leakage rato not
In excess of 1% of the sampling rate ls ac-
ceptable. Carefully release tbo probe Inlet
plug and turn off the pump. Place crushed
Ice around the Implngcrs. Add more Ice dur-
ing the run to keep the temperature of the
Cases leaving tho lost Implnger at 70* 7. or
Ic&s.
4.1.8 Sample collection. Adjust the siun-
plo flow rato proportional to tbo stack gut
velocity. Take readings at least every five
minutes and when significant changes in
stack conditions necessitate additional ad-
justments in flow rato. To begin sampling,
position tho tip of the probe at the first
sampling point and start the pump. Sam-
ple proportionally throughout the run. At
the conclusion of eaeh run, turn off the
pump and record the anal readings. Rcmovs
the probe from the stnck nnd disconnect it
from tho train. Drain the tea bnth and purge
the remaining port of the train by drawing
claim ambient air through the syetom (or 16
ml mites.
44 Sample recovery. Disconnect tbo Im.
plngers after purging. Discard tho contents
of the midget bubbler. Pour the contents of
the midgot Implngers Into a- polyethylene
shipment bottle, ninoo the three midget Im-
nnd tlin connecting tubes with dls-
ui
-------
B-l
APPENDIX B
HIGH-VOLUME TEST METHODS
-------
B-2
OREGON DEPARTMENT OF ENVIRONMENTAL QUALITY
AIR QUALITY CONTROL DIVISION
May, 1972
STANDARD SAMPLING METHOD
DETERMINATION OF PARTICULATE EMISSIONS FROM CYCLONES
(High Volume Method)
1. Principle and Applicability
1. 1 Particulate matter is withdrawn from the source under isokinetic
conditions as a component of the flowing gas stream. The particulate
is removed from the sample stream by filtration through a glass fiber
filter. Particulate mass is determined gravimetrically.
1. 2 This method is applicable for the determination of particulate
emissions from cyclones exhausting directly to the atmosphere.
1. 3 It is recognized that this sampling method is not necessarily consistent
with other standard methods of source testing. The Department of
Environmental Quality and other agencies may re-evaluate this method
in comparison with other methods, as more data becomes available,
and will revise this method as required.
2. Range and Sensitivity
2. 1 The range of the method is dependent upon the sampling time and
flow rate. To obtain the minimum required sample weight of 100 mg.
on a filter in the minimum acceptable time of 15 minutes, the particulate
concentration must be at least . 002 gr. /scf. when sampling at 50 cfxn,
or 0. 02 gr. /scf. at 5 cfm. The maximum practical loading on the
filter is not known at this time.
2. 2 The sensitivity of the method is better than 1. 0% of the measured
concentration, based on the ability to discriminate an increment of
sample weight of 1. 0 mg. in a sample of at least 100 mg.
3. Interferences
3. 1 Particulate matter present in a gaseous phase at the filtration temperature
will probably not be collected. This method should not be used if
significant quantities of condensible particulate matter are expected,
unless the temperature of the sample gas can be reduced to approximately
70* prior to filtration.
4. Apparatus
4. 1 Sampling - A schematic diagram of a typical sampling train is shown in
Figure 1. The recommended design of this device is described in Reference
10. 1, and commercial models are available.
-------
B-3
-2-
4. 1. 1 Nozzle - The sampling nozzle shall be made of metal, and shall
be sharp-edged. Nozzle diameter shall be such that isokinetic
conditions can be maintained at flow rates to be used on each
test. The typical range of nozzle sizes is from 1/2 inch to 2
inches in diameter. The nozzle shall be connected to the probe
by means such that deposition of particulate matter in threads
or joints is minimized.
4. 1. 2 Probe - The probe shall attach to the filter holder either directly
or by the shortest possible length of smooth-walled tubing.
4. 1. 3 Filter Holder - The filter holder shall be air tight. A quick
disconnect assembly is recommended for ease of changing filters.
4. 1. 4 Metering System - The filter holder shall be followed by a
calibrated orifice, a thermometer or thermocouple, a flow
control and a high volume blower capable of 60 cfzn free air capacity.
4. 2 Sample Recovery
4. 2. 1 Probe Brush - It should be of a length at least equal to that of the
probe and any tubing connecting it to the filter holder.
4. 2. 2 Clean manila envelopes for handling and storing filters.
4. 2. 3 Wash bottles and storage containers for liquid samples.
•'. 1 Analysis - The laboratory in which samples are to be analyzed shall
include standard laboratory equipment such as glass weighing dishes,
an analytical balance accurate to the nearest milligram, and other
necessary equipment.
5. Reagents
5. 1 Sampling
5. 1. 1 Filters - Glass fiber filters, type A, type E, or equivalent,
shall be used. Prior to sampling, each filter shall be exposed
to a lightsqurce and inspected for pinholes, particles, or other
imperfections. Filters with visible imperfections shall not be
used. A small brush is useful for removing particles. Filters
shall be pre-conditioned by equilibrating at 20-24 C, 50% relative
humidity or less, for a minimum of 2 hours. Filters shall be
numbered for identification, and p re-weighed to the nearest 1
milligram. Filters shall not be folded before collection of the
sample.
5. 2 Sample Recovery - Reagent grade acetone, methanol, or other suitable
solvent shall be used for cleaning up the sampling train.
6. Procedure
6. 1 Determination of Air-Flow Rates and Properties
6. 1. 1 A sampling site for determining system flow rate shall be
selected in the approach-duct to the cyclone. The point
selected shall be as close as possible to the ideal location described
in standard source sampling literature (e. g. , eight diameters
downstream, and two diameters upstream, from the nearest obstruction
or bend in the duct), keeping in mind the necessity of having an
accessible location. The cross section of the duct shall be divided
into equal areas and a velocity traverse conducted according to standard
sampling methods (Reference 10. 2).
-------
F" E 1
SCHEMA- DIAGRAM
TYPICAL HIGH VOLUME PARTICULATE SAMPLING TRAIN
COMPONENTS:
1. Attached pitot tube - "P" type or "S" type.
2. Nozzle
3. Probe
4. Pressure gauge or manometer
5. Filter holder
6. Calibrated orifice
7. Pressure gauge or manometer
8. Thermometer or thermocouple
9. Control valve or damper
10. Optional flexible coupling
11. High volume blower
CO
-------
B-5
-3-
6. 1. 2 Either an S-type or P-type pitot tube may be used in conducting
a velocity traverse, according to Reference 10. 2.
6. 1. 3 Temperature, static pressure, and moisture of the gas stream
shall be measured in order for duct flow rate to be corrected
to standard conditions.
6. 1.4 Record data on Form 2 of Appendix I, or equivalent.
6. 1. 5 A pitot traverse is probably impractical for high pressure
pneumatic conveying systems, in which case it is permissible
to use manufacturer's data relating air flow to pressure drop
and rpm at the blower.
6. 2 Sampling
6. 2. 1 Velocity Survey at Cyclone Exhaust
6. 2. 1. 1 Use a pitot tube to roughly map the velocity
distribution across the cross section of the
cyclone exhaust.
6. 2. 1. 2 At each point at which velocity is measured,
measure the flow in the direction that gives
maximum deflection on the manometer or
pressure gauge. Record data on Form 3,
Appendix I.
6. 2. 1.3 Select six points at which emissions will be
sampled. Each of these points shall be in an
area of positive (out-going) flow. One point
shall be near the point at which maximum
velocity occurs. The points shall provide
a representative sample of the flow pattern,
and shall be numbered and referenced on the
sketch of the exit cross section flow pattern.
If six points with positive flow cannot be obtained,
use the maximum number possible.
6. 2. 2 Preparation!of Sampling Train - The following steps shall
be conducted prior to each run.
6. 2. 2. 1 All parts of the sampling train shall be cleaned
and properly calibrated as directed in Paragraph 7.
6. 2. 2. 2 Place a filter in the filter holder, coarse side
facing the flow, being careful not to damage it.
6. 2. 2. 3 Perform a leak check by plugging the nozzle, turning -
on the blower, and observing the deflection, if any,
on the flow orifice pressure gauge. Leakage shall
not exceed 5% of the expected sample flowrate.
6. 2. 3 Sample Collection
6. 2. 3. 1 With the probe out of the exhaust stream, turn on
the blower and adjust the flowrate so that the velocity
at the sampling nozzle corresponds with the velocity
at first point to be sampled in the cyclone exhaust.
When the correct flowrate has been established, turn
off the blower. Note- This step should not be done
if the local environment is extremely dusty and there
is danger of extraneous particulate contaminating
the sample. The same applies to similar steps Below.
-------
B-6
-4-
6. 2. 3. 2 Locate the probe at the first point to be sampled and
move it around until the velocity pressure matches
that for jA'hich the sampling flowrate was pre-set.
6. 2. 3. 3 Turn on the blower and sample for the desired period
of time. The sampling period at each point should
be such that total run time for the cyclone is at
least 15 min.
6. 2. 3. 4 Continually monitor velocity during the sampling
period and move the probe around as required to
keep it in an area where the velocity matches the
rate needed to match the pre-set sampling flowrate.
Record on the data sheet (Form 4, Appendix I) the
temperature and pressure drop at the orifice meter.
6. 2. 3. 5 At the conclusion of the sampling period for the
first point, move the probe to the next point and quickly
readjust the flowrate to the previously established
isokinetic rate. In the event conditions make it
impossible to adjust the flowrate rapidly enough, the
probe should be removed from the cyclone exhaust
and the rate pre-set as in 6.2. 3. 1. Note on the data
sheet (Form 4, Appendix I) the exact time of the sample
period.
6. 2. 3. 6 Repeat steps 6. 2. 3. 1 through 6. 2. 3. 5 until all points
are sampled. If excessive loading of the filter should
occur or the pressure drop should increase such that
isokinetic conditions cannot be maintained, replace the
filter and continue the test.
6. 2. 3.7 Extreme caution should be taken that the nozzle does
not touch the walls of the cyclone. Doing so may
dislodge the deposited material from the wall and cause
it to enter the sampling nozzle, thus invalidating the
sample. If there is reason to believe this has happened,
discontinue the sample, clean the train and start over again.
6. 3 Sample Recovery
6. 3. 1 Immediately upon removing the probe from the stack after completing
the final point, plug the nozzle uatil it can be cleaned. Take the sampler
to a reasonably clean area, turn on the blower, insert the probe
brush into the probe and brush the particulate from the nozzle and
probe into the filter. *Using a preweighed wash solution, rinse the
probe section into a clean container.
6. 3. 2 Open the filter holder, and use a fine brush to brush any partieulate
matter deposited on the front side of the holder onto the filter. Fold
the filter once length-wise, with the dirty side in, and place in a
folded «majj}pa tagboard, folded edge down. Put a paper clip on the
outside edge of the tagjboard, and place in a manila envelope.
-------
B-7
-5-
6. 3. 3 At the conclusion of testing of each cyclone, or more frequently
if desired, wash the inside of.the nozzle, the probe, the front
half of the filter holder and the probe brush with solvent. Place
the •washings in a labeled container for gravimetric analysis.
6.4 Analysis
6. 4. 1 Filter - Equilibrate the sample for at least 16 hours at 68-75° F
(20-24° C) and 50%, or less, relative humidity. In the case of
extremely wet particulate, oven drying at 150° F (65° C) may
precede equilibration. Weigh the filter to a constant weight to the
nearest mg.
6. 4. 2 Solvent Wash - Transfer the solvent washings to a tared beaker,
and evaporate to dryness at room temperature and pressure.
Alternatively, the solvent may be evaporated in an oven at 150 F
(65° C) or less. Equilibrate for at least 16 hours. Weigh to a
constant weight and report the results to the nearest mg.
6. 4. 3 Blanks - At least one filter for each four filters used in the
field shall be inserted in the filter holder, a leak check performed,
and removed and returned to the laboratory for analysis as a
blank. A portion of the solvent used for field clean-up shall
also be analyzed as a blank. Results from field samples shall
be adjusted according to the blank values.
alibration
The pitot tube, orifice flowmeter, pressure gauges and temperature measure-
ment devices shall be calibrated at least once a year against a primary standard
or a device which has been calibrated against a primary standard. The date and
method of calibration of these instruments shall be recorded on Form 1, Appendix I.
8. Calculations
Total particulate emissions from the cyclone shall be calculated by multiplying the
particulate concentration measured at the cyclone exhaust by the flow measured at
the inlet duct.
8. 1 Particulate Concentration - The following calculations shall be conducted
for each run.
9. 1. 1 Total Sample Weight - Calculate the total sample weight from
laboratory results by adding the net weight gain of filter samples,
adjusted for blank value, to the net weight or particulate matter in
the acetone washings. If the solvent washings represent more
than one run, they should be pro-rated for each run according to
the relative net weights of particulate collected on the filters. Record
results on Form 4, Appendix I.
8. 1. 2 Total Sample Air Volume - Calculate the sample volume for each
sample point by multiplying the duration of the sample in minutes,
times the average flowrate (cfxn). Add the volume of all sample
points to get the total sample air volume for the run. If each
point was sampled for an equal period of time, the total flow can
-------
-6-
be calculated as simply the total sample time multiplied by the
average flowrate for all sample points. Flowrate for each point
shall be determined from the calibration curve for the flow orifice,
corrected to standard temperature and pressure. Express the
results in the space provided on Form 4, Appendix I to the nearest
cubic foot, both on a wet basis (cu. ft.} and on a dry basis (scf. ),
using 60 F and 29. 92 in. Hg. as standard conditions.
8. 1. 3 Calculate the particulate concentration in gr. /scf. by the following
equation:
Cg= 0. 0154 x W
Where Cg = Particulate concentration, gr. /scf.
W= total particulate sample weight, mg.
Q= total volume of gas sample, scf.
Record this result in the space provided on Form 4 and Form 5,
Appendix I.
8. 2 Total Flowrate
Use data from the velocity traverse of the approach duct to calculate flow
through cyclone in scfm, using the tabular computing equations in Form 5,
Appendix 1. For some cyclones, the total flow may be adjusted to account
for air purposely vented out the bottom of the cyclone.
. 3 Total Emissions
Calculate the total particulate emission in Ib/hr by the following equation,
using Form 5, Appendix I:
E » . 00857 (CgHQa)
Where E = total emission, Ib/hr
Qa= total cyclone flowrate, scfm.
8.4 Percent Isokinetic
Use the tabular computing equations in Form 5, Appendix I to compute the
percent isokinetic (1), defined as the ratio of the average velocity of the
sample gas entering the sample nozzle to the average local velocity at the
sampling points. In order to achieve acceptable results, the value of this
parameter must be between 82% and 120%. Test results falling outside this
range shall be discarded and the test repeated.
9. Test Reports
The following outline shall be considered the minimum acceptable contents of a source
test report for a cyclone or group of cyclones at a plant site.
I. Introduction and Summary of Results and Conclusions.
II. Description of Source(s) - may be in tabular form for a large number of
cyclones.
A. A plant site plot diagram and a process flow diagram in which each
cyclone is clearly identified.
B. Process equipment involved - type, size.
C. Process material flow rates, fuel rates, etc.. .include assumptions
used in evaluating process variables.
D. Cyclone system design, type, size, cfm. etc.
E. Special conditions occurring during the source test period.
-------
-7-
III. Sampling and Analytical Methods.
A. Field equipment - include dates of calibrations.
B. Field procedures - describe deviations from the standard method,
if any.
C. Analytical methods - describe deviations from the standard method,
if any.
D. Special problems or considerations.
IV. Detailed results - may be in tabular form for a large number of cyclones.
A. Emissions in gr. /scf. and in Ib/hr.
B. Gas volume, temperature and moisture content.
C. Percent isokinetic sample rate.
D. Other results - particle size analysis, chemical analysis, or other
optional data that may have been obtained.
V. Appendix
A. Forms, 1, 2, 3, 4, and 5, or equivalent.
B. Other field or laboratory data.
10. References
10. 1 Boubel, Richard W. , "A High Volume Stack Sampler, " APCA Journal,
Vol. 21, No. 12, December, 1971.
10. 2 "Methods for Determination of Velocity, Volume, Dust and Mist Content
of Gases, " Bulletin WP-50, Western Precipitation Group, Joy
Manufacturing Company,
-------
C-1
APPENDIX C
REFERENCE METHOD 6
-------
APP.A
Till* 40—Protection of Environment
tilled: water and add tbme washings to the
tame storage container.
4J Sample analysis. Tronafor the content*
of the itorago container to • 60 ml. voiu-
metrto Hook. Dilute to tho mark with de-
lonlxed. distilled water. Plpelte e> 10 ml.
aliquot of this tolutton into a 125 ml Erten-
meyer flask. Ad4 40 ml. of Isopropanol awl
two to tout drain of taorln indicator. Titrate
to a pink endpolnt using 0.01 N barium
pefchlorate. Bun a blank with each series
of samples.
8. OoHbrotfcm.
6.1 Use standard method! and equipment
which bam been approved by the Adminis-
trator to calibrate the rotameter, pltot tub*,
dry IM meter, and probe neater.
U, atandardUw the barium perehlorato
against 95 ml. of standard •ulfnrio add con-
taining 100 mL of Isopropanol.
«. Oofculatloiu.
8.1 Dry gas volume. Correct the umple
volume BMaiurM by the dry gas meter to
where:
Osv» Concentration of sulfur dioxide
at itandard conditions, dry
basis, Ibycu. ft.
7.03X10-*—Conversion factor, Including the
number of grams per gram
equltalent of sulfur dioxide
(89 gVg.-eq.), 463.8 gylb, and
1.000 ml/L. lb.-U/g.-ml.
V,—Volume of barium perohlorate
Utrant uied for toe sample,
ml.
V .•.volume of barium perohlorate
Utrant used for the blank, ml.
N—Normality of barium perchlorate
Utrant, g.-eq./l.
V..I.—Total tolntlon volume of sulfur
dioxide, 60 mL
V.—Volume of (ample aliquot ti-
trated, ml.
V«,,4—Volume of gas sample through
the dry gM meter (standard
eondltloni), eu. ft* set equa-
tion 8-1.
7. Jteftreneet.
Atmospheric Emissions from Bulfurlo Add
Uanufseturing Processes, 0JS. DBEW, PBS,
Division of Air Pollution, Public Health Serv-
ice Publication No. BB9-AP-13, Olnolnnatl,
Ohio, 1886.
Oorbttt, P. f.. The Determlnstlon of SO,
and SO: In Flue Oases, Journal of the Insti-
tute of Puel. 94:8*7-343,1881.
Matty, R. & and X. R. Dleht, Meaiurlng
Plue-Ou SO, and SOr Powir /Oii04-87, No-
vimber, 1867.
Patton, W. P. and J. A. Brink, Jr., New
Equipment and Techniques for Sampling
Chemical Process dues, J. Air Pollution Con-
trol Amoclatlon. 13. in (1888).
standard condition* C70* P. and 30.03 Inches
Hg) by using equation S-l.
Pb
17.71
in. Ug
equation 6-1
where:
V«.a— Volume of gai umple through the
dry gai meter (standard condi-
tions), en. ft
7.,—Volume of gas umple through the
dry gai meter (motor condi-
tion*), cu. ft.
T.u—Absolute temperature at itandard
eondltloni. 630* B.
TM~ Average dry gas meter temperature^
PWr—Barometric prenure at the orifice
meter. Inches Bg.
P,a— Absolute pressure at standard con-
ditions, MM inches Hg.
93 Sulfur dioxide concentration.
equation 6-2
METHOD T—OmSMtWiTIOM Of NTrSOOXN OXBX
VCOM STATMMABT SOUSCES
1.1 Prlnclplo. A grab sample Is collected
In an evacuated flask containing a dilute
eulfurlo aeM-bydrogen peroxide absorbing
solution, and the nitrogen oxides, except
nitrous oxide, are measure ootorimeMeally
using the phenoldlsulfonlo add (PUS)
procedure.
Id Applicability. This method ls applica-
ble for the measurement of nitrogen oxidai
from stationary soureei only when specified
by the test procedures for determining com-
pliance with New Source Performance
Standards.
9. Apparatus
9.1 Sampling. See Figure 7-1.
9.1.1 Probe—Pyrex' glut, heated, with
filter to remove psrtleulste matter. Beating
le unnecessary If the probe remain* dry dur-
ing the purging period.
9.1.9 Collection flask—Two-liter, Pyrex,1
round bottom with short neck and 94/40
standard taper opening, protected against
Implosion or breakage.
9.1.8 Plask valve—T-bore stopcock con-
nected to a 34/40 standard taper Joint.
9.1A Temperature gauge—Dill-type ther-
mometer, or equivalent, espable of meaiur-
Ing 9* P. intervals from 96* to 198* P.
a.1.8 Vacuum line—Tubing capable of
withstanding a vacuum of 8 inches Bg abso-
lute pressure, with "T" ooanectloa and T-bore
stopcock, or equivalent.
a.1.8 Prenure gauge—U-tube manometer,
so inches, with O.l-lnoh divisions, or
equivalent.
• Trade name.
Chapter I—Environmental Protection Agoncy
App.
3.1.7 Pump—Capable of producing a Tac-
iturn of 3 Inchea Ug nbwluta pressure.
9.1.8 Squeeze bulb—One way.
a.2 Sample recovery.
3.2.1 IMpotto or dropper.
3.3.3 Glass atorage containers—Cushioned
for ihlpplng.
OJKMNO-GMMIOCI
5 NO. 12
MOM. i. ma.
fUNOMO I AMI
{ SUIVI NO. M/M
M"l« IOUHO«inOU. IHQUTWOl.
•II, JBIIVI NO. «•«
O
/,
Fljurc7-l. Sampling luln
Glass wash bottle.
9.3 Analysis.
33.1 Steam bath.
9.8.9 Beakers or casseroles—960 ml., one
for each sample and standard (blank).
333 Volumetric pipettes—1.9. and 10 ml.
944 Transfer pipette—10 ml. with 0.1 ml.
divisions,
9.8.8 Volumetric flask—100 mU one for
each sample, and 1,000 ml. for the standard
(blank).
9,3.8 Speetropbotometer—To measure ab-
lorbnnce nt 430 nm.
92.7 Graduated cylinder—100 ml. with
1.0 ml. divisions.
33.6 Analytical balance—To measure to
0.1 mg.
8. Retgenti.
8.1 Sampling.
9.1.1 Absorbing solution—Add 9.8 mL of
eonoentrsted B£O. to 1 liter of distilled
waiter. Mix well and add 8 mL of 8 percent
hydrogen peroxide. Prepare a fresh solution
weekly and do not expose to extreme heat or
direct sunlight.
8 J Sample recovery.
32.1 Sodium hydroxide (IN)—Dissolve
40 g. NoOH in distilled water and dilute to 1
liter.
8.9.9 Red litmui psper.
3.9.8 Water—Delonlaed, distilled.
3.3 Analysis.
3.3.1 Fuming STUTurtc add—is to 18% toy
weight free sulfur trtoxlde.
3.8.9 Phenol—White solid reagent grade.
933 BuJTurlc add—Concentrated reagent
grade.
3.3.4 Standard solution—Dissolve 0.6496 g.
potassium nitrate (KNO,) In distilled water
and dilute to 1 liter. Por the working stand-
ard solution, dilute 10 ml. of the resulting
solution to 100 ml. with distilled water. One
ml. of tlie working standard solution ls
equivalent to 2S14. nitrogen dioxide.
8.».0 Water—Dolonlccd, distilled.
3J.O Phenoldlsulfonlo add solution—
Dissolve 96 g. of pure white phenol in 160 ml.
concentrated sulfurto add on a steam bath.
Oool, add 78 ml. fuming sulfurio add, and
heat at 100* O. for 9 hours. Store la a dork,
stoppered bottle.
4. Procedure.
4.1 Sampling.
4.1.1 Pipette 98 ml. of absorbing solution
into a sample flask. Insert the Bask valv*
stopper into the flask with the valve in the
"purge" position. Assembls the sampling
train as shown la Figure 7-1 and place the
probe at the sampling point. Turn tho flask
valve and the pump vdve to their "evacuate"
positions. Evacuate the flask to at least 8
Inches Kg absolute pressure. Turn tho pump
63
-------
App. A Tlllo 40—Protection of Envlronmont
«.fl.3 Concentration In ibycu. ft.
Vb»«:
(••Cenemlrstlofi of partloulata maltar In Hack
Ilg-cu.ft
r-T r/0-
T-IA
«d aquation 0-5
M.-Tolal Amount e' parllculnto manor eollMtod
V»(U-Volume of KM mnpla tlirniiKli dry JM molar
(ilnildtirj condllloiu). (Ml. ft.
8.7 Isoklnetio variation.
H
I-Pwe«it ofboklnttlo umpllnii.
V|,-Tqi»l-»olum« of liquid collKttd In Urplngcn
nnd illlrn eel (Sro Fly. »-»}, nil.
Mi,n-I>ci»lty of w.iior, I r./inl.
R>Idi«l r-n constant, 51.83 Inches II»-«u. lt.flb.
M 11,0- Molecular vHfliloJwatfr, IS IMb.-mola.
V,-Volmiiotf(hWKiinpln lluouph the Urv eu motor
(ni«nr ruiiiluMw), cu. ft.
.-.Misoluu nver.ti.-a U.y pui mrlur lunpanture
'MRiire-'J,t.
Pk»-l)uiaiiirlr|e prntura lit mnitillnu nlln, IncliM
•I |f*
MI»Aymic« nrtttura tlren acroM ilia orlH« (IN
T.-AliMl'utt tvtreca Mack can tamprrntur* (IM
r U. fr-I), *H.
"-Toi.il miupllne lima. inln.
V.nSiack f>a v<&irliy ralcnloted by Method J,
K(|ii:iiion S-2, ft.&ru.
P.-Alnnliitrilurk irn» primurn. liirhn lie.
A.-Cn.«.j«llonnl urn of noulg, sq. fl.
0.8 Acceptable results. The following
range cets the limit on acceptable Iioklnetle
sampling results:
U 60% £ 110*. the results are acceptable;
otherwise, reject toe remit* and repeat
the teit.
7. Reference.
Addendum to Specifications for Incinerator
Testing at Federal Faculties. FHS, NCATC,
Deo. 0. 1007.
Martin, Robert M., Construction Detain of
Xioklnatlo Source Sampling Equipment, En-
Vlronmentol Protection Ageotsy, APTD-BOB1.
Rom, Jerome J., Maintenance, Calibration
and Operation of Xsoklnetle Source Sam-
pling Equipment. Environmental Protection
Agency, APTD-0578.
Smith, W. 6., n. T. Shlffehara, and W. 7,
Todd. A Method of mtorprotlng stack Sam-
pllng Data, Paper presented at the 03d An.
nual Meeting of the Air Pollution Control
A«oclallon. St. Louis, Mo., June 14-18, 1070.
Smith, W. 3., ot at.. Stack Qas Sampling
Improved and Slmplinod with New Equip-
ment. APOA paper Ho. 07-110, 1007.
Specification* for Incinerator Testing at
Federal Faculties, FHS, NOAPA, 1007.
equation 8-6
MTTTCOD S—DrrmMIHATION OP BULrUK DIOXIDE
EiasaiONS rnou STATIONARY SOURCES
1. Principle and applicability.
1.1 Principle). A gas sample Is extracted
from the sampling point In the stack. The
acid mist. Including sulfur trtojdclo, is sepa-
rated from the sulfur dioxide. The sulfur
dioxide fraction Is measured by the barium-
thorln tltrntton method.
1.3 Applicability. This method Is appli-
cable for the determination of sulfur dioxide
emissions from stationary sources only when
specified by the test procedures for determin-
ing compliance with Mow Source Performance
Standards.
3. Apparatus.
3.1 Sampling. See Figure o-i.
3.1.1 Probe—Pyrex > gloss, approximately
8 to 0 nun. ID, with a heating system to
prevent condensation and a filtering medium
to remove paniculate matter including *ul-
furto add mist.
9.1.3 Midget bubbler—One, with glas*
wool packed In top to prevent eulfurlo aold
mist carryover.
3.1.3 Glass wool.
9.1.4 Uidget implngei*—Three.
3.1.8 Drying tube—Packed with 8 to 10
mesh ladlcatlng-type silica gel, or equivalent,
to dry the sample.
9.1.0 Valve—Needle valve, or equivalent,
to adjust flow rate.
3.1.7 Pump—Leak-free, vacuum type.
9.1.8 Rate meter—Rotsmoter or equiva-
lent, to measure a 0-10 s.c JJa, flow range.
9J.9 Dry gas meter—Sufficiently accurate
to measure the sample volume within 1%.
9.1.10 Pitot tube—Type 8, or equivalent
necessary only if a sample traverse Is re-
quired, or If stock gas velocity varies with
time,
33 Sample recovery.
3J.1 alas* wash bottles—Two.
9O3 Polyethylene storage bottles—To
•tore Implngor samples.
34 Analysis.
'Tf«d8 nnmea.
80
Chapter I—Environmental Protection Agency App. A
f-nODE (END PACKED SILICA C.EL OnYING TUBt
WITH QUARTZ OR V/3JACK WAU
PYREX HOOL| • [^ MIDGET BUOOUR MIDGET IKPINGEfSS
GIASS WOOL
TYPE S PITOT TUDC
THERKOMETER
DRY GAS MCTER ROTAMETEII
Floure 6-1. 302 Campling train.
PUMP
3.8.1 Pipettes—Transfer type, 8 ml. and
10 ml. sizes (0.1 ml. divisions) and 36 ml.
tlse (OJl ml. divisions).
9.3.9 Volumetric flasks—BO ml., 100 ml.
and 1.000ml.
9.3.3 Burettes—8 ml. and BO ml.
3.3.4 Erlcnmeyer Bask—198 ml.
3. Reagenti,
3.1 Sampling,
8.1.1 Water—Dslonlced, distilled.
s.1.3 Isopropnnol, 80%—Mix 80 ml. of Iso-
propanol with 10 ml. of distilled water.
9.1.3 Hydrogen peroxide, 3%—dilute 100
ml. of 30% hydrogen peroxide to I liter with
distilled wntcr. Prepare fresh dally.
8 J> Sample recovery.
3J.I Water—Delontoed. distilled,
8.3.3 Isopropano!, 00%.
S3 Analysis.
8.3.1 Water—Deloateod. distilled.
3.3J isoproponoL
843 Thorln Indicator—l-(o-anonophen»
ylnco)-9-naphthol-8^>dlsuUonle acid, dlso-
dlum salt (or equivalent). Dissolve 0.20 g. In
100 ml distilled water.
8J.4 Barium perchlorate (0.01 /T)—Dis-
solve 1.06 g. of barium perohlorate
|Ba(01O,).« 8H.OI In 200 ml. distilled water
and dilute to 1 liter with Isoproponol. Stand-
ardise with rulturla aold. Barium chloride
may be used. .
8.8.8 Bullurle aold standard (041 W)—
Purchoss or itandardlte to ±0.0003 W
aenlnrt 0.01H NaOH which hoi prevlonsly
been standardised against potassium aold
phthalate (primary standard grade),
4. Procedure. ., •.
4,1 Sampling.
4.1.1 Preparation of collection train. Pour
IB mi. of 80% laopropa.no! luto the mldRot
bubbler and IS ml. of 3% hydrogen peroxide
Into each of the lint two midget Imploncra.
Leuvo tbe anal midget unplngcr dry. Assem-
ble the train as shown In Figure 0-1. Leak
chock tbe sampling train at tho sampling
site by plugging tho probe Inlet and pulling
a 10 Inches He vacuum, A leakage rate not
In excess of 1% of the sampling rate Is ac-
ceptable. Carefully release tho probe Inlet
plug and turn off the pump. Plaeo crushed
lee around the Implngcn. Add more Ice dur-
ing the run to keep the temperature of tbe
eases leaving tbe lost unpinger at 70* F. or
lets.
4.1.3 Sample collection. Adjust the sam-
ple flow rate proportional to the stack gns
velocity. Take readings at least every flvs
minutes and when significant changes la
itaek conditions necessitate additional ad-
justment* In flow rate. To begin sampling,
position the tip of the probe at tho first
sampling point and start the pump. Sam-
ple proportionally throughout the run. At
the conclusion of each run, turn off the
pump and record the anal reading*. Remove
the probe from the stnck and disconnect It
from the train. Drain tbe lea bath and purge
the remaining part of tbe train by drawing
clean ambient air through the system for 18
minute*.
4.9 Sample recovery. Disconnect tbe lm«
pingtrs after purging. Discard the contents
of the midget bubbler. Pour the contents of
the midget Implnger* into a. polyethylene
shipment bottle. Rinse tbe three midget Im-
and tho connecting tube* with dlv-
o
i
r-o
-------
D-l
APPENDIX D
REFERENCE METHOD 7
-------
AW..A
Title) 40—Protection of Environment
tilled water and add these washings to the
maw storage container.
4.3 Sample analysis. Transfer the contents
of the Btoroge container to a 60 ml. volu-
metrlo flask. Dilute to tho mark with de-
lonlcod, distilled water, ripe tie a 10 ml.
aliquot of this solution Into a 126 ml. Brian*
meyer flask. Add 40 ml. of Isopropanol and
two to four1 drop* of tborln Indicator. Titrate
to * pink endpoint using 0.01 N barium
peiehlorate. Bun a blank with each series
ol samples.
6. CaMbrofton.
6.1 TJeo standard methods and equipment
which have been approved by the Adminis-
trator to calibrate the rotometer, pltot tube.
dry gas meter, and probe heater.
6A. Standardize the barium perohlorate
against 36 m^- of standard sulf urio add con*
taming 100 mL of Isopropanol.
e. Calculation*.
8.1 Dry gas volume. Correct the sample
volume measured by the dry gas meter to
where:
C«o,— Concentration of sulfur dioxide
at standard conditions, dry
basis, Ib./cu. ft.
7.05X10-"— Conversion factor, including the
number of grams per gram
equivalent of sulfur dioxide
(33 gy£?.-eq.). 453.0 g^lb, and
1,000 ml./l., lb.-L/g.-ml.
V,—Volume of barium perchlorate
titrant used for the ^mrd*,
ml.
V,,—Volume of barium perohlomte
titrant used for the blank, ml.
W—Normality of barium perehlorate
titrant, g.-eq./l.
V,.i.—Total solution volume of sulfur
dioxide, 60 ml.
V—Volume of sample aliquot ti-
trated, ml.
V«.u—Volume of gas sample through
the dry gas meter (standard
conditions), ou. ft., see Equa-
tion 0-1.
7. Refereneei.
Atmospheric Emissions from Bulf urio Acid
Manufacturing Processes, U.S. DHEW. PBS,
Division of Air Pollution, Public Health Serv-
ice Publication No. 999-AP-13, Cincinnati,
Ohio. 1066.
Oorbett, P. P., The Determination of SO,
and SO, in Flue Oases, Journal of the Insti-
tute of Fuel, 34:387-343,1061.
Matty, B. B. and E. K. Dleht, Measuring
Flue-Gas SO, and SO,, Power JW:04-fl7, No-
vember, 1067.
Fatten, W. F. and J. A. Brink, Jr., New
Equipment and Techniques for Sampling
Chemical Process Gases, J. Air Pollution Con-
trol ASHOClatlon. 13,162 (1B63).
standard conditions (70* F. and 30.02 luchei
Hg) by using equation 9-1.
l7<71ta7Hi("7Er) equation ft-l
where*
V«,u— Volume of gas sample through the
dry gas meter (standard condi-
tions), cu. ft.
VM» Volume of gas sample through the
dry gas meter (motor condi-
tions) , cu. ft.
T...— Absolute temperature at standard
conditions. 830* 8, '•
TM— Average dry gas meter temperature,. .
B.
PM>— Barometric pressure at the orifice
meter, inches Hg.
Pi(4— Absolute pressure at standard con-
ditions, 29.93 inches Hg.
6.9 Sulfur dioxide concentration.
equation 6-2
MXTHOD T—DrnmMiNAnoj* or trmoanf oxmi
EMISSIONS rOOM STATIONAST 8O0OCIS
1. Principle and applicability.
1.1 Principle. A grab sample is collected
in an evacuated flask containing a dilute
sulfurlo acid-hydrogen peroxide absorbing
solution, and the nitrogen oxides, except
nitrous oxide, are measure colorlmetrlcally
using the phenoldlsulfonio add (PDS)
procedure.
1.2 .Applicability. This method Is applica-
ble for the measurement of nitrogen oxides
from stationary sources only when specified
by the test procedures for determining com-
pliance with New Source Performance
Standards.
3. Apparatus.
2.1 Sampling. See Figure 7-1.
3.1.1 Probe—Pyrex» glass, heated, with
filter to remove partlculata matter. Heating
is unnecessary if tho probe remains dry dur-
ing tho purging period,
2.1.3 Collection flask—Two-liter, Pyre*.1
round bottom with short neck and 34/40
standard taper opening, protected against
implosion or breakage,
3.1.8 Flask valve—T-bore stopcock con-
nected to a 34/40 standard taper joint.
3.1.4 Temperature gauge—Dial-type ther-
mometer, or equivalent, capable of measur-
ing 3* F. Intervals from 98' to 125* P.
3.1.8 Vacuum line—Tubing capable of
withstanding n vacuum of 3 inches Hg abso-
lute pressure, with "T" connection and T-bore
stopcock, or equivalent.
2.1.6 Pressure gauge—tJ-tubo manometer,
SO inches, with 0.1-lnoh divisions, or
equivalent.
1 Trade name.
Chaptor I—Environmental Protection Agency
App. A
3.1.7 Pump—Capable of producing a vac-
uum of 3 Inches Hg absolute pressure.
3.1.8 Squeeze bulb—Oneway.
3.2 Sample recovery.
3.2.1 ripotto or dropper.
2.3.2 Glass storage containers—Cushioned
for shipping.
EVACUATE
STANDARD TAfM. GMXINIMH.ASS
{ SU EVE NO. M/W SOCKT. $ NO. 114
FMU INCAStUEHir
••UllINO HAM •
MUCH. MnmO-OOnOU. SHOUT MCI.
Will 1 Silt Vt NO. J«MO
O
INi
Figure 7-1. Sampling train,
-------
App. A
Tillo 40—Protection of Environment
valve to Us "vent" pi>:illlon CIIH! turn olf tlio
pinup. ciii-cU llio iiuiuumuUT fur uiiy lluclii-
ulloii In 1:10 iiit-iciivy lovul. if Uuiro la A vlnl-
olo chiuiK" ovor tho upau of ono nitnulo,
chuck (or K-olu*. Record tlio lulllul volume,
U-mpcruuuo. and barometric prosuuro. Turn
the flask valve to Us "purge" po.tltlon, and
then do tho same with tho pump valve.
Purge the probe and tho vacuum tube using
the squeeze bull). It condensation occurs In
the probo and flnsk valvo area, heat tbe probe
and purge until tho condensation disappears.
Then, turn the pump valvo to Its "vent" posi-
tion. Turn tho fln.sk vnlva to Its "sainplu"
position and allow sample to enter tho dusk
Tor about 16 seconds. After collecting the
cample, turn tho fltvsk valvo to its "purge"
position and disconnect tho flnsk from tho
sampling train. Shnko the flask for 8
minutes.
4.2 Sample recovery.
4.3.1 Let the flask set tor a minimum of
10 hours and then shake tho contents for 2
minutes. Connect the flask to a mercury
filled U-tubo manometer, opc'n tho valve
from the flask to the manometer, and record
the flask pressure and temperature along
with tho barometric pressure. Transfer the
Cask contents to a container for shipment
or to a 260 ml. beaker for analysis, ntnsa the
flask with two portions of distilled water
(approximately 10 ml.) and add rinse water
to tho eamplo. For n blank use 25 ml. of ab-
Eorblng solution nud the oiimo volume of dis-
tilled water us used In rinsing the flask. Prior
to shipping or analysis, add sodium hydrox-
ide (IN) dropwlso Into both tho sample and
tbe blank until alkaline to litmus paper
(about 25 to 35 drops In each).
4.3 Analysis.
4.3.1 If the sample has been shipped in
a container, transfer tbe contents to a 260
ml. beaker using a small amount of distilled
water. Evaporate thu uohitloii to drynais on a
oU'imi Uutli mill thoii cuul. AilU :\ nil. pliuiiol-
dlimlrullld m:li| uulllllull to tho drlud ruulUua
mid trlturulo tlioioni;lily with n tfmut rod,
Mako sure thu uolullon uoutnotH all thu rust-
due. Ada 1 nil, dlstlllud water and four drops
of concentrated sulfurlc acid. Heat tho solu-
tion on a steam bath for 3 minutes with oc-
casional stirring. Cool, add 20 ml. distilled
water, mix well by stirring, and Add concen-
trated ammonium hydroxide dropwlso with
constant stirring until alkaline to lltruui
pupcr. Transfer tho solution to a 100 ml.
volumetric flask and wash tho beaker thrco
times with 4 to 6 ml. portions of distilled
water. Dilute to tho mark and mix thor-
oughly. If tho saniplii contains solids, trans- .
fcr a portion of the solution to a clean, dry
centrifuge tubo, and centrifuge, or fllter a'i
portion of tho solution, Measure tho absorb-
once of each sample at 420 nm. using the
blank colutlon as a zero. DUuto tho sample
and thu blank with n suitable amount of
distilled water if absorbanco fails outside tho
range of calibration.
5. Calibration.
6.1 Flask volume. Assemble the flask and
flask valve and fill with water to tho stop-
cock. Measure tho volume of water to ±10
ml. Number and record tbe volume on th»
flask.
0.2 Speotropbotometer. Add 0.0 to 10.0 ml.
of standard solution to ft scries of beakers. To
each beaker add 20 nil. of absorbing colutlon
and add eodlum hydroxide (IN) dropwltte
until alkaline to litmus paper (about 25 to
35 drops). Follow thu analysis procedure of
section 4.3 to collect enough data to draw •
calibration curve of concentration In *g. NO>
per sample versus absorbanoe.
0. Calculations.
9.1 Sample volume.
where:
V,.-
*.,«-
V,
v.
Sample volume at standard condi-
tions (dry basis), ml.
Absolute temperature at standard
conditions. 530* B.
Pressure at standard condltlOM,
20^2 Inches Hg.
Volume of flask and valve, ml.
, Volume of absorbing solution, 25 mL
F,™ Final absolute pressure of flask,
inches Bg.
P,-Initial absolute pressure of flask.
Inches Hg.
T!—Final absolute temperature of flask.
•R.
T,—Initial absolute temperature of flask,
93 Sample concentration. Bead *g. NO,
(or coca sample from tbe plot of MS- NO,
versus absorbonee.
0-Concentratlon of NO, ss NO. (dry
basts). IbywJ.
,, m-Uoss of NO, In gas sample, *g.
, V.t=SAmple volume at standard eondl-
tlons (dry basis), ml.
equation 7-2
1, Jte/ermeei.
Standard Methods of Chemical Analysts.
8th ed. New York, D. Van Nostrand Co., too,
1062, vol.1, p. 320-330.
Standard Method of Test for Oxides of
Nitrogen in Gaseous Combustion Products
L J
Chapter I—Environmental Protocllon Agoney
App. A
(I'huiiolrtluullonlo Aciil Procedure), In: I'JOU
lloulc of AU'I'M rllnmlnrd.-i, 1'urt i!il. riilliulul-
|)lili\. 1M. 1UOU, A3TM UculL'imlluii U-1UUU-UO.
P. ii!s-Y29.
Jacob, M. n,, Tho Chomlcnt Analysis of Air
Pollutanta, Now York, N.Y., Iivtorsclciico I'ub-
Ushon, Inc., 1000, vol. 10, p. 351-300.
METHOD e—DmcaitiNATioH or BVLrunia ACID
IdlflT AMP BULTO* DIOXIDI SMIBSIONB TSLOU
BTATJOHAllY SOOTCEU
1. Principle and applicability.
1.1 Principle. A gas sample Is extracted
from a sampling point in tho stack and the
acid mist Including sulfur trloxldc Is sepa-
rated from sulfur dioxide. Both traction* ore
measured separately by tho barlum-Uiorlu
tltratlou method.
1.3 Applicability. This method Is applica-
ble to determination of sulfurlc acid mist
(Including sulfur trloxldc) and sulfur dlox-
Ido from stationary sources only when spc-
clllod by tho test procedures for determining
compliance with tho New Source Torform-
nnco Standards.
2. Apporalta.
2.1 Sampling. See Figure 8-1. Many of
tho design specifications of this sampling
train are described In APTD-0501.
PROSE
REVERSE-TYPE
PITOT TUBE
3.1.1 Nuwle—Sl:illilrvi i,I PC I (DID) with
2.1.'.! I'n.lji' I'yri-:: ' r.l:u.-i with n limiting
uy.lum to prevent vl:.ll>lo iviuUu: ullou Uur-
lni: unnxpllun.
tt.1.3 PItot tube—Typo 3, or equivalent,
attached to probo to monitor stack gu*
velocity.
2.1.4 Fitter holder—Pyres'glass,
2.1.5 Implngeru—Four as r.hown In Figure
8-1. The flrst and third are of the Orcenburg-
Entlth dOiilcu with stnndard tip, Tho aocoud
mill fourth nrc of thu CiR'OiiDurg-Sinllli do-
sl|»n, iniidincil by replacing Uio uinndurd lip
with a {6-lncb. ID gln.su tube extending to
ono-bnU Inch from tlio bottom of tho 1m-
plncer flink. Similar' collection systems,
which have been approved by tho Adminis-
trator, may bo used.
3.1.0 Metering system—Vacuum gauge,
leak-free pump, thermometers capable of
measuring tcmperutura to within 8* P., dry
gn.i meter will) 2% accuracy, and related
equipment, or equivalent, as required to
maintain au isokmcttc snmpllng rate and
to determine sample volume.
2.1.7 Barometer—To measure atmocpherle
pressure to ±0.1 Inch Hg.
1 Trade name.
FILTER HOLDER ^THERMOMETER ^
CHECK
VALVE
VACUUM
LINE
VACUUM
GAUGE
IN VALVE
IB-TIGHT
PUMP
DRY TEST I
. Figure 8-1. Sulfurlc acid mist sampling train.
85
-------
E-l
APPENDIX E
COMPLETE SUMMARY OF METHOD 5 SAMPLING
-------
PLANT: A
Tm, °R
T °R
TS> K
PB, IH. Hg
Ps, IfJ. Hg
(T, MIN
AS« Fl2
AN, FT2
Cp
Vj , ml
i. T
MN» mg
% 02
% co2
X CO
% Ng
fZ? Ave
£.H Ave
Vnstd- DSCF
Vystd- CF
Di t/\
WO
EA, %
Mj, Ib/lb mole
MS Ib/lb mole
Vs, FPS
Qs, DSCFH
A j to
C$> Ib/SCF
i
Cs, gr/SCF
PMR, Ib/hr
1
536
790
29.08
29.07
80
28.274
1.36 x 10~3
0.78
30.1
157.29
17.4
2.2
0.2
80.2
" 0".122
0.50
29.13
1.427
0.047
447
29.048
28.529
7.939
501 ,862
90.48
1.191 x 10"5
0.083
5.975
E-2
TEST 2
525
785
29.12
29.11
80
28.274
1.36 x 10"3
0.78
32.0
175.39
19.2
1.8
0.2
78.8
0.122
0.351
34.61
1.52
0.042
1121.0
29.06
29.59
7.91
505,910
106.7
1.331 x 10'5
0.093
6.732
3
. 532
585
29.40
29.39
80
28.274
1.36 x 10"3
0.78
29.3
195.8
18.4
1.4
0.2
80.0
0.141
0.69
34.613
1.389
0.039
649
28.960
28.533
9.096
589,935
99.4
1.247 x 10'5
0.087
7.358
-------
PLANT: B
Tm, °R
TS, °R
PB> IN. Hg
Ps, IN. Hg
-------
PLANT: C.
Tin, °R
TS, °R
PB» IK- Hg
Ps, IN. Hg
V, MIN
AS, FT2
V FT
cp
V , ml
HN, mg
%02
% coz
% CO
% Ng
(2Tp Ave
iH Ave
VMStd> DSCF
vWStd' CF
Oi IA
WO
EA, %
Md, Ib/lb mole
MS Ib/lb mole
Vs, FPS
Qs, DSCFH
I, %
Cs, Ib/SCF
1
Cs, gr/SCF
PMR, Ib/hr
567
950
29.25
29.22
60
7.86
7.669 x 10-4
0.78
64.90
528.10
14.8
5.0
—
80.2
~ "0.373
1.378
34.277
3.076
0.082
232
29.392
28.458
26.582
378,124
93.44
3.397 x 10"5
0.237
12.778
E-4
TEST
2
541
940
29.25
29.22
60
7.86
7.669 x 10'4
0.78
68.2
213.2
15.8
4.2
—
80.0
0.396
1.539
39.387
3.233
0.076
298
29.304
28.445
28.079
404,151
99.87
1.194 x 10'5
0.083
4.824
3
. ' 550
945
29.25
29.22
60
7.86
7.669 x 10-4
0.78
52.30
343.9
15.2
5.2
—
79.6
0.374
1.392
36.631
2.479
0.063
261
29.440
28.719
26.462
384,194
97.77
2.07 x 10-5
0.145
7.953
-------
PLANT: D
Tm, °R
TS, °R
PB» IN. Hg
Ps, IN. Hg
(T, MIN
As> FT2
AN> Fl2
cp
Vj , ml
MN» mg
% 02
% co2
% CO
% Ng
iCTp Ave
£H Ave
VMStd» DSCF
Vystd' CF
Bwo
EA, %
Md, Ib/lb mole
•Ms Ib/lb mole
Vs, FPS
Qs, DSCFH
I, 5.
Cs, Ib/SCF
€5, gr/SCF
PKR, Ib/hr
E-5
, TEST 2
550 547
720 725
30.12 30.12
30.11 20.11
60 60
12.566 12.566
1.36 x 10"3 1.36 x lO'3
0.78 0.78
44.3 51.6
29.8 51.84
14.4 15.2
6.2 5.8
— —
79.4 79.0
"0.155 0.182
1.133 1.45
33.199 37.02
2.10 2.446
0.059 0.062
219 269
29.568 29.536
28.885 28.82
9.403 11.09
388,080.0 346,166.5
103.5 98.8
1.979 x 10-6 3.09 x 10"6
0.014 0.072
0.768 1.07.
3
535
730
30.16
30.15
60
12.566
1.36 x 10"3
0.78
50.8
23.48
15.0
5.4
—
79.6
0.138
0.89
32.28
2.413
0.070
249
29.464
28.66
8.46
260,238.7
114.5
T.603 x 10-6
0.0112
0.417
-------
PLAi, l : t
Tm, °R
T °R
'$'
PB, IN. Hg
Ps, IN. Hg
IT, MIN
AS, FT2
AN, FT2
Cp
Vj , ml
MN» nig
% 02
% co2
% CO
% Ng
GlP Ave
AH Ave
VMStd, DSCF
vwstd- CF
^0
EA, %
Mj, Ib/lb mole
Ms Ib/lb mole
Vs, FPS
Qs, DSCFH
I, %
C$> WSCF
i
Cs, gr/SCF
pMR, Ib/hr
1
542
810
29.70
29.69
108
5.585
3.41 x 10~4
0.78
51.0
390.94
18.4
2.2
0.4
79
'0.432
0.61
46.845
2.42
0.049
685
29.088
28.54
28.159
349,516
122
1.84 x 10'5
0.129
6.431
E-6
TEST 2
546
835
29.70
29.69
108
5.585
3.41 x lO'4
0.78
50.2
338.93
17.6
3.2
—
79.2
0.431
0.61
46.39
2.38
0.049
532
29.216
28.67
28.46
342,717.5
123.16
1.611 x 10~5
0.113
5.521
3
545
840
29.70
29.69
108
5.585
3.41 x 10"4
0.78
48.1
369.58
16.6
4.2
—
79.2
0.437
0.63
46.20
2.28
0.047
385
29.336
29.80
28.88
346,356
121.3
1.764 x 10'5
0.123
6.109
-------
PLANT: F
Tm, °R
TS. °R
PB> IN. Hg
Ps, IN. Hg
cr, HIM
As, FT2
V Fl2
cp
Vj , ml
MN» m9
% 02
% co2
% CO
% Ng
£2"P Ave
£H Ave
VKStd> DSCF
vWStd> CF
R
Dwo
EA, %
Md, Ib/lb mole
Ms Ib/lb mole
VS. FPS
Qs, DSCFH
I, %
Cs, Ib/SCF
C$, gr/SCF
PMR, Ib/hr
1
554
760
29.97
29.96
60
12.57
1.36 x 10~3
0.78
32.6
342.1
17.0
3.4
0.2
79.4
0".193
1.48
! 38.598
1.545
0.0385
416
29.224
29.792
12.083
366,839.94
97.2
1.954 x lO'5
0.136
7.168
E-7
TEST 2
553
705
•?q '-7
29.96
60
12.57
1.36 x 10'3
0.78
31.3
278.8
16.6
4.0
0.2
79.2
0.227
2.07
45.465
1.484
0.032
374
29.304
28.942
13.647
449,911
93.4
1.352 x 10'5
0.094
6.083
3
541
650
29.97
29.96
60
12.57
1.36 x 10
0.78
30.1
414.74
17.6
2.8
0.2
79.4
0.206
1.83
43.996
1.427
0.031
505
29.152
28.806
11.920
426,655
95.36
2.079 x 10-5
0.145
8.868
-------
PLANT: G
Tm, °R
T °R
TS» K
PB> IN. Hg
Ps, IN. Hg
(T, MIN
AS> FT2
AN, FT2
Cp
Vj , ml
MM, mg
X 02
% co2
% CO
% Ng
&P Ave
A\\ Ave
VMStd> DSCF
vWStd» CF
Bwo
EA, %
Md, Ib/lb mole
MS Ib/lb mole
Vs, FPS
Qs, DSCFH
I, %
CS, Ib/SCF
Cs, gr/SCF
PMR, Ib/hr
J_
524
900
29.31
29.29
60
9.33
7.67 x 10~4
0.78
41.3
150,49
13.0
6,8
—
80.2
'""b.245
0.75
29.680
1.958
0.062
159
29.61
28.888
16.847
305,920,2
110.6
1.118 x 10'5
0.078
3-42
E-8
TEST 2
535
910
29.31
29.29
60
9.33
1.36 x 10'3
0.78
77.1
227.05
12.0
8.2
0.2
79.6
0.271
2.74
49.185
3.655
0.069
130.6
29.792
28.978
18.709
333,490
101.2
1.018 x 10'5
0.071
3.395
3
. 536
910
29.31
29.29
60
9.33
1.36 x TO"3
0.78
60.5
212.55
11.2
9.0
0.2
79.6
0.195
1.45
37.58
2.8677
0.0709
112
29.83
28.99
13.46
239,435.3
107.7
1.247 x 10"5
0.087
2.986
-------
PLANT : H
Tm, °R
TS. °R
,-B. IN. Hg
Ps, Hi. Hg
ir, KIN
Ac, FT2
o
AN, FT2
Cp
VT , ml
I
MN» m9
% 02
% CO-
d.
% CO
% Ng
ffip Ave
^H Ave
VMStd> DSCF
vWStd' CF
BWO
r~ n o!
EA, %
MJ, Ib/lb mole
d
Ms Ib/lb mole
Vc, FPS
o '
Qo, DSCFH
X5
I, %
Cs, Ib/SCF
1
Cs, gr/SCF
PMR, Ib/hr
E-9
TEST
i
540
750
29.98
29.27
108
12.566
7.67 x ID'4
0.78
102.9
2697.6
12.2
7 /L
/ . H
n 4
U « H
80.0
0.209
0-62
46 n
"TV * I •
4.878
0 095
\J * \J -/ +s
132
29.56
OQ nc.&.
cy .'tot
13.067
378,369.0
no
1.29 x 10"4
0.901
48.81
2
542
750
29.98
29.27
108
12.566
7.67 x 10'4
0.78
94.9
1988.28
14.2
5.8
0.3
79.7
0.208
0.60
46.853
4.498
0.088
201
29.412
28.41
13.02
380,072
112
9.357 x 10'5
0.654
35.564
3
546
740
29.98
29.27
108
12.566
7.67 x lO'4
0.78
90.6
2964.37
13.7
6.6
0.2
79.5
0.205
0.59
45.122
4.294
0.087
184
29.55
28.54
12.712
376,628.6
109.2
1.449 x ID'4
1.012
54.573
-------
PLANT: I
Ira, °R
TS, °R
PB, IN. Hg
Ps, IN. Hg
IT, MIN
AS. FT2
V FT2
cp
Vj , ml
HN» mg
% Q2
% co2
% CO
% Ng
#P Ave
&H Ave
VKStd> DSCF
VilStd' CF
Buo
EA, %
Md, Ib/lb mole
Ms Ib/lb mole
Vs, FPS
Qs, DSCFH
I, %
CS, Ib/SCF
Cs, gr/SCF
PMR, Ib/hr
i
551
870
30.08
30.06
120
7.069
1.36 x 10~3
0.78
107.1
795.9
16.6
3.8
0.2
79.4
"0.301
1.40
75.265
5.077
0.0632
370
29.216
28.507
13.503
196,981.7
99.3
2.331 x 10'5
0.163
4.593
E-10
TEST
2
548
860
30.08
30.06
120
7.069
1.36 x 10-3
0.78
104.4
2777.55
16.0
4.2
0.2
79.6
0.213
1.46
77.526
4.949
0.060
311
29.312
28.633
14.190
210,123.4
95.9
7.9 x 10'5
0.552
16.60
3
.552
860
30.08
30.06
72
7.069
1.36 x 10~3
0.78
69.6
431.9
16.8
3.6
—
79.6
0.201
1.37
45.761
3.299
0.067
399
29.250
28.494
13.43
197,731.5
100.4
2.08 x 10'5
0.145
4.105
-------
PLANT: J
Tin, °R
TS, CR
PB, i;;. Hg
Ps, Hi. Hg
v, i:iu
As, FT2
AN, FT2
Cp
Vj , ml
HN» ^9
%02
% co2
% CO
% Ng
^P Ave
£H Ave
VMStd, OSCF
BWO
EA, 5
Md, Ib/lb mole
Ms Ib/lb mole
Vs, FPS
Qs, DSCFH
I, o
CS, Ib/SCF
Cj, gr/SCF
PMR, Ib/hr
1
556
970
29.25
29.23
60
7.07
1.36 x 10"^
7.67 x 10~4
0.78
62.4
399.78
14.4
6.0
0.2
79.4
"0.261
1.720
*i37.556
2.958
0.073
215
29.536
28.694
18.714
235,642.6
2.347 x 10'5
0.164
5.53
TEST 2
549
935
29.25
29.23
60
7.07
7.67 x 10~4
0.78
53.4
394.75
13.4
7.0
0.2
79.4
0.281
0.91
31.803
2.531
0.074
174
29.656
29.793
19.747
257,680
m5
•
-------
F-l
APPENDIX F
STATE OF VERMONT WOOD BOILER REGULATIONS
-------
F-2
VERMONT ADMINISTRATIVE RULE
Cv"l
1*1 Adopted Rule
TITLE OR SUBJECT: Air Pollution Control Regulations - Wood-Fired Boilers
AGENCY: Environmental Conservation
AGENCY'S REFERENCE NUMBER FOR RULE (IF ANY): Regulations 5-101, 5-211,
5.-231-3 and 5-408-11
EFFECT ON EXISTING RULES: New Material: 5-101-10 Amends: 5-211
5-101-22 5-408-11
5-101-45
5-211-3
5-231-3.b.
STATUTORY AUTHORITY: 10 VSA 554 and 558
Effective Date: tfi
-------
F-3
a shade, or density, greater than 40S opacity (No. 2 of the Ringelmann
Chart).
At no time shall the visible air contaminants have a shade, density, " "
or appearance greater than BQ% opacity (No. 3 of the Ringelmann Chart).
2. Installations^ constructed subsequent to April 30, 1970
No person shall cause, suffer, allow or permit the emission of any visible
air contaminant, from installations constructed subsequent to April 30, 1970.
[after the effective date of these regulations] for more than a period or periods
aggregating six (6) minutes in any hour, which has a shade, or density, greater
than 20^ opacity (No. 1 of the Ringelmann Chart).
At no time shall the visible air contaminants have a shade, density, or
•
appearance greater than 60S opacity (No. 3 of the Ringelmann Chart).
_3^ Exceptions - Wood Fuel Burning Equipment
a_)^ During normal startup operations, emissions of visible air contaminants
in excess of thelimits specifiedin subsections1 & 2 above may be allowed
for a period not to exceed one (1) hour.
f
bj During normal soot blowing operations, emissions of visible air
i
contaminants in excess of the limits specified in subsections 1 & 2 above
may be allowed for a period not to exceed 30 minutes during any 24 hour
period.
-,&
_
subsection have a shade, density, or appearance greater than 80% opacity
(No. 4 of the Ringelmann Chart).
clj Any wood I'ucl burniny equipment thai has a rated output of 40 II.I1. ;
or less shall not be subject to this regulation (§5-211).
Action 3. Section 5-231-3, entitled "Prohibition of Particulate Matter - Combustion
Contaminants", is amended by adding the following new subsection:
-------
F-4
5-231 PROHIBITION OF PARTICULATE MATTER
3. Combustion Contaminants
b. A person shall not discharge, cause, suffer, allow, or permit the
emission of particulate matter caused by the combustion of wood fuel in
fuel burning equipment from any stack or chimney:
1). in excess of 0.45 grains per dry standard cubic foot (gr/DSCF) of
exhaust gas corrected to 12% CO? in any combustion installation that has a
rated output of greater than 90 H.P. which commenced operation prior to
December 5. 1977.
2j_ in excess of 0.20 gr/DSCF corrected to 12* C02 in any combustion
installation that has a rated output of greater than 90 H.P., but less than
r
1300 H.P., which commences operation after December 5. 1977.
11 in excess of 0.10 gr/DSCF corrected to 12% COg in any combustion
Installation that has a rated output of greater than 1300 H.P. which commences
operation after December 5, 1977.
Any wood fuel burning equipment that has a rated output of 90 H.P. or less
shall not be subject to these particulate emission standards.
When any fossil fuel is burned in combination with wood fuel . and the fossil
fuel contributes less than 50% of the total BTU Input, the above particulate
standards shall apply. If the fossil fuel contributes more than 50% of the
total BTU input, subsection 3. a. of this regulation shall apply.
Uhen a soot blowing cycle exceeds 15 minutes, separate emissions testing
for particulate emissions during the soot blowing cycle may be required in
addition Lo emissions testing during normal operating conditions pursuant to
Regulation 5-404 below. In this event, the emission 'rate calculated for the
soot blowing cycle shall be prorated over the time period between soot blowing
cycles.
-------
Action 4. Section 5-408-11, entitled "Installations Requiring That Information
Be Submitted To The State Air Pollution Control Agency Prior To Construction",
is amended to read:
5-408 INSTALLATIONS REQUIRING THAT INFORMATION BE SUBMITTED TO THE STATE AIR
POLLUTION CONTROL AGENCY PRIOR TO CONSTRUCTION
The following types of installations are required to submit to the State Air
Pollution Control Agency information regarding the air pollution potential of their
proposed new construction, new installation, or modification:
11. Fuel Burning Installations:
a^ Fossil fuel burning equipment of greater than 10 million BTU's
per hourrated heat input.
JK Wood fuel burning equipment of greater than 90 H.P. rated output.
[11. Fuel burning installations greater than 10 million BTU's per hour rated.
heat input]
[ ] = Deletions
= Additions
-------
G-l
APPENDIX G
STATE OF VERMONT - CYCLONE REGULATIONS
-------
G-2
5-231 PROHIBITION OF PARTICULATE MATTER
i
1. Industrial Process Emissions
a. No person shall discharge, cause, suffer, allow, or permit in any
one hour from any stack whatsoever particulate matter in excess of the
amount shown in Table 1. For purposes of this regulation the total process
weight entering a process unit shall be used to determine the maximum
allowable emissions of particulate matter which may pass through the stack
associated with the process unit. When two or more process units exhaust
through a common stack, the combined process weight of all of the process
units, served by the common stack, shall be used to determine the allowable
particulate emission rate.
b. In cases where process weight is not applicable as determined by
the Air Pollution Control Officer, the concentration of solid particulates
in the effluent gas stream shall not exceed 0.14 grams per cubic meter
(0.06 grains per cubic foot) of undiluted exhaust gas at standard conditions
on a dry basis. In the case of wood processing operations, process weight
is not applicable and, instead, the concentration standard specified in this
subsection shall apply.
«U.S. GOVERNMENT PRINTING OFFICE: 197» — 600-923/343
------- |