EPA-650/2-74-003
January 1974
Environmental Protection Technology Series
w
'4
UJ
(3
i! f-ark
-------
EPA-650/2-74-003
A STUDY
OF AIR POLLUTANT EMISSIONS
FROM RESIDENTIAL HEATING SYSTEMS
by
R. E. Hall, J. H. Wasser, and E. E. Berkau
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
ROAP No. 21ADG-AO
Program Element No. 1AB014
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, N.C. 27711
January 1974
-------
This report has been reviewed by the Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency, nor does
mention of trade names or commercial products constitute endorsement
or recommendation for use.
ABSTRACT
This document presents a comprehensive collection of recent EPA
research work into the problem of air pollutant emissions from small scale
combustion systems. Major factors for controlling emission levels were
found to be: excess air, residence time at high temperature, combustion-
air-handling components of burners, and burner maintenance. Recommendations
for minimizing emissions from new and existing equipment are given, based
on the research results obtained. Data illustrating the effects of
combustion parameter changes on emission levels are given both for experi-
mental combustors and for residential heating equipment currently in use
in the U. S. Future work directed toward reduction of emissions is also
outlined.
11
-------
CONTENTS
LIST OF FIGURES v
LIST OF TABLES vii
ACKNOWLEDGEMENTS viii
SUMMARY 1
Major Results 1
Minimizing Emissions from Existing Equipment 2
Future Work 4
INTRODUCTION 5
EXPERIMENTAL EQUIPMENT 7
Experimental Furnace 7
Conventional Residential Furnace 7
Test Fuel 11
Combustion Chambers 11
Combustion Improvers 11
Flame Retention Burners 13
Gas Burners 23
Other Oil Burners 23
Ignition Systems 23
Oil Nozzles 23
ANALYTICAL INSTRUMENTATION AND PROCEDURES 25
EXPERIMENTAL RESULTS 27
Summary of Results 27
Estimate of Experimental Error 31
DISCUSSION OF RESULTS 37
Oxides of Nitrogen (NOX) 37
Oxides of Sulfur (SO) 37
A
Carbon Monoxide (CO) 37
Hydrocarbons (HC) 37
Smoke and Particulates 38
Air/Fuel Stoichiometry 38
Residence Time 50
m
-------
CONTENTS (Continued)
Page
Combustion Chamber Effects
Combustion Improving Devices
Flame Retention Burners
Cyclic-Based Emissions
Natural Gas Burners
Other Distillate Oil Burners
Ignition Systems
Nozzle Effects
Effect of Time on Tuning
Burner Maintenance
PROCEDURES FOR REDUCING POLLUTANT EMISSIONS FROM CURRENT
DISTILLATE-OIL-BURNING HEATERS
CONCLUSIONS AND RECOMMENDATIONS
BIBLIOGRAPHY
Appendix A. BURNER ADJUSTMENT AND COMPARISON
Appendix B. FUEL ANALYSIS
Appendix C. COMPARISON OF ALL OIL BURNERS TESTED
Appendix D. OTHER DISTILLATE OIL BURNERS
Appendix E. CONVERSION FACTORS
61
62
69
71
75
79
80
81
83
86
89
91
94
95
103
105
109
113
IV
-------
LIST OF FIGURES
Fig. No. Title^ Page
1 Experimental Furnace, Interior Detail ' 6
2 Experimental Furnace, Schematic 8
3 Conventional Domestic Furnace 9
4 Domestic Furnace Instrumentation Schematic 10
5 Typical Combustion Chambers 12
6 ABC Model 45 Burner 14
7 Monarch G-81-C Combustion Head 15
8 Delavan FlameCone 16
9 Shell Combustion Head 17
10 Gulf Econo-Jet 18
11 Union (Pure) Flame Control Device 19
12 ABC Mite Burner (Model S) 20
13 Beckett Bantam Burner (Model AF) 21
14 Union Flame Control Device Installed on ABC
Model 45 Burner 22
15 Nozzle Distributors 24
16 Critical Parameters Affecting Pollutant Formation 39.
17 Combustion Chamber Temperatures vs Stoichiometric
Ratio 40
18 Furnace Temperatures at Air/Fuel Ratio of 1.75 42
19 Particulate Emissions vs Stoichiometric Ratio 43
20 Smoke Emissions vs Time for Various Stoichio-
metric Ratios 44
21 Carbon Monoxide Emissions vs Stoichiometric Ratio 45
22 Gaseous Hydrocarbon Emissions vs Stoichiometric
Ratio 46
23 Nitrogen Oxide Emissions vs Stoichiometric Ratio 47
24 Sulfur Dioxide Emissions vs Stoichiometric Ratio 48
25 Experimental Furnace Heating Efficiency 49
26 Effect of Residence Time on Carbon Dioxide and
Oxygen 52
27 Effect of Residence Time on Particulate Emissions 53
28 Effect of Residence Time on Smoke Emissions 55
-------
LIST OF FIGURES (Continued)
Fig. No. Title Page
29 Effect of Residence Time on Carbon Monoxide
Emissions 56
30 Effect of Residence Time on Gaseous Hydrocarbon
Emissions 57
31 Effect of Residence Time on Nitrogen Oxides Emissions 58
32 Nitric Oxide Equilibrium Flame Curve for No. 2 Oil
with 70 F Air Inlet Temperature 59
33 Effect of Residence Time on Sulfur Oxides Emissions 60
34 Average Smoke Emissions of Combustion Improving
Devices vs Stoichiometric Ratio 63
35 10th Minute Smoke Emissions of Combustion Improving
Devices vs Stoichiometric Ratio 64
36 Carbon Monoxide Emissions of Combustion Improving
Devices vs Stoichiometric Ratio 65
37 Gaseous Hydrocarbon Emissions of Combustion
Improving Devices vs Stoichiometric Ratio 66
38 Nitrogen Oxides Emissions of Combustion Improving
Devices vs Stoichiometric Ratio 67
39 Overall Heating Efficiencies of Combustion
Improving Devices vs Stoichiometric Ratio 68
40 Hydrocarbon and Carbon Monoxide Trends During
Cycle 72
41 Smoke and Particulate Trends During Cycle 73
42 Nitric Oxide Trend During Cycle 74
43 Carbon Monoxide Emissions for Gas-Fired Units 76
44 Carbon Monoxide Emissions for Oil-Fired Units 76
45 Hydrocarbon Emissions for Gas-Fired Units 77
46 Hydrocarbon Emissions for Oil-Fired Units 77
47 Nitric Oxide Emissions for Gas-Fired Units 7"8
48 Nitric Oxide Emissions for Oil-Fired Units 78
49 Change in Pollutant Emissions with Time--Light-Oil
Residential Heater 84
50 Deterioration of Furnace Efficiency with Time 85
A-l Burner Operating and Emission Characteristics 96
A-2 Determination of Stoichiometric Ratio for No. 1
10th Minute Smoke 97
VI
-------
LIST OF FIGURES (Continued)
Fig. No. Title Page
A-3 Determination of Carbon Monoxide for Normal
Operating Conditions 98
A-4 Determination of Hydrocarbons for Normal Operating
Conditions 99
A-5 Determination of Nitric Oxide for Normal Operating
Conditions 100
A-6 Determination of Average Smoke for Normal Operating
Conditions 101
A-7 Determination of Efficiency for Normal Operating
Conditions 102
LIST OF TABLES
Table No. Title Page
1 Air/Fuel Stoichiometry 26
2 Residence Time 28
3 Combustion Chamber Effects 29
4 Combustion-Improving Devices 30
5 Flame Retention Burners 32
6 Natural Gas and Oil-Fired Burners 33
7 Effect of Ignition Systems on Nitric Oxide Emissions 34
8 Nozzle Effects 35
9 Experimental Error 36
10 Furnace Residence Time Comparison 51
11 Comparison of Average Emissions from Residential
Oil Burners 87
vii
-------
ACKNOWLEDGEMENTS
It is our pleasure to acknowledge the Environmental Protection
Agency employees who contributed to various aspects of the program.
R. P. Hangebrauck and G, B. Martir assisted with the air/fuel stoichiometry
and residence time studies, respectively. Project engineers for the work
with combustion improving devices and many of the flame retention burners
were M. H. Hooper and D. P. Howekamp. We also express appreciation
to C. H. Bernhardt, N. L. Butts, R. A. Mueller, and R. E. Thompson, the
engineering technicians who helped perform the tests.
viii
-------
SUMMARY
MAJOR RESULTS
This study showed that excess air, residence time, flame retention
devices, and maintenance are major factors in the control of air pollutant
emission levels and equipment performance of residential heaters. It was shown
that carbon monoxide (CO), gaseous hydrocarbons (HC), smoke, and particulate
emissions pass through a minimum point as excess air is increased from
stoichiometric. Oxides of nitrogen (NO ) behave in the opposite manner,
however: as excess air is increased, NO emissions pass through a maximum
A
point.
A longer residence time was found to significantly reduce emissions
of CO, gaseous HC, smoke, and particulates. However, NO emissions were
increased slightly.
A study of combustion improving and flame retention devices showed that
flame retention can be used both to reduce total emissions and to increase
furnace efficiency. Although most flame retention devices tested increased
emissions of NO , one device reduced them. Combustion improving devices,
other than one utilizing flame retention, had little effect on pollutant
emission levels.
This study and related field tests showed that burner and furnace
maintenance affect burner performance and emission levels. Old, worn out,
poorly constructed, or maladjusted burners are responsible for unnecessarily
high levels of air pollutant emissions.
Ignition systems, nozzles, and combustion chamber shape and material
were found to be less significant variables in the control of air pollutant
emissions. NO w.as.- the only pollutant affected by ignition systems. It
A
was found that some ignition systems increased nitric oxide (NO) emissions
by about 10 percent, but one unit tested had no effect on them.
-------
Nozzles were found to have a small effect on emissions of smoke,
gaseous HC, CO, NO, and carbon dioxide (C02). However, differences for
nozzles within the same brand as well as for nozzles of different brands
indicated that burners should always be readjusted when nozzles are replaced.
Combustion chamber material was found to affect all emissions. When
firing into a steel-lined chamber, the excess air had to be increased to
obtain acceptable levels of CO, gaseous HC, and smoke. Thus, the efficiency
was reduced. The combustion chamber shape had little effect on the emission
levels, as long as the chamber dimensions were not changed significantly.
Gas burners, which had a rating equivalent to the oil burners discussed
above, were also tested. These tests provided a comparison between emissions
of gas and distillate oil burners. It was found that emissions from gas
burners are about the same as those from most equivalent-size, htgh-pressure,
atomizing-gun oil burners.
MINIMIZING EMISSIONS FROM EXISTING EQUIPMENT
The results discussed above can be used to minimize air pollutant
emissions from existing equipment. Excess air, one of the major variables
in reducing emissions, should be set as low as possible to provide high
efficiency; however, it should not be set so low that it creates excessive
amounts of CO, gaseous HC, smoke, and particulates. The burner should be
set so the smoke level at hot running condition is no higher than No. 1 on
the Bacharach scale*. A more detailed explanation is given in Appendix A.
A refractory-lined combustion chamber, as opposed to a steel-lined chamber,
will allow burner operation at a lower excess air level.
Since longer residence time has a positive effect in reducing overall
emissions, existing furnaces should be underfired; i.e., a nozzle, slightly
*The Bacharach Smoke Number to which reference is made throughout this paper,
is used for convenience and because most readers are familiar with it. The
Bacharach Smoke Number is actually the "Smoke Spot Number" described in
ASTM D 2156-65.
-------
smaller than the one which originally came with the furnace, should be
installed. The smaller nozzle will help reduce emissions in two ways:
it will provide longer residence time; and it will reduce cyclic-based emissions
since the burner will have to remain on longer to provide a given heat load.
The number of cycles per unit time will be reduced. Of course the nozzle
must have the capacity to supply a sufficient quantity of oil when the heating
demand is greatest.
Since flame retention devices were found to improve furnace efficiency
and reduce overall emissions, flame retention should be utilized. The flame
retention concept is not limited to new or replacement equipment. The compon-
ents which create the flame retention effect (e.g., the end cone and the
retention ring or cone) can easily and inexpensively be installed on existing
burners in the field. One such device, which increased efficiency and re-
duced all air pollutant emissions except NO , was the Union flame control
/\
device. Of the flame retention burners tested, the ABC Mite and Beckett
Bantam burners increased efficiency and reduced emissions most effectively.
The ABC Mite was the only burner tested which significantly reduced NO
/\
emissions, however.
Since air pollutant emissions can be reduced significantly if all burners
are properly maintained, boilers and furnaces should be serviced by an
authorized serviceman at least once a year, normally just prior to the heating
season. Nozzles should be replaced yearly, with readjustment of the burner.
If the heater malfunctions, the serviceman should be recalled. The service-
man should always adjust the furnace by checking the COp level, draft, and
the Bacharach Smoke No. with proper instruments, not by "eyeing" the flame.
Old, worn out units which cannot be adjusted properly should be replaced.
The above recommendations can result in the following new-burner emission
levels:
CO: 0.5 g/kg fuel
Gaseous HC: 0.06 g/kg fuel
NO; 0.8 g/kg fuel
Bacharach Smoke No.: 1.0
(after 10 minutes
of operation)
If the heater is properly serviced these levels should be maintained.
3
-------
FUTURE WORK
These studies indicate that further work is required in several areas.
Long-term performance tests are needed to accurately determine the effect of
time on burner emissions once the burner is adjusted. Tests are also needed
to more accurately determine the effect of underfiring burners both to
increase residence time and to lower cyclic emissions.
A study is needed to determine whether improvement of furnace efficiency
by using better heat exchangers is an economical way of reducing air pollutant
emissions. If better heat exchangers are used, less fuel will be required
for a given heat load; thus, the total emissions will be reduced.
Studies of methods to reduce cyclic emissions are also needed. A pilot
or modulating burner should be investigated.
Critical burner and furnace design factors need further investigations.
Such a study is presently being performed by the Rocketdyne Division of
Rockwell International for distillate oil burners, under EPA Contract 68-02-0017.
Similar programs are needed for other types of burners; e.g., residual oil
burners and mixed fuel burners.
Burners or components which offer possible reductions in air pollutant
emissions and improved efficiency should be investigated further. This
includes burners such as "blue flame" and other unique burners, compact heating
systems, and components such as sonic nozzles.
More refined instrumentation is needed for servicemen when adjusting
burners. For example, a portable instrument which could measure COp, stack
temperature, draft, and smoke at the same time would be beneficial. Perhaps
a rating (such as good, fair, or poor) could be provided which would integrate
the CQy* stack temperature, draft, and smoke measurements. This could either
be provided as part of the instrument or on a separate chart. Reasonably
priced and portable instruments for accurately measuring gaseous HC, CO,
and NO should be developed. Such instruments would help the serviceman
/\
to adjust heaters for low emissions and maximum efficiency. As a followup,
training should be provided for the serviceman. If he is not familiar
with the instruments and does not know either how to use them properly
or how to evaluate the results, they will be of little use.
-------
INTRODUCTION
The residential heater project was established because seasonal
and geographic surveys indicate that significant amounts of air pollution
1 2
result from domestic heating ' . It has been estimated that air pollution
from residential and commercial heating sources constitutes approximately
10 percent of the total air pollution in the United States. However, since
pollution from these sources occurs only during the heating season at
ground levels and in highly populated areas, the pollution problem is more
significant than indicated by the 10 percent estimate. Therefore, the
U. S. Environmental Protection Agency (EPA) initiated several projects
related to the reduction of air pollution from residential heating. The
early work was performed with an experimental furnace to determine the effects
of air/fuel stoichiometry and residence time on air pollution emissions.
Later, specific commercially manufactured furnaces, combustion improvers,
flame retention burners, and prototype burners were investigated to identify
the designs with low emissions. Most of this work was performed with a
commercially available furnace.
The study concentrated on distillate oil and natural gas heaters which
account for over 90 percent of the residential heating units in the United States.
Specifically, this study was planned to determine: the burner design variables,-
components, and process conditions which are critical for control of air
pollutant emissions; what can be done with present technology to control
pollutant emissions; and research requirements to eliminate pollutant emissions
from residential heaters.
-------
EXHAUST
SAMPLING PROBE
INLETS
NOTES:
1. UNITS ARE
INCHES UNLESS
OTHERWIDE
INDICATED.
2. POINTS TT
THROUGH T8
ARE THERMOCOUPLE
LOCATIONS.
HEAT EXCHANGER
TUBES: 7x7
%O.D., 0.58 I.D.
COOLING
AIR IN
COOLING
AIR OUT
INSULATING
REFRACTORY
WING WALL,
26.5-DEGREE ANGLE
FROM WALL
FUEL
OIL
COMBUSTION
AIR
Figure 1. Experimental furnace, interior detail.
-------
EXPERIMENTAL EQUIPMENT
EXPERIMENTAL FURNACE
The experimental furnace shown in Figures 1 and 2 was built for the
initial studies. The design was dictated by several criteria: the equipment
must be able to control the fuel, air, and other process variables; the
internal geometry of the combustion chamber and heat exchanger must be simple
enough to permit the high-temperature residence time of the combustion gases
to be estimated and varied, and deposits to be removed; and the unit must
be flexible enough to allow for changes in equipment and methods of operation.
An ABC Model 55J-1 oil burner equipped with a 1 gph*, 80-degree hollow-cone
nozzle, was chosen for these studies. It is a popular-make, high-pressure
atomizing-gun burner: in 1962, over 87 percent of the domestic oil burners in
3
the United States were of this type and 1 gph was an average domestic furnace
firing rate; and by 1970, about 95 percent of the domestic oil burners were
4
the high-pressure atomizing-gun type . The combustion chamber for this burner
was designed so the interior dimensions and wing walls conformed to recommended
sizes for this capacity burner ' ' . The air-cooled, steel heat exchanger was
a shell-and-tube type in which the combustion gases passed through the tubes.
CONVENTIONAL RESIDENTIAL FURNACE
The part of the program in which commercially available and prototype
equipment were tested required a furnace that was both representative of
commercially available furnaces and adaptable to a large percentage of the
burners to be tested. A survey was made to determine the burner size most
widely used and the furnaces adaptable to that burner size. As a result of
the survey, it was decided to test burners with a 0.75 gph firing rate, using
a Williamson Temp-0-Matic Lo Boy Furnace for the tests. The furnace is shown
schematically in Figure 3. Instrumentation used with the furnace is shown in
Figure 4. The Williamson burner which originally accompanied the furnace (a
conventional high-pressure atomizing-gun ABC Model 45 burner with a 0.75 gph,
80-degree hollow-cone nozzle) was used as the standard of comparison.
*Although it is EPA policy to use the metric system, this publication uses
certain non-metric units for convenience. Those more familiar with metric
units should refer to Appendix E for the proper conversion factors.
7
-------
FLUE GAS
00
COOLING
AIR IN
COOLING ,
AIR OUT
FLAME
THERMOCOUPLE
ENTRY POINTS
t
FILTER
CONTINUOUS
AIR BLOWER
LAMINAR
FLOW
ELEMENTS
COMBUSTION
AIR IN
"ON" CYCLE
J / COMBUSTION
/ AIR BLOWER
MAIN
FUEL
SUPPLY
"ON" CYCLE
COMBUSTION AIR
BLOWER CONTROL
POWERSTATS
CONTINUOUS^
'CHECK AIR RATE
VALVE MANOMETER
CONTINUOUS AIR
BLOWER CONTROL
"ON" CYCLE
COMBUSTION
AIR RATE
MANOMETER
ffl
WEIGH
SCALE
STAINLESS STEEL
FUEL SUPPLY
SYSTEM
Figure 2. Experimental furnace, schematic.
-------
RETURN
DUCT
WARM AIR
COOL AIR RETURN
WARM AIR
COMBUSTION GASES
POWER
FUEL LINE
COMBUSTION
AIR INLET
o
INCHES
I.D. O.D.
BURNER
2 ) COMBUSTION CHAMBER
HEAT EXCHANGER
OUTER RADIATOR
INTER CHAMBER
WARM AIR BLOWER
12
H
15
Q
0
y2 2
i
o ^^ i
HEATING SURFACE
Figure 3. Conventional domestic furnace.
9
-------
PUMP
NEEDLE VALVE
TOGGLE VALVE
V
ROTAMETER
NDIR
NO
ANALYZER
PUMP
NEEDLE VALVE
TOGGLE VALVE
ROTAMETER
POLAROGRAPHIC
ANALYZER
TEFLON TUBING
STAINLESS STEEL
TUBING WITH
HEATED TAPE
SINTERED METAL
FILTER
MOLECULAR
o S|EVE
(3A- CLAY BASE)
0
TRAP
(MIDGET IMPINGER)
FIBERGLASS
PARTICULATE
FILTER
PUMP
© TOGGLE VALVE
FLAME
GLASS
55,
F'LTER
FLUE
GAS
STACK
^tv^
SEQUENTIAL
TAPE SMOKE
SAMPLER
SIEVE
PUMP
NEEDLE VALVE
TOGGLE VALVE
ROTAMETER
co2
ANALYZER
ANALYZER
Figure 4. Domestic furnace instrumentation schematic.
-------
An operating cycle of 30 minutes with a burner on-time of 10 minutes
o
was selected as a result of a previous field test which showed that oil
burners operate cyclicly and burn about a third of the time during the
heating season.
TEST FUEL
The test fuel, No. 2 distillate fuel oil, was a blend of catalytically
cracked and straight-run stocks derived from a Gulf Coast crude. It had an
API gravity of 35 degrees, aromatic content df 25 percent, carbon/hydrogen
ratio of 6.62:1, and a nitrogen content of less than 0.01 percent. A
complete fuel analysis can be found in Appendix B.
To provide a fuel of uniform quality, the oil was stored under a blanket
of pure nitrogen.
COMBUSTION CHAMBERS
In order to determine the effect of combustion chamber configuration
and material, three different combustion chambers were tested. Two were
refractory lined; the third was steel with no refractory. As shown in
Figure 5, one refractory lined chamber was cylindrical, fired into radially,
and lined with a soft refractory material. The square chamber was lined with
light brick refractory. The steel cylindrical chamber, with no refractory,
was fired into axially.
COMBUSTION IMPROVERS
Combustion improving devices are designed to improve performance of
older burners by providing better air/fuel mixing. It was desired to determine
their effect on new, more efficient burners. Five commercially available
combustion improving devices for high-pressure atomizing-gun burners were
chosen for the study.
The devices were installed on the standard ABC Model 45 burner. In
each case, the same standard burner chassis was modified by installing the
combustion improving device to be tested. The tests were made in the con-
ventional domestic furnace described above.
11.
-------
CYLINDRICAL (ROUND)
(SOFT REFRACTORY)
SQUARE
(LIGHT BRICK REFRACTORY)
21/2"
'-1
I
9"
10"
CYLINDRICAL HORIZONTAL
(STEEL LINED)
12
r\
Figure 5. Typical combustion chambers.
12
-------
A schematic of the fuel nozzle and air/fuel mixing assembly of
the unmodified ABC Model 45 burner is shown in Figure 6. Schematics of the
Monarch G-81-C combustion head, Delavan FlameCone, Shell combustion head,.
Gulf Econo-Jet, and Union (formerly Pure) flame control device are shown
in Figures 7 through 11, respectively.
Each of these devices except the Delavan FlameCone utilized swirl
mixing to some degree. The Delavan FlameCone and Union device used flame
retention. Thus the Union device was the only one which incorporated both
swirl and flame retention. It was also the only device which controlled
inlet air by sliding an air shield within the blast tube rather than by
manipulating shutter vanes on the burner housing.
FLAME RETENTION BURNERS
High-pressure atomizing-gun burners which utilized flame retention were
also tested. All of these burners except one were available as an entire
burner. The one exception was the Union device which was actually a combustion
improving device that utilized flame retention, the same device that was
described previously as a combustion improving device. As with the other
combustion improvers, the Union flame control device was installed on the ABC
Model 45 burner. This device should not be confused with the Union flame
retention burner, which actually is a Beckett burner fitted with the Union flame
control device.
The flame retention burners tested included: ABC Mite, Beckett Bantam,
Esso Model 40, Sun-Ray, U. S. Carlin, Union, Union flame control device installed
on an ABC Model 45 burner, Wayne, and White-Rogers. Schematics of the ABC Mite,
Beckett Bantam, and Union flame control device are shown in Figures 12 through
14, respectively.
Each device utilized flame retention to hold the flame to create a more
stable, compact, intense flame. The ABC Mite and Union device also incorporated
swirl mixing.
13
-------
1-1/8"
4-1/8" O.D
ELECTRODES
NOZZLE
ADAPTER
NOZZLE
CHOKE
STATIC PRESSURE DISC , 31/2" dia
\\\\\\\\\ \\\ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ m\ \ \ \ \ \ \
NOZZLE
80-DEGREE
HOLLOW-CONE
Figure 6. ABC Model 45 burner.
-------
\\\\\\\\\\\\\\ \\x\\\\\\x\\\\\\\\\\\\\\\\\\\\\
AIR TUBE
\\\\\\\\\\\\\\\\
ELECTRODES
SUPPORT ASSEMBLY
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
FAN WHEEL
U*- 1/8" GAP
CHOKE
NOZZLE
80-DEGREE
SOLID-CONE
Figure 7. Monarch G-81-C combustion head.
-------
1" MIIM.
FLAMECOIME TURBULATOR DISC
V\V\\\\\\\YV\\ V_\_ A.. X \ \\\ \ \ \\\ \\\
NOZZLE
70-DEGREE
HOLLOW-CONE
Figure 8. Delavan FlameCone.
-------
4 HOLES, 7/16" dia. 90°
(AIR INLET TO PRIMARY
V AIR CAN)
\\v\\\\\\\\\\\ \\\Y\
FRONT VIEW
END CONE
REMOVED
NOZZLE
80-DEGREE
HOLLOW-CONE
Figure 9. Shell combustion head.
-------
AIR TUBE
4-1/8" O.D.
oo
STATIC DISC
STATIC DISC
ELECTRODE SUPPORT
NOZZLE ADAPTER
NOZZLE
NOZZLE
80-DEGREE. WITH SIX
ASPIRATOR HOLES
Figure 10. Gulf Econo-Jet.
-------
2-3/4'
tv v * V v \ v v V * v ^ vv *** * v »
AIR TUBE
ELECTRODES
4-1/8" O.D.
COMBUSTION AIR ADJUSTING ASSEMBLY
WITH AIR SHIELD ATTACHED
MOVEMENT
AIR FLOW INCREASED ^BH
AIR FLOW DECREASED fr
^ ^ vvx \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\x\\\\\\\\\\\\\\\\\\\\\\\'
END CONE
AIR SHIELD 2-3/4" dia.
NOZZLE
60-DEGREE
HOLLOW-CONE
Figure 11. Union (Pure) flame control device.
-------
ELECTRODES
AIR TUBE
NOZZLE
ADAPTER
CHOKE
Figure 12. ABC Mite burner (model S).
-------
AIR TUBE
ELECTRODES
NOZZLE
ADAPTER
COMBINATION END CONE _ _ _
AND FLAME RETENTION SHIELD
Figure 13. Beckett Bantam burner (model AF).
-------
AIR TUBE
ELECTRODES
NOZZLE
ADAPTER
AIR
SHIELD
Figure 14. Union flame control device installed on ABC model 45 burner.
-------
GAS BURNERS
Two types of natural gas burners were chosen for tests in which gas
burner emissions could be compared with oil burner emissions. A Williamson
Mono-Port burner and a Bryant Sectionalized burner were tested. Both were
rated at 100,000 Btu, the same as the oil burners.
OTHER OIL BURNERS
Distillate oil burners other than the high-pressure atomizing-gun
type were also tested. The objectives of these tests were to compare operation
and emissions of the various types of burners with those of conventional burners,
Other burners tested included four low-pressure burners, one vaporization
rotary-type burner, and four blue flame burners. Two of the blue flame
burners utilized induced internal recirculation; the third utilized external
recirculation of flue gases; the fourth did not have recirculation.
IGNITION SYSTEMS
Three ignition systems were tested to determine their effect on oil burner
emissions. Two of the systems were manufactured by the France Manufacturing
Company: Franceformer (Cat. LKJ) ignition system was installed on an ABC Model
45 burner, and a Franceformer (Cat. 4LACYU-4) was installed on a Beckett Bantam
burner. The third ignition system tested was a Prestolite 0-120 installed on
an ABC Mite burner.
OIL NOZZLES
Four major brands of oil nozzles (Delavan, Monarch, Hago, and Steinen)
were tested for variation of emissions and flow rate of new nozzles. All were
0.75 gph, 80-degree, hollow-cone nozzles. Variations in the nozzle distributors
can be seen in Figure 15.
23
-------
DELAVAN
MONARCH
HAGO
STEINEN
Figure 15. Nozzle distributors.
24
-------
ANALYTICAL INSTRUMENTATION AND PROCEDURES
Emission measurements were made over a wide range of stoichiometric
air/fuel ratios for each burner. Automatic analyzers and recorders contin-
uously monitored temperatures (inlet, outlet, and flue), Op, COp, CO, NO,
and gaseous HC (as methane). Bacharach smoke spots were taken once a minute
during the on-period and measured on a reflectance photometer. S0? and
particulate weight were only measured during tests with the experimental
furnace.
The actual data were the average of the emissions from the entire
on-cycle and thus included any startup or shutdown peaks. Emissions were
not measured during the off-period of the operating cycle. A detailed
description of the equipment and methods used for analysis can be found in
Reference 9.
Oxygen was measured with a pplarigraphic analyzer; NO was measured by
the phenoldisulfonic acid (PDSA) method, as described in Reference 9, during
the earlier stages of the program. Later NO was measured with a long-path
non-dispersive infrared (NDIR) analyzer. Since results between the two
methods were in agreement, the wet chemical method was replaced by long-path
NDIR analysis. Hydrocarbons were measured by flame ionization, and NDIR
analyzers were used for measuring both CO and COp. The efficiency of each
burner was calculated by dividing the total amount of heat (Btu) coming from
the heat exchanger by the net heating value (Btu) of fuel burned for each
cycle.
Smoke numbers are reported as 10th minute and average. The 10th
minute smoke number refers to the one smoke spot taken over a 1-minute
period after the burner had been on for 9 minutes. In other words, this
is the level of smoke produced at "steady state" (or hot running) conditions.
It was used to determine operating air settings for burner comparison (see
Appendix A). However, to determine a number indicative of the total amount
of smoke or carbon particulate produced, an average smoke number was cal-
culated by averaging Bacharach values for the ten measurements taken during
the entire on-period.
25
-------
Table 1. AIR/FUEL STOICHIOMETRY
Combustion product
Particulate, g/kg fuel
Filterable
Condensable
Carbon monoxide, g/kg fuel
Gaseous hydrocarbons, g/kg fuel
Oxides of nitrogen, g/kg fuel
Sulfur dioxide, g/kg fuel
Sulfur tri oxide, g/kg fuel
Oxygen, vol %
Carbon dioxide, vol %
Smoke No., Bacharach
Smoke density, Cohs/1000 ft
Range of emission
Minimum
0.04
0.17
0.86
0.03
1.08
1.46
<0.02
5.5
5.9
0
7.0
Maximum
33.40
1.70
96.60
17.00
2.41
1.96
<0.02
12.8
9.6
9 +
>1230
26
-------
EXPERIMENTAL RESULTS
SUMMARY OF RESULTS
The experimental data is summarized in Tables 1 through 8. All tests
were made with No. 2 oil or natural gas.
Table 1 contains the results from the initial studies of the effects
of air/fuel ratio on air pollutant emissions (CO, HC, SO^, NO, N02» smoke
and particulates) from an oil-fired test furnace. The range of air/fuel
ratios investigated was 1.0 to 2.5, corresponding to excess air levels of
0 to 150 percent, at a constant fuel rate of 1.0 gph. The gaseous measurements
were obtained in units of parts per million (ppm) but were converted to
emission factors in units of grams of pollutant per kilogram of fuel burned.
The results of the tests to investigate residence time are contained
in Table 2. Data for the short residence time were obtained from the tests
to determine the effects of air/fuel stoichiometry described above. A longer
residence time was achieved by increasing the height of the combustion chamber,
thus increasing the volume by a factor of 1.8. With that exception the tests
were identical to the air/fuel stoichiometry tests.
Table 3 contains the results of tests with various combustion chamber
configurations and materials to determine the effect on COp, 0^, and emissions
of CO, HC, NO, and smoke. The data presented in this table are based on burner
adjustment to a No. 1 smoke number at hot running conditions (see Appendix A).
The range of air/fuel ratios for these tests was 1.1 to 2.6, at a constant
fuel rate of 0.75 gph.
The results of tests to determine the effect of combustion improving
devices on burner performance (furnace efficiency, C02 and Op levels, and
emissions of CO, HC, NO, and smoke) are given in Table 4. Five combustion
improving devices were compared to a standard high-pressure atomizing-gun
burner, using the method described in Appendix A. The burners were operated
27
-------
Table 2. RESIDENCE TIME
Pollutant emission
Filtered participate,
g/kg fuel
Carbon monoxide,
g/kg fuel
Hydrocarbons,
g/kg fuel
Nitric oxide,
g/kg fuel
Sulfur dioxide,
g/kg fuel
A/F ratio
1.00
1.25
1.50
1.75
1.00
1.25
1.50
1.75
1.00
1.25
1.50
1.75
1.00
1.25
1.50
1.75
1.00
1.25
1.50
1.75
Residence time
Short
33.40
5.40
0.34
0.06
96.60
8.87
0.86
1.20
17.00
1.82
0.03
0.07
0.70
0.81
1.05
1.30
1.46
1.80
1.86
1.96
Long
10.840
0.146
0.024
0.027
1.380
0.458
0.612
1.107
0.288
0.068
0.059
0.103
0.61
0.88
1.18
1.37
1.56
1.75
1.84
1.94
28
-------
Table 3. COMBUSTION CHAMBER EFFECTS
Burner
Williamson
Pure
Monarch
Pollutant emissions
and Stoichiometric Ratio
Stoichiometric ratio
10th min smoke, Bacharach
Avg smoke, Bacharach
HC, g/kg fuel
CO, g/kg fuel
NO, g/kg fuel
Stoichiometric ratio
10th min smoke, Bacharach
Avg smoke, Bacharach
HC, g/kg fuel
CO, g/kg fuel
NO, g/kg fuel
Stoichiometric ratio
10th min smoke, Bacharach
Avg smoke, Bacharach
HC, g/kg fuel
CO, g/kg fuel
NO, g/kg fuel
Cylindrical
refractory
1.52
1.0
2.6
0.02
0.6
1.26
1.18
1.0
1.1
0.08
0.6
1.63
1.63
1.0
2.4
0.04
0.5
1.26
Square
refractory
1.52
1.0
3.5
0.14
0.7
1.32
1.20
1.0
1.1
0.11
0.4
1.92
1.38
1.0
4.0
0.08
0.6
1.08
Horizontal
steel
1.65
1.0
2.2
0.08
0.4
1.55
1.37
1.0
3.5
0.06
0.3
1.76
2.03
1.0
1.6
0.15
2.3
0.94
29
-------
Table 4. COMBUSTION-IMPROVING DEVICES
A/F ratio
producing
No. 1 smoke
Air setting,
% co2
Efficiency
of furnace, %
Gaseous HC,
g/kg fuel
CO, g/kg fuel
NO, g/kg fuel
Ave smoke,
Bacharach No.
Standard
ABC
burner
1.53
9.9
76.6
0.06
0.5
1.11
2.9
Monarch
combustion
head
1.66
9.1
71.5
0.06
0.6
1.25
2.0
Delavan
Flame-
Cone
1.80
8.2
70.5
0.03
0.6
1.30
1.3
Shell
combustion
head
1.60
9.4
76.0
0.06
0.3
1.68
2.0
Gulf
Econo-
Jet
1.40
10.8
75.0
0.06
0.6
1.69
3.0
Union (Pure)
flame
retention
head
1.20
12.6
83.0
0.06
0.5
1.25
1.2
30
-------
over a range of air/fuel ratios from 1.0 to 2.6, at a fuel rate of 0.75
gph.
Table 5 contains the results of the studies which were designed to
determine the effect of flame retention devices on burner performance. These
tests were identical to those with combustion improving devices.
The results of comparing emissions of CO, HC, and NO from natural gas
burners with emissions from equivalently rated oil burners are given in
Table 6. Since the air/fuel adjustment on the gas burners was limited, the
range of air/fuel ratios for the gas burners was very narrow. The gaseous
measurements were made in units of parts per million (ppm) but were converted
to emission factors in units of grams of pollutant per million calories
input.
Table 7 indicates the effects of ignition systems on emissions of NO.
The tests were made with three different ignition systems, with and without
combustion taking place. Therefore, the NO emissions are given in parts
per million (ppm).
Table 8 indicates effects of nozzles on air pollutant emissions (CO^,
HC, NO, and smoke) and furnace efficiency. The tests were made at a constant
air/fuel ratio of 1.60 to allow more reliable comparison and analysis of the
data. Estimated experimental errors are included in the table.
ESTIMATE OF EXPERIMENTAL ERROR
To estimate the experimental error in the test data, a statistical
analysis was performed on one combustion improving device for 10th
minute and average smoke, CO, gaseous HC, NO, and efficiency. The
product of this analysis is an estimate of the standard deviation,
defined as S, and is shown for each set of data in Table 9. To de-
termine if the experimental error had changed after a period of 2
years, another analysis was made as part of the nozzle testing program.
This data is also given in Table 9. Since the difference between the
31
-------
Table 5. FLAME RETENTION BURNERS
Air setting,
% co2
Efficiency
of furnace, %
Gaseous HC,
g/kg fuel
CO, g/kg fuel
NO, g/kg fuel
Ave smoke
Bacharach No.
ABC
Model 45
9.9
75.0
0.06
0.5
1.10
2.9
ABCh
Mite
10.9
79.5
0.06
0.5
0.77
2.0
Becketg
Bantam
11.6
81.1
0.06
0.5
1.40
2.5
Union .
modification
12.6
83.0
0.06
0.5
1.25
1.2
Conventional burner
""Flame retention burner
32
-------
Table 6. NATURAL GAS AND OIL-FIRED BURNERS
Burner
Gas-fired:
Williamson
furnace
Bryant
boiler
Bryant
furnace
Oil-fired:
Union (Pure)
ABC Mite
ABC Standard
(Model 45)
Stoichiometric
ratio
1.20
1.40
1.60
1.20
1.38
1.53
NO, g/106
cal input
-
0.084
0.115
0.112
0.115
0.071 .
0.102
HC, g/106
cal input
0.0007
0.0014
0.0075
0.0055
0.0055
0.0055
CO, g/106
cal input
0.022
0.099
0.032
0.046
0.046
0.046
33
-------
Table 7. EFFECT OF IGNITION SYSTEMS ON NITRIC OXIDE EMISSIONS
Ignition system
Franceformer
(Cat. LKJ)
£
Franceformer
(Cat. 4LACYU-
4)
Presto! 1te
(0-120 Ignition
system)
Burner
Williamson
(standard
ABC)
Williamson
with Union
(Pure) device
Beckett
Bantam
ABC Mite
Run No.
1
2
1
2
3
4
5
1
2
1
2
Status
ON
OFF
ON
OFF
ON
OFF
ON
OFF
OFF
ON
OFF
ON
OFF
ON
ON
OFF
ON
OFF
NO level,
ppm
67.0
60.0
123.0
113.25
117.5
111.0
118.0
111.5
65.5
73.5
2.25
11.75
3.00
11.75
88.5
79.5
74.0
74.0
Reduction
ppm
7 n
9.75
6.5
7.0
6.5
8.0
9.5
8.75
8.75
9.0
0.0
%
10.4
7.9
5.5
5.9
5.5
10.9
10.2
0.0
Average
Reduction
Ppm
7.0
7.8
9.0
9.0
0.0
%
10.4
7.1
10.2
0.0
Combustion
Yes
Yes
No
Yes
Yes
-------
Table 8. NOZZLE EFFECTS
lOth minute
smoke,
Bacharach No.
Average
smoke,
Bacharacii No.
Gaseous HC,
g/kg fuel
NO,
g/kg fuel
CO,
g/kg fuel
rn °/
LUp » k
Efficiency
of furnace, %
Nozzle
Delavan
1.9
2.93
0.080
1.13
0.34
9.20
68.16
Monarch
1.73
3.03
0.082
1.11
0.34
9.58
64.78
Hago
2.47
4.59
0.086
0.96
0.32
9.37.
72.21
Steinen
4.61
6.19
0.113
0.95
0.42
9.28
74.52
Experimental
error
estimate^
Standard dpviatipn)
0.24
0.23
0.01
0.06
0.05
0.19
3.11
35
-------
Table 9. EXPERIMENTAL ERROR
Data set
10th minute smoke,
Bacharach No.
Average smoke,
Bacharach No.
CO, g/kg fuel
Gaseous HC,
g/kg fuel
NO, g/kg fuel
Efficiency, %
Standard deviation(S)
Combustion improving
device tests
0.13
0.15
0.05
0.01
0.09
2.17
Nozzle
tests
0.24
0.23
0.05
0.01
0.06
3.11
S value is very small for each data set, it is assumed that the ex-
perimental error given in Table 9 is representative of the error for
the entire program.
The standard deviation can be used to make a confidence statement
about the average response. Defining x as the average of several
readings taken at the same excess air setting, one can state with
95 percent confidence that the true average of the data points at
this setting will lie within the the interval x plus or minus approx-
imately 2 standard deviations.
36
-------
DISCUSSION OF RESULTS
OXIDES OF NITROGEN (NO)
A
In each of the studies discussed below it is important to understand
the mechanism of formation of oxides of nitrogen (NO ) which represent the
/\
combination of NO and NO,,. NO is formed from both free nitrogen in the
£ A
atmosphere at high temperatures and from bonded nitrogen in the fuel.
Atmospheric nitrogen reacts with oxygen at elevated temperatures to form
NO and to a lesser degree NOp. Nitrogen which is bonded in the fuel reacts
as part of the fuel and is not considered to be as temperature dependent.
Since the nitrogen content of the fuel used in this work is less than 0.01
percent, the maximum amount of NO formed from fuel nitrogen is about 13 ppm
^
at 3 percent 02(16 percent excess air), or 0.23 g NO/kg fuel. This assumes
100 percent conversion which may not occur in actual practice. Therefore,
it is assumed that any change in NO emissions in subsequent discussions is
A
related only to the NO formed from high-temperature fixation of atmospheric
/\
nitrogen.
OXIDES OF SULFUR (SO )
rt
Oxides of sulfur (SO ) are formed from chemically bonded sulfur which
A
reacts with oxygen during the combustion process. SO is present as S09
f\ £
and S03. During the combustion process S02 is much more prevalent than
SOV Since 95 percent or more of the fuel sulfur is converted to SO , SO
«j XX
was not measured during most of this work. The sulfur which does not
oxidize to SO is emitted with the particulate matter.
/\
CARBON MONOXIDE (CO)
CO is a product of incomplete combustion. If combustion is complete,
the carbon in the fuel will be oxidized to CO^. Therefore, properly
designed and well-maintained burners will not emit very high levels of CO.
HYDROCARBONS (HC)
HC emissions are also a product of incomplete combustion. If combustion
is complete, the hydrogen will be oxidized to form H90 and the carbon will
37 . £
-------
be oxidized to form CCL. As with CO, HC emissions should be very low
if the burner is properly designed and maintained well.
SMOKE AND PARTICULATES
Smoke is generally considered to consist of carbon particulates and
is therefore a product of incomplete combustion. However, some particulates
are the result of non-combustible material in the fuel. Smoke was measured
for all tests made during this study. However, particulate by weight was
only measured during tests made with the experimental furnace. Particulates
were not measured during the entire program because particulate sampling
is time consuming and difficult.
AIR/FUEL STOICHIOMETRY
The initial studies with oil burners were performed in the experimental
furnace with the ABC Model 55J-1 burner described earlier. The main
objective was to establish the effects of air/fuel stoichiometry on air
pollutant emissions. Four critical parameters (oxygen concentration, flame
temperature, inlet combustion air velocity (turbulence), and mean gas
residence time) which are known to affect the quantity of pollutants formed
are plotted for a range of stoichiometric ratios in Figure 16.
Flame temperature was measured by dividing the combustion chamber into
three vertical zones: A, B, and C (Figure 2). Entry ports are positioned
along the centerlines in each of these three zones and labeled W, X, Y, and
Z from top to bottom (row W was not used during these tests). Temperatures
were recorded at 1-inch intervals through the chamber from wall to wall at
each entry port. Every height level was traversed with three probes
operating simultaneously. Fine wire thermocouples (0.003 inch, iridium
and iridium with 40 percent rhodium) were used to minimize error due to
conduction and radiation. These temperatures and maximum refractory
temperatures are shown in Figure 17. Flame temperatures, except for point
maximums, were below the theoretical adiabatic flame temperature. The
point maximums were above adiabatic at stoichiometric ratios greater than
1.6 in Figure 17 because of poor mixing in localized regions. Even though
the overall ratio was 1.6 or more, the air/fuel ratio at points can be
much lower or higher.
38
-------
<
2400 0.4
2300
2200 0.3
21001
.
2000 Q 0.2
g 1900
LLJ
>
1800 0.1
1700
1600
OXYGEN
15 40
10 30
u
-------
3500
3000
u. 2500
o
UJ
Q.
2000
1500'
1000
NOTE:
ZONES ARE
DEFINED IN
FIG. 2.
MAX
. REFRACTORY
ITEMP: .
1.00 1.25 1.50 1.75 2.00 2.25 2.50
.STOICHIOMETRIC RATIO
Figure 17. Combustion chamber temperatures
versus stoichiometric ratio.
40
-------
The temperature of the refractory, flue gas, tube sheet, and heated
air typically increased over the burner-on cycle, as shown for an air/
fuel ratio of 1.75 in Figure 18. Note that only the refractory temperatures
do not reach steady state during the 10-minute on cycles.
The trends for emissions of particulates, smoke, CO, and gaseous HC
can be seen in Figures J9 through 22, respectively. These emissions were
all minimized at air/fuel ratios between 1.65 and 2.00. Emissions of NO
/\
and SO,,, Figures 23 and 24, respectively, decreased as the air/fuel ratio
decreased. The reduction in S02 at low air/fuel ratios is attributed to
the lack of available oxygen and sorption by the large amounts of carbon
soot produced.
Relative curves for heat balance and operating efficiency of the
experimental furnace are shown in Figure 25. Efficiency was calculated
by subtracting the heat lost both in the flue gases and through incomplete
combustion based on a carbon balance, from the net heat input. This calcula-
tion resulted in a maximum heating efficiency of 73.6 percent at an air/fuel
ratio of 1.25. The maximum C02 reading was obtained at this same ratio, which
verifies the setting for maximum heating efficiency.
Adding the amount of heat gained by the cooling air passing through
the heat exchanger to these losses in Figure 25 leaves some heat which is
not included. This heat is considered as being absorbed by and radiated
from the combustion chamber refractory and is included as a positive factor
in computing efficiency.
By comparing the trends of each pollutant with the four parameters
shown in Figure 16, the levels of air pollutant formations can be explained.
Combustion was complete at levels between 1.65 and 2.00. At levels above
2.00 the flame temperature and combustion gas residence time in the furnace
were too low for complete combustion. At low air/fuel ratios, the high
flame temperatures together with poor air/fuel mixing due to low inlet
air velocity resulted in thermal cracking of the fuel droplets, thus yielding
carbon soot, CO, and unburned HC. NO emissions were minimized at the
41
-------
cc
<.
cc
LLJ
0.
1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
I I I f
REFRACTORY:
FLUE GAS
WARM AIR OUTLET
J L
01 234567
ELAPSED TIME, min.
8 9 10
Figure 18. Furnace temperatures at air/fuel ratio of 1.75-
42
-------
40.0
10.0
5.0
_3
01
CO
g
CO
CO
1.00
0.50
o
<
a.
0.10
0.05
KEY;
O TOTAL
D FILTERED
CONDENSED
0.01
1.00 1.25 1.50 1.75 2.00 2.25
STOICHIOMETRIC RATIO
Figure 19. Participate emissions versus
stoichiometric ratio.
2.50
-------
600
o
o
£
co
ac
O
34567
ELAPSED TIME, min.
10
Figure 20. Smoke emissions versus time for various
stoichiometric ratios.
44
-------
10,000
9,000
1,000
0 1.80: 1.25 1.50 1.75 2.00 2.25 2.50
STOICHIOMETRIC RATIO
Figure 21. Carbon monoxide emissions versus stoichiometric ratio.
45
-------
25
20
o>
3
CO
O
CQ
CC
<
O
2
Q
15
10
3000
2700
2400
2100
1800
1500
1200
900
600
300
n.
Q.
O
OQ
OC
oc
Q
0 1.00 1.25 1.50 1.75 2.00 2.25 2.50
STOICHIOMETRIC RATIO
Figure 22. Gaseous hydrocarbon emissions versus stoichiometric ratio.
46
-------
2.0
01
J£
\
O)
CO
LLJ
O
x
O
g
1.5
1.0
0.5
NOTE: NOX INCLUDES APPROX.
10% NO,,90% NO.
60
50
40
30
20
10
0 1.00 1.25 1.50 1.75 2.00 2.25 2.50
STOICHIOMETRIC RATIO
Q.
Q.
CO
LU
Q
cc
i
Figure 23. Nitrogen oxides emissions versus stoichiometric ratio.
47
-------
CD
3
2.0
1.5
1.0
0.5
1.0
1.2
I
THEORETICAL
1.4
1.6
1.8
2.0
STOICHIOMETRIC RATIO
Figure 24. Sulfur oxides emissions versus stoichiometric ratio.
48
-------
TOTAL HEAT CONTENT OF FUEL
O
O
LL.
U.
LLJ
O
<
\\\\\\\\\\\\\\
\ \HEAT ABSORBED & RADIATED
WARM AIR HEAT GAIN
INCOMPLETE
COMBUSTION
HEAT LOSS
FLUE GAS HEAT LOSS
1.00 1.25 1.50 1.75 2.00 2.25 2.50
STOICHIOMETRIC RATIO
Figure 25. Experimental furnace heating efficiency.
49
-------
lowest air/fuel ratio because the oxygen concentration is low and mixing
is poor due to low turbulence. As the air/fuel ratio increases, the NO
^
emissions increase until a peak is reached at an air/fuel ratio of 2.25.
Beyond that point, lower temperatures and shorter residence times (because
of the high air/fuel ratios) cause a decrease in NO emissions.
/\
Note that the air/fuel ratio values obtained during this work are
unique to the burner tested. However, this study did establish general
trends which are applicable to most high-pressure atomizing-gun burners.
As a result of the work relating air pollutant emissions to air/fuel
stoichiometry, it was evident that better combustion was needed. Emissions
of smoke, particulates, CO, and HC were too high at air/fuel ratios below
1.6, but NO and S02 emissions were reduced. These results indicated that
it was desirable to operate at a lower air setting to get higher furnace
efficiency and lower NO and SOp emissions if combustible emissions could be
reduced. Increased residence time and/or combustion modification were the
most apparent methods to improve both combustion and furnace efficiency
and to reduce NO and SOp emissions.
RESIDENCE TIME
This study was performed to determine the effects of residence time on
air pollutant emissions . The experimental furnace used in the stoichiometric
studies was also used for these tests, except that the height of the
combustion chamber was increased from 15 to 27 inches (a factor of 1.8), to
allow a significant variation in combustion product residence time.
Residence times for the 27-inch and 15-inch combustion chambers, and
a typical domestic furnace (Williamson Temp-0-Matic Lo Boy) are compared
in Table 10. The residence times were calculated for each air/fuel ratio
by dividing the volume of the combustion chamber by the flue gas flow rate
at the average combustion chamber temperature.
50
-------
Table 10. FURNACE RESIDENCE TIME COMPARISON
Excess
Air, %
0
25
50
75
100
125
150
Residence time, seconds
Typical domestic
Furnace combustion
chamber
0.429
0.342
0.287
0.253
0.232
0.216
0.209
Experimental furnace
(15-inch combustion
chamber)
0.398
0.315
0.265
0.233
0.213
0.199
0.188
Experimental furnace
(27-inch combustion
chamber)
0.727
0.578
0.485
0.427
0.390
0.365
0.345
Theoretical values of C02 and 02 (dry) in the flue gas were
calculated based on the carbon/hydrogen ratto of the test fuel. These
theoretical values appear in Figure 26 along with data taken for the studies
with long and short residence times. As shown in Figure 26, actual C02-02
curves for the longer residence time conform much more closely to the
theoretical curves than the curves for shorter residence times, reflecting
the degree of incomplete combustion with shorter residence times.
Particulate emissions were reduced drastically by increased residence
time as shown in the semi-log plot of Figure 27. The curve which describes
the emissions at a longer residence time (dotted line) shifted down and
to the left of the curve representing a short residence time (solid line).
With a longer residence time the minimum is moved to a lower air/fuel
ratio, and the quantity of emissions at the minimum point is reduced.
51
-------
en
ro
X
O
LU
O
X
O
s
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
SHORT RESIDENCE TIME
(15-in. chamber)
LONG RESIDENCE TIME
(27-in. chamber)
1.2
1.4
1.6
1.8
2.0
2.2
2.4
STOICHIOMETRIC AIR/FUEL RATIO
Figure 26. Effect of residence time on carbon dioxide and oxygen.
-------
0)
-*
o>
OC
(J
Q
LU
OC
SHORT RESIDENCE TIME
(15-in. chamber)
LONG RESIDENCE TIME
(27-in. chamber)
0.02
0.01
1.0 1.1
1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1
STOICHIOMETRIC AIR/FUEL RATIO
Figure 27. Effect of residence time on participate emissions.
2.2 2.3 2.4
53
-------
As expected, the smoke results shown in Figure 28 are similar
to those for particulate weight. The minimum again occurred at a lower
air/fuel ratio. Also shown in Figure 28 are data obtained in a con-
12
current project . The smoke curve for the standard equipment furnace
falls between the curves for the different residence times. (This
standard equipment furnace was used as the basis for calculation of
residence times in a typical domestic furnace as reported in Table 10.)
It is clear that smoke was reduced significantly by longer residence
time.
CO and gaseous HC data are shown in Figures 29 and 30, respectively.
In both cases increased residence time shifted the minimum toward a lower
air/fuel ratio. Also, the quantity of emissions at the minimum was
reduced significantly.
NO emissions are plotted in Figure 31. (These values, reported as
NOX, include about 90 percent NO and 10 percent NO^.) NO emissions from
longer residence times were somewhat greater than those from tests at
shorter residence times. At an air fuel ratio of 1.4 the increase was about
16 percent. However, NO emissions were 17 percent less for the longer
A
residence time when compared to the NO emissions at the short residence
time when the excess air in both cases was adjusted to give equivalent
Bacharach smoke indices of 1, as is the practice of burner-furnace service
men. (See Appendix A for explanation.)
Since NO levels are well below equilibrium, as shown in Figure 32,
the formation reaction rate is controlling the NO quantity. This explains
why an increase in residence time can be expected to result in an increase
in NO emissions.
Emissions of SO were virtually unaffected by the change in residence
J\
time, as shown in Figure 33. This indicates that sulfur in the fuel
oxidizes very rapidly.
54
-------
Chapter 2 BACKGROUND
For a number of years estimates of concentra-
tions were calculated either from the equations of
Sutton (1932) with the atmospheric dispersion
pnrameters C,, C«, and n, or from the equations of
Bosanquet (1936) with the dispersion parameters
p and q.
Hay and Pasquill (1957) have presented experi-
mental evidence that the vertical distribution of
spreading particles from an elevated point is re-
lated to the standard deviation of the wind eleva-
tion angle,
-------
30
25
20
0)
UJ
O
15
10
LONG RESIDENCE TIME
(27-in. chamber)
SHORT RESIDENCE TIME
(15-in. chamber)
"I l"=""
1.0
1.2
1.4 1.6 1.8 2.0
STOICHIOMETRIC AIR/FUEL RATIO
Figure 29. Effect of residence time on carbon monoxide emissions.
2.2
2.4
56
-------
SHORT RESIDENCE TIME
(15-in. chamber)
LONG RESIDENCE TIME
(27-in. chamber)
1.0
1.4 1.6 1.8 2.0
. STOICHIOMETRIC AIR/FUEL RATIO
Figure 30- Effect of residence time on gaseous hydrocarbon emissions.
57
-------
2.0
1.5
3
O>
-*
\
O>
CO
g i.o
X
O
LONG RESIDENCE TIME
(27-in. chamber)
SHORT RESIDENCE TIME
(15-in. chamber)
O
DC
NOTE: NOX INCLUDES APPROX.
10% N02. 90% NO.
0.5
1.0
1.2 1.4 1.6 1.8 2.0
STOICHIOMETRIC AIR/FUEL RATIO
Figure 31. Effect of residence time on nitrogen oxides emissions.
2.2
2.4
58
-------
-------
2.5
2.0
THEORETICAL
J 1.6
CO
LU
Q
X
O
CO
1.0
0.5
LONG RESIDENCE TIME
(27-in. chamber)
SHORT RESIDENCE TIME
(15-in. chamber)
1.0
1.2 1.4 1.6 1.8 2.0
STOICHIOMETRIC AIR/FUEL RATIO
Figure 33. Effect of residence time on sulfur oxides emissions.
2.2
2.4
60
-------
In general, the increase in residence time provided a noticeable
decrease in total pollutant emissions and allowed the furnace to be operated
thermally more efficiently; i.e., at a lower air/fuel ratio.
Although these studies show that residence time is an important factor
in the control of air pollutant emissions, further work is needed to determine
the extent to which residence time can be used to control emissions. An
optimum residence time would allow enough time for approaching complete
combustion (i.e., for minimizing CO, HC, and particulate emissions), and
simultaneously be short enough to minimize NO formation.
A
COMBUSTION CHAMBER EFFECTS
Before studying combustion modification, it was decided to find the
effects of combustion chamber configuration and material on air pollutant
emissions. A refractory cylindrical chamber (fired into radially), a steel
cylindrical chamber (fired into axially), and a refractory square chamber
were tested under the same conditions. Combustion was poor in the steel
chamber, presumably because it was non-refractory, not because of its shape.
It was assumed that the non-refractory steel chamber would create a cold
wall effect which would quench the flame before combustion was complete,
thus producing excessive amounts of CO, HC, and smoke. In order to achieve
acceptable combustion with the steel chamber, the burner had to be fired at
a higher stoichiometric ratio, which reduces efficiency. Combustion in
refractory-lined combustion chambers was good; there was very little difference
in results with different chamber configurations. It is thus concluded that
combustion chamber configuration has little effect on emissions if the
chamber volume and dimensions are similar. If, however, the volume or
dimensions are significantly different, chamber configuration can become
very important with respect to the production of pollutant emissions .
It was also determined that the material used in the combustion chamber can
have a great effect on emissions of CO, HC, smoke, and particulates. NO
/\
emissions were unaffected in all cases, indicating that most of this
pollutant is formed early in the combustion process.
61
-------
COMBUSTION IMPROVING DEVICES
Since the air/fuel stoichiometry and residence time studies indicated
a need for improved burner performance, combustion modification was considered
as a possible method of reducing air pollutant emissions. Five commercially
available combustion-improving devices were tested to determine their
effects on furnace efficiency as well as on emissions of smoke, CO, total
gaseous HC, and NO . Each device was designed to improve the combustion of
A
high-pressure gun-atomizing burners used in domestic oil-fired furnaces by
improving the air/fuel mixture.
Plots comparing emissions for a range of air/fuel ratios are shown
for average smoke, 10th minute smoke, CO, gaseous HC, and NO in Figures 34
A
through 38, respectively. Comparisons of efficiency are shown in Figure 39.
Compared with the standard burner, only the Union flame control device
substantially reduced average smoke levels. The Gulf Econo-Jet reduced
average smoke levels slightly with greater reduction at air/fuel ratios
above 1.8. However, when the burners are compared at actual operating
conditions, as shown in Table 4, only one burner tested had higher average
smoke than the standard, and several burners significantly reduced the
level of smoke emissions. For example, the Union device reduced smoke by
almost 60 percent when compared at a No. 1 smoke level at "steady state"
(hot running) conditions (see Appendix A).
The Union device reduced emissions of both CO and gaseous HC over most
of the operating range of air/fuel ratio. At low air/fuel ratios all
devices produced roughly the same levels of CO and HC, generally lower than
those of the standard burner.
Even though the Monarch and Delavan devices slightly reduced NO
/\
emissions at air/fuel ratios above 1.45 and 1.75, respectively, only the
Union device produced substantially less NO than the standard equipment,
A
and then only at air/fuel ratios above 1.65. When the devices are compared
at actual operating conditions (see Appendix C) all of them produce levels
62
-------
10
00
5 5
i4
DEVICE
STANDARD ABC
MONARCH COMBUSTION HEAD
DELAVAN FLAMECONE
SHELL HEAD
GULF ECONOJET
UNION (PURE) FLAME RETENTION HEAD
1.0 1.2 1.4 1.6 1.8 2.0 2.2
STOICHIOMETRIC AIR/FUEL RATIO
2.4
2.6
2.8
Figure 34. Average smoke emissions of combustion improving
devices versus stoichiometric ratio.
63
-------
DEVICE
STANDARD ABC
MONARCH COMBUSTION HEAD
DELAVAN FLAMECONE
SHELL HEAD
GULF ECONOJET
UNION (PURE) FLAME RETENTION
1.4 1.6 1.8 2.0 2.2
STOICHIOMETRIC AIR/FUEL RATIO
Figure 35. 10th-minute smoke emissions of combustion
improving devices versus stoichiometric ratio.
64
-------
10
01
6
X
O
O
O
o>
cc
<
O
DEVICE
STANDARD ABC
MONARCH COMBUSTION HEAD
DELAVAN FLAMECONE
SHELL HEAD
GULF ECONOJET
UNION (PURE) FLAME RETENTION HEAD
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
STOICHIOMETRIC AIR/FUEL RATIO
Figure 36. Carbon monoxide emissions of combustion
improving devices versus stoichiometric ratio.
2.6
65
-------
2.25
2.00
DEVICE
STANDARD ABC
MONARCH COMBUSTION HEAD
DELAVAN FLAMECONE
SHELL HEAD
.- GULF ECONOJET
UNION (PURE) FLAME RETENTION HEAD
0.25
1.0
1.2
1.4 1.6 1.8 2.0 2.2
STOICHIOMETRIC AIR/FUEL RATIO
Figure 37. Gaseous hydrocarbon emissions of combustion
improving devices versus stoichiometric ratio.
66
-------
DEVICE
STANDARD ABC
\\ _
.. MONARCH COMBUSTION HEAD
DELAVAIM FLAMECONE
- SHELL HEAD
GULF ECONOJET
UNION (PURE) FLAME RETENTION HEAD
1.4 1.6 1.8 2.0 2.2
STOICHIOMETRIC AIR/FUEL RATIO
Figure 38. Nitrogen oxides emissions of combustion
improving devices versus stoichiometric ratio.
67
-------
100
90
80
70
60
iii 50
o
40
30
20
10
1 I I I
I I
DEVICE
STANDARD ABC
MONARCH COMBUSTION HEAD
DELAVAN FLAMECONE
SHELL HEAD
GULF ECONOJET
UNION (PURE) FLAME RETENTION HEAD
I I I I I I
D 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8
STOICHIOMETRIC AIR/FUEL RATIO
Figure 39. Overall heating efficiencies of combustion improving devices
versus stoichiometric ratio.
68
-------
of NO higher than those of the standard burner.
/\
When compared with the standard burner, the Shell and Gulf devices
had only slightly higher furnace efficiencies, whereas the Union device
increased efficiency appreciably since it operates at a lower air/fuel
ratio while producing a No. 1 smoke at "steady state" conditions.
During these tests, problems were encountered with defective burner
equipment: various brands of nozzles showed non-uniform spray characteristics,
one nozzle was defective; and in another case, other parts of the combustion
improving devices were defective. After replacement or repair of the
defective parts, the burners performed correctly, indicating that product
consistency and quality control are critical in the manufacture of burner
equipment, especially nozzles.
These tests indicated that only the Union flame control device sub-
stantially reduced smoke and increased efficiency when compared to the
standard equipment (ABC Model 45 burner). None of the devices reduced NO
A
when compared under actual operating conditions (see Appendix C). CO and
gaseous HC emissions were about the same for all burners at low excess air
levels.
Although these devices were designed to improve combustion in older
inefficient furnaces (rather than in new, more efficient ones), one of the
devices (the Union flame control device) reduced the smoke and increased
the efficiency of a new, well-designed burner!, This work indicates that
certain combustion-improving devices offer potential for reducing levels
of one or more pollutants and for improving combustion efficiency.
FLAME RETENTION BURNERS
The studies with combustion improving devices identified the Union flame
control device, which utilizes flame retention, as being most effective of
those tested in reducing pollutant emissions and increasing combustion efficiency.
To verify these results and to further investigate flame retention, nine different
69
-------
burners, all featuring retention-type end cones, were tested: the ABC
Mite, the Beckett Bantam, the standard burner (ABC Model 45) modified with
the Union flame control device, the Sun-Ray, the White-Rogers, the Wayne,
the Union, the U. S. Carlin, and the Esso Model 40.
The ABC Mite, Beckett Bantam, and Union device are shown in Figures 12
through 14, respectively. The ABC Mite differs from the others in that
its blast tube diameter is much smaller. The Union device differs in that
inlet air is controlled by sliding an air shield within the blast tube
rather than by manipulating shutter vanes on the burner housing. The
dimensions and operation of the Beckett Bantam, however, are the same as
those of conventional burners. Each burner employs some form of shield,
cone, or ring to which the flame "attaches" thus creating the flame retention
effect.
The performance of the ABC Mite, the Beckett Bantam, and the standard
with Union flame control device was superior (i.e., higher furnace efficiency,
lower smoke, and ho increase of CO and HC) to that of the standard and other burners
featuring flame retention. Results of tests on the standard and superior
performing burners are listed in Table 5 with the emissions of each.
The Union device had very high efficiency and low smoke emissions, but
high NO emissions. High NO emissions would be expected from flame retention
A X
devices because of their more compact, intense flame resulting in higher
flame temperatures. However, the ABC Mite had both low smoke and low NO
/\
emissions. The Mite's efficiency was higher than that of the standard but
lower than that of the burner modified with the Union device. The reason
for the ABC Mite's lower NO emissions is not known.
A
Table 5 shows that there is no difference between burners in CO or HC
levels. All three flame retention burners and the standard produced 0.06
grams of HC and 0.5 grams of CO per kilogram of fuel burned. These levels
are quite low compared to other sources (e.g., automobiles have HC and CO
emissions of approximately 18 and 188 grams per kilogram of fuel, respectively).
70
-------
These tests indicated that flame retention burners can be operated
at low excess air levels without producing excessive smoke. This results
in an increase in combustion efficiency and a corresponding reduction in
total air pollutant emissions since less total fuel is required for a given
heat load. The most important result of this work was that one burner
(ABC Mite) was identified as reducing both smoke levels and NO . A study
/\
is now underway to define critical variables in burner design and maintenance
which affect combustion efficiency and air pollution emissions. From this
work a comprehensive manual will be written for distillate oil burner manu-
facturers and servicemen. This manual, which will describe the correct design,
operation, and maintenance procedures to control air pollutant emissions
while maintaining the highest combustion efficiency,is expected to be available
by early 1974 (Contract No. 68-02-0017 with Rocketdyne).
CYCLIC-BASED EMISSIONS
The study of air/fuel stoichiometry effects on air pollutant emissions
indicated that some pollutants have sizeable peaks during ignition and/or
shutdown. In some cases these peaks account for most of the pollutant
8 9
emissions. These findings were confirmed during subsequent studies ' .
CO and gaseous HC emissions both peak during ignition and after burner
shutoff as shown in Figure 40. HC emissions return to near-zero after the
initial peak and remain low until the burner goes off. Peaks are also responsible
for much of the CO emissions. However, in this case the emissions tend to
reach a measurable "equilibrium" value between peaks.
Both the smoke spots and particulate matter peak during ignition and
taper off continuously for the remainder of the cycle. Figure 41 shows that
by the end of the cycle very little smoke or particulate matter is emitted.
NO emissions do not peak like the other pollutants. Figure 42 shows
that, after the initial jump, the emission level rises at a fairly steady
rate. Although the NO emissions would eventually reach equilibrium and level
off, the on-time of most domestic burners during a cycle is not long enough
7.1
-------
BURNER
ON
HYDROCARBON
CONCENTRATION
BURNER
OFF
CO CONCENTRATION
TIME
Figure 40. Hydrocarbon and carbon monoxide trends
during cycle.
72
-------
BURNER
ON
SMOKE
CONCENTRATION
BURNER
ON
PARTICULATE
CONCENTRATION
TIME
Figure 41. Smoke and paniculate trends during cycle.
73
-------
BURNER
OFF
BURNER
ON
NO
CONCENTRATION
TIME
Figure 42. Nitric oxide trend during cycle.
74
-------
for this to occur.
These peak emissions during ignition and/or shutdown are caused by
variation in the combustion chamber temperature. At ignition, a cold refractory
will not support complete combustion, thus producing peaks of CO, HC, and
smoke. The source of post-burn emission is fuel leakage from the nozzle.
The nozzle absorbs heat from the hot refractory and its temperature increases
rapidly after the oil flow is shut off. The oil in the nozzle expands and flows
from the nozzle. As the fuel drips into the hot combustion chamber it vaporizes
and is partially oxidized, thus producing HC and CO.
The peak emissions during ignition could be eliminated or reduced by
keeping the refractory warm during the burner-off period. The most apparent
possibilities are pilot or modulating burners. However, a more sophisticated
control system which cycles the burner more frequently, to maintain a hot
refractory, is also a possible solution for ignition emissions. Post-burn
emissions might be eliminated by: adding a solenoid cutoff valve in the
fuel line, using a clutch to stop the fuel flow before the fan stops, or cooling
the combustion chamber refractory rapidly after the burner is shut off.
It is important to recognize, however, that cyclic emissions mainly affect
emissions of CO and HC both of which are very low for oil burners. Smoke and
particulates can be reduced more effectively by means other than those mentioned
above; e.g., the to-be-developed optimum distillate oil burner discussed earlier.
NATURAL GAS BURNERS
Both a Williamson Mono-Port and a Bryant Sectionalized gas burner were
compared with oil burners. Each was rated at 100,000 Btu/hr, equivalent to
that of the oil burners tested previously.
The tests showed that natural gas combustion with residential burners
produce emission levels similar to those from distillate oil combustion.
Results of these tests are shown in Figures 43 through 48.
75
-------
0.32
0.28
0.24
0.20
S 0.16
x
o
o
CO
0.12
0.04
0
1 I T
GAS BURNERS
1 T
O BRYANT SECTIONALIZED (BOILER)
A BRYANT SECTIONALIZED (FURNACE)
O WILLIAMSON MONO - PORT (FURNACE)
c?
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
STOICHIOMETRIC RATIO
Figure 43. Carbon monoxide emissions for gas-fired units.
0.32
0.28
0.24
0.20
S 0.16
o
z
o
o
CO
a:
0.12
0.08
0.04
OIL BURNERS i
STANDARD ABC MODEL 45 I
UNION FLAME CONTROL DEVICE |
I _
1.0 1.2 1.4 1.6 1.8 2.0
STOICHIOMETRIC RATIO
2.2 2.4
Figure 44. Carbon monoxide emissions for oil-fired units.
" 76
-------
8
o
3
'g, .0160
s
ul
| .0120
i
LU
5=
3 .0080
c-o
0
CO
« .0040
0
o
cc
o
= n
1 1 1 1 1 1 1
GAS BURNERS
O BRYANT SECTIONALIZED
JL A BRYANT SECTIONALIZED
l_
a WILLIAMSON MONO-PORT
A
A
A
CD O
1 ^*efr- tP 1 1 1 1
1
(BOILER)
(FURNACE)
(FURNACE)
1
1.0 1.2 1.4 1.6 1.8 2.0 2.2
STOICHIOMETRIC RATIO
Figure 45. Hydrocarbon emissions for gas-fired units.
2.4 2.6
oo .
-------
CO
1
3
^
DO
LLj
Q
X
O
O 1
fc1
lUb
0.14
0.12
0.10
0.08
0.06
0.04
0.02
n
1 1 1
O
O
& ^
_ 0 A _
a
9
CD
an
a
GAS BURNERS
O BRYANT SECTIONALIZED (BOILERS)
A BRYANT SECTIONALIZED (FURNACE)
D WILLIAMSON MONO-PORT (FURNACE)
1 1 1
1.0 1.2 1.4 1.6 1.8 2.0 2.2
STOICHIOMETRIC RATIO
Figure 47. Nitric oxide emissions for gas-fired units.
2.4
0.16
0.14
0.12
0.10
0.08
x
O
0.06
0.04
0.02
I -LI I
OIL BURNERS
STANDARD ABC MODEL 45
UNION FLAME CONTROL DEVICE
ABC MITE
1.0 1.2 1.4 1.6 1.8 2.0
STOICHIOMETRIC RATIO
2.2 2.4
Figure 48. Nitric oxidj^emissions for oil-fired units.
78
-------
NO emissions from the natural gas burners were about the same as those
/\
from most high-pressure atomi zing-gun oil burners. The exceptions to this
were the Williamson Mono-Port gas burner and the ABC Mite oil burner whose
NO emissions were lower than emissions of the others tested.
A
Hydrocarbon emissions from the natural gas burners were generally slightly
lower than those from oil burners. CO emissions were about the same with gas
firing as with oil firing. Of course, smoke emissions from properly adjusted
gas burners are negligible.
OTHER DISTILLATE OIL BURNERS
The domestic heater studies also included testing of experimental
burners and commercially available burners other than the high-pressure
atomi zing-gun type. Two vaporizing burners were tested. One such burner
was a vertical wall flame burner which vaporized fuel prior to combustion;
the other was a blue flame burner which used a combination of low pressure
atomi zati on and vaporization. All other burners tested atomized the fuel
and most of those were the low-pressure type. Several high-pressure atom-
i zati on blue flame burners were tested; all but one utilized internal
recirculation where the combustion products were recycled within the
combustion chamber. (The exception incorporated external recirculation
where the combustion products were taken from the combustion chamber and
returned externally to the air inlet section.) Each of these burners
is discussed in Appendix D; results of their tests are given in Appendix C.
Of the other oil burners tested, the Rocketdyne Una Spray burner shows
some promise of commercial development. It has a unique method for atomizing
the fuel. A thin film of oil flows over a small hollow glass sphere in which
there is a very short narrow slit. Air, forced into the sphere at low
pressure, atomizes the oil film as it emerges through the slit. The burner,
more compact than other burners with the same rating, was designed for easy
servicing. The efficiency of the prototype unit which was tested was high
and smoke emissions were acceptable, but NO emissions were extremely high.
79
-------
The blue flame burner also shows promise. It offers high efficiency, low
smoke, and low NO emissions; however, thus far none have performed well during
A
ignition. There are two main causes for poor combustion during ignition with blue
flame burners: (1) most of these burners rely upon recirculation which is not estao-
lished immediately; and (2) blue flame burners usually rely upon a hot surface sur-
rounding the flame envelope, also not established immediately. Because most blue
flame burners have either internal or external recirculation: smoke, NO , and noise
level are reduced and the burner can be operated at low air/fuel ratios.
IGNITION SYSTEMS
Tests were made to determine the effect of ignition systems on pollutant
emissions from domestic burners. Most ignition systems supply a high-energy
arc to ignite the oil spray as it leaves the nozzle; ignition is usually on
continuously during the burning cycle. Even though modern burners do not
require a continuous ignition arc to stabilize the flame, most burners operate
with the ignition on during the on-period. The main reason is that controls
for turning the arc off after the flame is established add a few dollars to
the cost of the burner.
It was found that NO was the only pollutant affected. Smoke, CO, HC,
Op, and COp were not changed.
Three different ignition systems were tested as shown in Table 7. A
Franceformer (Cat. LKJ) ignition system was tested with a Williamson (ABC
Model 45) burner. First, the burner was run with the ignition on. It was
then run at the same air/fuel ratio with the ignition disconnected after
the flame was established. With the ignition off, there was a reduction in
NO of 7 ppm (approximately 10 percent). This system was also tested with a
Union (Pure) flame retention device: the average reduction during the first
four runs without ignition was 7.8 ppm (about 7 percent); during the fifth
experiment, the oil was turned off, thus no combustion. Therefore, any NO
produced had to come from the electrical discharge of the ignition system.
In these tests there was an average NO emission of 9.0 ppm.
The second ignition system tested was a Franceformer (Cat. 4LACYU-4)
on a Beckett Bantam burner. The burner was first run with the ignition on.
80
-------
Then, at the same air/fuel ratio, with the ignition turned off, NO was
reduced by 9.0 ppm (about 10 percent).
The third unit tested was a Prestolite 0-120 ignition system, on an
ABC Mite. This system showed no reduction in NO when the ignition was
turned off. It is assumed that NO emissions were not affected by this
ignition system because it had a lower power output than the others.
These tests indicated that ignition systems have no effect on smoke,
CO, HC, Op, or COp emissions. Only NO emissions were affected, and even
then the increase was relatively small.
NOZZLE EFFECTS
Because of problems encountered with nozzles during tests with combustion
improving devices and since approximately 95 percent of the domestic oil
burners in the U. S. are high-pressure atomizing-gun burners which use
nozzles to atomize fuel oil, it was necessary to determine the effect of
nozzles on air pollutant emissions. Four different brands of 80-degree,
hollow-cone nozzles (Delavan, Monarch, Hago, and Steinen) rated at 0.75 gph
were tested. Although all nozzles were new, each was examined microscopically
before testing: all flaws were photographed. The tests were designed to
determine if there was any difference in emission levels within a given type
and brand of nozzle, and between different brands of nozzle to obtain
minimum emission levels.
To obtain a random sample, two nozzles of each brand were bought from
five different stores. The tests were made with a Williamson (ABC Model 45)
burner at an air/fuel ratio of approximately 1.60. By operating at a constant
air/fuel ratio the emissions of each nozzle can be directly compared. A
statistical analysis of the data was performed upon completion of the tests.
The results are given in Table 8.
The average fuel rate for all nozzles was 0.82 pounds/10 minutes
(0.70 gph). The analysis showed that measurements of smoke, gaseous HC,
NO, CO, C02> and efficiency were significantly different when comparing
nozzles within a brand and when comparing different brands of nozzles.
All tests were made at a confidence level of 95 percent. Since emissions vary
81
-------
significantly between nozzles, burners should always be readjusted
when the nozzle is replaced.
The pre-test photomicrographs of all nozzle flaws qualitatively
suggest a basis for the results that were obtained from the data. As
shown in Table 8, emissions of 10th minute smoke, average smoke, and HC
were lowest from Delavan; Monarch was second, followed by Hago. Steinen
had the highest emission levels.
In the following discussion reference to smoothness and cleanliness
is in relation to the nozzle distributor (or metering disk) and the
nozzle tip around the orifice. These observations were made with a
microscope, thus a surface which appears smooth without the aid of
magnification may appear very rough under the microscope. The photo-
micrographs showed that Delavan nozzles were smoothest and were
relatively clean. Monarch nozzles were rough but, although oily, were
clean. Hago nozzles, more smooth than Monarch and less smooth than
Delavan, were very dirty: there were many small particles, chips, and oil
droplets on the nozzle surfaces. Steinen nozzles were rough and their
surfaces were covered with many small particles, chips, and oil droplets.
Because of the design of these tests it is not possible to recommend one
brand of nozzle over another. Even though some nozzles had better
performance than others, those with poorer performance may have been
improved if the excess air had been adjusted. The following conclusions
can be drawn, however:
1. There is a significant difference between nozzles of the
same brand.
2. There is a significant difference between nozzles of
different brands.
3. The furnace or boiler burner should always be readjusted
when changing nozzles.
It would be beneficial if nozzle manufacturers would conform to standards
strict enough to guarantee to the consumer that a nozzle with a given rating
(flow rate, spray angle, and spray pattern) will not vary among brands or within
the same brand over a period of time. A too-strict standard, hr^ever, would
result in an unnecessary increase in nozzle price to the consumer.
82
-------
EFFECT OF TIME ON TUNING
Tests were also made to determine the effect of time on burner per-
formance; i.e., after a burner is adjusted, how rapidly its performance
deteriorates. The same furnace was used for these tests as was used with
the combustion improving and flame retention devices described previously.
With each new nozzle tested, the burner was adjusted for a No. 1 tenth
minute smoke spot. The burner was then cycled randomly to simulate the
cycle of a furnace in a home. The test was duplicated to establish
variability and reliability.
The results of the tests are presented in Figures 49 and 50. As shown
in Figure 46 neither CO nor gaseous HC changed during the 10-week test
period, and the change in NO was very small. However, the most dramatic
change was with smoke, which changed from just below a No. 1 Bacharach
number to near a Bacharach No. 6 in the 10-week period. As shown in
Figure 50 the heating efficiency dropped from 78 percent to 68 percent
during the same period.
This shows that air pollutant emissions, at least of smoke and
particulate, can be significantly reduced and heating efficiency can be
increased by adjusting the burner periodically. To determine whether
the results of these tests were typical of results which would be
obtained in actual practice, Battelle made follow-up measurements at
2-month intervals twice during the heating season as part of a field
test program (Contract No. 68-02-0251).14
Of the four residential units on which follow-up measurements were
made, the emissions of two remained essentially unchanged. On one unit
C02 dropped by 8 percent and emissions of CO and HC increased, but not
to significant levels. Smoke, particulate, CO, and HC increased on the
fourth unit but the change was due to nozzle clogging which resulted from
dirty fuel.
As a result of Battelle's field tests it is felt that the in-house
results are not typical of most burners, but after tuning, the perform-
ance of some burners does deteriorate at a faster rate than the performance
of most burners. To determine the seriousness of this problem further
field tests are needed.
83
-------
1.4
1.3
1.2
1.1
1.0
0.9
I 0.8
O!
j:
\
S ,_ 0.7
5
O
O
a.
0.6
0.5
0.4
0.3
0.2
0.1
O NO
SMOKE
D CO
A HC
5 6
TIME, weeks
10
CO
^
(0
-C
o
o
6 3S
tu
I-
5 I
o
4
10
Figure 49. Change in pollutant emissions with time -- light-oil residential heater.
-------
80 h
70
o
o
o= I
50!
. 2
4 5 ,6
TIME ELAPSED, weeks
.7
.10
Figure 50. Deterioration of furnace efficiency with time.
-------
BURNER MAINTENANCE
The importance of burner maintenance was shown in the field test study
14
performed by Battelle. It was found that tuning effectively reduces
emissions of smoke and CO, but has little effect on the mean values of other
pollutants. The most significant finding was that a major reduction in CO,
gaseous HC, and particulate can be achieved by identifying and replacing or
repairing units in bad condition. To exemplify this, some typical emission
levels for the various types of residential equipment tested are shown in
Table 11. The top line gives the average emissions for all units tested
whereas the data from units in need of replacement have been excluded from
the second line. By eliminating the three bad units the CO emissions were
reduced from greater than 3.07 to 1.08 g/kg and the gaseous HC emissions
were reduced from 0.79 to 0.10 g/kg, whereas NO remained essentially
/\
unchanged. Filterable particulate was reduced from 0.40 to 0.33 g/kg and
total particulate was reduced from 1.24 to 0.83 g/kg. In other words,
simply by eliminating equipment in need of replacement in the Battelle
study, the following reductions could be achieved: CO by at least 65 percent
(the exact percentage is not known since the instrumentation went off scale),
gaseous HC by 87 percent, filterable particulate by 17 percent, and total
particulate by 33 percent. Even these numbers are conservative since
Battelle only included burners which were under a service contract.
Therefore, instead of accounting for about 10 percent of the total burner
population, burners in need of replacement probably account for 20 to 30
percent of the total population. This strongly indicates that emissions
from residential sources could be reduced to an insignificant level simply
by using proper maintenance procedures which would identify units in need
of replacement and by providing proper tuning for the remainder.
86
-------
Table 11. COMPARISON OF AVERAGE EMISSIONS FROM RESIDENTIAL OIL BURNERS
14
Units
All units
All units
23 except those
in need of
replacement
Percentage reduction
when 3 bad units
were eliminated
Condition
As found
As found
Number
of units
in
sample
32
29
Bacharach
smoke
number at
5th minute point
(a)
3.2
Emission
Gaseous emissions
CO
>3.07(b)
1.08
>65
HC
0.79
0.10
87
N0x
2.7(
2.1't
factors, g/kg
Parti cul ate emissions
(modified EPA
Filterable
0.40
0.33
17
procedure)
Total
1.24
0.83
33
(a) Oily smoke spots from the 3 units in need of replacement prevented obtaining a meaningful average.
(b) The analytical instrumentation went off scale at 3.07 g/kg.
-------
PROCEDURES FOR REDUCING POLLUTANT EMISSIONS
FROM CURRENT DISTILLATE-OIL-BURNING HEATERS
2
As part of one study , emission levels were calculated using fuel
usage data for a typical Northeastern city. During the peak winter months
domestic and commercial distillate-oil-fired units produced approximately
13 percent of the particulates emitted in that city. If all burners were
operated at an air/fuel ratio resulting in a Bacharach smoke level of no
greater than ^, the particulate produced from domestic and commercial heaters
would be less than 1 percent of the total. Obviously, to account for 13
percent of the total particulates many burners being used must be worn out,
poorly built, or maladjusted. Proper maintenance and quality control can
reduce this type of pollutant emission. It is recommended that all burners, boilers,
and furnaces be serviced by an authorized serviceman at least annually, normally
just prior to the heating season.
All domestic burners can be operated at lower total emission levels with
some sacrifice of efficiency, by operating the burner at a slightly higher than
normal air/fuel ratio. As shown in Figure A-2 (Appendix A), the standard
Williamson (ABC Model 45) burner would normally be operated at an air/fuel
ratio of 1.53. If it were operated at a higher setting, 1.60 for example,
emission levels of average smoke, CO, and HC would be lowered or remain constant.
This higher setting will also allow for a small amount of drift after the
burner is adjusted. If this burner were set at an air/fuel ratio of 1.53 and
later drifted to a setting of 1.50, the smoke emissions would increase
significantly, as shown in Figures A-2 and A-6.
14
As indicated in a related report by Battelle, which was jointly sponsored
by the Environmental Protection Agency and the American Petroleum Institute
under Contract No. 68-02-0251, a serviceman should not increase the excess
air level to unnecessarily high levels to reduce the smoke emissions. Besides
reducing efficiency unnecessarily the CO and gaseous HC emissions can rise
sharply at high air settings as shown in Figures A-3 and A-4. Since CO generally
increases before gaseous HC as excess air is increased, a serviceman- could be
89
-------
relatively certain that the gaseous HC emissions are low if the CO
emissions are low. In order to measure CO emissions on a routine basis,
an accurate, portable CO monitor must be developed.
Instrumentation for burner servicemen can be improved. If present
instruments for measuring COg. stack temperature, draft, and smoke could be
combined, the serviceman would be more likely to use the equipment. Better
training is needed for the serviceman; not only in how to use the equipment
properly, but in how to interpret the results. Better training will enable
him to adjust heaters for maximum efficiency with minimum levels of air
pollutant emissions.
In essence, the responsibility of reducing emissions from existing
burners lies with the individual (homeowner and serviceman). He should
utilize the methods described above and keep informed about future develop-
ments.
90
-------
CONCLUSIONS AND RECOMMENDATIONS
Tests showed that low air/fuel ratios provide maximum efficiency,
and that higher settings minimize emissions. Therefore, a compromise is
necessary to obtain a setting which will provide an acceptable level for
emissions without lowering the efficiency appreciably. The technique for
finding this setting is discussed in Appendix A. The excess air should be
set as low as possible without producing a smoke spot number greater than 1
under hot running conditionsJ i.e.,after 10 minutes of operation. A re-
fractory-lined combustion chamber, as opposed to a steel-lined chamber,
will allow burner operation at a lower excess air level.
Combustion chamber configuaration effects and combustion improving
devices (other than flame retention) were found to have little, if any,
effect on pollutant emissions. The study also indicated that some ignition
systems are capable of producing small amounts of NO ; whereas others, with
/\
lower power, have no effect. It was also shown that nozzles can be respon-
sible for unnecessary pollutant emissions. This can usually be corrected
by adjusting the burner.
It was found that residence time and flame retention devices have the
greatest effect on air pollutant emissions. A longer residence was shown
to reduce emissions of smoke, particulate matter, CO, and HC. However,
NO emissions may be increased slightly. One drawback to using a longer
/\
residence time for lowering pollutant emissions is that equipment manufac-
turers will have to make larger, bulkier furnaces or use lower fuel rates
in present designs. Also, it may be that longer residence time would have
a much smaller effect when more efficient burners with better performance
are used.
Studies showed that most flame retention burners have better performance
characteristics (higher efficiency with lower smoke and/or NO emissions)
/\
than those of a conventional high-pressure atomizing-gun burner. The ABC
Mite was the only flame retention burner tested that reduced both smoke and
NO emissions.
X
91
-------
These studies Indicate that several areas of further research are
necessary. \More work with the effects of residence time is needed to find
the best residence time for low emissions with high efficiency. Also, more
studies are needed for optimizing the burner design to improve fuel/air
mixing. A contract has been awarded to the Rocketdyne Division of Rockwell
International to design and develop an optimum distillate oil burner; i.e.,
a burner with high efficiency and low pollutant emissions. The results of
this work will be available for oil burner manufacturers and servicemen
early in 1974.
Field tests are needed for burners which were most promising in order
to demonstrate long term effectiveness. As a result of an in-house study
which indicated that burner performance can deteriorate significantly over
a period of 10 weeks after tuning, Battelle performed a limited number of
follow-up measurements during their field study . Measurements were made
at 2-month intervals twice during the heating season on four units. The
tests indicated that some units are more prone to performance deterioration
than others, but further tests are needed before precise conclusions can
be made.
Research is also needed to find a method of controlling cyclic-based
emissions. This could reduce pollutants such as smoke, particulates, HC,
and CO. Reducing or eliminating the ignition and/or shutoff peaks could
reduce pollutant emissions by 50 percent or more. The Rocketdyne work may
result in a significant reduction in cyclic emissions. Cyclic emissions
could also be reduced by utilizing modulation, or by using an undersized
oil nozzle in a furnace or boiler, since the burner would have a longer
on-time to meet a given heat load.
Tests in which natural gas was used as the fuel indicated that the
level of air pollutant emissions from residential gas burners is about the
same as that from equivalent-size oil burners. The exceptions to this were
the Williamson Mono-Port gas burner and the ABC Mite oil burner whose NO
A
emissions were lower than emissions of others tested.
It is also important to note that pollutant emissions from existing
domestic heaters could be reduced significantly by proper maintenance, and
by replacing poorly performing units as emphasized in the Battelle study .
92
-------
This includes servicing by an authorized serviceman at least once each
heating season, preferably at the beginning. The nozzle in an oil burner
should be replaced each season and the burner should be readjusted by using
proper equipment for measuring smoke and COo- The furnace air filters
should be changed several times during the heating season to avoid appreciable
reduction in furnace efficiency. Also, better instrumentation is needed for
burner and furnace servicemen. As a followup, a program should be initiated
for training servicemen to use the equipment properly. Organizations such
as the American Boiler Manufacturers Association (ABMA), the American
Petroleum Institute (API), the American Society of Heating, Refrigerating,
and Air-Conditioning Engineers (ASHRAE), the Hydronics Institute, the National
Oil Fuel Institute (NOFI), and the National Oil Jobbers Council (NOJC)
could play important roles in such a program.
93
-------
BIBLIOGRAPHY
1. Heller, A. N. Impact of Changing Patterns of Energy Use on
Community Air Quality, 57th Annual APCA Meeting. Houston,1964.
2. Howekamp, D. P. Flame Retention-Effects on Air Pollution.
Presented at 9th Annual NOFI Convention. Atlantic City,June 1970.
3. Dunphy, B. Oil Heating 1962 Estimated at 497,772. Fuel Oil and
Oil Heat.p.37, January 1963.
4. Mantho, M. Oil Heating Gains in 1970. Fuel Oil & Oil Heat. p. 38-41,
January 1971.
5. Steiner, K. Oil Burners, 3rd Ed. Heating Pub., Inc., N. Y.,
p. 335, I960.
6. Burkhardt, H. Domestic and Commercial Oil Burners, 2nd Ed.
McGraw-Hill, N. Y., p. 112-130, 1961.
7. Home Heating Efficiency Improvement Manual, Esso Standard Oil
Company, Newark,1959.
8. Wasser, J. H., R. P. Hangebrauck, and A. J. Schwartz. Effects of
Air-Fuel Stoichiometry on Air Pollutant Emissions from an Oil-Fired
Test Furnace. APCA Journal. 18:332-337, May 1968.
9. Martin, G. B., D. W. Pershing, and E. E. Berkau. Effects of Fuel
Additives on Air Pollutant Emissions from Distillate-Oil-Fired Furnaces.
U. S. Environmental Protection Agency, Research Triangle Park, N. C.
Publication No. AP-87. June 1971.
10. Dorsey, J. A. and J. 0. Burckle. Particulate Emissions and Process,
Chem. Eng. Progr. 67:92-96, August 1971.
11. Wasser, J. H., G. B. Martin, and R. P. Hangebrauck. Effects of
Combustion Gas Residence Time on Air Pollutant Emissions from an Oil-
Fired Test Furnace. Presented at NOFI Workshop, (September 17-18, 1968).
12. Hooper, M., D. P. Howekamp, and R. P. Hangebrauck. Effects of
Combustion Improving Devices on Air Pollutant Emissions from Residential
Oil-Fired Furnaces. Presented at NOFI Workshop, (September 24-25, 1969).
13. Dickerson, R. A. and A. S. Okuda. "Design of an Optimum Oil Burner
for Control of Pollutant Emissions," Rocketdyne Division-of Rockwell
International, Twenty-Third Technical Progress Narrative for EPA Contract
No. 68-02-0017, June 1, 1973.
14. Barrett, R. E., S. E. Miller, and D. W. Locklin. Field Investigation of
Combustion Emissions from Space Heating Equipment. Report No.
EPA-R2-73-084a, NTIS No. PB 223-148, June 1973.
94
-------
APPENDIX A
BURNER ADJUSTMENT AND COMPARISON
To determine the operating and emission characteristics of the
burners, emission measurements were made over a wide range of air
settings. Typical results for tests of this type are shown in
Figure A-l. Burner X appears to have lower emissions than burner Y.
When comparing the two at one air/fuel ratio, that is true; however,
when comparing them under actual operating conditions (i.e., at an
air/fuel ratio where each operates efficiently with low smoke
emissions) the burners may operate at different air/fuel ratios.
In the case mentioned above, burner X may normally operate at an
air/fuel ratio of 1.2 and burner Y at 1.6. The dotted lines show
that burner Y actually produces lower emissions than burner X, on
an actual operating basis.
For this reason a method was chosen which permitted the air/
fuel ratio to be found at which each burner would operate normally.
This setting was determined by using a technique employed by oil
burner servicemen in adjusting furnaces. Since heating efficiency
increases as the air/fuel ratio decreases, the air settings are
usually set as low as possible without producing excessive smoke
(>No. 1 smoke spot) at hot running conditions. Therefore, for
purposes of comparison, the stoichiometric ratio was found for
each burner at which a Bacharach No. 1 smoke spot was recorded at
"steady state" conditions (after 10 minutes of operation). This
procedure is illustrated in Figure A-2 for the Williamson (ABC
Model 45) burner.
The various burners can be conveniently compared by reading the
values of CO, HC, NO, average smoke spots, and efficiency at the
95
-------
(D
0>
a>
0.
I
BURNER Y
BURNER X
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
STOICHIOMETRIC RATIO
Figure A-1. Burner operating and emission characteristics.
96
-------
10
cc
LU
CD
O
I 4
O
<
"
I |_J
0 I.Oi 1.2 1.4, 1.6 1.8 2.0 ;2.2 2.4 2.6
STOICHIOMETRIC RATIO
Figure A-2. Determination of stoichiometric ratio
for No. 1 10th-minute smoke.
97
-------
stoichiometric ratio at which a No. 1, 10th minute smoke spot was
produced. Figures A-3 through A-7 show this procedure for the
Williamson (ABC Model 45) burner.
Results of the emissions tests for all burners are in Appendix C.
All burners tested or investigated are classified as high-pressure or
low-pressure atomizing, air atomizing, blue flame, internal recirculating,
or external recirculating. Those which utilized a combustion improving
or flame retention device are indicated.
10
^ 5
D)
-*
0>
(D 4
O
0 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
STOICHIOMETRIC RATIO
Figure A-3. Determination of carbon monoxide for normal
operating conditions.
98
-------
O)
2.00
1.75
1.50
1.25
1.00
0.75
0.50
0.25
1 T
1 I I T
0.00
0.0 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
STOICHIOMETRIC RATIO
Figure A-4. Determination of hydrocarbons for normal
operating conditions.
99
-------
2.0
1.6
1.2
-------
10
oo 6
13
2
5 si
00
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
STOICHIOMETRIC RATIO
2.6
Figure A-6. Determination of average smoke for normal
operating conditions.
101
-------
100
90
80
70
5?
O
LLJ
U
60
40
30
20
10
1.0 1.2 1.4 1.6 1.8 2.0
STOICHIOMETRIC RATIO
2.2
2.4 2.6
Figure A-7. Determination of efficiency for normal
operating conditions.
102
-------
FUEL ANALYSIS
Fuel: Gulf Coast oil, refined at the Toledo Refinery
Description: No. 2 distillate heating oil
Use: Burner, boiler, and furnace studies
Location: Combustion Research Section3
Fairfax Laboratories
3914 Virginia Avenue
Cincinnati, Ohio 45227
Viscosity, kinematic (D-445): 2.42 centistokes. 100°F
Gravity API:
Aniline point (D-611):
Total sulfur:
Ash % (D-482):
C/H ratio:
Heat of combustion (D-240):
Distillation:
Initial, °Fi
10%:
50%:
90%:
End point:
36.4
142.5°F
0.098 wt %
0.003 on 10% residuum
6.62
18,443 Btu/lb
360
424
465
519
591
Now: Combustion Research Section
Control Systems Laboratory
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
103
-------
APPENDIX B (CONTINUED)
Hydrocarbon analysis:
Paraffins, vol %: 38.4
Naphthenes, vol %: 35.3
Aromatics, vol %: 26.3
1-ring aromatics, vol %: 12.1
2-ring aromatics, vol %: 13.9
3-ring aromatics, vol %: 0.3
Olefins: vol % <1.0
104
-------
Appendix C
COMPARISON OF ALL OIL BURNERS TESTED
(These comparisons were made by
determining the stoichiometric
ratio for each burnerburning
No. 2 oilat a No. 1, 10th
nrinute smoke spot on the
Bacharach scale. See Appendix
A, Burner Adjustment and Comparison,
for more detail.)
The data included in this report
is only representative of the
burner models tested. Thus, the
data is not necessarily repre-
sentative of a manufacturer's
entire model production.
105
-------
Appendix C. COMPARISON OF ALL OIL BURNERS TESTED
Burner
1 . Standard burner
Will iamson
(conventional ABC
Model 45)
2. Combustion improving
devices
Delavan FlameCone3
- Gulf Econo-Jeta
0
O)
Monarch combustion head3
Shell combustion head3
Union (Pure) flame
control device3'"
3. Flams retention devices
ABC Mite (Model S)
Beckett Bantam (Model AF)
Date(s)
tested
5-3-68
6-24-68
6-25-68
12-16-68
12-17-68
1-16-69
1-24-69
1-29-69
11-26-68
12-2-68
2-13-69
2-14-69
2-17-69
7/19-22/68
3/10-11/69
4-28-70
4-29-70
5/1-4/70
Characteristics
Conventional ,
high-press. ,
atom. -gun
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
Results
Effic-
iency
%
76.6
70.5
75.0
71.5
76.0
83.0
79.5
81.1
10th
min
smoke
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Avg
smoke
2.9
1.3
3.0
2.0
2.0
1.2
2.0
2.5
I
NO,
g/kg
fuel
1.11
1.30
1.69
1.25
1.68
1.25
0.77
1.40
HC,
g/kg
fuel
0.06
0.03
0.06
0.06
0.06
0.06
0.06
0.06
CO,
g/kg
fuel
0.50
0.6
0.6
0.6
0.3
0.5
0.5
0.5
Stoich
ratio
1.53
1.80
1.40
1.66
1.60
1.20
1.38
1.31
Air
setting
% co2
9.9
8.2
10.8
9.1
9.4
12.6
10.9
11.6
aCombustion-improving device
Veil iz;.-d fl.-m.. rotanti.n
installed on Williamson standard burner.
-------
Appendix C- COMPARISON OF ALL OIL BURNERS TESTED (CONTINUED)
Burner
3. Continued
Flame retention devices
Esso (Model 40)
Sun-Ray (Model DC-1)
Union burner (Model AFC)
_,
o
^ U.S. Carlin (Model 150N-2R)
Wayne (Model ER)
Wayne (Model M-SR)
White-Rogers (Model FR-B)
4. Miscellaneous devices
Cyril Meenan
Combusto-Jetb
Stewart-Warner
burner/boiler
Date(s)
tested
11-16-70
12-3-70
3-5-69
2/17-19/71
2/23-24/71
3-9-71
3/18-19/71
3-18-69
3-19-69
6-9-69
6-10-69
10-18-73
5-26-69
5-27-69
4-13-70
4-14-70
1-29-70
Cliaracteri sties
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
High-press. ,
atom.
Results
Effic-
iency
%
80.2
73.0
80.0
70.3
75.0
82.5
78.0
- - -
No
data
10th
min
smoke
1.00
1.00
1.00
1.00C
1.00
1.00
1.00
. . _
1.00
Avg
smoke
2.6
1.1
3.5
2.2
1.7
1.4
2.1
~ *.
1.80
NO,
gAg
fuel
1.76
1.16
1.00
1.48
1.14
1.17
1.45
0.89
1.30
HC,
g/kg
fuel
0.02
0.06
0.08
0.08
0.04
0.01
0.02
0.17
0.13
CO,
g/kg
fuel
0.28
0.6
0.25
1.0
0.5
0.21
0.4
3.89
0.31
Stoich
ratio
1.44
1.63
1.16
1.86C
1.42
1.18
1.48
1.47
1.53
Air
setting
% co2
10.7
9.3
13.5
7.9
10.7
12.2
10.2
10.3
9.9
Data incomplete: not comparable to other burners since 10th minute smoke is unknown.
cExtrapolated data.
-------
Appendix C. COMPARISON OF ALL OIL BURNERS TESTED (CONTINUED)
Burner
4. Continued
Miscellaneous devices
Stewart-Warner burner
Master air-atomizing
space heater"
Rockeydyne Una Spray6
o Torrid Heat Wall
00 Flame burner/furnace
Auburn blue flame
Bailey-OOHA blue flame5
3ailey-OOHA blue flame0
Blue-Jet blue flame
Vapo-Product blue flame
Date(s)
tested
6-18-70
7-11-70
10-8-69
6-9-70
8-28-69
10-6-70
10-7-70
9-2-70
2-6-73
7-20-73
3-10-70
Characteristics
Low-press. ,
air-atom.
Low-press. ,
air-atom.
Low-press. ,
liquid-film,
air-atom.
Rotary, verti-
cle wall , flame
Blue flame,
internal
recirc
Blue flame,
external recirc
Blue flame,
external recirc
Blue flame,
combination low-r
press, air atomi-
zing and vaporiz-
ation
Blue flame,
internal recirc
Results
Effic-
iency
%
73.5
No
data
91.0
90.5
81.8
83.8
- - _
10th
min
smoke
1.00
<0.05
1.00
1.00
1.00
0.0
Avg
smoke
1.45
<0.10
2.50
out of
range
of data
1.85
0.2
W,
j/kg
fuel
1.24
1.3+
1.1
1.74
f
3.15C
0.46
0.39
0.41
1.20
0.49
HO,
g/kg
fuel
0.11
0.1
0.20
p
0.0
0.16
0.22
0.04
0.25
CO,
g/kg
fuel
2.80
0.6-
0.5
0.50
r+
7.50C
2.5
0.56
0.57
2.01
0.45
Stoich
ratio
1.86
1.45-
1.8
1.20
1.11
1.20
1.06
1.06
1.25
1.68
Air
setting
% co2
8.0
10.4-
8.2
12.6
13.7
12.6
14.4
14.4
11.8
8.9
Uata incomplete; not comparable to other burners since 10th minute smoke is unknown.
°Extrapolated data.
Space heater; not comparable to other burners.
Efficiency not accurate; data from latest test 6/9/70.
-------
APPENDIX D
OTHER DISTILLATE OIL BURNERS
Most of the burners discussed in
this appendix were experimental or
prototype designs. Therefore, future
designs may have different performance
characteristics and/or different
emission levels.
109
-------
Appendix D
OTHER DISTILLATE OIL BURNERS
1. Stewart-Warner Low-Pressure Burner
Tests showed that performance of this burner is inferior to that
of high-pressure burners. The one advantage of the burner is that it
does not require as much maintenance as a high-pressure unit.
2. Torrid Heat Wall Flame Burner and Furnace
Emission characteristics from this furnace were quite different
from those of the high-pressure units. It had a very narrow range
of excess air settings for good combustion and pollutant emissions
were excessive at higher or lower settings.
3. Rocketdyne Una Spray Burner
This prototype burner has a unique air-atomizing system. The
tests showed excellent combustion with low levels of emissions except
for NO . Tests of a later model verified these results.
A
4. Blue Flame Burners
Of the blue flame burners tested only the Bailey-OOHA and the Blue-
Jet performed satisfactorily. The others performed poorly during
ignition, i.e. they had high smoke, CO, and HC emissions. The OOHA and
Blue-Jet burners had higher CO and HC emissions during ignition than
those of a conventional burner but they were much lower than those of
the other blue flame burners tested. The main advantage of the blue
flame burner is its low level of NO emissions. However, the NO emis-
X A
sions of the Blue-Jet burner were as high as those of a conventional
burner, probably because the others utilized some form of recirculation
of the combustion products and the Blue-Jet burner did not.
a. Auburn Blue Flame Burner
This burner is the Shell Ventres design which recirculates the
flue gas through the flame. The burner performed erratically and
110
-------
had a very narrow operating range for good combustion. The one
good feature of this burner was its low NO level of 25 ppm (about
/\
half the emission level of a high-pressure burner).
b. Vapo-Products Blue Flame Burner
This rather crude prototype burner, after a very noisy and smoky
startup, burned quite well with an almost perfect blue flame. After
ignition it produced zero smoke and only 30 ppm NO (at 3 percent 0?, dry
A £
basis). If the ignition problems can be corrected, the unit may be
marketable.
c. Bailey-OOHA Blue Flame Burner
This blue flame burner was built to operate with a hot water
generator. Initial tests were made in September 1970 at five excess
air settings: three of the ignitions were very smoky. A complete
set of data was not obtained because of some difficulties with the
air pump. The low NO emissions of about 20 ppm were impressive.
y\
In February 1973, tests were made on a new blue flame burner
developed by Mr. Frank Bailey, research consultant of Operation Oil
Heat Associates (OOHA). This burner design had corrected the problems
of the earlier burner: NO levels were about 0.4 g/kg with accompanying
/\
low levels of CO and HC even at excess air levels as low as 5 percent.
Since the new design is relatively simple, retrofit to many existing
furnaces is possible and new units could easily be designed without
adding significantly to costs. This burner has excellent potential
for practical application.
d. Blue-Jet Blue Flame Burner
The Blue-Jet burner utilizes a combination of low pressure atomization
and vaporization to prepare the fuel oil for combustion. During the
ignition period (about 1 minute) the oil is atomized and ignites at an
electrode located about midway from the nozzle and the burner grid. After
the burner grid is sufficiently hot the ignition is switched off, and
the flame jumps to the burner grid. Its appearance is then very similar
111
-------
to that of a natural gas burner. There is also a small flame inside
the burner ahead of the grid which aids in vaporizing the fuel.
The performance of this burner was similar to that of a conventional
burner, with one exception. The CO and gaseous hydrocarbon emissions
had high peaks during the ignition period (2.16 g HC/kg and greater than
32 g CO/kg). The efficiency was about 84 percent, which is better
than that of a conventional burner. NO emissions were about 1.2 g/kg which
/\
are similar to those of a conventional burner. Most blue flame burners
have much lower NO levels but they usually utilize some form of
/\
combustion product recirculation and the Blue-Jet does not. This burner
has excellent potential for practical application.
5. Master Air-Atomizing Space Heater
The burners from these units were tested: all produced relatively
low emissions. However, they cannot be validly compared to other
burners since the Master units can only be operated at one excess air
setting.
6. Stewart-Warner Oil-Fired Boiler
A boiler was tested to determine if emission levels from boilers
were significantly different from those of warm air furnaces. The cests
showed no significant differences. Test results were confirmed by nearly
identical data from the Stewart-Warner Laboratory in Lebanon, Indiana.
7. Cyril Meenan Combusto-Jet Burner
This prototype burner was a complete failure when tested the first
time. It produced excessive amounts of NO (>150 ppm) and caught fire
A
after 30 minutes of operation. The manufacturer returned with an
improved design which operated satisfactorily but had no significant
effect on emissions. This device may perform better in the larger
range for which it was initially designed.
112
-------
Appendix E. CONVERSION FACTORS
MULTIPLIERS TO CONVERT EMISSION FACTORS FROM
g/kg TO OTHER UNITS FOR NO. 2 OILa
To obtain emission factor
in these units
Gaseous pollutants and parti cul ate:
kg/ 1000 liter fuel
g/10 calories input
lb/1000 Ib fuel
lb/1000 gal
lb/106 BTU input
Gaseous pollutants :
ppm at 3% Op, dry basis
ppm at 0% Op, dry basis
ppm at 12% C02
Parti cul ates:
lb/106 scf flue gas at 3% 02
lb/106 scf flue gas at 0% 02
lb/106 scf flue gas at 12% C02
Multiply emission factor in
g/kg fuel by
0.862
0.092
1.000
7.194
0.051
1770
~MW~
2065
MW
1597
MW
4.58
5.27
4.13
Typical No. 2 fuel oil having 33 API gravity
DMW = molecular weight of pollutant
113
-------
Appendix E.
CONVERSION FACTORS (Continued)
MULTIPLIER TO CONVERT EMISSION FACTORS
REPORTED AS NO TO EMISSION FACTORS REPORTED AS N02
Emission factors for NO are often reported as NOp because a major portion
of the nitrogen oxides is oxidized to N02 in the atmosphere. In this
report, however, emission factors for NO are reported as NO. To convert
emission factors reported as NO to N02 multiply by 1.53, which is the ratio
of the molecular weights of N0? and NO.
Example: 1.11 g NO/kg fuel x 1.53 = 1.70 g N02/kg fuel.
MULTIPLIERS TO CONVERT FROM THE
ENGLISH SYSTEM TO THE METRIC SYSTEM
To convert
from to
Btu/hr
Btu/lb
°F
ft/sec
gph
in.
lb/10 min
cal/hr
cal/g
°C
m/sec
liter/hr
cm
kg/10 min
Multiply English units by
251.98
0.56
5/9 (°F-32)
0.30
3.79
2.54
0.45
114
-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
EPA-650/2-74-003
3.S^ecipient's Accession No. i
--J
i
4. Tide and Subtitle
A Study of Air Pollutant Emissions from Residential
Heating Systems
Report Date
January 1974
6.
7. Author(s)
R. E.Hall, J.H.Wasser, and E. E. Berkau
8. Performing Organization Rept.
No.
9. Performing Organization Name and Address
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, NC 27711
10. Project/Task/Work Unii No.
ROAP 21ADG-AO
11. Contract/Grant No.
In-House
12. Sponsoring Organization Name and Address
NA
13. Type of Report & Period
Covered
Final
14.
15. Supplementary Notes
16. Abstracts ^he report presems3 results of recent EPA research into the problem of air
pollutant emissions from small-scale combustion systems. Major factors for contr-
olling emission levels were found to be: excess air, residence time at high tempera-
ture, combustion air handling components of burners , and burner maintenance.
Recommendations for minimizing emissions from new and existing equipment are
given, based on the research results obtained. Data illustrating the effects of comb-
ustion parameter changes on emission levels are given both for experimental comb-
ustors and for residential heating equipment currently in use in the U.S. Future
work, directed toward reduction of emissions, is also outlined.
17. Key Words and Document Analysis. 17o. Descriptors
Air Pollution
Combustion Control
Gas Burners
Oil Burners
Combustion Products
Nitrogen Oxides
Sulfur Oxides
Carbon Monoxide
Hydrocarbons
17b. Identifiers/Open-Ended Terms
Air Pollution Control
Stationary Sources
Residential Heating Equipment
Particulates
Residence Time
Smoke
Maintenance
Stoichiometry
17c. COSATI Field/Group
13B, 21B
18. Availability Statement
Unlimited
19..Security Class (This
Report)
UNCLASSIFIED
20. Security Class (This
Page
UNCLASSIFIED
21. No. of Pages
123
22. Price
FORM NTIS-33 (REV. 3-72)
THIS FORM MAY BE REPRODUCED
115
USCOMM-OC 14952-P72
------- |