-------
The importance of particulate emissions from oil-fired sources, in general, has recently
been estimated for several northeast metropolitan areas (Reference 1-7). High volume sampler filters
(used to measure ambient concentrations) collected during 1974 were examined by optical microscope
to provide a semi quantitative estimate of the oil soot present. Filters in these areas (Washington,
D.C., Providence, Philadelphia, Baltimore, Chattanooga) showed that oil soot particles were esti-
mated to comprise from 8-13 percent by weight of the total mass visible under the optical micro-
scope. This represented the average estimated percent visible mass of from 20 - 30 filters per
area. They were selected to represent several seasons of the year.
While this is not a quantitative technique, it does provide some perspective on the extent
of the problem - a significant though not overwhelming fraction of the mass was believed to be oil
soot. One should realize that over half of the typical ambient concentration in these areas is not
amenable to traditional control strategies (i.e., resuspended street dust, natural background, auto-
mobile exhaust particles, and particles formed by transformation of gases in the atmosphere). Thus,
the importance of identifying the impact of controllable emission categories is most important.
It is important to note that this same study found that while all of these areas are violating
the primary standard at one or more sites, the reasons for these violations are varied and in some
cases identifiable as localized problems such as resuspended dust or construction activity. The
control agency has the responsibility of weighing these factors in the development of a new control
strategy.
The control techniques and regulatory programs discussed here are directed towards the main-
tenance and attainment of current NAAQS for particulate. Compliance with these ambient air stan-
dards is based on measurements by High Volume Samplers (Hi Vols) that are placed at strategic loca-
tions throughout the region. Therefore, this report deals mainly with efforts to reduce mass emis-
sions as measured in the conventional manner - i.e., either by opacity, Smoke Spot Number according
to ASTM Procedure D2156-65, or particulate loading according to EPA Method 5.* The EPA recognizes
that these standards may not be sufficient to protect health and welfare from particulate pollution
and is currently evaluating the need for an additional air quality standard for fine particulates.
Some of the potential shortcomings associated with the current standard are attributed to the fol-
lowing concerns:
*The Smoke Spot Number is frequently reported as the Bacharach-Shell Smoke Spot Number, the
Bacharach Smoke Spot Number, or simply the Bacharach Number. All four descriptions refer to
the same measurement technique.
1-5
-------
Heavy particles contribute most of the mass measured by the Hi Vols although they
account for only a few percent of the material collected on a number basis. According
to the results of one analysis of an ambient sample by microscopic count, 60 percent of
the particles by number were under 3 microns in diameter (i.e., in the respirable range)
and accounted for about 4 percent of the total mass; in contrast, 4 percent by number
were over 11 microns in diameter and contributed nearly 80 percent of the total weight
(Reference 1-4).
There is some evidence to indicate that toxic materials tend to concentrate on the small
particles (Reference 1-5). Hence, reductions in mass emissions of particulates to the
point where the NAAQS are not exceeded may not be sufficient.
Secondary particulates (those generated in the atmosphere) may be more hazardous than
primary ones (those emitted into the atmosphere from combustion, industrial, transpor-
tation, and natural sources).
Although a standard for fine particulates would probably require the implementation of additional
controls beyond those required to meet the primary standards, many of the controls that can be
used to meet the primary standard also reduce the emissions of submicron particles.
The body of this report lays the groundwork for the final chapter, which presents control
strategies that could be used by air pollution control authorities in the metropolitan areas
and the States to reduce particulate emissions from oil-fired boilers and furnaces. Thus, in Sec-
tion 2, we present descriptions of the equipment under consideration and inventories of their in-
stalled capacity, by region within the United States. In that discussion, as in the remainder of
the document, boilers and furnaces are subdivided further into four user categories based on size:
residential, commercial (e.g., apartment buildings, institutions, and public buildings), industrial
and utility.* The equipment descriptions will include a discussion of the particulate emissions from
these units and the factors which affect those emissions. Next, in Section 3, we consider the ques-
tion of fuel, such as the effects of fuel properties on emissions, the costs of various fuels, and
the effect of fuel additives or water/fuel emulsions on particulate emissions. All techniques other
than fuel switching and fuel additives which can reduce particulate emissions from oil-fired burners
are discussed in Section 4. These techniques tend to fall into one of the following categories.
*These categories are distinguished by size because it is easier for enforcement purposes to sub-
divide sources on that basis. The NEDS system, for which Table 1-1 was developed, theoretically
identifies sources on the basis of their application; thus the "Electric Generation" category may
contain steam power plants that are located in an industrial facility and supply electricity only
to that facility.
1-6
-------
Operation and maintenance procedures
Design changes to the burner and/or combustion chamber
Flue gas treatment
A brief review of industry standards for the construction and performance of boilers and
furnaces is then presented in Section 5 to show the relationships, albeit limited, between these
standards and particulate control strategies. Since any control program that is recommended should
take past experience into account, Section 6 is devoted to a summary and review of existing regula-
tory programs for the control of particulates from oil-fired units. This summary includes a compila-
tion of pertinent regulations throughout the United States and a more detailed review of the regula-
tions and enforcement procedures used by nine metropolitan or State control authorities to reduce
particulate emissions from indirect heat fuel burning combustion sources. Based on this review of
the characteristics of the sources, their utilization patterns, existing particulate control tech-
nologies for these sources, and current regulatory practices, we then list and rank control measures
(by user group) in Section 7 that local authorities could adopt to reduce particulate emissions with-
in their areas of jurisdiction.
Thus, this material is discussed in detail in Sections 2 through 7; it is summarized in the
following three subsections of this Introduction and Summary.* Appendix A contains recommendations
for research and development programs aimed at reducing particulate emissions from oil-fired boilers
and space heating furnaces. Suggested topics include equipment redesign, demonstration tests on
suggested improvements, development of new monitoring systems, etc. Appendix B presents a brief
comparison of the various methods used to measure smoke and particulate emissions, and Appendix C
lists 'the conversion factors between engineering and SI (System International) units. Appendix D
presents sample calculations analyzing costs. Appendix E lists the members of the Ad Hoc Advisory
Committee.
*Since the following subsections merely summarize the remainder of this document, detailed refer-
ences are given only in the main body of text.
1-7
-------
1.2 SUMMARY Of EXISTING CONTROL PROGRAMS
This subsection presents a summary of existing control programs for particulate emissions
from oil-fired boilers and furnaces. The summary begins with tabulation of the emission limits
by state and then continues with a discussion of the regulations and enforcement procedures used
by nine control authorities which are believed to have active control programs for oil-fired sour-
ces. A summary of these regulations as they specifically relate to each user category is pre-
sented in the next subsection at the end of the summary of technical background information for
each category.
Before preceding to the discussions of regulatory programs, we note that certain nonregulatory
programs exist which aim to reduce particulate emissions. Chief among these are the various private,
industry, and trade association sponsored training programs for burner servicemen and boiler operators/
maintenance staff. Voluntary operational changes that result in energy conservation are indirect
particulate control programs since they cause a reduction in fuel consumption. This category of
activities includes the'use of more building insulation, more attention to energy efficient opera-
tion, the installation of energy recovery devices such as waste heat recovery units, and a shift /to
greater recycling of metals, glass, and paper products. And, finally, information campaigns ("PR")
can educate both the public and the owners of commercial or small industrial boilers about the value
of proper, periodic maintenance programs. Such campaigns seem to be most effective if they stress
the fuel savings that come from correct maintenance practices in addition to describing environmental
benefits.
Table 1-2 presents the particulate limits by each State for new oil-fired indirect heating
fuel burning sources.* This table shows that there is considerable variation among the regulations
from State to State. Some States do not place emission limitations on smaller units, and where all
the States restrict emissions from a given size source, these restrictions can sometimes differ by
an order of magnitude. Typically regulations range from about 0.6 Ib/MBtu (258. ng/J) for units
with a heat input of 10 MBtu/hr (2.93 MW) (in those States which regulate units of that size) to values
of 0.1 to 0.2 Ib /MBtu (43 to 86. ng/J) for the very large utility boilers."^ Some States are more re-
stricting, with limits as low as 0.06 Ib/MBtu (25.8 mg/J) for 10 MBtu/hr (2.93 MW) boilers and
Tables 1-2 through 1-4 are identical to Tables 6-2 through 6-4, respectively, except that the foot-
notes for Tables 6-2 and 6-4 are not repeated here.
^Throughout this report the letter M represents 106 in accordance with System International practices.
Thus MBtu = 106 Btu.
1-8
-------
TABLE 1-2. STATE REGULATIONS FOR PARTICULATE EMISSIONS FROM
OIL-FIRED INDIRECT HEATING SOURCES3, Ib/MBtu
" 1
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
D.C.
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa " .
Kansas
Kentucky
Louisiana
Maine .
Maryland'
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey .
New Hex 1 co
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia .
Washington
West Virginia
Wisconsin
Wyoming
A. Samoa
Guam
Puerto Rico
Virgin Islands
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
>
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Source Heat Input Rate, MBtu/hr
<1
0.083
0.167
0.5
0.1
0.3
0.1
0.10
0.6
0.6
0.4
0.6
0.6
0.6
0.6
0.33
0.60
0.3
0.4
0.33
0.15
0.1
0.1
0.3
<10
0.6
,0.25
0.21
0.13
0.50
0.6
0.6
0.6
0.56
0.60,
0.06
0.1
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.005
0.6
0.6
0.4
0.6
0.4
0.20
0.60
0.6
0.5
0.34
0.6
>10
»
---»
0.05
»
»
0.1
*
20
0.51
Q.23
0.51
0.56
--»
0.4
tf
0.41
50
0.4
0.18
0.41
0.38
0.46
0.41
0.40
0.22
0.486
0.24
100
0.35
0.15
0.35
0.33
0.4
0.35
0.35
0.15
0.443
0.17
<200
B
0.3
(
>200
0.30
0.13
0.30
0.375
0.1
<250
0.12
»
-
»
>250
0.28
0.12
0.10
0.28
0.10
,
0.36
--
0.10
0.1
-.
0.10
0.10
<500
*
0.24
0.30
0.24
0.359
>500
0.243
0.1'
»
»
»
0.1
1000
0.21
0.21
0.26
0.21
0.328
'0.10
0.20
2500
0.17
0.17
0.23
0.17
0.291
~*
*
5000
0.13
0.14
0.2
0.14
0.266
7500
0.11
0.13
0.19
0.13
*
0.252
~*
10000
0.09
0.02
0.12
0.12
0.19
0.18
0.12
0.12
0.10
0.242
0.12
0.12
> 10000
»
*
0.1
0.09
50000
'0.036
0.197
>10!
^
0.025
^
'
0.02
0.180
_^
__
*
*
-0.10
__
a,b,*
Footnotes See Section 6. Limits are for new units 1f state has separate limits for 1
new and existing sources. See Appendix C for conversions to SI units. 1
1-9
-------
0.02 Ib/MBtu (8.6 ng/J) for utility boilers in Maryland. New Mexico currently limits oil-fired
boilers even further to 0.005 Ib/MBtu (2.15 ng/J) for all units greater than 10 MBtu/hr (2.93 MW)
heat input, but they are considering revising these limits upwards. Conversion factors between
SI and the more familiar engineering units are presented in Appendix C. New boilers greater than
250 MBtu/hr (73.25 MW) must also comply with an NSPS of 0.1 Ib/MBtu (43. ng/J). Since the regula-
tions differ so much from State to State, a condensed version of this large table has been prepared
to facilitiate comparisons among the states (Table 1-3). In this abbreviated version of the State
regulations, each column approximately represents one of the four user categories discussed in this
report. A quick review of this table shows, for example, that 29 states and territories (i.e.,
slightly more than half of the 55 states and territories included on the table) do not restrict par-
ticulate emissions from residential and small commercial units (<1 MBtu/hr or 0.293 MW). Limitations
for large commercial and industrial units (nominal 10 and 100 MBtu/hr or 2.33 and 29.3 MW, respec-
tively) range from 0.06 to 0.6 Ib/MBtu (25.8 to 258. ng/J) and those for utility boilers (nominal
1,000 MBtu/hr or 293 MW) range from 0.02 to 0.6 Ib/MBtu (8.6 to 258 ng/J) (excluding New Mexico,
which may revise its limits soon).
Opacity limits are presented in Table 1-4 for fuel burning sources. Many states now limit
opacity to 20 percent (Ringelman #1), although some restrict visible emissions to 10 percent and
three even permit no visible emissions. Frequently a more lenient limit is placed on existing
sources than on new sources. In addition, sources are allowed to emit more than the limits shown
on this table for short periods of time, e.g., 3 to 6 minutes during each 1-hour period. In some
states there is no restriction on the opacity of the plume during these exception periods, whereas
in other states the plume can obscure no more than 40 or 60 percent of the transmitted light. Some
states apply these opacity limitations to all sources, irrespective of their size.
More detailed information on the regulations and enforcement procedures used by local author-
ities was obtained by direct contact with personnel in nine agencies which were believed to have
active control programs. The enforcement procedures in eight of the nine districts visited ap-
peared to depend heavily on the use of construction and operation permits. These permits are used
to identify sources of emissions. The applications for construction permits are generally reviewed
by the engineering and/or enforcement staff in the control agency to determine whether the equipment
and the proposed operating procedures will enable the source to comply with local regulations. After
the new source has been installed, the control district inspects the facility to check its perfor-
mance with the application and to observe the procedures used by the operating personnel. If the
inspector's report is favorable, an operating permit is then issued to the installation. Enforce-
ment through the use of permits is generally supplemented by roaming inspectors who look for viola-
tions of the visible emission limits, by stack testing upon the request of the control district if
1-10
-------
TABLE 1-3. PARTICULATE LIMITS AT SELECTED HEAT RATES*, LB/MBiu
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
D.C.
Florida
Georgia
Hawai i
Idaho
Illinois
Indiana
Iowa . ' ' .
Kansas
Kentucky
Louisiana .
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island .
South Carolina
South Dakota
Tennessee
Texas
Utahb
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Samoa
Guam
Puerto Rico
Virgin Islands
Source Input Heat Rate, MBtu/hr
<1
0.5
0.083
0.167
' 0.5
0.1
0.3
0.1
0.1
0.6
0.6
0.4
0.6' '
,0,6
0.6
0-6
0.33
0.6
0.3
0.4 "
0.33
0.15
0.10
0.10
0.3
-------
TABLE 1-4. VISIBLE EMISSION LIMITATIONS FOR OIL-FIRED INDIRECT HEATING SOURCESa
State
Alabama
Alaska
Arizona
Arkansas
California
Col orado
Connecticut
Del aware
D.C.
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Opacity
20
20
40
20b
20C
20
20
20
0
20d
20e
20f
20g
30h
40
40
201
20
20
40
0
20
20
20
40
20
20
20
State
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyomi ng
A. Samoa
Guam
Puerto. Rico
Virgin Islands
Opacity
20
20J
20
20
20k
20
201
20
20
20m
20
0
20"
20
20
20
20°
20P
20
20q
10
10r
20s
20
20
20
20*
Footnotes: See page 6-10.
1-12
-------
a problem is suspected, by permanent installation of continuous opacity monitors in units which
exceed a certain size or which use residual fuel, and, in the case of Connecticut, by annual stack
tests for all units which emit more than 100 tons-per year (TPY) total pollutants.
The regulations were generally stringent in these nine regions by comparison with many of
the states. For example, two of the regions prohibited all visible emissions, and the other seven
only allowed up to 20 percent obscuration, whereas several of the other states permitted plumes to
have opacities as great as 40 percent (Ringelman #2). With respect to particulate emissions the
regulations were strict in that they imposed relatively low limits on all sources, and that limits
were applied even to the smaller units in the district. Thus, in some regions all units had to con-
form with the particulate loading limits, irrespective of size. In all regions which were studied
in detail, burners that were larger than 5 MBtu/hr (1.5 MW) had to comply.
Most of the regions contacted also had stringent restrictions on the allowable sulfur content
of the fuel. Since desulfurization also results in the removal of some of the solid matter in the
fuel and in a decrease .in weight (increase in API gravity), the sulfur restriction has a beneficial
side effect by reducing sources of particulate emissions. Three of the nine control districts also
prohibited the use of residual oil in small units.
Different regions need different control strategies because of differences in the distribu-
tion of sources. For example, New York City needs to stress emission reductions in commercial sized
units as a result of the larger number of apartment buildings in that area, whereas a less densely
populated, more industrialized area would emphasize emissions from industrial and utility boilers.
The major differences between the control programs in the metropolitan and in the state agen-
cies visited during the course of this study appears to be in the stringency of the limitations,
especially those for sulfur in the fuel. As one would expect, the metropolitan regions, with their
high concentration of sources and human receptors, generally place tighter restrictions on the sul-
fur content of the fuel and lower emission limits on the sources than do the states. There is also
a tendency within metropolitan control agencies to place greater emphasis on the control of emissions
from commercial size units than is the case in the states. And finally, the size of the enforcement
staff in the metropolitan control districts is generally larger than it is in the state regions.
For example, Philadelphia had approximately the same number of inspectors as did the State of Con-
necticut. The need for a large staff is a consequence of the need to control the many smaller units,
along with large units, which exist in the metropolitan areas. Beyond these three areas, the major
1-13
-------
differences are Individual and probably due to the approaches preferred by the people who establish
the programs. That is the use of an equipment standard in Maryland and a performance standard in
New York City is probably not at all related to the fact that one is a state and the other a metro-
politan area.
1.3 SUMMARY OF BACKGROUND INFORMATION
1.3.1 Residential Units - 0 ^ 0.4 MBtu/hr (0 - 0.12 MM)
1.3.1.1 Usage and Emissions
Oil-fired units in residences tend to be nearly equally divided between warm air and hot
water or steam systems. Together they account for slightly over one-quarter of all residential
heating units in the United States, and their market share may increase in the future as oil be-
comes less expensive than electricity (total life cycle costs) and more available than gas. Four
times as much energy (all fuels, including electricity) is used for space heating as for the gen-
eration of hot water for washing. The most common units are lined with refractory brick and use
conventional high pressure burners. The average age of residential oil-fired furnaces and boilers
is about 15 years, and about one-quarter of these have had the original burner replaced.
Mechanical atomization is the primary atomization method used in domestic sized furnaces.
Other methods u:sed are wall flame rotary, vaporizing and low pressure burners, but the combined
usage of these types account for only about 10 percent of the total.
Filterable particulate emissions from currently installed, well maintained residential
furnaces average around 0.018 Ib/MBtu (7.74 ng/J); and those for boilers 0.013 Ib/MBtu (5.56
ng/J). The current recommended emission factor for all residential units, as a group, is 0.017
Ib/MBtu (7.31 ng/J). This factor applies to areas where the units receive periodic inspection and
maintenance service by qualified personnel.
Particulate emissions from newer burners are lower than those from older ones, primarily
due to improved design practices in recent years. However, particulate emissions from properly
maintained oil burners should not increase as the burner ages.
1.3.1.2 Fuels and Fuel Additives
Virtually all oil-fired residential units burn distillate oil. No. 1 distillate is used
in burners which prepare the fuel for burning solely by vaporization, while No. 2 oil is used in
burners which prepare the fuel for burning by a combination of vaporization and atomization. The
majority of residential units are fired with No. 2 oil.
1-14
-------
According to one study no significant difference in participate measurements was found by
firing a wide range of No. 2 oils. The API gravity of the oils ranged from 30 to 37, sulfur con-
tent from 0.05 to 0.3 percent, and carbon residue from 0.1 to 0.3 percent. The influence of fuel
gravity on emissions varied, depending on whether the unit was tested in an as-found or tuned con-
dition. In well-tuned units, a change to a heavier fuel of the same grade (i.e., increasing
specific weight within the allowable limits for No. 2 oil) did not automatically cause an increase
in smoke or solid particulate emissions.
Based on one study only metallic additives containing cobalt, iron or manganese appreciably
decrease particulate emissions from distillate oil-burning units. Another study found that certain
proprietary organic additivies reduced particulates but only when the weight added, approached one
percent of the weight of the fuel. The known toxicity of the metallic emissions produced from all
but the iron-based additives makes their use questionable. Moreover, there is some concern that
the iron oxides which are discharged when iron-based additives are used may interact with hydro-
carbons in the atmosphere to become potentially carcinogenic.
1.3.1.3 Particulate Control by Improved Burner Servicing
Since the main fuel used in oil-fired residential heating units is No. 2 distillate, and
not residual oil, high smoke readings in the exhaust are due mostly to improper air-to-fuel ratios,
inadequate draft, or worn-out burner components. Therefore, an effective and practical technique
to reduce particulate emissions from residential oil burners is to keep serviceable units well
maintained and to replace worn-out components or the entire burner. Significant improvements can
also be obtained through the use of new, well designed burners and combustion chambers.
Average smoke emissions from a number of properly adjusted burners can be at mean Smoke
Spot Numbers as low as 1.1 to 1.3. These levels were reported by two separate teams who inves-
tigated representative samples of currently installed units. Individual systems had Smoke Spot
Numbers as low as 0. They were able to obtain smoke reductions of 40 to 60 percent through
proper servicing. Moreover, all units in one of these samples could have complied with a
standard of Smoke Spot No. 3 when tuned according to commercial practices, and nearly 90 per-
cent could have complied with a level of No. 2. Alternatively, over 90 percent of the units
could achieve a level of 3 at exhaust gas C02 concentrations of at least 8 percent (correspond-
ing approximately to a thermal efficiency of 77.5 percent) and three-quarters even could achieve
Smoke Spot No. 1 at 8 percent C02 or No. 2 at 9 percent C02 (about 79 percent efficiency).
1-15
-------
Tuning produced little change in NOX, while filterable particulates were reduced 24 percent, from
0.021 to 0.016 Ib/HBtu (9.03 to 6.88 ng/J). The results also indicated that some 71 percent of
the reduction in particulate emissions came from the replacement of malfunctioning units. These
burners accounted for 10 percent of the units tested, a condition indicating the desirability of
a periodic (e.g., yearly) inspection of all burners by a qualified serviceman. In both samples
the thermal efficiency increased an average of 1.7 percent, to about 79 percent, when the burner
was tuned for minimum smoke.
The value of a systematic, district-wide burner maintenance program for some areas is demon-
strated further by the results of a study in the Maryland and Washington, D.C., areas. Here inves-
tigators found that less than 42 percent of the units tested (over 400 total) complied with the local
Bacharach No. 2 standard. Moreover, thermal efficiency ranges from 54 to 80 percent. However, per-
sonnel who surveyed the particulate emissions problems in the Boston region concluded residential
and commercial burners were already well maintained in that area, and, therefore, no further improve-
ment could be obtained by a regulation that mandated periodic inspection. This conclusion was based
on discussions with oil suppliers and burner servicing companies, which reported that most domestic
and commercial units were serviced at least annually in an effort to conserve energy. No emission
tests were conducted to substantiate this belief.
The importance of annual servicing is not so evident if one just looks at burner performance
before and after a tune-up of one unit that customarily receives yearly maintenance. Burners which
do not receive yearly maintenance will begin to perform progressively worse during the second, third,
and succeeding years. Thus, periodic service is preventive maintenance; it generally does more to
keep burners from becoming high emitters than to actually reduce emissions from these burners.
Usually, burner maintenance service can be obtained from an oil supplier on a contractual basis.
The cost for such service is approximately $30 to $35 per year, excluding parts. Although peri-
odic replacement of the filter for the house air or cleaning of the boiler surfaces does not di-
rectly affect particulate emissions, it does so indirectly by maintaining the system's efficiency.
In light of the importance of proper burner maintenance in reducing air pollution, upgrading
the competence of servicemen and supplying them with good diagnostic instruments can be considered
a particulate control technique. Guidelines have been published recently by EPA to aid servicemen
in maintaining burners for high efficiency and low emissions. Training programs for maintenance
personnel are available in some localities where oil is in heavy use. These programs are usually
offered in community colleges or trade schools. A few private or trade organizations in the United
1-16
-------
States also^specialize in setting up training programs and preparing textbooks for oil burner ser-
vicemen. A typical course takes 3 days and-costs $45. Diagnostic instrumentation to measure CO-,
CO, Smoke Spot No., stack temperature, and draft are commercially available. Kits that include
units to measure,all five of these parameters cost about $300 to $500.
1-3.1.4 Particulate Control by Equipment Design Changes
Nearly all of the design variables for warm air furnaces or boilers affect particulate
emissions to some degree. The more significant effects are associated with the burner, specifical-
ly, its design and nozzle type. Additionally, tuning, usage patterns, and combustion chamber
materials can cause significant variations in particulate emissions. The burner design features
* .
which most strongly affect particulate emissions include:
a Fuel atomization techniques
Combustion aerodynamics (including the use of flame retention devices, swirl, and/or
recirculation)
Combustion chamber shape as it affects residence time
Combustion chamber materials (and hence wall temperatures)
New burner developments center mainly on the improvement of oil atomization and vaporiza-
tion schemes and on the optimization of the combustion aerodynamics, particularly by adaptation
of the internal or flue gas recirculation concepts. According to one source (Reference 1-6),
"Commercially used atomization nozzles which are available at low cost (less than about $1.00
per nozzle, nominally) do an excellent job of atomization". Therefore, most development work in
atomization is directed at improvements which reduce maintenance requirements for the nozzle or
allow optimization of the fuel spray pattern for given combustion aerodynamics.
Optimization of combustion aerodynamics is considered by many experts in the field to be
the best approach toward achieving the triple goals of low particulate, low NO , and high thermal
, , - ' . X
efficiency. Such an optimization attempts to control mixture (i.e., flow patterns and turbulence
levels) and residence time to achieve complete combustion at a temperature that, locally or through-
out the combustion chamber, never falls below the level required to sustain combustion nor rises
above the level at which NOX formation becomes significant. Although most of the research in this
area has been directed at NOX reductions, with smoke merely a limiting factor, several burners have
been developed which reduce smoke concurrently with lowered NO .
1-17
-------
One such unit is a "blue-flame" burner, which has just become available commercially for
V
single-family warm air furnaces. Because this burner has to be matched to the combustion volume
to obtain the desired (internal) recirculation pattern, it can be purchased only as a burner-
furnace package. In laboratory tests on this unit, a thermal efficiency of over 80 percent was
achieved simultaneously with a Bacharach Smoke Number close to zero and NOX emissions just over
20 ppm (typical distillate burners emit about 75 to 90 ppm). Its only drawbacks are high CO emis-
sions for 3 to 4 seconds during ignition and 10 percent higher initial costs than for a furnace
with a conventional burner. Because of its high efficiency, this first cost .difference would pay
back in a fairly short time.
Another new "optimized" burner is nearing commercialization. It produces negligible quanti-
ties of smoke (less than Bacharach No. 1) and nitric oxide emissions that are about one-half to
two-thirds the amount generated in typical, new commercially available burners. This is accom-
plished while operating at only 10 percent excess air (in laboratory tests) instead of typical
values of 25 percent for other new burners. This burner was designed as a retrofit (see next
paragraph) and is expected to be available from several manufacturers within 1 to 3 years. A
matched burner and combustion chamber that operates on the same principle but can achieve still
lower NO emissions and higher thermal efficiency with no increase in smoke is expected to be
X
developed within the next 3 to 6 years.
Retrofitting of existing units which are difficult to maintain or defective in design is
another method of reducing particulate emissions. These retrofits generally involve replacement
of malfunctioning burners at a cost of approximately $300. In the future they may also involve
the replacement of operational burners, with the "optimized" burners just mentioned above. An-
other presently used retrofit is to add one of the commercially available combustion-improving
devices to an existing burner for the purpose of improving the mixture of air and fuel. One such
add-on, a retention head device, was tested on a standard, high pressure atomizing burner and
found to reduce smoke emissions almost 60 percent (average Bacharach No. of 1.2 with the reten-
tion head versus 2.9 without it). In addition, thermal efficiency was nearly 10 percent higher
for the retention head unit (83 versus 76 percent). Data from other tests on retention heads
showed an average particulate emission rate of 0.0056 Ib/MBtu (2.39 ng/J) for a burner with a
flame retention device versus 0.012 Ib/MBtu (5.16 ng/J) for one with a conventional head
1-18
-------
(i.e., a 54 percent reduction). The flame retention burners also produced lower CO, HC, and
NOX emissions than did the conventional types. However, laboratory experiments have shown that
not all flame retention burners simultaneously reduce particulates and NO . These combustion
improving devices cost about $30 installed.
One of the major factors contributing to high particulate emissions from domestic burners
is on-off cycling. The high emissions during the transients are caused mainly by variations in
the combustion chamber temperature. Since a cold chamber wall will not assist the combustion,
peaks of CO, HC, smoke and particulate are produced -just after ignition. Moreover, the oil nozzle
may dribble during ignition and shutdown, and the large droplets which then fall out result in high
particulate emissions. Modulation and underfiring have been suggested as methods of reducing the
significance of the startup transient. When a system is controlled by modulation, the burners are
run continuously during peak heating periods at firing rates that vary between a low level and the
rated one, rather than alternating between operation at rated condition and shutdown. Underfiring
consists of using a smaller (undersized) unit which has to stay on longer and, therefore, cycles
less. However, no data are available on the effectiveness, cost, an.d public acceptability (more
complex system or inability to heat house as comfortably during very cold periods) of these pro-
posals.
To summarize, commercially available burners, add-on devices, and/or optimized combustion
chamber geometries can be used in distillate-fueled residential furnaces and boilers to achieve
low smoke emissions. New units can operate with Smoke Spot Numbers of less than 1 at steady-
state thermal efficiencies of more than 80 percent. Based on a sample of units presently in-
stalled in the field, most furnaces and boilers can ahieve a Smoke Spot Number of 2 at an ef-
ficiency of 79 percent, or better, after being tuned according to normal practices. Signif-
icant reductions in area-wide particulate emissions (about 20 percent of those due to residential
units) can be obtained by replacement of malfunctioning burners. Reductions of over 50 percent
in particulate mass emissions from residential units should be achievable by installing a pro-
perly selected flame retention head (one that is suited for the burner in.question and has been
shown to reduce particulates without increasing NO ). Such devices usually also decrease fuel
consumption.
1-19
-------
1.3.1.5 Particulate Control by Conservation
Conservation measures, and especially improved building insulation, reduce particulate emis-
sions indirectly by reducing fuel consumption. In the long run - i.e., after most of the homes in
an area have been properly insulated - particulate emission reductions of up to 1.4 percent of the
area total could be achieved in northeastern U.S. areas with cold climates and heavy reliance on
oil-fired home furnaces.
1.3.1.6 Industry Standards
Residential and small commercial boilers and furnaces are generally designed and fabricated
to meet Underwriters Lab (UL) and/or American National Standards Institute (ANSI) standards. Al-
though the primary aim of these standards is consumer protection, several features impact emis-
sions. Distillate fired units will soon be required to emit no more than a Smoke Spot No. 1,
concurrently with a C02 concentration of at least 10 percent to obtain UL and/or ANSI certifica-
tion, and the burner ignition and control system must satisfy certain requirements designed to
minimize "ignition puffs" and par'ticulate emissions caused by misaligned or malfunctioning units.
1.3.1.7 Existing Regulatory Programs
Host States do not limit mass emissions from boilers or furnaces that are classified resi-
sidential. Usually control is imposed on units which are in'l to 2 family residences.or rated at
less than 0.35 MBtu/hr (0.10 MW), but some states imposed no mass limits on systems which are
smaller than 1.0 MBtu/hr (0.29 MW). However, most areas require that all sources, including resi-
dential boilers and furnaces, obey the visible emission limits. Residential units should have
no trouble satisfying even a "zero visible emissions" limit, as in the Maryland, Washington, D.C.,
and Rhode Island areas, but units which are not well maintained may not meet Maryland's stringent
smoke spot limits (Smoke Spot No. 2). Several areas prohibit the use of residual oil in resi-
dential and commercial units smaller than about 3 to 5 MBtu/hr (0.88 to 1.5 MW) and place limits
on the sulfur content of the distillate oil (e.g., 0.2 to 0.3 percent by weight). None of the
air pollution control authorities contacted required mandatory service or maintenance procedures
for residential units. However, several of the staff members in the control districts believe
such a requirement would be a good one, especially if it were also combined with a mandatory
licensing and certification program for servicemen.
1-20
-------
1.3.2 Commercial Units 0.4 to 12.5 MBtu/hr (0.117 to 3.66 MW)
1.3.2.1 Usage, Fuels and Emissions
Commercial oil-fired furnaces and boilers are used for space heating and hot water in
apartment buildings, public offices, and commercial areas. Warm air furnaces are used mainly
in moderate climates if comfort heating is the only requirement. Elsewhere, boilers which can
supply hot water or steam for multiple purposes are used. Nationwide, in this size category about
one-sixth as much energy is consumed to generate hot water for washing as is used to provide space
heating.
Warm air furnaces are generally upgraded, roof mounted versions of residential ,units. Al-
through the burners tend to be more versatile than the ones used in residential units (i.e., can
operate properly over a much wider range of firing rates), burner and combustion chamber modifica-
tions generally have the same effect on emissions and efficiency in both size ranges.
Boilers can be cast iron or packaged firetube (watertube boilers comprise only 5 percent of
this category). Smaller units tend to be cast iron, whereas larger ones are most often firetube.
The most popular firetube boilers now are the compact, efficient Scotch and firebox designs. Com-
mercial boilers are approximately equally distributed between oil and gas firing, regardless of
size. The trend is strongly toward dual fuel (oil and gas) capabilities. Among the oil users,
approximately 95 percent of the smaller units burn distillate, whereas 50 percent of the larger
ones use preheated Nos. 5 and 6. Overall, fuel grades Nos. 2 through 6 are used with distillates
and residuals being almost equally in demand. No. 4 oil can be a distillate or a mixture of dis-
tillate and residual oils. It is used in some schools and apartment buildings and in situations
where the equipment cannot handle higher viscosity oils such as Nos. 5 or 6.
Filterable particulate emissions from commercial boilers in steady-state operation range
from an average of 0.0086 Ib/MBtu (3.70 ng/J) for those fired on No. 2 oil to an average of
0.24 Ib/MBtu (103 ng/J) for those on No. 6 oil. Discharge rates as low as 0.0036 Ib/MBtu
(1.54 ng/J), and as high as 0.38 Ib/MBtu (163 ng/J) were measured during the tests from which the
above averages were derived. Recently revised EPA emission factors for commercial units range
from 0.014 Ib/MBtu (6.02 ng/J) for distillate fired units to 0.053 Ib/MBtu (22.9 ng/J) for units
fired on No. 6 with 0.5 percent sulfur content and 0.22 Ib/MBtu (94.6 ng/J) for those using 3
percent sulfur resid. These factors assume the burners receive proper maintenance. Ash content
increases as the fuel becomes heavier (lower API gravity), but not enough to account for the
1-21
-------
higher participate levels when tbjase heavier fuels are used. A 1 percent sulfur residual oil (LSR)
was also investigated in the above-mentioned tests and found to yield filterable particulate levels
about equal to those from No. 4 oil and only one-third of those from No. 6 oil. In units which are
correctly designed for their intended fuel, the single most important fuel property influencing
filterable particulate is carbon residue. API gravity and sulfur content also have a significant
effect.
The sulfur content of domestic No. 6 oil ranges from 0.2 to 3.5 percent, depending on the
crude source and the method of processing it. With the advent of oil desulfurization processes
the r'efining industry can now make low sulfur residuals, but not yet in unlimited quantities.
Moreover, a 1 percent sulfur resid costs about 13 percent more than a high sulfur resid (i.e.,
unspecified sulfur content about 1 percent), and an 0.3 percent sulfur resid costs 16 to 24 per-
cent more than the high sulfur fuel. In addition, the energy requirements of the desulfurization
process are about 3 to 10 percent of the refined fuels heat content, depending upon the process,
the sulfur content of the crude oil, and the desired sulfur content of the refined products.
Particulate emissions from commercial units may depend upon atomizing methods (as in industrial
and small utility boilers). The trend in small commercial units is towards mechanical atomizing
(for simplicity) and in large units towards air atomization. Both types are gaining in popularity
as the more difficult to maintain rotary burners are phased out.
1.3.2.2. Particulate Control by Improved Burner Servicing
The most commonly proposed techniques for the control of particulates for commercial oil-
fired furnaces and boilers are the widespread implementation of, and adherence to, proper operating
and maintenance procedures for existing units, the use of new improved burners or combined burner/
combustion chamber designs, and the prohibition of residual oil as an allowable fuel. Although
particulate collectors can be used in the exhaust system ;of commercial units, they are generally
not used for this purpose because of their cost (abou't 12 percent of the initial cost of the boiler
or, when annualized, 11 percent of the fuel costs) and need for greater attention by operators than
is normally given to boilers in commercial applications.
Several studies have shown that smoke emissions can be reduced by as much as 40 percent in
existing units by proper burner tuning, cleansing of dirty burner cups or nozzles, and/or replace-
ment of work or damaged components. Thermal efficiency is generally improved by these adjustments,
1n some cases significantly, but on the average slightly less than 2 percent. After adjustment
1-22
-------
boilers fired on low sulfur residual (No. 6) can achieve Smoke Spot Number 3 at steady-state
thermal efficiencies of at least 80 percent whereas those burning fto. 6 oil of unknown sulfur
content (presumably greater than 1 percent) can achieve No.4 at the same efficiency (based on
data from a sample of furnaces and boilers that was selected to be representative of the in-
stalled population). Distillate units can be tuned to emit less than Bacharach No. 2 at the
same thermal efficiencies.
The cost for proper burner maintenance is slight compared with the cost of annual fuel
usage. Normally, this service requires 4 to 8 hours of labor at approximately $25 an hour.
At a median cost of $150, this would represent about 5 percent of che annual fuel bill for a
10 gph burner (assumed to use 10,000 gal/yr) neglecting any credit for improved fuel consump-
tion. The effective cost would be reduced by about 2 percent, on the average, due to fuel.
savings. ~
Proper training is essential for boiler operators and maintenance personnel in order to
realize the potential reductions cited earlier. Guidelines have been published recently by EPA
to aid servicemen in maintaining burners for high efficiency and low emissions. A few local
agencies have done the same and/or have instituted brief courses for boiler operators. Training
programs for maintenance and operating personnel are also offered by community colleges, trade
schools, or private trade organizations in some localities where oil is in heavy use. The same
diagnostic instrumentation can be used on commercial boilers as on residential units. Kits that
contain instruments to measure C02, CO, Smoke Spot Number, draft pressure, and stack temperature
are available commercially at a cost of $300 to $500.
1.3.2.3 Particulate Control by Equipment Design Changes
Proper oil atomization is essential for complete, smoke-free combustion. Conventional
atomizers can perform satisfactorily to give low smoke and high thermal efficiency. However,
one manufacturer claims that solid particulate emission reductions of 90 percent and combustion
efficiencies of over 83 percent can be realized with the use of an acoustic nozzle. Unfortun-
ately, no quantitative emission data are available to substantiate this claim.
Researchers have attempted to optimize the combustion aerodynamics in commercial burners as
well as in residential ones. These efforts have led one team of investigators to the development
of a 9-gph distillate oil burner that can operate "smoke-free" (i.e., Bacharach No. 1) with as
little as 2 percent excess air (typical new commercial burners require up to 25 percent excess
air). In addition, this burner produces about one-half as much NOX as do commercially available
1-23
-------
units. It should be on the market within three years. Another group has developed a new burner
for No. 6 oil which uses swirl, recirculation, and a multistage atomizer to operate over a wide
range of oil viscosity and at excess air levels as low as 2.4 percent without exceeding Bacharach
No. 4 or emissions of 0.2 Ib/MBtu (86 ng/J). Although this burner does not achieve lower smoke
levels than other commercially available units, it reduces fuel consumption (and, hence, par-
ticulates indirectly) and, therefore, could be used in areas that do not need to achieve significant
particulate reductions.
The retrofitting of existing units which are difficult to maintain or defective in design
is another way to reduce the particulate emissions. Such retrofits usually consist of burner con-
trol replacements, minor modifications to the combustion chamber, changes to the draft system, or
even burner replacement. Both new burners described above are designed to be used on existing
boilers.
At present one commonly practiced retrofit is replacement of rotary cup burners with pres-
sure or air atomizing units. Experience has shown that rotary burners usually emit more particulates
than other types, especially when burning residual oil. Therefore, at least one regulatory agency
has prohibited the use of these burners. The cost to replace a rotary cup burner with an air
atomizing type varies with burner size from about $1500 installed (part's and labor) for a 10 gph
(10.5 cmVsec) burner to about $5200 for a 50 gph (52.5 cm3/sec) unit. These costs are equal to
about 10 to 14 percent of 1 years total annualized costs (amortized initial cost, fuel, and main-
tenance) and would be repaid in 3 to 4 years if the new unit reduces fuel consumption by 10 per-
cent, as expected (through operation at lower excess air without violating smoke regulations).*
Another potential retrofit is the addition of a modulating unit to eliminate cyclic, or
on-and-off, operation in furnaces and boilers that provide mainly space heating. It has been well
documented that the transient nature of cyclic operation contributes significantly to total par-
ticulate emissions because the emission rates are very high during the starting transient. The
modulating operation should eliminate the peak particulate, smoke, HC, and CO emissions associated
with on-and-off or step turn down controls (the effect on NOX emissions is minimal). Unfortunately,
data to substantiate this claim are not available (i.e., comparison of total particulate emis-
sions from one days operation). Most new burners over 10 gph (10.5 cmVsec) can be ordered with
modulating control at an additional cost of about $500. Labor charges to retrofit a modulating
control to an existing burner add an additional $300 (for a total,of $800).
Based on fuel consumption of 10,000 gal/yr at a cost of 40£/ga1.
1-24
-------
Improperly designed combustion chambers can also contribute to particulate emissions, and
some of these problems can be corrected in the field (especially if they were caused by an ad hoc
field modification, as is often the case).
1-3.2.4 Particulate Control by Fuel Additives and Emulsions
Additives for distillate oil are covered under residential burners. Additives for residual fuels
have been studied by a number of investigators.* One study found that chelates of iron and cobalt,
and also hydrazine, reduced smoke in heavy fuel oils. The additives were used in rather high con-
centrations, ranging from 0.01 to 0.1 percent. Iron chelate was most effective.. At a concentration
of 0.01 percent it reduced Bacharach Smoke Number from 2.6 to 0.6. Other additives achieving varying
degrees of smoke reductions were manganese compounds and copper, iron, manganese inorganic salts and
some organic compounds (at concentrations approaching 1 percent). Even though additives may decrease
smoke emissions, some (especially the metallic-based ones) might create potentially harmful new emissions.
Moreover, additives which are compatible with one'fuel may be incompatible with fuel of the same grade
that comes from a different crude. Sludge which can form due to interactions between the additive and
the fuel can cause burner plugging.
Both water/distillate and water/residual oil emulsions were tested on a packaged commercial
boiler using (1) low pressure air atomization and an ultrasonic energy emulsifier and (2) high
pressure mechanical atomization and a high pressure emulsifier. In the distillate case, an emul-
sion with 25 percent water allowed the first unit to be run with approximately 4 percent less
excess air without increasing smoke emissions, but the resulting gain in thermal efficiency was
offset by heat lost to vaporization of the water, water supply problems, and emulsifier energy
requirements firing residual. The results depended significantly on the emulsifier used, and base-
line oil-only results varied significantly due to the difference in atomization technique used.
In the first of the two systems tested the smoke density of the boiler with oil only and 0.9
percent sulfur content in the fuel tended to stabilize at about Bacharach No. 6 beyond 40 per-
cent excess air. It could achieve the same smoke level with only 15 percent excess air or it
could achieve Bacharach No. 3 with 32 percent excess air if it burned a 25 percent water emulsi-
fication. Particulate emissions at a stoichiometric ratio of 1.47 were cut in half, from about
Additives for use with distillate fuel were discussed in Subsection 1.3.1,2, under residential
systems.
The iron oxides which are emitted when iron chelate is added to the fuel are not thought to be
harmful by themselves. However, recent studies suggest that they may combine with polycyclic
organic matter in the atmosphere to become potentially carcinogenic substances.
1-25
-------
0.064 to 0.032 Ib/MBtu (27.5 to 13.8 ng/J). With the other system, smoke with oil only was
No. 3.5 and could not be reduced much below No. 3 by the addition of water. Baseline particulate
emisssions of 0.12 Ib/MBtu (51.6 ng/J) (at a stoichiometric ratio of 1.35) were reduced to 0.042
lb/Hl!tu (18.1 ng/J) with 25 percent emu!sification. These baseline (oil only) variations were
attributed to the different atomizer used in each test; one resulted in unusually high smoke,
even with oil only and the other in unusually high particulate mass emissions - 0.127 Ib/MBtu
(54.5 ng/J) as compared with the EPA emission factor of 0.08 Ib/MBtu (34.4 ng/J) for this 0.9 per-
cent sulfur resid. In both cases, the thermal efficiency first-increased 1 to 2 percent when 15
percent water was added and then decreased as more water was added reaching a loss of 4 to 5 per-
cent with a 25 percent water emulsion. In conclusion, emulsions were optimized with approximately
15 percent water, had little effect on thermal efficiency, and reduced particulate mass emissions
significantly.
1.3.2.5 Particulate Control by Conservation
Just as in residential units, energy conservation practices can reduce particulate emissions
by reducing fuel consumption.
1.3.2.6 Industry Standards
Both the Hydronics Institute and the American Boiler Manufacturers Association (ABMA) rate
boilers in an attempt to provide a uniform method of determining boiler capacity. In order to be
listed with the ABMA, a boiler must be capable of continuous operation at its stated capacity with-
out violating smoke ordinances. The Hydronics Institute specifically requires compliance with
Bacharach No. 2 for distillated-fired boilers and No. 4 for residual-fired units. The American
Society of Mechanical Engineers (ASME) publishes detailed maintenance and testing procedures
which, if followed, lead to optimized combustion for that system and, hence, minimum particulate
production. These procedures are in addition to the design standards (structural, electrical,
etc.) which were developed for safety and do not impact particulate emissions.
1.3.2.7 Existing Regulatory Programs
All but three states limited mass emissions from units that were rated <.10 MBtu/hr
(2.9 MW). These limits ranged from 0.06 to 0.6 Ib/MBtu* (25.8 to 258 ng/J). All the control dis-
tricts contacted during this study had both opacity and particulate limits which applied to at
Excluding New Mexico which may soon revise its limits upwards.
1-26
-------
least some boilers and.furnaces 1n this user category. In a few regions units which were smaller
than 1 to 5 MBtu/hr (0.29 to 1.46 MW) were not subjected to the particulate loading regulations.
New York City requires commercial, boilers to pass a Bacharach smoke test with a reading of No. 3
or less and a stack thermal loss no greater than 20 percent of the heat input. This joint
stipulation of a smoke reading and a thermal efficiency is important because many people else-
where satisfy Bacharach limits by the use of excess air. Most of the regions contacted also
restricted the sulfur content in the fuel to 0.3 to 0.5 percent by weight.
Maryland is the only region of the nine studied in detail which attempts to limit emissions
by "an equipment standard. They recently banned the installation of rotary cup burners in new units
in the 1 to 13 MBtu/hr (0.29 to 3.8 MW) size category, and required that all units throughout the
state'phase out rotary cup burners by 1976. This equipment standard was based on their observa-
tions that rotary cup burners require more maintenance to insure low particulate emissions than
do other burners, and that the operators of commercial size boilers are not likely to give these
units the required maintenance. Therefore, they decided to solve the problem by simply prohibit-
ing the use of these burners. New York City also requires that new and upgraded boilers meet
performance specifications and installation criteria.
The enforcement procedures used in these nine control districts for commercial boilers and
furnaces varied from no activity after the granting of the permit to equipment restrictions to
periodic inspections. Generally, districts which use periodic inspections restrict their review
to operating procedures and equipment. They only resort to stack tests if they suspect a problem
with the installation. Only New York City requires a test for smoke, temperature, draft, and C02
for every permit renewal (3 years).
In order to supplement the enforcement program, which relies heavily on the skill of boiler
operators, New York City offers a basic boiler operator course to all boiler operators.- Erie
County requires the operators of the boilers in the public school system to attend a similar class.
1.3.3 Industrial Boilers - 12.5 MBtu/hr to 250 MBtu/hr (3.7 to 73 MW)
1.3.3.1 Usage and Emissions
Industrial boilers cover the size range from commercial units on the small side to utility
boilers on the large side. Therefore, the smaller industrial boilers are similar to large commer-
cial units, and the large industrial boilers have much in common with small utility boilers. Thus,
steam generators in this size category are either of firetube or watertube construction.
1-27
-------
The former design predominates the market among the smaller size units, whereas the latter prevails
among larger boilers (above 70 MBtu/hr or 20.5 MW). Units that produce steam above about 150 psig
(1.03 HPa) are generally also watertube boilers. Although most installed units are designed to
burn only one fuel, the trend is toward a dual fuel design (e.g., oil and gas). The smaller in-
dustrial boilers are now distributed about equally between gas and oil (90 percent residual),
whereas the larger units are distributed evenly among coal, gas, and residual oil. Overall, in this
size range, more than 60 percent of the fuel oil used is residual. Air and rotary cup atomizing
predominate among the lower capacity boilers in use today, although steam atomizing is also com-
mon. Rotary cup type burners are expected to be replaced by air or steam atomizing types in the
near future. Steam is, and apparently will remain, the major atomizing medium in the large units.
Air atomization produces about one order of magnitude less particulate than does mechanical atom-
ization. Moreover, particulates from an air atomized unit contain proportionally less combustible
matter than those from the other unit.
Although many industrial boilers were field erected in the past, projections suggest that
by 1990 over 90 percent of the industrial units sold will be packaged boilers. Boilers greater
than about 30 MBtu/hr (8.8 MW) are equipped with soot blowing equipment to clean the heat transfer
surfaces. Soot blowing procedures (e.g., frequency and duration of each blow) are known to affect
visible and particulate emissions, at least during the blow. However, data are not available to
show the effects of different soot blowing procedures on particulate emissions averaged over com-
plete operating cycles.
As with commercial boilers, particulate emissions from heavy oil-fired industrial boilers
are higher than those from units burning lighter oils. Thus, uncontrolled industrial boilers emit
0.0186 to 0.0371 Ib/MBtu (8.0 to 15.9 ng/J) when burning No. 2 oil, 0.037 to 0.112 Ib/MBtu (15.9
to 48.2 ng/0) when burning No. 5 oil, and 0.042 to 0.103 Ib/MBtu (18.1 to 44.3 ng/J) when con-
suming No. 6 oil (0.10 Ib/MBtu corresponds approximately to 43 ng/J which is the maximum allow-
able limit in many states for large boilers - see Tables 1-2 and 1-3). Emission rates decrease
with increasing size for steam atomized No. 6 fired boilers. However, for units that burn No. 2
or No. 5, the emission rates appear to be more a function of atomizing type than of size. Re-
'cently revised emission factors for industrial boilers depend upon the fuel used, ranging in
value from 0.013 Ib/MBtu (5.6 ng/J) for distillate fired boilers to 0.220 Ib/MBtu (94.6 ng/J)
for units which burn No. 6 oil containing 3 percent sulfur.
1-28
-------
Participate emissions increase with increasing carbon residue in the fuel oil. The limited
data available on particule size distribution suggest that over 90 percent of the number of particles
emitted from correctly adjusted No. 6 oil-fired industrial boilers are less than 6 microns in di-
ameter. The open literature does not contain any reports on tests with additives specifically for
particulate control from industrial boilers but an EPA sponsored study in this area will be com-
pleted soon by Battelle Columbus Labs.
1.3.3.2 Particulate Control
Large industrial boilers are generally well maintained to minimize fuel consumption. There-
fore, particulate collectors have been viewed as the only method of controlling particulate emis-
sions from these sources. However, a recent study has suggested that the maintenance procedures
designed to minimize costs (labor for maintenance versus fuel penalty) do not minimize emissions.
This study concluded that particulate emission reductions of up to 30 percent could be obtained by
reducing intervals between successive shutdowns for system inspection and cleaning (i.e., quarterly
instead of annually). Since the smaller industrial boilers are similar to the large commercial ones,
control strategies for the latter, such as improved maintenance procedures and operator/serviceman
training, could be applied beneficially to these industrial units, too.
When low sulfur oils are used, smoke is frequently eliminated by the use of high excess air
(e.g., 15 percent instead of the 3 to 5 percent desired for minimum fuel consumption). New, better
designed burners or closer combustion control should be used in these boilers to improve thermal
efficiency by operating with less air. Moreover, when high sulfur fuels are used, low excess air is
necessary to minimize the formation of SOg. This effluent not only leaves the stack as a visible
plume but also accumulates on the heat transfer surfaces, causing reduced heat transfer and tube wall
corrosion.
Electrostatic precipitators are the most common collectors. When properly designed and
maintained, the electrostatic precipitators can capture up to 95 percent (by weight) of the solid
particulates emitted from oil-fired boilers. At this efficiency smut fallout and plume visibility
can be eliminated. ESP's are limited more by their inability to reduce emissions below about 0.01
Ib/MBtu (4.3 ng/J) than by their maximum collection efficiency. The cost to own and operate preci-
pitators designed to capture 90-95 percent by weight is less than 2.5 percent of the cost to fuel
an industrial boiler rated at 150 MBtu/hr (44 MW) heat input (about 120,000 Ib/hr steam output).
Draft losses across the precipitator are only about 2 inches of water (498 Pa).
1-29
-------
Mechanical cyclone collectors are the other type of collectors used for oil-fired boilers,
but they are not effective for particles smaller than 10 microns. However, this type of collector
is useful to control acid smut emissions during soot blowing, and the large particles that are
emitted from units which do not receive adequate maintenance or careful combustion control.* Total
annual costs for mechanical collectors are about one-fifth those for precipitators, and draft losses
are only slightly higher.
The other two types of collectors, bag filterhouses and wet scrubbers, have been rarely used
in oil-fired boilers. However, particulate or combined particulate and SO scrubbers are being de-
veloped and prototype units have recently been installed. Although some problems remain to be
solved, including disposal of the spent scrubber solution, these preliminary results suggest that
wet scrubbers can remove 99 percent of the particulate (by weight) at costs which are comparable to
those of a precipitator with 90 percent collection efficiency. Bag filterhouses for oil-fired units
also await the solution of some difficult problems, but it is believed that they could collect 99
percent of the particulate emissions at about the same cost as a 90 percent efficient precipitator.
Virtually all watertube boilers greater than 30 MBtu/hr (8.8 MW) and some large firetube
boilers are equipped with soot blowers to remove ash and slag deposits which accumulate in the heat
transfer surfaces. There is no reason to believe that equipment characteristics or medium used
(steam or air) affect the particle emissions since all systems are designed to remove the deposits
effectively. Soot blowing frequency can alter the plume visibility, grain loading, and size dis-
tribution during the blow. Therefore, it can affect the impact of the particulate emissions on the
ambient air if the boiler is uncontrolled. If the boiler is equipped with a sufficiently large
precipitator, the outlet grain loading will be the same during a blow as during other periods.
Opacity monitors are available commercially for the continuous checking of particulate emis-
sions from stacks. Units which satisfy EPA specifications for in-stack continuous monitors on
boilers smaller than 250 MBtu/hr (73 MW) cost $1000-1500 installed. They have generally been con-
nected to alarm systems and/or continuous recorders but not to automatic burner shut-off systems.
(It is also not clear that automatic shutdown of an entire boiler is necessary every time exces-
sive smoke appears.) Oxygen meters (flue gas concentration) have also been relegated to a mon-
'itoring function in most cases, but as they now are quite reliable, more and more will probably
be used in closed-loop burner control systems.
This can be a problem with units smaller than 25,000-100,000 Ib steam/hr (25-100 MBtu/hr, or
7 to 29 MW, heat input), especially when the facility has only one boiler. Usually plants
with many boilers maintain them well, even if each individual boiler is small.
1-30
-------
Opacity monitors which cost only about $500-600 are available for the smaller industrial
boilers and can be used as an alarm system to notify the operator of a problem.
Industrial boilers typically operate at between 60 and 100 percent of their design capacity,
depending on the load requirement. However, information about the effect of boiler load on particu-
late emissions is scarce. The limited available data suggest, for example, that a 40 percent re-
ducation in particulate emissions could be realized by decreasing the boiler load to 50 or 60 per-
cent of rated conditions. However, the capital expenditure penalties of prohibiting boilers from
operating above 60 percent of their design capacity are significant and most probably would not
be acceptable given current shortages in both capital and electric generation capacity. If addi-
tional tests confirm the limited data which show decreasing particulate emissions with reduced
load, mandatory output restrictions could be used during air pollution episodes.
Although there have been claims of up to 50 percent reduction of S03 in the exhaust system
by the use of flue gas additives (e.g., limestone, dolomite, magnesium oxide, or ammonia), total
mass emissions actually increased when these additives were used. The difference was that the
particulate was now in the form of dry, .powdery solid sulfates instead of acid smut. In general,
post-combustion additives are used only to reduce cold-end corrosion and smut deposits inside
the furnace. When resids with high vanadium content must be burned, flue gas additives have to
be used to protect the superheater tubes.
1.3.3.3 Industry Standards
The industry standards that were summarized under the section on commercial boilers also
apply to units in this size category.
1.3.3.4 Existing Regulatory Programs
Virtually all the states regulate mass emissions from industrial boilers, with limits ranging
from 0.04 to 0.6 Ib/MBtu (17.2 to 285 ng/J) for a 100 MBtu/hr (29.3 MW) boiler (or equivalent
grain loading expressed in terms of grains/scf.)* Maryland also requires the use of a dust col-
lector on all resid-fired units greater than 13 MBtu/hr (3.8 MW) heat input and specifies the
collection efficiency for the dust collector. Naturally all boilers in the size group are
allowed to burn residual oil, but the sulfur content in the fuel is generally restricted to less
than 1 percent by weight except for heavily impacted regions. In Massachusetts the allowable
Excluding New Mexico, which expects to relax its limits.
1-31
-------
limits were 0.5 percent in nonmetropolitan areas and 0.3 percent in metropolitan areas such as
Boston. In 1975 these limits were temporarily relaxed to 2.2 and 1.0 percent, respectively, for
boilers that could satisfy certain conditions (see Table 6-5). The issue may be reexamined in
mid-1976. Four of the nine districts contacted require the installation of continuous monitoring
smoke meters in all industrial (and utility) boilers, and one additional region requires it
only for utility boilers. One state, New Jersey, prohibits any visible emissions from boilers
with heat input less than 200 MBtu/hr (58.6 MW) or with a stack whose diameter is less than 60
inches.
The enforcement procedures for industrial and utility boilers were generally the same in
all the regions contacted. Units in this size are the relatively few, easily identified ones.
Any« differences in enforcement procedures between the various regions are probably more subtle
than in the other user groups because the practices used for large sources tend to depend mainly
on the specific experiences gained by the enforcement staff in their dealings with industrial or
utility boiler operators.
1.3.4 Utility Boilers - over 250 MBtu/hr (73 MM)
1.3.4.1 Usage and Emissions
Boilers in the utility size range are all of the watertube type. They are classified by
their firing type i.e., the particular way in which the oil is injected into the boiler. The
three primary firing types for oil-fired units are single wall (or front face) firing, in which
a bank of burners mounted on a plane wall ejects fuel in one direction only; opposed firing, in
which two banks of burners eject fuel toward one another from opposing walls; and tangential firing,
in which the burners are located at the corners of a square and impart a rotational motion to the
combustible mixture. Three others, vertical, turbo-fired, and cyclone firing, are primarily used
with coal, but can be used with oil. The majority of the oil-fired utility boilers in use today
are opposed or tangential fired, although approximately one-half of the smaller ones employ single
wall firing. Coal is the primary fuel, especially in the very large units (>600 MWe). Approxim-
ately one-half of the boilers which are not restricted to coal are configured to burn gas and
oil. Virtually all of the power generated in 1973 by utilities in New England came from oil-fired
boilers. The Middle Atlantic and Pacific regions also depend upon oil to a large extent. Some
of this oil-fired capacity is being shifted to coal in response to oil shortages. Approximately
96 percent of the oil used by utilities is residual oil, and over half of this fuel (55.5 per-
cent) is burned in units which are equipped with particulate control devices.
1-32
-------
Participate emissions from utility boilers are a function of boiler size more than anything
else. Measuremeats indicate that participate emission levels from uncontrolled boilers range from
0.01 to 0.26 Ib/MBtu (4.3 to 112 ng/J) for 10 to 100 MWe units (approximately 100 to 1000 MBtu/hr)
but only from 0.03 to 0.06 Ib/MBtu (12.9 to 25.8 ng/J) for 300 to 600 MWe units (3000 to 6000 MBtu/
hr). The average emission rate for these uncontrolled boilers is about 0.07 Ib/MBtu (30.1 ng/J)
for the smaller units (<100 MWe) which account for about 40 percent of the fuel consumed, and approxi-
mately 0.047 to 0.053 Ib/MBtu (20.2 to 22.8 ng/J) for the,larger units (350 to 600 MWe).
EPA has proposed emission factors for utility size boilers that depend upon fuel grade and,
in the case of No. 6 oil, sulfur content. For a boiler firing a No. 6 oil with 0.5 percent sulfur
content (by weight), the emission factor is 0.053 Ib/MBtu (22.8 ng/J); if the oil contains 3 per-
cent sulfur, the emission factor is 0.22 Ib/MBtu (94.6 ng/J).
1.3.4.2 Particul ate Control
The techniques that were described for the control of particulates from industrial boilers,
especially the large ones, can be applied equally well to utility boilers. The only difference is
that utility boilers can generally reach lower emission levels (per unit heat input) than industrial
units fired on the same fuel because of the size effect. The most promising control techniques are
frequent cleaning of the fuel and air handling equipment, the use of electrostatic precipitators
designed for oil-fired boilers, matching of soot blowing procedures with the particulate collection
equipment, continuous (or at least frequent) flue gas measurements ((X>2 or Og) to maintain optimum
combustion conditions, and the use of opacity monitors to warn the operator of abnormal combustion
conditions that result in smoke and other particulate emission. The effectiveness of the last three
approaches are not known.
It has been suggested that emission reductions of up to 30 percent could be achieved by re-
ducing the interval between major cleaning and inspections from one year to three months. Electro-
static precipitators have been shown to be capable of reducing mass emissions by 50 to 99 percent,
on the average, and eliminating any visible plume. Reduction of over 90 percent (by weight) can
be obtained with correctly sized precipitators that are designed specifically for oil or converted
from a coal system to an oil system. Such precipitators would add about 0.5 percent to the charge
for electricity.
Emissions from controlled units (at an estimated 50 percent collection efficiency) range
from about 0.07 Ib/MBtu (30.1 ng/J) for small units to 0.01 Ib/MBtu (4.3 ng/J) for the largest
units when no additives are employed to reduce cold-end corrosion and slagging. Additives must
1-33
-------
be used when fuels with high vanadium and sodium content are burned, but they result in con-
trolled emissions which are approximately twice as large as they are from boilers which do not
need additives. An average emission factor of 0.025 Ib/MBtu (10.8 ng/J) has been suggested for
controlled utility boilers of all sizes. This average factor is based on existing units, many
of which were designed and first installed on the boiler when it was firing coal. Since the
precipitator was not changed when the boiler was converted to oil, its collection efficiency
is only about 50 percent. If one specifies instead an efficiency of 90 percent, all utility
boilers (>100 MWe) should be able to meet a limit of 0.01 Ib/MBtu (4.3 ng/J)., The large
(>300 MWe) units should even reach levels as low as 0.004 to 0.006 Ib/MBtu (1.7 to 2.6 ng/J).
1.3.4.3 Industry Standards
It is believed that the only industry standards that apply to utility boilers are the ASME
design codes, which do not impact emissions. These large boilers are generally custom built to
the utility's own specifications.
1.3.4.4 Existing Regulatory Program
State regulatory programs for utility boilers are very much like those for industrial
units, except that the mass emission limits are frequently lower for the utilities. Thus 1000
MBtu/hr units (approximately 100 MWe) and larger are limited to values as low as 0.02 Ib/MBtu
(8.6 ng/J) in some states.* New steam generating units greater than 250 MBtu/hr (73 MW) must
also comply with a federal standard of performance of 0.1 Ib/MBtu (43 ng/J) or, where more
stringent, the applicable state or local agency standard. Since many uncontrolled boilers emit
less particulate, the federal standard does not represent the best available control technol-
ogy for particulate from oil-fired units.
1.4 SUMMARY CONTROL MEASURES
This subsection presents a summary of control measures which local or state air pollu-
tion control authorities can implement to attain and maintain particulate national ambient air
quality standards (NAAQS) in their region. These control measures are expected to be applicable
mainly in major urban centers where the need for control of particulate emissions is more de-
manding.
Since the contribution of any source category to the total emissions in an area varies from
region to region, and since the seriousness of the problem (e.g., the current and anticipated
ambient air concentrations of particulate matter) also varies from region to region, ,a control
Excluding New Mexico, which expects to reduce its limits.
1-34
-------
program should be area specific. That is, the user categories which need to be controlled and
the degree to which each of these need to be controlled will depend upon the specific problems
of a given area. Therefore, we shall present separately controls for each user category (resi-
dential, commercial/institutional, industrial, and utility). Within each category, we shall then
rank the control measures in order of increasing effectiveness, taking into account the emphasis
on future sources of emissions, cost, and public acceptance.
The control measures are presented here in tabular form only (text to explain and justify
these accompanies the tables in Section 7). A separate table is used for each size category.
For regulatory purposes the class of boilers and furnaces is divided into four categories based
on size; the names residential, commercial, etc., are used here merely to facilitate reference
to a particular size category.
The format of the tables which are used to present the control measure is shown and ex-
plained by Table 1-5. The tables themselves are repeated here as Nos. 1-6 through 1-9. In most
cases, the control measures would be applied cumulatively. That is, if a local air pollution
control agency decides that measure No, 1 is not sufficient to solve its ambient air problem,
it should consider next the use of both measures No. 1 and 2. If these two still did not re-
duce emissions enough, the agency should then investigate the simultaneous implementation of
measures No. 1,2, and 3, etc. Where measure No. 2 is simply a more stringent version of
measure No. 1, the second one automatically includes the first one.
Air pollution control authorities are encouraged to engage in public information campaigns
("PR") which stress the value to individual home owners of having their oil burner serviced an-
nually. Periodic burner servicing can be a useful way to reduce emissions because tuned and
well maintained burners emit less than those just left by themselves. This is especially true
if burners which are found to be damaged or worn out are replaced. Despite the general aware-
ness of increasing air pollution, individuals still seem to be reluctant to spend funds vol-
untarily to reduce their own contribution; therefore, programs which stress economic gains and
treat pollution reductions as a side benefit will more likely be accepted than those which do
not. All publicity campaigns for improved burner maintenance should be accompanied by exhorta-
tions to conserve energy. Energy conservation is a valid control strategy because particulate
emission rates are generally a direct function of fuel consumption.
1-35
-------
TABLE 1-5.
POSSIBLE PARTICULATE
BOILERS AND FURNACES
CONTROL MEASURES FOR
"SAMPLE FORMAT"'
Rank
1
2
Control Measure
I
'5.
a
«j
£.2
=> 4»>
s!
£ **" i-
i O OJ
e 3 v
C t? 4-»
*»-*J u
O CX 3
V O
** 4J VI
i*i
4J c in
*) id */»
ill
1
111
£42 «
*>» o
3 «*- 4J
§**" o
«j *-o
e « «j
Kf
*o^ «
"c-^5
Ki^+J
*J*O QJ
"in c o c
v «>4-i cn
4-» e c ai
« o eu i_
J^ 4-* (J
"cn
o
13
Qi
S!
c
O)
4J
r- O)
4J (/)
U «J
"ia ^~
T> 4->
r- C
1|
O
tn
35
R)
3 «J
cn en
01 Q)
O£ ^
Effectiveness
Reduction
per unit.
X
J
C
o
J~ ^T
tn cn
in u
g(M
2
c *n
o £
u o
3 U
it!.
ox,
CltJ
O. QJ
fl> O
cn QJ
«3 **-
!_«*-
Q> N)
Estimated a\
each source
t
3
«J
E
1
A
4-*
5
S
XJ
1
"o
i
*f-
o
4J
U
s.
~
Reduction
Per A1r
Region6
,
C U
*W 01 £
S^ 01
x:
15 ° >>
d> s-
f- 3 -Q
4-> VI O
o S u
*" £
O *
"S «£*
tn - o
§"*" "^" » O
IQ i~
4-> O (V
4-> 3
C A3 O
"o S-^ oi
0 1--C S-
O. 3 4J 3
E O ^
^ ^'g g
T3 «n E
ssg^
JE^^
t!E2§
Cost Impact*1
User,
*
"O
C
cn
c
03
§
tO
3
C
C
&-
3 Q>
o «
4-> &.
O Q.
O> O
U- O.
A3 O.
R3
.C
u m
o>
2 3
4-» 4J
o c
APCD
S
f^
&
o
^3 O)
Ol L>
L. 3
11
£. B
ts
O 4J
<^- C
"*- O
fc *-»
O
12
M-^
O =3
Indication <
sources reg
linerav
Impact,6
X
o!
i_
g c
c"
0 VI
E O r-
3 3-^-
tn tn o
O *n M
OJ «^-
>» Cf» t-
cn c 3
1- «J<4-
m x: r
C ^ l^
C QJ
0 Oi^C
, >
4J OJ E
003
O. 3 C
E O O
1 V> t>
Public
Acceptance
j: i c
4J r in o a>
«D tf» *- E
S*^ tO 4-» O
4-* o. o in
o -o »n oe »-
T- -r- O
O OI - > >> >
O) -O .c: o>
I. 4J 4->
3 T3 rcf *O
O J3 i o
a.Z£ tn t
^- cn c »n C
« OJ O 3 3
J_ 4^ u o "O
OJ (O L. (U (U
Q> J3 CL U
« O (O «-
JC-O -M
(O 3 **~ E *-*
E t- cn cn
4J *= O O -r-
c « -o cn
OJ ^- QJ J= OJ
E O t- CHr QJ
O> l_ =J r i
cr>4-> in O
-------
o
o
a;
a:
"fl-
ea
P
o
10
o
«^
O
03
4/7
LU
a:
o v in
«3 '
11
1 >
i
0
c
C 4J 1/1
Bsl
»!!!
Building coc
tives, coope
standards Is
APCD regulat
-*j -i- i-. e a
IM^= =
1 i«8'|si
1
^
£K
ifll}
^ c - « ^ .
4J 3 .,- .r- I_ 1
j
voluntary too
Force. Probably;
f fenders.
? 0 IA
JE *» i.
,
°
!
5
o
1
a
1
PJ
1
t-
§ w
S I/!
II
C r
o .a
I 0
^- c
is
(M
, especially on
>ome users to
l|s
j aJ «
111
'-
a
s
E
=
o cn i
» !-
5"°-'
" <2 o
S
S
S
0
i
~
S
c
o
O)
£
a
o
£
2-tj
trt S.|
£ c a
3 '"i-
o»1o £
^ i K
" "a
o £"52
o a
C *J ,ESS§
C =1 0 -
SJ3 U >,+J
«fl^3
J U U _U 1-
< 41 3 3 0
UJ U Q. D. *J
O
+J
f
IU
"i=s
S^'5-"
°l 0
cn
S
s
3j
-
f
1
i
w s=
^ to
« 1
i s
13 5
c
*£ ra
17
c: S >i
°£^
« +J 0
4J C *
^ «J 0
IS
Hi
a-
rner. Energy ir»-
Effectiveness
d resid.
Ill
Isss
aJ la u
§
s
eo"
o
n
1
oHs
ISs
SB-:
oo£
J
_l
§
1
IO
1
~
o
(O
5
«j
F-
«
g.
o
4-J
restricted
li
in
1 « 1
i« s°S
!isf ss
;u! *J a S-£
*j at S-X> £ -o
s5IS|S
Vf C vl *« 4-> 4-)
u 5 -i- t- y
2 £> > fc J= £
-> 0."- 0 4J tt) ^j
:P.5||^
tlS-s?l=
§
o
1
0
s^
s£
g,
s
o
g
It) D>
.= g
"S £
IS
v» c:
is!
IS
U CL
o a
^4J
_E *
W i
0 *
"3 O
S -o t-
«3 -O c
isl
ill
IO
1
I
at
1
+j
D
1
|
^
S
T3
^
o
S
1
g
c
o
s.
M Q
>> -^
1 1
1 1?
5 l!
s - 1
= s^
_o c ^
i ^11
w» E g _O
E ^ 4J t £
<4- U CJ IO
C o E" -^ =
2 In **~ SJ *J
U ^ u^ « J3
ai e o oj 3
> t. S o t- «
C «J O
"11 ° I <
0 ^00^
o a. t « "oT ^
i'CoSS'1
1 « - -5 b i-
1: * > - 1
1-37
-------
CO
p
I
in
CM
CO
U4
O
Ul
I
o
to
2
V
if
at;
sF
|
u
f
I ,
=
Ji",:
1 Pf
*8
t)§»
"2 *~
«j
c
S.
I
=
»
e
1
.*
Assumes new burners will cost no
more than existing designs. Ap-
proach ts almost volunUry for
existing units too wny sources
to enforce. Probably ignore* by
worst offenders. EPA presently
considering certification agree-
ment with ABHA or HytJrorics Inst.
!
o
3
-
s
s
0
Si
3.3
isl
-EJS
Pi
F.ll
ill!
0 ,
J «*-
*"£ Si uH-s
w O «S] «> MM "**** S
WC 3 **S
V«1 8 ^£^ .
S *° u> =*»
" J I 45 V- 'e: in
S >» S **" c CX
HI ° las?
Host of cost bom by user. First
step to strengthen 1, above, with-
out excessive enforcerent costs
!
c7
2
S
X
Jfof
t^M .
S^SS
-111
£
'o,i
Isi
SS^
35eT
""-S
sisi
ss&^
£
£ sl
Ss"fi
Sfe"
"£ £
|P|
illsl
4-> oi~n
Has noticeable impact only in those
regioris with many rotary cup burners
i
"o"
s
1
§>oio E1
ss«'i
llsl
3
?
s
o
I
i,
£
c ,
S IS
21
Ss
E **
tL!
3 3
hS,
Prohibit rota
prove cooiplia
ing condition
. ... ,
Impact depends on current fuel
usage pattern. Energy impact at
refinery. High cost for cases
where need new burner. Subject
to availability of fuel.
1
CO
B
3
sllss
"
s
i
1
£
I
^
to
"g
u.
^ en
£ n
11
1 1
t
lEpacts winy units. Hay force sore
to replace burner. States are
encouraged to irake reciprocal agree-
rent for licensing servlceren.
1
eg
2
S
!
s
If 00
^ in vl<-*~~
sl"ll
S
S
(0
.
J
"£
V !-
Ct
SI
&
s
S 01
'o.S
|ic
"II
als
S£,
111
..... .... .
Similar to above, but impacts more
sources and will force more to
change burners
i
oo
o
3
-~.
§£{£
§g
Ol
«
g
F
o
a:
*r
u
«st
o
Q
a
c
^
o
o !-
K O .
Requires burner replacement in
addition to above
!
o
CO
o
+J ,
O)
!!!
^-'~-«
31
S
5
S
D)
+J
J|
3S
?".
sl
3:5
o c
ta *
s^l
TJ C3 in
O ro
1
>v
I
£
"
,.
g
«
i-
S
£
g
«
=
O
to
>1 .*<
11
c §. =
K S |
» 5 "
*-» !-
i ll
I "Is
"* » J
S | g i
0 ti S " "
C O E -r- "c
1- v- ut !5 ^
o g. S S T ^
=S .S 3 £ S ..
tl 01 S C C
Irt ^ *-
-------
CM
ro
o
LO
CM
O
O
CQ
cc
Z3
to
UJ
O
a:
o;
a:
a-
oo
CO
o
Lul
_1
CO
$
J3 Q.
SS
11
-o °
t; <
a.
M
o s-
=
g.o
i lii
S §u
is tJi«
Is
>,s
'o ^
c
Control Measure
c
MO 1 a! « t-J
= = fe=S B^S:
«;;; AS" SS" =
"Eu-SfiCTSS-gfl.;
5S..E * = "«»»««
0 u .- 5 a- t: vj.
ISJ: g" ° 8 S| = S
.E \. C-. >, 4J 4J I/I < .£> U +J >l
: o". wS a o» 1- *> C .- **
, +J ^ -0 ~ ~ h ** m "7 1
JESgg ..gfi'SSJ..
5s1 S-» 8+ .S£So
l|»§s ^sgs5!.
°>,)3ao ^""^t^J
lo-j^&I^JJJ^^J:
illlsflll^If
1
S
JOl
=E
_
**
S
i
s ,
llll
fl
c 5 2
(0 Ci
ill 'I
-£ = .,
£ gas- u>
ti E «^= 5
o = c +J *J *^
;*:*' =2
Llfi II
^s si w^- -5
g=-£^| ==^r ;
s?£eo t°t/>° t
*.<«£:** *-
4-io£-OCin ^
i'l^o'r.r *r- o £
^ o *u
tTSwt-«'s 5<£ 2S
Il&£^° a&
-
ti
!M
|!i
en's o
gjoo
"*" 3'* g
lilt.
i
j
3
a
1
^
«, g
§ 3S
1'i"
s
1
VI
i
>^
Cn
O
-A >,
x *io
Require operating penn'ts for (
ing boilers, to be renewed anni
if comply v/itli following:
Distillate SSN 2
Residual SSN 4
~
8 «J *
Lsi!
sE-.g|^
_ O 4->*r- E3 Ol
" £ f S oi"«
ta
o cn.c'a ja
tlflil
3 U »- +J w>
O i- > C C
U I- 3 I- --
E
O
-5 -5
§ §
a>
«
o
s:
s
« = i£ £
SJ-J S- C cn
ro o o c
§ f £ £?'£
^ 5 3s
s i ^^-s^^
Specify burner servicing and c
ing frequency:
Units >30 HBtu/hr (10,000 Ib
stean/hr) on residual
3 months
All other units this categor
6 wonltis
Require operator and maintenan
servicencn to be licensed. Own
advise APCO of service perform
and post-service measurement o
and thermal efficiency (COg an
stack temperature).
*£.
..l-fa'
:Iill^
j^JHj f
E 4->
in -r- a. 4-" J3 > M
-o fc. v
i -r-' 3 O a* o v> o
r- C "O -E 4-> .C O
j-S c5o 8'£-£|£
)
3
U
i
1
ss
o
4->
si
1
^
£
S
s i'S>-4 ^ ""5
0S ^ t!£'B S m4JQ)
1 sill sllssl fs ^-i
vi i. =:f.
= o ,0 , S*.£^
CT ll- U- .,-. 3 .E
^ t/> E.-U 0
2
1-39
-------
IT)
-o
- VIU
I
O
|
1
s
~
o
1
o
o
10
i,
£
O
it
ectors on residua
HBtu/hr xlth de-
fficiency >75X
10 urn.
i^gs
|5s3
fill
^
^ll
Zla-3^
a: (/> S.*2 -tJ
o
£
B
|
"S
0
H
p-
S
1
(O
cn
£
1
nti-corrosion
umulation
i-
ii!
Ill
to
of distillate fuel.
fuel and distillate
&£
r- 5
IS
> c
oc i
sff
₯s«
5SS
o
£
O
3
1
s
M
Cn
rc
0
o
a
cn
£
i
i
3
!E
o
1 =
^
t on enforceable
fl treated optimally.
Kimended at this
s toxic effects and
;h fuels.
41 3 0. T-
t3 *« Q 0 -°
Sou
3 W O J-
O wi 3 4t
< " "S § ?
c ""^ »**"
01
s
«£i,
|| !|
'SE^'s
||li
i!
c
v> o
II
S£
*j t! S .
Ill
CO
1,
V
s
1
§ ,-:
*o
! ,
E
>
£
g
1
1
£
«
: !:
';: ° !
£ ' £ * '
= S -
Ml -
~ S T,
C O Q>
| s| x-
1 1 ° c
g ! .M i
'M (/» 4J Si ti
> t. S S £ 5
o ja 0) **-
X "* " S S i
P a> o o a»
5 § «) 3 c t=
o. «- o> >^ C
"* " « "a -M en
Sa: ^ F _ -o
c/> a> 5 c 3
1-40
-------
O
to
10
LU
CJ
ca
2:
a:
o:
o
^
«c
eC
=3
C_>
It
fe
LU
CO
I(
to
g
O1
I
CO
, * " .
1 i «J U C
iS i etJ JT'^'S
1
7
1
si
1- Ift
O 3
M "
1
s
c
ra
*5
Cl
* =1
0 l;e?
5 M o
3 ?°" o "°
a ° o o
g fe 'o * " S
° °" X JC "~ ^
£g Of- 0 ^
3^- -^ A &
*»l Is
*J O ^ 3 -W
1 S. c. £ j
-j o £; "- -J
*~
1
C9
U
5
1
1
fc
CJ
to
QJ
E
*
to
o
*J
"3
t-
(_>
t- u-
°rf J.
CJ in
a! -a °
c
"i OT J
O-S !-
g"S
g ,
o§
O 3
O r
U)
^.
*.
c
o
J^
^
£
s
t- t- C f.
** 'iJ J_
IfcSfi
S'EC*J
S S 'o S
r£ &g
w> 1 UJ
o 01 "o
11 oS
I ? 0 1
^ o o S -"^
1 o -2 = I
a: t- 'i « o
«
ly. During
ould be
respectively.
«$*
35 &s
£.-?
0 ifl WLO
"£5:^:
°5S
i c c
gogg
.-S^l
illl
1
*v
|
a«
s^l
ssl
1
c
11
Kg
g
TO
"3
£
t^
3
"""a SI
|f BS
i^cj Be
JjO OiC
«s^s
^_ ."^S.
Ilii
s? . !
= o|S
E gs«
r- I/I
**"
e »
§.!,> 8
^] *J & __>, M
O 1- t/1 u»^ o>
.xS.o.Sg.S
a:rsii
C nj 3 irt
°|^=,&S
4J ct^"r- 0) «*-
JSu^tlo01
° ° ^"xZ^
S'u ai<«- ai
>,+j «i +j
fe=B£SS
^iLgli
£RSs|gS
1
;
I
!
"o i
°-l
** S 0
I
en
.£«
R^.
SS
c
o
re
*a
£
S
sS"
S5°
00°.
O 4J^-
« , >^
**" ° 2! S.'al^ £
PjJijj
a c c S. - -5
>, C U-l O T3 TJ
. ai «
|f^|l|.
tslcgssl
o
o
0
s
ro
|
S
o
K
C"
s
1
^
*5
£
S
H
c >
o -
"uj-^
£§
0
V o
o 3
4->
M 3
! !
a. «
to
o
*
i
!
!c
o
i
c[
a
4J
£ '
o
c
s
I
1
g
to
- X
"c =
S £!
I IS
K 3 |
° -S 5i
4J T-
C 0 "S
0 I. N
Irt O i
i ill
E*" c "o
o
= i 1
£ " ^
1 fc " °
S "e -S "o
fe g f g
0 §. '? S
""""?
1 | « 1
J1 5 i " S
s s ; t ? f
X/) LO Q£ M S *^
« £t O T3 O( »-
- -
1-41
-------
REFERENCES FOR SECTION 1
1-1. Monitoring and Air Quality Trends Report, 1973. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina. Publication No. EPA-450/1-74-007. October 1974.
As updated per telecom with T. McMullen, Monitoring and Reports Branch, Monitoring and
Data Analysis Division, EPA, Durham, North Carolina, December 16, 1975.
1-2. Nationwide Emissions Report, National Emissions Data System. Computer printout of emis-
sions file as of June 19, 1975. Obtained from Monitoring Data Analysis Division, OAQPS,
U.S. EPA, Research Triangle Park, North Carolina, September 12, 1975;
1-3. Revised "Compilation of Air Pollutant Emission Factors," Office of Air Quality Programs
and Standards Publication AP-42. In draft review. Revised emission factors received by
telephone from T. Lohre, National Air Data Branch, OAQPS, EPA, Durham, North Carolina,
November 21, 1975.
1-4. Hemeon, W.C.L., "A Critical Review of Regulations for the Control of Particulate Emis-
sions," JAPCA, 23, No. 5, pp. 376-387, May 1973.
1-5. Fennelly, P.F., "Primary and Secondary Particulates as Pollutants -a Literature Review,"
JAPCA 25, No. 7, pp. 697-704, July 1975.
1-6. Dickerson, R.A. and Okuda, A.S., "Design of an Optimum Distillate Oil Burner for Control
of Pollutant Emissions," Rocketdyne Division, Rockwell International, Prepared for U.S. EPA,
Report No. EPA-650/2-74-047, June 1974.
1-7. Data to be published. Received from T. Pace, Control Program Development Division, EPA,
Durham, North Carolina, January-February 1976.
U42
-------
SECTION 2
SOURCE CLASSIFICATION
One of the essentials of an effective air pollution control plan is an appreciation of the
sources to be controlled.. It is necessary,to understand which types of equipment are particularly
large contributors to area-wide'emissions in order that regulatory action can be properly applied.
By "large contributor" is meant not only those units which have high emission factors, but also
those which may have moderate emission factors, but comprise a large fraction of the source popula-
tion. This chapter, therefore, presents both a population distribution for oil-fired units and a
description of each unit type considered.
A 1970 report on particulate emissions from stationary sources (Reference 2-1) indicates
that oil combustion accounts for less than 1 percent of the total .nationwide particulate emissions
from industry and utilities, and slightly more than 2 percent of the emissions from all industrial
and utility sized combustion sources. Table 1-1 also shows that all oil-fired indirect heating com-
bustion sources account for about 1 percent of the total 1975 particulate emissions. However, this
national average does not apply to individual regions. In the northeast United States oil is a
primary fuel, and the contribution of oil combustion to the total particulate emissions is consid-
erably higher than the national average. In the Mew Jersey-New York-Connecticut area, for example,
oil combustion accounts for nearly 20 percent of the combustion generated particulate, and about
one-half of this oil generated pollution is produced by domestic and commercial sized units which
are located in or near high population density areas. On the other hand, in the southeast Florida
AQCR, utility boilers account for about 20 percent of the particulate emissions from all sources
and 90 percent of those from oil burning boilers and furnaces. Nationwide, most of the particulate
emissions from oil-fired sources come from commercial and small industrial boilers, up to about
50 MBtu/hr or 14.6 MW* (Reference 2-27).
Conversion tables between conventional engineering and SI (System International) units are con-
tained in Appendix C. Throughout this document the prefix M represents million in accordance
with SI practices. Also watts, W, or megawatts, MW, refer to heatjnput rates unless subscripted
with an "e" to designate electric output.
2-1
-------
What is indicated here is that effective control programs for participate must be sensitive
to geographical variations in emission sources. Consequently, this chapter gives population dis-
tributions in terms of geographical location for the cases where data are available.
Size Categorization
EPA classifies boilers and furnaces by application, independent of size, for their emissions
inventory. However, for the purposes of regulatory programs, oil-burning units are commonly divided
into the following four categories depending on their size (based on heat input).
Residential -0-0.4 MBtu/hr (0 - 0.12 MW)
t Commercial - 0.4 - 12.5 MBtu/hr (0.12 - 3.66 MW)
Industrial - 12.5 - 250 MBtu/hr (3.66 - 73.25 MW)
Utility - Over 250 MBtu/hr (over 73.25 MW)
Although this categorization suggests a distinct transition from one group to another, a fraction
of the units which belong to one category on the basis of application fall in the size range of
another group. Moreover, in the course of this study, it was discovered that an excellent source
of information exists on the population distribution of the boiler community for two size ranges:
0.3 - 10 MBtu/hr (0.088 - 2.93 MW) and 10 - 250 MBtu/hr (2.93 - 73.25 MW). In light of the small
differences between these size ranges and the ones selected for the present document on the basis
of common usage patterns, the boiler population given for the 0.3 - 10 MBtu/hr (0.088 - 2.93 MW)
group was used for the commercial category and that given for the 10 - 250 MBtu/hr (2.93 - 73.25 MW)
class for the industrial group. The remainder of this chapter presents population distributions
and emission characteristics separately for each one of these four user categories.
2.1 RESIDENTIAL UNITS
Residential units are those boilers and furnaces whose rated heat input (fuel consumption)
is less than 0.4 MBtu/hr (0.12 MW). Equipment in this size range is primarily warm air furnaces
and small cast iron or steel boilers, although about 19 percent of the oil-fired units consist of
stoves, fireplaces, and miscellaneous homemade units (see Reference 2-17 and Figure 2-1). Due to
the diversity in the design of these units and to the uncertainty in their number, they will not
be discussed here.
2-2
-------
in
0)
J/l
O)
u
It)
(/I
OJ
O
4->
00
fO
01
o>
O
Q.
<0
OJ
*J
l/l
S-
01
*->
(O
-M
o
CO
VD
oo
I
CD
ro
O)
c
0)
T3
I/)
o>
s-
c
O
-o
J-
fO
c
o
(O
(J
-------
2.1.1 Residential Equipment Description
2.1.1.1 Warm Air Furnaces
A warm air furnace is a self-enclosed appliance used for heating air for a house or build-
ing. It discharges hot air directly into the space'being heated or more commonly through ducts
which transport the warm air to the area to be heated. Some furnaces operate by gravity (buoyancy
forces), but the majority are forced air systems, using a blower to move the air through the ducts
and back to the furnace.
The furnace consists of a burner with related piping and controls, a heat exchanger, and a
blower for forced air systems. The furnace package is enclosed in a steel casing and the heat ex-
changer compartment is insulated to improve efficiency and to limit the outer casing temperature
for personnel safety. The combustion takes place in the primary combustion space of the metal-
walled or refractory-lined heat exchanger, and the combustion products pass through secondary flue
gas passageways of the heat exchanger to exit through a flue to the atmosphere. The air to be
heated circulates over the outside of the heat exchanger to the warm air discharge, where it leaves
at a temperature of 100°F to 150°F (38°C to 65°C). Furnaces are fired by natural gas and distil-
late oil and (rarely) by coal.
Domestic or residential warm air furnaces are available in capacities under 0.4 MBtu/hr
(0.12 MW) and are generally classified into four types:
Downflow, in which the.blower is mounted above the heat exchanger, the flue gas dis-
charges at the side or top, and the warm air discharges from the bottom of the unit
(counter-flow).
Upflow or "high-boy," which is similar to downflow except that the blower is mounted
below the heat exchanger and both the warm air and flue gas discharge from the top of
the unit (co-flow). Figure 2-2 shows a typical oil-fired upflow furnace.
Horizontal, in which the blower is mounted next to the heat exchanger, the flue gas
discharging at the top, and the warm air at the end, of the unit.
Basement or "low-boy," which is very similar to the horizontal unit except that the
flue gas discharges at the side and the warm air from the top. In addition, the heat
exchanger may be higher and narrower than for a horizontal furnace.
All of these units are available in a wide range of capacities. The choice of the appro-
priate type is based most often on available installation space. For example, horizontal units
are intended for use in situations where headroom is limited, such as in attics or crawl spaces.
2-4
-------
LD
U_
Belt drive blower
Flue
Burner
Heat
Exchanger
Blower
Direct drive blower
Figure 2-2. Forced warm air oil-fired furnace upflow
(Courtesy: 'Lennox Industries).
2-5
-------
Two types of blowers are found on modern forced warm air furnaces; direct drive and belt
drive. Direct drive blowers are of the multiple speed type and usually are found on combination
heating and cooling furnaces. However, they are also found on units used for heating only. When
properly connected electrically, the blower will run at slower speeds in the winter months and at
a faster speed during the summer months.
Belt drive blowers operate at one speed, which is dependent on the adjustment of the pulley
system. This speed is set by the serviceman during furnace installation and can have a major im-
pact on efficiency if done improperly.
The heat in a typical warm air furnace is supplied by an oil burner of the type pictured in
Figure 2-2. These burners consist of a combustion air blower, motor, fuel pump, spark ignition
system, main air tube and swirlers, and fuel nozzle. The fuel flow rate is determined by the oil
nozzle orifice size, and the total air flow rate by the blower and a damper in the exhaust flue.
The proper air-to-fuel ratio is obtained by adjusting the damper until the optimum C02 and smoke
levels are achieved.
The burner is mounted in a refractory or refractory-felt lined combustion chamber which is
cooled by air drawn from the house. From the combustion chamber the flue'gas passes through a
heat exchanger and out the stack. The burner blower may supply the full pressure to exhaust the
gas from the stack or it may just assist the buoyancy forces in the stack. Table 2-1 lists typi-
cal operating conditions for this type of furnace.
The procedure for the design of a warm air furnace is empirical and relies on prototype
development, experience, and testing. Standards for construction and thermal efficiency are set
by Underwriters Laboratory and American National Standaards Institute (ANSI).
TABLE 2-1. CHARACTERISTICS OF A TYPICAL 0.1 MBtu/hr (29.3 kW)
WARM AIR OIL-FIRED FURNACE
Excess Air: 10 - 100 percent
Flue Exit Diameter: 5" to 7" (130 - 180 mm)
Heat Exchange Area; 20 ft2 to 30 ft2 (1.8 - 2.8 m2)
Overall Heat Transfer Coefficient: 2 to 3 Btu/hr-ft2-°F (11 - 17 watts/m2-°C)
Exit Flue Gas Temperature: 500°F - 900°F (260°C - 482°C)
Combustion Chamber Pressure: 0.05" W.6. to 0.2" W.6. (12 - 50 Pa)
Recirculating Air Flow: 800 to 1300 scftn (0.38 - 0.61 m3/sec)
Temperature Rise on Air Side: 75°F to 80°F (41 °C - 45°C)
Overall Steady State Efficiency: 70 to 80 percent
Common Operating Mode: on/off
Ignition System: Spark
Draft System: Forced
2-6
-------
Nearly all of the design variables for a warm-air furnace affect participate emissions to
some degree. As will be discussed below the more significant effects are associated with the burner,
specifically, Mts design and nozzle type. Additionally, combustion chamber materials, usage pat-
terns, and tuning can cause significant variations in particulate emissions.
The burner design features which most strongly affect particulate emissions include:
" Fuel atomization techniques
Relative fuel droplet/air velocity
Flame retention devices
Combustion chamber.material
Combustion chamber wall temperature
These design parameters will be discussed briefly in the order presented. Throughout the
discussion, it should be noted that the adjustment of a burner parameter to obtain low particulate
emissions often results in a corresponding increase in NO due to improved and hotter combustion,
« \
or a decrease in efficiency due to the use of too much excess air. Consequently, a burner designed
for low total emissions may involve a compromise among individual species emissions. The desirable
situation is one in which combustion takes place over a long enough time in. a hot enough environ-
ment to completely oxidize all hydrocarbons to C02 and water with little CO produced, and yet one
in which the temperature is not so high as to provide significant amounts of nitrogen oxides. In
other words, a long, slow (but complete) combustion is called for.
Fuel Atomization Techniques
Mechanical atomization is the primary method used in domestic sized furnaces. Other methods
used are wall flame rotary, vaporizing, and low pressure burners, but the combined usage of these
types accounts for only about 15 percent of the total. For a complete discussion of fuel atomiza-
tion in domestic and commercial burners, the reader is referred to Reference 2-3.
High pressure mechanical atomization accounts for by far the majority of domestic and com-
mercial fuel atomization systems. These units operate by supplying oil at high pressure (100 psig
or 690 kPa, nominally) to one or more orifices in the nozzle. The design of the orifice influences
the spray pattern and, to some degree, the droplet size. The flow channel to the orifice is routed
in such a way as to impart a high degree of swirl to the exiting oil. Combustion air, which flows
in an annular region around the gun, is also given swirl before it mixes with the fuel spray out-
side the nozzle.
2-7
-------
In general, a higher atomization pressure leads to a more finely atomized spray which should
lead to cleaner combustion. However, Sjorgren (Reference 2-4) has shown that although higher fuel
pressures achieve better atomization, the relative droplet/air velocity may decrease resulting in
inferior combustion (see internal recirculation, Section 4.1.4.2). Dickerson, et al. (Reference
2-5) determined that "commercially used atomization nozzles which are available at very low cost
(less than about $1.00 per nozzle, nominally) do an excellent job of atomization," and parti oil ate
emissions from these nozzles can be very low.
The firing rate of a high pressure gun burner is determined by the oil pressure and the
orifice size. The firing rate in turn dictates the air flow rate through the burner and, conse-
quently, the residence time (i.e., the time during which a volume of fuel is subjected to a com-
bustion environment). Short residence time is a major cause of high CO, unburned hydrocarbon,
and particulate emissions (as shown again by Reference 2-6). Underfiring the burner by instal-
lation of a smaller orifice nozzle (at the nominal cost of one dollar) represents an elementary
way of increasing the residence time and reducing emissions, provided the heat load can be carried.
Underfiring also reduces emissions in another way. Particulate emission rates are higher immedi-
ately after ignition, due to the poor combustion which occurs at low temperatures. Underfiring
causes the burner to remain on longer to provide a given total heat output and, consequently,
the burner is turned on and off fewer times, resulting in a reduction in the number of high emis-
sion periods.
Low pressure mechanical atomization systems account for only about 8 percent of all burners
in the residential section, and there is a continuing trend toward their elimination. Low pressure
guns are unique with respect to other domestic burners in that oil and a portion of the combustion
air are mixed inside the nozzle. Once this mixture is ejected from the nozzle, it contacts a
highly turbulent supply of oxidizing air and combustion commences. The operation of this burner
requires a supply of air at 1 to 15 psi (6.9 - 103 kPa) and oil at 1 to 5 psi (6.9 - 35 kPa). The
need for pressurized air is an additional complication compared to high pressure nozzles.
Vaporizing burners are becoming rare although they are still common in the Southeast and
some other regions with mild climates. They consist of a small container in which a fuel is heated
to a highly volatile condition. The resulting vapor is mixed with air and burned. A small, con-
tinuously burning pilot flame is generally utilized for ignition purposes, and this flame can be a
source of smoke.
2-8
-------
Flame Retention Devices . . . .
Many small combustors use flame retention devices in the form of plates, grids, or cones
to produce local recirculation zones and thus reduce smoke emission. These devices also allow
the burner to be run on much less excess air than would otherwise be required. Operation at low
excess air improves thermal efficiency (less heated air is-discharged out the stack) and, there-
fore, reduces fuel consumption. In a study by Beach and Siegmund (Reference 2-11) a typical home
heater oil burner required 85 percent excess air to achieve a Bacharach No. 2 Smoke reading, where-
as with the flame retention device the burner required less than 10 percent excess air to achieve
the same smoke number. Other studies along these same lines include those by Walsh and by Howekamp
and Hooper (References 2-12 and 2-13). Howekamp and Hooper studied five combustion improving de-
vices for domestic home heaters which produce either a swirling flow or recirculation zones with a
flame retention device. Not all of these devices were effective in reducing smoke over the whole
range of air-to-fuel ratios. These results are also summarized in Reference 2-6. A paper by T. D.
Brown (Reference 2-14), which also compares a flame retention head to a swirl head in a domestic
oil burner supports the theory that there exists an optimum amount of swirl to achieve minimum
smoke for a given excess air level (see Section 4.1.4.2).
Barrett, et al. (Reference 2-15) measured particulate emissions from several oil burners in-
stalled.for domestic and commercial use. The data indicated an average particulate emission rate
of 0.012 Ib/MBtu (5.16 ng/J) for a burner with a conventional head, versus 0.0056 Ib/MBtu (2.39
ng/J) for one with a flame retention device.* The flame retention burners also produced lower CO,
HC, and NOX emissions than did the conventional types. Laboratory experiments (References 2-5 and
2-6) have shown that not all flame retention burners simultaneously reduce particulates and NO ,
, X
and in fact, some showed marked increases in NO . Therefore care should be exercised in extrapo-
lating the average reductions measured by Barrett, et al., to a special flame retention device.
Combustion Wall Temperature .
Cool walls in a combustion chamber increase particulate emissions due to poor combustion in
low temperature regions. The coupled effects of cool walls and a short residence time result in the
incomplete combustion of carbon particles. A typical increase of 1 to 2 Bacharach smoke numbers is
found to occur when the wall temperature is decreased from 2600°F to 1600°F (1430°C - 870°C) (Ref-
erence 2-10).
These numbers are for tuned units and are inferred from measurements made by the "Modified EPA"
(MEPA) method of Barrett, et al., using an average correlation factor of 1.67 to relate MEPA data
to that obtained by the standard EPA sampling train.
2-9
-------
While hot walls promote complete combustion of participates (and at the same time reduce
CO and unburned hydrocarbon emissions), they result in an increase in NOX emissions. Again, a com-
promise must be made.
Combustion Chamber Material
Combustion chambers of domestic oil burners are usually cylindrical and are made of hard
(noninsulating) refractory, soft (insulating) refractory, felt ceramic, or steel. Light materials,
such as the soft refractory or felt ceramic, with low conductivity are preferrable, especially in
burners which see cyclic operation, because their surface temperature reaches equilibrium with com-
bustion gases faster than does that of a hard refractory or steel combustion chamber. As previously
mentioned, a high combustion chamber wall temperature is essential to low particulate combustion.
During the time in which the chamber temperature is low, excess air levels must be increased to
counteract the poor combustion due to low temperature. Since it has not been practical to allow
for altering the amount of excess air during a heating cycle, the initial high air level is main-
tained throughout, resulting in decreased efficiency. However, in units larger than 20 gph (21 cm3/
sec), combustion volume temperatures generally exceed 3000°F (1650°C) and, therefore, hard ceramic
must be used.
Burner Use Pattern
The on-off cycling of an oil burner has a serious effect on particulate emissions as well as
other pollutant species. Particulate produced during the incomplete combustion which occurs during
ignition and shut-down contribute a large percentage of the total from a complete cycle. Therefore,
short cycles, which turn the burner on and off many times a day, should be avoided
The dependence of emissions on cycling is discussed more completely in Section 4.2.2.1.
Burner Age
Barrett, et al. (Reference 2-15) measured emissions from a set of older oil burners and com-
pared the results with those from newer units. They concluded that particulate emissions from newer
burners are lower than from older ones, and they attribute this not to an aging effect, but to the
improved design practice of recent years. Particulate emissions from a properly maintained oil
burner should not increase as the burner ages.
2-10
-------
Tuning
The importance of tuning as a means of controlling particulate emissions is witnessed by
References 2-15 and 2-16 in which particulate emissions were measured for a number of oil burners
in domestic and commercial installations.
The oil burners used were chosen as typical of the national distribution of equipment types,
and particulate emissions were measured both in the "as found" condition and after tuning. (The
tuning procedure was directed toward reduction of both CO and particulate while maintaining a high
efficiency.) Results show an average emission of 0.021 Ib/MBtu (9.03 ng/J) in the "as found" con-
dition as compared to 0.016 Ib/MBtu (6.88 ng/J) in the tuned condition. Tuning resulted in a 24
percent overall reduction, a condition indicating the desirability of a periodic (e.g., yearly)
inspection of all burners by a qualified serviceman.
Tuning an oil burner consists primarily of cleaning and inspecting the pump, nozzle, and
combustion chamber and readjusting the air flow to the optimum air/fuel ratio (i.e., excess air
level). The variation of emissions as a function of excess air is shown in Figure 2-3. As in-
dicated, low excess air is required for maximum efficiency. However, reduction of excess air be-
low about 60 percent (for the burner whose characteristics are given in Figure 2-3) causes a marked
increase in smoke and, to a lesser degree, CO and HC. The tuning must then be a compromise between
emissions and efficiency, the determination of which involves several measurements (e.g., CO, C0?,
stack temperature^ draft). The need for qualified service personnel is indicated. Tuning is dis-
cussed more fully in Chapter 4.
2.1.1.2 Steam and Hot Water Units
Approximately 36 percent of the residential heating in this country is supplied by steam or
hot water units using a boiler. Heat for the production of steam or hot water is transferred from
the combustion products to the boiler in the firebox. The boiler heat transfer .surfaces usually
consist of steel or cast iron tubes that are exposed to, or carry, the hot combustion gases.
Cast Iron Boilers -Capacities up to 13.5 MBtu/hr (3.95 MH)
Many domestic and small commercial boilers in use today have heat exchangers that are con-
structed of cast iron. They are designed for supplying low pressure steam at about 15 psig maxi-
mum (103 kPa) or hot water at 160 psig maximum (1.1 MPa) and 250°F (121°C) and are used primarily
for hydronic heating systems. These boilers consist of an assembly of cast iron sections. Generally
2-11
-------
luaoaad '(,.OL x)
ipejeippg '.adorns
6 /6m '03
CM
O
O
O) E
0)
C 'I-
nj w
01
» s-
s_
O
3 S-
O
0) O
- E evj
(U o
\^ tM- ni
O -C O
E O £=
""52
<(- 4-> 0)
O 1/1 4-
OJ (/I
s. s- s.
4-> 0)
(0 >, 3
S- 0 J3
(1) C
C OJ i
-------
cool water enters at the bottom of the sections, and hot water or steam exits at the top. The pro-
ducts of combustion are conducted through labyrinth passages cast into the sections. The capacity
of a sectional boiler may be varied by the addition or deletion of sections.
Figure 2-4 shows an oil-fired cast iron boiler, typical of those quite common in older homes
in the Northcentral and Northeast states. These boilers are similar in some respects to oil-fired
Warm air furnaces in that they may employ a refractory-lined combustion chamber followed by the heat
exchange surface.
It has been reported that poor matching of burner and combustion chamber requirements occurred
frequently in the past, partly because the burner and the furnace manufacturers did not work to-
gether to avoid this problem (Reference 2-5). However, the recent emphasis on emission reductions
has induced burner and furnace manufacturers to coordinate their designs, and mismatching is no
longer a problem. Typically, because of careless installation considerable leakage of secondary
air into the firebox is allowed, which serves to lower the flame temperature. Leakage probably
results in overall lower boiler efficiency. Typical operating conditions and design characteris-
tics for an oil-fired system are given in Table 2-2.
Cast iron boilers are rated by the Hydronics Institute (IBR ratings). They must be designed
and constructed in accordance with the ASME Code, Section IV, Heating Boilers. The IBR Testing and
Rating Code specifies the test conditions and such things as minimum efficiency, percentage of car-
bon dioxide in the flue gas, flue gas temperature, and smoke reading. The ratings cover a range of
0.048 MBtu/hr (0.014 MW) to 13.5 MBtu/hr (3.9 MW). Many of the design features associated with fur-
naces are also directly applicable to boilers. For instance, flame retention heads are used on
burners in boilers as well as in furnaces, and they are as effective in boilers as in furnaces as
a means of reducing particulate emissions. In fact, because of their relatively high noise level,
flame retention head burners are less common on furnaces than on boilers. Similarly, considera-
tions of nozzle type, residence time, combustion wall material and temperature, cycling patterns,
tuning, etc., are equally applicable.
2.1.2 Emissions
Emission measurements from 15 residential cast iron boilers are reported in Reference 2-15
and compared to similar measurements on 11 warm air furnaces. These tests showed that mean CO and
particulate emissions for the furnaces (0.018 Ib/MBtu (7.74 ng/J)) were consistently and appreciably
higher than for the boilers; however, the mean CO levels for the furnaces were not considered exces-
sive. The higher CO level with furnaces could not be explained in view of higher combustion chamber
2-13
-------
Flue
Burner
Heat
Exchanger
Figure 2-4. Oil-fired cast iron boiler (courtesy of the
Utica Radiator Company).
2-14
-------
TABLE 2-2. CHARACTERISTICS OF A TYPICAL 0.1 MBtu/hr (29.3 kW)
OIL-FIRED CAST IRON BOILER
Recirculating Water Flow
Excess:Air .: '
Flue Exit Diameter
Exit Flue Gas Temperature
(Upstream Draft Control)
Combustion Chamber Pressure
Temperature Rise on Water Side
Overall Steady State Efficiency
3-15 gpm (189 - 945 cm3/sec)
30 - 100 percent
3.5-5 inch (89 - 127 mm)
400°F- 600°F (204°C - 315°C)
0.02 inch W.6. (4.98 Pa)
10°F - 40°F (5.5°C - 22°C)
80 - 85 percent
2-15
-------
temperatures generally expected in furnaces. Mean NOX emissions for the furnaces were slightly, but
consistently, higher than for the boilers, probably reflecting the i.'.qher combustion chamber tempera-
tures in the furnaces.
2.1.3 Population Distribution (Residential Units)
The population of oil burning residential units is categorized in the following subsections
in terms of unit type, burner type, combustion chamber material, and average age with size and geo-
graphic location as parameters. According to data obtained from the Gas Appliance Manufacturers
Association (GAMA), the trend in the last 10 years had been toward gas-fired forced warm air sys-
tems. The 1970 U.S. Census data substantiated this finding by indicating that almost 30 percent
of the warm air furnaces were oil-fired in houses built prior to 1939 compared to approximately
15 percent in houses built during the period 1960 to 1970. Increasing gas shortages are expected
to reverse this trend.
2.1.3.1 Population Distribution by Unit Type
Of the 53.7 million residential heating units in operation in the United States in 1970, 26.3
percent were oil-fired, 57.7 percent were gas-fired, and 14 percent used other fuels such as propane,
coal, and wood (Reference 2-17). The generation of water for washing consumes about one-quarter the
energy used for space heating (Reference 2-25). The population distribution of the primary types of
oil-fired residential heating units in use in the United States is shown in Table 2-3.
TABLE 2-3. RESIDENTIAL OIL BURNING UNITS (1970) (SOURCE: REFERENCE 2-17)
Type of Unit
Steam or hot water
Warm air furnaces
Other
Percent of Total
36
45
19
In computing these percentages from census data, the following two assumptions were made:
All steam units and central warm air furnaces listed as servicing buildings containing
more than five residential housing units were assumed to have capacities over 0.4 MBtu/hr
(0.12 MW), and were therefore not included in the residential category.
All direct space heating systems used in residences of any size were assumed to have a
capacity of less than 0.4 MBtu/hr (0.12 MW).
2-16
-------
Some comment is necessary concerning the category denoted as "other." Statistics from the
United States Census (1970) show this category to include approximately 19 percent of all oil burn-
ing units, and it therefore comprises a significant fraction of the total population. Units in
this category include stoves, fireplaces and a variety of homemade units. Due to the unavailability
of data on the population of units of this type, they have been omitted in the breakdown by size.
Table 2-5 shows.the population distribution for oil-fired residential units by unit type for
three size ranges. Si/ice some of this information is based on 1970 sales data, which are not neces-
sarily identical to sales in previous years, this table provides only a reasonable estimate of the
1 '
current distribution. Gas burning units are included for comparison.
2.1^3.2 Population Distribution by Burner Type
Table 2-6 shows the population distribution of oil-fired units categorized by burner type,
with size and geographic location as parameters. Unfortunately, published data are not available
in the same size subcategories as were used in Section 2.1.2.1. However, the data here indicates
the differences in the type of burner used in the "large" and "small" categories.
Approximately 10 percent of the burners now in use are of the low pressure type, 5 percent
are rotary, and 1 percent are of the vaporizing variety. The majority of the burners are of the
high pressure atomizing type.
Recent sales statistics (Table 2-4 below) indicate a significant trend toward the more effi-
cient high pressure gun burners with a marked decrease in the rotary and vaporizing types.
TABLE 2-4. SALES TRENDS FOR RESIDENTIAL OIL BURNERS BY TYPE
(SOURCE: REFERENCE 2-16)
Oil Burner Type
High-pressure
Low-pressure
Wall flame rotary
Vaporizing
Misc. types
1969 Percent
95.0
3.4
0.4
1.2
1941 Percent
71.7
6.2
11.0
10.8
0.3
2-17
-------
K
u
0.
CO LU
o
LU Z
O LU
t-t C£
> UJ
C£ LL.
LU LU
CO Qi
3 LU
os:
z Q.
ii
CO n>
I o-
H-1 LU
LU
o a
I I LU
co o:
LU HH
LL, CO
0-=?
o u.
>> o
CQ LU
co
CO
I LU
So
i a:
rs zs
a. o
o co
LO
t
CVJ
LU
o ,__^
d£!
1 1
£=*
°~
-5'
°* .
1 *****
,
" CNJ
^^ ^^
_
"
o ^
CM
1 1
°s
$9"
re^1
O-
re i-
CJ-C
T3 3
CU +->
4-> CO
10 S
Di
CO CVJ «^3 «d-
VD CO LO *
«d- in co i
LO LO O 0
O CD «d- VO
UD co oo r--.
co i^. r en
CO i IO CO
CM !». CT>
S- S-
O) O)
fO (O
s S
o o
£: J^
(O (/)
S_ S- S- J_ S_ S-
^.
en
c:
re
S-
cu
c
CD
en
c
1
X5
CU
3
C/)
0
r-
-P
0.
ci
13
(/)
C/)
«=c
2-18
-------
TABLE 2-6. POPULATION DISTRIBUTION OF RESIDENTIAL OIL BURNERS BY BURNER TYPE*.,
New England
Mid-Atlantic
South Atlantic
Midwest .
West
All Sections
New England
Mid-Atlantic ,.
South Atlantic
Midwest
West
All Sections
All Sections
High Pressure
Conventional
44.9
66.3
74.8
63.0
80.3
63.1
61.4
69.0
89.4
73.3
80.3
71.2
.
68.3 ,
Shell**
Head
Retention
Head
Low
Pressure
Rotary
Percent of Total
Less
7.2
8.4
1.9
9.4
0.6
7.2
0.120
7.9
5.2
2.3
13.1
0.6
6.9
than 0.120 MBtu/hr (35 kw)
14.7
3.8
13.0
13.1
5.2
9.1
- 0.360 MBti
13.6
8.2
7.3
6.2
2.9
8.3
26.4
10.6
0.8
5.6
8.1
11.5
i/hr (35 - 1C
9.0
10.7
0.8
6.0
13.2
8.6
5.8
8.5
0.1
8.6
0.5
6.6
)5 kw)
7.6
6.8
0.2
1.3
1.8
4.7
Ml Oil Burners up to 0.360 MBtu/hr (105 kv
7.0
8.6
9.6
5.4
Vaporizing
1.0
2.4
9.4
0.3
5.3
2.5
0.5
0.1
0.1
1.2
0.3
)
1.1
Source: Reference 2-15.
**A combustion improving device which utilizes swirl
mixing to promote clean combustion.
2-19
-------
The majority of wall flame rotary burners and vaporizing burners now in service are installed in
older units; they are expected.to be almost nonexistent (in residential equipment) in the future
unless newly designed vaporizing burners for No. 2 oil gain widespread acceptance.
2.1.3.3 Population Distribution by Combustion Chamber Material
The type of material which lines the combustion chamber of an oil burner has been shown to
have an effect on flame properties, and consequently, on emissions (see Section 2.1.1.1). For this
reason, Table 2-7 is included showing the population distribution of oil burner combustion chamber
material in two size ranges and several geographic areas. A majority of the chambers are con-
structed from firebrick (hard and soft), but new ceramic materials with improved characteristics
are expected to gain a larger share of the market in years to come. The rate at which this dis-
tribution is changing due to current sales trends is unknown.
2.1.3.4 Average Age of Oil-Burning Equipment
The average age of various types of residential oil-burning equipment is shown in Table 2-8
with geographic location as a parameter. It should be noted that in constructing this table, a
distinction has been clearly made between burners and furnaces or boilers (i.e., complete units).
The age of a unit is related to emissions in the usual sense of increasing discharge due to
nozzle clogging, pump wear, etc. Moreover, since many residential units are seldom serviced, the
age of a burner is often coincident with the length of time to the last maintenance. The small
average age of retention head burners is due to their rather recent inception.
2.2 COMMERCIAL UNITS - 0.4 - 12.5 MBtu/hr (0.12 - 3.66 MW)
Commercial oil-fired systems are used mainly for space heating and hot water just as with
residential equipment. Therefore, this category can also be subdivided into warm air furnaces
and boilers. Unlike residential units, however, several different kinds of commercial boilers
are manufactured (see Figure 2-5).
The distribution of equipment types for heating commercial buildings varies around the
country. On the West Coast, or in other moderate climates, about 85 percent of the heating units
are warm air furnaces. Nationwide, 11 percent of the housing units listed in the 1970 U.S. Census
and estimated to be heated by units of commercial size utilize warm air furnaces. Warm air fur-
naces have a lower initial cost than boilers, and the latter require more elaborate piping systems.
Therefore, if comfort heating is the only requirement, and if mildly fluctuating delivered air
2-20
-------
*
5:
LU
CQ
I
o
00
CQ
a:
LU
o
_j
=C
LU
O
O
I I
a
s:
o
Jl
p
a
E
o
+5 *f
3 E
co1-
o
II
p
T3
0) i
C
LU
(/I
,_ 0
01
OO
c Location
Geographi
Is*
CO t-. in p
CO CM OO O
Is*"
- 'in ' * g
U3 CM r- |°
a- co co o
Is*
(43 CM CM O
ID CM r- lO
i CT> O 2
U3 i CM ~
O CM CO O
ID CM i p
CO *
CO vx
' o
S- Z S-
.c ja 0)
cu »-> S- '<£
+-> CQ O
10 E: +-> o
az o-i-
CM re E i
Oil i. 10 0)
C r- 14- S- 0)
i- 01 oi -p re
s- o o: o oo -P
i- V O
.u- 1-
10 0 «* 10
U3 r CM O
r r CO O
(^3 CO O
CO f-» O O
lO CO O
.
CO CM O O
^ ^ r- JO
Is*
<3* ID O |O
CO
CO ^ - .
O
s- si s-.
^: ja o
C£ O-i-
CM (O E i
Oil S- (O <1J
C r If- S- O) i
r- 01 01 -P re
s- o o; o oo -p
-A 0
U- h-
O
s-
JD
LO
r- -P
1 <)-
CM O
VI
0)
o -a
c c
re
a) -a
1- S-
at re
*
Source:
**
Includes
2-21
-------
CO
I
o
u.
CO
CM
^
£
U>
.g
r-
U
-g'43
"i
o
E «
p
O
CU r
C
UJ
U)
_§
(U
CO
c
o
+J
re
u
_j
u
jr
0.
2
en
r-. o i~~ ID (--
CM 1 CO CO CO
i co r~- co co o
^* co co i""* Is*" r**>
r O O CO I-.
0 r- co en co
«3- O1 CM CM r>- CO
CM co "* co i-. in
i in * co
co en «J- -3- ID
r ^ ' "
CD i~~ cn in in en
CM co co co r~. co
1
T3
I i 0> N
r '£ _J D; >
b
o in co co
CO CO CO CM
> O Ol O
^: i± 5: £2
i r-* in en
i r-. ^ CM en
r~. * «i- «*
d- i~- co CM
CTl O CT* CT»
3- CO VD r
CO CO CO CM
co i-~ in CM
I.
u
0
CQ
en **^
.? s
4-> re
re c
01 S-
r- in 13
o +* O) (/>
c 0 s- c s.
i
-------
Commercial
Equipment
-
_ Warm Air
Furnaces
Tube! ess
Packaged
(wall -fired)
Firetube
(wall -fired)
-Coil
- Firebox
Scotch
Horizontal Return Tube
Firebox
Vertical
- Cast Iron
Figure 2-5. Commercial heating equipment types.
2-23
-------
temperatures can be tolerated, warm air furnaces are chosen over boilers. High rise and spread-
out buildings, however, almost always require boilers due to a demand for both hot water and warm
air, to space limitations for ducting, and to equipment size limitations. Warm air furnaces are
available in sizes up to approximately 5 MBtu/hr (1.46 MW) while boilers theoretically can be as
large as central station power plants. Commercial warm air furnaces and boilers are fired pri-
marily with distillate fuel oil and natural gas, although some boilers in this size group burn
residual fuel oil and, to a small degree, process gas. Nationwide, about one-sixth as much energy
is used to generate hot water for washing in commercial facilities as is used to provide space heat-
ing (Reference 2-20).
The type of boiler chosen for a given commercial heating application depends mostly on the
size of boiler required or its function. Generally speaking, cast iron sectional boilers are used
for supplying most low pressure steam or hot water in small to medium size projects. Firetube
boilers, most of which are packaged, are available in sizes up to 50 MBtu/hr (14.65 MW), while
packaged* watertube boilers are available in capacities between 0.4 MBtu/hr and 250 MBtu/hr (0.12 -
73.25 MW).
2.2.1 Commercial Equipment Description
Commercial size heating equipment is available as warm air furnaces, cast iron boilers, fire-
tube boilers, and watertube boilers. Commercial size cast iron boilers are identical to residential
units except for size, Since cast iron boilers were discussed in Section 2.1.1.2, they will not be
described here. Watertube boilers, typical of operations larger than commercial size, comprise
only about 5 percent of all commercial boilers and will be described in Section 2.3.1.1. Therefore,
the remainder of this subsection is restricted to firetube boilers arid commercial furnaces.
2.2.1.1 Commercial Warm Oil Furnaces
The commercial warm air furnace design concept resembles that used for residential heating
purposes except that the larger scale equipment dictates that the unit usually be mounted on the
roof of the building. The designs of these units vary considerably among manufacturers, unlike
the domestic sector. The apparent standardization of the latter is due mainly to the large sales
volume as well as the extremely competitive marketplace. These conditions differ sufficiently in
the commercial heating sector to allow some design innovations, especially in gas-fired units.
A packaged boiler is a unit which is engineered, built and warranteed by one manufacturer. Thus,
the unit is assembled (packaged) at the manufacturers plant and fire tested before shipment.
2-24
-------
The burners in these units tend to be inherently more versatile than in residential units.
For example, many have the ability to operate at a greatly reduced load (large turndown capability).
Novel heat exchangers are also more common in these units. These may include finned heat exchanger
tubes, different heat exchanger materials, and more flexibility in design. Higher quality con-
struction materials and larger interior volumes are other factors that distinguish commercial-size
from domestic-size furnaces. In general, burner and combustion chamber design modifications have
the same effect on emissions and efficiency in commercial furnaces as they do in the smaller .domes-
tic units discussed earlier. Therefore, that entire discussion applies to commercial size units
and will not be repeated here.
2.2.1.2 Firetube Boilers - Capacities up to 25 MBtu'/hr (7.3 MM)
In a firetube boiler the products of combustion are directed through tubes which are sub-
merged in water. The tubes are normally straight and may be horizontal, inclined, or vertical,
with one or more passes. They are held in place inside the boiler shell by tubesheets at either
end of the bundle. Except for small domestic units, many of which are vertical, most units have
horizontal tubes. Firetube boilers have larger water storage capacity than cast iron boilers which
effectively dampens wide fluctuations in steam demand. In addition, they are more efficient than
simple shell boilers because heat is absorbed by the tubes as well as the shell.
Firetube boilers are used where steam demands are relatively small. The principal uses
of firetube boilers are for heating systems or industrial process steam. The heating boilers are
restricted to 15 psig (0.1 MPa) steam pressure or 30 psig (0.2 MPa) hot water pressure. Power
boilers in this class are usually limited to about 350 psig (2.4 MPa) steam pressure.
Most Scotch and firebox firetube boilers are now constructed with an internal furnace, which
is substantially surrounded by water-cooled surfaces. The internal furnace in a firebox firetube
boiler, described below, is surrounded by water-cooled surfaces except at the bottom.
There are many firetube boilers in operation which have been constructed with an external
furnace, usually of brick construction. The brickset type is not very suitable if water scale or
silt is to be expected.
The following types of firetube boilers are ;now commercially available or installed in exist-
ing buildings:
2-25
-------
0 Horizontal return tubular -Capacities up to 22 MBtu/hr (6.4 MW)
The horizontal return tubular (HRT) is a two-pass (one under the shell and one through the
tubes) power boiler. It was formerly the most popular type of firetube boiler. At present, approxi-
mately 10 percent of all commercial-size boilers are HRT. Water circulation tends to be sluggish,
contributing to a relatively low boiler efficiency.
Firebox - Capacities up to 20 MBtu/hr (5.8 MW)
The two major types of firebox boilers are the short and the compact, the former employing
two passes and the latter three. These units are constructed with an internal, steel encased,
water-jacketed firebox. Since the furnace size is limited, proper matching of burner flame length
and combustion volume is a critical factor. The greatest advantage of this type of boiler is its
efficiency and the minimum floor space required for installation.
Vertical -Capacities up to 3.5 MBtu/hr (1.02 MW)
Vertical boilers are normally single-pass boilers and are used for power boiler applications
or as commercial hot water and steam boilers. Power boilers are designed for up to 3.5 MBtu/hr
(1.02 MW) and high head room is required for these units. They have a limited steam release sur-
face, are restricted to a maximum pressure of 150 psig (1.0 MPa) and, since the upper ends of the
tubes are steam-cooled, require a slow startup to prevent tube damage by overheating.
t Scotch - Capacities up to 25 MBtu/hr (7.3 MW)
This type of boiler was developed 30 years ago based on the Scotch marine boiler and has
since become one of the most popular firetube boilers. The trend toward increased use of this
boiler type continues.
Figure 2-6 shows a typical oil-fired Scotch boiler. Its design emphasizes compactness and
limited head room. It is capable of gas, oil, or combination firing. Inside shell diameters are
normally limited to about 95 inches (2.4 m) and operating pressures to 350 psig (2.4 MPa) although
some units are designed for 600 psig (4.1 MPa).
Characteristics of a typical Scotch boiler are given in Table 2-9. They can have either
two, three, or four passes. The burner flame is contained in an elongated, water-cooled combustion
chamber which also acts as the first pass. This characteristic is unique to this type of boiler.
The rear wall of the furnace is either refractory-lined ("dry-back") or water-cooled ("wet-back")
in the later versions. Due to the relatively small diameter firetubes and concomitant large pres-
sure drop, three and four pass units generally require provision for mechanical draft.
2-26
-------
Flue
Firetubes, showing flue gas
directions
Blower
Figure 2-6. Four-pass scotch firetube boiler (courtesy of the
Cleaver Brooks Company).
2-27
-------
TABLE 2-9. CHARACTERISTICS OF A TYPICAL 3.2 MBtu/hr (0.94 MW)
PACKAGED SCOTCH FIRETUBE BOILER
Excess air: 5% to 40%
Heat exchange area: 400 ft2 (37 m2)
Combustion chamber dimensions: D = 1 to 2 feet (0.3 - 0.6 m)
L = 10 to 12 feet (3 - 3.6 m)
Firetube diameter: 1 to 2 inches (25 - 50 mm)
Common operating mode: Hi/Low fire modulation
Ignition system: Spark
Draft system: Forced
Maximum gas side heat exchanger temperature 900DF (482°C)
Overall steady state efficiency: 80%
Exit flue gas temperature: 450°F to 650°F (232°C - 343°C)
Natural gas fuel consumption: 3350 ft3/hr (0.03 m3/sec)
Light fuel oil consumption: 24 gal/hr (25 cm3/sec)
2-28
-------
Most Scotch boilers are designed with as little heating surface as is practical for firetube
boiler construction. Typically, 5 square feet of heating surfaces are required to produce 1 boiler
horsepower. Improved heat transfer Ms obtained by one or more of the following design features:
High flue gas velocities throughout the boiler due to the use of a steadily diminishing
number of constant-diameter firetubes in each subsequent pass
a Turbulators in the firetubes
e Improved water-side circulation by internal baffling and other means
Due to their compact design and multiple tube passes, the packaged Scotch boiler is more
difficult to clean than other firetube boilers.
All firetube boilers must be designed in accordance with the ASME Power Boiler Code (Section I)
or Heating Boiler Code (Section IV). They are rated by the American Boiler Manufacturers Associa-
tion.
2.2.1.3 Atomization Methods
As was the case with domestic sized furnaces, a variety of burner types are used for com-
mercial sized boiler applications. The main distinguishing feature is the method of atomizing the
oil. Four techniques can be used:
Air atomization
a Steam atomization
e Mechanical atomization (oil pressure atomization)
e Rotary cup burners
Most air atomizing oil burners used in commercial sized boilers are high pressure units.
These are similar to the low pressure mechanical burners used in domestic furnaces, the oil flow-
ing through a central tube in the nozzle and the atomizing air flowing in an annular tube around
the oil passageway. The difference is that the mixing of oil and atomizing air is more violent
than in the mechanical burners in order to atomize the oil into fine drops. This mixing can take
place either outside or inside the nozzle, though a lower air pressure can be used if the former
method is chosen.
Steam atomized burners are identical to high pressure air atomized burners except that steam
is used in place of the pressurized air. The use of steam is often desirable since a supply of
2-29
-------
high pressure steam is readily available at the boiler. Steam atomization is limited to use 1n
boilers which attain steam pressures in excess of 100 psi (690 kPa).
Mechanical (pressure) atomizing oil burners have been discussed previously (see Section
2.1.1.1). Rotary cup burners utilize a rapidly rotating cup to which oil is supplied. The rotary
motion of the cup causes the oil to form a thin film on its sides. Centrifugal force pushes the
oil to the lip of the cup at which point it is flung off in the form of tiny droplets. The mean
droplet size decreases as the rotary speed is increased. Older rotary burners are being phased out
at present, primarily because of their high maintenance requirements (see Section 4.1.2), although
some new designs may be able to operate cleanly with no more than normal maintenance.
All four types of atomization can be used with fuels ranging from distillate to No. 6 resid-
ual oil as long as proper oil preheat is applied to reduce the viscosity of the heavier oils before
they enter the nozzle.
2.2.1.4 Tuning
The tuning of commercial units is no less important than it is for domestic burners, and the
procedures and effects are similar in both cases. In addition, oil preheat must be adjusted in com-
mercial units which use heavy oil.
2.2.1.5 Cycling and Modulation
The operation of a boiler differs from that of a warm air furnace in that control of water
temperature and steam pressure can be achieved by modulation of burner output as well as by cycling
the burner in an on/off mode. The modulation method is preferable, especially in the case of resid-
ual oil firing. Data from Reference 2-16 indicates that particulate emissions from some boilers
firing No. 6 oil can decrease by a factor of two as the load is reduced from 80 percent of rated
conditions to normal low-load firing.
2.2.2 Emissions from Commercial Oil Burners
Particulate emissions from oil-fired commercial boilers vary over a wide range due to the
variety of fuels burned. Barrett, et al. (Reference 2-15) have related carbon residue, carbon con-
tent, viscosity, and API gravity to particulate emissions. Since the first three properties are
all related to the API gravity, particulate emissions are too. The net result is that as API grav-
ity decreases, particulate emission increases. Average emission factors for the sample of commer-
cial boilers tested in Reference 2-15 are given as:
2-30
-------
FueVOiV
Grade
No. 2
No. 4
No. 5
No. 6
LSR*
API
, Gravity
35
22
19
16
23
Filterable
Parti cul ate
Ib/MBtu (ng/J)
0.011 (4.7)
0.047 (20.2)
0.087 (37.4)
0.25 (107.5)
0.087 (37.4)
Larger commercial boilers than those tested in Reference 2-15 will have emission characteristics
similar to industrial size boilers.
The effect of boiler load on participate emissions is shown in Figure 2-7. In general, in-
creasing load does not significantly affect filterable particulates except for firing of No. 6
fuel oil. Increases in particulate emissions for four boilers fired on No. 6 ranged from 0.001 -
0.008 Ib/MBtu (0.43 - 3.44 ng/J) for each 1 percent increase in load.
Barrett, et.al. (Reference 2-15) have suggested appropriate emission factors based on emis-
sions data from 27 boilers in steady operation at 80 percent load and 12 percent C02 in the flue
gas. These numbers listed below represent a second degree curve fit to a plot of emission data
versus API gravity.
Fuel
Grade
No. 2
No. 4
No. 5
No. 6
No. 6
(1% S)
Suggested
Emission
Factor
Ib/MBtu (ng/J)
0.0086 (3.7)
0.093 (40.0)
0.18 (77.4)
0.24 (103.2)
0.08 (34.4)
Low sulfur residual typical of that marketed on the East Coast in 1971-72. Most heavily impacted
regions already require residuals to contain no more than 0.3 or 0.5 percent S.
2-31
-------
100
O No. 2
X No. 4
-}- No. 5
No. 6
20
40 60
Boiler load, percent
80
Figure 2-7. Effect of load on filterable participate emissions
from commercial boilers (participate emissions by
Battelle Modified EPA Procedure, Reference 2-15).
2-32
-------
They represent an improvement over the customary method of assigning one emission factor to resid-
ual oils as a whole in that the.considerable effect of API gravity on emission is included.
2.2.3 Population Distribution
In this section, the population of commercial boilers are characterized with respect to
boiler type, fuel capability, and burner type. Recent and estimated future trends in boiler and
burner type and fuel capability are also presented.
2.2.3.1 Population Distribution by Boiler Type
Table 2-1.0 shows the population distribution by boiler type for all commercial boilers cur-
rently in use. A review of the design features of the boiler types listed is given in Section
2.1.1.2, 2.2.1.2, and 2.4.1.1.
It is evident from Table 2-10 that watertube boilers comprise only a small fraction of the
total commercial boiler population. A large fraction of the smaller size units are cast iron
boilers, whereas most of the larger commercial ones are firetube boilers.
The sales data from 1975 (Reference 2-28) show that a large percentage of the commercial
size boilers sold in that year were Scotch firetube units. This would indicate that Scotch boil-
ers account for a larger fraction of the 1975 population than indicated in Table 2-10. This
growing popularity of Scotch boilers is indicated in Table 2-11 which shows past and estimated
future trends in the number of commercial boilers -in' service for the years noted. Thus, pack-
aged Scotch boilers appear to have replaced horizontal return tubular boilers in the two larger
size ranges. Other than this historical shift, no major changes have occurred since the 1950's
nor are any predicted for the next 15 years.
2.2.3.2 Population Distribution by Fuel Capability
Table 2-12 shows the population distribution of commercial sized units in terms of fuel
capabilities. The majority of the units are fueled with oil or gas, coal being used in only
1 or 2 percent of the systems. Less than 10 percent of the units have multiple fuel (oil/gas)
capability, though 1975 sales data indicates a large number of oil/gas units sold in that year
(Reference 2-28).
In the oil-fired units listed, the fraction of residual oil burned increases from 5 to
50 percent as the size of the unit increases from 0.3 - 10 MBtu/hr (88 kW - 2.93 MW). The rela-
tive amounts of the various types of oil used is important because particulate emissions are a
strong function of fuel characteristics.
2-33
-------
TABLE 2-10. POPULATION DISTRIBUTION OF COMMERCIAL BOILERS BY BOILER TYPE (1972)*
Rated Capacity MBtu/hr
Size Range (HW)
WATER TUBE
Coil
Firebox
Other
FIRE TUBE
Packaged scotch
Firebox
Vertical
Horizontal Return Tubular (HRT)
Misc. (locomotive type, etc.)
CAST IRON
MISC. (TUBELESS, ETC.)
TOTAL
0.33 - 1.7
(0.096 - 0.50)
2
1
3
15
25
1
5
2
45
1
100%
1.7 - 3.3
(0.50 - 0.96)
3
1
4
20
25
0.5
10
3
33
0.5
100%
3.3 - 10.0
(0.96 - 2.93)
2
1
2
30
30
nil
15
. 5
15
ml
100%
Source: Reference 2-20
2-34
-------
*
oo
UJ
D_
O
CO
CJ
cr:
UJ
o
CJ>
LU
a:
o
LU
CM
LU
1 '
"*
IO
00 01
oo o
1 1
t^ o
LO
o
~s
I-N LO
r-^ 0
i i
oo vo
00 01
o
CD
O
s-
JZ
5 2:
s*~"
r- 0}
U Ol
<~* (O
o
01
O N
0) -t-
+-> 00
to
o;
O LOLO^- OO( I I LOi O
CTt . , «d" OO 'I 'I 'r- i O
' ' . ' C C C ' i
O «* «* CM i LO r < .CM O r O
r^ >=i- oo -i- r o
C r
CD OO OO LO r O CM LO LO UO ^ O
LO CM 00 CM O
O r OO CM CM O LO O CO r r O
OO -i- CM ^O -I ! O
- c c =
O «l-c r-~ ID OO CM r t IO r O
CT1 CM CM !- OO T- O
C C r
O LOOOCO CM r- OO i r- O r O
r^ CM i -i- * o
O OOLOCO «d- .LO LO CM i LO CM O
O I COLO i OOOLOLOOOO O
OO -r- CM LO O
C r
O LOi LO OOVOi CMOi O
Ol i i CM "- ^J" CD
C ' i ~
O 00000 r CO 10 r- 00 LO i O
O CMUOCO Oi UDi LOOCM O
LO i ' LO O
O r COLO i UD LO LO CO O OO O
OO - -r- «3 O
C C "
ac "^
s- cu
1 1
CO) '-^
J= 3 -r- 1
O -P +J UJ
-M CU O
O S- E "
t/1 i (J OO
ra O UJ
UJ T3 r -P i _!
CQ x uJoiXiac^-'ZCQ
Z3 O CQO1OOO O=>
1 J3S-^><8J3-r-N.Ci;|
a:'r-s-J=io&-s-&.t/i i
UJO -(-> LUlO-i-010^: I CJ cS
K O i ' O O? O- U- i* ZC " OO OO ]
^ Lu 0 S °-
o
CM
1
CM
01
o
1
Ol
M-
oi
o
s-
3
o
*<"
2-35
-------
*
c\T
en
I
g
ca
to
UJ
_)
S
m
CO
HH
o:
D.
£
CM
UJ
CQ
«C
°s"
JB CM
, 1
«g
"-in
"" o
o
1
«£
°s
s-
e
J^* ._
CD-
1?
r- > l/l to
0 !-
. Z lf>
« -a
O) CO
4-> , as 21
£^ > l->
r "O » »r- O
10 tn z: n: i
r- 0) «t
_i a a: t
l-H O
0 h-
o
CM
i
CM
01
o
C
ai
ai
f-
tu
(LI
o
2-36
-------
Table 2-13 shows the past and estimated future trend in commercial boiler fuel capability.
In response to recent gas and oil shortages, many units which now burn only gas are being replaced
by dual fuel (oil and gas) systems, and some of these which now use only oil are being replaced
by coal-fired boilers. This trend is projected for all commercial boiler sizes, although in vary-
ing degrees. Sales data for 1975 show that almost one-third of all units sold have multiple fuel
capabilities (Reference 2-29).
2.2.3.3 Population Distribution by Burner Type
Table 2-14 gives the population distribution of .commercial sized units in terms of burner
type. Mechanical atomization predominates in the smaller size boilers, while air atomization and
rotary type burners comprise a large fraction of the bigger units.
Classification of oil-fired units in terms of burner type is particularly important in any
study dealing with particulate emissions. The size of the atomized oil droplets, which is a func-
tion of the type of atomization employed, has a direct effect on the amount of particulate emis-
sion, as discussed previously.
Table 2-15 shows past and estimated future trends in burner types in terms of the number of
units of a particular type in service for a given year. The past trend away from rotary burners
is projected to continue until they disappear entirely in the near future. In large units the
rotary burners have been replaced mainly by air atomizing types whereas in the smaller boilers and
furnaces they have been replaced by both air and mechanical aton.izers.
2.3 INDUSTRIAL SYSTEMS - 12.5 - 250 MBtu/hr (3.7 - 73.25 MW)
Industrial size heating units are composed entirely of firetube and watertube boilers.
Firetube boilers were discussed in Section 2.2.1.2, and watertube boilers are described below.
2.3.1 Equipment Description
2.3.1.1 Design Features of Watertube Boilers
Above a capacity of 30 MBtu/hr (8.8 MW), industrial boilers are almost exclusively of the
watertube type. They also dominate the market where design pressure exceeds 150 psig (1.0 MPa).
As the name implies, watertube boilers are designed to flow water through the heat transfer tubes,
instead of combustion products as in the firetube design. Because of the smaller diameter of the
pressurized components and the ability of tubes to accommodate expansion, watertube boilers are
2-37
-------
CO
S
S
01
LU
O
CO
O
a*
LlJ
oo
o
LLJ
OH
O
LU
CO
= s
r^ 2^.
^
>>
r- CU
U cn
O. C/1
O
(71
O
"
0
in
O
CO
o
CO
o
0
in
o
CO
0
CO
o
g
o
CO
J_
1
1
14
CO
S
LU
^
LL.
o in o o LO o
i OJ tO CD
CD CD LO O LO CD
CO CO CO O
"""as
o in o o m CD
3- OJ OJ O
o in o r- LO o
i co T- o
O O LO O LO O
LO CO LO O CM O
OJ CO CO O
""a*
o o LO o in o
CO CO r c O
c3 o in r LO o
i i r-^ ! CD
C r
o in o o in o
OJ oj i *d- o
r
o m LO co oj o
CO «3* I O
o o o in in o
3- CO OJ O
O O LO r in O
OJ r 10 !- O
C i
to
1 "cu
3
M-
cu
(O
^ " p
o cu
4J
IS
cn 4->
-a -i
c
i O
>! >> i IO 3 CU
i i c cn 4- d
CEO 0
0 ° ^ °° J 75
I W n3 r l/> ' 1
r- CO IO O
O O O i O
r^. co co *i o
CD O O O O
*3- 10 in i o
X ""**
O CD O O CD
co r^ in oj CD
o o o o o
oj co in co o
0 ^ r r 0
O !- <- -r- O
i C C C i
O O LO LO O
r~ co oj o
o "S o o o
LO LO <* r O
___ "~
o o o o o
^" to ^J" OJ CD
IO
_c +J
cu id
i- CU
a. J=
JV
C 0.
' ^
LO to
0 0
OJ Z. Z
+J oB
o ^:
? cn in
» 'r^
CU "~ CD
4J o3 Z
IO
r- * >, I
1 > I-H
r- T3 IO O
4-> T- O CU
o) to z. 3; _i
r- CU =C
i a cc (
t < O
0 ' 1
CD
oo
1
OJ
cu
U
c
cu
i-
£
cu
c£.
cu
o
S-
3
o
*
2-38
-------
:M
i
en
Q-
>-
CQ
>-
ca
OO
C£.
LU
CD
CQ
CJ
o:
o
z.'
CD
II
I
CO
II
t
It
Q
O
0.
o
D_
CM
CO
' vo
«2
'«f
i ,
"" CD
CD
' CD
,
1 '
sl
S-
.£=
33"
-PS:
03
>,
i- CO
O CD
<0 C
O- fO
(O C£
O
CO
a N
-------
o;
LU
a;
uu
o
CO
eC
i<
O
CO
a;
f ^
°s
°.CM'
, i
^01
^S
co^
co'°
1 '
^ s
~ o
-
o
r-. in
r-^ 0
1 1
CO <£>
CO CT1
. 0
O
O
i.
.c
3 3
CO-
>^
r- 0)
O Cn
a c
O- to
e i
tt)
O N
1/5
n3
0
cn
o
o
in
CD
CO
CD
cn
o
r~
o
in
o
CO
o
cn
o
r"-.
CD
in
CD
CO
LU
a:
CO
_j
i i
o
O CD . CD
UD -3- M- CD
in o in o
in «d* o
s«
CD CD O CD
CO ^ CO O
000 0
oj in co CD
O CD i O
CO f"--. -i , O
o in in o
CO UD O
&a
o o o CD
CM IO CM O
in in o o
r in co o
in in r co
r co !- o
c *~"
in in i o
i oo <- o
C i
69
in in CD o
r- t^ r 0
o o o o
i r^ CM o
cn
r-
N
cn 'E
r* | '
N (d
i
E OJ
O S-
+* 3 >i
(O !/) &
S- 0) -4-> _1
r- S- 0 *£.
< Q. o: H-
O
1
O
CM
1
CM
CO
O
c
CO
01
M-
o>
ce:
CO
(J
S-
0
00
*
2-40
-------
better able to contain the high pressure than are firetube units, and, hence, they are an inher-
ently safer design. Efficiencies attained in watertube boilers are about 80 percent without heat
recovery.
Watertube boilers can be either field-erected or packaged. Common sizes for the former type
fall between 50 and 500 MBtu/hr (14.6 and, 146 MW) while the capacity range for the latter is 10 to
350 MBtu/hr (2.9 - 102 MW).
Since the industrial-sized watertube packaged boiler was first introduced in the early 1940's
they have become very popular. In the period of 1930 - 1950, almost 95 percent of the units between
10 and 100 MBtu/hr (2.9 - 29 MW) were field-erected. However, it is anticipated that by 1990 about
99 percent of this class will be packaged. Similarly, until 1950 all of,the watertube boilers in
the range of 100 to 500 MBtu/hr (29 - 146 MW) were field-erected. Forecasts indicate that by 1990
about 90 percent of the sizes up to 250 MBtu/hr (73.25 MW) will be packaged.
-Field-erected boilers are usually balanced draft* and, therefore, require both forced draft
and induced draft fans. Field-erected boilers are commonly fired with coal, gas, oil, and/or waste
fuels. Many,such boilers exist today, but very few new applications have capacities lower than
200 MBtu/hr (58.6 MW), except for solid fired units. The packaged boiler's domination of the oil-
and gas-fired market below 200 MBtu/hr (58.6 MW) is due to their low capital cost.
Packaged watertube boilers are used for gas and oil-firing applications. They are not used
for coal-firing because they have much smaller furnace volumes than are permissible. They are de-
signed to be rail-shipped as a complete, single package with minimum fieldwork. The furnaces are
also designed to operate under positive pressure versus the balanced or slightly negative pressure
found in larger field erected units.
The two general types of watertube boilers are horizontal, straight and bent tube boilers.
Straight watertube boilers are no longer manufactured, having been completely supplanted by fire-
tube boilers in the smaller sizes and bent watertube boilers in the intermediate sizes. There are,
however, a large number of straight, tube boilers still in operation.
Straight Tube Boilers
The straight tube boiler (Figure 2-8) owed its early popularity to its low draft loss, good
tube visibility for inspection and cleaning, design that facilitated tube replacement, and low
Balanced draft refers to a boiler where static pressure at a specified location is equal to atmos-
pheric. The top of these boilers is at a slightly subatmospheric pressure.
2-41
-------
STEAM SOOT BLOWER SAFETY-VALVE
OUTLET CONNECTION CONNECTION
Sectional Header, Long Drum
Figure 2-8. Horizontal straight tube boiler
(courtesy, the Babcock & Wilcox Co.)
2-42
-------
headroom requirements. However, it is often subject to leaks around the handholes, considerable
labor is required to open a sectional header for inspection, and it has poor water distribution
and low circulation rates.,
These boilers are normally baffled to create two or three flue gas passes across the water-
tubes. The tubes are grouped into sections and expanded at the ends into headers. The headers
are connected to the drum by downcomers for supplying water to the tubes and by risers for dis-
charging water and steam from the tubes. The "horizontal" tubes are inclined at an angle of
5° to 15° for natural circulation of the water. The flue gas is normally moved by a forced cir-
culation fan..
Horizontal straight tube boilers can be subdivided by type of header and/or type of drum.
One variation of this type of boiler is a straight watertube boiler built with a steel encased,
brfcklined firebox with water cooled walls. The water and steam are contained in an enclosure
above the crownsheet. It is designed for 1 to 18 MBtu/hr (0.293 - 5.3 MW) and 15 psig to 250 psig
pressure (0.1 - 1.7 MPa).
Horizontal Bent Tube Boilers
Horzontal bent tube boilers, the other major type of watertube boiler, are classified by the
number of drums, headroom, and tube configuration, the latter of which is the most important dis-
tinguishing factor. In the boiler each tube is connected to the steam drum and to at least one
lower drum or header (see Figures 2-9 and 2-TO). The tubes enter a drum radially and are designed
to allow for the anticipated expansion. The furance is of the waterwall type backed up with re-
fractory. The tubes, which form the furnace waterwall, are an integral part of the boiler. The
steam drum contains internals, such as separators and cyclones, to facilitate steam-water separa-
tion. Bent tube boilers have the advantage of great design flexibility, making them readily adapt-
able to space limitations by either extending their length or, more practically, the wdith. They
provide good accessibility for inspection and cleaning, which normally entails mechanical cleaning
from inside the steam drum.
Most packaged units are oil-, gas-, or combination-fired, although there are some coal-fired
packaged units. The furnace wall cooling tubes are usually oriented vertically. Vertical furnace
tubes are particularly desirable in coal-firing since they are less susceptible to slag adherence.
Heat recovery from the flue gas is practical and often desirable with bent tube boilers,
particularly in the intermediate and large sizes. The recovery unit may consist of an economizer
which preheats feedwater, an air preheater to heat combustion air, or both.
2-43
-------
A-type has two small lower drums
or headers. Upper drum is larger
to permit separation of water and
steam. Most steam production
occurs in center furnace-wall
tubes entering drum.
D-type allows much
flexibility. Here
the more active
steaming risers enter
drum near water line.
Burners may be lo-
cated in end walls or
between tubes in
buckle of the D, right
angles to drum.
0-type is also a com-
pact steamer. Trans-
portation limits
height of furnace, so,
for equal capacity,
longer boiler is
often required.
Floors of D and 0
types are generally
tile-covered. ;
Figure 2-9. Types of bent tube packaged watertube boilers.
2-44
-------
Figure 2-10. D-type bent tube configuration (courtesy of the
Cleaver Brooks Company).
2-45
-------
The larger industrial-size, field-erected watertube boilers, that is, larger than 25 MBtu/
hr (7.3 MW), are identical to utility boilers used for generating electrical power. These multi-
burner facilities are available in the tangentially-fired design as well as the wall-fired type
found in the smaller units.
2.3.1.2 Atomization
The type of atomization employed in a given unit has been shown to affect both the total
amount of particulates emitted and their size distribution. McGary and Gregory (Reference 2-19)
compared particulate data for steam, air, and mechanical atomization techniques from the effluent
of three large industrial boilers. They showed that air atomization produced an order of magnitude
less particulate than the mechanical approach (Table 2-16). In addition, analysis of the particu-
late for combustibles showed that the lower emitters also produced particulate with proportionally
less combustible content. Both these results indicate that air atomization leads to more complete
combustion than steam, which is already much better than mechanical atomization.
Further evidence of the superiority of air to steam atomization is found in Reference 2-22
where a single boiler fueled with No. 5 oil was alternately fired using both types of atomization.
Air atomization produced 32 percent less filterable particulates than did steam (total particulates
were reduced by 16 percent.
2.3.1.3 Soot Blowing
In large boilers which see continuous operation, it is necessary to periodically remove the
soot which accumulates on the heat exchanger tubes. This soot layer effectively decreases the heat
transfer to the working fluid and, consequently, the overall boiler efficiency. In addition, it
can cause "hot spots" or corrosion which eventually will cause the boiler tubes to fail. Nearly all
industrial watertube boilers that produce more than about 30 MBtu/hr heat input (8.8 MW) are equipped
with soot blowers, and occassionally units that are one-half this size also have them. Some fire-
tube boilers that burn residual oil use soot blowers, too, but this practice is not common. One
important reason is that firetube boilers generally do not produce the high pressure steam needed
by the blowers.
The soot blowing operation is accomplished by activating a high pressure nozzle which travels
the length of the boiler and directs a high pressure stream of air or steam onto the boiler tubes,
thus dislodging the soot. The operation generally consumes less than 1 percent of the boiler's
thermal output and lasts for 2-5 minutes, depending upon the size of the boiler, its design, the
fuel, and the soot blowing procedure used. Frequently a boiler operator will clean one section
2-46
-------
*
LU
CL.
o;
LjJ
M
UJ
O
CM
CO
S-
O 00 r
*^" r*1*
r F*»t \D CTi 00 LO
OCOCMCTli IOOCOO
OO * VO CM CM r i OO
10 0
OO CM CF> CM if) LO
*
O^ t.p t-O CO CO CF% 4iO OO 1^3
VD i CM OO CO CM «*
cvj co
(O
s- . -
^~ t ^
4-* t/1 '5*$
3 to CU
s: cu 4J ' .«
4-> CU
^^ tHJ^ -[-*
CD en >
cu > ia. -a. ia. to N 3. CO CO Q.
+J O) -r- CD O
'3 T i co co co co E
2: Q. (0 -r-
~-- E E V V V V
(OS- co
T3tO OCUCUCUCUCU
O if_ ro- ' !-> S- S- S- S- E
O 3 O O
CO
O)
"^
o
en
en
c
'43
(O
J ^
r
c
3
CU
'M
CU
Q.
CO
to
g
'i
II
cr
E
s_
JC
*^"^
s:
o
2
E
o
s-
M-
T3
CU
(O
^
3
o
p~>
(O
o
to
cu
3
i
-o
O)
IM
^
to
£H
0
(O
cr>
CM
CU
^
/n
CU
CU
"
CU
U
_ *"
o
co
X
2-47
-------
of his boiler with .a 2-minute blow and then the other section about 1 hour later with another 2-
minute blow. Another approach is to use frequent 10 - 20 second puffs. Both air and steam can
be used, and they are equally effective (when measured by effectiveness per unit energy input).
The choice depends upon the economics of the situation at the facility where the boiler is located.
If high pressure steam is readily available, as in a utility boiler, this medium will be used. If
however, the boiler is relatively small, so that an ample steam supply is not available, and if
the plant already has air compressors to supply "shop air" or processes which need high pressure
air, then the boiler operator will probably opt for an air soot blower system. In any case, the
system is always designed to effectively remove the accumulated deposits from the heat transfer sur-
faces. Therefore, there is no reason to suspect that differences in soot blowing equipment or con-
figurations will affect the mass of particulate emitted during the cleaning cycle.*
Small oil-fired boilers may have only two soot-blowing heads, each servicing half the boiler
heat exchanger section. When one of these blowers is activated, it cleans a large fraction of the
total surface being blown and, consequently, the smoke emanating from the stack may have an opacity
of 80 to 100 percent (Reference 2-21). Large boilers can have as many as 25 soot blowing heads which
operate sequentially. In such a system, only a small fraction of the total blown surface is being
cleaned at a given time, and, consequently, a less severe change in Ringelmann number is seen at the
stack. In very large utility boilers, the soot-blowing operation may take place cyclically, with
operation of the last blower in the sequence being followed by the first, again.
2.3.1.4 Emissions
There appears to be little available data on filterable particulate emission levels from
industrial size boilers. However, emission levels have been measured for some of the smaller
boilers in this category operating at baseline conditions (Reference 2-22). These data are shown
in Figure 2-11. Industrial units in the size range 15 - 120 MBtu/hr (4.4 - 35.1 MW) fueled with
distillate have emissions comparable to that of commercial units, that is 0.01 - 0.03 Ib/MBtu
(4.3 - 12.9 ng/J). As boiler capacity increases, this level appears to decrease:
A similar trend is seen in the case of those units tested which burned residual oil, for
the mass emissions decreased from 0.1 to 0.04 Ib/MBtu (43 - 17.2 ng/J) as boiler capacity increased
from 50 to 130 MBtu/hr (14.7 - 38.1'HW).
Since there is not data to substantiate this statement several boilers should be tested to deter-
mine how total emissions and particulate size distributions vary with time, both during and be-
tween soot blowing operations, and with soot blowing equipment and frequency.
2-48
-------
0 10
3
+J
CO
s:
O
to
dJ
+-J
o
i.
a
"(0
£
Om -
or
/r\
^
(T|
1 51
^^^
@
_
vi/
CD
Arji
£2LJ
O Steam
Q - Air
/\ - Mechanical
(Numbers indicate oil
©
grad
0
C2)
V^
e fired)
, '
20 40 60 80 100
Boiler load (15 steam/hr x. 10"3)
120
140
Figure 2-11. Particulate Emissions vs. Boiler Load
2-49
-------
The larger Industrial boilers, which are similar to the smaller size utility boilers, ex-
hibit lower emissions than the smaller industrial boilers for two primary reasons. Due to the
large quantity of fuel consumed in these units, any reduction in efficiency due to substandard
combustion becomes very expensive. Therefore, the larger units often receive better maintenance
and are operated under more carefully controlled conditions than the smaller units. Secondly,
there is a certain degree of custom engineering involved in the design of utility and large in-
dustrial boilers, including expensive control systems. This often results in reduced emissions.
Therefore, particulate emission concentrations from large industrial''boilers are comparable to
those from utility boilers and significantly lower than those from commercial and smaller indus-
trial units.
EPA has recently revised the emission factors for particulates from oil-fired boilers and
furnaces (Reference 2-26). These new factors recognize the effect of fuel on particulate emissions
by specifying a separate factor for each residual fuel grade and, further, by relating emissions
to sulfur content in No. 6 oil. Table 2-17 presents the new factors and also shows the old ones
for comparison. The new emission factors are averages for each fuel grade and do not distinguish
among different size units. Thus, they do not show that emissions decrease as the boiler size
increases (see Reference 2-23).
2.3.2 Population Distribution
This subsection presents the population distributions of industrial size oil-fired firetube
and watertube boilers. Separate tabulations are given for population by boiler type, fuel capa-
bility, and burner type.
2.3.2.1 Population Distribution by Boiler Type
Table 2-18 shows the population of industrial boilers classified in terms of boiler type.
Firetube boilers account for more than three-quarters of the population in the size range below
16 MBtu/hr (4.7 MW), while nearly all boilers over 100 MBtu/hr (29.3 MW) are of the watertube
type.
As shown in Table 2-19, firetube boilers have become increasingly popular in the smallest
size range, and this trend is expected to continue. There is also an increasing preference for
packaged, as opposed to field-erected, watertube boilers in the large size categories.
2-50
-------
*
CO
UJ
CO
OO
13
CQ
UJ
o:
I
1
II
o
o
o
Si
CO
CO
CM
UJ
_1
CO
1-
4-
S-
O
u
a
u_
c
o
t/)
t/>
^
uS
-a
o
*t~
<1>
r
I
-3
en
c:
^
CQ
JO
t
1
C
3
CO
2:
;2
"co
3
4J
C
D.
3
CT
UJ
*
K
in o co co
O ID in CM
CO «* «3 CM
^^----N
X
*
r r-. ro ro
t-, o LO un
Or- r- 0
o o o o
, ~^*r^
ID
CO
51- CM
I-. O CM CO CO 1
r-I ID o co x
CM CM
co
.CO
CM
o
CM
O
o
CO ^" 1^ 1"^ -i-
r- r- * 10
o o o o co i
C3 O O O X
ID
o
o
Ol (U
rtj (O «=j- to ^o
!^ ^ o o o i
U)
'r~\ **"
rr- -
o
^
4->
*rr
. .
i (O lO ZD 13 ^3
C O S- + + *t* >}
OJ S- 4J -I-1
T3 0) > Q) (U > > i
(A g TD O O - O !-
O) O C JO -Q JO -P
U)
flj
r-
^~
3
S-
O
(U
0
-o
o
CO
(ft
to
3
cr
S.
O
+J
u
4-
T3
CO
trt
>
s-
0)
^
in
CD
.
"+? o
J=
en X
o> -a
S c
i 0-
H O.
i (U
O) S- (U U
O 3 "r~ fO
g M- 4-> 4-
(1) i '1-
S- 3 i C
OJ t/J -r- . O
(Ji! 4-> -r-
m 4-> 3 tn
Cd C S-
Qj en aJ
u c >
S- !- C
0) ' 0) T3 O
O O- 3 O
3 II "3 S-
O X O
CO CO UJ U.
* - 5 f
2-51
-------
TABLE 2-18. POPULATION DISTRIBUTION OF INDUSTRIAL BOILERS BY
BOILER TYPE (1972)*
Rated
Capacity
Size Range
WATER TUBE
Packaged
MBtu/hr
(MM)
Field erected
FIRE-TUBE
Packaged scotch
Firebox
Horizontal return tubular (HRT)
Misc. (locomotive type, etc.)
TOTAL
10-16
(2.9-4.7)
15
7
30
25
20
3
100%
17-100
(5.0-29.3)
55
24
10
10
1
nil
100%
101-250
(29.6-73.3)
25
75
~
100%
*Source: Reference 2-20.
2-52
-------
*
OO
a.
>-
o;
te
=3
a
00
o
Q
UJ
CM
UJ
CQ
=C
If? 8 S ! ! ! ! §
CO
o
to co
CM r~
I i
i 10
o
T~ CM
CO
0
o en
1 CM
i i
r~- O
to
r
to
T i
o
CM
f
i-
-E
^»^*~ *»
3 3
+-> S
§i
g. s s : s : i °
g ° § i : i i |
r~
S ° § .! ! ! ! S
a s- ^--^^ §
O O«=J- LOr * r O
(^. CO P"~ *CZ "« ^^
C C *"""*
o cocr> CVJr~"*CT §
to co . P 2
~
0 0 «=f ' ,- r- .- 10 t- 0
co en £ -^ 0
s ° ° s «: i. i 1.
O COr OOrr- O
r- * * : £ . °^
o eMto motoco o
m co . o
i -^
x_x O
S- 0)
.Q f*^~
3 >5
1 c
T- 0)
o en
a c
O. CO
re o:
(L)
-0 N
4J OO
1 -= = "
1 o -M *-*
n 4J 0) o
S o s- E
0 « ^0
dj ro o
UJ -O S- T3 4J r
CQ S Q) LU O) X C
^3 cn CQ en o o
j-y* O tt) 1 (_>*->- " "5
t L) (Q .|__ |jj (tJ *t O *r ^t
J CL. l-'-r (y, Ci- Li- ^- S- ^~
5 S C F
__J ___
C3
CM-
CM
CD
O
C
01
i.
£
S
-------
2.3.2.2 Population" Distribution by Fuel Capability
«
Table 2-20 shows the population distribution of industrial boilers in terms of fuel capa-
bility. Oil and gas are the primary fuels for the-smaller size units. However, coal is of almost
equal importance in the larger sizes.
Of the oil-fired units smaller than 16 MBtu/hr (4.7 MW), approximately 30 percent use dis-
tillate, No. 4, or unpneheated No. 5. Boilers in the larger size ranges, however, are fired almost
exclusively on prehea'ted residual.
Table 2-21 shows past and estimated future fueling trends. Until recently, coal experienced
a decrease in popularity, while at the same time, distillate use had been on the rise. Ever tight-
ening air pollution regulations caused both trends, but oil and gas shortages are expected to stimu-
late a shift back to coal in all size ranges.
The types of fuel used depends to a large extent on the geographic region in which the user
is located. Table 2-22 shows the national distribution of fuel usage for industrial sized boilers
(Reference 2-21). The data is in terms of percentages of tots! fuel consumed in a particular geo-
graphic region based on the conversion of all fuel consumption into units of "ton equivalent coal."*
A map showing the related geographic regions is presented in Figure 2-12.
The regions using the largest percentage of oil to fuel their industrial boilers are the
Atlantic Section (AT), where oil accounts for over 92 percent of the total fuel usage, and the
Rural North (RN), where oil accounts for 84 percent of this total. This reliance on oil, coupled
with the large amount of industry located in those areas, makes them a particularly large source
of oil-derived particulate emissions.
2.3.2.3 Population Distribution by Burner Type
As indicated in Table 2-23, all burner types are employed to a significant degree in small
sized boiler applications. In the larger boilers, where production of sufficient quantities of
high pressure atomizing air is a problem, steam atomizing burners predominate in spite of evidence
(Reference 2-19} that air atomization produces signifiantly lower particulate levels. Rotary type
The conversion was based on a heat content of 12,776 Btu/lb (29.7 MJ/kg) for coal, 19,570 Btu/lb
(45.5 MJ/kg) for distillate oil, 19,050 Btu/lb (44.3 MJ/kg) for residual oil, and 20,000 Btu/lb
(46.5 MJ/kg) for gas.
2-54
-------
TABLE. 2-20. POPULATION DISTRIBUTION OF INDUSTRIAL BOILERS BY FUEL CAPABILITY (1972)*
Rated Capacity MBtu/hr
1 Size Range (MW)
FUELS
Oil only
Gas only
Coal only
Oil & gas and gas & oil
Oil & coal and coal & oil
Gas & coal and coal & gas
Misc. fuels
(alone or with alternate fuels)
TOTAL
OIL
Distillate, No. 2
Res id
No. 4 and light No. 5 (no preheat)
Heavy No. 5 and No. 6 (preheated)
TOTAL OIL
*
Source: Reference 2-20
10-16- '
(2.9-4.7)
35
45
3
16
1
100%
10
(90)
20
70
100%
17-100
(5.0-29.3)
35
35
10
18
2 .
100%
2
(98)
2
96
100%
101-250
(29.6-73.3)
30
22
18
26
0.5
0.5
3
100%
2
(98)
nil
98
100%
251-500
(73.5-146)
22
22
22
23
3
3
5
100%
2
(98)
nil
98
100%
2-55
-------
K
UJ
£
s
CM
I
CM
(JO
g*
in t
i in
in co
CM r-»
*~*
^ .
in co
CMI-.
i i
r IO
O
"""CM
o cn
rCM
1 1
t-» O
in
10^
i-*
i i
CD cn
CM
^"*
j-
***%*"^
li
r- Ol
O CD
S.S
Id C£
O N
-
H- 1
§
0.
g
_J
UJ
=D
U-
r t o o o in in o
r- !- ID CM r CD
O O O O O O r O
CM CM CM CM i i -r- O
in in o in co CM i o
r i IO -i O
E r
in in o r- i . i o
C C C C i
i i oooinin o
r- ! in CO i CD
C C i
CM CM r CM O
r~*
oocooinincM o
CM CM CO i o
in in o i . i t o
Cn *r~ *r~ *r~ 'p~ CD
C C C C ,r
O «3- O O ID O
r "5j- "Si" O
r
o o in o in o
CO CO CO O
&5
o o o in in o
CO CO CO CD
r~
co o in i CM o
r i r^ ( o
C t
co ID o in ID o
r CO * O
'
o o in o in o
CO CO CO O
n
co o o in CM o
«* CM r- CM O
r-~
r>, in in r ro o
i !>.!- O
V)
"oi
£
i o cn e
O «3 oa a)
08 i
id 10 to
U) 0 O
id o o j=
TO +J
T3 T3 !-
a c c s
c: id id
id to t-
i i i O
^> t/i id id o
>>>>i idoosa)
r-?i-?C TOO 01- C
ceo o
O O 08 o3 08 r
r O Id
i w rd i i w 1/1 - I
r- Id O * "r- Id 'r- «C
OCDOOOtDS K-
f^i
2 § ^ § 8
X C r
LO
ir> a> i LO o
*>«-'x T O^ ^3
_«. C r^
i i -E ° °
r S' r- 0 0
i- O -r- O O
C r C 1 r
0 0~ r- 0 0
^ cn *r- cn o
X = r-
m iS r in o
2. - cn o
e~~ «. C ^~
^ | -T: g g**
C C- = r- -
i O r O O
i- O (- O O
C n-_ C r t
O O i O O
CM CO -r- CO 0
x c <
0 § ,- 0 0
r- Ci T- CO O
t~ f
r-\ ^&
i is in in o
r- " CO O
C O "
r O i O O
r- O !- O O
C C ^- t
O O* r O 0
CO P^. "r- h^ O
O O O O O
i CO r CO O
^-^ r^
CM co co in o
cn CM r-^ o
* ' r
in in o in o
CO CM f~ O
% -
.c -S
£ S
O. -C
t 1
r- -a Id O
+J -r- O
-------
TABLE 2-22. DISTRIBUTION OF FUEL USAGE IN INDUSTRIAL BOILERS
BY REGION PERCENT ON A "TONS EQUIVALENT COAL"
BASIS (Reference 2-21)
Total
Region
AT
GL
WS
cu
's' SE
RN
USA
Coal
5.7
24.5
0.9
22.1
29.2
10.9
11.9
Distillate
Oil
25.9
25.3
5.4
9.1
13.5
26.6
20.9
Residual
Oil
66.9
38.3
62.7
32.5
43.8
57.8
57.3
Gas
1.5
11.9
30.9
36.4
13.5
4.7
10.8
2-57
-------
-o
c
1
HI
Qi
I
CM
0)
13
cn
+J wo
*fO c/) t^*
I i co ra LLJ
et C3 :x o oo
2-58
-------
TABLE 2-23. POPULATION DISTRIBUTION OF INDUSTRIAL BOILERS BY BURNER TYPE (1972)"
Rated Capacity MBtu/hr
Size Range (MW)
OIL BURNERS
Air atomizing
Steam atomizing
Pressure or mechanical atomizing
Rotary
TOTAL OIL
10-16
(2.9-4.7)
40
20
10
30
100%
17-100
(5.0-29.3)
15
70
10
5
100%
101-250
(29.6-73.3)
5
85
10
100%
Source: Reference 2-20
2-59
-------
burners are expected to be replaced by air or steam atomizing types in the near future (see
Table 2-24).
A description of the various types of oil burners presently in use was given in Sections
2.1.1.1 and 2.2.1.3.
2.4 UTILITY - OVER 250 MBtu/hr (73.25 MW)
2.4.1 Equipment Description
Boilers in the utility size range are all of the watertube type, previously described in
Section 2.3.1.1. Utility boilers are classified by their firing type - i.e., the particular way
in which the oil is injected into the boiler. The three primary firing types for oil-fired units
are single wall (or front face) firing, in which a bank of burners mounted on a plane wall ejects
fuel in one direction only, opposed firing, in which two banks of burners eject fuel toward one
another, and tangential firing, in which the burners are located at the corners of a square and
eject oil in such a direction as to give a macroscopic rotational motion to the combustible mixture.
These firing types are shown schematically in Figure 2-13.
In addition to these firing methods, two others, vertical and cyclone firing, are primarily
used for coal firing but can be used with oil. These are included in the population distribution
for completeness.
2.4.2 Emissions
Particulate emissions from utility boilers are a function of boiler size more than anything
else. Measurements made in a recent study (Reference 2-23) indicate that the average particulate
emission levels from a sample of uncontrolled boilers decreased from 0.085 to 0.045 Ib/MBtu (36.6 -
19.4 ng/J) as the boiler size increased from 1 to 500 MWg (see Figure 2-14). This sample includes
boilers which emit between 0.01 and 0.26 Ib/MBtu. Emissions from large boilers (>200 MW ) are rela-
tively low because these boilers tend to be new and equipped with sophisticated combustion controls.
EPA has proposed an emission factor for utility size boilers that depends upon fuel grade and, in
the case of No. 6 oil, sulfur content (see Subsection 2.3.2 and Table 2-17). For a boiler burning
No. 6 oil v/ith 0.5 percent sulfur content (by weight), the emission factor is 0.057 Ib/MBtu (24.5
ng/J); if the oil contains 3 percent sulfur, the emission factor is 0.235 Ib/MBtu (101 ng/J).
2.4.3 Population Distribution
Population data for utility boilers were obtained mainly from the Edison Electric Institute
which maintains a data bank containing information on approximately 900 power plants. A total
2-60
-------
o
cn
r in
i- cn
o
o
o
in
o
ro
o
cn
ii
ro
o II o
in co t~,
CM t-» || -
o
in
o
o
co
cn
Lf)
cn
*
LU
o.
o:
H-
t/5
to
o
2:
LU
o
UJ
co
o
o cn
i CM
I I
r- o
o
Cn
o
ro
r- o cn r
.<* . £
cS
ro o «* co
CO r
o
o
o
0
o
o
0
o
0
CD
0
o
0
CM
o
0
r-
l£>
T i
o cn
O
0
in
0
CO
U 1
ro
O
CM
O
u/
ro
O
ro
o
ro
CM
o
CM
in
CM
-
o
ro
un
ro
0
0
o
0
o
^-1 -
"=!
*-"£.
>>
r- 03
cn
C
N
o cn 1 co -g
ii II
c_> DC n3
CO
Nl
r
It)
1
A3
O
r-
c
c~
O
s-
0
Pressure
Rotary
s
OJ
OJ
QJ
O
1
M-
I-V*
Source: I
T«
2-61
-------
SINGLE WALL
OPPOSED WALL
TANGENTIAL
Figure 2-13. Standard firing types for oil-fired utility burners.
2-62
-------
-V
»..«% v
o
CD
O
O
=]-
(O
O
S g>
ro .JE
Q-CO
fd CSJ
O 1
1/1 QJ
> U
O OJ
M- 4-
CO OJ
O Ol
1- >>
i- O
+-> I
O O-
OJ E
i 0)
O CD
OJ >
o
o o
c c
CNJ
O
O SO
r O
O
O
o
o
O
o
'SUOLSSLIUS
2-63
-------
sample of 650 utility boilers was considered, 307 of which were fired on oil or oil with an alter-
nate. Additional data came from the Electric World Magazine (principal design features of elec-
tric utility boilers under construction), the Electric World Directory of Electric Utilities (a
list of power plants operating in the U.S.), and Hoody's Public Utilities Manual (annual data on
regional fuel consumption). The population distributions are presented by firing type and fuel
capability.
2.4.3.1 Population Distribution by Firing Type
Table 2-25 shows the population distribution in terms of firing type. The majority of
these oil-fired boilers use opposed or tangential firing, though approximately half of the
smaller ones employ single wall firing.
*t
2.4.3.2 Population Distribution by Fuel Capability
Table 2-26 gives a population distribution of utility boilers by fuel capability. No dis-
tinction is made between primary and secondary fuel in the case of multifuel plants. As one might
expect coal is the primary fuel, especially in the very large units (>600 MW). Approximately one-
half of the boilers which are not restricted to coal are configured to burn gas and oil.
A geographical distribution of 1973 fuel consumption by type is given in Figure 2-15. The
source of the data, Moody's Public Utilities Manual, also includes annual data of this type for
the last 25 years. Some mild trends are evident in that data, but they are not presented here
because they will be altered by "the need to rely more on coal.
What should be noted in the chart is the high percentage of oil used in New England. Ref-
erence to NEDS data (see Table 1-1) indicates that oil combustion does in fact, account for a sig-
nificant fraction of the particulate emissions in some AQCRs within that area.
2.4.3.3 Population Distribution by Fuel Usage and Particulate Control
Table 2-27 shows how many oil-fired utility boilers are equipped with particulate control
devices (usually electrostatic precipitators). These data were taken from the NEDS system and
include only those units that satisfied at least 10 percent of their fuel demands'with oil (Ref-
erence 2-23). Nearly 60 percent of the residual oil is burned in units whose heat input rate is
larger than 1000 MWe). Only about one-half (55.5 percent) of the residual oil consumed by utili-
ties for the generation of electricity is burned in units that are equipped with particulate conr
trols. Apparently many of these boilers were converted from coal, and, therefore, the electro-
static precipitators were designed for coal flyash. As noted in Section 4.3.1, precipitators
2-64
-------
UJ
Q.
CD
Z
14
cc
CQ
oo
I
UJ
U.
I
u_
o
CQ
11
C£.
CO CO
t1 II
Q GO
«ac
z: CQ
et UJ
_J O
ID o;
Q. UJ
O Q-
D-
UO
(NJ
C\J
UJ
__1
CO
0
0
VO
s-
o>
o
o
o
VO
1
o
o
CO
o
o
oo
1
o
in
I ;
o
!*
c
"^
o
o
o
<£>
s_
J"
r^
+l_
JZ
=5
CQ
S.
'SsS
oo r- 1 1 g
oo vo II ;±
r-^. VO CO «* O
^t- oo r o
^^ - i
co i i co oj o
r 00 in O
Cxi ' 0
TI **
H s - :
/*? 3 t3 +> (C 0- OJ C 0 C
_ 4-> (/) OJ -I- 0 £ -J
Jr. 3 o en 4-> r c fC
?j 0 Q. CS- 0 0 h-
^<-C3.(C OJ >>5-0
,g 0 1- > 0 U- I
- .
c
0
i.
0)
n.
CO
q-
o
o
c
(U
"r"
^
f .
o5
! . =
I '
(0
S~
o'
(0
O)
+-> c
3 0
r- T3
4J O)
CI f«J
0 +->
i* 13
i. a.
<-> E
O T-
0)
r- >,
UJ O1
S-
- 0 =
tO O)
5 o>
UJ -M
_i
O) X
0 Q
S_ S-
3 Q.
0 Q-
OO
-------
a.
o
to
o
to
0
D
«
o
o
CD
to
S-
0)
0
0
o
o
to
o
0
CD
CO
o
o
0
co
1
o
o
LO
1
CD
CD
LO
^_
1
o
f«*^
4-
4-
.c
3
CD
>
3
Q-
O
tO tO CO r- r- r-, CO
COCOP^rv.1^. COO
tO 1 r O
r""" ^O r" d? ^^ ^^ ^3
* * *
t^t *S^ r f"" CO ^^ §- c~a
r * CO O
'
co LO r~- LO co to i
CO i O «d- tOi CM O
LO. CM r O
r^. co LO r~> i CM
rj_ - .
r LO COCOCO r tO O
«3- CM c O
^«
^~T
0
0
1108
>, »a -
>> >> 1 . !- O O
O 0 08 08 1
r f « <
(O -i O
t3 Q.
- ^.
c co
o
U 4-
o
)
(C >J
3: o
c
I- >
+J 0
<0
4-> C
i/l ro
OJ
4-> r C
3 0
r £ X)
4-> tO C/) (U
to a> to
C >, CD CO
1 1 r tO J3
a> <->
0 > C +J
r- T- CO 3
s_ +j o a.
4-> CO S- C
O i a) a.
i S- >>
UJ CO CD
CO t/) 5-
C (DO)
O -t-> -C C
(A CO -P (U
i .C
"O +J CD <1)
UJ C 4->
cyj -i cO
O) «-> E
.. +J td !
CD fO i X
O O 3 O
S_ T- O S-
3 -o a.
o c ca a.
co t-i 0
-------
r
ta
c
o
4_>
ta
^
II
J_
2:
o
J3
c
ra
^~
*
-o
i
E:
n
ct
E:
c
0)
-E
+J
S-
O
z
IO
Ul
II
CJ
s
5
.E
4->
S_
o
2:
0)
3
II
O
§
o
r-
C
10
4J
a:
J=
o
oo
it
eC
l^
C
O)
o
JZ'
4->
3
0
o
tu
2
II
0
to
3
c
^
-t->
3
O
«^
II
E
O
r
14-
O
rt)
O.
II
Q_
o
C
^_
Dl
LU
s
OJ
2:
II
LU
kN\\\\N\\\\\\\\\\\VOsX\\\\\\\
C
o
-13
N
+-> I
3 ta
3
r C
««
I/)
f 3
3 O.
J3
r- (/>
-
u
S- S_
en 3
o o
0) oo
o-
2-67
-------
UIH--K
CM
CM
LU
_]
CO
in
01
r-
O
CO
C
o
r-
*"^
JO
E
O
O
r
ro
£
cu
X
LU
o
£.
r
LU
^
O
p
r-
«r
U
^.
4->
0
CJ
UJ
0
Ol
,
O
S-
4J
C
0
u
r^
r-
O
ra
3
o
in
c c: 4J
i i O O T- o c tn
0 >,=3 S- CU
h- i ja cu cn in
Q) 08 r C OJ
r 3 C * rO -r
10 u- OTJ o ce s-
3 -I- CU CQ O
C 1 4-> r CU CO
C 10 CLr- T3 N
«t 3 E O CU 1-
T3 3 S-iCO
4-> *s.
ex
i E tn
10 3 I- >>
3 in cu s-
t) Cr CU 0^
r- O T- N Cni
in o o i- ai ro
CO CO tO 4-> Cn
cc i to
i- M- J= O O
r O O O O
10 10 CU O
3 i C LU COi
c: cu o c^^
C 3 T- C IO
4- S- CO rd
CU 0) O
JO I J=
E T- o cu
3 o (O cn
z: CQ LU c:
to
ce
^
i-
a> o
M CO f *
r- CU ' CU
to 4J S- 3
to J=E
S- CJ ^
CU 3 i
i CU +->
r- COCO O
0 C -Z. !l
CQ IO
ce
1
i .C 1 CU
r O O O T- to
ro CQ U O
3 I C to
c: co >,=> s- cu
c =j-Q co co in
-I- 0) CO O
+J to +J t i
o in f= c o i
r- O O S- r
3« Q O CJ 4-> =C
CO 1
-t-> a.
10 E in
31- >>
i in co s-
^ Ci CU 0--~
4-> O 'r- M COi
in o o -i- cu to
r CQ CO 4-3 CO
O I IO
r CD O 0 CD
IO IO CU O
3 C LU COi
c cu o c-
C 3 T- C CO
^
c o
4- -i CU CO
O N CO
in T- 4->
S- S- OO IO
CU 0) O
E -^ <-> co
3 o ro cn
z: CQ LU c
IO
CU «i-
N in--
r-: cu i-
co -i jz:
i. O 3
co cn4->
i CU CQ
i- 4-> E:
O IO ^-^
CQ CJ
r CO IO
o
O I-. to
i CM
CM CM i CO
LO CM LO to
cn o to
A r> n
r^* CM co
CM tO
r^* f)
n n
CM «3-
CM CM LO CT>
CO LO
r I
0 S
o o
i O
O 0
r 1 1 CD
V r
O CD A
i O
0 0
co r**
p_
CO ^f
P^ CO
O CD IO O
«t rt
r-. co
CO
O CD r CM
0
O CD
CD O
r 0
CD CD
1 1 CD
V i
O O A
t
in
S-
cu
>t_
o
CO
c
0
r
in
3
Q
o
IO
S-
cu
4J
X
LU
o
cu
s-
f
LU
^
CD
4->
r-
J ^
o
1 ^
0
CO
UJ
-a
CO
"~~
o
&.
I '
c
o
o
c"
^^ r^ ^^
o
1 r ' CM
CM CO
LO CO CD IO
r^. CM LO co
co to o i
» n n ft
to o o to
I i LO
CM co on
co «d-
LO 0 «* CO
IO CM
CM O CM i
co r- CD cn
LO CO i i
i CO CM i
CO CM
CM r
tn r r-~ co
CD
0 0
CD O
i i O
0 0
i 1 1 O
V i
0 0 A
i O
^~
>^
u
c:
cu
o
H-
cu
^
IO
cu
1 *
IO
i-
cu
0
c:
cu
o
s-
01
a.
to
CO
in
FS
3
tn
in
to
.
cu
.,_
CO
^*
s-
ai
Q.
3
O-
c
^J
10
ro cu
CM jz:
1
CM S-
0) ^~-
O 3
C 4->
co ca
i- E:
CO
CO
at: cn
c
o
cu
O "O
!- CO
3 to
O IO
to co
* +-
2-68
-------
that are built for coal, and are later used on oil, Tose much of their efficiency. It is esti-
mated that the average collection efficiency of the predpltators on these boilers is 40 - 55
percent. .
2.4.4 Effects of NOX Control Modifications on Particulate Emissions
Since large utility boilers are major contributors to the total nationwide emissions of NOX
newly installed units are required to meet stringent N0x limits,* and many older units are also
being regulated by local air pollution control authorities. Therefore, boiler manufacturers and
operators are modifying their units to achieve the required NOX reductions, and many of these
modifications have an impact on particulate emissions. The following approaches are currently
receiving the most attention in NO control programs for oil-fired utility boilers:
X
. Optimum combustion aerodynamics (especially turbulence level and swirl)
Low excess air
Off-stoichiometric combustion
Flue gas recirculation
Reduced wal1 temperatures
The flow patterns in the firebox determine how the air and fuel mix and how long the gases
remain in the high temperature region of the boiler. That is, the aerodynamics affect the tem-
perature-time history of the gases and, along with the primary and secondary air flow settings,
dictate the amount of oxygen locally available to the combusting mixture. These are precisely
the parameters which control the production of NOV and particulate. Unfortunately, however, the
X
optimum conditions for low particulates, namely intense, high temperature flames, as produced by
high turbulence and rapid air-to-fuel mixing, are not the optimum conditions for low NOV produc-
A
tion. The latter pollutant is minimized in cooler, longer flames that come from low turbulence
levels and slow mixing of the air with the fuel. Therefore, most attempts to reduce NOV emissions
A
through a redesign of the combustor aerodynamics have had to devise a flow pattern that resulted
in a well-controlled flame whose temperature was high enough to avoid smoke and low enough to
minimize NO generation.
X
Swirl is one of the most important variables in determining the aerodynamics of the com-
bustor, but the effect on particulate emissions of optimizing swirl for low NO is unclear. Since
X
New Source Performance Standards for boilers >250 MBtu/hr (40 CER 60.42), which also limits par-
ticulate emissions to 0.1 Ib/MBtu (43.0 ng/J).
2-69
-------
every burner has an optimum swirl ratio (a measure of ratio between circulatory and throughput veloc-
ities) for minimum NO and one for minimum particulate, the impact of NO reduction efforts on par-
X ' X
ticulate depends upon the swirl ratio in the "uncontrolled" burner and the relationship between the
two optima.
Since smoke and particulate emissions tend to increase as the available oxygen is reduced,
the degree to which the excess air can be lowered to control NO is usually limited by the onset
X
of smoke. The limiting excess air value depends upon the burner, and many modern systems can oper-
ate with as little as 3 to 5 percent excess air.
Off-stoichiometric combustion can be used to control NO from large boilers with multiple
burners arranged in rectangular matrices. In general, this method employs locally fuel-rich com-
bustion in the burner region. Combustion is then completed as the bulk gases rise to lower tem-
perature regimes and are mixed with additional air.
Off-stoichiometric combustion for NO control has been variously achieved through (1) oper-
ation of all burners fuel rich with injection of additional air through overfire air ports, (2)
operation of the lower array of burners fuel rich with upper burners operated on air only (burners
out of service), (3) operation of burners in a fuel-rich or air-rich mode in a staggered pattern
(biased firing). The degree to which off-stoichiometric combustion can be employed is frequently
limited, by particulate emissions which increase as the air supply to the primary fuel-rich stage
decreases. To suppress particulate emissions it is crucial that the second stage air be well mixed
with the primary stage products and that the residence time following mixing of second stage air
be sufficient for carbon and CO burnout prior to quenching. It is particularly important to main-
tain the burners in good condition when operating them fuel-rich to avoid smoke and high particu-
late emissions.
Flue gas recirculation tends to decrease both NOV, and particulate emissions; hence, the use
X
of this technique to comply with a NO limit should not interfere with attempts to reduce particu-
late emissions. The last technique, reduced wall temperatures, is probably the least likely to
be used, which is fortunate because it increases particulates as it reduces NO .
2-70
-------
REFERENCES FOR SECTION 2
2-1. "Emissions, Effluents, and Control Practices for Stationary Participate Pollution Sources,"
Midwest Research Inst., Kansas City, Mo., 1970.
2-2. Emission data from the National Emissions Data Service (NEDS), Environmental Protection
Agency, Research Triangle Park, North Carolina, September 12, 1975.
2-3. Burkhardt, Charles H., Domestic and Commercial Oil Burners, 3rd ed., McGraw Hill, New York,
1969,
2-4. Sjogren, Arne, "Soot Formation by Combustion of an Atomized Liquid Fuel," in: 14th Symposium
' (International) on Combustion. Pennsylvania State University, Universtiy Park, Pa., August
20-25, 1972, The Combustion Institute, Pittsburgh, 1973, pp. 919-927.
2-5 Dickerson, R. A. and Okoda, A. S., "Design of an Optimum Distillate'Oil Burner for Control
of Pollutant Emissions," Rockwell International, Rocketdyne Division, Prepared for U.S. EPA,
Report No. EPA 640/2-74-047, June 1974.
2-6. Hall, R. E., Wasser, J. H., and Berkau, E.E., "A Study of Air Pollutant Emissions from Re-
sidential Heating Systems," EPA 650/2-74-003, January 1974.
2-7. Schindler, R. E. and Ranz, W. E., "Recirculation in a Vortex-Stabilized Oil Flame," CP 65-1,
API Research Conference on Distillate Fuel Combustion, 1965.
2-8. Gills, B. G., "Production and Emissions of Solids, SOX and NOX from Liquid Fuel Flames,
Paper 9, Journal of the Institute of Fuel, Vol. 46, February 1973, pp. 71-76.
2-9. Brown, T. D. and Haneby, V. I., "High Intensity Combustion," Paper No. Inst. _F-NAFTC-2
presented at the North American Fuel Technology Conference, Ottawa, Canada, May 31 to June 3,
1970.
2-10. Barrett, R. E., Hazard, H. R., and Locklin, D. W., "Design and Preliminary Combustion Trails
of a Burner for Firing No. 6 Fuel Oil at Low Flow Rates," presented at National Fuel Oil
Institute, Inc., Third New and Improved Oil Burner Equipment Workshop, September 23-24, 1970.
2-11. Beach, W. A. a/id Siegmund, C. W., "Flame Studies Leading to Improved Combustion in Domestic
Burners," CP 62-13, API Research Conference on Distillate Fuel Combustion, June 19-20, 1962.
2-12. Walsh, B. R., "Burner Research Related to the API Program," CP 62-13, API Research Confer-
ence on Distillate Fuel Combustion, June 19-20, 1962.
2-13. Howekamp, David P. and Hooper, Mark H., "Effects of Combustion Improving Devices on Air
Pollutant Emissions from Residential Oil Fired Furnaces,: APC No. 70-45, presented at the
Annual Meeting of the Air Pollution Control Association, St. Louis, Missouri, June 14-19,
1970.
2-14. Brown, T. D., "The Performance of Vane Swirlers in Domestic Oil Burners," Divisional Report
FRC 72/59-CCRL Fuels Research Center, Mines Branch, Department of Energy, Mines and Resources,
Ottawa, Canada, July 1972.
2-15. Barrett, E. R., Miller, S. E., and Locklin, D. W., "Field Investigation of Emissions from
Combustion Equipment for Space Heating," Battelle, Columbus Laboratories, Prepared for U.S.
EPA, Report No. EPA R2-73-084a, June 1973.
2-16. Levy, A., et al., "A Field Investigation of Emissions from Fuel Oil Combustion for Space
Heating," Battelle, Columbus Laboratories, API Publication 4099, November 1, 1971.
2-17. "1970 Census of Housing," Table A-4, U.S. Bureau of Census, Washington, D.C.
2-18. Gas Facts. American Gas Association Inc., New York, 1970.
2-19. McGarry, Frederick J. and Gregory, Constantine, J., "A Comparison of the Size Distribution
of Particles Emitted from Air, Mechanical and Steam Atomized Oil Fired Burners," JAPCA
.Vol. 22, No. 8, August 1972, pp. 636-639.
2-71
-------
2-20. Locklin, D. W., et al., "Design Trends and Operating Problems in Combustion Modifications
of Industrial Boilers," Battelle, Columbus Laboratories, Prepared for U.S. EPA, Report No.
EPA-650/2-74-032, April 1974.
2-21. Background Information for Establishment of National Standards of Performance for New
Sources; Industrial Sized Boilers," Walden Research Corp., Cambridge, Mass. 1971.
2-22. Cato, 6. A., et al., "Field Testing: Application of Combustion Modifications to Control
Pollutant Emissions from Industrial Boilers -Phase 1," KVB Engineering, Inc., Prepared
for U.S. EPA, Report No. EPA-640/2-74-078-a, October 1974.
2-23. Sahagian, J., Dennis, R., and Surprenant, N., "Particulate Emission Control Systems for
Oil-Fired Boilers," 6CA Corporation, Prepared for U.S. EPA, Report No. EPA-450/3-74-063,
December 1974.
2-24. Barrett, R. E., Locklin, D. W., and Miller, S. E., "Investigation of Particule Emissions
from Oil-Fired Residential Heating Units," Battelle, Columbus Laboratories, Prepared for
U.S. EPA, Report No. EPA-650/2-74-026, March 1974.
2-25. "Technical Options for Energy Conservation in Building," National Bureau of Standards Tech-
nical Note 789, July 1973.
2-26. From draft of revised AP-42, "Compilation of Air Pollutant Emission Factors" as transmitted
over telephone to Aerotherm by T. Lahre, National Air Data Branch, EPA, Durham, N.C., Novem-
ber 21, 1975.
2-27. Putnam, A. A., Krupp, E. L., and Barrett, R. E., "Evaluation of National Boiler Inventory,"
Battelle, Prepared for U.S. EPA, Report No. EPA-600/2-75-067, October 1975.
2-28. Axtman, W., ABMA. Private communication, February 9, 1976.
2-72
-------
, SECTION .3 .
FUEL OIL UTILIZATION AND IMPACT. ON .EMISSIONS,.,
In order to properly assess the problem of participate emissions from oil-fired furnaces and
boilers, it is necessary to look at the types and quantities of oils used today and the, impact of
their utilization on particulate emissions. Therefore,.we discuss beloW the availability and costs
of various fuel oils, their properties and property variations,' their impact on particulate emissions
and the potential effects of fuel oil additives and pretreatment.
3.1 FUEL OIL PROPERTIES, AVAILABILITY, AND COSTS ' ' .
The various grades of fuel oil available today are No. 1, No. 2, No. 4, No. 5 (light) No. 5
(heavy), and No. 6. The first two are distillates; the last three, residual oils.. Grade No. 4 oil
can be a distillate or a mixture of distillate and residual oils (Reference 3-24). These oils are
classified according to their physical characteristics by the specifications in ASTM standard D-396.
These specifications are presented in Table 3-1. Even though refiners and suppliers have agreed to
conform to these specifications (see footnote g in Table 3-1), there are still wide variations in
the properties of U.S. fuel oils. Table 3-2 shows the ranges of some selected fuel oil properties.
These ranges change only slightly from year to year. From Table 3-2 it is evident that fuel oil
properties can vary from below the minimum to above the maximum limits set forth in Table 3-1.
The amount of sulfur in fuel oils is related to the crude oil from which they are derived
and to the additional processing they receive. These factors are responsible for the rather wide
variation in the sulfur content of available fuel oils (see Table 3-2)'. Although the sulfur limit
proposed by ASTM for Nos. 1 and 2 oil is 0.5 percent unless local regulations specify a lower level
(see Table 3-1), the sulfur content of the distillate oils burned in most states is well below 0.5
percent. From Table 3-2, in 1975 an average sulfur content for No. 2 oil, for example, was 0.23
percent.
With the advent of oil desulfurization processes, the refining industry can produce low sul-
fur residuals at some increase in cost but not in unlimited quantities. Desulfurization to a level
between 0.3 and 1.0 percent requires an equivalent of 3 to 6 percent of the quantity of oil desul-
furized as energy for the desulfurization process (References 3-26 and 3-27). Oils with such low
3-1
-------
to
CO
flj
0
_J
LU
LL*
S
CO
LU
LU
cr
LU
DC
a
LU
pj
s
a
oo
LU
J
CO
jfs
is^g
t/)«2 fi.8
oS.'= oj
O u5U §
> . 01,
.t?"O ^
i
£
1
>
U
S
g
2
^
2"
1
,S
i
n
VI
,1 3^
||3
* £
o 5 o*
H"S
u.^
<
u._
Q O
_* **
<~
Sri^
3 **
u. ""
la
|g
-)-
||
s0"
2 x 5-1
ffi!
i ' O i uT
5111 gS.ll |s.l
ra^li I'D 1 u g
g . ot u r"T
o-t-o2
.cs-u.C
11-11
K
S
K
fl
S
C
2
X
s
-E
X
i
I
X
a
S
c
5
s
s
c
s
X
a
a
X
S
X
X
n
s
X
X
C3
2
c
I
O
s
iT
a
S
o
^'3t>'3- n ^
So.* doi" «p jf jj> Ji
o* ...
S S ' : :
«*
^^ oo
JS *°
: : : : S A
3 S *
ir«i m " "
«
S «N -x S A :.
A .
: : : : f §
"^ -5
: : : : S A
*"'
S" S"
; £ 1 1 1 I
Co S- 8 §
: S § A A A
si ll ; _ ; : ;
: |S : : : :
*~^
O ...
O O O
i ; o o o *
o o : : : i
S s S 8 8 8
0 o 0 r-»
& 1 & 1 ' *
& ^ ^ M ~ : :
o_ e_ o o o_ oS1
ill III §11 §11 §11 ~~
SCM 13=- i«. -Of 'S--o
p^'5 corf £ '5' '3^^
I*? r:« i| f| «s| ^^
c g- 2 (£ < 'C ^ «i JS Ji -E u '5 c
'~r|- .- «'| -a >." >,-o" 1"!
seot."5 ^-=2 <= "EooM,- «, S £ Z < o.^ <2 Zo.o' ZoI-oS" z££E^ Zf£.c
w) u ,'. iL "* ^
8 "S | I'l 1
E 0 1 g ^ J2
« & < 1% 3
*2 >. c X c =
'^ S v S 2 g
5j ") o, e .k *"*
^ > 0 u S> 5
c c «rj ** C
=> | es &j2 ?
i 1 i It
I -s i i'! 1
K -8 =3 ; § r
8 i 1 £i °
V V *" {H* b. U
** t/i 'S £ -.» *"
.S - S » .5 S
n c 5 i" °° ° *
a - "«> = ^ J
U g .5 1 73 * S
f ._! c J-
>! " c "^ "" 7 c
15 ^ O S "O v, o
U '='< E 3 C
g. a o * o- =-
« " e _ o p *-
£ ^ M C *-* *~ 3
*" * js si ^ "S co
1 1 ! Js S-i =S
1 s"88 ^ &% 1
"S ^'52 8 1 = I
s o*£ 6^8,r
2 »Ke S-. "fc"-
= rju !i i
a ^ " S -S "S *"22o
» ocE'gg Jggo.
o &O ~ o -^ °" jj c Q
1 |iSfil11|s
fc S §." | J o 2 -3 til
1 inf-jjvfij!
^ « * 2 ~o s"'«c:ji
i ?f:itil=5ip
1 :|Nj5!ii-|fll
K 5^'gS-.2.E.i = -32j
'Ti :5-n=-EE8»!Ess10
tS O-2O* & CZ JZ u a
f s |-| g'Z I si? ?^
c u yO=3**-^'8«'ai>'£7
,O 'S^^*)e:^**tjo^*"
a w a" *o y ^* 'J '5 § £L'^
1 E1? i ^ c £ i ^ ~ S S -
"*j£ uS'5S"°i *eT!
f || irl.slff ||t
^ H « o 5*Z"3!= *^ 2
g o o "5»°o S °*S 2 '§ c 2
S £ .| ^ g > g >* * a.'~ > '
l|S|iif!|ilil-
f s:rr1<^t'?r'!* i-^«
^ Z -5 "S «
3-2
-------
TABLE 3-2. RANGES OF SELECTED FUEL OIL PROPERTIES*
Gravity, °API
Viscosity at 100°F, cST
Sulfur Content, wt. %
Carbon Residue3, wt. %
Ash Content, wt. %
Fuel Oil Grade
No. 2
30.0 - 46.8
1.3 - 3.6
0.004 - 0.51
0.005 - 0.31
No. 6
2.4 - 25.7
32 - 750
0.23 - 3.15
0.35 - 18.0
0 - 0.15
aCarbon residue on 10 percent bottoms for No. 2 and on 100 percent
for No. 6.
*Source: Compiled from data for 1975 in References 3-23 and 3-25.
3-3
-------
sulfur content are used, for example, in some areas of the northeast. Boiler operators are re-
quired to burn these low sulfur oils to reduce S02 emission; as will be discussed later (subsection
3.3.1), there is no unique relation between sulfur content and particulate emissions.
Approximately 70 percent of the U.S. domestic crude is low sulfur (less than 1 percent).
Venezuelan and Middle Eastern crudes, on the other hand, are generally sour (greater than 1 percent
sulfur). The Bureau of Mines publishes monthly statistics on fuel availability by surface level
(Reference 3-31).
Recent prices for various grades and sulfur levels of fuel oils are presented in Table 3-3.
The No. 6 oil price ranges in this table were used to produce the curve shown in Figure 3-1. It is
clear from this figure that the prices for No. 2 and No. 6 oil are comparable only for residuals
with a sulfur content less than about 0.5 percent. In order to show delivered fuel prices, we pre-
sent in Table 3-4 recent tank wagon prices in Chicago, Note that the distillate oil prices in this
table are higher than those presented in Table 3-3.
Owing to the increasing scarcity in some areas (e.g.., New Jersey, Boston, etc.) of naturally
low sulfur resid and the greater cost of desulfurized residual oil, limits on the sulfur content
of this oil could be relaxed in the near future.
According to the most current data on U.S. annual fuel consumption, use in oil-fired furnaces
and boilers comprises approximately two-thirds of the domestic demand (4.5 x 1010 gal/yr) for dis-
tillate oil and virtually all of the demand (4.0 x 1010 gal/yr) for residual oil (References 3-2
and 3-3).* The remaining demand for these fuel oils consists of on-highway, off-highway, railroad,
vessel bunkering and miscellaneous uses (Reference 3-4). Table 3-5 shows the annual fuel consump-
tion for the four user categories of oil-fired furnaces and boilers in the U.S.
The following subsections present data on the grades of fuel oil burned in each of the four
user categories, and discuss for each category the effects of fuel oil properties, additives and
pretreatment on particulate emissions.
3.2 RESIDENTIAL UNITS
Virtually all oil-fired residential units burn distillate oil. (Residual-fired residential
units are classified as commercial for this study.) No. 1 distillate is used in burners which pre-
pare the fuel for burning solely by vaporization, while No. 2 oil is used in burners which prepare
the fuel for burning by a combination of vaporization and atomization. The majority of residential
units are fired with No. 2 oil (see Table 3-6).
Distillate is used in this study to mean ASTM Grades No. 1, No. 2 and No. 4. Residual refers to
ASTM Grades No.'SL, No. 5H and No. 6.
3-4
-------
o
CO
CO
I
CO
ro
r-
CL
01
o
ro
e~
0-
co
ro
c
1
E
s-
01
t
- ^J
ro
LLj
O
CO
ro
c:
r.
O
s-
ro
O
O
CD
re
o
JZ
s-
o
01
01
-o
re
s_
cs
o
LO
CVJ
CO
1
CD
CO
CO
in
o
CO
CO
0
o
CVJ
CO
0
CO
CO
CO
1
in
CVJ
CVJ
CO
o
LO
CO
CO
1
o
o
p__
CO
o
LO
CO
CO
1
LO
,
CO
o
0
LO
,«
CO
1
CD
CO
O
CO
o
CO
CVJ
CO
1
LO
CVl
CD
CO
LO
LO
CVl
CO
1
CD
LO
O
CO
o
o
CVJ
CO
1
0
o
o
CO
o
LO
CVJ
CO
LO
r-
CD
CO
CVJ
o
2:
CO
o
en
r^.
,_!
CO
i
co
.
O
CO
|Q
f>^
f
, 1
oo
o^
OJ
1 .
f-*.
LO
CO
OJ
-""*
t
r
CO
en
OJ
i
f-^
to
CO
OJ
|
f«»
CO
o
CVJ
CO
CO
en
,
CO
o
1
-Q JD
LO r
. CVJ
CVl «=J-
LO l~~
en i
CVJ CVJ
1 1
r-~ o
LO en
CO CO
CVJ CVJ
JD.
Cvl
,3-
[«*,
p^
CVJ
1
CVJ
o
f.^
CVl
ff- *.
o
^
LO
CVJ
CD
CO
1
r~~i
O
O
CO
|
LO
O
z.
JD JO
LO O
O i-^
LO r^.
en in
ood
CO CVl
i i
CVl O
r LO
O r-
CO Cvl
-Q
3-
CVJ
o
o
10
Cvl
1
CD
«^-
CVI
JD
CVl
CD
CO
LO
CVJ
1
CD
^~
CVJ
,^^
o
^
in
CVJ
en
CVl
i
o
o
en
CVl
-Q J3
CO CD
O r
^- co
CVl CO
CO CVl
1 1
i LO
CO *
i-^ CO
CO CVl
CD
O
CO
i.
CL c
r-
01 !->
o o
3
cr o
CU
^J c5
r
t-> r
ro ra
en
01 S_
3 (U
1- CL
CU CO
J= +>
CU
C O
O
&_ 4J
t- r
i CU
3 S-
CO !-
ro
c:
01 S-
O CU
S- Q.
01
s~
e
O E
CL O
CO O
S- CU
S- S-
O CU
<-> s
CO CO
01 01
CO O
-C S- LO
t-> CL 1
C CO
cu -o
S- CU CU
ta ^ u
Q- o c:
3 CU
c cr s-
r- 01
i 4_
CO rO <1J
S- = C£
01 -r-
f^ Ol
3 S- CU
cr o o
i.
01 CU 3
-C ^C O
1 ( OO
ro JQ
3-5
-------
vB
O
33 T
32-
31-
30
IO
3
c
o
29
e-J
a 23
-------
TABLE 3-4. CHICAGO TANK WAGON PRICES (IN CENTS
PER GALLON) FOR FEBRUARY 27, 1976a
Grade
No. 1
No. 2
No. 5
No. 5Lb
No. 5Hb
No. 6b
Minimum Size
Shipment (gallons)
150
400
1500
6000
6000
6000
Price
(cents/gallon)
39.5
37.2
31.9 (1.0)
30.0 (1,0)
29.5 (1.0)
29.0 (1.0)
aThe numbers in parentheses correspond to the
maximum percent sulfur in the fuel at the
quoted price.
bThe prices quoted for these oils are truck
transport prices.
Source: Reference 3-5.
3-7
-------
O
1I
I
CO
o
o
LO
I
CO
UJ
_l
CQ
r
to
O
t
£
r
r-
4-5
si
to
o
E
t l
"«
i
O
S-
CU
E
o
co
(O
I
c
cu
-a
CO
cu
C£
CO
o
CO
en
O
o
si-
o
fs^
I-HS.
*
O
CO
"^
CU
to
r
r"
±i
CO
r"
a
CTk
co
CO
,_!
CO
CM
r
JD
co
t^
o
.a
1
r_
to
a
r-
CO
CU
c£-
0
^
r-.
o
CM
CM
o
CM
o
LO
I
CO
^
,
to
t-i
o
1
oo
o
o
o
i.
o
co
CO
JZ
p
i-
o
o
cu
o
o
co
re
-a
cu
CO
CO
to
cu
to
c c
O 3
I
CO I
s- to
CU -r-
d)
a
CO
cu
T3
cu
a.
ex
cu
cu
co
to
to
CO
CM
I
co
o
o -a cu
cu s-
x f- tu
cu
cu
o
3
O
C/)
3-8
-------
TABLE 3-6. CONSUMPTION OF ASTM FUEL OIL GRADES FOR HEATING IN
RESIDENTIAL AND COMMERCIAL UNITS3
No. 1
No. 2
No. 4
Total Distillate
No. 5
No. 6
Total Residual
Totals
Residential
10l°gal/yr
0.037 (25)b
1.11 (75)
~
1.48 (100)
Commerical
1010gal/yr
0.67 (45)
0.10 ( 7)
0.15 (10)
0.15 (10)
0.58 (38)
0.73 (48)
1.50 (100)
Residential +
Commercial -
1010gal/yr
0.37 (12)
1.78 (60)
0.10 ( 3)
0.15 ( 5)
0.15 ( 5)
0.58 (20)
0.73 (25)
2.98 (100)
aSee Appendix C for conversion to SI units.
bNumbers in parentheses are approximate percent of totals.
Source: Compiled from data in References 3-2 and 3-5.
3-9
-------
The latest particulate emission factor for distillate fired residential units (Reference 3-1)
is 0.018 Ib/mBtu (7.74 ng/J).* Using the distillate fuel oil consumption given in Table 3-6 for
residential units, the total particulate emission rate amounts to 18.5 x 103 tons/yr.
3.2.1 Impact of Fuel Oil Properties on Particulate Emissions
The impact of fuel on emissions from oil-fired residential systems was studied in detail by
Barrett, et al. (Reference 3-6) for the E.P.A. and A.P.I. The program covered emissions from 33
residential heating units and included the effects of various fuel oil compositions.
No significant difference in particulate measurements was found by firing a wide range of
No. 2 oils. The A.P.I, gravity of the oils ranged from 30 to 37, sulfur from 0.05 to 0.3 percent,
and carbon residue from 0.1 to 0.3 percent. The influence of fuel gravity on emissions varied,
depending on whether the unit was tested in an as-found or tuned condition. For units in the as-
found condition, lighter fuels produced lower smoke and total particulate, but heavier No. 2 fuels
produced lower filterable particulate emissions. For tuned units, however, the heavier fuels pro-
duced less smoke and filterable particulate but more total particulate. Thus, if a unit is well-
tuned, a change to a heavier No. 2 will not automatically cause an increase in smoke or solid par-
ticulate emissions.
3.2.2 Impact of Fuel Oil Additives and Pretreatment on Particulate Emissions
The effect of additives to distillate oils on pollutant emissions has been investigated in
numerous studies (cf, References 3-7 to 3-11). Fuel oil additives have aroused considerable interest
as they provide one expeditious remedy to the problem of smoke emissions. Martin, et al. (Reference
3-8) screened and tested over 200 distillate oil additives and found that only 17 reduced particu-
late emissions. Of these 17, only 7 were identified as substantially reducing total particulate
emissions. In these cases, total particulate was reduced by 30 to 50 percent (see Table 3-7).
From this study it was concluded that only metallic additives containing cobalt, iron or manganese
appreciably decrease particulate emissions from distillate oil-burninc1 units. However, the unknown
toxicity of the metallic emissions produced from these additives makes their use questionable.
Although the iron oxides which are emitted when iron based additives are used are not considered
to be toxic by themselves, "recent study suggested that these oxides may combine with POMs in the
ambient air to form a potentially carcinogenic substance" (Reference 3-28).
Assuming 140,000 Btu/gal distillate
3-10
-------
>-
_1
a:
ii
h-
oo
en
OOcO
OO
100
oo
oo i i
uj s:
> UJ
» I
I UJ
"-* I
Q<
Q _l
<£. =>
O
I <
_1 Q.
UJ
UJ CD
_l UJ
i i O
1 =3
OO Q
I I UJ
Q o:
co
UJ
_j
CO
r »*
r:; -tt£ cu
^P -i
o .E:
E O>
O T~
O CU
3
cu
>r_
] *
r
-Q
^^
CO CO ^^ t~~" ^st" CO O">
to LO LO co ^o co co
CD CD CD CD CD CD O
E CU
E (dnJEO SIO U_
. S CJ CJ S U_ 0
CU S5 ^«
1 ' ^^ ^^ ^Q ^^ ^^ LO ^5 LO
O CO r CM r CM CTl r
^^ *« »
ooo o o o o o o o
CM r
O CO O O O O LO
LO CO LO LO i «* O
* * a a
O O O CD O CD O
O O
LO 0 CD O C3 O O
i CD LO r O CD O
r~. i i LO CM LO
0 * D 0* *
Q. CO
r S- r r-
CU CO O (O (O
E O 000
CU T- 'I !-
0 E C E E
O O) O CU CU
i. _E <- J= Q JE
S- O -4-5 CO <_3 OO O
CU 4-> CO CD 1 CD
U. CM i O 30Oi r CO
1 rO O -Q n3 tO (0 i
CU 1 1 ! CO E O «i i 'r-
OO OO i OO O3 S-O
J= S-S- ECO Or S-E 4->O
nji CUO.CUCM cu cys- co-t->
Q. >, EE i Q. r 3 EO 3fC
(O J= E>-< EQ CULL, EU_ T33
S- 4-> O OS 3 O E
eC UJ O CD U_ O i i
OO n3
1 S-
co o>
o
CU r
O !-
E ^t
CU
i- i.
CU CU
1- O-
cu
a: co
cu
r
O
CU E
O T-
S- r
3 r
O T-
oo s:
3-11
-------
The Influence of additive concentration is also very important. For example, the effect of
concentration on participate emissions for the Ethyl CI-2 additive is shown in Figure 3-2. From
this figure we se'e that total particulate emissions can decrease with increasing additive concen-
tration up to a point. Beyond this point, even though the carbon particulate continues to decrease,
the total particulate emissions increases as a result of the increasing concentration of additive.
Other tests on the effects of distillate additives on particulate emissions produced simi-
lar results (cf, References, 3-7, 3-10, and 3-12). At a given operating condition, the use of cer-
tain metallic additives can reduce the amount of particulate emitted from a distillate oil burner.
However, burner modifications can produce an even greater reduction in particulate. The unknown
toxicity of the new emissions which these metallic additives create makes their immediate and wide-
spread use questionable. The cost of the additive required to achieve a 35 percent reduction is
0.3 to 3 cents per gallon. This corresponds to a 0.7 to 7 percent fuel cost increase to the con-
sumer (based on additive costs ranging from $0.60 to $10 per pound) or about $3.00 to $30.00 per
year (Reference 3-8).
In addition to additives for improving combustion and reducing pollutant emissions, other
additives are frequently used to improve noncombustion functions, e.g., storage stability, handling
characteristics, etc. In most cases, however, these additives do little to influence particulate
emissions from distillate oil-burning residential units (cf, Reference 3-8).
In a recent study conducted by Hall (Reference 3-11), the effect of water/distillate oil
emulsions on pollutants from residential heating systems was investigated during tests on a resi-
dential sized warm air furnace. When firing the water in oil emulsion, slightly higher Bacharach
smoke numbers were measured, whereas the furnace heating efficiency remained unchanged and NOX de-
creased as the water percentage increased. The percent water in the emulsion varied from 0 to 32.
3.3 COWERCIAL UNITS
Both distillate and residual oils are fired in commercial systems (see Table 3-6). Fuel
grades No. 2 through No. 6 are used, with distillates and residuals being almost equally in demand.
No. 4 oil is used in some schools and apartment buildings and in situations where the equipment
cannot handle higher viscosity oils such as No. 5 or No. 6.
The latest particulate emission factors for commercial units and the corresponding total
emission rates using the data of Table 3-6 are presented in Table 3-8.
3-12
-------
1.0
0.0
Figure 3-2.
0.010 0.020 0.030
Additive, wt. %
Participate ratio (participate
vjith additive/particulate with-
out additive) versus CI-2 addi-
tive concentration (Reference
3-8).
3-13
-------
TABLE 3-8. PARTICULATE EMISSION FACTORS AND TOTAL EMISSION
RATES FOR COMMERCIAL UNITS9
ASTM
Fuel Oil
Grade
No. 2
No. 4
No. 5
No. 6
Total
Emission
Factor
Ib/MBtu (ng/J) .
0.014 (6.02)
0.050 (21.5)
0.067 (28.8)
0.067S + 0.02
(28. 8S + 8.6)C
Annual .
Emission Rate
103tons/hr
6.7
3.5
7.5
37. 7d
55.4 :
Emission factors assume 140,000 Btu/gal
distillate and 150,000 Btu/gal residual.
These values were obtained by multiplying the
emission factors by the fuel consumption data
in Table 3-6. See Appendix C for conversion
to SI units.
CS = percent by weight sulfur in fuel.
For purposes of illustration, S was chosen as
1.0 percent
3-14
-------
3.3.1 Impact of Fuel Oil Properties on Particulate Emissions
Commercial units were included in the recent study by Barrett, et al. (Reference 3-6) men-
tioned during the discussion of residential burners. Five types of fuel oils were fired in the com-
mercial boilers, No. 2, No. 4, No. 5, No. 6 and a low sulfur (1 percent) No. 6 (LSR). Significant
effects of fuel properties were observed in the commercial boilers. Particulate emissions and
smoke increased with increasing fuel grade when the above oils were fired at the same operating
conditions. Figure 3-3 illustrates typical curves of smoke versus excess air for one commercial
boiler fired at 80 percent load with three oils and natural gas. This figure shows that for existing
units neither of the No, 6 oils can reach the low smoke levels obtainable with No. 2. Moreover,
even if the goal were to comply with a smoke reading of Bacharach No. 3, residual-fired units would
have to use much more excess air than distillate ones and, hence, would operate less efficiently.
Since residual oils contain greater amounts of micellular clusters of true organic molecules, one
would expect them to tend(to burn less efficiently. In addition, heavier oils are higher in molecu-
lar weight, have lower,percentages of hydrogen and would, therefore, have a greater tendency to coke.
Residual oil typically has an ash content of 0.1 percent. The particulate emissions result-
ing from highly efficient combustion of residual oil are constituted almost entirely of inorganic
ash which occurs as oxides, chlorides or sulfates. Residual oil combustion products, however, are
more often found to contain about 50 percent by weight of sooty organic material. Frequently, this
material consists of unburned carbonaceous-solids which tend to be sticky and hygroscopic. The
latter condition probably arises from the presence of calcination products and condensed sulfuric
acid (Reference 3-19).
The effects of fuel grade .on filterable and total particulate emissions are summarized in
Figures 3-4 and 3-5 for all the commercial boilers tested. Ash content tends to be higher for fuels
of low API gravity but is not sufficient to account for higher particulate levels with heavier fuels.
The band of ash content for the fuels is shown in Figure 3-4. The 1 percent sulfur residual oil
(LSR) was closer in performance to a No. 4 or No. 5 oil; it yielded filterable particulate levels
about equal to those from No. 4 oil and only one third of those from the No. 6 oil. Low sulfur
residuals tend to produce less particulate than straight residual oils, since they form less sul-
fates, are typically lower in ash content, and have a lower viscosity (thus, better atomization).
However, no quantifiable relationship exists between fuel sulfur content and particulate emissions
when a wide variety of fuels are considered (see below).
3-15
-------
-------
Mo. 4~-|-No.2«~|
t-N°-HFuel oil grades
100
^r No. 6~{
CD
O
O
O
ro
13
-------
TOO'
90--
fB
D)
O
O
O
c
O
in
O
80--
70
60--
50--
40--
30--
20--
10--
-i
10 15
20 25 30
API, gravity
35
40
Figure 3-5. Relation of total participate emissions
to API gravity for commercial boilers
(Reference 3-6).
3-18
-------
A regression analysis performed in this study indicated that the single most important fuel
property influencing filterable particulate was carbon residue.* API gravity also had a signifi-
cant effect, but viscosity at firing temperature was relatively insignificant. An index combining
carbon residue, viscosity at firing temperature, carbon content and API gravity yielded a good cor-
relation with filterable particulate (see Reference 3-6). Since carbon residue cannot be reduced
by filtering or centrifuging, the only method which can decrease the carbonaceous content of re-
sidual oils is mixing or blending with an oil which has a lower carbon residue.
A brief review of the No. 6 fuel oil analyses presented in Reference 3-25, however, shows
that there is no direct relationship between sulfur content and carbon residue. Some low sulfur
fuels have higher carbon residues than those with higher sulfur content; fuels with essentially
the same sulfur content may vary widely in carbon residue, etc.
Desulfurization of a specific high sulfur fuel oil to a lower sulfur level will lower its
carbon residue. How.ever,.it should be remembered that two high sulfur fuels (of the same sulfur
content) will not necessarily have the same carbon residue and thus, even though they are desul-
furized to the same sulfur level, there may be an appreciable difference in their final carbon
residue contents (Reference 3-29),
Making a residual oil to a specified carbon residue would be unfeasible with current refining
practices and equipment. Doing so would require the addition of another process (e.g., vacuum dis-
tillation, de-asphalting, etc.) and would result in a product much like a heavy No. 2 oil.
3.3.2 Impact of Fuel Oil Additives and Pretreatment on Particulate Emissions .
"Fuel washing" has been suggested as. a potential method of reducing particulate emissions
from oil-fired furnaces and boilers (Reference 3-18). This process has been developed mainly to
pretreat residual oils before firing them in gas turbines. As such it reduces turbine blade dam-
age by removing ash and sediment in the oil (by approximately 70 percent in a typical one
percent sulfur residual oil) without altering the viscosity or gravity. However, fuel washing does
not remove carbon residue, which is the main source of particulate emissions from well maintained
residual oil fired burners. Moreover there are no data (especially in stack emission test results)
*Carbon residue is the amount of carbonaceous residue left after burning a sample of oil in the _
absence of air (Conradson or Ramsbottom carbon test). The percentage carbon rescue does not give
an actual value for the formation of carbon or coke in actual practice, .but only a relative value
of this formation in an improperly designed or inefficiently operated oil burner insta nation .No
trouble should be experienced if .the correct grade of oil 1S used. However, if the oil is causing
carbon trouble, this condition can be eliminated by decreasing the carbon residue with blending
Mixing and blending with a No. 2 oil, which has very little carbon residue, or with any oil that
has less carbon residue than the troublesome fuel, will reduce the total carbon residue.
3-19
-------
that demonstrate the effectiveness of the washing technique in reducing particulate stack emis-
sions.
The effects of additives on smoke and particulate emissions from distillate oil-fired units
were discussed in the section on residential boilers and furnaces. Additives for residual fuels
also have been studied by a number of investigators (e.g., References 3-10, 3-12, 3-14, 3-17). One
study found that chelates of iron and cobalt, and also hydrazine, reduced smoke in heavy fuel oils.
The additives were used in rather high concentrations, ranging from 0.01 to 0.1 percent. Iron
chelate was most effective. At a concentration of 0.01 percent it reduced Bacharach smoke number
from 2.6 to 0.6. Other additives achieving varying degrees of smoke reduction were manganese com-
pounds, copper, iron, and maganese inorganic salts, and organic compounds.
Fuel oil additives can produce additive-fuel interactions resulting in the formation of solids
which could plug up the system. These interactions are so sensitive to the fuel properties that
when one simply changes to a different refiner's product of the same grade or when different sources
of crude are refined by the same refinery, these interaction problems can easily occur. It therefore
seems necessary for the fuel supplier to demonstrate the compatibility of a particular additive with
his fuel before adding the additive to the fuel.
A study on the effectiveness of additives in reducing particulate emissions from commercial
boilers was recently performed by Battelle (Reference 3-30), It was observed that additives con-
taining certain alkaline-earth and transition metals in concentrations in the range 20 to 50 ppm in
residual oil were effective in reducing carbon particulate by as much as 90 percent. While similar
low carbon particulate performance can be achieved with good burner design and adjustment, these
results suggest that additives might be considered as a candidate control technique of the "insur-
ance type" for the many commercial and small-industrial boilers operating in the field that are
marginal in design or do not get the needed service adjustment. Thus, it may be possible for addi-
tives to give some tolerance to "sloppy adjustments."
Even though smoke-reducing additives may decrease visible emissions and, at the same time,
not cause any adverse side effects on operation, environmental health considerations make the use
of additives in commercial boilers questionable at this time. Further studies on these problems
are needed and are currently underway. Thus, the use of these additive materials might create po-
tentially harmful new emissions (even though total emissions might decrease); and therefore, none
of the additives can be recommended as a means of controlling particulate emissions without further
investigation.
3-20
-------
In two recent studies by Hall (References 3-11 and 3-15), the effects of water/distillate
and water/residual oil emulsions on pollutants from commercial heating systems were investigated.
In general, both types of emulsions reduced smoke (see Figures 3-6 and 3-7) and total particulate
from a packaged commercial boiler using (1) low pressure air atomization and an ultrasonic energy
emulsifier and (2) high pressure mechanical atomization and a high pressure emulsifier. The energy
requirements for emulsification are generally less than one percent of the energy content of the
fuel to be emulsified.
In the distillate case, an emulsion with 25 percent water allowed the first unit to run with
approximately 4 percent less excess air without increasing smoke emissions, but the resulting gain
in thermal efficiency was offset by heat lost to vaporization of the water, water supply problems
and emulsifier energy requirements. When firing residual oil, the emulsification results depended
significantly on the emulsifier used, and baseline (oil only) results varied due to the different
atomization technique used. Only the 25 percent emulsification test data reached as low a. smoke
level as Bacharach No,. 3, and this only at about 40 percent excess air. The boiler's smoke inten-
sity tends to stabilize at about Bacharach No. 6 beyond 40 percent excess air when firing only re-
sidual oil (no water); it could achieve the same smoke level with only 15 percent excess air if it
burned a 25 percent emulsification. If the burner were operated on oil only at this last air-to-fuel
ratio, its smoke level would be near Bacharach No. 8, which is marginally visible, depending upon
the character of the plume. In conclusion, emulsification of a residual oil can enable a unit to
achieve a significantly lower Bacharach No. than it could without emulsification, or to achieve an
equivalent Bacharach No. with significantly less excess air.
3.4 INDUSTRIAL UNITS
Over 60 percent of the fuel oil used in industrial boilers is residual. Of the residual oils,
No. 5 (light) usually does not need preheat, while No. 5 (heavy) and No. 6 do require it. Virtually
all the distillate used in the industrial sector (see Table 3-5) is No. 2. The EPA particulate
emission factors for industrial boilers are identical to those for commercial units and are given
in Table 3-8.
3.4.1 Impact of Fuel Oil Properties on Particulate Emissions
The effects of fuel on particulate emissions from uncontrolled* oil-fired industrial boilers
were investigated in a comprehensive field testing study (Reference 3-16). Industrial boilers firing
*No particulate control devices
3-21
-------
8
7-
6-
o 5. .
o
-------
u
O
ro
CQ
O)
J*i
O
CO
8-
Percent water
in emulsion
00
A 10.0
Q 17.5
GJ 25.0
1.00
Figure 3-7.
1,10 1.20 1.30 1.40
Stoichiometric ratio
1.50
Smoke emissions versus Stoichiometric ratio
for water/residual oil-fired commercial
boiler (Reference 3-15). Newer units emit
less smoke.
3-23
-------
a variety of Nos. 2, 5 and 6 fuel oils were included in the sample. The emission tests with No. 2
oil resulted in particulate levels of 0.02 to 0,04 Ib/MBtu (8.6 to 17.2 ng/J), those with No. 5
oil 0.04 to 0.12 Ib/MBtu (17.2 to 51.6 ng/J), and those with No. 6 oil 0.045 to 0.11 Ib/MBtu (19.3
to 47.3 ng/J). Emission rates decreased with increasing size for steam atomized boilers fired with
No. 6 oil (from about 0.1 Ib/MBtu (43 ng/J) at 50 MBtu/hr (14.6 MW) to 0.05 Ib/MBtu (21.5 ng/J) at
130 MBtu/hr (38.1 MW)). However, for units that burned No. 2 or No. 5 oil, the emission rates
appeared to be more a function of atomizing type than of size. Particulate emissions were also
found to increase as the carbon residue of the oil increases (see Figure 3-8).
A limited amount of particle size classification was also performed in these tests. For No.
6 fuel oil, over half the number of particles were less than 4 microns in diameter, and about 90
percent of the particles had diameters less than 6 microns. All particles were less than 50 microns.
Thus, most of the particulate is in the respirable range, 0.5 to 5.0 microns..
3.4.2 Impact of Fuel Oil Additives and Pretreatment on Particulate Emissions
The impact of additives on particulate emissions from industrial boilers has been investi-
gated only to the extent that other distillate and residual oil-fired units have been studied. Some,
of these studies v/ere discussed in the two sections on residential and commercial units.
Magnesium oxide, manganese, and combinations of the two have been used as fuel oil additives
to prevent corrosion and slagging in industrial and utility boilers. The specific choice of addi-
tive depends upon the needs of the particular boiler, as well as environmental considerations. These
additives are used when required to control corrosion even though they generally cause total partic-
ulate emissions to increase.
Fuel washing may offer one possible solution to the problem of corrosion (thus, no metal
containing additives would be necessary and no new emissions would be created). Vanadium, one of
the primary substances in oil responsible for corrosion, is slightly water soluble. Thus, it may
be possible to decrease the vanadium content of oils by fuel washing. This, of course, remains to
be tested. Fuel washing was discussed in more detail in Section 3.3.2.
3.5 UTILITY UNITS
The major grade of fuel oil burned in utility boilers is No. 6, with lesser quantities of
No. 5, No. 4 and No. 2 used (see Table 3-5). A few utility boilers today burn crude oil instead of
the refined products.
3-24
-------
0.5
-!->
CO.
tn
O)
(O
S-
(O
Cu
fO
4J
O
2468 10 12
Fuel oil carbon residue, percent
14
Fiqure 3-8. Effect of fuel oil carbon residue on particulate
emissions from industrial boilers (Reference 3-16;
numbers for identification-in original test program)
3-25
-------
Approximately 55 percent of the oil consumed by utility boilers is burned in units that are
equipped with particulate control devices (Reference 3-19). The EPA particulate emission factors
for utility boilers are identical to those for commercial systems, and are presented in Table 3-8.
These, of course, are average uncontrolled emission rates that assume some mix of controlled and
uncontrolled units; emission rates from controlled boilers will of course, be less.
3.5.1 Impact of Fuel Oil Properties on Particulate Emissions
Since the oils fired in utility boilers are the same as those fired in industrial boilers,
one might to expect to find the same dependence of particulate emissions on API gravity as that dis-
M
cussed in Section 3.4. One recent study (Reference 3-19) investigated the effects of fuel ash and
sulfur and fuel additives on particulate emissions from utility boilers. It has been postulated
that increased sulfur in the fuel can lead to increased SOg adsorption, and hence, a greater mass
accumulation on particulate sampling filters. The net result would appear as an increased solids
emission rate. However, no positive correlation between sulfur content and particulate emissions
was found in the data taken from both controlled and uncontrolled power plant boilers in the size
range over 70 MW .* In addition, no correlation was found between fuel ash content and particulate
emissions from controlled boilers. This was partly due to the fact that organic material (carbon
residue) may have constituted a large fraction of the total emissions. As noted in subsection 3.3.1,
unburned carbon from this residue can be an important constituent of particulate emissions.
3.5.2 Impact of Fuel Oil Additives and Pretreatment on Particulate Emissions
The effects of fuel oil additives on particulate emissions were also investigated in this
study (Reference 3-19). Magnesium and calcium additives are commonly used to improve boiler heat
transfer characteristics and reduce corrosion problems, especially when the fuel contains large
concentrations of vanadium and sodium (References 3-20 and 3-21). In these cases, additives are
the only known means of avoiding serious corrosion problems (other than switching to a lower vanadium
fuel). Since desulfurization also reduces the vanadium concentration, low sulfur residual oils can
usually be burned without additives. Also, it may be possible to decrease the vanadium concentra-
tion by fuel washing (see Section 3.4-2).
By virtue of the fact that they are ash-forming substances, these fuel additives were found
to increase the solid particulate emissions from residual oil-fired utility boilers included in
Reference 3-19. It was found that sulfates consituted approximately 35 percent of the filterable
solids generated by the combustion of high sulfur (>2.5 percent) residual oil using MgO additives.
*A controlled boiler is used here to mean one having a particulate control device (e.g., ESP, scrub-
ber, fabric filter, cyclone, etc.).
3-26
-------
If no additive is- used, the total sulfate decreases due to S03 penetrating the sampling filter,
and the sulfate which is captured occurs as sulfuric acid. Figure 3-9 shows the amount of solid
particulate as a function of boiler size for both controlled and uncontrolled systems. If one
assumes that the efficiencies of the electrostatic precipitators used to control particulate emis-
sions from oil-fired systems are essentially constant, the higher effluent concentrations shown
in Figure 3-9 for systems using MgO additives are readily explained.
3-27
-------
p>7>
/ 1
e s > . .. / 1
O) O) T- / ' '
.»-> (/| 4J +J '1
to -a ai co / 1
r -0 r- E / . |
i rO r -r /.
0 0 10 ' /
1- CO S- 3 /I
,1} r-* [ ^ y 1
C T- E «-> . ' I
O Irt O O / '
CJ =3 CJ E ' 1
0 0 / / -
ia0 n
' /^WV
i § vjf^/
/ f
f
^x ^/
'^/ .*^ 1
>^» "5" '
/*"
/>
^ ^ /
/~~
/ ** i
~& * ^ /
/~ »
~* ~*3 *
o / ^7
c?/ ^/
/ (S /
/ 6-
/ O
0 $ /
\ 9 ' \ n .1.11<..,Q/.£., i i
OkO^t-CMO CXDIO«^-CVJ C
r o
o
- o
in
0 3
" § s
ft
s.
1
o
fO
ex
to
o
- § £
M =
(O
s-
ai
Q.
o
S-
cu
O i
- O -r-
OJ O
CO
o
o
C3
D
i i i i OOOO
o o o
'SUOLSSLIU3
o
3
CU
r-
4-
I
r
r
O
(D
r
O)
cr>
Q- T
CO CO
UJ
0)
s- o
a> c
»->
O -r-
O "O
j~ flU
o
S- 4->
M- 3
o
to J^
to
to T3
cu
r
5- -r-
s_
3
CD
3-28
-------
REFERENCES FOR SECTION 3
3-1. Lahre, T. (NADB, EPA). Private communication to G. R. Offen (Aerotherm/Acurex) concerning
revised emission factors for AP-42, November 21, 1975.
3-2. Fuel consumption data from the National Emissions Data Service (NEDS), Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, June 1975.
3-3. "Crude Petroleum, Petroleum Products, and Natural Gas Liquids, December 1974," Mineral Industry
Surveys, Bureau of Mines, April 1975.
3-4. "Sales of Fuel Oil and Kerosene in 1974," Mineral Industry Surveys, Bureau of Mines, September
1975.
3-5. "The Oil Daily," No. 6092, Monday March 1, 1976.
3-6. Barrett, R. E., Miller, S. E. and Locklin, D. W., "Field Investigation of Emissions From
Combustion Equipment for Space Heating," Battelle, Columbus Laboratories, Prepared for U.S.
EPA, and American Petroleum Institute, Report No. EPA-R2-73-084a.
3-7. Martin, G. B., "Use of Fuel Additives and Combustion Improving Devices to Reduce Air Pollu-
tion Emissions from Domestic Oil Furnaces," presented at Third New and Improved Oil Burner
Equipment Workshop of the NOFI, Hartford, Connecticut, September 23-24, 1970.
3-8. Martin, G. B., Pershing, D. W. and Berkau, E. E., "Effects of Fuel Additives on Air Pollutant
Emissions from Distillate-Oil-Fired Furances," AP-87, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, 1971.
3-9. Barrett, R. E., Moody, J. W. and Locklin, D. W., "Preparation and Firing of Emulsions of
No. 2 Fuel Oil and Water," NTIS Report No. PB189075, prepared by Battelle Memorial Institute,
Columbus, Ohio, 1968.
3-10. Salooja, K. C., "Burner Fuel Additives," Combustion. January 1973, pp. 21-27.
3-11. Hall, R. E., "The Effect of Water/Distillate Oil Emulsions on Pollutants and Efficiency of
Residential and Commercial Heating Systems," presented at 68th Annual Meeting of APCA, Paper
No. 75-09.4, June 15-20, 1975, Boston, Massachusetts.
3-12. Finfer, E. Z., "Fuel Oil Additives for Controlling Air Contaminant Emissions," JAPCA, 1_7_,
No. 1, January 1967, pp. 43-45.
3-13. Schmidt, P. F., Fuel Oil Manual (Third Edition), Industrial Press Inc., New York, .1969, p. 124.
3-14. Exley, L. M., "A Practical Review of Residual Oil Firing Problems and Solutions," Combustion,
March 1970, pp. 16-23.
3-15. Hall, R. E., "The Effect of Water/Residual Oil Emulsions on Air Pollutant Emissions and Effi-
ciency of Commercial Boilers," presented at Winter Annual Meeting of ASME, Paper No. 75-WA/
APC-1, November 30 to December 4, 1975, Houston, Texas.
3-16. Cato, G. A., et al., "Field Testing: Application of Combustion Modifications to Control
Pollutant Emissions From Industrial Boilers," KVB Engineering, Prepared for U.S. EPA, Report
No. EPA-650/2-74-078a, October 1974.
3-29
-------
3-17. Pershing, D. W., et al., "Effectiveness of Selected Fuel Additives in Controlling Pollution
Emissions from Residual-Oil-Fired Boilers," EPA-650/2-73-03^, Environmental Protection Agency,
Research Triangle Park, North Carolina, 1973.
3-18. Siegel, R. D., Rich, E. W. and Morgenstern, P., "An Evaluation of Control Strategies for Sta-
tionary Fuel Burning Sources in the Thirty Inner Cities and Towns of the Metropolitan Boston
Intrastate Air Quality Control Region," prepared by Walden Research, Cambridge, Massachusetts
for EPA, Raleigh, North Carolina, June 1973.
3-19. "Particulate Emission Control Systems for Oil-Fired Boilers," GCA Corporation, Prepared for
U.S. EPA, Report No. EPA-450/3-74-063, December 1974.
3-20. Exley, L. M., Tamburrino, A. E. and O'Neal, A. J., "LILCO Trims Residual Oil Problems," Power,
April 1966, pp. 69-72.
3-21. Kukin, I., "Additives Can Clean Up Oil-Fired Furnaces," Environmental Science and Technology,
7., No. 7, 1973, pp. 606-610.
3-22. "Standard Specification for Fuel Oils," Annual Book of ASTM Standards, Part 23, D396-75,
October 1975, pp. 219-220.
3-23. "Burner Fuel Oils, 1974," Mineral Industry Surveys, Bureau of Mines, Petroleum Products Sur-
vey No. 86, August 1974.
3-24. Schmidt, P. F., "Knowing How Oil Behaves Makes For Better Operation," Power, October 1975,
p. 28.
3-25. "Heating Oils, 1975," BERC/PPS-75/2, Bartlesville Energy Research Center, ERDA, Bartlesville,
Oklahoma, August 1975.
3-26. Weiland, J. H. (Environmental Protection Department/Texaco). Private communication to 6. 6.
Poe (Aerotherm/Acurex), December 16, 1975.
3-27. Granville, M. F. and McCay, R. C., "Potential for Energy Conservation in the United States:
1974 - 1978," A Report of the National Petroleum Council, September 10, 1974, p. 29.
3-28. Moran, J. (Health Effects Research Laboratory/EPA). Private communication to G. G. Poe
(Aerotherm/Acurex), December 8, 1975.
3-29. Weiland, J. H. (Environmental Protection Department/Texaco). Private communication to G. R.
Offen (Aerotherm/Acurex), February 13, 1976.
3-30. Giammar, R. D., et al., "The Effect of Additives in Reducing Particulate Emissions from
Residual Oil Combustion," presented at Symposium on Stationary Source Combustion, U.S. EPA,
September 24-26, 1975, Atlanta, Georgia.
3-31. "Availability of Heavy Fuel Oils by Sulfur Levels," Mineral Industry Surveys, Bureau of Mines,
U.S. Department of the Interior, Washington D.C. 20240. Published monthly. Contact: S. K.
Patterson, (202) 634-1088.
3-30
-------
SECTION 4
EXISTING CONTROL METHODS FOR PARTICIPATE EMISSIONS
A variety of methods are known to reduce particulate emissions from oil-fired boilers and
furnaces. '- For economic, technical, or institutional reasons some of these are not applicable to all
the categories of users. For example, since fuel accounts for 80 to 90 percent of the annualized
costs of owning and operating a boiler, these units are nearly always well-maintained in order to
minimize the consumption of costly fuel. On the other hand, fuel costs may represent only 5 to 10
percent of a homeowner's living expenses (in a cold climate). Therefore, many individuals do not
give high priority to activities such as maintenance, which may reduce this cost by 5 to 10 percent.
Hence a mandatory periodic inspection and maintenance program could be an effective particulate con-
trol tactic if applied to residential, commerical, and small industrial units, but would be less
effective for large industrial or utility boilers than are particulate collection devices.
The organization of this section takes these differences into account by discussing sepa-
rately the controls that have been suggested for each user category. Because we feel that greater
emphasis and interest will be placed on control of commercial systems than on residential units,
and because the same control techniques are generally recommended for boilers and furnaces in both
size categories, we have chosen to discuss particulate controls for commercial systems first. That
discussion will be followed by a description of possible controls for residential units, with fre-
quent reference being made to the material just presented for commercial systems. Then we will
present a summary of potential control techniques for industrial and-utility boilers. Since many
of the same techniques are used in both categories, the discussion has been combined in one sub-
section.
The discussion of each control technique will start with a description of the system and its
operating principles. We will then present its reported effectiveness. Any problems that may be
associated with the use of this control system and the cost of owning and operating it will be sum-
marized too.
4-1
-------
The effectiveness and cost of many, but by far not all, of the control techniques discussed
in the literature are well known. For example, some sources claim significant reductions in partic-
ulate emissions firom specific control techniques, but they do not report emission levels, boiler
types on which the unit has been tested, boiler types which cannot accommodate the system, etc.
Therefore, we do not have sufficient data to recommend such systems as a particulate control tech-
nique. To clearly separate these potentially valuable systems from the currently demonstrated
techniques, we include only demonstrated and thoroughly documented technologies in the first sub-
sections under each user category (commercial, residential, and industrial/utility); the last sub-
section is then devoted to emerging technologies, which includes both these existing, but not fully
tested, systems as well as ideas that are currently being developed.
4.1 COMMERCIAL FURNACES AND BOILERS
The most commonly proposed techniques for the control of particulates from commerical oil-
fired furnaces and boilers are the widespread implementation of, and adherence to, proper operating
and maintenance procedures and the use of new improved burner or combined burner/combustion chamber
designs. In the following three subsections, therefore,-we discuss the methods and results obtained
with proper operation and maintenance procedures (Subsection 4.1.1), with the use of redesigned
burners and/or combustion chambers (Subsection 4.1.2), and with dust collectors (Subsection 4.1.3).
Subsection 4.1.4 deals with emerging technology. The effects of fuel oil grade variations on par-
ticulate emissions have already been discussed in Section 3.
4.1.1 Operation and Maintenance Procedures
Interest in proper operation and maintenance procedures as a particulate control strategy
for commercial oil-fired units has received renewed emphasis as a result of recent surveys of the
condition of these units in selected northeastern areas (New York - Reference 4-2; Maryland - Ref-
erence 4-1). These investigations concluded that many units were not maintained, but that those
Which were did not emit excessive quantities of particulate. The following three subsections
describe the strategy of periodic inspection and proper maintenance as it relates first to the
operation of the burner, then to the rest of the system (fuel handling as well as firebox), and
finally to the administrative aspects of such a strategy.
4.1.1.1 Burner Tuning or Replacement
Smoking fires, or those with high smoke spot number in the absence of a visible plume, are
usually caused by improper burner adjustment, dirty burner cups or nozzles, or damaged burner
4-2
-------
components (Table 4-1). Several studies have shown that smoke emissions can be reduced significantly
by proper burner tuning and replacement of worn or damaged components.
In one of these studies (Reference 4-1) 400 tests were conducted on 310 burner units. Of
these, 160 were in residential units that burned distillate, 48 were in commercial units fired with
distillate, and the remaining 102 were in commercial units that used residual oil. Measurements
were taken of smoke emissions (Bacharach Number), flue gas composition, and temperature before and
after the burner units were adjusted. The adjustments included any of the following: replacement
or repair of burner nozzle or cup, adjustment of primary or secondary air, repair or replacement of
burner parts affecting oil distribution, and repair of the combustion chamber. Of these, nozzle
replacement and air flow adjustment were found to have the greatest impact on smoke emissions.
According to this study, the service procedures described above reduced smoke emissions
nearly 40 percent. The average Bacharach smoke number for all units was reported to be 2.8 before
adjustment and 1.7 afterwards. This post-adjustment figure includes lower values for residential
units, an average number of 2.2 for all commercial units (distillate and residual), and a mean
reading of 2.6 for all residual fired commercial systems. As mentioned in Section 3, residual-
fired units generally emit more particulate matter than do distillate fired burners because of the
higher solid carbon content in the oil.
Thermal efficiencies were also determined for each burner unit before and after adjustment.
The results from all the units showed that, on the average, only slight improvements were obtained
(from 78.2 to 79.2 percent).
Another study (Reference 4-3), undertaken by the City of New York, found similar results.
Twenty-seven residual oil burners were tested during this program, and all but three were able to
attain stack thermal losses of less than 20 percent and a Bacharach Number no greater than 3 with
only minor adjustment. The problems with the remaining three units were attributed mainly to de-
sign flaws, such as flame impingement on combustion chamber walls, overfiring of the boiler, or
improper combustion chamber modifications.
The effect of excess air on smoke emissions from six commercial boilers was reported in
another study (Reference 4-17). These results will be mentioned here briefly even though they come
from units tested in the "as found" condition because adjustment of excess air is part of the tuning
i
procedure. They show that all of the units could achieve a level of Bacharach No. 2 at an excess
air setting of no more than 30 percent when burning No. 2 oil. Four of the six units could even
achieve Bacharach No. 1 at excess air levels below 30 percent.
4-3
-------
TABLE 4-1. COMMON OIL BURNER OPERATIONAL PROBLEMS IN COMMERCIAL AND
RESIDENTIAL BOILERS AND FURNACES (Reference 4-54)
Burner Type
Commercial & Industrial (steam boilers)
Horizontal Rotary Cup
Steam Atomizing
Air Atomizing
Domestic (residential furnaces, water
heaters)
Rotary Cup
Vaporizing
Oil Type
Usually Used
No. 4,. 5
No. 4, 5, 6
No. 5, 6
No. 5
. No. 1 or 2
No. 1 or 2
No. 1
Defects Which May Cause
Odors and Smoke
Oil preheat too low or too
high; nozzle wear; nozzle
partly clogged; impaired
air supply; clogged' flue
gas passages; pobr draft;
overloading
Oil preheat too low or too
high; burner partly clogged
or dirty; impaired air sup-
ply; clogged flue gas pas-
sages; poor draft; over-
loading
Oil preheat too low or too
high; burner partly clogged
or dirty; impaired air sup-
ply; clogged flue gas pas-
sages; poor draft; over-
loading; insufficient
atomizing pressure
Oil preheat too low or too
high; burner partly clogged
or dirty; impaired air sup-
ply; clogged flue gas pas-
sages; poor draft; over-
loading; insufficient
atomizing pressure
Viscosity of oil too high;
nozzle wear; clogged flue
gas passages or chimney;
dirt clogging air inlet;
oil rate in excess of de-
sign; loose or bad flue
pipes; poor mixing of air
& fuel; flame impinging on
firebox or boiler surfaces;
absence of, or faulty baro-
metric damper
Viscosity of oil too high;
clogged nozzle or air sup-
ply; oil rate 1n excess of
design
Fuel variations; clogged
flue gas passages or
chimney; clogged air
supply
4-4
-------
These studies, therefore, lead one to conclude that proper maintenance procedures can produce
the following results:
1. A decrease in smoke emissions per unit of fuel consumed
2. An additional decrease in annual particulate emissions because of the lower fuel con-
sumption that comes from improved efficiency
3. A consequent cost saving to the users by the slightly improved thermal efficiency
It is important to note that while it is desirable to keep the excess air to a minimum, the
efficiency gained (only a few percent) is small compared to the efficiency (10-15 percent) that can
be lost due to fouling of heat transfer surfaces by soot. Consequently after cleaning the C02 con-
centration should be set 1-2 percent lower than the minimum attainable from the burner to avoid
sooting with time. '
4.1.1.2 Additional Maintenance Requirement
There are other operation and/or maintenance practices that can reduce smoke emissions and
increase thermal efficiency. These include:
i
1. Cleaning soot from the boiler tubes to improve heat transfer effectiveness (i.e., with
vacuum cleaner)
2. Adjusting oil preheat system to give correct oil temperatures for optimum atomization
3. Ensuring that combustion chamber is in correct working condition; there should be no
flame impingement or cold surfaces
4. Sealing all oil or air leaks
5. Ensuring that proper maintenance is performed on any dust collectors
6. Verifying that burner controls are providing correct primary air-to-fuel ratio with
changing load
7. Cleaning sludge from oil storage tanks to reduce burner plugging, and, consequently, in-
creased emissions from poor atomization
Negligence in some of these practices contributed to the failure of many burner units tested
in the New York City study to meet the minimum thermal efficiency and maximum Bacharach Smoke Number
criteria when tested in the "as found" condition.
The cost for proper burner maintenance is slight compared with the cost of annual fuel usage.
Normally, this service requires 4 to 8 hours of labor at approximately $25 an hour.
4-5
-------
4.1.1.3 -Procedures and Training for Boiler Operators and Maintenance Personnel
The investigations discussed above have shown that significant reductions in particulate
emissions can be obtained by sound maintenance practices. In order to realize this potential,
proper training is essential for boiler operators and maintenance personnel.
The New York City Department of Air Resources, in 1971, developed an air pollution control
guidebook for boiler and incinerator operators (Reference 4-4). This guidebook is used in class-
room training for the boiler operators as well as in the field for day to day guidance. It clearly
presents the precautions to be taken to prevent excessive smoke, and the steps to be taken when ex-
cessive smoke does occur.
The EPA currently is also developing guidelines for use by service technicians when maintain-
ing and adjusting oil burners to minimize pollutant emissions, especially particulates. A separate
guideline is being prepared by a contractor to the EPA for each user category, including commercial
oil-fired boilers (Reference 4-5).
4.1.2 Burner and Combustion Chamber Redesign and Retrofit
Continuing research efforts into both the fundamentals of combustion and the operation of
boilers and furnaces have uncovered the relationships between various combustion parameters and the
emission of air pollutants. Of these parameters, oil atomization method, excess air, combustion
chamber temperature, flue gas recirculation, residence time, and air-fuel mixing have been found to
affect the formation of particulates. In addition, investigators have noted that some burners, such
as the rotary cup, need to be maintained daily to operate at high efficiency and low smoke.
Therefore, the use in new installations of burners, combustion chambers, and/or associated
systems which are designed on the basis of these recent findings could reduce particulate emissions.
Retrofit with new burners of boilers or furnaces found to have high emissions could have an impor-
tant impact also. New systems and retrofits are discussed separately in the next two subsections.
4.1.2.1 Burner and Combustion Chamber Redesign
In this subsection we describe the state-of-the-art in combustion system design practices
for low emissions. Atomization and combustion chamber size are treated here, while a discussion
of more recent innovations is deferred to Section 4.1.4.
4-6
-------
Oil Atomization Methods ; ,-
Proper oil -atomization is essential for complete, smoke-free combustion of oil. The oil
droplet resulting from atomization must be vaporized before the oil is burned. Since large droplets
will not be completely vaporized by the time they have traveled through the flame, they will leave ,
as intermediate products, such as coke particles. Very large oil droplets could even pass completely
through the flame in the liquid stage and spatter on the relatively cool heat transfer surfaces caus-
ing smoke (Reference 4-7). Therefore, fine oil droplets from the atomization process are a prereq-
uisite to..smoke free combustion.
Conventional burners employ one of four different methods to produce fine oil droplets: high
pressure air or steam atomization, low pressure air atomization, mechanical (or oil pressure) atom-
ization or rotary cups. These methods are described in Sections 2.1.1.1 and 2.2.1.3.
Conventional atomizers can perform satisfactorily to give low smoke and high thermal effi-
ciency. However, frequent maintenance is necessary to ensure this condition. New methods of atom-
ization being developed to minimize the cost associated with these maintenance requirements should
also be beneficial in limiting smoke emissions. Presumably, smoke emissions are reduced and high
thermal efficiency is maintained by these new burners because they remain tuned longer. Several
new successful methods of atomization are described in Section 4.1.4.
Combustion Chamber Size
Residence time also has been shown to affect the amount of pollutants emitted. Investigators
(e.g., Reference 4-13) have found that the longer residence time obtained by increasing the size of
a combustion chamber has resulted in lower particulate emissions. The effect of residence time on
particulate and gaseous emissions from residential units will be discussed quantitatively in Sub-
section 4.2.2.1.
4.1.2.2 Retrofit of Existing Burners and Combustion Chambers
Improved burner maintenance has already been shown to be an effective particulate reduction
technique (see Section 4.1.1). Retrofitting of existing units which are difficult to maintain or
defective in design is another way to reduce the particulate emissions. Such retrofits usually
consist of burner or control replacements, minor modifications to the combustion chamber, changes
to the draft system, etc.
One commonly practiced retrofit is replacement of rotary cup burners with pressure or air
atomizing units. Although the rotary cup burner can function well when it is properly maintained,
4-7
-------
experience has shown that this burner usually emits more particulates than other types, especially
when burning residual oil (Reference 4-14). The burner is very sensitive to changes in viscosity
and is difficult to adjust (Reference 4-15). When No. 6 oil is used, the rotary cup usually must
be cleaned daily (Reference 4-16). Therefore, at least one regulatory agency (Maryland) has pro-
hibited the use of these burners.
The cost to replace a rotary cup burner with an air atomizing type varies with burner size.
A 10 gph (10.5 cm'/sec) burner costs about $1000, and the, labor to install it adds approximately
$500. If such a unit consumes 10,000 gal/yr at 40
-------
chambers could frequently reduce particulate emission and increase thermal efficiency of other/vise
well maintained units. Presumably,'such alterations would bring the emissions and efficiency into
line with those achieved by well designed existing boilers and furnaces.
4.1.3 Dust Collectors
Although dust collectors can be used in the exhaust system of commercial units, they are
generally not used for this purpose because of their high cost relative to the, rest of the system and
their need for greater attention than the boiler usually receives. 9ne study has shown that they
can be effective in reducing emissions from some commercial boilers, but the investigation did not
report on the condition of these burners, or on the attention paid by the operator to the collector
(Reference 4-14). The same report suggests that dust collectors on commercial boilers and furnaces
may be a liability rather than an asset in many cases because of insufficient or improper mainten-
ance (collection bag or hopper not emptied and cyclone not washed).
The cost of a cyclone collector (see Subsection 4.3.1.2 for a description of a cyclone collec-
tor) is about $4000 installed for a 20 gph unit (about 3 MBtu/hr or 0.88 MW) and $5800 for an 80 gph
one (10.4 MBtu/hr or 3.05 MW). These figures are for systems that are designed to collect 80 per-
cent (by weight) of the particles over 10 microns. If one annualizes the cost of the collector by
using a carrying charge of 20 percent (which might be appropriate to cover interest, maintenance,
and related costs for an amortization over 10 years), and if one assumes that the 20 gph (0.88 MW)
boiler consumes 20,000 gal/yr at $0.40/gal, then the cost of the collector represents about 10 per-
cent of the annual fuel bill. Alternatively, the cyclone would add nearly 12 percent to the capital
cost of the boiler (about $35,000, including installation, fuel supply systems, and controls, but
excluding the building).
Electrostatic precipitators are not available in this size range, wet scrubbers have waste dis-
posal problems that make them unsuitable for typical commercial boiler applications, and fabric fil-
ter baghouses have a tendency to become plugged or damaged by corrosion when used to capture particu-
late emissions from oil-fired boilers (all four kinds of.particle collectors are discussed in greater
detail in Subsection 4.3.1).
4.1.4 Emerging Technology
Several oil burners using improved atomization techniques or optimized flow patterns have re-
cently been marketed. Some data are available for many of these, but not enough to determine their
applicability to general service or their effectiveness if included in a regional control strategy.
4-9
-------
4.1.4.1 Novel Atomization Techniques
One novel method of oil atomization makes use of pressure oscillations at ultrasonic fre-
quency. Simple atomizers that are based on this principle have been developed by attaching ceramic
piezoelectric crystals to a stepped horn amplifier containing a fuel feed tube which discharges at
the tip of the horn. The unit is tuned to its resonant frequency with the vibration amplified by
the horn. The liquid fuel at the tip then breaks into droplets. Extremely fine particles can be
produced at ultrasonic frequencies with low-level power supplies. These designs use relatively large
oil passages, even at low flows, which explains the improved maintenance characteristics shown in
several long-term tests. The same approach has been applied to larger atomizers. Fuel rates up to
approximately 20 gph (equivalent to approximately 3 MBtu/hr or 0.9 MW heat input) have thus- far
been attained (Reference 4-9).
Progress also has been made in the field of acoustic nozzles by units incorporating cavity
resonators. Steam or air is used as the driving medium which, in conjunction with the resonant
cavity, generates an intense sound field that shears the fuel into atomized particles. A wide
range of spray patterns and flame types result, depending on the cavity and fuel-nozzle configura-
tion. In several applications such nozzles have achieved rates in excess of 1000 gph (1050 cm2/sec)
with appreciable turndown. Manufacturers claim that, when compared to conventional nozzles, reso-
nant-cavity nozzles permit the use of lower excess air, have reduced clogging tendencies, and can
operate stably over a wider heat release range (Reference 4-9). One manufacturer claimed that
solid particulate emissions were cut by 90 percent and combustion efficiencies raised to over 83 per-
cent with the installation of an acoustic nozzle (Reference 4-8). Unfortunately, quantitative data
on particulate mass emission rates or smoke spot numbers could not be obtained to substantiate this
claim. The approximately 3000 units which have been sold to date, in the range of 50 to 1000 gph
(52.5 - 1050 cmVsec) have been marketed on the basis of their higher thermal efficiency (83 to 87
percent) and reduced maintenance (one-third to one-half the effort required for normal burners) at
"no smoke" conditions. These atomizers cost about $1400 for parts and labor when installed as a
retrofit.
When considering any new nozzle, including a sonic one, care must be exercised to insure that
the flame pattern produced by the nozzle does not cause flame impingement on the walls of the fur-
nace and, therefore, poor combustion.
A third novel atomization technique is incorporated in an experimental burner, the Babington
Burner, which is discussed in Section 4.2.3.
4-10
-------
4.1...4.2 Combustion Aerodynamics
An important parameter in controlling particulates is the amount of swirling (circulatory)
motion given the gases in the combustion region (Reference 2-7 to 2-10). In general, there is an
optimum value of swirl which yields low particulate emission for a given combustion system, as shown
in Figure 4-1. Swirl can also influence the particle size distribution in the following ways:
Increasing swirl reduces the quantity of large particles
Overswirling increases the production of submicron soot and reduces the burnout of the
largest particles.
Other aerodynamic factors influencing particulate emission are recirculation in the gas flow
and relative air/fuel droplet velocity. The combustion of an atomized liquid fuel occurs in the
homogeneous mixture of fuel vapor and combustion air; however, partial combustion, or cracking of
the liquid, fuel droplets can also occur. Such droplet combustion results in soot production, but it
can be avoided if the relative velocity between the air and the fuel droplets is above a threshold
value known as the "extinction velocity." This threshold velocity increases with both increasing
droplet diameter and oxygen concentration. Internal recirculation of the combustion gases reduces
the local oxygen concentration and, consequently, the extinction velocity, and so promotes soot-free
combustion.
The investigator who obtained the results that have been reproduced in Figure 4-1 also found
that at optimum swirl the fuel spray angle had little effect on particulate emissions. The swirl is
a part of the whole aerodynamic flow pattern of the burner/combustion chamber configuration and gener-
ates recirculation into the oil jet. As the swirl is increased both the internal and external entrain-
ment to the jet increases. However, at excessive swirl the external recirculation between the jet and
the wall of the furnace decreases (Reference 2-9) due to the influence of the wall, and soot production
is increased.
While astute selection of the swirl rate in a burner (by the manufacturers) can result in low
particulate emissions, increasing swirl in general causes a corresponding increase in NO . Therefore,
this parameter must be determined with low total emissions in mind.
Optimization of combustion aerodynamics is considered by many experts in the field to be the
best approach toward achieving the triple goals of low particulates, low NO , and high thermal
efficiency. Such an optimization attempts to control mixture (i.e., flow patterns and turbulence
levels) and residence time to achieve complete combustion at a temperature that, locally or globally,
4-11
-------
1.0
0.8
Solids
Burden
0.6
% by wt
of fuel
0.4
0.2
4% 0
I j
I _ I _ 1 _ I
0.6 0.7 0.8 0.9
Tangent of swirl air angle
1.0
Figure 4-1. Effect of combustion air swirl on
solid emission (Reference 2-8).
4-12
-------
never falls below the level required to sustain combustion nor rises above the level at which NOX
formation becomes significant. Although most of the research in this area has been directed at NOX
reductions, with smoke merely a limiting factor, several investigators have been able to reduce smoke
concurrently with lowered NOX-
One such study began by investigating the pollution characteristics of high-pressure atomized
distillate oil burners when fired coaxially into cylindrical combustion chambers (Reference 4-11).
The results of these tests were analyzed and the findings used to design an "optimized" 9 gph (9.45
cmVsec). burner (approximately 1.3 MBtu/hr or 0.4 MW). The objective was to make minor changes in
the burner blast tube end which would result in a reduction in emissions and improvement in efficiency.
These changes were intended to be of a kind that could be retrofitted onto existing burners without
requiring the use of new or special servicing requirements.
The resulting "optimized" burner employs peripheral swirl vanes oriented at 25° relative to the
blast tube axis. This swirl vane angle gave the best compromise between smoke emissions and nitric
oxide emissions. A flame retention device was not used because the pollution characterization tests
had indicated that these devices promoted mixing and recirculation in the flame zone which .was con-
ducive to NO formation -under efficient, near-stoichiometric combustion.
A
Nitric oxide emissions from this burner were found to be very low, at a level of about 0.6 mg
NO/g fuel (approximately 35 ppm), compared to 1.2 mg NO/g fuel for similar sized representative com-
mercial burners when fired into the same cool wall combustion chamber. The test burner also was
found to be capable of operating smoke free at as low as 2 percent excess air, whereas the commer-
cial burners required from 5 to 25 percent excess air to eliminate smoke. Simulated field testing
of the burner for a total of 112 hours showed little variation in performance and no noticeable de-
gradation in emissions. When the burner was fired into a hot wall (refractory lined) chamber, NOX
emissions rose to about 1.0 mg/g fuel during smoke free operation. Conventional burners generated
over 1.5 mg NO/g fuel under similar conditions. This burner should be commercially available with-
in 3 years.
The above burner was designed only for the lower size range of commercial boilers, which
nearly all use distillate.' However, another new burner has been developed for residual fired units.
It uses both swirl and recirculation, as shown on Figure 4-2a. This burner has a multistage atomizer,
utilizing shear forces in the first atomizing stage, and vibrational or acoustic energy in the final
stage (see Figure 4-2b). Field performance has shown that the burners are able to operate over a
wide range of oil viscosity and at excess air levels as low as 2.4 percent (for No. #6 oil).
4-13
-------
Recirculating eddy Air leaving
Combustion air \Stagnation «J*«t1on block
/ V, / belt I
Burner /
housing
I
Atomizer
N
f-
«
M
; \
Backflow (if any) Recirculating
centre core
Figure 4-2(a). Typical flow patterns in vortometric
burner (Reference 4-12).
Steam
Oil C
Generator
Oil tube
,Steam inlet
Annul us
Oil tube
Cross-section
Figure 4-2 (b). Atomizer is key component of vortometric
burner (Reference 4-12).
4-14
-------
The particulate emissions at full load were under 0.14 Ib/MBtu (60.2 ng/J) for a 2 MBtu/hr
(0.59 MW) unit, although the Bacharach Number was slightly above 4. At 85 percent load, the par-
ticulate loading was about 0.22 Ib/MBtu (94.6 ng/J) while Bacharach Number was between 3 and 4
(Reference 4-12). Both these emission rates are slightly below the emission factor of 0.24 lb/
MBtu (103 ng/J) proposed in Section 2.2.2 to describe existing commercial units fired on No. 5 oil.
Thus the advantage of this burner is that it can achieve higher efficiencies than most other designs
without increasing particulate emissions. It is also a relatively Tow NOX emitter.
4.1.4.3 Modulation
A potential particulate control technique, which can be added to new units or retrofit to
existing ones, is a modulating unit to eliminate cyclic on-and-off operation. It has been well docu-
mented that the transient nature of cyclic operation contributes to particulate emissions (e.g.,
Reference 4-17). When a burner is turned on, a few minutes are required for it to attain steady
state conditions (see Section 4.2.2.1), and investigations have shown that particulate emissions,
modulating burner control can b,e utilized. With modulating control, the burner is adjusted con-
tinuously to meet changing boiler or heat load requirements. The system never cools because the
burner is always on. Thus, the modulating operation eliminates the peak particulate, smoke, HC, and
CO emissions associated with on-and-off or step turn down controls and, coincidentally, prolongs
system life. The effect on NOX emissions is expected to be minimal. Unfortunately, no quantitative
data are available on the difference in particulate mass emissions over an extended period of time
(e.g., one day) between a standard unit and one equipped with a modulator.
Most new burners over 10. gph (10.5 cm3/sec) can be ordered with modulating control. Therefore,
when an existing burner needs to be replaced, a unit with modulating control should be considered.
The cost involved is about $500 more per burner for the modulating unit. If the modulating unit is
to be installed on an existing burner in the field, the extra labor charge is approximately $300
(i.e., total cost to retrofit is $800).
4.1.4.4 Retrofit Combustion Chamber Liners
Particulate emissions from some existing commercial and residential boilers can be reduced by
placing a ceramic felt liner just inside the existing inner surface of the combustion chamber. This
liner surrounds the combustion volume with a thin layer of insulation whose inner wall reaches tem-
perature equilibrium with the combustion gases more rapidly than the older insulation, thereby re-
ducing the particulate emissions during the start-up transient. However, the felt cannot withstand
4-15
-------
very Ivigh temperatures i.e., those corresponding to combustion volume temperature above 2600°F
(1427°C). Moreover, no specific data are available to show how much this retrofit reduces mass
emissions.
4.2 RESIDENTIAL
The main fuel used in oil-fired residential heating units is No. 2 distillate oil. Since par-
ti cul ate emissions from the burning of this grade of oil are small compared to those from units fired
with residual oil, high smoke readings are mostly due to improper air-to-fuel ratios, improper draft,
worn-out burner components, or poor burner and combustion chamber design.
As with commercial units, an effective and practical technique to reduce particulate emis-
sions from residential oil burners is to keep them well maintained. Procedures and programs direc-
ted toward this goal are discussed in Section 4.2.1, below. Significant improvements can also be
obtained through the use of new, well designed burners and combustion chambers. These, and other
potential control methods are discussed in subsequent sections.
4.2.1 Burner Maintenance
4.2.1.1 Burner Tuning or Replacement
Several investigators have shown that particulate emissions can be reduced significantly by ad-
justing, or tuning, existing burners and by replacing worn out components or the entire burner.*
One team of investigators studied the emissions from 33 residential units during the 1970 to
1972 heating seasons (Reference 4-17). Pollutant emissions were measured in the "as found" condi-
tion and then after proper tuning, which included:
Cleaning and adjusting the ignitor electrodes
t Cleaning the blast tube and blower shell
Cleaning or replacing the nozzle
0 Cleaning or replacing the oil filter
Sealing air leaks at the inspection door, around the blast tube, or at other easily acces-
sible locations
The relation between smoke level, gaseous pollutants and C02 concentration in the stack are shown
in Figure 2-3. All these levels are affected by tuning.
4-16
-------
Changing the draft regulator setting (replacing regulator if necessary)*
Three of the 33 units tested were found to be in such poor condition that they had to be replaced.
Table 4-2 summarizes, the reduction in mean pollutant emissions that could be accomplished by
the following steps (assuming a typical distribution of 3 units in poor condition out of each sample
of 33 units):
',. 1. Identifying and replacing the, units obviously in poor condition
2. Completing Step 1 and, in addition, tuning the remaining units
Tuning of the serviceable units reduced the mean values of Bacharach Number considerably as
indicated by Figure 4-3. For the "as-found" condition, the mean Bacharach Number was 3.2. After
tuning, this number was reduced to 1.3. Moreover, all of the serviceable units could have complied
with a standard of Bacharach No. 3 and nearly 90 percent could have even complied with a level of 2.
Tuning produced little change in NOX, while filterable particulates and HC emissions decreased only
slightly (see Table 4-2). The relationshop between smoke emissions and C02 concentration in the
exhaust gas is shown in Figure 4-4. This plot shows that 11 of the 12 units tested could achieve
a smoke level of Bacharach No. 2 with C02 concentration greater than 8 percent (Bacharach No. 2 is
the Maryland regulation, and they recommend that it be achieved at C02 levels of 8 to 10 percent -
see Chapter 6). Similarly, three-quarters of the units (9 out,of 12) could achieve either a Bacharach
No. 1 at 8 percent C02 or a No. 2 at 9 percent CO,,. Since thermal efficiency is directly related to
excess air (and, hence, indirectly to C02) and stack temperature, the C02 concentrations can be
viewed as a measure of efficiency if one assumes constant stack temperature. On this basis, 8 percent
C02 represents about 77.5 percent efficiency and 9 percent, 79.1 percent for a typical stack tempera-
ture of 500°F (260°C).
Two other studies also concluded that burner tuning produced a marked reduction in smoke emis-
sions (Reference 4-1, 4-18). In one of these studies, the mean Bacharach Number of 200 burners was re-
duced slightly more than 40 percent from above 2.0 before adjustment to 1.1 afterwards (Reference
4-1). Reference 4-17 also showed that tuning increased the overall thermal efficiency for 8 of the
13 units tested. The average increase for all 13 units was 1.7 percent. This average increase is
similar to the one reported in Reference 4-1 (from 78.2 to 79.2 percent) and discussed earlier
(Section 4.1).
Periodic replacement of the filter for the air that is circulated through the house will affect
particulate emissions indirectly by reducing the loss in furnace efficiency due to the restricted
air flow.
4-17
-------
3
a
>=!
o o E
s- £
Q. CU '
11 Q
C3. .
sr~-
>* r
CJ I
cu
(J
s-
c
en i o 2
in co en rj
A
gp!
o o
i" z:
-i C3
s: o
ijj _
gg
UJ O
E:Q-
TZi ~
o z:
cu
-(J
(O
1
s-
O-
a>
cu
4-18
-------
As found
Tuned 0
s
Mean Values
As Found
Tuned
O0Q3
0OO
O3
GOOGOQOCD
60
«u
Percent of units having emissions less than or equal to stated value.
(Mean values exclude units in need of replacement)
Figure 4-3. Distribution of smoke emission for
residential units (Reference 4-17).
4-19
-------
Efficiency computer
on basis of 500°F
stack temperature
7 8 9 10
Efficiency, percent
Figure 4-4. Composite of smoke emissions from
tuned residential units as a func-
tion of efficiency and C09 in flue
gas (Reference 4-17).
-------
The value of a burner maintenance program is shown by the experience of the Maryland Bureau
of Air Quality Control, which conducted a survey of over 400 residential burner units in the Balti-
more and Washington Metropolitan Areas during 1972 to 1973 (Reference 4-10). They found that only
42 percent of the units had Bacharach Numbers less than 2 (the standard according to their regula-
tions), and that the average thermal efficiency was about 70 percent. The poorest burners were
only 54 percent efficient, whereas the best achieved 81 percent thermal conversion. However, in-
vestigators who studied the problem of particulate emissions in the Boston area concluded that most
residential and commercial units in that area received good maintenance service (Reference 4-55).
This conclusion was based on contacts with oil suppliers and burner servicing companies, who repor-
ted that most residential and commercial systems in the Boston area were serviced at least once each
year (References 4-60 and 4-61). The adherence of a large number of home-owners and building land-
lords to a periodic service program is attributed to public relations compaigns by the oil suppliers .
that stressed burner maintenance for energy conservation. Since energy observation was the criterion
used to adjust the burners, some reductions in smoke emissions are still possible by requiring that
boilers and furnaces be tuned to both a smoke spot number and a C02 concentration.
These results again show the value of proper maintenance both in directly reducing emissions
and in indirectly reducing them as a by-product of decreased fuel consumption. In fact, since they
showed that all properly operated and maintained units could achieve a Bacharach Number of 3, and
nearly all could reach 2, these results imply that particulate emissions from residential sources
could be reduced to an acceptable level in many regions by a control strategy that relies on mainte-
nance procedures to identify units in need of replacement and on proper tuning for the remainder.
Usually burner maintenance service can be obtained from oil suppliers on a contractual basis.
The cost for such services is approximately $30 to $35 per year.
In order to determine how rapidly burner performance deteriorated after a burner adjustment,
follow-up measurements were made by the investigators who studied the 33 units described at the
begining of this subsection (Reference 4-17). These later tests were performed twice during the
heating season at 2-month intervals for four units. They showed that in only one unit did smoke,
CO, and HC increase. In another unit, CO and HC increased, while the other pollutants remained
constant. Emissions remained nearly constant for the other two units. However, results of tests
in a laboratory burner unit (Reference 4-20) indicated that smoke emissions increased from a
Bacharach Number of 1 to nearly 6 while NOX, CO, and HC remained essentially constant over a 10-
week period. The inconsistent results demonstrate the need for further study of the effects of
time on emissions.
4-21
-------
One other study (Reference 4-63) was conducted to determine how smoke and efficiency vary
between burner servicing?. Tuning resulted in an average reduction of smoke from Bacharach No. 1.9
to 0.9 and a 1.9 percent increase in efficiency. When the burners were next serviced, 3 months
later, efficiency had not deteriorated, but the average smoke number had increased to 1.8.
The study just referenced above also concluded that the importance of annual maintenance is
not so apparent if one only looks at emissions and efficiency of a well maintained burner before
and after a given servicing. However if the unit is not serviced annually, the annual decrease in
performance becomes additive until very poor operation is experienced. One reason is that smoke
implies the presence of soot, some of which adheres to the heat transfer surface. After a boiler
has operated for some time with even a moderately low level of smoke, the accumulation of soot will
reduce its efficieny.
Periodic servicing is, therefore, a type of preventative maintenance that keeps the equip-
ment operating near its optimal level and, hence, minimizes emissions. It also avoids system failures
during the heating season.
4.2.1.2 Training of Servicemen
In light of the importance of proper burner maintenance in reducing air pollution in many
areas, up-grading the competence of servicemen and supplying them with good diagnostic instruments
can be considered a particulate control technique. Unfortunately it is difficult to assess the
effectiveness of a program which requires that all oil burner servicemen receive a certain training
and/or use specified instruments to adjust the burners. However, it is the uniform opinion of air
pollution control authorities that the quality of service currently available varies from very good
to very poor; therefore, they all believe such a control strategy would help. However, they con-
sider it to be either unenforceable or not cost-effective, and none of them have instituted such
a program.
Training programs for maintenance personnel are available in some localities where oil is in
heavy use. These programs are usually offered in community colleges or trade schools. A few pri-
vate or trade organizations in the United States also specialize in setting up training programs and
preparing textbooks for oil burner servicemen. One such organization, the Petroleum marketing
Education Foundation, based in South Carolina, conducts training programs throughout the United
States. These programs are usually .sponsored by trade organizations. Class durations run from two
days to one week. Tuition is approximately $15 per day per student, including textbooks. The most
popular program run by this organization is a 3-day course in which the servicemen are trained to
adjust burners for high efficiency and low smoke.
4-22
-------
The trainees in the above program are encouraged to utilize proper instruments when adjusting
burners instead of merely "eyeballing the flame" (Reference 4-21). Instruments that can be used to
measure C02, CO, Bacharach No., stack temperature, and draft press'ure are available commercially at
a cost of $300 to $500 for the set.
One of the documents in the burner maintenance and adjustment guideline series currently being
developed by EPA is concerned with residential oil burners (Reference 4-29). This guideline is in-
tended to be both a service guide and a training guide'.
4.2.2 Burner and Combustion Chamber Redesign and Retrofit
Fundamental combustion research has stimulated new developments in small oil burning equip-
ment, especially as applied to residential space heating using No. 2 distillate. Both the govern-
ment and industry have sponsored programs to develop low pollution and high thermal efficiency re-
sidential oil burners. Some of the development efforts have resulted in commercially available units,
while others are still in the experimental stage.
Improved systems designed for new installations will be described below; possible retrofits
to existing systems will be discussed in the following subsection.
4.2.2.1 Burner and Combustion Chamber Redesign .
New burner developments center mainly on the improvement of oil atomization and on the optimi-
zation of the combustion aerodynamics, particularly by adaptation of the flue gas recirculation con-
cept.
Combustion Aerodynamics
Reductions in smoke and other pollutants have been achieved by improved design of the flow
patterns in the burner and combustion chamber (see Section 4.1.4.2). Most of these systems rely on
optimal choices for swirl air velocity to obtain the desired turbulence level and residence time. In
addition, many use either internal or external flue gas recirculation (References 4-24 and 4-25).
One such development is now available commercially from the Blueray Company. This "blue
flame burner" is sold in a warm air furnace in two firing ranges, 0.6 gph and 0.75 gph (0.63 - 0.79
cmVsec). Combustion gases amounting to 50 percent of the stoichiometric air requirement are re-
circulated internally to produce smoke-free, blue-flame combustion with essentially zero excess air.
Most of the incandescent carbon particles are eliminated during the reburning. Because this burner
has to be matched to the combustion volume to obtain the desired (internal) recirculation pattern,
it comes only as a burner-furnace package.
4-23
-------
About 300 units have been sold so far (Reference 4-26),, but they are expected to be widely
available by mid-1976. In laboratory tests, the thermal efficiency of the burner has been shown to
be over 80 percent (Reference 4-27). A Bacharach Smoke Number close to zero, and NOX emissions just
over 20 ppm were achieved over a range of 11 to 14 percent COg. In addition it is purported to be
quieter and smaller than comparable forced air heating systems.
The CO concentrations during ignition were'1200 to 1300 ppm (Reference 4-28). These high CO
levels occur for about 3 to 4 seconds. After this initial ignition period, CO'emissions drop down
to about 30 ppm, which is normal for oil burners. Although these high peak CO emissions are not
desirable, the benefits of reduced NOX and particulates, increased efficiency, quieter operation,
and reduced size far outweigh'this disadvantage.
The cost for the burner package is about $550 excluding installation, or approximately 10 per-
cent more than a conventional oil-powered warm-air furnace (Reference 4-27). This cost differential
is quickly recovered through fuel savings and, in a new home, through lower costs for sound insula-
tion and building space to house the unit.
The utilization of a flame retention device to create.a more stable, compact, and intense flame
is another area of development. Many such units are available currently. Studies have shown that
most flame retention burners perform better - i.e., at higher efficiency with lower smoke and/or NO
X
emissions than do those with a conventional burner geometry. One of the flame retention units
tested in Reference 4-20 emitted almost 60 percent less smoke than the conventional unit (average
Bacharach No. of 1.2 for the retention head versus 2.9 for the others). In addition, thermal
efficiency was higher for the retention head unit (83.0 percent versus 75.0 percent). A set of field
test results (Reference 4-17) also showed that smoke, particulate, CO and HC emissions were lower
for retention head burners than conventional ones, while NO emissions were essentially the same (see
Table 4-3).
4.2.2.2 Retrofit of Existing Burners
Proper burner maintenance has already been shown to be effective in reducing particulate emis-
sions from residential burners (see Section 4.2.1). As with the commercial sector (Section 4.1.2.2),
the retrofitting of existing units which are difficult to maintain or defective in design is another
method of reducing particulate emissions. These retrofits generally involve replacement of the
burner units and controls at a cost of about $300, including parts .and labor. However, sometimes
commercially available combustion improving devices are added to the existing burners for the purpose
of improving the mixture of air and fuel.
4-24
-------
ro
E
a.
o
CO
z
C3
to
UJ
a
15
3C
I
Z
O
§
CJ
oo
rs
o
l-H
o;
CO
CO
2:
o
II
CO
,_>
"-3
Dl
C
=3
CO
s:
CO
o
ro
Lu
C
O
CO
CO
r
UJ
O
CO
o
CO
CO
UJ
S
re
o
+J
5-
rO
a.
o
CO
CO
e
UJ
rs
o
CO
re
CU
-C
<_
£
JZ
lv
re
CC
M-
s.
fl
_c
E
z
10
o
l^
CU
J3
re
S-
aj
U_
CM
^ CO
re
0
1
J-J
o
° =
^ E:
jji u>
£«
00
+31"
; C OO
\ rD
JD
C
o
I
-a
cz
o
CJ
c:
o
CO ro
J3 3:
S
O
CJ
VO OO **~ ^O ^"^ C3
O ^D LO O CM
CM i i CM f^. i .
co en uo oo co oo
»^ CO OO * i CVJ
ooo ooo
o o o o o o
, . ^^ . . . .
' » co oo r-^ r^
o co r-~ --^ co oo
... 10
en vo ^t- oo oo ill
. -_x ^_, CO * ~-- I I I
i to r ' cn en
CM i r CM O O
OOO OOO
O O O C3 O CD
cocncvj r~~oor rocxico
cncoo cocooo ocnco
^- r CM CM i r- CMrr-i
ocnoo cnocn 5r^rrJ
r-^toto LDIOOO COCOLO
OOCD OOO OOO
cncnoo cncni covoi
cnOLO inoi ID i i
i-. co i ror-~o OICDCO
OOCMCM iii CD i CD
I^OOCO COOOCO COCOCO
1 '
a) . £ > ("" Frr > 1~~ C
EOre crjjto CCUfc
rSo^LZ cjood cjcou-
co
ro
c
0
CJ
to
c;
o
CO
+J
c
^
to
CO
CU
+J
o
"O
CU
CO
3 -
CU
t-
cu
c
cu
S-
4-
OJ
.
c a>
cu s-
E: 13
cu ~a
o cu
ro O
i O
0- S-
CU D-
S- +->
=t C
4- 0. T-
O UJ O
a.
, CU CU (O
| CU !- 4->
e- c: 4- re
r- -O
CU C T3
U -r- 0 >i
CU 't~ -Q >^
CU 3 1) C
rv 4-> O
cn ro
r- "5 O
CU "O CJ
O 3 .r- -O
s_ ^_ +j cu
3 CJ S- CO
OX ro ro
C/1 UJ D_ CQ
ro -Q O T3
4-25
-------
The ability of combustion-improving devices to reduce emissions and increase thermal efficiency
has been investigated (Reference 4-20). When added to existing burners, these devices modify the
combustion aerodynamics in the furnace. Of the five devices tested on a standard, high-pressure
atomizing burner, only one, the flame retention head, was found to reduce smoke emissions significantly
in comparison with the standard burner (see Figure 4-5). At a stoichiometric air-to-fuel ratio of
1.6 (minimum smoke for the retention head), the reduction was from slightly more than Bacharach No. 1
to almost 0, while the NO emissions were unchanged. More importantly, the retention head could
achieve a smoke reading of No. 1 at an air-to-fuel ratio of less than 1.2 and No. 2 at 1.06, where-
as the standard burner could only reach these levels at 1.64 and 1.5, respectively. Therefore,
efficiency could be increased from 76.6 percent to 83.0 percent with this flame retention head.
However, none of the other four devices gave the same encouraging results.
4.2.3 Emerging Technology
Current research directed toward reduction of particulate emissions from residential oil bur-
ners is aimed primarily at the following areas (excluding modifications to the fuel): improved
atomization, optimization of the combustion aerodynamics, system use patterns, and modifications to
the combustion chamber length and material. These are discussed below.
Improved Oil Atomizers
The high-pressure air atomizing gun-type burner is the most common one used for residential
heating. Even though the basic principles and component functions have generally remained the same,
marked advances have been made recently in gun burners. These improvements have resulted in more
compact burners, having improved efficiency and flame stability. The improved efficiency reduces
fuel consumption, which in turn, should reduce the total annual particulate-emissions. The improved
flame stability allows more uniform burning and, hence, fewer localized cold spots or temporary fuel
rich zones where smoke or particulates are formed.
Ultrasonic and acoustic atomization, as discussed in Subsection 4.1.4.1, are two schemes
presently under development. Another is the system used in the "Babington Burner" (Reference 4-62),
financially backed in part by the National Oil Fuel Institute (NOFI), the trade association for
home heating oil dealers. In this novel atomizer, liquid fuel washing over the outside surface of
a small glass or plastic bulb forms a thin surface film. Air forced through one or more slots
in the bulb breaks the liquid into a fine spray. Proponents of the burner claim uniformity of par-
ticle size, elimination of clogging, and ability to burn a variety of fuels (e.g.: diesel, distillate,
kerosene, etc.) as advantages over conventional burners. Fifty of the units have been field tested
and reportedly produced low smoke levels, but no numerical emission data are available.
4-26
-------
DEVICE
Standard ABC
,._i -Monarch Combustion Head
Delavan Flamecone
Shell Head
Gulf Econojet
.-Union (Pure) Flame Retention Head
1.2 1.6 2.0 2.4
Stoichiometric Air/Fuel Ratio
Figure 4-5. Average smoke emissions of combustion improving
devices versus Stoichiometric ratio (Reference
4-20).
4-27
-------
Vaporizing Burners
Vaporizing burners have been used with kerosene and No. 1 heating oil for many years.. How-
ever, with No. 2 oil vaporization may not be complete leaving oil deposits on the metal surfaces.
The deposition rate can sometimes be as high as 200 grams of carbonaceous matter per cubic meter of
oil consumed, and this material can later be emitted as particulate. A vaporizing burner for No. 2
oil that could overcome this problem would be of considerable interest because of its potential ad-
vantages: simplicity, quiet operation at low-capacity, low smoke, and high efficiency.
A prototype vaporizing burner for No. 2 distillate oil has been developed and tested (Refer-
ences 4-22, 4-23). Oil is introduced into a hot cast steel chamber where it vaporizes quickly. The
vapor is then conducted to a combustion chamber, where it is mixed with air in a swirl-type burner.
Such burners have performed satisfactorily on No. 2 oil for periods equivalent to a heating season,
firing 0.5 gallon per hour (0.53 cm3/sec) on an intermittent basis. Installed in a storage type
water heater, one of these units logged over 1000 hours firing time, and gave satisfactory combus-
tion performance of 12 percent C02 with 0 to 1 Bacharach Smoke Number.
Combustion Aerodynamics
An improved burner design for residential units evolved from the study discussed in Subsection
4.1.4.2 to design an optimum distillate oil burner for retrofit into existing warm air furnaces (Ref-
erence 4-10). In addition to the 9 gph (9.5 cm3/sec) burner described there, a 1 gph (1.1 cm3/sec)
was also designed on the basis of the tests with existing units. The burner was fired in a refractory
lined, coaxial combustion chamber. Smoke emissions were less than Bacharach No. 1 and only negligible
quantities of unburned hydrocarbons and carbon monoxide were produced. Mass particulate emissions
were not measured. Nitric oxide emissions were about 1 mg NO/g fuel burned (about 50 to 60 ppm),
compared with nominal levels of 1.5 to 2 mg NO/g fuel from typical new commercial burners fired into
the same combustion chamber. The NO emissions increased to 2.0 mg/g fuel at a smoke level of Bach-
arach No. 1 when the burner was positioned at right angles to the combustion chamber. This geometry
is commonly found in residential furnaces. Other commercially available burners either smoked or
emitted about 3 mg NO/g fuel when fired in the same facility. Additionally, the "optimum geometry"
burner was capable of operating (in the laboratory) at only 5 to 10 percent excess air without smoke,
whereas typical new commercial burners required up to 25 percent excess air to achieve smoke free
operation.* Simulated field tests were conducted in a commercially available furnace for a total of
128 hours of cyclical on and off operation. The burner performed over the entire test period with
*Burners in the field usually need to be set at 10 to 20 percent more excess air than in the labo-
ratory to avoid sooty operation due to dirt accumulation on air blower fans and the like. Never-
theless, the optimized burner can still run with less excess air than other currently available
units, and these, in turn, are an improvement over the home heating systems that were installed in
the past, which require up to 60 percent excess air for smoke free combustion.
4-28
-------
little variation in performance and no noticeable degradation in emissions. The stoichiometric ratio
was about 1.10, Bacharach No. 0, and NO emissions 0.95 mg/g fuel. This burner should be available
commercially within 1 to 3 years.
Results from this study also suggested that an optimally designed burner which was specifi-
cally matched with a combustion chamber could achieve even lower NO levels at higher thermal effi-
ciency with no increase in smoke. This new system is being designed with a sealed combustion air
and draft control system that uses outside air'instead of previously heated indoor air. The result-
ing conservation of heated indoor air should reduce fuel consumption. Work is currently underway to
develop this matched system. Such a system may be commercially available within 6 years.
Burner Use Pattern
One of the major factors contributing to high particulate emission, particularly in domestic
burners, is cycling. The on-off mode is a dominant characteristic of warm air furnaces, and its
importance as a cause of increased emissions has been well documented (References 2-6 and 2-15).
A typical furnace cycle is shown in Figure 4-6. The burner is -ignited at a point off the
chart and after a minute or so, the blower starts circulating air over the heat exchanger. Four
minutes later the burner is extinguished, and about 5 minutes after that the blower stops. Inves-
tigations on a model residential heating system indicated that the sizeable peak emissions measured
during ignition and shutdown can account for most of the total emissions. Figure 4-7 shows qualita-
tive emission traces from an oil burner during a typical cycle. CO and HC emissions peak at ignition
and shutoff. HC concentration drops to insignificant levels between the peaks, while CO emissions
tend to flatten out at a measurable level. Particulate emissions continuously taper off after the
ignition-induced peak, whereas NO emissions first rise rapidly for a short period and then continue
to rise at a more moderate rate as the combustion chamber temperature increases. The operating time
of most domestic burners seemingly is not long enough for NO to reach equilibrium levels.
The transient emissions are caused mainly by variations in the combustion chamber temperature.
At ignition, a cold refractory will not assist complete combustion and, therefore, peaks of CO, HC,
and smoke can occur. In older units the oil nozzle tended to dribble during ignition and shutdown
due to momentary low atomization pressure. This dribble is composed of large droplets which result
in high particulate emissions. The problem has been minimized in new systems by the use of solenoid
valves in the fuel tube (delayed on - instant off), the placement of this valve nearer to the burner
tip, and the use of smaller diameter tubes.
4-29
-------
i.
CD
O1
C
o
X
O)
fO
O)
<=c
Area represents
unused heat
Extrapolated
to zero
temperature
rise
15
20
Blower time in minutes
Figure 4-6. Temperature rise across an oil-fired warm air
furnace heat exchanger during a typical cycle.
4-30
-------
Filterable
Particulate
NO
Burner
On
Burner
,0ff
N
vfi
Time
Time *
HC
Burner
On
Burner
Off
CO
Burner Burner
|0n I Off
Time
Time
Figure 4-7. Characteristic emissions of oil burners
during one complete cycle.
4-31
-------
It should also be mentioned here that cycling has a deleterious effect on overall efficiency.
As indicated in Figure 4-6, the fire is extinguished at the peak temperature rise across the furnace,
and it is at tljis point that the maximum certified efficiency is probably achieved. For the remainder
of the cycle the air blower remains on, but the flue gas temperature and flow rates through the
burner have dropped to zero. Perhaps what is more important, and not shown on this curve, is that
the burner has been firing for 1-3 minutes prior to blower start-up. Again, this means that there
is a significant period of time when there is very little flow on the air side, implying ineffec-
tive heat exchange. Of course, some of this heat is stored in the furnace structure to be recov-
ered eventually, but much of it escapes out the stack under inefficient conditions.
Peak smoke emissions resulting from the on and off cycle of boiler and furnace operations is
a more serious problem for the residential sector than for other sectors (see also Subsection 4.1.2.2).
Nevftrtheless, modulating burners are rarely used in residences because the peak oil consumption rate
is usually less than 3 gph (3.15 cm3/sec). To modulate these burners, they would have to perform
effectively at the extremely low oil consumption rates characteristic of low load firing. Much more
work would be needed to make burner modulation applicable to residential burners. Hopefully, the
continuous development of the newer burners, such as the blue-flame (Reference 4-27) and optimum
burners (Reference 4-10), could minimize the cyclic emission problem.
Combustion Chamber Size and Material
Residence time affects particulate and NO emissions from residential boilers and furnaces
just as it does commercial units (see Subsection 4.1.2.1). Figure 4-8 shows particulate emissions
as a function of excess air from two refractory-lined combustion chambers, one a 15-inch (380 mm)
chamber and the other a 27-inch (680 mm) chamber. The longer chamber was created by adding a 12-
inch (300 mm) refractory-lined section between the end of the 15-inch combustion chamber and the
metal heat transfer surfaces. The increase in chamber size effectively increases the residence time.
This figure shows that particulate emissions can be reduced substantially if a short residence time
unit is enlarged. For example, at an air-to-fuel ratio of 1.6 the reduction is ten-fold, from 0.20
to 0.02 mg/g fuel. The emission rate from the long chamber is less than one-quarter the rate from
a well maintained currently installed unit which has^been retrofitted with a flame retention device
to improve combustion and reduce particulate emissions (see discussion of flame retention devices
above). The curves on Figure 4-8 also indicate that the particulate emissions from the long chamber
at typical stoichiometric air-to-fuel ratios (i.e., between 1.45 and 1.85) are below the lowest
levels attained by the short chamber, even when the latter is operated under very lean conditions.
4-32
-------
Short residence time (15-in. chamber)
Long residence time (27-in. chamber)
0.01
1.2
1.4 1.6 1.8 2.0
Stolchiometric Air/Fuel Ratio
2.2
2.4
Figure 4-8. Effect of residence time on participate emissions
(Reference 4-20).
4-33
-------
Figure 4-9 shows the NO emissions from the two chambers. At a stoichiometric air-to-fuel ratio of
1.6 (i.e., minimum smoke for the longer unit), the NOV emissions are slightly higher for the larger
A
chamber (1.25 mg/g fuel) than for the shorter one (1.1 mg/g fuel). Due to space limitations in homes,
it may be difficult to apply this technique to commercially available systems. Moreover, data on
emissions under cyclical operation are not available.
The material used in the combustion chamber has been shown to affect smoke emissions (Refer-
ence 4-20). When a burner was fired into two chambers which"were similar except for liner material,
the refractory-lined chamber gave lower smoke readings than the steel-lined one. It was assumed that
the steel created a cold wall effect which quenched the flame before combustion was complete, thus
producing excessive amounts of CO, HC, and smoke. Hence, refractory-lined combustion chambers are
preferred for lower pollutant emissions, at least in steady-state operation. The potential use of
ceramic liners as retrofits to existing units is discussed in the commercial section (4.1.2).
4.3 INDUSTRIAL AND UTILITY
In the industrial and utility sectors fuel accounts for 80 to 90 percent of the annual opera-
ting and maintenance costs. Therefore, the large boilers in these sectors are generally well-maintained
to minimize fuel consumption, and state-of-the-art burner/combustion chamber design is used in new sys-
tems. Hence, particulate emissions from these sources are generally controlled by collection devices
in the exhaust stream.
A recent study on particulate control strategies for the Boston area, however, has concluded
that reductions of up to 30 percent could be achieved on industrial and utility boilers by more fre-
quent inspection and improved maintenance procedures for boilers larger than 30 MBtu/hr (8.79 MW)
(Reference 4-55). This strategy was chosen over fuel switching, fuel washing, and the purchase of
steam from a central station for residential and commercial space heating on the basis of its techni-
cal feasibility, acceptable capital and operating cost increases, projected fuel availability, and
enforcement requirements. The basis of this strategy is the set of procedures listed in Table 4-4.
This table was an attachment to a compliance schedule submitted by the Boston Edison Company to the
Metropolitan Boston Air Pollution Control District in support of their claim that their boilers were
in compliance with particulate loading limitations.* The key, unique features of this program are
their commitment to continuous sequential operation of soot blowers (Item IV) and their program to
*They stipulated frequent testing of the oxyger concentration in the flue gas rather than continuous
monitoring (see Item I) because they felt the oxygen monitors which were available at that time
were not reliable enough to use (Reference 4-57). Many large boilers, however, are equipped with
automated air-to-fuel ratio controllers which respond to stack measurements of oxygen concentration
and smoke.
4-34
-------
2.0
Long residence time (27-in. chamber)
- Short residence time (15-in. chamber)
CH
I
<
1.2
1.4 1.6 1.8 2.0
Stoichiometric Air/Fuel Ratio
2.2
2.4
Figure 4-9. Effect of residence time on nitrogen oxides emissions.
(Reference 4-20).
4-35
-------
TABLE 4-4. PROCEDURES FOR CONTROL OF PARTICIPATE EMISSIONS
FROM THE STACKS OF THE BOSTON-EDISON COMPANY
I.
II.
Ill,
IV.
V.
VI.
Frequent testing to maintain proper fuel/air
ratio for combustion.
Purge cleaning of fuel-oil burning guns at
each shutdown.
A. Routine disassembly and cleaning of
fuel oil burning guns to determine
possible plugging of orifices.
B. Renew worn parts of fuel oil burning
guns as soon as wear is noted.
Complete overhaul and repair of all fuel oil
burning equipment and all flue gas passages of
the boiler, during every annual outage for
inspection.
Continuous sequential operation of soot
blowers.
Fire sides of boilers, boiler duct work and
ash hoppers are to be cleaned at periodic
intervals during the year.
The frequency of cleaning will be de-
termined by the results of the test
program described in Item VI.
A program of periodic testing of particulate
emissions for representative Boston Edison
boilers is under development. The object of
this program is to determine the frequency
of cleaning necessary to maintain particu-
late emissions in compliance with the Bureau
of Air Quality Control's regulations.
4-36
-------
develop a maintenance schedule designed to maintain all units in compliance (Item VI). Soot blowing
is discussed in Subsection 4.3.2.1 below. Since the submission of this plan, they have adopted a
schedule of boiler cleaning (especially Items II and V) every 3 months (Reference 4-56). This inter-
val is based on'tests of emissions versus time since the last cleaning. These measurements showed
that the particulate emission rate increased with time and approached the regulatory limit about 3
months after the previous cleaning.
In general, some control over particulate is available by adjusting the air-to-fuel ratio, but
other factors constrain the use of this technique. If low sulfur fuels are burned, the particulates
consist mainly of ash (noncombustibles in the fuel) and smoke (partially burned fuel). Boilers which
are fired with such a fuel are generally adjusted to the lowest air-to-fuel ratio at which no visible
smoke is produced. This practice has the dual benefit of minimizing both fuel consumption and NOX
emissions. Although it is desirable to operate at an air-to-fuel ratio that represents about 3 per-
cent excess air, many industrial and some utility boilers still require up to 15 percent excess air.
New, better designed burners would have to be used in these boilers to permit the use of less air.
When high sulfur oil is used, the particulate can take on a different character with the pres-
ence of significant quantities of SO.,. These sulfates not only increase the rate of deposit accumu-
lation on the heat transfer surfaces but also leave the stack as a visible plume (References 4-6 and
4-12). Virtually all the sulfur in the fuel is converted to S02 or S03, and the ratio of S03 to S02
increases with available oxygen. Therefore, low excess air firing is a necessity when high sulfur
fuels are used. This requirement, more than the desire to reduce NO and increase thermal efficiency,
A
was, in the past, responsible for the development of burners that could combust fuel with only 3 per-
cent excess air.
Burners for packaged boilers have undergone almost continual refinement in the past 15 years
to narrow the flame further for higher-capacity furnaces and to optimize efficiency and reliability.
These burners now provide for complete burnout of combustibles with low excess air. (In addition to
optimizing boiler performance, low-excess-air operation helps maintain emissions of nitrogen oxides
at low levels.) However, high boiler efficiencies cannot be achieved without close control of the
combustion process. Vendors generally recommend maintaining excess air in packaged oil-fired steam
generators between 10 and 15 percent. Although efficiency can be increased by dropping excess air
further, operation at lower levels requires extremely close attention on the part of operators to
prevent incomplete combustion, and the release of substantial quantities of soot, during rapid changes
in boiler load. Industry's drive to conserve fuel and minimize pollution control problems also has
4-37
-------
led to the development of burners that can handle a wide variety of waste fuels - tar, pitch, hydro-
gen, naphtha, coke-oven gas, carbon monoxide, sander dust, refinery gas, sewage sludge, etc. (Power
Special Report, >Power from waste, February 1975).
To summarize, when low sulfur oils are burned, excess air can be used to eliminate any visible
emissions. However, this is not a useful approach because it also causes NO emissions and fuel con-
X
sumption to increase. Moreover, when high sulfur fuel is used, low excess air combustion (e.g., 3-5
percent) is necessary to minimize particulate. Therefore, boilers should be equipped with burners
that are capable of operating at low excess air to minimize simultaneously particulates and fuel con-
sumption.
Since the smaller industrial units span the same size range as the larger commercial ones,
they show many of the same operational characteristics. Therefore, control strategies suggested for
the commercial units, such as improved maintenance procedures, could be applied to these industrial
units. In the following subsections, only the control techniques for the larger boilers are discussed.
4.3.1 Particulate Collectors
Four basic types of particulate collectors are available for industry and utility applications:
electrostatic precipitators, mechanical collectors, bag filterhouses, and wet scrubbers. Of these,
the electrostatic precipitator and the multicyclone, which is one type of mechanical collector, are
used in oil-fired facilities. Bag filterhouses and wet scrubbers are generally not used at this
moment.
Particles that are emitted from large oil-fired units range in size from 0.1 to over 10 microns.
If the burner is well maintained and properly adjusted, total particulate emissions should be low, but
a substantial portion of those emitted will probably be less than 1 micron. However, it is common to
find oil-fired sources which are not serviced frequently enough and whose emissions, therefore, are
dominated by the larger particles. The plume from such a boiler would normally not be visible unless
it was the discharge from a very large boiler with a correspondingly large diameter stack. Fine par-
ticles can cause the stack discharge to be visible because their size approaches the wavelength of
visible light, and hence enables them to scatter light effectively. Particulate collectors have been
used to minimize this problem (References 4-30 and 4-31).
"l
Since particle size is an important factor in assessing the impact of emissions on various
receptors, one needs to consider the relationship between collection efficiency and particle size when
one discusses a control system. As shown on Figure 4-10, this efficiency decreases rapidly as the
particle size approaches the submicron range.
4-38
-------
I
0)
i-
> o
o c
'
t- S_
O O
O O
a) a)
4-39
-------
The following subsections describe each of these participate, or dust, collection devices,
give their control effectiveness, and present some cost information.
4.3.1.1 Electrostatic Precipitators
Electrostatic precipitators have been used by the electric power industry for coal flyash
collection since 1923. Their use in this application has increased rapidly since 1950, from about
80 x 10s cfm (3.8 x 10" m3/sec) installed capacity then to over 500 x 106 cfm (2.3 x 10s m3/sec) in
1970 (Reference 4-35). Although they are now the most common collectors on oil-fired boilers, this
application accounts for less than 10 percent of the total installed capacity.* Their predominance
among control devices for oil-fired units can be attributed partly to the fact that many boilers
which were at one time coal-fired have since converted to fuel oil. They originally were equipped
with a precipitator to abate coal flyash and have kept it in operation after the conversion, with
certain modifications to increase collection efficiency. In general, it is conceded that electro-
static precipitators are the most effective collectors in use today.
The operating principle consists of electrostatically charging the particulates by a corona
discharge and then passing these charged particles through an electrical field which drives them to
collecting plates. These plates are rapped periodically to dislodge the collected particulates,
which then fall into a hopper.
The efficiency of electrostatic precipitators has been described quantitatively by simplified
equations (Reference 4-30). These relationships are useful for preliminary design and systems analy-
ses purposes, for they show the impact of the significant parameters. Among these, the most impor-
tant are the effective collecting electrode area of the precipitator (A), the gas flow rate through
the precipitator (Q), the strength of the two electric fields (charging and collecting), and the
particle diameter. The A/Q ratio must be large for high collection efficiency; thus velocities
should be low and collection plates large. In addition, the collection efficiency is proportional
to the strength of the electric fields. However, experience has indicated that the effect on collec-
tion efficiency of particle size above the submicron range is small. Other factors that influence
the precipitator performance are the gas viscosity, particle resistivity, sulfur content of the fuel,
flue temperature, carbon content of the particulates, particulate loading and ability to get particles
off collecting plates and into the hoppers (Reference 4-30).
*Based on Reference 4-35 for total installed capacity and Reference 4-36 for quantity of fuel burned
annually in utility boilers that are equipped with precipitators.
4-40
-------
For oil firing, field experience has shown that collection efficiencies between 50 and 95 per-
cent by weight can be obtained. Normally, the precipitators are designed to remove about 90 percent
of the participate emitted (by weight) (References 4-31, 4-32, 4-33). However, it is not entirely
meaningful to talk about efficiency alone because it is a function of particulate loading, increasing
as the input particle concentration rises. Reference 4-31 claims that the outlet loading from a pre-
cipitator is nearly constant at'0.02 to 0.05 Ib/MBtu (8.6 - 21.5 ng/J) independent of the inlet con-
ditions (0.05 to 0.33 Ib/MBtu (21.5 - 142 ng/J) in this case, including periods of soot blowing).
One precipitator vendor states, further, that control performance is limited more by a lower bound
of 0.005 gr/scf (0.009 - 0.011 Ib/MBtu, depending on the excess air used) than it is by collection
efficiency (Reference 4-68). For example, precipitators are in use which collect over 99 percent
of the large amounts of flyash emitted by some coal-fired power plants; equivalent units designed
for oil-fired units, whose.uncontrolled emissions are much lower, can only achieve 90 to 95 percent
collection. However, the outlet loading can be the same in both cases. At these typical efficien-
cies, well-maintained and correctly operated units should not emit a visible plume from the stack,
given the ash and sulfur content of most oils. If one uses 0.01 to 0.26 Ib/MBtu (4.3 - 112 ng/J)
as representative data (Reference 4-36) for uncontrolled emissions from utility boilers and speci-
fies a 90 percent efficient precipitator, then these boilers should emit less than 0.03 Ib/MBtu
(12.9 ng/.J). The larger boilers (>300 MW) only emitted 0.04 to 0.06 Ib/MBtu (17.2 - 25.8 ng/J) un-
controlled; therefore, they should be able to achieve the 0.01 Ib/MBtu (4.3 ng/J) lower bound from
precipitators.
Precipitators that are originally designed to collect coal flyash can experience a drop in
efficiency down to 45 percent when used on oil. Apparently the changes in particle resistivity,
size distribution, and surface properties are the primary reasons for this reduction. Therefore,
units which were installed on coal-fired boilers should be modified if the plant is converted to
oil. With proper modifications, efficiencies approaching 90 percent can be achieved. However,
once a precipitator which was designed for coal has been used on an oil-fired source without modi-
fication, its efficiency is reduced even when switched back to coal (Reference 4-36).
Costs of precipitators in 1973 are shown on Figure 4-11(a). These figures are for new units
to be used on oil-fired sources and for the modification to change from coal to oil firing.
Erected costs are highly variable, especially on existing installations. For new installations,
the erected, or installed, costs run about 1.7 times the FOB costs. Installed 1975 costs for both
medium and high efficiency precipitators (all applications, not just oil-fired boilers) are shown
in Figure 4-ll(b). According to Figure 4-ll(a), the purchase cost of ESP's designed for oil-fired
4-41
-------
2.50,
Coal to Oil
Firing Modi
0.00
50 TOO 150
GAS VOLUME, ft3/m1n/lOOO
200
250
Figure 4-11(a). Electrostatic precipitator cost comparisons for
90 to 95 percent collection efficiency (1973
dollars see Reference 4-31).
4-42
-------
1/1
o
CJ
o.o-
I.U
.10-
1H
Indicates
data spread
High
efficiency
Medium
efficiency
»
X
-
x
^*
x
^-
X
X1
X
X
X
X*
x'
X
X
X
X
X
X
/
X
>/
/X
x'X
^
.01
0.1 1.0
Gas volume through precipitator, TO6 acfm
10.0
Figure 4-11(b).
Electrostatic precipitator installed costs in 1975.
Costs based on all units 'sold; ESP's for oil-fired
boilers cost about 50 percent more (Reference 4-69),
4-43
-------
units is about 50 percent more than that for ESP's used on coal units; hence the installed costs
shown on Figure 4-11(b), which are heavily weighted by the preponderance of units sold to coal-fired
boiler operators, should also be multiplied by 1.5 when used to calculate costs for oil-fired units.
Operating cost is very low for electrostatic precipitators, amounting to about $0.03 to 0.05/yr/cfm
(Reference 4-31). Likewise, maintenance costs are about $0.03/yr/cfm. Thus as shown by the calcula-
tions in Appendix D (page D-3), costs for high efficiency units are about $0.8 x 10" /lb steam ($1.8
x 10 /kg steam) for an industrial boiler with 125 MBtu/hr (36.6 MW) heat input (lower limit of data
on Figure 4-11 and corresponding to about 100,000 Ib steam/hr) and 0.008<£/kw-hr fora utility system.*
Some recent data suggest that these costs may be as high as 0.02£/kw-hr for large oil-fired boilers
(see Appendix D). This means that the high efficiency precipitator cost represents nearly 2.5 percent
of the fuel costs for the industrial boiler and 0.4 to 0.8 percent of the typical charge for electricity
to a consumer (5
-------
tangentially or axially over swirl vanes (see Figure 4-12) thus creating a high velocity in the cylin-
drical portion of the device. Particulate matter, which is forced to the outer wall by the centrif-
ugal forces, drops to the bottom of the cyclone and into the hopper. Axial collectors and centrifugal
separators are both variations of mechanical cyclone collectors. These devices can also be built as
multiple-cyclones, which consist of a number of small-diameter cyclones operating in parallel with a
common gas inlet and outlet.
Typical efficiencies of the mechanical collectors are 75 to 85 percent by weight for particles
10 microns or larger (Reference 4-32). This efficiency is much lower for small particles. Therefore,
the cyclones would be ineffective in oil burning installations if all burners were properly maintained
and adjusted because small particles account for more than one-half of the particulate emissions by
weight from such sources (References 4-30> .4-31, 4-37, and 4-38). However, if the burners receive
only nominal attention, so that many larger particles are also emitted, or if the pollution problem
is due to acid smut alone (typically large particles), and not opacity, a mechanical collector can
provide adequate control. It is also useful in capturing the large particles that dominate the dis-
charge during soot blowing.
The draft loss for mechanical collectors in normal boiler operation ranges from 2 to 5 inches
of water. Capital, maintenance, and operation costs are approximately one-fifth of the correspond-
ing costs for electrostatic precipitators (Reference 4-40).
Mechanical collectors have been used both before and after electrostatic precipitators. When
installed ahead of the ESP, the purpose is to reduce the load on the precipitator, and when used
afterwards, it is to catch re-entrained material that was dislodged from the collecting plates dur-
ing rapping. Although these combinations are unnecessary with properly designed precipitators (Ref-
erences 4-32 and 4-41), they may be cost effective in certain instances.
4.3.1.3 Bag Filterhouses
In a bag filterhouse, the particle laden gas is filtered as it is forced through fabric tubes,
or bags. The filtering is extremely efficient and normally results in better than 99 percent re-
moval by weight of the entrained particles. The bags must be purged of the collected material peri-
odically, and the method and frequency of cleaning characterize one type of filterhouse from another
(e.g., mechanical shaker-type, reverse flow types, etc.). The pressure drop across the bags is
usually greater than 6 inches of water.
This type of particulate collector has not been used in the control of particulate emissions
from oil burning facilities. In fact, only one full-scale bag filterhouse is known to have ever been
4-45
-------
Flue Gas Inlet
Downwardly Directed
Peripheral Flow
Clean Gas
Outlet
Whirl Vanes
Central
Vortex Sink
Particulates
To
Hopper
Figure 4-12. Principle of cyclone separator operation
(Reference 4-30).
4-46
-------
installed on an oil burning boiler. The purpose of that filterhouse was to eliminate a visible plume,
and this requirement has now been eliminated by the availability of low sulfur, low ash oil. There-
fore, the filterhouse has been removed from service (Reference 4-32). Many problems were encountered
when the filterhouse was in service, the most prominent being the deterioration of the bags by the
acidic oil and ash and the plugging of the bags due to the hygroscopic nature of the oil ash.
The costs of bag filterhouses vary appreciably depending upon the operating conditions for
which they are designed. A fairly representative estimate for a typical installation, including
amortized capital, installation, auxiliary power, operation, and maintenance costs is 2.2
-------
The capital cost of the wet scrubber is approximately one-quarter of that for an electrostatic
precipitator. However, the operation and maintenance costs could be as high as five times that of
the precipitator (Reference 4-40). On the basis of these figures'the total annualized cost for the
scrubber would be about 70 percent that of the precipitator before consideration of the differential
power requirements to overcome the additional draft loss.
4.3.2 Operational Methods
4.3.2.1 Soot Blowing
Soot blowers are used in virtually all large industrial and utility boilers to remove ash and
slag deposits from the heat transfer surfaces. These deposits come from both the noncombustibles in
the fuel (especially the silica solids, sulfur, and vanadium) and the heavy hydrocarbon molecules
that are difficult to burn (i.e., the carbon residue). Some lighter combustibles will also appear if
the combustion is not controlled well. Moreover, the ratio between those particulates which leave the
boiler directly and those which are first deposited on the heat transfer surfaces, and only leave when
dislodged from these surfaces by the soot blowers, can depend upon the combustion characteristics, the
configuration of the flow passages in the boiler, the shape of the heat exchangers, and the time since
the last blow. Since soot blowers can do no more than remove material which has been deposited on the
internal surfaces of the boiler, the major effect of varying soot blowing procedures is to change the
r
manner in which particulates leave the boiler; little can be done directly with soot blowing to reduce
the total mass of parti oil ate created.
Specific soot blowing procedures can be used, however, to indirectly reduce total particulate
emissions from a plant. The most important procedural change that impacts emissions is the interval
between successive blows. Since soot blowing requires the use of either pressurized steam or air,
both of which cost money and fuel to generate, the frequency of soot blowing has been dictated in the
past by an economic trade-off analysis. The cost of blowing was balanced against the gains from im-
proved heat transfer. On the basis of these guidelines, and also to minimize "hot spots" and buildup
of corrosive deposits on the metal tubes, boiler operators generally established soot blowing programs
with 4- to 8-hour intervals between blows on oil-fired boilers. One consequence of this approach was
the periodic, short-duration discharge of a heavily dust-laden, visible plume. These emissions were
probably dominated by large, heavy particulates that settled back to the ground near the plant and
were a frequent source of complaints by neighbors. In an attempt to overcome these two problems (vis-
ible plumes arid localized fallout), some utility and industrial owners of large boilers have adopted
continuous, or continuous-sequential, soot blowing procedures. Boston Edison, for example, now blows
4-48
-------
each boiler in one of their plants in sequence (i.e., first one, then another, etc., and then return-
ing to the first after the last of the boilers has been cleaned) rather than only two times per shift
as they used to when guided only by boiler economics (Reference 4-56).
Although many blowers are manually operated, soot blowing-manufacturers,will now supply pro-
grammable blowers that can be controlled by a minicomputer to any desired schedule (e.g., one dictated
by economics based on real-time analysis of the boiler performance, knowledge of the energy required
to blow the surfaces, and the capacity of the particle collection equipment).
Unfortunately very little quantitative data are available on particulate emission rates or size
distributions during soot blowing, and apparently none on total emissions during a complete cycle (the
period between the start of two successive blows). One source reports that a particular residual oil
fired utility boiler emitted between 0.05 and 0.08 Ib/MBtu (21.5 - 34.4 ng/J) during normal opera-
tion and about 0.15 Ib/MBtu (64.5 ng/J) during soot blowing (Reference 4-30). Since this boiler was
equipped with an electrostatic precipitator, and since the collection efficiency of precipitators in-
creases with increasing inlet loading and particle size (see Subsection 4.3.1.1), the emissions from
the plant (i.e., downstream of the precipitator) were no higher during the soot blowing period than
during the other times - about 0.02 to 0.05 Ib/MBtu (8.6 - 21.5 ng/J). Apparently this precipitator
was sized conservatively enough to handle the higher particulate loadings during the blow. In plants
where this is not the case, alterations to the soot blowing frequency could reduce total emissions
from the plant by insuring that the emission rates during a blow never exceed the capacity of the
particle collection device.
If the plant has no collection device (recall from Subsection 2.4.3.3 that about one-half the
residual oil consumed by utility boilers is burned in plants which do not have a particulate control
device), then, as noted in the Boston Edison example cited above, the implementation of a reduced
interval between successive blows changes the characteristics of the plume. It is possible that fre-
quent soot blowing causes an increase in total suspended particles because it results in the frequent
discharge of smaller particles rather than the occasional discharge of the larger particles which set-
tle rapidly to the ground. Furthermore, the total mass emissions to the atmosphere probably do not
change. Research is required to determine more clearly the relationship between soot blowing fre-
quency and impact on total suspended particles in the ambient air.
Increased soot blowing frequency is the only variable that has been considered to date to alter
particulate emissions. Spray angles, blower locations, fluid injection ratio, etc., are designed to
remove deposits effectively and efficiently. Moreover, air and steam are considered to be equally
4-49
-------
effective when driven by the same power; the choice of which fluid to use is based strictly on indi-
vidual plant economics (availability of steam at the desired pressure, need for compressed air else-
where, etc.). Therefore, none of these parameters need to be considered when developing a particulate
control program.
4.3.2.2 Low Load Versus High Load Operation
Industrial and utility boilers typically operate at between 60 and 100 percent of their design
capacity, depending on the load requirement. However, information about the effect of boiler load on
particulate emissions is scarce. Examples of the limited available data are presented in Figures
4-13 and 4-14 for two boilers, 250 MW and 600 MW, respectively (Reference 4-36). Both boilers em-
ployed fuel additives and operated with an electrostatic precipitator. If one ignores the scattered
points at the 44 percent load level for the 250 MW unit and at full load for the 600 MW unit, then
both plots show that particulate emissions increase with load.* These graphs suggest, for example,
that a 40 percent reduction in particulate emissions can be realized by decreasing the boiler load
from rated conditions to 50 or 60 percent of rated conditions. However, the capital expenditure
penalties of such an approach are significant and most probably would not be acceptable, especially
given current shortages in both capital and electric generation capacity.
4,3.2.3 Flue Gas Additives
Previous investigators have shown that S03 in the exhaust stream contributes significantly to
the formation of visible plumes (Reference 4-45). Upon emerging from the stack the SO, forms a fine
aerosol mist which increases the plume's opacity. One of the methods that can be used to reduce the
amount of S03 emitted from the stack is to treat the flue gas with additives. This approach has been
used primarily to reduce cold-end corrosion and smut deposits inside the furnace. The corrosion is
caused by sulfuric acid, which is formed on these relatively cold surfaces by the condensation of SO,
(claims of up to 50 percent reduction in S03 with use of flue gas additives appear in the literature -
see Reference 4-49).
If the fuel contains high concentrations of vanadium and sodium, compounds of these metals
will contribute significantly to the corrosion in utility boilers. These deposits generally cannot
be removed from the heat transfer surfaces by soot blowers. Therefore additives must be used or a
different fuel must be obtained."''
*A similar result was obtained for commercial boilers that were fired on No. 6 oil (see Figure 2-7).
""In some cases the corrosion problem can be solved by reducing the superheater temperature from about
1000°F to 900°F (537 - 482°C). However, this causes a fuel penalty or decreased output of 2 to 4
percent and even then may not eliminate the problem (depending on fuel composition, superheater mate-
rial, and boiler utilization).
4-50
-------
D)
= o
3 «0
+->
' >, to
U »->
(0 0
CL HI
(Or
0 0)
c-o
Ol C S-
r- T-
3 -r- 0.
r- (J
O T3 0)
IT) T3 S-
CM 10 a.
o
N
o
o
0
o
3
o
CO
0
o
0
LO
03
o
(O
O
O 01
JD I
*
CO
3
c
i- D;
30
U ID
r- OJ
4->
<_ i-
-------
i- O.
OT3 <-
03 C O
CLrt) 0
m .^
(/IT- OS
T3T3 (0
ro o
io,i;^
C U 0
OT- ID
3 CO
CD
O
I i-
w
-------
Additives in dry powder, slurry, or gaseous form are injected into boiler installations either
inside the firebox or the breeching. The most commonly used additives are limestone, dolomite, mag-
nesium oxide, and ammonia. These additives react with the S03 to form solid sulfates (References
4-42 and 4-45) and, therefore, increase the emission of solid particulates.. However, they turn the
wet acid smut and sticky hygroscopic oil ash into a dry powder (References 4-42, 4-46, 4-47). Al-
though precipitators and cyclones can collect a greater percentage of this dry powder than the acid
smut or oil ash, the collection efficiency is not improved enough to offset the increased mass emis-
sions. Therefore the total mass emissions are actually increased, even from boilers that are equipped
with particulate control devices.
A coating of magnesium oxide on the superheater tubes has also been shown to effectively in-
hibit the conversion of S02 to S03. Apparently, the MgO coating prevents the vanadium deposits and
the iron oxide from acting as catalysts for the S02 to S03 conversion (Reference 4-28).
From the cost-effectiveness standpoint, this approach probably should not be applied if the
only concern is the reduction of plume visibility caused by SOg. However, investigations have indi-
cated that flue gas additives can be beneficial in reducing low temperature corrosion caused by the
condensation of SO, and.may be the only means available of burning residue with a high vanadium
«J
content.
4.3.2.4 Opacity Monitoring and Burner Controls
Instruments are available commercially for the continuous monitoring of stack opacity. In
general these opacity monitors are connected to visible or audible alarm systems which give warnings
should a preset level of opacity be reached. Many power plants and industrial operators of large
boilers now use such a system, and some states require its use in residual-fired units or boilers
larger than a certain size (see Table 6-5). These monitors are now also being connected to continu-
ous recorders to provide a permanent record of boiler performance to both plant supervisory engineers
and air pollution control authorities.
Until recently only one unit was available in the United States which satisfied EPA specifica-
tions for continuous opacity monitors on new steam generators >250 MBtu/hr (73.25 MW). These sources
have to comply with a standard of performance which includes a continuous monitoring requirement.
However, other transmissometers which will also satisfy these specifications (especially spectral re-
sponse in the visible range and restricted divergence of the light beam) are now appearing on the
market. The first unit mentioned above consists of a light source and photocell mounted on one side
of the stack with a reflector mounted diametrically opposite. Since the light beam traverses the
stack two times before it is received by the photocell, the sensitivity is twice what it would be if
the photocell were directly opposite the light. Dust loading, or particulate density, is measured
4-53
-------
by the attenuation of the light beam as it traverses the stack. This attenuation is related loga-
rithmically to the optical density, which, in turn, is related linearly to the dust loading, assuming
a fixed particle size distribution and path length.* Therefore, once calibrated for a given source
the instrument will read dust loading directly if the particulate characteristics do not change. As
shown on Figure 4-15 for the emissions from a coal-fired utility boiler, the monitor's output can
correlate well with outlet particle loading.
This particular system can be used in stacks whose diameter is between 1.2 and 60 feet (0.36 -
18.3 m). The monitor can be installed at a representative location of the exhaust-system. If this
location happens to be away from the stack exit, the output can be corrected to the stack exit condi-
tions. Calibration is done remotely and, therefore, as often as'one wants. The monitor costs about
$8,800, including the unit startup, on-site calibration, alignment, repair kit, spare parts, and ser-
vicing for the first year of operation. The price does not include a recording device such as a strip
chart recorder ($440). The newer units cost about one-third as much as this one, but they do not have
as good a purge system as the more expensive monitor, and maintenance requirements for the system are
somewhat greater. Ambient air is continuously blown across the optical surfaces to purge them. Main-
tenance after the first year includes cleansing of the optical surfaces, calibration, and trouble-
I
shooting. This is normally performed four times a year and takes 2 to 4 hours each time. The cost
of this service is about $800 per year, including labor and spare parts (Reference 4-51).
In addition to these accurate, but expensive, monitors, there are also quite a few less costly
units available. These systems sell for $600 to $1200. Although the lower cost ones usually do not
meet EPA specifications for units less than 250 MBtu/hr (73.25 MM), the more expensive ones generally
do satisfy these requirements (Reference 4-67). In some cases the main difference between a $600 and
a $1200 unit is its linearity - the cheaper system is linear only up to about 80 percent opacity where-
as the more expensive one is linear all the way to 100 percent opacity (as required by the EPA speci-
fication). Both respond to the obscurations in the spectral range specified in the Federal Register.
Some of these units can be used only on stacks whose diameter is less than 10 ft (3 m) and are, there-
fore, not appropriate for large boilers. However, they can be particularly useful when installed on
commercial or small industrial boilers to notify the operator (by an alarm) of a high smoke condition
so that he can correct it immediately. The advantages of such a warning system in reducing localized
*According to the Beer-Lambert law, the transmittance, or fraction of the intensity of a light beam
that passes through a particulate-laden flow, is given by T = e-kcd, where k is a measure of the
light absorbing capability of the particles, c = the concentration of particles (mass per unit vol-
ume), and d = the stack diameter. The optical density, D, is defined by the equation D = - log T.
By combining these two equations, one can show that D = Kc, where K = kd/ln (10) = constant for
constant particle size distribution and light scattering characteristics.
4-54
-------
0.05 0.10 0.15 0.20
Grains/cu ft
0.25
0.30
0.35
Figure 4-15. Optical density versus outlet grain loading from a
coal-fired utility boiler (Reference 4-53).
4-55
-------
pollution and saving fuel are obvious. These opacity monitors need frequent cleaning of the optical
surfaces because they do not have an air purging system.
Certain opacity monitors can' be connected to relays which shut off burners when excess smoke
is encountered. However, boiler manufacturers and operators generally have not used this automated
shutdown approach because they felt that the smoke meters have not been reliable enough (References
4-57 and 4-58). Moreover, high opacity readings in the stack can come from a variety of problems
other than the burners themselves, such as failure of a particulate collection device or accidental
shutting of an air damper. The temporary and occasional localized nuisance caused by the smoke
during such malfunctions probably does not warrant a complete shutdown of a large, multiburner
industrial or utility boiler.
Oxygen concentration in the exhaust gas is a better parameter to monitor than smoke for opti-
mizing the combustion process. Typically a boiler can operate over a range of excess air levels with-
out smoking, especially when it is on the lean side of the optimum oxygen concentration. Therefore,
much more information about the combustion process is obtained from the exhaust gas oxygen concentra-
tion than from the plume opacity.
Host utilities now use automated controls to minimize operator error. The primary control
parameters are load and oxygen content (as measured in the firebox), and the system is designed to
ensure that sufficient oxygen is always present. As load increases, the automatic controls cause the
air flow increase to lead the fuel flow rate increase; conversely, when power demand decreases, fuel
reductions occur before air flow is decreased.
4.3.3 Emerging Technologies - Viscosity Control
Although the viscosity of the fuel at the burner head has been shown to affect particulate
emission from commercial boilers (Reference 4-17), no data are available that demonstrate the effects
of varying only viscosity. Nevertheless it has been suggested that improved viscosity control could
reduce particulate emissions.
Oil-fired industrial and utility boilers, which generally burn residual oil, are equipped with
fuel preheaters. These systems heat the fuel to reduce its viscosity to a level which is compatible
with the burner nozzle. Most preheaters are controlled by the temperature of the heated oil, but a
new device is now available which measures viscosity continuously, in real-time, with an accuracy of
1 percent (Reference 4-66). Output from the device can be used to regulate the preheat level. The
value of this device is that it delivers the fuel at a specified viscosity independent of the fuel's
viscosity-temperature relationship (which is generally different for each batch of fuel).
4-56
-------
4.4 ENERGY CONSERVATION
Energy conservation is at once the most satisfying and the most difficult approach to reducing
particulate emissions. Effective conservation practices would reduce the quantity of fuel burned and,
hence, the generation of ccpbustion derived pollutants. Unlike most other particulate control tech-
niques, this one does not involve any trade-offs between particulates, fuel consumption and NOX emis-
sions; instead all are reduced simultaneously and proportionally.
Particulate emissions from industrial and utility boilers could be reduced by improved
utilization of process steam, by the application of waste heat recovery systems to preheat combus-
tion air or feedwater, by increased reliance on recycled materials, and by reduced consumption of
electricity throughout industry, commercial or public buildings, and private homes. The high cost
of energy is sufficient inducement for most industrial and commercial enterprises to introduce energy
* conservation procedures, and these vary considerably from one facility to another. In addition, many
utilities have active information programs that describe techniques for conserving electricity to
their residential customers. Therefore, these approaches will not be discussed further here.
In the residential sector a recent study (Reference 4-63) has shown that fuel consumption
could be reduced by replacing existing fuel nozzles with properly sized ones. This conclusion is
based on the observation that 97 percent of the units surveyed during the study were overfired - i.e.,
were equipped with a nozzle whose capacity is larger than necessary to heat the residence. When 27
of these units were fitted with smaller nozzles, their efficiency increased an average 1.6 percent.
Unfortunately the average smoke number also increased, from Smoke Spot Number 0.8 to 1.2.
The most direct approach to energy conservation involves the design of new and more efficient
heating equipment. In the past, the design of heat exchanges, for example, has been based on a com-
promise between cost and efficiency. However it is estimated (Reference 4-64) that improved designs
could reduce energy demand by as much as 15 percent while maintaining a good cost/efficiency trade-
off.
Other examples (Reference 4-64) of new and improved equipment design include the use of auto-
/
matic combustion air control (potential energy reduction 10 - 27 percent), modulating burners (10
percent) and oil heat pumps (50 percent). Of these, the heat pump and improved heat exchangers seem
to have the best cost/efficiency trade-off and, therefore, have the highest potential for particulate
reduction through conservation.
One area of energy conservation which has received considerable interest recently is building
insulation. This practice is directly relevant to the current study because improved insulation re-
duces the load on oil-fired furnaces. Space heating accounts for about 80 percent of the consumption
of energy for comfort heating and hot water in residences and about 86 percent in commercial buildings
(Reference 4-57). It has frequently been stated that proper insulation could reduce the nation's
4-57
-------
energy consumption for residential and commercial space heating by 40 percent. If these figures apply
to the large northeastern cities, and if a 40 percent reduction in energy consumption is equated to a
40 percent reduction in particulate emissions, then the immediate application of proper insulation
would reduce the particulate emissions contribution from these sources by up to 5 percent of the
total in these areas. Since immediate application of improved insulation to all buildings is unre-
alistic, the actual reductions in particulate emissions from an organized building insulation program
will be less.
Several documents have been published recently by national agencies which describe how to re-
duce energy consumption for space heating and give the associated costs and benefits. These reports
are listed in Table 4-5. Suggested energy reduction programs for winter heating are presented in
Table 4-6. These approaches generally fall into three categories: life-style changes (e.g., lower
thermostat settings), building improvements (insulation, weather stripping, storm windows), and
proper maintenance of the heating equipment.
TABLE 4-5. SELECTED REPORTS ON ENERGY CONSERVATION
Technical Options for Energy Conservation in Buildings
National Bureau of Standards Technical Note 789, July 1973. (Avail-
able from the Superintendent of Documents 6PO, for $2.35.) Contains
material distributed at the Joint Emergency Workshop on Energy Con-
servation in Buildings, June 19, 1973, under the sponsorship of the
National Conference of States on Buildings Codes and Standards and
of the National Bureau of Standards.
Energy Conservation Program Guide for Industry and Commerce
National Bureau of Standards Handbook 115, September 1974. (Avail-
able from the Superintendent of Documents GPO, for $2.90.) Prepared
in cooperation with the Federal Energy Administration/Conservation
and Environment as an energy management tool for intermediate to
small sized firms.
Potential for Energy Conservation in the United States: 1974-1978
National Petroleum Council (Committee on Energy Conservation),
September 1974. Presents petroleum industry assessment of potential
for energy reductions through conservation. Gives estimates of
energy savings from a variety of conservation measures.
Mineral Resources and the Environment
Committee on Mineral Resources and the Environment, Commission on
Natural Resources, National Research Council, National Academy of
Sciences, 1975. (Available from Printing and Publishing Offices,
NAS, 2101 Constitution Avenue N.W., Washington, D.C. 20418.) Con-
tains a section on material conservation.
4-58
-------
TABLE 4-6. REDUCTION OF HEATING ENERGY CONSUMPTION
without Extra Cost . . ,
Set your thermostat lower . ,
Close off rooms not used and turn off heat
Let the sunshine in on winter .days; pull shades at night
Reduce air leakage and ventilation
Be careful about open windows and doors
Reduce temperature in public spaces, lobbies, etc.
Institute rigorous schedules for planned operation of ventilation
Wear heavier clothing
Maintain an efficient heating plant
Turn off - turn down lights and electric appliances except when needed
Concentrate evening work or meetings in a single heating zone
' ' With Extra Cost
Add a clock thermostat
Add insulation, as much as .feasible
Add ins.ulating glass or storm windows and doors
Caulk and seal around windows, doors and other openings
a
a
a
Insulate heating ducts and seal against air leakage into nonheated spaces
(attics, crawl spaces)
a Maintain heating equipment - clean heat transfer surfaces, set flame and
combustion air
a Install heat recovery and conservation devices
a Install automatic pilot light;
a Adjust ventilation system
a Avoid use of portable electric heaters by improving main heating systems
a Replace defective or inefficient heating systems with systems of higher
efficiency
a Modify systems for zone control using systems of higher efficiency
a Provide means to transfer heat from the core of a large building to the
cool periphery needing heat
Install automatic door closers '
Source: Technical Options for Energy Conservation in Buildings, NBS Technical
Note 789, 1973
4-59 '
-------
REFERENCES FOR SECTION 4
4-1. Bunyard, F. L. and Copeland, J. 0., "Soiling Characteristics and Performance of Domestic and
Commercial Oil Burning Units," January 28, 1968, Air Pollution Technical Information Center
Report No. 76132.
4-2. "Report on Combustion Testing Program Concerning Residential and Commercial Oil Burning Equip-
ment in New York City," Department of Air Resources, The City of New York, February 1974.
4-3. Ibid.
4-4. "How To Stop Smoking," An Air Pollution Control Guidebook for Boiler and Incinerator Operators
from the New York City Department of Air Resources, June 1971.
s~
4-5. Locklin, D. W. and Barrett, R. E., "Guidelines for Burner Adjustments of Commercial Oil-Fired
Boilers," Battelle, Columbus Laboratories, Prepared for U.S. EPA, Report.No. EPA-600/2-75-
069b, October .1975.
4-6. Monroe, E. S., Jr., "Improving Combustion with Heavy Fuel Oil," The Oil and Gas Journal, Vol-
ume 64, No. 48, November 28, 1966, pp. 66-69.
4-7. Safford, D. E., "Burner Changes Vs. Fuel Changes," Hydrocarbon Processing, Volume 49, No. 2,
February 1970, pp. 99-100.
4-8. Korn, Joseph, "Sonic Fuel Atomization," Heating, Piping and Air Conditioning, Volume 43, No. 4,
April 1971, pp. 84-86.
4-9. Schreter, R. E., et al., "Industrial Burners -Today and Tomorrow," Mechanical Engineering,
June 1970, pp. 23-29.
4-10. Drake, P. F. and Hubbard, E. H., "Combustion System Aerodynamics and Their Effect on the Burn-
ing of Heavy Fuel Oil," Journal of the Institute of Fuel, Volume 39, No. 302, March 1966,
pp. 98-109.
4-11. Dickerson, R. A. and Okuda, A. S., "Design of an Optimum Distillate Oil Burner for Control of
Pollutant Emissions," Rocketdyne Division, Rockwell International, Prepared for U.S. EPA,
Report No. EPA-650/2-74-047, June 1974.
4-12. Lambert, L. C., "Versatile Oil Burner Offers High Efficiency and Reduced Emissions," Chemi-
cal Engineering, September 2, 1974, pp. 51-54.
4-13. BUnz, P., et al., "Influences of Fuel Oil Characteristics and Combustion Conditions on Flue
Gas Properties in Watertube Boilers," Journal of the Institute of Fuel, September 1967,
pp. 406-416.
4-14. Robison, E. B., "Application of Dust Collectors to Residual Oil Fired Boilers in Maryland,"
Bureau of Air Quality Control, State of Maryland, BAQC-TM-74-15, December 1974.
4-15. Personal Communication. Maryland State Department of Mental Health and Hygiene, June 1975.
4-16. Personal Communication. Ray's Burner Company, San Francisco, California, June 1975.
4-17. Barrett, R. E., et al., "Field Investigation of Emissions from Combustion Equipment for Space
Heating," Battelle, Columbus Laboratories, Prepared for U.S. EPA, Report No. EPA-R2-73-084a,
June 1973.
4-18. Taylor, F. R., et al., "A Survey of Emissions from Residential Oil-Fired Heating Units,"
Proc. Mid-Atlantic States Section - APCA, Philadelphia, Pennsylvania, October 24, 1963.
4-60
-------
4-19. "A.Study of Residential Oil Burners in the Baltimore and Washington Metropolitan Areas During
the Winter of 1972 - 1973," Bureau of Air Quality Control Data Report, State of Maryland,
September 1974.
4-20. Hall, R. E., et al., "A Study of Air Pollutant Emissions from Residential Heating Systems,"
EPA-650/2-74-003, January 1974.
4-21. Personal Communication. Petroleum Marketing Education Foundation, South Carolina, October 1975.
4-22. Locklin, D. W., "Recent Research and Development in Residential Oil Burners," Combustion Tech-
nology: Some Modern Developments, ed. H. Palmer and J. M. Beer, Academic Press, New York and
London, 1974.
4-23. Hein, G. M., et al., "A Progress Report on the API Vaporizing Burner Program," Paper CP66-7,
1966 API Research Conference on Distillate Fuel Combustion, Chicago, Illinois, June 13-15, 1966.
4-24. Cooper, P. W., et al., "Re<;irculation and Fuel-Air Mixing as Related to Oil-Burner Design," API
Publication 1723, May 1964.
4-25. Monaghan, M. T. and McGrath, I. A., "The Influence of Flue Gas Recirculation on the Combustion
of Fuel Oil," Journal of the Institute of Petroleum, Volume 55, No. 545, September 1969, pp.
302-321.
4-26. Personal Communication. Blueray System, Weston, Massachusetts, August 1975.
4-27. "14 Years, $l-Million Later, OOHA's New Burner-Furnace on Market: Is This IT?," National Petro-
leum News, January 1975, pp. 34-35.
4-28. Personal Communication. R. E. Hall of EPA, August 1975.
4-29. Locklin, D. W. and Barrett, R. E., "Guidelines for Residential Oil-Burner Adjustments,"
Battelle, Prepared for U.S. EPA, Report No. EPA-600/2-75-069a, October 1.975.
4-30. Fernandes, J. H., et al., "Boiler Emissions and Their Control," Paper presented at Conference
on Air Pollution Control, Mexico City, April 28, 1966.
4-31. Hartshorn, W. T., "Electrostatic Dust Collection from Oil-Fired Boilers," TAPPI, Volume 56,
No. 6, June 1973, pp. 84-86.
4-32. Bagwell, F. H. and Velte, R. G., "New Developments in Dust Collecting Equipment for Electric
Utilities," JAPCA, Volume 21, No. 12, December 1971, pp. 781-782.
4-33. Pinheiro, George, "Precipitators for Oil-Fired Boilers," Power Engineering, Volume 75, No. 4,
April 1971, pp. 52-54.
4-34. Derby, K. and Whitehead, C., "The Performance of Electrostatic Precipitators in Relation to
Low Sulfur Fuels," Proceedings of the Second International Clean Air Congress held at Washing-
ton, D.C., December 6-11, 1970, pp. 911-922.
4-35. White, Harry J., "Role of Electrostatic Precipitators in Particulate Control: A Retrospective
and Prospective View," JAPCA, Volume 25, No. 2, February 1975, pp. 102-107.
4-36. "Particulate Emission Control Systems for Oil-Fired Boilers," GCA Corporation, Prepared for
U.S. EPA, Report No. EPA-450/3-74-063, December 1974.
4-37. McGarry, F. J. and Gregory, C. J., "A Comparison of the Size Distribution of Particulates
Emitted from Air, Mechanical, and Steam Atomized Oil-Fired Burners," JAPCA, Volume 22,
No. 8, August 1972, pp. 636-639.
4-38. Burdock, J. L., "Centrifugal Collectors Control Particulate Emissions from Oil-Fired Boilers,"
TAPPI, Volume 56, No. 6, June 1973, pp. 78-82.
4-39. Barrett, R. E., et al., "Investigation of Particulate Emissions from Oil-Fired Residential,
Heating Units," Battelle, Columbus Laboratories, Prepared for U.S. EPA, Report No. EPA-
650/2-74-026, March 1974.
4-61
-------
4-40. Booth, J. B., "Particulate Control Equipment for Industrial Power Boilers," ASME Paper 75-
IPWR-4, Hay 1975. .
4-41. "Control of Air Pollution from Fossil Fuel-Fired Steam Generators Greater than 250 Million
Btu per Hour Heat Input," EPA, for technical review only - not for publication.
4-42. Bagwell» F. A., et al., "Design and Operating Experience with a Filterhouse Installed on an
Oil-Fired Boiler," JAPCA, Volume 19, No. 3, March 1969, pp. 149-154.
4-43. Statm'ck, R. H. and Drehmel, D. C., "Fine Particle Control Using Sulfur Oxide Scrubbers,"
JAPCA, Volume 25, No. 6, June 1975, pp. 605-609.
4-44. Accortt, J. I., et al., "Fine Particulate Removal and SO? Absorption with a Two-Stage Wet
Scrubber," JAPCA, Volume 24, No. 10, October 1974, pp. 966-969.
4-45. Austin, H. C. and .Chadwick, W. L., "Control of Air Pollution from Oil-Burning Power Plants,"
Mechanical Engineering, Volume 82, No. 4, April 1960, pp. 63-66.
i
4-46. Kukin, Ira, "Additives Can Clean Up Oil-Fired Furnaces," Environmental Science and Technology,
Volume 7, No. 7, July 1973, pp. 606-610.
«
4-47. Morse, W., "Smutting of Oil-Fired Boilers and Its Correction by DCP System," The Plant Engi-
neer, Volume 14, No. 8, August 1970, pp. 191-192.
4-48t Exley, L. M., "A Practical Review of Residual Oil Firing Problems and Solution," Combustion,
Volume 41, No. 9, March 1970, pp. 16-23.
4-49. Salooja, K. C., "Burner Fuel Additives," Combustion, January 1973, pp. 21-27.
4-50. "Smoke Stacks Signal Poor Combustion Control," Process Engineering, April 1974, pp. 76-77.
4-51. Personal Communication. Lear Siegler, Inc., Los Angeles, California, September 1975.
4-52. lammartino, N. R., "Technology Gears Up to Control Fine Particles," Chemical Engineering,
Vol. 79, No. 18, August 21, 1972, pp. 50-52.
4-53. Schneider, W. A., "Opacity Monitoring of Stack Emissions: A Design Tool with Promising Re-
sults," Generation Planbook 1974, by the editors of Power Magazine, 1974.
4-54. "Inspection of Oil-Fired Boilers for Air Pollution Control," Bureau of Air Quality Control,
State of Maryland, BAQC-IB 73-102, June 1973.
4-55. Siege!, R., et al., "A Strategy of Reduction of Particulate Emissions in the Boston Area,"
JAPCA, Volume 25, No. 3, March 1975, pp. 256-259.
4-56. Private Communication. From C. Dolloff, Boston Edison, October 16, 1975.
4-57. Private Communication. From F. Gottlich, Boston Edison, December 3, 1975.
4-58. Private Communication. From S. Potterton, Bacock & Wilcox, December 5, 1975.
4-59. Private Communication. From R. H. Boll, Bacock & Wilcox, December 5, 1975.
4-60. Private Communication. From R. Siege!, Walden Research, December 3, 1975
4-61. Private Communication. From T. Parks, Mass. Air Quality Control, December 4, 1975.
4-62. Metzger, "Clog-Proof Superspray Oil Burner," Popular Science, January 1976, pp. 64-66.
4-63. Katzman, "A Study To Evaluate the Effect of Performing Various Energy Saving Procedures on
Residential Oil Burner Installations in the New England Area and To Gather Information on the
Steady-State and Dynamic Performance of These Installations," Walden Research Division, Abcor
Inc., Cambridge, Massachusetts.
4-64. Private Communication. From William Axtman, ABMA
4-65. "Potential for Energy Conservation in the United States: 1979 - 1985," National Oil Jobbers
Council, Washington, D.C., 1975.
4-62
-------
4-66. Manufacturers' Literature. Automation Products, Inc., Houston, Texas.
4-67. "Standards of Performance for New Stationary Sources," Federal Register, Volume 39, No. 177,
September 11, 1974.
4-68. Private communication from 6. Pinheiro, Research Cottrell, May 7, 1976.
4-69. Vandegrift, A. E., et a!., "Particulate Pollutant Systems Study, Volume III, Handbook of Emis-
sion Properties," Midwest Research Institute, Prepared for the U.S. National Air Pollution
Control Agency, NTIS Report No. PB 203 522, May 1, 1971.
4-63
-------
-------
SECTION 5
INDUSTRY STANDARDS FOR CONSTRUCTION, SAFETY, AND
PERFORMANCE OF BURNERS, BOILERS, AND FURNACES
As seen in Chapter 2, participate emissions from an oil-fired furnace or boiler are a func-
tion of the design characteristics of the unit. This chapter gives an overview of the rules and
guidelines used by thevfive principal private and trade organizations which set design standards
for furnaces and boilers:
American Boiler Manufacturers Association (ABMA)
e The Hydronics Institute (Formerly the Institute of Boiler and Radiator Manufacturers,
or IBR).
« The American Society of Mechanical Engineers (ASME).
The American National Standards Institute (ANSI).
9 Underwriters Laboratories, Inc. (UL).
The ABMA is primarily concerned with providing a uniform method of testing commercial and
industrial boilers to determine their rated capacity and efficiency (Reference 5-1). In order
for a manufactured boiler to be listed with the ABMA, it must be capable of continuous operation
at its stated capacity "without objectionable smoke as defined by existing smoke ordinances."*
Consequently the ABMA serves an initial screening function against potentially large polluters.
The boiler must also be in compliance with UL safety standards for construction.
The Hydronics Institute performs a function similar to the ABMA, but mainly for residen-
tial units. In order to be listed with the Institute a boiler must maintain a Bacharach smoke
reading of less than No. 2 for distillate firing or No. 4 for residual oil firing (Reference
5-2). Consequently the Hydronics Institute also serves to screen out designs which are inher-
ently heavy particulate emitters.
The manual did not specify whose smoke ordinances should be obeyed; presumably, they mean the
regulations of the air pollution control district in which the boiler will be located although
they could refer to those of the district in which the boiler manufacturer resides.
5-1
-------
In addition to supplying a testing service and a boiler rating guide, the Institute pub-
lishes instructions for the operation and maintenance of large and small boilers. -These manuals
are generally educational and are written in non-technical language so as to be of use to oper-
ators of small boilers with little knowledge of boiler fundamentals.
The ASHE publishes detailed recommended rules for the operation of both heating and power
boilers (References 5-3 and 5-4). These rules include several practices which, if followed,
would lead to reduced particulate emission levels. They are as noted below:
A detailed maintenance procedure, including a list of boiler parts which typically
require frequent inspection (e.g., nozzles and oil line strainers, both of which cause
increased particulate emission when dirty),
x
A detailed description of an annual combustion efficiency test, including measurements
of draft, smoke, and COg. Although target C02 concentrations are suggested, no limi-
tations are placed on Bacharach smoke numbers.
General guidelines for soot-blower operation, which may reduce particulate emissions,
or at least avoid the notorious short duration, high grain loading puffs.
The ASME also publishes many nationally recognized codes, such as the other sections of the
pressure vessel code, which set design standards for the structural components of the system and
operating specifications for the controls. These were developed to ensure the safety of such sys-
tems and are not affected by any proposed particulate control techniques.
The tv/o organizations which exercise the most authority over furnace and boiler design stan-
dards are UL and ANSI, and their primary aim is toward consumer protection. Both organizations
publish standards for construction and operation of furnaces and boilers. These deal mainly with
structural, electrical, and control requirements. In the case of the UL standards, any references
to combustion systems duplicate their standard No. 296 for oil burners (Reference 2-5). This
standard, which is also approved as an ANSI standard, includes the following requirements that can
impact particulate emissions:
Emissions may not exceed a Bacharach smoke number of 2 if distillate is used or No. 4
if resid is used. These requirements must be met even if a burner is allowed to con-
tinue operation after failure of a forced draft fan.
Ignition systems must activate before, or simultaneously with, main burner fuel delivery.
This reduces "ignition puffs" to a minimum.
5-2
-------
t Burner designs must include a self aligning feature so that replaced nozzles and firing
assemblies will be automatically aligned to their correct position. This stipulation
eliminates high particulate emissions caused by leaking nozzles, loose pump connections,
etc. ' -.--...-
Burners must automatically shut off if a low atomizer pressure is detected. This re-
quirement reduces dribble to a minimum. Additionally, the burner is tested over its
entire range of operation under "worst case" conditions (for example, it is fired
during testing with the heaviest recommended fuel oil), and it must perform satisfac-
torily under simulated emergency conditions. i ,
A new ANSI standard, Z91.2 for automatic pressure atomizing oil burners of the mechanical
draft type, has recently received mail ballot approval and will be published in early 1976. This
standard is directed at residential units (<7 gph or about 0.3 Mw) and stipulates that they must be
capable of producing and maintaining a CCL concentration in the exhaust of no less than 10 percent
and a Smoke Spot Number of no greater than shade No. 1 during steady operation. These limits are
considered to be a reasonable standard for the industry at this time, since they represent perfor-
mance and emission levels that the products from many, but not all, manufacturers can meet now.
Moreover, they are more demanding than the previous levels of 8 percent C02 and Smoke Spot Number 2.
In general, the rules, procedures and guidelines proposed by each of the above organizations
seem to lead incidentally to reduced particulate emissions; no guideline was found that promoted
increased emissions as a by-product of encouraging safety and standardized rating methods. These
new equipment standards, though not required by pollution control regulations, appear to be bene-
ficial in terms of limiting particulate emissions. In addition, the ASME boiler operating rules
represent a basic standard of maintenance that should ensure minimal emissions from existing equip-
ment.
5-3
-------
REFERENCES FOR SECTION 5
5-1. Packaged Firetube Boiler Ratings (1971 Edition). Prepared by the Technical Committee of
the Packaged Firetube Section, ABMA, 1180 Raymond Blvd., Newark, N.J. 07102.
5-2. Testing and Rating Standard for Cast Iron and Steel Heating Boilers, First Edition, The
Hydronics Institute, 35 Russo Place, Berkeley Heights, N.J7 07922, March 1975.
5-3. ASME Boiler and Pressure Vessel Code Section VI, Recommended Rules for the Care and Opera-
tion of Heating Boilers, The American Society of Mechanical Engineers, United Engineering
Center, 345 East 47th Street, New York, N.Y. 10017, July 1, 1974.
5-4. ASME Boiler and Pressure Vessel Code, Section VII, Recommended Rules for Care of Power
Boilers, The American' Society of Mechanical Engineers. United Engineering Center, .345
East 47th Street, New York, N.Y. 10017, July 1, 1974.
5-5. Standard for Safety, Oil Burners. UL 296, Underwriters Laboratories, 1285 Walt Whitman
Road, Melville, N.Y. 11746.Approved as American National Standard ANSI Z96.2-1974.
September 19, 1974.
5-4
-------
SECTION 6
SUMMARY OF EXISTING CONTROL PROGRAMS
6.1
INTRODUCTION
The major objective of this document is to propose model control programs that can serve as
guides to State and local air pollution authorities when they prepare regulations for the control of
particulates from oil-fired burners. Since any proposed control program should take advantage of
the experiences gained with the existing control programs, we present here a summary and review of
control programs that are currently in use for particulates from oil-fired equipment. This sum-
mary deals mainly with existing emission limits and enforcement procedures that are used by several
control authorities. Unfortunately very little can be said about the effectiveness of these control
programs, both because many of them are quite new and because the relationship between ambient air
particulate concentrations and emissions from a given group of sources is not always clear. Occa-
sionally, we have received suggestions for additional or new control programs by -staff members of
various local and state control authorities, and these are included in our review.
Two approaches have been used to evaluate the existing control programs. The first consists
of the review and compilation of regulations dealing with particulates. These regulations were
obtained from the Environment Reporter - State Air Laws (Reference 6-1). Our review is restricted
to those regulations which control parti cul ate .loadings and visible emissions from indirect heat-
ing oil-fired fuel combustion sources.
The other approach was to meet directly with the enforcement and engineering staff in nine
State and local air pollution control agencies which are believed to have pursued active programs
for the control of particulate matter from oil burner equipment. Four control agencies were con-
tracted which had jurisdication over the entire State and five were contacted which had jurisdiction
only over a metropolitan area (see Table 6-1). Los Angeles is included in the list of agencies con-
tacted, despite the relatively low utilization of oil in the past, because of its history of leader-
ship in control programs. The list of agencies contacted was purposely selected to include both
State and metropolitan agencies. This was done to determine whether each one of these types of
agencies had specific and unique control problems that the other agencies did not encounter.
6-1
-------
TABLE 6-1. AIR POLLUTION REGULATORY AGENCIES VISITED DURING REPORT
States
Connecticut
Mary!and
New Jersey
Rhode Island
Metropolitan Areas
Boston
Erie County, New York
Los Angeles
New York City
Philadelphia
In the next section we present the review of the particulate and visible emission limits for
all the states. The following section then looks in greater detail at the emission limits and en-
forcement procedures used in the nine selected control districts. Before preceding to these dis-
cussions of regulatory programs, we note that certain nonregulatory programs exist which aim to
reduce particulate emissions. Chief among these are the various private, industry, and trace asso-
ciation sponsored training programs for burner servicemen and boiler operators/maintenance staff.
Voluntary operational changes that result in energy conservation are indirect particulate control
programs since they cause a reduction in fuel consumption. This category of activities includes
the use of more building insulation, more attention to energy efficient operation, the installation
of energy recovery devices such as waste heat recovery units, and a shift to greater recycling of
metals, glass, and paper products. And, finally, information campaigns ("PR") can educate both the
public and the owners of commercial or small industrial boilers about the value of proper, periodic
maintenance programs. Such campaigns seem to be most effective if they stress the fuel savings
that come from correct maintenance practices and only secondarily describe the consequent environ-
mental benefits (see, for example, the discussion in subsection 4.2.1.1 about burner servicing in
Boston).
6.2 PARTICULATE LOADING AND VISIBLE EMISSION LIMITS
Table 6-2 presents the particulate limits for each state for oil-fired indirect heating fuel
burning sources. As mentioned earlier these limits were taken from the State Air Laws volume of a
publication called the Environment Reporter (Reference 6-1). This multi-volume document is pub-
lished in loose-leaf format and updated periodically. Other volumes report state laws for other
media or federal regulations. Since this publication is directed in large part to environmental
lawyers, it is frequently available at large law libraries. Most of the regulations are in terms
of mass of pollutants per unit heat input (Ib particulate/MBtu), but some are in terms of mass
emissions per unit exhaust flow (grains particulate/standard cubic foot of exhaust). Whenever the
6-2
-------
TABLE 6-2. STATE REGULATIONS FOR PARTICULATE EMISSIONS FROM OIL-FIRED
INDIRECT HEATING SOURCES3, Ib/MBtu
AlabMB
Alaska
Art MM
Arkansas
California
Colorado
Connecticut
Delaware
O.C.
Florida
Seorgla
Hwall
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
Mew Hampshire
New Jersey
New Mexico
Raw York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Mtode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Veroont
Virginia
Washington
Best Virginia
Wisconsin
Wyoming
A. Saaoa
Sun
Puerto Rico
Virgin Islands
fc
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
"ft
ft
*
ft
ft
*
ft
ft
ft
ft
ft
ft
*
ft
ft
ft
ft
ft
*
ft
Footnotes follow.
Source Neat Input Rate. HBtu/hr
<1
0.083
0.167
0.5
0.1
0.3
0.1
0.10
0.6
0.6
0.4
0.6
0.6
0.6
0.6
0.33
0.60
0.3
0.4
0.33
0.15
0.1
0.1
0.3
<10
0.6
0.25
0.21
0.13
0.50
0.6
0.6
0.6
0.56
0.60
0.06
0.1
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.005
0.6
0.6
0.4
0.6
0.4
0.20
0.60
0.6
0.5
0.34
0.6
See Appendix C
>10
0.5
*
a,
0.05
0.1
20
0.51
0.23
0.51
*
'0.56
--*
0.4
«>
0.41
so
0.4
0.1S
0.41
0.38
0.03
0.46
0.41
0.40
0.22
0.436
0.24
100
0.35
0.15
0.35
0.33
0.4
0.35
0.35
0.15
0.443
0.17
iZOO
ft
0.3
B
>200
B
0.30
0.30
0.375
0.1
£250
0.12
»
>250
0.2S
0.12
0.10
0.28
0.10
0.36
»
0.10
0.1
»
0.10
0.10
<500
»
0.24
0.30
0.24
0.359
>500
0.243
0.1
f
0.1
1000
0.21
0.21
0.26
0.21
0.328
0.10
2500
0.17
0.17
0.23
0.17
0.291
5000
0.13
0.14
0.2
0.14
0.266
7500
0.11
0.13
0.19
0.13
0.252
10000
0.09
0.02
0.12
0.12
0.19
0.18
0.12
0.12
0.10
0.242
0.12
0.12
> 10000
Bt
.
0.1
0.09
50000
0.036
0.197
>10*
9»
0.025
jr
^.
. £»
0.02
0.180
for conversion to SI units.
6-3
-------
FOOTNOTES FOR TABLE 6-2
a
Source: Environment Reporter: State Air Laws. Volume 2, Parts 1 and 2. A loose!eaf document
published and periodically updated by the Bureau of National Affairs Publishers, Washington, D.C.,
20037. Based on update as of November 1975.
Footnotes follow by states in alphabetical order for those with *. E = emission limits, Ib/MBtu;
H - input heat rate, MBtu/hr. Table contains limits for new units whenever there are different
limits for new and existing sources. Limits for existing sources in the footnotes. >- means
constant emission limit from left end of line to tip of allow. »- means continuously decreasing
limit between the values at each end of the dashed line (limits in state regulations either given
as a graph or as an equation. See the footnotes by state below).
Alabama
Alaska
Arizona
California
Connecticut
Delaware
D.C.
Florida
Georgia
Hawaii
Idaho
Indiana
Tabulated emissions are for Class 1 counties.
H-o-589 for 10 < H < 250 MBtu/hr.
For Class 2 counties E = 3.109
Based on regulation of 0.05 grains/scf flue gas @ standard conditions or 0.10
grains/scf for units operating prior to July 1, 1972.
Based on 24 hour arithmetic average.
Based on grains/scf corrected to 12 percent C02- Valid for 14 counties. One
additional county limited new sources to 0.167 Ib/MBtu and existing sources to
0.333 Ib/MBtu. Another county limited sources to 0.263 Ib/MBtu and 12 counties
limited sources to 0.5 Ib/MBtu. In addition the South Coast Air Basin limits all
new sources to 10 Ib/hr.
0.2 Ib/MBtu for existing sources.
No restrictions if heat input less than 1 MBtu/hr. Restrictions exclude start-ups
and shutdown periods. Limits based on a 2 hour average.
-0.13 Ib/MBtu for units less than 3.5 MBtu/hr.
Based on a 2 hour average.
For units in operation or under construction on or before January 1, 1972:
Less than 10 MBtu/hr - 0.7 Ib/MBtu
Greater than 10 but less than 20,000 MBtu/hr - E = 0.7 (10/H)0'202
Greater than 2,000 MBtu/hr - 0.24 Ib/MBtu
No general regulation stated in final SIP.
Sources less than 1 MBtu/hr are exempt; sources greater than 10 MBtu/hr and less
than 10,000 MBtu/hr have allowable emis'sion rate determined by Log E = -0.233 Log H
+ 1.0112.
For existing sources:
H MBtu/hr
H £ 10
10 < H < 10,000
H > 10,000
E Ib/MBtu
0.6
0.87H"0'16
0.2
Iowa
0.6/MBtu for any new source and any source in any standard metropolitan area; 0.8
Ib/MBtu outside metropolitan area.
6-4
-------
FOOTNOTES FOR TABLE 6-2 (Continued)
Kentucky
Maine
Maryland
For existing sources the emissions are limited to:
MBtu/hr Priority
<_ 10
50
100
250
500
1000
2500 '
5000
7500
10000
- Based on 2 hour average ,
Based on regulations in gr/scfd at 50 percent excess air for new units. Limits
for existing residual-fired units greater than 200 MBtu/hr are 0.03 Ib/MBtu. No
mass limits on smaller existing units or on any sized distillate units.
1 '
0.56
0.38
0.33
0.26
0.22
0.19
0.15
0.13
0.12
0.11
2
0.75
0.52
0.44
0.35
0.30
0.26
0.21
0.18
0.16
0.15
3
0.80
0.57
0.49
0.40
0.34
0.30
0.24
0.21
0.19
0.18
Massachusetts 0.15 Ib/MBtu for existing sources greater than 3 MBtu/hr except 0.12 Ib/MBtu in
critical areas of concern (e.g., Boston). New units greater than 250 MBtu/hr can
emit 0.10 Ib/MBtu if also equipped with an S02 control device.
Minnesota
Missouri
-All old installations 0.6 Ib/MBtu.
Montana
Nevada
New Hampshire
- For existing sources:
10 MBtu/hr
100 MBtu/hr
1000 MBtu/hr
2000 MBtu/hr
0.6 Ib/MBtu
0.43 Ib/MBtu
0.29 Ib/MBtu
0.26 Ib/MBtu
10000 MBtu/hr- 0.18 Ib/MBtu
- For existing sources:
£ 10 MBtu/hr
100 MBtu/hr -
1000 MBtu/hr -
> 10000 MBtu/hr -
0.6 Ib/MBtu
0.40 Ib/MBtu
0.28 Ib/MBtu
0.19 Ib/MBtu
-For 2.5 to 1000 kcal/hr, E = 1.34H"°'231(1.02H~°-231)kg/106 kcal.
E = 13.9H"0'568 (17.0H~°-568).
For existing sources
Less or equal to 10 MBtu/hr - 0.6-Ib/MBtu
50 MBtu/hr - 0.46 Ib/MBtu
100 MBtu/hr - 0.40 Ib/MBtu
500 MBtu/hr - 0.31 Ib/MBtu
1000 MBtu/hr - 0.28 Ib/MBtu
2500 MBtu/hr - 0'.24 Ib/MBtu
5000 MBtu/hr - 0.22 Ib/MBtu
>7500 MBtu/hr - 0.20 Ib/MBtu
>10000 MBtu/hr - 0.19 Ib/MBtu
For > 103 kcal/hr,
6-5
-------
FOOTNOTES FOR TABLE 6-2 (Concluded)
New Mexico
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
South Carolina
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Wisconsin
Wyoming
Guam
Virgin Islands
c- Based on operation greater than 8000 hours/year. Revision upward expected in 1976.
Use Figure 2 in state regulations for interpolating between given values.
For existing equipment E = 0.80 Ib/MBtu
- E = -0.15 log H + 0.55 for 10 MBtu < H < 1000 MBtu/hr.
For old equipment less than 0.6 Ib/MBtu.
For old equipment less than 0.66 Ib/MBtu.
-2.5 < H < 50 MBtu/hr
50 < H < 600 MBtu/hr
H > 600 MBtu/hr
E = 0.4 Ib/MBtu
E = 3.6 H"0-56
E = 0.1
For all equipment emission limits depend upon stack height for heat inputs greater than
500 MBtu/hr. For units in use or under construction before February 11, 1971, limits are
0.8 Ib/MBtu for those less than 10 MBtu/hr and the same as new units for those >10 MBtu/hr.
For existing sources built or under construction on or before July 1, 1975:
H £ 10 MBtu/hr E = 0.6 Ib/MBtu
10 < H <_ 250 MBtu/hr E = 0.6 (lO/H)0-259"
H > 10000 E = 0.1
Federal regulation (0.1 Ib/MBtu) extended to all oil-fired sources in the state
unless sources are less than 500 HP. This regulation was obtained from phone
conversations with Texas Air Control Board on December 3, 1975. For existing
sources E (Ib/hr) = 0.048 q°-62 for H < 10000 MBtu/hr where q is effluent flow
rate in ACFM, and E = 0.1 Ib/MBtu for H > 10000 MBtu/hr.
No limits.
0.10 Ib/MBtu for H > 300 MBtu/hr. 0.020 Ib/MBtu for sources constructed after
July 1, 1971 and H > 1000 MBtu/hr.
For heat inputs H, where 25 < H < 10000 MBtu/hr, the emission of particulate is
E = 0.8425 H~°-231*.
For all old equipment, limit is 0.66 Ib/MBtu.
Limit for units built or under construction on or before April 1, 1972 is 0.6
Ib/MBtu except 0.3 Ib/MBtu in Subregion 1 of Lai'e Michigan Interstate AQCR and
0.15 Ib/MBtu for units greater than or equal to 250 MBtu/hr in Southeast Wisconsin
Interstate AQCR.
- For old sources H <. 10 MBtu/hr, E = 0.6 Ib/MBtu.
No general regulation stated in final plan.
-Regulation obtained from APTD-1334, "Analysis of Final State Implementation Plans -
Rules and Regulations".
6-6
-------
regulations were given in the later form, they were converted to a heat input basis by the
following relationship:* . .
1 grain/scf = 1.67 Ib/MBtu = 718.1 ng/J
Table 6-2 shows that there is considerable variation among the regulations from state to
state. Some states do not place emission limitations on smaller units, and where all the states
restrict emissions from a given size source, these restrictions can sometimes differ by an order
of magnitude. Typically regulations range from about 0.6 Ib/MBtu (258 ng/J) from units with a heat
input of 10 MBtu/hr (in those states which regulate units of that size) to values of 0.1 to 0.2
Ib/MBtu (43 - 86 ng/J) for the very large utility boilers.f Since the regulations differ so,much
from state to state, a condensed version of this large table has been prepared to facilitate com-
parisons among the states (Table 6-3). In this abbreviated version of the state regulation's, each
column approximately represents one of the four user categories discussed in this report. A quick
review of this table shows, for example, that 29 states and territories (i.e., slightly more than
half of the 55 states arid territories included on the table) do not restrict particulate emissions
from residential and commercial units smaller than 1 MBtu/hr (0.293 MW). Only three states or
territories do not limit particulate emissions from units larger than 1 MBtu/hr (0.293 MW). The
table also shows that the restrictions for industrial boilers of nominal 100 MBtu/hr (29.3 MW)
capacity range from 0.04 to 0..6 Ib/MBtu (17.2 - 258 ng/J) and those for utility boilers of nominal
1000 MBtu/hr (293 MW) capacity range from 0.02 to 0.6 Ib/MBtu (8.6 - 258 ng/J) with the exception
of New Mexico, which expects to reduce their limits.
Opacity limits are presented in Table 6-4 for fuel burning sources. Many states now limit
opacity to 20 percent (Ringelmann #1), although some restrict visible emissions to 10 percent and
three even permit no visible emissions. Frequently a more lenient limit is placed on existing
This conversion factor was used in "Analysis of Final State Implementation Plans" (Reference 6-9).
It assumes 5-10 percent excess air as in large finely tuned boilers. At 50 percent excess air
the converstion becomes 1 grain/scf = 2.2 Ib/MBtu.
t
For comparison, standards of performance for new boilers >250 MBtu/hr limit particulate emissions
to 0.1 Ib/MBtu (40 CFR 60.42). These standards also limit visible emissions to 20 percent opacity,
except for a period of up to 2 minutes during each hour when they may discharge a plume with 40
percent opacity. It has been stated that any boiler which burns an oil whose ash content is less
than 0.45 percent (by weight) should need no particulate control device to meet the gravimetric
standard, and if it meets the gravimetric standard, it should also meet the opacity limit, except
maybe during soot blowing (Reference 6-2). Moreover, many units have been shown to be capable of
emitting less than 0.1 Ib/MBtu (43 ng/J) as indicated in Sections 2.4.2, 4.3.1.1, and 4.3.2.1.
These low emitters are primarily those boilers which are larger than 1000 MBtu/hr (293 MW). Hence,
control technology exists to set limits which are lower than the standard of performance.
6-7
-------
TABLE 6-3, PARTICULATE LIMITS AT SELECTED HEAT RATES*, LB/HBru
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
D.C.
Florida
Georgia
Hawaii b
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
Hew Hampshire
Hew Jersey
Hew Mexico
Hew York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utahb
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Samoa
Guam"
Puerto Rico
Virgin Islands
Source Input llsat Rate, MBtu/hr
<1
0.5
0.083
0.167
0.5
0..1
0.3
0.1
0.1
0.6
0.6
0.4
0.6
0.6
0.6
0.6
0.33
0.6
0.3
0.4
0.33
0.15
0.10
0.10
0.3
£10
0.5
0.083
0.6
0.25
0.167
0.21
0.1
0.3
0.13
0.1
0.50
0.6
0.1
0.6
0.6
0.6
0.56
0.6
0.6
0.06
0.1
0.3
0.4
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.005
0.6
0.6
0.4
0.6
0.33
0.4
0.2
0.6
0.3
0.6
0.1
0.5
0.4
0.33
0.34
0.15
0.10
0.10
0.3
0.6
100
0.5
0.083
0.35
0.15
0.167
0.15
0.1
0.3
0.06
0.1
0.16
0.35
0.1
0.6
0.6
0.35
0.33
0.6
0.32
0.03
0.1
0.3
0.4
0.41
0.28
0.35
0.35
0.35
0.35
0.15
0.005
0.33
0.44
0.2
'0.35
,0.33
0.27
0.2
0.6
0.3
0.17
0.1
0.1
0.29
0.33
0.166
0.15
0.10
0.10
0.3
0.352
1000
0.12
0.083
0.21
0.10
0.167
0.10
0.1
0.3
0.035
0.1
0.10
0.21
0.1
0.1
0.6
0.21
0.10
0.6
0.3
0.02
0.05
0.19
0.4
0.26
0.14
0.2
0.21
0.20
0.10
0.10
0.005
0.10
0.18
0.33
0,1
0.20
0.33
0.1
0.1
0.6
0.3
0.10
0.1
0.02
0.17
0.33
0.10
0.15
0.10
0.10
0.3
0.21
'Abstracted from Table 6-2. Limits are for new sources whenever a state
has different regulations for new and existing sources. For further
detail see Table 6-2 and accompanying footnotes.
No limits on paniculate emissions from oil fired sources 1n state
regulations.
6-8
-------
TABLE 6-4. VISIBLE EMISSION LIMITATIONS FOR OIL-FIRED INDIRECT HEATING SOURCES8
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
D.C.
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Mary! and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Opacity
20
20
40
20b
20°
20
20
20
0
20d
20e
20f
209
30h
40
40
201
20
20
40
0
20
20
20
40
20
20
20
State
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
A. Samoa
Guam
Puerto. Rico
Virgin Islands
Opacity
20
20j
20
20
20k
20
201
20
20
20m
20
0
20n
20
20
20
20°
20p
20
20*
10
10r
20s
20
20
20
20*
Footnotes follow.
6-9
-------
FOOTNOTES FOR TABLE 6-4
aFrom Environment Reporter: State Air Laws. Volume 2, Parts 1 and 2. A looseleaf document pub-
lished and periodically updated by the Bureau of National Affairs Publishers, Washington, D.C.,
20037. Based on update as of November 1975.
^Emission limitation is 40 percent opacity for existing equipment.
cEach APCD has own regulations. Limits are 20 percent opacity, except for 3 minutes in any hour,
in South Coast Air Basin and San Francisco Bay Area Air Pollution Control District.
20 percent limit also for existing units (built or in operation on or before July 1, 1975) larger
than 250 MBtu/hr. Existing and new sources less than 250 MBtu/hr have 40 percent opacity limitation.
eUm'ts in operation or. under construction on or before January 1, 1972 are limited to 40 percent
opacity, except 60 percent for a total aggregate time of 3 minutes/hour.
^Existing units have a limit of 40 percent.
SExisting units have a limit of 40 percent.
"New sources larger than 250 MBtu/hr have a 20 percent limitation except 40 percent for aggregate
periods of 3 minutes during 60 continuous minutes, 3 times daily from one source location within
1000 feet radius from center point of any other stationary source.
Units in operation or under construction on or before January 1, 1971 are limited to 40 percent.
^Existing units have a limit of 40 percent.
k
Units in operation or under construction on or before February 1, 1975 are limited to 40 percent.
Existing units have a limit of 40 percent.
"Units in operation or under construction on or before March 1, 1971 are limited to 40 percent.
"Existing units have a limit of 40 percent.
°Existing units have a limit of 40 percent.
PUnits in operation or under construction on or before April 30, 1970 are limited to 40 percent.
^Existing units have a limit of 40 percent.
rUn1ts in operation or under construction on or before April 1, 1972 are limited to 40 percent.
sExisting units have a limit of 40 percent.
Existing units have a limit of 40 percent.
6-10
-------
sources. In addition, sources are allowed to emit more than the limits shown on this table for
short periods, of time, e.g., 3 minutes during each 1-hour period. In some states there is no re-
striction on the opacity of the plume during these exception periods, whereas in other states the
plume can obscure no more than 40 or 60 percent 'of the transmitted light. Some states apply these
opacity limitations to all sources, irrespective of their size.
6.3 SUMMARY OF SELECTED REGULATIONS
The last three tables presented regulations for all the states and other governmental juris-
dictions of the United States. The particulate loading and visible emission limitations represent
only a portion of the particulate-related regulations for each state (or each local air pollution
control authority, where the state was subdivided). More detailed information on the regulations
and enforcement procedures used by local authorities was obtained by direct contact from personnel
in nine agencies which were believed to have active control programs, as mentioned earlier. The
quantitative information obtained from these visits is summarized in Table 6-5, which includes
requirements for permits, continuous monitors, dust collectors, and specified fuels, in addition to
the particulate and vfsible emission limitations.
The enforcement procedures in eight of the nine districts visited appeared to depend heavily
on the use of construction and operation permits. These permits are used to identify sources of
emissions. The applications for construction permits are generally reviewed by the engineering
and/or enforcement staff in the control agency to determine whether the equipment and the proposed
operating procedures will enable the source to comply with local regulations. After the new source
has been installed, the control district inspects the facility to check its conformance with the
application and to observe the procedures used by the operating personnel. If the inspector's
report is favorable, an operating permit is then issued to the installation. Enforcement through
the use of permits is generally supplemented by roaming inspectors who look for violations of the
visible emission limits, by stack testing upon the request of the control district if a problem is
.suspected, by permanent installation of continuous opacity monitors in units which exceed a certain
size or which use residual fuel, and, in the case of Connecticut, by annual stack tests for all
units which emit more than 100 tons per year (TRY) total pollutants.
The regulations were generally stringent in these nine regions by comparison with many of
the states (compare Table 6-5 with Tables 6-2 through 6-4). For example, two of the regions pro-
hibited all visible emissions, and the other seven only allowed up to 20 percent obscuration,
whereas several of the other states permitted plumes to have opacities as great as 40 percent
(Ringelmann #2). With respect to particulate emissions the regulations were strict in that they
6-11
-------
ta
z:
CQ
UJ
=>
U.
13
E:
g
u_
ce:
C3
i
LU
CO
m
i
LO
Ul
I
C
re
o
o
t,
4-*
4/1
4-*
S
in
£
n
c
o
_J
re
!rr
0.
(U
o
re
Q.
^
^
^
QJ
01
'3
CD
to
_J
^
E= i.
3 O
O >
,
!~
"~3
1)
1
re
*>»
u
40
U
(U
c
o
5-
E OJ
JZ OJ fO
O fc. t r
r- re 3 .
.C -*=tU ,
Or- Or--^
to E E OJ E E
co re o in re o
M- JZ CO 4- £
0
U
m co
r- .
in
,
yj re (/) rij
C 4-> 1 C ,2 1
.° 3 3 .° o 3
l/J r U) TO (/) "^ ^ trt
r- Q. 0. C -r- >* CL C
EH- re e ^ «
j- C° 're s- °° 1o
0 A 0 A
(do C
Hj Q «- o
a. o~ 4-> aj
LJJ V) CL
Q£ c CD
S
I ' CO CO
-1
^,
^- CNJCO" 5
- - CO
ID
CO
O
'
O)
4-) * «
i v- CO
C
3
L.
^ 3
r CMC-o" ^
O
in
CM
L.
J=
^^ *5
, r CM <0 4->
* * 9^
o
QJ1 -r-
Ifl -r- i CO 3
r- If) , * - 4J
> V) CO
O'E ' " E
E S ^~
1-
"CJ t/l -C
<1) QJ
O) +J 4-J OJ
*~ ]i E ^
'c E ai
3 'CO Q.
CU £ (C (0
r- in i re c »
(/> C > 4-> O CM ^~
r- O Z3 I/) -r- - -
> -r- J3 C 4->
V) -i-
O VI >j
C «- C fc.
E « O
S-
o s-
-C
i CM in 3 40
(/J
OJ LO
a
(/> O)
4- C - . 3 S- OJ t.
LULOr OOJC OS-OISl-r-
COO S- 4j5-r- C4-J-aviaP
i i i OJ O. re 4-J -r- ( C <1)
coco c tuSro 4-icnJ.j-
»-«ootu uos-coc
^= ^ 5^-g 3ErSl £
LU re 3
-------
o
t-J
in
i
10
(M
=C
CO
ro
OJ
S-
E
ro
r-
'o
CX
O
£
O)
E;
to
O)
-p
ro
CO
.
ro
Q.
CD
o
ro
x:
Q_
o
s
CD
Z
10
OJ
*OJ
o>
<
CO
o
s^
E t-
ZJ O
O >-
4-3 ^
CD CO
i- ^
LU
C
O
P
(/I
O
CQ
ra
in
11
CD
|
C(L
CD
t/1
S-
CD
CU
s
JZ
ro
ZJ
'.p
g
c:
o
c
o
->
T3
L)
a
_i
" ' 1
CO CO CO CO
CD CD CD O
Q. LO CD
*3" *d- CM i
CD CD CD CD
CT
LO LO LO LO
O O O O
a.
CD
o
CM CM CM
000
CM CM CM .
O O CD CD
LO
CD CD O O
CO
O
CD
| CM CM CM
O CD O
-p
UJ . ' -i S_
H- C -C
CM
O
CM
CD
CO
O
p
CQ
z.
x7
LO
LO
CD
CM
O
CO
CD
Distillate
by weight)
&«
CM
O
CO
XI
f
C
10
>,
CM
O
CO
XI
CM
CD
CO
XI
C
C
OJ
CM
0
CO
XJ
L
L
C
ue Cleaning
u_
ZJ
23
)
ZJ
-P
CQ
E:
3
CQ
O
D
3
CQ
s:
o
o
~5
SM~ -r-
O O
£
D 3 CO
- CD
6-13
-------
TABLE 6-5. Continued
Footnotes
aMinirmim size (MBtu/hr) above which permit is required.
bRemoval of rotary cup in sources >13 MBtu/hr by 1976. Also, no new rotary cups after July 1974
anywhere in the state.
cList of exemptions includes space heating furnaces and gass or LP6 boilers <250 MBtu/hr.
Also require source testing at least once every two years for all sources which emit more than
100 TPY total all pollutants.
eRenewed every three years if pass stack test.
Limits in Ringelmann Number (U.S. Bureau of Mines Chart) unless otherwise stated. Most regula-
tions allow the use of equivalent opacity. Time for exception is number of minutes exception is
allowed during any consecutive 60 minute period.
%) visible emissions for sources <200 MBtu/hr or from stacks with internal cross-sectional dimen-
sion of less than 60 inches.
Commercial boilers limited to Bacharach No. 3 at a maximum 20 percent thermal loss through stack.
Photo-electric monitor with audible signal.
JSmoke meter with audible signal at Ringelmann Number 1.
Unless only gas is burned.
Most regulations also required the use of best available technology (control practices) to prevent
fugitive dust (e.g., from collector solid discharge).
"Vor units larger than 5 MBtu/hr.
nDust collectors also required for new residual fired units larger than 13 MBtu/hr and existing
units larger than 200 MBtu/hr. Collection efficiency requirements specified (Table 1, Booklet
10.03.39). Users of residual fuel during gas interruptions only do not need dust collectors.
°For units larger than 3 MBtu/hr.
*V\ source is exempt if it can be proven that it will not exceed the established ambient air quality
standards. Erie County limit of 0.1 Ib/MBtu applies to all units >250 MBtu/hr.
%ased on regulations given in grains per cubic foot of gas (calculated to 12 percent C02 with
assumption that fuel is CH« o) anc' 10 Ib/hr limit independent of size.
Based on regulations given in lb/1000 Ib stack gas adjusted to 12 percent CO, and assumption that
fuel is CH2 2. i
S0.10 Ib/MBtu if S02 control device installed.
In Baltimore and Washington, D.C. metropolitan area.
"Given as 10 gph for No. 4 and 20 gph for No. 6 oil.
6-14
-------
TABLE 6-5. Concluded
Footnotes (concluded)
- *i
'Provided the service retains -its capability to fire a lower sulfur content fuel , does not create a
localized problem due to downwash, and is equipped with monitors. Relaxatnon from 0.5 to .IX sulfur
I UL.Q I I i.CU IJI \ju i tin vi wv. \t*j MWIII !---J _-.. - - ,---, ,
content due to expire on July 1, 1977 unless extended.
wNo source may discharge sulfur exceeding 0.2 percent by volume as S02 at the point of discharge.
6-15
-------
imposed relatively low limits on all sources, and that limits were applied even to the smaller units
in the district. Thus, in some regions all units had to conform with the particulate loading limits,
irrespective of sMze. In all regions which were studied in detail, burners that were larger than
5 MBtu/hr (1.46 MW) had to comply. Some states still differentiate between new and existing sources,
whereas others did for only the first few years after the enactment of the regulations. In these
cases a compliance schedule was included in the regulations giving existing sources a fixed time to
bring their units down to the same level as new sources.
Most of the regions contacted also had stringent restrictions on the allowable sulfur con-
tent of the fuel. Since desulfurization also results in the removal of some of the solid matter in
the fuel, and in a decrease in weight (increase in API gravity) the sulfur restriction can have a
beneficial side effect by reducing sources of particulate emissions.* Three of the nine control
districts also prohibited the use of residual oil in small units. Combustion cannot be controlled
as well in these small units as it can in the large ones, nor is the temperature in the combustion
volume as high. Therefore, it is more difficult to obtain low particulate combustion in small
burners using residual fuel. Thus, this restriction on the use of residual oil eliminates one
potential source of large .particulate emissions.
In the remainder of this section we discuss the regulations in these nine control districts
as they affect each of the four user categories.
6.3.1 Residential Units
In general, the emission regulations in these nine districts did not apply to residential
units. For example, permits were not required for units smaller than 1 MBtu/hr (.293 MW) in most
districts: only New York City required them for units that were less than 1 MBtu/hr (.293 MW), and
their minimum limit was .35 MBtu/hr (.10 MW). Thus, in New York City only the very largest of the
units which have been classified as residential in this study would be subjected to the require-
ment for permit. However, if the unit was in a one or two-family home, the owner would still not
have to obtain a permit for it, even if it exceeded the .35 MBtu/hr (.10 MW) limit.
As noted in Section 3, particulate emissions from correctly maintained and operated oil fired units
depend largely on the carbon residue content of the fuel. Although desulfurization of a given oil
will reduce its carbon content, the amount of carbon residue in any given oil depends more on the
nature of its crude source than on the sulfur content of the refined product. There is no known re-
lationship between sulfur and carbon residue content of crudes.
6-16
-------
The exception to this general lack of regulations for residential units is in visible emission
limits. Most air pollution control authorities include residential units in their opacity regula-
tions (that is, they use such words as "no person" or "no building or -installation" when they identi-
fy who must comply with the regulations). Maryland also limits smoke from residential units to a
Smoke Spot Number of 2 or less. The other regulation that can apply to residential units is the
fuel restriction. As mentioned earlier, several regions prohibit the use of residual oil in resi-
dential size units.
None of the air pollution control authorities contacted required mandatory service or main-
tenance procedures for residential units. However, several of the staff members in the control
districts believe such a requirement would be a good one, especially if it were also combined with
a mandatory licensing and certification program for servicemen. Both Boston and New York City
would like to follow such an approach, but they cannot do it now because of a lack of funds. Their
interest in this control strategy stems from their observations of existing service practices for
residential units. According to staff members from several of the control districts, some service-
men provide good service, and some do not. Moreover most residential units are serviced only when
the owners experience a problem with them. Such programs may be the next step which cities in the
New Jersey, New York, Connecticut, Massachusetts area need to adopt. Although their current regu-
lations for all other sources are generally considered to be quite stringent, many of the large
cities in this area have not been able to meet the NAAQS for particulates. Space heating is be-
lieved to be one of the sources which could be controlled more efficiently.
6.3.2 Commercial Boilers and Furnaces
All the control districts contacted during this study had both opacity and particulate limits
which applied to at least some boilers and furnaces in this user category. In a few regions units
which were smaller than 1 to 5 MBtu/hr (.293 - 1.46 MW) were not subjected to the particulate load-
ing regulations. New York City requires commercial boilers to pass a Bacharach smoke test with a
reading of Number 3 or less and a stack loss no greater than 20 percent of the heat input. In
addition, new units must show that they can achieve an overall thermal efficiency of at least 75
percent and a C02 reading of 12-1/2 percent without exceeding the Bacharach No. 3 limit. This
joint stipulation of a smoke reading and a thermal efficiency is important because many people
elsewhere satisfy smoke limits by the use of excess air. Such an approach reduces the thermal
efficiency and hence increases fuel consumption. At best it leaves mass emissions unchanged, but
6-17
-------
J
it probably increases them as a result of the increased fuel consumption.* New York City also
requires the use of a forced draft system and a continuous smoke monitor. By comparison Erie
County, in New York, which follows the state regulations, has no smoke number or thermal efficiency
limits, no requirement for forced draft fans, and only requires continuous monitors for boilers
with capacity greater than 250 MBtu/hr (73.25 MW). As mentioned earlier most of the regions con-
tacted in this study did have sulfur restrictions on the fuel that could be burned in their areas,
and hence indirectly reduced the emission of particulate matter from oil burning equipment.
Maryland is the only region of the nine studied in detail which attempts to limit emissions
by an equipment standard. They recently banned the installation of rotary burners in new units in
the 1 to 13 MBtu/hr (.293 - 3.8 MW) size category, and required that all units throughout the state
phase out rotary burners by 1976. This equipment standard was based on their observations that
rotary burners require more maintenance to insure low particulate emissions than do other burners,
and that the operators of commercial size boilers are not likely to give these units the required
maintenance. Therefore, they decided to solve the problem by simply prohibiting the use of these
burners.
New York City also requires that new and upgraded boilers meet performance specifications and
be installed according to certain criteria (Reference 6-3). For example, all units must be able to
demonstrate adequate draft over a range of outdoor air temperatures from 11 to 94°F (-12 to 34.2°C),
and they may not be constructed with air-cooled walls. Moreover, if the boiler system is for an
installation where the maximum load is going to be greater than 50 gph (52.5 cmVsec) and if the
load will vary to less than 50 percent of the maximum, then multiple boilers must be used. Maximum
and minimum heat release rates are stipulated by boiler type, and performance specifications are
given for burner flow rate control, interlocks with the air fan, oil preheat equipment and tempera-
tures, combustion air availability, and system controls, among others. These design criteria should
insure that all newly installed units will meet the emission and thermal efficiency specifications
(i.e., they should reduce the possibility that some units will fail to pass the Bacharach and stack
loss tests when installed) and that they will be built in such a way as to improve the chance that
they receive proper maintenance during the three years between permit renewals. In effect these
criteria show in a published document what standards the Department of Air Resource engineers will
use when they review an application for a construction or operation permit.
Paniculate loadings are generally corrected to a standard C0? concentration to overcome
this problem of dilution, neither Bacharach readings nor opacity readings are corrected in this
l^hi9"-Quantlta*ivf.°Pacity measurements could be corrected to a standard CO, or 0, concentra-
tion if the concentration of C02 or 02 in the exhaust is measured in addition t§ opac?ty
6-18
-------
'The enforcement procedures used in these nine control districts for commercial boilers and
furnaces varied from no activity after the granting of the permit to equipment restrictions to peri-
odic inspections. For example, Maryland, as just mentioned, requires permits and prohibits rotary
burners with the hope that by prohibiting hard-to-maintain units and reviewing all new installation
designs they will reduce emissions as far as practical. However, they do admit concern about the
effectiveness of their program due to their experience with boiler operators who do not service
their units as often as they should. New York City, on the other hand, sets emissions standards,
checks each unit when it grants the initial operation permit, and then checks it periodically again
(every 3 years) to insure that the facility's operating procedures conform to the permit. These
tests are in addition to their published engineering criteria. The other districts who use periodic
inspections also restrict their review to operating procedures and equipment. They only resort to
stack tests if they suspect a problem with the installation. During the inspection the enforcement
officer observes the boiler operator to see if he is following the correct procedure; he also checks
such operating parameters as stack and/or preheat temperature, draft pressures, functioning of the
forced draft fan, if required, and flame color, instability, and impingement. In New York City each
inspection also includes a test for Bacharach smoke reading and stack losses. Although several of
the regions require the use of continuous smoke monitors, most of them need not be of the recording
type and in any case these requirements are relatively new. Therefore it is too early to judge the
effectiveness of this requirement.
In order to supplement the above mentioned enforcement procedure, which relies heavily on the
skill of boiler operators, New York City offers a basic boiler operator course (see Reference 6-4
for the text). However, due to funding limitations, this is a very elementary course which only
teaches the operators how to run the boilers and perform routine maintenance; it does not instruct
them how to adjust or repair their burners. In a similar vein Erie County requires the operators
of the boilers in the public school system to attend a class on the subject of boiler operations.
In general the agencies feel that most commercial boilers and furnaces are in compliance with the
regulations.
To summarize, most control districts enforce their regulations by a combination of permit
reviews, a small group of roving inspectors who look for violations of visible emission standards,
on-site inspections, and tests if they suspect a problem or have to respond to a complaint. The
enforcement staff in the agencies contacted typically find 10 to 20 violations per month.
The differences among the control programs of the various regions seem to rest more on the
emphasis given to commercial units versus industrial and utility boilers than on what regulations
6-19
-------
they use. For example, in Connecticut, where enforcement is a state function and is based on
regulations that were passed in 1971, the approach chosen was to first address themse.lves to
the largest individual emitters. By controlling these large sources, Connecticut-officials felt
they could obtain the greatest'leverage in terms of potential emission reductions per man-hour of
enforcement time spent. New York City, on the other hand, has an extensive program for the reduc-
tion of emissions from commercial boilers. They stress control of these units because they are
used in the many apartment houses and office buildings within the densely populated city and, hence,
contribute significantly to the total mass of particulates emitted into the city's air.
6,.3.3 Industrial and Utility Boilers
These two user categories will be treated together because the regulations and enforcement
procedures are virtually identical for both. The only significant difference is the quantitative
emission limits that units in the two categories must attain. As shown on Tables 6-2 and 6-3 most
air pollution control authorities impose lower limits on the larger units (based on Ibs of emissions
per unit input). Thus, typical values range from 0.06 to 0.6 Ibs/MBtu (25.8 to 258 ng/J) for in-
dustrial size boilers and down to 0.02 Ibs/MBtu (8.6 ng/J) for utility boilers larger than 10,000
MBtu/hr (2930 MW). Maryland also requires the use of a dust collector on all units greater than
13 MBtu/hr (3.8 MM) heat input and specifies the collection efficiency for the dust collector.
This requirement is based on the results of an evaluation they conducted of installed equip-
ment (Reference 6-5). They measured the solid matter that had been collected by a dust collector
on a sample of boilers in Maryland and found that these collectors were actually preventing the
emission of up to 50 lb/1000 gal (143 ng/J) in commercial size units.*
Naturally, all boilers in this size group are allowed to burn residual oil, but the sulfur
content in the fuel is generally restricted to less than 0.3 to 0.5 percent by weight. In Massa-
chusetts the allowable limits are 0.5 percent in nonmetropolitan areas and 0.3 percent in metro-
politan areas such as Boston. Four of the districts require the installation of continuous monitor-
ing smoke meters in all boilers in these two categories, and one additional region requires it only
for utility boilers. One state, New Jersey, prohibits any visible emissions from boilers with heat
input less than 200 MBtu/hr (58.6 MW) or with a stack whose diameter is less than 60 inches (1.5 m)
across.''"
For comparison the EPA emission factor for units burning No. 6 oil with 7 percent sulfur is 13 lb/
1000 gal (37 ng/J), at 3 percent sulfur, it is 33 lb/1000 gal (94.6 ng/J).
'''According to the Beer-Lambert law, opacity depends upon the concentration and the path length of
the dust-laden air through which the light must pass (i.e., the diameter of the stack). There-
fore, emissions at a given concentration are more visible in a large stack than in a small one.
6-20
-------
The enforcement procedures for industrial and utility boilers were generally the same in all
the regions contacted. Units in this size are the relatively few, easily identified ones. Moreover,
due to their^size, each source has a bigger impact on air quality-than does the average commercial
or residential unit and, hence, every region with a particular problem limits emissions from large
oil fired boilers. Since the large boilers are usually operated and maintained as economically as
possible by qualified, full-time staff, particulate emissions are usually not due to poor operation
or maintenance procedures. Instead, if they exist, they are due to characteristics of the combustion
process, itself, the ash or sulfur content of the fuel, or a lack of flue gas cleaning equipment.
Any differences in enforcement procedures between the various regions are probably more subtle than
in the other user groups because the practices used for these sources tend to depend mainly on the
specific experiences gained by the enforcement staff in their dealings with industrial or utility
boiler operators.
The development of sophisticated techniques to identify those sources which are contributing
to high ambient particulate levels is also a distinguishing factor among the control districts. In
this regard Erie County is probably the most advanced, using a network of 32 manned high volume
samplers. The readings from each of these samplers are fed into a computer, which then generates a
pollutant rose plot. Inspection of this plot enables the control district to rapidly identify large
offenders. The ability of the system to discriminate among sources is limited more by their radial
separation, relative to each ambient air sampling station, than by their size. That is, the network
can only identify sources that are 30° to 40° apart when viewed from one station unless these sources
happen to be located in a part of the region where they have a large number of stations. Then they
can sometimes detect individual sources that are closer together by triangulation. In general, how-
ever, they could not detect an industrial sized oil-fired boiler which is located in the midst of a
large manufacturing complex.
6.3.4 General Comments
It is difficult to evaluate the effectiveness of the particulate control programs described
above for the nine selected regions for the following reasons:
0 Oil burners are just one among many other sources in a given region. A variety of indus-
trial processes (e.g., the iron and cement industries, transportation, fugitive dust
from industrial activities, and agricultural and natural sources, etc.) all contribute to
the ambient air concentration of particulates.
e Many of the particulate control programs are relatively new; therefore, they are not yet
fully enforced and historical data are not available to separate the effects of the
6-21
-------
control program from the other factors which cause ambient pollutant concentration levels
to fluctuate from year to year.
Particulate emissions from residential and commercial boilers and furnaces fluctuate sig-
nificantly during the cycle of each unit. They are high at the start and stop of the
cycle and lower inbetween. They are also non-uniform in larger units, being unusually
high during soot blowing operations. Therefore, one cannot assess the effectiveness of
the control program for oil-fired boilers and space heating furnaces on the basis of re-
ductions achieved in steady state emissions.*
Different regions need different control strategies because of differences in the distribu-
tion of sources. As mentioned earlier New York City needs to stress emission reductions in commer-
cial sized units as a result of the large number of apartment buildings in that area, whereas a less
densely populated, more industrialized area would emphasize emissions from industrial and utility
boilers.
One measure of the effectiveness of a control program is the stringency of emission limits
that are set by the local regulations. Inspection of Tables 6-2 through 6-4 shows that many states
do not impose as stringent emission limits on their fuel burning sources as do others. Therefore,
if those states with the more lenient emission limits are not able to attain the national ambient
air quality standards for particulates, they should at least consider the advisability of setting
new standards which are as stringent as those used elsewhere.
Since the purpose of this report is to develop model control strategies, we asked the nine
air pollution control authorities contacted during the course of this study whether they were con-
templating new regulations in order to improve their own control programs. Our survey showed that
this was generally not the case. Rhode Island had contemplated a regulation which required the in-
stallation of an electrostatic precipitator for oil-fired utilities. However, since most of these
boilers are capable of firing either oil or coal, since they already have an ESP for coal operations,
and since the same ESP cannot be used effectively for both oil and coal, it was decided not to im-
pose the economic burden of requiring two electrostatic precipitators on these sources. Los Angeles
The standard test procedure for residential and commercial units, which operate in cyclical
fashion, is to measure the smoke emissions during the 10th minute of operation after startup.
Thus, these measurements are made when the unit has reached a steady state operation and all
the surfaces inside are warm.
6-22
-------
is currently thinking of requiring utility boiler operators to install dust collectors if they are
not, themselves, able to solve the fallout problem that currently exists when they fire oil.*
The major differences, between the control programs in the metropolitan and in the state
agencies visited during the course of this study appeared to be in the stringency of the limita-
tions, especially those for sulfur in the fuel. , As one would expect, the metropolitan regions,
with their high concentration of sources .and human receptors, generally placed tighter restrictions
on the sulfur content of the fuel and lower emission limits on the sources than did the states.
There was also a tendency within metropolitan control agencies to place greater emphasis on the
control of emissions from commercial size units than was the case in the states. And finally the
size of the enforcement staff in the metropolitan control districts was generally larger than it was
in the state regions. For example, Philadelphia had approximately the same number of inspectors
as did the State of Connecticut. Of course, the need for a large staff is a consequence of the
need to control the many smaller units which exist in the metropolitan areas. Beyond these three
areas, the major differences were individual and probably due to the approaches preferred by the
people who establish the programs. That is the use of an equipment standard in Maryland and a per-
formance standard in New York City is probably not at all related to the fact that one is a state
and the. other a metropolitan area.
The relationship between metropolitan and state control authorities also differ from state to
state. Thus in Maryland and Massachusetts the state regulations include special provisions for the
metropolitan districts. However, enforcement is frequently accomplished by local districts (e.g.,
Maryland and Erie County in New York) who accommodate local requirements by adapting their enforce-
ment procedures to those requirements (e.g., by developing an effective, computerized monitoring
system to identify problematic sources). On the other hand, New York City, Los Angeles, and
Philadelphia all developed their own regulations for emission limits and enforcement practices
themselves.1" No attempt was made during the study to investigate the potentially sensitive area
of the working relationship between metropolitan control districts and the state air pollution
authorities, and particularly not the state agency's role as responsible body for the quality of
the air within the state and as intermediary between the local air pollution authorities and the
U.S. Environmental Protection Agency.
Recall that Los Angeles has not had to deal with the problem of particulate emissions from oil-
fired equipment until recently because most of their major sources used to burn the cleaner
natural gas. - - , . ,
^Philadelphia actually enforces the State regulation whenever it is more stringent than the
city's own laws.
6-23
-------
This concludes our summary and review of existing control strategies for particulate
emissions from oil-fired boilers and furnaces. In the next section we present possible control
measures which draw from the experiences gained by agencies which have been most active in control-
ling particulate emissions as well as from the knowledge about the characteristics of equipment to
be controlled (see Section 2), the fuel normally used by this equipment (Section 3), and the con-
trol techniques available for reducing particulate emissions from oil-fired boilers and furnaces
(Section 4).
6-24
-------
REFERENCES FOR SECTION 6
6-1. Source: Environment Reporter: State Air Laws. Volume 2, Parts 1 and 2. A looseleaf
document published and periodically updated by the Bureau of National Affairs Publishers,
Washington, D.C.,'20037. Based on update as of November 1975.
6-2. "Control of Air Pollution from Fossil Fuel-Fired Steam Generators Greater than 250 Million
Btu per Hour Heat Input," EPA, for technical review only not for publication.
»
6-3. "Engineering Criteria Fuel Oil Burning Equipment," City of New York Department of Air Re-
sources, May 23, 1972. Available from The City Record Sales Office, 31 Chambers Street,
New York, N.Y. 10007, for $2.25.
«
6-4. "How To Stop Smoking," An Air Pollution Control Guidebook for Boiler and Incinerator Opera-
tors from the New York City Department of Air Resources.
6-5. Robison, E. B., "Application of Dust Collectors to Residual Oil Fired Boilers in Maryland,"
Bureau of Air Quality Control, State of Maryland, BAQC-TM-74-15, December 1974.
6-6. Barrett, E. R., Miller, S. E., and Lock!in, D. W., "Field Investigation of Emissions from
Combustion Equipment for Space Heating," Battelle, Columbus Laboratories, prepared for U.S.
EPA, Report No. EPA R2-73-084a, June 1973.
6-7. Levy, A., et al., "A Field Investigation of Emissions from Fuel Oil Combustion for Space-
Heating," Battelle, Columbus Laboratories, API Publication 4099, November 1, 1971.
6-8. Barrett, R. E., Locklin, D. W., and Miller, S. E., "Investigation of Particulate Emissions
from Oil-Fired Residential Heating Units," Battelle, Columbus Laboratories, prepared for U.S.
EPA, Report No. EPA-650/2/74-026, March 1974.
6-9. Duncan, L. J., "Analysis of Final State Implementation Plans Rules and Regulations," The
Mitre Corp., prepared for U.S. EPA, Report No. APTD-1334.
6-25
-------
-------
SECTION 7
POSSIBLE CONTROL MEASURES FOR FUEL OIL COMBUSTION
7.1
INTRODUCTION
This section presents control measures for fuel oil combustion which local or state air
pollution control authorities can implement to help attain and maintain particulate national ambi-
ent air quality standards (NAAQS) in their region. These control measures for oil-fired boilers
and furnaces are expected to be needed in major urban centers where particulate matter emissions
from the use of oil are an important part of the total inventory. Continuing economic and demo-
graphic growth in all regions and the trend in some regions of the country from natural gas to
oil will increase the impact on the ambient air due to oil burning installations, and, therefore,
use of the control measures described herein will likely expand in the future.
Since the contribution of any source category to the total emissions in an area varies from
region to region, and since the seriousness of the problem (e.g., the current and anticipated ambi-
ent air concentrations of particulate matter) also varies from region to region, control programs
must be area specific. That is, each local or state control agency must evaluate its own situa-
tion (source and population distribution, ambient monitoring stations where NAAQS are being ex-
ceeded or are expected to do so in the future, characteristics of the particles captured by these
samplers, standard that is exceeded (24-hour or annual), meteorology, topography, projected growth,
etc.) to determine which user categories need to be controlled and the degree to which each of
these needs to be controlled. Therefore, possible controls for each user category (residential,
commercial/institutional, industrial, and utility) are presented separately. Within each cate-
gory, the possible control measures are ranked in order of increasing effectiveness, taking into
account the emphasis on future sources of emissions, cost, and public acceptance. This set of
ranked control measures are intended to serve as guidelines to local and state regulatory agencies
who must, however, evaluate, their applicability in light of the conditions in their own district.
Presentation of these control measures in the above format provides a control district with
information that they can use in one of two ways. If they already know which user category they
have to control and by how much (i.e., from analysis of Hi Vol catches, as noted above), they can
7-1
-------
use the information in the lists of possible control measures as input to a systems analysis and
participate dispersion model to determine the most cost-effective control program for their region,
considering all sources of particulate emissions within the region, the current and projected am-
bient air concentrations, and the projected economic and population growth rates.
The possible control measures are presented in tabular form with text to explain and justify
them. A separate table with accompanying text is used for each size category. For regulatory pur-
poses the class of boilers and furnaces is divided into four categories based on size; the names
residential, commercial, etc., are used here merely to facilitate references to a particular size
category. Thus the controls which are applicable to boilers and furances in the 0.4 - 12.5 MBtu/hr
(0.12 - 3.66 MW) category apply to any unit in this size, irrespective of whether it is located in
a commercial or an industrial facility. This practice differs from the one used to classify boil-
ers for the NEDS emission survey. That system attempts to categorize boilers on the basis of their
application with no regard to size. However, a division according to size is more meaningful for
regulatory purposes because emissions and emission control techniques depend more upon size than
application. Moreover, a distinction by size facilitates the enforcement of regulations. The limi-
tation of two-family homes for residential units is an exception to this general rule.
The control measures described in the following sections exclude most surveillance and en-
forcement procedures that are either not specific to particulate control (e.g., stack testing of
large sources) or that are applicable to all size categories. In particular, they do not mention
the use of visual observations of plume opacity by roving inspectors. These inspectors can be
mobile, stationed on the top of large buildings, or airborne in light planes or helicopters, and
they can be equipped with cameras for later analysis and use in notifications of violations.
The format of the tables which are used to present the potential control measures is
shown and explained by Table 7-1. In most cases the intent is for the control measures to be
applied cumulatively. That is, if a local air pollution control agency decides that measure No. 1
is not sufficiently effective to solve their ambient air problem, they should consider next the use
of both measures No. 1 and 2. If these two still do not reduce emissions enough, they should then
investigate the simultaneous implementation of measures Nos. 1, 2, and 3. Where measure No. 2 is
simply a more stringent version of measure No. 1, the second one automatically includes the first
one.
All the control measures discussed here are based on data that were presented in Sections 3
and 4.
7-2
-------
TABLE 7-1. POSSIBLE PARTICULATE CONTROL MEASURES FOR
BOILERS AND FURNACES: "SAMPLE FORMAT"9
Rank
1
2
Control Measure
o.
OJ
t- S
3
0> U
3 -M
u» U
£-S£
§*S
H- (-
o a. 3
QJ O
4J 4J CA
C M
IE tn Ic
+-> t- 4->
#0 »-
VI O
> t>
o ««*-
4* C tA
*>£ £
.Q C f*
S i i
^r E aj
i
CX (A
E 4J i
-88
t- cn
O (A
M- *- W
£-1
3 »*- 4-*
I! aT"S
l-i
§-".2
o
C OJ
4-> O -C
S -M
S "^ O)
M
cn c o c
T- 4-» O
« § 5 £
*> *£» *O OJ
2«O *- -C
4-> **- 4->
.£) C <»-
Policy Instrument
>
O)
S'
o
-s
0)
0)
c
CA
T>
^0)
- 3
i i-
*» 4->
r- C
Q
L. +J
O -
S- V)
3 tO
cn cn
Q) Q>
Q£ i
Effectiveness
Reduction
per unit,
*
m
c
a
§s
"r"^_
4-* C
(J O
3 O
TJ
aj in
_g
4-1 4->
c
ai >>
CJ "O
a. a>
OJ O
cn a>
ra **-
S- "4-
OJ rg
nj Ol
U
11
ra cn
E
,p j_
4-* U
CA *t»
UJ 0>
£
3
K
|
C
O
u
(A
4-*
S
s
o
cr
o
u
3
a
I
o
*J
u
Reduction
Per Air
Region0
i
c: u
r- QJ
tn o> i»
ffi
» o
m
>> u
O (U
CD U
QJ i.
4-> 3
<9 O
(J (A
Q) CO
3 4^ 3
: O «
tA *4- tl
o QJ
(A E
5-2°
£ U 4J
| ^^ 0
*"
Cost Impact
User,
5s
c:
cn
c:
E
Of
CL
O
3
c:
to
L.
O)
4/j, »
3 QJ
a to
4-) *C
Ql O
(t- Q.
(J tA
it! fO
a»
I/) »-
o-
APCO
V
u
o
Q.
O
4-*
-o t- o
(U U
0 >,
S- J3
M- T3
C QJ
QJ *J
(13
(*-
O 3
cn
C OJ
o s.
-*J Wl
o u
^- i-
O 3
C 0
Pnppnu
Impact,6
%
>» Of
^ c
QJ
C
O (A
r- (O
§0^"
3 -
cn cn o
O - OJ
(j tn N
X cn t-
cn c 3
QJ x: »
C (J 3
»
^ *^" QJ
o cn c
U C *r-
CA "O QJ
r- 3 1-
4J (J >)
C -O
4-> Q) E
(J L) 3
rO L. (A
O. 3 C
E O O
Public
Acceptance
S .1 c
4-> »~- O 4J
O « 4-^0
4-» 4J (J CA
C QJ QJ *4~
o -a GZ '-
4-> (/I C
(JO) QJ
(0 i. >-, >
QJ -C QJ
4- 4-> 4->
3 ro -a
E 4-»
(A >» 0
ft r- <*-
O- t- W>
QJ O. CJ
< ra QJ a.
M- -r-
QJ t-H >> (A >,
4J 4-> O i
E 'o E "c
cn ^ -Q *r-
(S f^~ o tn o a> ex
T3 C «- 1- -Q O
3 O \C J= fO QJ
"S-
_,
aSee Appendix C for conversion to SI unit.
bSSH « Smoke Spot Number
"Relative to area-wide emissions from all sources 1n this category only; Impact on
region depends on relative importance of this source category.
dSamp1e calculation of cost Impact on annual 1 zed basis given In Appendix D.
'(Increase) or decrease 1n efficiency.
Judgement; no data to substantiate.
7-3
-------
7.2 RECOMMENDED CONTROL STRATEGIES
7.2.1 Residential Boilers and Furnaces - 0 to 0.4 MBtu/hr (0 to 0.12 MM)
For the purposes of these recommended control strategies residential units will be defined
as those hot water or steam boilers or space heating furnaces whose rated heat input is less than
0.4 MBtu/hr (0.12 MW) or which serve only one or two family residences. Nationwide, approximately
one-quarter of all residential units are oil-fired, and this number will probably grow as gas be-
comes more scarce. The average age of the oil-fired units is about 15 years, and approximately
one-quarter of these units have had their original burner replaced. Even though this source cate-
gory does not now contribute significantly to the total particulate emissions in any ACQR (the larg-
est contribution is 3.4 percent of the particulate emissions from all sources in one ACQR - see
Table 1-1), they are concentrated in densely populated areas and are used mainly during the winter
months of the year. Therefore, their impact is greater than the annual area-wide emission averages
would indicate. They do pose an enforcement problem, however, since there are so many units, they
are widely dispersed, and they are owned by people who are not generally knowledgeable about their
operation or about the relationship between their operation and emissions.
For the above reasons, air pollution control authorities are encouraged, as a minimum, to
engage in public information campaigns ("PR") which stress the value to the individual owner of
having his burner serviced annually. Periodic burner servicing can be a useful way to reduce emis-
sions because tuned and well maintained burners emit less particulates than those just left by
themselves (see Subsection 4.2.1). This is especially true if burners which are found to be dam-
aged or worn out are replaced. Despite the general awareness of increasing air pollution, individ-
uals still seem to be reluctant to spend funds voluntarily to reduce their own contribution; there-
fore, programs which stress economic gains and treat pollution reductions as a side benefit will
more likely be accepted than those which do not.
Since oil suppliers have ready access to individual homeowners (their customers), air pol-
lution control authorities should make a special effort to induce these suppliers to develop similar
information campaigns or to intensify existing ones. All publicity campaigns for improved burner
maintenance should be accompanied by exhortations to conserve energy. These campaigns should re-
mind the homeowner of the need to lower his thermostat setting in the winter, to use less elec-
tricity and hot water throughout his household, and to install or increase insulation of his home.
Energy conservation is a valid air pollution control measure because particulate emission rates
are generally a direct function of fuel consumption.
7-4
-------
All regulatory programs have two parts to them - a standard and an enforcement procedure.
Standards can be either an emission limit, a fuel specification, or an equipment standard, .and each
of these cart be used in combination with one or both of the others. No equipment standard is yet
workable for residential units because there does not appear to be any particular design feature
or equipment type that stands out as being necessary to the attainment of low emissions. For ex-
ample, as noted in Section 4.2, emissions from the integrated optimized burner/furnace combination
are expected to be as low as those from the furnace with the blue flame burner. Moreover, the
blue flame burner is a proprietary item and therefore could be produced by only one manufacturer.
Hence, if an APCD requires the use of a furnace with a blue flame burner in new buildings, it would
preclude installation of the.integrated optimized burner/furnace, which is equally effective, and
thereby prevent other manufacturers from entering this market. Even worse, such a specific equip-
nfent standard might inhibit the development of a better unit. Admittedly it is easier for a con-
trol district to plan and enforce an emissions reduction program if it can rely on equipment stand-
ards because then they do not have to test or analyze each different kind of unit. However it is
felt that the other factors outweigh this advantage.
The potential control measures for residential furnaces and boilers are presented in Table
7-2. Although all reasonable control measures have been included for residential units on this
table, there is serious doubt that any air pollution control agency will find it necessary or use-
ful to implement the lower priority ones.
The first regulation which air pollution control authorities might implement to reduce par-
ticulate emissions from oil-fired residential units is one which sets stringent limits for new and
replacement burners. This priority on new units is consistent with the basic philosophy of this
whole document, which is to emphasize emission reductions from new sources to offset growth and
allow an area to maintain the quality of its air. The control measure requires that smoke emis-
sions be limited to no greater than a Smoke Spot Number 1, and that this level be attained simul-
taneously with a flue gas concentration of C02 that is at least 10 percent. This C02 concentra-
tion corresponds to an excess air setting of about 50 percent and prevents the inefficient use of
additional dilution air to meet a smoke limit. Atrpresent these limits can be met fay the furnace
with the blue flame burner, by burners with flame retention devices, and by many other burners
which are now commercially available. In the near future the integrated optimized burner/furnace
combination will also be available and should be able to reach the same levels. Moreover, the ANSI
standard Z91.2, which has recently been approved, also specifies these levels as an industry goal.
7-5
-------
o
o
o
ce:
o
o
t/i
01
t-Lj
O
CO
ce:
o
CO
LU
CC
I
LU
O
O
Cu
LU
_I
03
HH
CO
CO
o
0.
cvl
_
CO
a:
o
a
i-
3
t/V
&
«i
u
HI
*2
i
u
fit
o
S 0.
"S
»%
ll«
"~
a
^
t-
J*^
O I. U
Isi'l
go. a:
ou
tj§«'
I b
cc a.
Bf
I
i
°s
3
i
I
c
s
i&
*~ S~.-
~ *j . ^
S"C 1
g" o
fll*r- 1. **~
r- 4J O
a. u e .g
m +j £^
i/5 tfl«£ O
,
J
O*
S
ifi
s
J
o ai in
^ 3 01
** " S
s^
s?
11
SI
O
ai
x *T
2 oS
SC3
M « O 0
3 O ^~ r
P" *5 -o
£E"o c
LU 01 0* O >£
t!g°
IA 3C
r- ,
U O ~Q
III
t-
is
o
g
~*
^
s
1
o
o
'M
cn
S
l«
'=
y>
i
«
il
W) -r-
X 01
lations, 1
1
Q
S:
^
o
S'o
j3 S" i
c »- 0
|| s
SE-S
O 01
C +J U
sli
r- tn 4)
Itf^- S
S o -^
g-ss
ro
Of
O > SI £
01 a'" ?"
> §" « j
o S
^ o «^ ti
*"« &
L. » u «-
ai vi c «3
c a u '"-t,
g J3 U >!*->
co ,2 |!Z |^ 3!
o. O..O A c:
X 01 3 3 O
LU i- o. n. u
i_
u-)
O
c:
&
3;
0>
to CT> C
we O
*~ m 3 §
I_ wi i ro
o"*oi *->
CO 3
<*- 0
01
o
1
X
C
o
TI
E
O
s
-
^ (/»
1
."^'oT
c >
c
g|;?
.S.HI
E nj O
irti s"
T IA"I»
££';?i
^
^-
^" iT c
_ QJ 01
01 C >
p UJ -f- T!
>\ F t *"
i|^|
V
/) TJ +J E
d c s.«
1
CO
o
LO
g
""*
+ O VJ
ggs
ss^
3
O O t«-
1
o
o
1
s
X
CM
^
z
5
i
*OJr~
3 -r-
U- O
u,
§y|°|
S LU ^ SI 3 -^
« S i"^ "c
5 S^ S *
S 3 -r- 1- U
SO -> *
r- . 4J U 01
Q VI -O r- « -r- D
S "~ (0 3 13 O
T3 C *- *) ** -C
O T) T) *J
r U E '"- Vt
Sw --^^: i5 >/> c
C CM *-" U W _O
t.
o
s
o
g
~*
o
O M~
+J
Q>
O
0
o
?£
0
O - XJ
M IT) "3
nj C -0
|p
Q C 01
< 4-> C
C (A
01 t-
E T-
S^
CL
"S3
4Ji
0" 3
> Of O)
o u
1 ol
'13 c ^
S5T
"§ « ti
o *j -5
to
1 5
1
1-
0
*0>
1-
g
-o
1
?
c.
4->
a
- 0
s i
m >» a
C en C
s -
s; s §
+J f O>
t/1 *J (/)
I ^ J
S^~ -o
01
O J- N
' g r- | 0*
(A ECU
ftJ O O *
1 i: w «;
CD. +J
o -^ -^ S
^ 5 « 2
0) 1= O 3
g t. S 0 >
O J3 01 «*- O
(j E -O O *J
O ^ 1 O *-*
0 & £ "5 ^ "^
X "^ "" *3 c
r- U) O U
S § 0) U C C
"* " nj "o. +J 01
t/i co oe to Z3 j*^
7-6
-------
A no-visible-emissions limit can be included for enforcement purposes. The appearance of
smoke from a residential furnace or boiler indicates that the unit is malfunctioning and broadcasts
this fact to anyone outside the house. A visible emissions limit gives APCD inspectors a firm
legal basis for their actions when they advise a homeowner that his unit is malfunctioning and re-
quire that he correct the problem and show compliance.
If the regulatory agency (or the EPA) can obtain the cooperation of the American National
Standards Institute (ANSI) or the Underwriters Laboratory (UL) in the form of an agreement to test
prototypes of new units for their ability to comply with the limits stated above, then the regula-
tion should include a requirement that all new or replacement burners carry a label from ANSI or UL
which states that the unit belongs to a family of burners which has been certified by one of these
labs as meeting the local emission limits. EPA is presently evaluating the practicality of such
agreements. Such a cooperative effort would reduce the burden on both the local regulatory agency
and the local burner suppliers. Moreover, if ANSI and/or UL are willing to measure particulate mass
emissions during burner certification, tests, the regulation should also include a mass emissions
limit (given the difficulty and expense of mass emission measurement, it would be unrealistic to
set such a limit unless coupled with a burner certification program). The value of 0.01 Ib/MBtu
fuel suggested in Table 7-2 is based on results with several retention heads.
Two methods are available to implement this regulation in addition to the cooperative effort
with the standard setting laboratories. Since architectural drawings for all new buildings have to
be approved by the city or county building department, performance standards for new burners could
be written into the building codes and their installation policed by the building inspectors (after
proper training). Replacement burners, however, are generally installed without the need to notify
any governmental agency. Therefore, it is suggested that homeowners be induced to install these
improved, low emission, high efficiency burners by means of a property tax reduction (which could
be offset by an equivalent general tax increase). To obtain this reduction, which would be some
fraction of the incremental cost of installing a low emission burner, the homeowner would have to
obtain a form from the local APCD stating that his unit has been approved by these authorities. He
could then include this form with his property tax payment to justify the reduction in taxes. Al-
ternatively, the local air pollution control authorities could enforce the requirement that the
replacement burners meet the stringent emission limits by spot-checking the parts storage facili-
ties of the local burner suppliers. The purpose of these spot checks would be to insure that these
burner suppliers stock only the approved kinds of burners. Similarly, the control agency could re-
quire service people to keep records of burner replacements and then spot-check them.
7-7
-------
This control measure can be extended to cover more sources by requiring that the burners
1n all houses that are sold meet the stringent limits. Since the seller presumably has established
a working relationship with his oil supplier or burner serviceman, the burden can be placed upon
him to prove that his system complies with this regulation. Proof of compliance would be a pre- ,
requisite to recordation of the transfer of title. Such a regulation would be similar, for ex-
ample, to one which pertains to automobiles in the State of California. In that case the seller
of a car, if it is in a certain age group, must obtain a certificate stating that his automobile
has been tested and found to comply with the air pollution control regulations before the Depart-
ment of Motor Vehicles will issue a registration in the name of the new owner.
The maximum estimated unit effectiveness of 50 percent shown in Table 7-2 for this control .
measure is based on the assumption that a reduction from Smoke Spot No. 2 to Smoke Spot No. 1 is
roughly equivalent to a 50 percent reduction in particulate mass emission rates. Virtually all
currently available residential burners (distillate oil-fired) can achieve a Smoke Spot No. 2 and
many can reach a level of 1. This proposed regulation, therefore, is intended to insure that all
new units which are installed are of the lower emitting variety. Since this control measure only
affects new and replacement units, its impact on the total emissions to the area will only be felt
in the future. By the time it has a widespread impact, however, it should be a significant one,
since the total emissions from the residential category would be reduced to half of its current
level. Alternatively, the number of residential oil-fired units in a given region could double in
the next 10 to 20 years, either because of population growth or conversions from electric or gas-
fired units to oil, without increasing the total particulate emissions from residences.
The estimated cost impact shown in Table 7-2 for this first control strategy is based on
the assumption that it will be achieved either by the use of the flame retention device or by the
use of a new burner/furnace combination, such as the blue flame or the optimized units. These
furnaces are estimated to cost up to $100 more than conventional furances. This extra initial
cost would be recovered within 2 to 5 years, depending on fuel costs and annual usage, due to in-
creased efficiency of the new burners. Since this control strategy relies heavily on cooperation
with the standards laboratories, the use of the building permit system, tax reductions, and the
city or county recorder or clerk, it should not impose a significant additional enforcement burden
on the local air pollution control district.
In addition to reducing emissions by mandating the use of improved burners, this regulation
should stimulate the conservation of petroleum because these new burners are 5 to 10 percent more
7-8
-------
efficient than the currently available ones. This improved efficiency also results in a secondary
beneficial impact on particulate emissions through the reduction in fuel consumption. The public
should be willing to accept this control strategy because the installation of a newer, more effi-
cient burner results in a fuel savings and because the additional cost impact usually comes at a
time when people are expecting to spend money and, hence, usually more willing to do so than nor-
mally. A distinction is being made here between a regulation that requires all units to convert
to a lower polluting, more energy efficiency system by a certain date and the proposed one, which
only affects new and replacement units. In the first case the people would not have factored
this expense into their budget. Moreover, the cost impact to them would be the total cost of the
burner, not just the differential cost between currently marketed units and the .improved ones.
The next likely step for a control program of increasing stringency is to impose moderate
emission limits on existing units. The possible limits are a Smoke Spot No. 2, a COg concen-
tration of at least 8 percent in the flue gas, and no visible emissions. Several investigators ,
have demonstrated that existing units can meet these limits (see Section 4.2.1.1 and References
4-2, 4-17, 4-18). However, since the residential heating sector represents such a large number
of widely distributed sources, it would be very difficult to enforce such a regulation. Therefore,
as stated control measure No. 2 almost implies voluntary compliance. Consequently, both the reduc-
tion per unit burner and the impact on basin-wide emissions from this source category would
probably be low. The cost impact has been estimated to be only $30 per year, which is the cost
of an annual service and does not include a potential burner replacement. Although an annual in-
spection and servicing should improve the efficiency of the burner, we have assumed that the burners
which would really benefit from such a program would not, in fact, be serviced. Therefore, zero
energy impact is shown for this proposed regulation.
In order to overcome the shortcomings of this control measure, the third measure then com-
bines the emission limits just proposed with a mandatory annual inspection and maintenance service
by a licensed serviceman. The requirement for a mandatory annual service is intended to force
each homeowner to bring his unit into compliance with the air pollution regulations at least once
a year. By further requiring that this service be accomplished by a licensed serviceman, who has
had to demonstrate competence as a burner serviceman (including the ability to work with CCL, smoke,
and similar test instrumentation) in order to obtain his license, the chances are greatly improved
of actually attaining throughout the region the reductions that test programs have shown to be pos-
sible by periodic burner maintenance. Implementation of this control measure mainly ensures that
burners are not allowed to deteriorate significantly. Although their performance usually does not
7-9
-------
change markedly during one heating season, they could rapidly become inefficient and high emitters
if allowed to operate for several years without competent service.
There are several ways of enforcing such a regulation, and any of these could be used singu-
larly, or in combination, with the others. The traditional approach would be for the air pollution
control district to spot check a sample of the homes in the area. If they find burners which do
not meet regulations, they could then fine the homeowner, or if he could prove his unit was mis-
serviced within the last year by a licensed serviceman, they could fine or revoke the license of
the serviceman. The effectiveness of this program would be enhanced by widely publicizing the
enforcement activities and clearly identifying the inspector's cars. Another approach would be to
lay the burden of compliance on the fuel oil suppliers (because they are fewer in number and sup-
posedly more competent burner technicians than the homeowners) and require that they demonstrate
annuVlly, to the satisfaction of the air pollution control district that the burners of all the
homes which they supply with fuel have been inspected and serviced. A third approach would be to
induce the homeowner to retain a licensed serviceman for the annual inspection and maintenance of
his unit by giving him a reduction in his property taxes if he can demonstrate that his unit has
been serviced and complies with the district's emission limits. Such a demonstration would be in
the form of a certificate provided by the serviceman on which he notes the service he conducted,
his charge, and the performance of the burner after it had been serviced (i.e., emissions and ther-
mal efficiency).
An annual inspection and maintenance program, in conjunction with the emission levels speci-
fied in the previous control measure, should reduce emissions from the average home furnace or
boiler by the amount reported in the literature cited earlier. Since virtually all units in the
region would be affected, the impact on the area-wide emissions from residential units should be
approximately equal to the average reduction in emissions from each unit. The cost associated with
this program is estimated to be about $30 per year for the annual service, plus up to $300 for the
replacement of malfunctioning burners. A portion of these costs will be offset by fuel savings
(nearly 2 percent, on the average, or about $8/yr). Investigation of a sample of residential units
has indicated that approximately 10 percent of the units are in need of burner replacement, and
that the replacement of these malfunctioning units can result in up to a 17 percent reduction in
participate emissions from residential burners.
APCD enforcement personnel will still have to conduct spot checks for this control measure
and this will involve some costs. However, the total financial impact on the district could be
reduced if it recovers the cost of training, licensing, and enforcing the servicemen from their
7-10
-------
license fees. Moreover, since both private agencies and public schools offer training programs for
boiler and burner servicemen, the control district could identify those courses'which are acceptable
to it and require a certificate of completion from these courses as.a .criterion for obtaining a
license. Such an arrangement would eliminate the need to conduct the courses themselves. It may
also be advisable to require the servicemen to pass a test once every 1 to 2 years, and to certify
to the district that they understand the regulations and the obligations imposed upon them by the
regulation.
Most burners should gain a few percent in efficiency after the service, but as noted earlier,
the reduced cost of the fuel does not offset the cost of the service itself (at least not at current
fuel prices). For most people, however, the cost impact will not be terribly great and, therefore,
it is expected that the public will accept this proposed control strategy, although possibly not
with great enthusiasm.
If the ambient air in a given region exceeds the NAAQS for particulates by a large amount,
and if residential units contribute significantly to the emissions in that region, then the con-
trol district could consider the stringent limits for existing units proposed in control mea-
sure No. 4. This strategy goes one step beyond the previous one by requiring that units emit no
more than a Smoke Spot No. 1. It also includes the annual inspection and maintenance by licensed
servicemen that was discussed in the previous control measure. Obviously, such a control measure
should have a greater impact on the average particulate emission reductions per unit, on the total
emissions from all residential units in the region, and on the costs to the air pollution control
district. It should also have a greater impact on the users as a group, because more would have
to replace their burners with new ones (the cost per burner replaced would be the same as in the
previous control measure). Since more burners would be changed under this control measure and
since they would be replaced by new, more efficient units, the average increase in thermal effi-
ciency would be higher than in the previous measures. However, the cost of a new burner is a sig-
nificant item for the typical homeowner, and the payback period from the reduced fuel consumption
is probably too long to interest him; therefore, the public reaction may be poor, unless the im-
plementation of this recommended control measure is accompanied by a massive publicity campaign
which stresses its energy conservation aspects.
In some regions of the country, a slight additional reduction in emissions can be obtained
by restricting all residential units to the use of No. 2 fuel oil. This recommendation is based
on experimental results which show that burners fired on No. 2 oil emit less particulates than
those fired on heavier fuels (see Figure 3-4). However, the impact of such a control strategy
7-11
-------
on an area-wide basis would probably be low because most residential units already use No. 2 oil.
For those that do not, the owners would probably incur a cost of up to $300 for a new burner de-
signed to operate on No. 2 oil, and they would have to use a fuel which is 10 to 15 percent more
expensive than the heavier fuel they are now using. If this proposed control measure affects
enough people that their voices are heard, public acceptance may be poor, despite the improvement
in air quality.
Combustion improving additives at optimum concentrations have been shown to reduce emissions
by 30 to 50 percent (see Section 3.1.3). Therefore, a requirement that they be added to home heat-
ing oil could be an effective particulate reduction measure. In order to insure a uniform and cor-
rect application, control measure No. 6 suggests that the chemicals be added by the fuel supplier
at his distribution center. Since this control measure would reduce emissions from each unit sig-
nificantly, and since it would impact all the units in the area, the total impact on emissions from
all residential units in the area would be high. Furthermore, these reductions would be obtained
without much additional effort by the enforcement personnel in the control district. However, the
generalized use of combustion additives for residential burners is not recommended at this time
because the metal compounds emitted from units which use metal based additives may be toxic or may
react with substances which are already in the ambient air, such as POMS, to create a potential
health hazard. Although some organic-based additives may not result in toxic emissions, they gen-
erally have to be added in relatively large doses (1 percent by weight of the fuel), and, hence,
would be expensive. Moreover, some additives may interact with the fuel to increase deposits in
oil lines, filters, or other burner parts. Therefore, this control measure would not likely be im-
plemented until these issues have been resolved.
7.2.2 Commercial Boilers and Furnaces -0.4 to 12.5 MBtu/hr (0.12 to 3.66 MM).
The commercial category of boilers and furnaces is very similar to the residential group in
that it is comprised of systems which provide space heating and hot water for washing. Boilers and
furnaces in this size category are found mainly in apartment houses, office buildings, shopping
centers, and similar commercial or institutional facilities. They differ from residential units,
however, because they are larger, they generally receive more maintenance, and they frequently
burn residual oil. The first two factors tend to make them lower emitters than residential units,
but that can be offset by the higher emissions due to the use of residual oil.
Particulate emissions from this category can account for as much as 9.5 percent of the total
from all sources in an AQCR. Moreover, since commercial systems are larger than residential units,
each individual source contributes more to the total particulate emissions in the region than does
an individual residential heater- And, finally, since there are fewer commercial units than
7-12
-------
residential ones, the enforcement program can be more comprehensive. Therefore, the impact of a
regulatory program for commercial boilers and furnaces should be more effective than one for resi-
dential units.
For the purposes of these potential control measures, the word "commercial" will be used as
an abbreviation to describe those boilers whose rated heat input is between 0,4 and 12.5 MBtu/hr
(0.12 - 3.66 MW), even-though some boilers in this size range are used by industry to generate
process hot water or steam. The control measures for this category are listed and ranked on Table
7-3.
The first measure that the control district might implement to reduce particulate emissions
from commercial systems is a set of emission limits for both new and old systems. These emissions
limits are noted under measure No. 1 in Table 7-3. They are stated in terms of Smoke Spot Number,
with separate limits for new and old units, and within each of these categories, by fuel. Values-
for existing units correspond to typical levels reported by various investigators for well main-
tained systems (see Section 4.1.1 and References 4-2, 4-3, and 4-17). Those for new units are
based on the lowest levels reported by any of these investigators and on the differences between
old and new residential units. These limits should be accompanied by a requirement that they be
satisfied at a minimum thermal efficiency with a (XL setting that allows for continued smoke-free
operation until the next annual service i.e., 1 - 2 percent lower than the value at which smoke-
free operation can be achived immediately after servicing. This margin ought to ensure a sufficient
supply of oxygen even after the eventual accumulation of dirt on the air supply system. The minimum
CQy concentration of 12.5 percent specified in Table 7-3 is intended mainly for burner certification
tests. As mentioned earlier during the discussion of residential units, an efficiency requirement
prevents the owner or operator from engaging in the wasteful practice of using too much excess air
to comply with the smoke limit. A "no-visible-emissions limit" is also appropriate to assist the
enforcement activities.
Although it is not reasonable to expect each owner or operator of a commercial boiler or
furnace to demonstrate compliance of his unit with a mass emissions limitation (as measured, for
example, by EPA Method 5), an initial emission rate can be inferred for each new unit if the ABMA
and the Hydronics Institute are willing to cooperate by including a measurement of mass emissions
as part of their rating procedure. EPA is presently evaluating the practicality of such agreements.
These organizations provide a rating service to boiler manufacturers. The particulate limit of 0.1
Ibs/MBtu (43 ng/J) for new units larger than 5 MBtu/hr (1.47 MW) is equal to the most stringent
limitation imposed by several states on this size category.
7-13
-------
US
CO
O
4J
CM
r-*
O
cc:
fl-
ea
o
CO
O
u.
LU
a:
I
LU
U
LU
o
cu
CO
I
r-.
2
'
"*.
u|
3 o.
<
x*^
ft-
s
2t
a
u
1
M
S fc
O I.U
S "5e
1 S^"
* !?"».
is.
«j
i
i
- >»
?
°g
i
£
"i
I
e
O 1 Cl X I *J
S a g « C £,~
8 « x"* o S *° o
cflfHSf
if"O*
S^I «i|e*
g 0 *" C *J t-
VI U O--- o a § £
1
U
0
g
-4
U
i
0)
£
*
o
£s«,
2 -'S
Isi
-S3
M C VI
*d X
£_ -+J is
S O O»
u i
M dO S
t5* o( .!="=§
sg q»^~ §!§
a! "' "^g«
**? 1 S^«"
WICJ i£J T"o m "
w *7| «SJ m CM *J :
S (M 25! Tl - « i
j;!^ ?i ||| !
.?|S i| «~ glcs
iirf=" nit
-
»
!.*-'*
i^ +j
"'5
b|s
3 "S
111
*J O» >
°si.
Ill
"g
o
o
Cvj
O
s
E
a
z:
^--^
w> cn"cx"ci.
01 C o cn
^*>oo
§-sgg
1O 3 in CM
Ills
"°~"~
f 8»
Ijj
c v)
>d S oj
|Bj|
o S£ I
O "o ei .Q
*
i >
£ S^
u w o
ex c
^ W 01
rero w'=
a; " S
o" -S 01
slsis
f- Ctf *-» t- -^
S ^ 1 1 ^
3r~ ai w CT.
L. t/> «-> -r-
~
in u
^ C
T- a.
^ £*
l|
jQ JC
S*
il
ui cn
IE 1.
O
S
o
s
X
o
_J
^-.0
"Q.'Q.' Si
cn cn o c
C3O ">
ggs^;
in CM 01 3
|
o
s
o
§
m
cn
S
«d *->
og
) 41
Ml Q.
0 0
3 Id
it
3 t-
J3 _g
ex c
sis
t- Q.+J
g-i-
o
li!
ife.5
«'
*^ u
= u Sg
V '*" O
t. >,»- j a
u SS E»-
gUJ U =0
C? C T- 0) T-
C U X C.-
0.*^ T3 S
01 +J ' 01 id
O « ^ OJ i
d. i- c *-
*J ft) «
u 01 c 01 >
i" | £-| o
|
CO
o
g
*
C3 J_ ^_
o a. ^."o.
«n c cn cn
^ 5 o 2 S
l^oo
01
s
01
o
3u
CD
CM
,1 :
.1
§1
1"
g ^
'Id
HI
fcs'E
a. a. ai
«f 1
||1
,
t
§
0
"~ a)
(U
o||
^H S en
III
O
CO
s
in
g
'
"S"^-
Cx cn 3
°^0
gS«
m 10 *
gg
in
f
in
:
o
rd
(1)
a
fe
CM
s
>*
c:
O
£>
c
3
1
C
a
O (-
z o
*
c
I
a.
^>
a> S
E"
a o
|l
o"5
(3"S
1
CO
s
in
.c
3:
^^.^^^
"a."o. 3
cn cn1*-
O O i*-
goS
cn
o
o
o
.1
o
i £
M C
O U
**~ c»
5 x
a id
?*:
si
o. o
o
E O in
^^d>
55-s
^ --.
So-g i
I5.ti 1
° TJ 15 E « W
WO 1- *- *J r-
f- ^C 3 to *X
J id U- O C
*J "O £> O
2 E 5 -S 01 S
p tx. tti i 'w oi
& A u O C
»
§*i-s
Sss
(d jc
3 -r- Ol
cn S M
01 -r-
i- C C
o cn
O -r- O
(_> 4-J U
"a)
£ 3
o-g :
to -o
nj
> 5 ' I
1| 1
O itj
1/1 c e !
P
Ss ,,
on
i.
o
cn
O)
«
u
S
3
O
trt
!e
**
o
01
3
0
g-
^
ai
a
£
C
O
o
I
a
1
S"
g
I
- o
g 1
s ^ ^
C Dl C
K S £
O « t-
1 £ 1
1 II
W) O f^
.s " g
E r C
S r- C
« EC
e £ ° ^
<*- U fO
1 1 - 1
aj i o 3
c a*
O jQ 01 it- O
JJ° 0 4->
J c «
O d "cx 4-> cn
ai i/i o) § e 3
) 10 a: oo ra "^
7-14
-------
As with the equivalent control measures for residential units, measure No. 1 can be imple-
mented by an APCD regulation. Its enforcement can rely in part on building codes, cooperation with
the standard seating laboratories, and inducements for voluntary compliance through the use of tax
incentives. The suggestion that a tax reduction inducement be incorporated into, this control mea-
sure, the estimated effectiveness of only 10 percent emissions reduction per unit shown on Table
7-3, and the pessimistic tone of the comments for this control measure all stem from the belief
that there may be too many commercial-sized installations in the highly urbanized northeastern
cities for the APCD enforcement staff to monitor on their own now. The additional procedures to
be described under control measure No. 2 below are intended to overcome this 'problem. Moreover,
if controls are placed on new equipment, this enforcement problem will disappear eventually.
Most local or state air pollution control authorities who have an active program for the
reduction of emissions in their region require an installation and operating permit for new sources
above a certain size. Such a requirement assists the district staff to locate all new sources and
enables them to provide a preconstruction review of the source. These early reviews lessen the
chance that the source will exceed the local emission limits after it has been built and set into
operation. By incorporating a requirement that the source owner renew his operating permit periodi-
cally (e.g., every 3 years as in New York City), and by further requiring that the renewal applica-
tion include the results from a stack test to demonstrate compliance with the emission limits, the
APCD can effectively enforce these limits at minimal additional cost and burden to itself. The
fee for the renewal application should include charges not only to process the application, but also
to cover the control district's cost of sending inspectors out occasionally to spot check emission
tests.
The cost impact of control measure No. 1 in Table 7-3 is based on a typical annual service
charge of approximately $150 for commercial-size units (including the cost of a smoke spot test)
plus $15 to cover the cost of a permit application. This latter figure is based on New York City's
projected triennial fee of $50.00, when distributed equally over a 3-year period for accounting
purposes. Since some boilers and furnaces will need to replace their malfunctioning'burners with
new ones to comply with the emission limits, the cost of a new burner has also been included in
the table.
If an air pollution control district feels that it needs to reduce particulate emissions from
commercial units even further and 'faster, and if that region contains many systems whose burners are
equipped with rotary cups, then the next step should be to prohibit the use of rotary cup burners.
7-15
-------
This control measure is based on the findings by the Maryland Bureau of Air Quality Control that a
large percentage of the units that fail to comply with their smoke limits are rotary cup burners.
Moreover, many of those which do comply do so only by using too much excess air and, therefore,
waste fuel. In order not to stifle the development and utilization of a rotary cup burner that
can be maintained smoke-free under typical operating conditions, it is recommended that the regula-
tions include a proviso which allows someone to use such an improved rotary cup burner if he can
demonstrate compliance under typical operating and maintenance conditions. Such a demonstration
would, of course, be at the expense of the proponent of the new system.
The impact on area-wide emissions of such a measure is expected to be relatively low because
not too many burners are currently equipped with rotary cups (for example, in the Northeast, where
they are most prevalent, they account for only 5 to 10 percent of the burners on existing furnaces
and boilers). For this reason, and because the prohibition can be enforced by spot checking oil
supplier records, it is estimated that the regulation will have only a minimal impact on the en-
forcement program of the APCD. The 5 to 10 percent gain of thermal efficiency shown on Table 7-3
for this control strategy is based on the observation of the Maryland Bureau of Air Quality Con-
trol that many owners of rotary cup burners now use large amounts of excess air to comply with the
Smoke Spot Number limitations. Presumably they would not need to follow this wasteful practice
with a new burner.
Particulate emissions are generally higher from units that burn a heavy grade of fuel, such
as No. 6 oil, than they are from those which burn a lighter grade of fuel, such as No. 2 oil. The
problem is generally considered to be worse in small units than in large units, because small units
typically do not receive the service and maintenance attention that the larger units do, and be-
cause the temperature in the combustion volume of small units is not always high enough to com-
pletely oxidize all the fuel when it is a heavy oil. Therefore, several regions, such as Maryland
and Boston, prohibit the use of residual oil in smaller units. Control Measure No. 4 is based on
this practice. Before adopting this measure, the control agency should determine that local fuel
suppliers can meet the increased demand for distillate fuel.
Since most burners and fuel handling systems are designed to operate with a fuel of a given
specific gravity and viscosity, several components would have to be replaced or readjusted if the
unit were changed from residual to distillate firing.* The cost of such a change is indicated in
The changes include disconnection of the heating system for the fuel tank and air lines, flushing
of the fuel systems and inspecting it for leaks, replacing the fuel filter and pump (or changing
its speed, if that is possible), replacing the nozzle, and retuning the burner.
7-16
-------
the cost impact column on Table 7-3. The lower estimate for smaller units is based on the assump-
tion that this conversion would cost about three times as much as a typical annual service call
(twice as much work plus parts). The corresponding figure for larger units comes from Reference
4-55. In both cases the upper cost impact is based on a complete burner replacement. It should
be noted that there is a substantial cost saving if a new installation is initially designed to
burn distillate rather than residual oil. The 15 percent fuel cost increase also noted on this
table is based on the average price difference between distillate fuels and the higher sulfur re-
siduals. The energy impact shown on the table for this control measure represents the energy re-
quirement at the refinery to refine the crude into a distillate. Since the cost impact is signifi-
cant, the public reaction may not b'e favorable.
Although boilers and furnaces which are located in commercial installations are frequently
monitored and cleaned by a boiler operator, this person is usually responsible for many other
aspects of the operation and maintenance of the building. The boiler is only a small portion of
his job, and he may not even be trained for this task. Therefore, the additional requirement of
a mandatory annual inspection and maintenance by a licensed serviceman, as included in control
measure No. 5 should be an effective supplement to the previous control measures. Gains would
come both from the use of specifically trained servicemen and from the guarantee that all units
were serviced annually. This guarantee could be useful if permit renewal is required only once
every 3 years, as included in control measure No. 2, because some boiler operators may call upon
the services of a qualified maintenance person only before they are required to demonstrate com-
pliance with the emission limits. The implementation and effectiveness of this control measure
are similar to that described in the residential sector for the corresponding measure.
Control measure No. 6, which provides that no residual fuel be burned in any commercial
boiler and furnace, is supported by the same reasoning as control measure No. 5, but simply goes
one step further. Therefore, the impact on emissions to the ambient air will be greater, but so
will the cost since the conversion charges will have to be born by more users.
In areas where the ambient air quality exceeds or is expected to exceed the NAAQS by a sig-
nificant amount, and where commercial boilers and furnaces are major contributors to the total
particulate emissions in that region, it may be necessary to go to control strategy No. 7 and apply
stringent limitations on the smoke emissions from all units, both new and old. The indicated
levels (Smoke Spot Number 1, minimum overall efficiency of 80 percent, or C02 flue gas concentra-
tions of at least 12.5 percent for burner certification) presume that the user will have to install
7-17
-------
a new> well designed burner such as the optimized one that has been developed for residential and
small commercial units. It also presumes that he will have to use distillate oil, at least until
a low particulate and N02 emitting burner is developed for residual fired units. The estimated
effectiveness of this control measure shown on Table 7-3 is based on several assumptions: (1) cur-
rent units emit between 0.053 and 0.067 Ib/MBtu particulate (22.8 - 28.9 ng/J) in accordance with
the EPA emission factor for a unit using No. 6 oil with 0.5 to 0.7 percent sulfur content; (2) dis-
tillate fired units with Smoke Spot Number 1 would emit about 0.007 Ib/MBtu particulate (3.0 ng/J)
because current units are characterized by Smoke Spot Number 2 and are estimated to emit about
0.014 Ib/MBtu (6.0 ng/J) (as per the EPA emission factor). That is, reductions from 0.014 to 0.007
Ib/MBtu (6 to 3 ng/J) are envisioned for units which currently burn distillate oil and from about
0.067 to 0.007 Ib/MBtu (28.8 to 3.0 ng/J) for those which now consume residual oil. The cost im-
pact shown on Table 7-3 for this control measure is based on the need to replace the burner and on
the price differential between No. 2 and high sulfur residual oils. The energy impact again repre-
sents the increased energy consumption at the refinery to refine the fuel. Although the use of
building codes for new installations and tax reductions as inducements for existing installations
are indicated as part of this control measure, the need to insure that all commercial boilers and
furnaces burn only No. 2 oil and are equipped with burners which are designed specifically for that
fuel would probably impose a significant burden on the enforcement staff of the control district.
Control measure No. 8 concerns the use of additives. Iron-based additives can reduce partic-
ulate emissions from commercial units by up to 50 percent, and some organic-based additives are
equally effective. The impacts of such a control measure, both beneficial and adverse, are the
same as those that were described under the residential sector. Implementation of this measure
should be delayed until more information is available concerning toxicity of resulting emissions
and possible adverse effects on fuel stability.
No control measure has been described that relies expressly on the use of emu!sification.
The available data come from only one set of tests on one boiler and are inconclusive. Of the two
types of emulsifiers tested, one caused smoke emissions to increase when it was installed on the
boiler and operated at zero or low water addition rates. Typical uncontrolled smoke levels were
only reached when significant quantities of water were emulsified with the fuel. However, it was
effective in reducing mass emissions. The other emulsifier did not affect smoke emissions signifi-
cantly, but it increased mass emissions at zero or low water addition rates and never reduced them
much below typical uncontrolled rates from standard systems. However, nothing in the control
7-18
-------
measures listed on Table 7-3 prevents an owner of a boiler or furnace'from using emu!sification to
achieve specific emission limits if he feels that this technique is the most practical in his case.
7.2.3 Industrial Boilers - 12.5 to 250 MBtu/hr (3.66 to 73.2 MW)
Industrial boilers are generally used to provide process steam for washing, cooking, heating
a substance to be dried, or heating a reaction vessel to provide the correct environment for a
chemical process as well as space heating in conjunction with one of the above. They can account
for as much as.9.5 percent of the particulate emissions from all sources in an AOCR. Some of the
larger watertube boilers in this size category are also used to generate steam that drives a steam
turbine which provides electricity to a plant (or a small utility). This size category of boilers
is somewhat of an anomaly in that it straddles the regime between commercial boilers on the small
size and utility boilers on the large size. The smaller industrial boilers generally have the
same emission characteristics as the larger commercial units and frequently recieve the same kind
of attention as do their commercial cousins. The large industrial boilers, on the other hand, are
mostly of the watertube design, as in the utility sector,, and are operated with the same care and
professionalism received by boilers in central power stations. A concrete example of the transi-
tion that takes place in this size category is given by the application trends among soot blowers;
units less than about 30 MBtu/hr (8.8 MW) usually are firetube boilers and are not equipped with
soot blowers, whereas those which are larger than this size tend to watertube designs and are
equipped with soot blowers. Maintenance practices vary widely in this size group and are frequently
more a function of the size of the total facility rather than of an individual boiler. For example,
some manufacturing facilities have several smaller boilers which each provide hot water or steam
for one process. In such a case, these smaller boilers may receive the same kind of attention
that a much larger boiler would normally receive.
The three major approaches to controlling particulates from boilers in this size category
are as follows:
« Specific emission limitations
e Required installation of particulate collectors with specified collection efficiency
Fuel restrictions
The first is used to insure that the burner is correctly designed to handle the fuel it receives
given the combustion volume in which it is placed. From a control point of view this can be
achieved by specifying emission limits for all boilers, and particularly all new ones. These
7-19
-------
limits could be set at levels which have been demonstrated by well designed units. Additional
servicing and monitoring requirements can be placed on the owner of an industrial boiler to insure
that his unit continues to emit no higher than its design levels. The second tactic can be invoked
either directly by requiring the use of a particulate control device (usually an electrostatic
precipitator or, in some cases, a multicyclone), or indirectly by specifying emission limits which
are so low that they can only be met by the use of a control device, or by both approaches. Since
the size, and hence the cost, of a precipitator are functions of the emissions reduction desired,
successive control measures from among a prioritized list can mandate lower and lower emission
limits at consequently increasing cost to the user (e.g., measure No. n could require 70 percent
reduction, No. n+1, 90 percent, and No. n+2, 95 percent). The third approach, which is to place
restrictions on the fuel used, attempts to insure that emissions will be reduced by removing the
source of much of the particulate matter. One such restriction would be to limit the concentration
of sulfur and vanadium in the fuel. The value of such a measure is that it eliminates the need for
additives to reduce deposit accumulation and corrosion on the superheater tubes. These deposits
tend to be more of a problem with utility boilers than with industrial units because the latter
usually operate at steam temperatures which are low enough that corrosion is not significant. Ac-
cording to measurements taken on utility boilers, those which fired a fuel that required the use of
additives emitted approximately twice as much particulate matter as those which burned a fuel that
did not require additives.* The other fuel restriction that could be imposed would be a prohibition
on the use of residual oil in any industrial boiler. As with the commercial units, distil late-fired
industrial boilers emit less particulates than do residual-fired units.
Table 7-4 contains possible control measures for industrial (or small utility) boilers
ranked in a suggested order of implementation. As before, the intent of this table is that con-
trol strategy No. 2 be applied only if control strategy No. 1 does not reduce particulate emissions
from industrial boilers sufficiently. Similarly control strategy No. 3 would be implemented only
if Nos. 1 and 2 together do not reduce particulate emissions far enough.
Control measure No. 1 contains two parts. The first is to require a construction and opera-
ting permit for all new or replacement boilers in this size category. Although most state and local
air pollution control authorities already use a permit system, this measure is included for those
which do not already have one. The significant part of this control measure, however, is the emis-
sion limits for new and replacement boilers or burners. The mass emission level indicated for
These test results come from boilers which are equipped with particulate collection devices whose
efficiency is approximately 50 percent.
7-20
-------
Id
oo
CO
O
IT)
CM
o
CQ
a;
o
O
D;
o
(-3
Qi
a;
D_
co
oo
o
CL.
LU
I
ca
8
U>
at
c
s
u.
' =
o
tj
*I
g!
S|-
O
a.
""
i_
=
0 J- U
o «-
"O ttl 11
QJ Q- o:
o:
O -M
U C
U O **
C
*o -«->
r:
JD
4)
=1
1
e
c
o
o
r
4->
TO o i a» TO o
.°^> ."*" 3 *° § 9 ^ g
-r~ ttTlT*,^ ^ £ TO ^ "c ^ tj
(!) 4J 4-J 'O>T~--^-I->it- £
c i- +J TO c »j c u- *c: 31 1/1 OJ »- V. '^- li
^ § !- *V1 3 C-r-C
o t- a» o an at at ,3 <" SJ
"5 ° "« -^ SI >>f- '.--£ *« 1-
ttt'fc c omSc. ° *" Tj ,^_ o ?
a
3
oj
o-
u
o
o
o
o
i S ,~Z
_5 aj > "TO
|i3
f
i||
|S-S
O) -0 ttl
O 0 TO C
0. O 4-> O
-£ (J W *J
"^g"
;£-;g-.£ S
t, ^~ t; in -C -c:
!I!P. 3|
"e . "TO e c g >
O fl] »> **-
3 l .»- O O
i S b fe c «
Ot ">
g, LJ_ J - -r- +J *-*
0 Cl *-> I- « ° "-
u 1 1'' w £ - 3 1 *- -
a> -^- 1 *-> ^-t)«i ^
^OTO'IOCC" tow ai -
lll«ll 5'£ II
-
"o. Q.
8° "
Based ori Maryland regulation as
Presumes units now emit @ SSt) 4
(barely visible). Cost for pern
plication.
a
V
O
4-1
TO
1
S
^
0
0 l£>
in ro
S
at
£
"i
1
4->
3
a
0
4J >>
"xla
V 3
**- ^
U S CD
PI
Is! ii
c. ai ;-, ^"ro
at *- f_ - i;
^ jQ O til 'in
3 U -r- at
«-"
«
,- i
8 « ,
E- .^e g
n. c -o«*- v-
States are encouraged to rake re
aoreerents for licensing service
Whenever e source can shew that
ent burner servicing and clesnin
quency caii maintain the same etrri
these provisions coold be waived
o
s
o>
o
1
i- i-
>> >i
& S
0 O
O CM
ai
CD
o
o
is§ i
« c S t- "
§»'-J3 i. c en
OJ ra O O C
cj at M c
a a. ~a i »-
C3 O TO >i TO
0- O i- 4-> I-
0 =
1 | ' "Sl-'o'0
*)
U C "i- OJ (-) M O
r- o J= -O o ra c:
*~ ^"Ci/i mi^ """aJ^'u'aj'
t- S 7? c "c "c -u <- -i- **- ,*^
^ o- A 5 ° & !^ i f1'' )O)l/i O tUCJ lrtH4->
!^ i £ -^ -^ wi R £ .*
U C ^ U > -r- 0
S.J? ^ S^ a; -u c c -*^
3
^f t C 4-> 4J '5 U *-*
o c 01 M e: c--r- u
O TO <4- TO "O
at o ai QJ i- . i.
3+j ataj'rot-FJn
O I- 3 l/> C 3
OJ3 4->l/lt,O"O4-'
c' I-5^°|S.
j c: at o o L. o
5 at cn-o X o TO
cu caJ54Ju_i2
3 S *J 3 £ 'c O M-
s? St". 8t;§
cc a. O to o
^C jz2'^^^
0 ° 3* 0) 3 Ot 3
jJ~T) C *-' O <-t 5 - *-
o
tD /
0)
o
a.
g
t-
S o
o
gl
in Q.
1
«
S
S
a.
' ^ 3 ° ^ 1
J£ 'i S o -^ ^S ^ ^
^ j_ c £ ai e u wi-
.. f f fsi"^ II 11
at- W t.' e w « "c <-S *J~~'
Dwi ^^^k^-o^lJsS ^^
r- t, t. O 3
3 O £ 4J £ ft)
S ^S.£§
co
7-21
-------
i? C
u< u at o
t * a~'5
at ui a> a ai
llSss
g
S
O
i
*
£
**-
o
CM
0
S
3
S
c
*9
1
t_
o
o
s
l-g
S5K
!-£ A
O *-»7j 3
O 0» A
*O A C i-
tf«*-«^
y,
Cl
0 U
1 I/I >
1^. v- O
O U U T3
«£M i
ggJl'TJ
"511
Iss^
C -^ ai *-
3; u ns
o x +*
C U> (U £
10 t- **
z°>j-s
c ut O w
Sai r^tn i- ^
>i<+- ai c
OQ o o .a ai
o
g
CD
CM
a.
1
"o
2:
51
V O
en
4J
I
IO
at
£
O
=
s
A
§
tn en
^|
S *
O
is
^.
S^-u"
CL-r- »«- 44
u *" "*" Si
fl ** wX
4J J3 *J
SI t
lisS
i/i v- o d
Same coiment a
efficiencies o
approxlrrately
reductions are
mav be oosslbl
o
g
V
s
I
"§
z
wS
in<»-
-^«*-
cn
z
o
s
g
I
fO
1
I
o
A
I
rtJ
11
c-in
+JO
^o
+J
g|
JE
in
&lg-o^,
P^£
4-» VI
OC ro 10 Ol
to exj= Ei
-J.g*'0,
' J^ ttJr^-S
"O 4> T3 «
d) - Jrf V l/»
ui X <- C) dl
a i r- u c
c cr o o
C 0=3 3
O -i- v» W
Oi 3
"SS^fe.2
S .E^S
^a X ai «- *~
+j 5 ^
S
o
£
CO
o
tn
I
i
^
o
s
1
p-
0
I
IQ
g
g
|g
?2
11
U- iO
o
a|
_*^ -^ ai
5 c -
|-g|
«5
s^
4> -M
1^ -u
4-1 C
1 3
r- C
O C3 *J
Subject to av
Cost depends
as replacenen
I
CO
s
1
1
o
^n
f
a:
s
1
1
o
(J
r-
1
*i5
i
1
OJ,
dl
*j t: S
M O >-
i§i
CO
I
5
U
s
5
o
a*
u
rt>
4->
1
1
£
g
w
"c
£.
10
0
S1
*
o
1
g-
7. °
t -O
0 £
t i IO C
in u ai
3 ^ P>
U) 4-1 U)
*- C u)
1 II
i " i &
"3 S S
It- U IU 0
'° s a = s
C O E " C
* i. '§ 8 £ "S
o Jo di »- 01 o
£ lj * ° ° "to
0 3 *-M - T3
x ^ *° 'S g c
1 1 « *s 1 S
S *" * dl ^ E
** " "^ "o. +-> o>
1 « i 1 I 1
7-22
-------
residual-fired units is based on the most stringent State regulations which do not envision add-on
controls. Depending on design and fuel characteristics, some boilers can reach even lower emission
levels. Thus, technically, this emission level may be revised downward if necessary.
An additional restriction is included in this control strategy by requiring that the plume
not be visible. The 5 percent opacity limit is intended for those boilers which may be equipped
with opacity monitors. A "no-visible-emissions" limit is also included for use by roving enforce-
ment personnel. These two limits are essentially equivalent because plumes begin to be visible as
their opacity approaches 5 percent. The Smoke Spot Number restrictions are the same as those given
for new commercial boilers; their application to industrial boilers can be justified on the same'
grounds as the application of commercial mass emissions limits to industrial units. Although it is
not common practice to measure Smoke Spot Numbers on units as large as the ones in this size cate-
gory, enforcement personnel in Maryland use such tests to check for compliance with their regula-
tions (which include Smoke Spot Number limits for all boilers). They feel that their measurements
are meaningful as long as they insert the probe far enough into the stack to avoid wall effects.
This control measure recognizes the difficulty and expense of making mass emission measure-
ments, particularly according to EPA Method 5. Hence, mass emission measurements are required only
to obtain the original construction and operating permit for the boiler. A Smoke Spot Number read-
ing would also be taken at this time at a representative location. The permit would be renewed
each year after a total engineering evaluation by the control district, including review of smoke
spot measurement operating parameters, condition of the equipment, etc. The use of smoke spot is
based on the assumption that mass emissions at any future date will be no higher than they were
during the original test if the Smoke Spot Number at the same sampling location is also no higher
than it was during the original test (presuming the firing rate and air-to-fuel ratio are also the
same). This assumption can be justified by data from residential and commercial units which show
that, for any given unit, as the mass emissions increase so do the Smoke Spot Number readings, al-
though not necessarily linearly. It is necessary to specify the sampling location as well as firing
conditions because the flow is frequently nonuniform in the large stacks on these boilers. An addi-
tional flexibility is included in this control measure. If a manufacturer builds several boilers
that are identical, he can ask the ABMA or Hydronics Institute to certify that this family meets
the emission limits. New owners of one of these boilers can then use this certification in their
permit application in lieu of another emission test.
7-23
-------
The uppermost value of potential emission reductions reported for this control measure is
based on the difference between the highest emissions levels reported for existing units of this
size category and the indicated control levels.
The impact of this control measure on the area-wide emissions from all industrial boilers
depends on the economic growth in the area'and the future fuel usage patterns (particularly the
degree to which coal replaces oil or oil replaces natural gas). Since the emission levels speci-
fied in this control measure are being achieved by existing units, no cost impact has been assigned
to this control measure for the purchase or operation of new burners. Costs shown are for permit
applications, including testing. The impact on energy consumption will be favorable if a unit
that can run at less than 15 percent excess air is chosen as a result of this control measure
rather than one which would have required more excess air.
Control measure No. 2 applies to existing boilers. It also introduces the requirement for
an operating permit, which would be renewed annually, in the event that the control agency in ques-
tion does not already use such a system. Since there are no data to substantiate mass emission
limits for existing boilers on the basis of combustion modifications (such as was available for
residential and commercial units from tests on the effects of tuning or the addition of a retention
head), and since the cost to replace a burner with a new one that would meet the emissions limits
specified in control measure No. 1 could be high, the first step indicated for existing units is
to place a Smoke Spot Number limitation on them. As with measure No. 1, the limits shown here are
based on the levels achieved by commercial units. They also conform to Maryland's regulations.
The cost impact is based on the assumption that servicing and tuning industrial boilers takes about
2 to 4 times as long as it does for a commercial unit. The emission reduction of up to 30 percent
that might be achieved by the implementation of this control measure is based on the assumption
that existing industrial boilers are operating with a plume that is just below the visible limit.
This limit corresponds to a Smoke Spot Number of 4 to 6, depending on the fuel and the combustion
characteristics of the boiler. If the reduction from Smoke Spot Number 6 to Smoke Spot Number 4
corresponds to an equivalent reduction in particulate mass emissions, the upper limit of 30 percent
reduction is justified.
The next control measure would be implemented if the air pollution control authorities pre-
dicted that they could not attain and maintain the NAAQS in their air basin despite the use of the
first two control strategies, and if they thought that their inability to achieve these goals was
due, in part, to the fact that many industrial boilers were not complying with the above regula-
tions in between permit renewal periods. Thus, part A of control measure No. 3 specifically
7-24
-------
requires the owner of a boiler to perform frequent maintenance rather than leaving it up to his ini-
tiative to do so in order to comply with the emission limit. The 3-month interval between successive
servicing and cleaning of the components that handle the fuel oil and combustion air is based on
the experiences of the Boston Edison Company, as reported in the introduction to Section 4.3. This
interval should be applied to residual-fired boilers that are larger than 30 MBtu/hr (8.8 MW).
Smaller units, or larger ones that burn distillate oil, would only be required to service and clean
their systems every 6 months. Although it is known that distillate oil does not create the same dirt
problem that residual oil does, there are no quantitative data available to establish that 6 months
is the optimum frequency'between successive fuel system cleanings. Therefore,, this control measure
indicates the inclusion of an option which allows a boiler owner to demonstrate to the local air
pollution control authorities that emissions from his unit will not increase substantially if it is
cleaned less frequently. In order to insure that the boilers are cleaned by servicemen who are
trained to optimize the performance of the boilers from both an energy and an emissions point of
view, this control measure includes a requirement that both operators and maintenance/servicemen
be licensed. Of course, the license would only be issued to people who could demonstrate that
they had been appropriately trained and were competent in the field.
Part B of this control measure adds the requirement that boiler operators install flue gas
monitors on the stacks of their boiler. These monitors would measure just smoke in the smaller
systems and both smoke and oxygen in the larger ones. The reason for including oxygen monitoring
on the larger boilers is to insure optimum combustion in those units. A distinction is made be-
tween the smaller and the larger boilers because the accurate transmissometers that meet EPA speci-
fications are expensive.* The main purpose of these smoke meters is to notify the operator very
rapidly if there is a malfunction in his boiler which causes smoke to be emitted. Such a condi-
tion could occur if a burner became plugged.or if, for some reason, an air damper fell shut. Since
smaller boilers emit less mass per unit time (i.e., g/s), then do larger ones, it seems reasonable
to allow the smaller units to use a less accurate and hence cheaper smoke detector. The alarm
level on such a unit should be set for about 15 percent opacity, or slightly higher than the level
which would be set on the more accurate transmissometer that would be installed on a larger boiler.
Unfortunately, data are not readily available in the open literature that document the frequency
For .units <250 MBtu/hr (73.2 MW), transmissometers that are purported to meet all EPA specifica-
tions (Federal Register, Vol. 39, No. 177, September 11, 1974) except linearity of response above
80 percent opacity are available for $500 - $600. Those which respond linearly up to 100 percent
opacity cost about $1000 - $1500 (in early 1976).
7-25
-------
and duration of typical upset conditions; therefore, one cannot estimate the effectiveness of this
part of the control strategy.
Oil-fired boilers that are well maintained and tuned should not emit any significant quantity
of large particulate matter (diameter greater than 10 microns). However, a survey by the Maryland
Bureau of Air Quality Control has shown that some boilers apparently do not receive the attention
they should because mechanical dust collectors in their exhaust do collect significant quantities
of particulate matter (see Reference 4-14). Since these devices do not collect small particles
efficiently one must assume that the exhaust from some boilers does contain significant quantities
of large particulate. Therefore, part of the approach which should be taken to insure compliance
with the control measures 1 and 2 is to require the use of mechanical dust collectors on all resid-
ual fired industrial boilers (except where a more efficient collector is used). Units are available
which can collect up to 75 percent of the particulate matter greater than 10 microns (see Section
4.3.1.2). As noted on Table 7-4 in the comments for this control measure, collectors probably help
more to alleviate local nusiance problems from the fall-out of large particulates than to reduce
the quantity of total suspended particulates in the atmosphere. Compliance with the collection ef-
ficiency requirement can be demonstrated by proof that the collector is designed to capture 75 per-
cent of the 10 urn particulates. This provision obviates the need for expensive EPA method 5 mea-
surements.
The next two control measures listed on Table 7-4, Nos. 4 and 5, are stated in terms of mass
emissions limits in order to allow the boiler owner maximum flexibility in meeting these limits. The
limits in measure No. 4 are based on Maryland's regulations (0.02 gr/scf, or 0.04 Ib/MBtu at typical
excess air levels of 20-30 percent). It is recognized that this limit may implicitly require the use
of a multicyclone or electrostatic precipitator (ESP) with collection efficiency of 50-70 percent.
3epending on emission characteristics, ESP's are capable of efficiencies of greater than 90 percent
to ultimate emissions of approximately 0.01 Ib/MBtu. Thus, if a control agency determines that
greater reductions are necessary to meet air quality standards, it may be possible to impose more
stringent limits than those in Table 7-4. However, such parameters as cost, fuel characteristics,
technical feasibility, etc., should be considered prior to imposition of more stringent limits. For
example, a lack of adequate space for ESP around certain facilities in densely populated urban areas
could contrain the use of this strategy.
The next two control measures listed on Table 7-4 (Nos. 6 and 7) deal with restrictions on
fuel and fuel treatment. Both of these techniques are very expensive and may create economic dis-
locations if premium fuels become more scarce than residuals. It is unlikely that these strategies
7-26
-------
would have to be implemented because emission reductions of over 70 percent should have already been
achieved by the use of the first five control measures.
Additives such as MgO are required in utility boilers if they are to burn fuel which has a
high vanadium, sodium, and sulfur content because compounds of these -elements would otherwise cause
cold-end corrosion and acid smut accumulation on the heat transfer surfaces. Although industrial
boilers generally operate at lower temperatures and, therefore,, are not affected as much by these
compounds, control measure No. 6, which prohibits the use of these additives, has been included
to cover units which might use additives. Measurements on utility boilers have shown that those
units which use additives emit twice as .much particulate matter as those which do not. Boilers
which are plagued by corrosion would have to switch to a lower sulfur, lower vanadium oil. Even
greater particulate emission reductions could be achieved by switching all industrial boilers to
distillate oil, providing the requisite fuels are available.
The final control measure is the possible use of combustion improving additives. The poten-
tial advantages and disadvantages associated with the use of additives are discussed under both
residential and commercial systems.
7.2.4 Utility Boilers -Over 250 MBtu/hr (73.2 MW)
Oil-fired utility boilers account for up to 19 percent of the particulate emissions from all
sources in an AQCR. Their impact is greatest in the urbanized northeastern areas of the country,
where many power plants have been converted from coal to oil over the years to reduce sulfur and
particulate emissions. They also are major contributors of particulate matter in the southern half
of Florida, and moderately significant in a variety of other AQCRs throughout the country (see
Table 1-1). Their impact in the future may depend more on the rate of oil to coal conversions in
the eastern portion of the country and gas to oil conversions in the western portion than on the
projected growth of electric power generation.
The same kinds of particulate controls can be used on utility as on large industrial units.
Therefore, only a short discussion is required here in support of the control measure presented in
Table 7-5. The major differences between the industrial and utility section are as follows:
0 Utility boilers are generally operated and maintained by technicians who are specifically
trained to optimize boiler performance. However, this does not always mean that they are
trained to minimize emissions.
Utility boilers show less variation in operating condition, maintenance practices, and
quality of the staff than do industrial boilers.
7-27
-------
CO
r-
o
in
CM
A
CO
ui
o
» -A
o o. ai *J *J t_ u
38s|-5!.sJ
S>K g 1 =-S u o
"g
&
g
*> 1.
s|
O 3
M °
72
1
o
LO
I
«
at
3
s
x g
S ~ |»|
3 ' ;2H
S JS gS
* .. *~° ^ ^
H ei °.~~. " *"
*?l'a «l^"* -E 3
§c " tn *cx
^ "^ *
~i- 3^. ^ "
§3 £ £li
"g
3
Of
1
1
.5
1
CNJ
«J
1
5
S
o
n>
CT
£
a
S
L *-
V O^^
C N
I- O o
= 0<->
=*fe
S S"
li°
i^"1!
S||
fl Ol >
C I-
£S?
01 t. w
a. ZI
-° .J2;2
- c c*i
W 30
-O OlO
O i-CU
4->O
i .s
el5
- «J i. g
3S^S
sources based on ESP of 90-
1c1ency for nonraT operations
roxinately 99* during soot
. For existing sources,
n 755 efficiency norrially
roxirnately 97. 5X during soot
S *- Q. cno a. o>
c Of ia-^-O S"-^
1- *«TJ 6 S? T) O
O LO C « rO C t
u- f-» « jan ro ja
"S
s
V
1
J:*3
** O Of
14- +J E
0 3
+J tn
"ss
1
C7>
II
£
IR
o
|Q
£
O
«t
^is
^*s°
. v: o
o*~*
4-> It.
"sE 3 D
s o.s
155
^'S's
iP?
2^=g
m
iL ,1 1
fsii^i!
t. > w 1 *-» o *-»
O. 41 ( nj *J
JJ Dl Cl Q)
S^aS K C.S
s^Ssc si-si
t-
1
O
s
ro
1
H
**-
o
J2
O" '
O
in
o
(S
£
C3
S
o
Sg
O "-
o -5
£5
.!.!
c
0.
o
£
O
. '44
3.
i" 0
» X
*i= c
in >) CL
S fe *
i S1 -
LO 0 41
O in f-
M S in
I ^ J
1 S "S
O J- N
§ ^ C C
i cj: ° £ 5
M- U O* n)
£ 0 1" - C
O -^* - *o
u E o ai u
> 1- S o t- tn
£ ai u
o ^3 a> *- a» o
O E T3 O T3 -»
1- K "5 C 1- «
o too*-1
o 3 +"» ' "a
O Q. 1- "O a;
5" M *> - O
a) t/i d> ia c 3
10 10 0- to ZJ *0 '
«» J3 0 -D O) "*-
7-28
-------
Utility boilers are larger than industrial units and have a higher load factor than
many of the latter system. Therefore, they have inherently lower particulate emissions
and can better absorb the costs of control devices and increased maintenance efforts.
Fuel costs account for a larger fraction of total operating costs in utility boilers
than in most industrial units.
Due to their size and continuous operation, each utility boiler has greater impact on
the ambient air than a typical industrial boiler even though the unit emission rate
(Ib/MBtu) is generally lower for the utility.
Because of these differences, the control strategies in Table 7-5 place less emphasis on mainte-
nance and operating personnel requirements and more on the use of high efficiency ESPs. Moreover,
emission limits are specified in terms of a continous variation with size rather than in discrete
groups.
The highest priority control measure is considered to be a low efficiency collector. Units
with 50 percent collection efficiency would make a significant impact on the basin-wide emissions
from the source category at a minimal cost. The control measure actually provides the boiler opera-
tor with some flexibilty in meeting the requirement because it is stated in terms of emission limits
rather than equipment standards. Some boilers already emit at lower levels without controls (see
Figure 7-1), either because of design features or the use of oil with low concentrations of ash,
salts, and/or carbon residue. Boilers that do not comply with this limit without controls might
use an existing ESP (e.g., one originally installed to reduce flyash emissions from a coal-fired
boiler before it was converted to oil) or perhaps in some cases, a high efficiency multicyclone.
The specified levels require a 50 percent reduction from the maximum emission rates reported for a
sample of uncontrolled oil-fired utility boilers (see Figure 7-1, which shows the upper envelope of
uncontrolled levels and curves which represents 50 and 75 percent control, respectively, from the
upper envelope). Similarly, the levels in measures No. 4 and 5 are based on increasingly greater
reduction from the curve of maximum emissions in Figure 7-1, with a lower bound at 0.01 Ib/MBtu;
precipitators generally cannot achieve lower levels (equivalent to 0.005 gr/scf at 3-5 percent ex-
cess air). The relationship and control measures in Table 7-5 correspond approximately to the
following emission levels in Ib/MBtu (ng/J).
50 MW
200 MW
£550 MW
Measures No. 1, 4 (50% collection)
Measures No. 4, 5 (75% collection)
Measure No. 5 (<95% collection)
0.08 (34.4)
0.04 (17.2)
0.02 (8.6)
0.05 (21.5)
0.03 (12.9)
0.01 (4.3)
0.02 (8.6)
0.01 (4.3)
0.01 (4.3)
7-29
-------
C
CO
o
oo
II
3
CO
/ .
/I- /
>;«
o
o
CU
>
to
O
o
Lf)
o
o
o
o
CO
o
o
C\J
o
to
Q-
(O
o
CO
QJ
Q.
O
S-
OJ
o
CO
T3
o
0} i
cu
o >
C Ol
CO r
i- O
a; t-
r 4J
i- C
o o
-Q U
>> 0)
r (/)
4-> CO . -
3 O IO
D.CV)
U I
r- 73 «3-
V- C
QJ
O
C
a>
S..
o a>
tO 4-
CL a;
i tO Of.
r O
£M- §
M O S_
C M-
O C
O O tO
C <- (->
3 4J to
O T3
g C
O 3 TJ
S- >4- Qi
<_)
«U
o
o
fO i
to O
c to s..
O fO )-'
r- C
tO ' - O
to "O U
r- 0) C
E tO 3
cu
3
CO
o
o
CM
O
CO
r
C3
CM
O
O
O
O
'suo.ss.iua
7-30
-------
It is indicated that the emission limits can apply during soot blowing as well as during nonsoot
blowing as long as these limits do not require more than 99 percent control (from an assumed con-
centration level of 20 times the normal one for the soot blowing period). One source has stated
that mass emissions during soot blowing can be'as much as 2-0 times higher than they are during nor-
mal operation (see Section 4.3.2.1 and Reference 4-31). He has also shown that moderate and high
efficiency ESP's allow no greater particulate mass to pass through during soot blowing than they
do during the other operating'periods. Therefore, the limit specified in control measures 1, 4,
and 5 also apply to soot blowing periods. If emissions during soot blowing are truly 20 times
higher than they are otherwise, the ESP would have to collect 98 percent of the emissions to comply
with this limit. Commercially available units that are designed for oil-fired boilers can achieve
this efficiency when the inlet grain loading is high (their major constraint is a minimum outlet
loading of 0.005 gr/scf, or about 0.01 Ib/MBtu).
Since these limits are based on reductions from the maximum reported levels and on the high
multiplier of 20 for soot blowing, no power plant should have technical difficulties in,complying
with them. Although the limits in measures No. 4 and 5 appear to be very stringent, one should
remember that the hourly'emissions from a power plant are much higher than they are from boilers
in the other source categories. For example, one 500 MW plant that is controlled to 0.01 Ib/MBtu
emits as much per hour as 12,000 homes with typical uncontrolled distillate-fueled furnaces.
Since these home heaters do not operate continuously or year around, the annual emissions from a
utility boiler are equivalent to those from nearly 100,000 homes.
For existing boilers it may be appropriate to consider the age of the unit when establishing
emission limits. Utility boilers generally have a finite life (e.g., circa 40 years), and it may
be most cost-effective for a region to spend more on controlling newer plants with many years left
than on older ones which will soon be removed anyway. The applicability of a system of reduced
control requirements for older sources is dependent on the severity of the local pollution problem
and on the distribution of sources in that area. Such reduced control requirements should be
negotiated on a case-by-case basis between the control agency and the source. Moreover, any regu-
lation which takes account of plant age should, however, not discourage a source from delaying the
replacement of an older, less efficient, higher emitting source with a new, controlled one. There-
fore, consideration should be given to the userof variances, bonds .(financial assurances .that the
old source will be removed or upgraded by a certain date), or substantial fines if the source is
not removed or upgraded by a certain date.
7-31
-------
The effectiveness of increased burner servicing (control measure No. 2) is given in terms of
a maximum expected value because this figure is based on a projection that has not yet been substan-
tiated. The requirement for continuous stack gas monitoring instrumentation on new units outlined
in measure No. 3 is redundant because they are already equipped with these devices in compliance
with NSPS. This control measure is included here in the event some control districts do not already
require it on their existing sources. The last control measure has been discussed in the previous
subsection on control of industrial systems.
7.3 ENFORCEMENT OF THE POSSIBLE CONTROL MEASURES
Before an agency decides what control measures to implement, it should investigate fully
how it may utilize measures presently required. For example, revisions were made to 40 CFR 51
"Requirement for Submittal of Implementation Plans," and 40 CFR 60 "Standards for New Stationary
Sources," October 6, 1975 (40 FR 4620) with regard to emission monitoring. These Federal regula-
tions set forth minimum requirements for continuous emission monitoring and recording that each
State Implementation Plan must include for specific existing source categories and that each
source category covered by New Source Performance Standards must include. Existing fossil fuel
steam generators are covered under 40 CFR 51, Appendix P for State Implementation Plans and new
fossil fired steam generators are covered in 40 CFR 60.45 under New Source Performance Standards.
The emission monitoring required by these regulations would satisfy the opacity monitoring require-
ments of control measure Nos. 2 and 3 in Table 7-5 for boilers and furnaces of greater than 250
MBtu/hr (73.2 MW). It should be possible for an APCD to obtain this information.
To judge the stringency of a regulation, Federal New Source Performance Standards (NSPS)
can be used as a guide. These standards are based on best available control technology consider-
ing costs. However, state and local regulations can be more stringent than NSPS and should be for
the case of fossil fired steam generators. This standard of 0.18 g/Mcal heat input (0.1 Ib/MBtu
or 43 ng/J) is meant for coal-fired generators, and much more stringent emission standards can
obviously be met with oil-fired boilers.
7-32
-------
APPENDIX A
POTENTIAL RESEARCH AND DEVELOPMENT TOPICS FOR PARTICIPATE
CONTROL FROM OIL-FIRED BURNERS AND BOILERS
Several areas of research into participate emissions, their control, and their fate are
currently being pursued. Successful conclusion of these research efforts will assist in the devel-
opment of more refined control programs that should lead to significant and identifiable improve-
ments in the air quality. The topics which are being considered fall into two broad categories
problem definition and control technology. In the first category the work that is underway and
/
needs to be continued is directed at the following topic areas:
The relationship between emission reductions from a given source category and consequent
decreases in ambient air concentrations of particulate matter
9 The relationship between particulate size, physical characteristics, and chemical com-
position on the one hand, and health hazard or welfare degradation, on the other
The relationship between opacity, mass emissions, and smoke spot number
The precursor/fate relationship for both primary and secondary particulates
Development of improved control technology is currently aimed at methods to collect fine particu-
lates and changes to burners and/or burner/combustion chamber systems. This latter work has tended
to stress smoke reductions (as measured by the Bacharach Smoke Spot Number) in residential and com-
mercial sized units and NO reductions in the larger burners found in industrial and utility boilers.
Since some combustion modifications for NO control can increase particulate emissions, additional
research appears to be necessary to develop methods which can reduce both NOX and smoke. In addi-
tion, studies are required to determine optimum maintenance practices (methods and frequency of
maintenance) and soot blowing procedures for industrial and utility boilers. These subjects are
discussed in somewhat greater detail below. It should be noted, however, the recommended R&D is not
necessarily all-inclusive or presented in order,of priority; it was not within the scope of this
task to go beyond a simple identification of potentially fruitful R&D topics.
Fundamental burner design research has resulted in the development of an "optimum" oil burner
for distillate-fired residential and commercial sized units (Section 4.2.3). These burners* need to
A-l
-------
be field tested to verify the results obtained in labora"->ry experiments. A similar program is
under way with residual oil-fired burners for commercial burners (under the auspices of 6. B. Martin,
Combustion Research Branch, Industrial Environmental Research Laboratory - RTP, EPA). Although the
main purpose of this task is to develop a low NO burner, a secondary goal is to minimize particulate
emission.
Several different atomization methods have been suggested for particulate reduction and/or
improved fuel economy. These include the acoustic and ultrasonic schemes (Section 4.1.4.1) and the
Babington burner (Section 4.2.3). Their capabilities should be evaluated by lab and field tests.
Similarly, tests should be conducted on industrial and utility boilers to determine the potential
for reducing particulate emission from residual oil-fired boilers when equipped with viscosity con-
trolled preheaters.
As noted several times in Section 3, further R&D is required on the emission reduction, health
Impact, and potential operational problems of additives. This could be a very fruitful area of work
because of the large emission reductions that may be achievable by cost-effective means that are
relatively easy to implement and enforce. On the subject of fuel treatment, some work may be justi-
fied to further evaluate the value of fuel washing, especially as a means of eliminating the need
for adding corrosion inhibiting MgO additives, which then become particulate emissions, themselves.
There appears to be little information on the effects of soot blowing on particulate emis-
sions. In order to realistically evaluate the impact (if any) of various methods of blowing soot,
measurements are needed of both particulate loading and size distribution in the stack during soot
blowing and nonsoot blowing operation. A preliminary evaluation should be conducted of a number of
boilers using different blowing procedures. More detailed measurements of emission characteristics
as a function of blowing frequency may then be in order. The emission test results should then be
related to control equipment capabilities and resulting ambient air levels (through the use of dis-
persion models) to determine the real impact of soot blowing procedures on the ambient particulate
levels.
Electrostatic precipitators (ESPs) are the most commonly used particulate control devices for
large boilers. Several operational problems were identified in Section 4.3 relative to the use of
ESPs on oil-fired units. Development work may be appropriate to overcome these potential deficien-
cies and, thereby, add to the confidence that the ESP will perform continuously at its design effi-
ciency. The discussion in Section 4.3 also pointed out that scrubbers and baghouses possibly could
be used effectively on oil-fired units, but here too more R&D is required.
A-2
-------
The difficulty in making accurate particulate emission measurements may prevent a control
agency from adopting the most cost-effective control strategy. EPA Method 5 is expensive because
it requires sophisticated sampling equipment and much time. Smoke spot methods, on the other hand,
are fast and relatively simple, but they give no quantitative emission data (see Appendix B). There-
fore, efforts should be made to develop a simpler mass emission sampler, or at least one which will
give approximate data. For example, tests could be conducted on a variety of residential, commer-
cial, and small utility systems to seek a "representative" sampling point that obviates the need
for a stack traverse.
Several local agencies, as well as the EPA, have initiated programs to more clearly identify
the causes of high ambient particulate levels. The microscopy studies mentioned in Section 1.1 are
an example of such efforts. More work is required to establish more positively the,relationship
between specific source emissions (both individual point sources and distributed area sources) and
ambient air quality. A typical question that needs to be answered is the relative impact of many
small, low emission point sources (e.g., home heaters) as opposed to a few large, tall stack sources
(e.g., power plants). Other questions deal with the relative importance of fugitive dust and natu-
ral sources of particulates, the generation of secondary particulate in the atmosphere, etc.
A-3
-------
-------
APPENDIX B
MEASUREMENT OF SMOKE AND PARTICIPATE EMISSIONS
Several techniques are commonly used to obtain an indication of combustion generated
particulate emissions. One approach is to determine the opacity of the plume, either by human ob-
server or transmissometer. Another is to obtain a measure of the soiling capacity of the particu-
lates in the exhaust by means of a Smoke Spot reading (e.g., Bacharach test). A direct measurement
of the particulate loading is obtained by EPA Method 5. All three approaches actually measure dif-
ferent properties of the exhaust. Opacity describes the light obscuring ability of the particulates
and is affected largely by the size of the particulates in the exhaust and their surface- character-
istics. EPA Method 5 measures mass emissions, but, unless the probe is fitted with additional
equipment, it does not differentiate between heavy particles that might fall out in, or near, the
private property surrounding the source and light particles which will remain suspended longer and
contribute to the TSP. The Smoke Spot Number of a discharge gives a measure of the soiling char-
acteristics of the particulate in the exhaust. Therefore, the reading obtained depends, in part,
on the "stickiness" and the color of the particles. In addition, it is a very useful diagnostic
for evaluating burner performance because it can measure the quality of the exhaust when the plume
is invisible the lower readings on the smoke spot scale (i.e., approximately 0 to 6 on a scale
that goes from 0 through 9) are obtained in plumes which are not visible. Since EPA Method 5 and
smoke spot are affected differently by the characteristics of the exhaust, we present below a brief
comparison of the two techniques.
EPA Method 5 consists of isokinetically withdrawing a volume of effluent from the gas stream
and collecting the particulate on a filter followed by a series of impingers (Federal Register,
Vol. 36, No. 247). A cyclone may also be used at the front end of the sampling system to collect
large particules which would otherwise clog the filter rapidly. The particulate loading for com-
pliance purposes is determined by the weight added to the filter (after it is dried). The impinger
is used for further characterizing emissions.
The accuracy of the method is dependent mainly upon the completeness of removal of the sample
from the sampling system, especially for sources where emissions are relatively low. Battelle has
B-l
-------
recently used a modification of EPA 5 which includes more thorough washing of the probe and im-
pingers (Reference B-1). This modification (subsequently referred to as MEPA-5) results in par-
ticulate measurements on residential units and commercial boilers which are approximately 1.7 times
greater than those made by the standard method.
Due to the complex and time consuming nature of Method 5, several simpler (and necessarily
more qualitative) measurement techniques are commonly used. The most important of these is mea-
surement of the Smoke Spot Number. In this test (defined precisely in ASTM Procedure D2156-65) a
sample of the effluent gas is withdrawn by means of a hand pump, and particulates are collected on
a spot on filter paper. The amount of pu. ciculate collected is inferred visually from the darken-
ing of the spot by comparison with a set of standard spots. The smoke concentration is then assigned
a number based on the corresponding standard spot.
Several attempts have been made to relate particulate mass emissions to Smoke Number, but
none have been very successful (see References Bl and B2). Data from one study (Reference B-1)
shows widely different particulate levels even at equivalent smoke numbers. The most important
reasons for this lack of correlation appears to be the dependence of Smoke Spot Number on particle
size distribution and physical characteristics. In addition, EPA 5 measurements are often averaged
over a long period of time (e.g., 10 minutes), while smoke spot determinations are essentially in-
stantaneous in nature.
REFERENCES FOR APPENDIX B
B-1. Barrett, E. R., Miller, S. E., and Locklin, D. W., "Field Investigation of Emissions from
Combustion Equipment for Space Heating," Battelle, Columbus Laboratories, EPA R2-73-084a,
June 1973.
B-2. Levy, A., et al., "A Field Investigation of Emissions from Fuel Oil Combustion for Space
Heating," Battelle, Columbus Laboratories, API Publication 4099, November 1, 1971.
B-2
-------
APPENDIX C
UNIT CONVERSIONS
The following presents multiplication factors to be used in converting from engineering to
SI units. This table is restricted to those physical quantities which are encountered frequently
throughout the text and in the tables and figures of this document. Additional conversion factors
can be obtained from "American National Standard E380-72, Metric Practice Guide."
To convert from
Ib/MBtu
MBtu
MBtu/hr
gph (gal/hr)
gr/SCF
To
nanogram/Joule (ng/J)
Giga Joule (GJ)
Megawatt (MW)
cubic meters/sec
g/m3 (0°C, 1 atm)
Multiply by
430
,1.06
0.293
1.05 x ID'6
2.42
The above conversion from MBtu/hr to MW is for thermal MW's - i.e., the burner input heat rate as
expressed in SI units. Frequently it is desired to convert boiler heat input rate to an equivalent
electrical output rate. This conversion is approximately 10 MBtu/hr per MW. The nuclear power
industry generally uses the subscrips t and e to distinguish thermal (input) from electrical (out-
put) power.
Other conversions within the engineering system of units are encountered often enough to be
included here.
To convert from
1000 Ib steam/hr
bhp (boiler hp)
9Ph
gph
Ib parti cul ate/1 000 gal
Ib parti cul ate/1 000 gal
psi
In. W.G.
°F
inches
feet
kinematic viscosity in centistokes (cSt)
To_
MBtu/hr
MBtu/hr
MBtu/hr
MBtu/hr
Ib/MBtu
Ib/MBtu
Pascal (Pa) or Newton/m2 (N/m2)
Pa
°C
meters
meters
m2/sec
Multiply by
1
0.0335
0.140 (No. 2 oil)
0.150 (No. 6 oil)
0.00714 (No. 2 oil)
0.00667 (No. 6 oil)
6.895 x 103
249
5/9 (°F-32)
2.54 x 10~2
0.3048
TO"6
The conversions that depend upon the oil are based on 140,000 Btu/gal for No. 2 oil and 150,000 Btu/
gal for No. 6 oil .
C-l
-------
-------
APPENDIX D
SAMPLE CALCULATIONS TO ANNUALIZE COSTS
One useful way of comparing the costs associated with a variety of emission control tech-
niques, especially when they involve differing initial and continuing costs, is to annualize all
the costs. We present here two examples of such a calculation, one using the costs associated
with control measure No. 4 for residential units (see Table 7-2) and one with measure No. 5 for
utility boilers (see Table 7-5).
Control measure No. 4 for existing residential boilers and furnaces specifies that they meet
stringent emission limits, which means that many owners will need to purchase a new burner and have
it serviced annually. A benefit of this control measure would be a cost saving due to the improved
fuel consumption of new burners relative to the older ones they usually replace. Based on the fig-
ures shown on Table 7-2, the owner could, therefore, incur a one-time cost of about $300 to pur-
chase the burner and an annual cost of $30 to service it. We will assume an intermediate value of
3 percent reduction in fuel consumption, an annual consumption of 1000 gallons (3.78 m3), and a
constant fuel cost of 40<£/gal. The initial cost is annualized using the capital recovery factor,
CRF, with an interest rate of 10 percent (as in a home improvement loan) and a life of 10 years.*
Then
Annual costs (AC) = CRF x Initial costs (I) + Annual service cost - Fuel savings
= 0.163 x 300 + 30 - 0.03 x 1000 x 0.40
or
AC * $67
The cost calculation for the addition of an electrostatic precipitator to a utility boiler
is more complicated because one must include.at least depreciation and the tax reduction associated
with the fact that normal expenses reduce the taxable income. For simplicity, we shall assume
CRF
= id + i)n , where i = interest rate (0.10 here) and n = number of years (10 here).
D-l
-------
straight line depreciation, instead of an accelerated approach such as sum-of-the-years, and ignore
any possible investment tax credit. We shall also use the same basic calculation procedure as in
the residential example instead of a discounted net cash flow computation. These simplifications
«
are acceptable for comparing area wide strategies, given the oth.r uncertainties in'the comparison,
even though they should not be used by any individual firm when comparing two clearly defined alter-
native proposals. An interest rate of 10 percent will be used again, but the life of the system
will be taken as 15 years, which is more representative of the life of large systems, at least for
accounting purposes. The computation for a high efficiency unit with a capacity of 300,000 acfm
(i.e., approximately 100- MW electric output) is then as follows (assuming a stack gas temperature
of 350°F, 3 percent excess oxygen, and a plant thermal efficiency of 38 percent).
1. Installed cost (see Figure 4-llb) = $825,000
« 2. Fixed cost = CRF x (1) = 0.131 x 975,000 = $108,000
3. Operating costs = $0.05/acfm x 300,000 acfm = $ 15,000
4. Maintenance Cost = $0.03/acfm x 300,000 acfm = $ 9,000
5. Before tax costs = (2) + (3) + (4)
6. Operating tax credit = 0.5 x (5)
7. Net annual costs = AC = (5) - (6)
= $132,000
= $ 66,000
= $ 66,000
A similar calculation for a medium efficiency precipitator (installed cost = $575,000, operating
costs » $Q.03/acfm) yields AC = $46,800. If these plants operate 8000 hours per year, the unit
costs of the precipitators are:
AC/000 MW x 8000 hr) = $0.083/MW-hr = 0.008
-------
One recent report suggests that ESP's designed for oil-fired boilers cost $5 - $8/acfm.* A
value of $3.3/gcfm was used in the above calculation for high efficiency units. At $8/acfm, the
net annual cost of the preci pita tor would be 0.02U/kW-hrj and the before tax costs would represent
0.4 percent of a 5£/kW-hr charge for electricity.
The calculation for an industrial boiler follows the same procedure, but here it is most rea-
sonable to relate the cost increase to the amount of steam produced. If one assumes 20 percent ex-
cess air, 350°F stack gas temperature, 80 percent thermal efficiency, and a 100,000 Ib steam/hr
industrial boiler, the exhaust flow is about 50,000 acfm. The cost calculation then proceeds as
above, using the following costs.
Cost
Installed ($)
Operating ($/acfm/yr)
Maintenance ($/acfm/yr)
High Efficiency Medium Efficiency
430,000 200,000
0.05 0.03
0.03
0.03
The results of these calculations yield annual costs of $30,200/yr and $14,600/yr, respectively,
for this 100,000 Ib steam/hr boiler. Assuming that this boiler is used only 4000 hr/yr (i.e., two
shifts of 5 days/week), the above costs become $0.37 x 10~Vlb steam for the medium efficiency pre-
cipitator and $0.75 x lO'Vlb steam for the high efficiency unit. Fuel costs of $2.5/MBtu are equi-
valent to $31.3 x lO'Vlb steam for an 80 percent efficient unit; therefore, the medium and high
efficiency precipitator costs represent 1.2 percent and 2.4 percent of these fuel costs, respectively.
"Partuculate Emission Control Systems for Oil-Fired Boilers," 6CA Corporation, Prepared for U.S.
EPA, Report No. EPA-450/3-74-063, December 1974. (Reference 4-36)
D-3
-------
-------
APPENDIX E
AD HOC ADVISORY COMMITTEE - PARTICULATE EMISSIONS FROM OIL BURNERS
Name
Affiliation
Specialty
Industry and Public Interest Groups
Ahmed, A. Karim
(Staff Scientist)
Axtman, William
(Assistant
Executive
Director)
Boll , Richard H.
(Research
Specialist)
Borger, Henry
(Executive
Secretary,
Federal Construc-
tion Council)
Craig, Glen
(Director,
Research and
Development)
Desmond, John A.
(District Director)
Fisher, Len
(Vice-President,
Engineering)
Lock! in, David W.
(Program Manager,
Combustion and
Energy Utilization
Research)
McMillan, R. E.
(Development
Engineer, Applied
Thermodynamics)
Robison, Ernest
(Chief, Division
of Engineering)
Rosche, Paul
(Assistant Head,
Fossil Fuels
Division)
Sommerlad, Robert
(Manager,
Development
Contract Operation
Department)
Wei land, John H.
(Coordinator,
Environmental
Protection Dept. )
Weller, Burton L.
(Staff Engineer)
Woodworth, John
(Technical Director)
Natural Resources
Defense Council
American Boiler
Manufacturers
Babcock & Wilcox
Alliance Research
Center
Building Research
Advisory Board,
National Academy
of Sciences
Aqua-Chem, Inc.
Clever-Brook Division
Metropolitan Boston
Air Pollution Control
District
The Carl in Company
Battelle-Columbus
Research Division,
Foster Wheeler Energy
Corporation
Maryland Bureau of
Air Quality and Noise
Control
Department of Air
Resources,
Environmental Pro-
tection Administration,
City of New York
Research Division,
Foster Wheeler
Energy Corporation
Texaco, Inc.
(Representing A.P.I.)
National Oil Jobbers
Council (NOJC)
Hydronics Institute
Public Interest
Group
Industry Expert
Particulates
and Soot
Blowers
Commercial
and Industrial
Construction
Industrial
Research
Regulatory
Official
Residential Oil
Burning Units
Residential
Units, Fuel
Additives,
Particle Size
Distributions
Power Plants
Regulatory Offi-
cial - Residual
Oil Boilers
Regulatory
Official
Power Plants
Fuel Oil
Utilization
Residential
Heating-
Residential and
Smal 1 Commer-
cial Units
E-l
-------
AD HOC ADVISORY COMMITTEE - PARTICULATE EMISSIONS FROM OIL BURNERS (Concluded)
Name
Affiliation
Specialty
Environmental Protection Agency
Finfer, Edward
(Senior Plan
Advisor)
Hall, Robert E.
(Research Mechanical
Engineer)
Harmon, Dale
(Chemical Engineer)
Iverson, Reid E.
(Chemical Engineer)
Pace, T. G.
(Environmental
Engineer)
Sableski, J. J.
(Chief, Plans
Guidelines Section)
Schueneman, J. J.
(Director, Control
Programs Development
Division)
Venezia, Ron
(Environmental
Engineer)
Walsh, Robert
(Chief, Control
Technology Office)
Air Branch
EPA, Region II
EPA (IERL)*
EPA (IERL)*
EPA (ESED)**
EPA (CPDD)***
EPA (CPDD)***
EPA (CPDD)***
EPA (IERL)*
EPA (ESED)**
Combustion
Applications
Combustion
Research
Combustion
Research
Combustion
Applications
Parti cul ate
Sources
General
General
Combustion
Research
Combustion
Applications
*
Industrial Environmental Research Laboratory
**
Emission Standards and Engineering Division
***
Control Programs Development Division
E-2
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO. -
EPA-450/3-76-005
3. RECIPIENT'S ACCESS! ON«NO.
TITLE AND SUBTITLE
CONTROL OF PARTICULATE MATTER FROM OIL BURNERS AND
BOILERS
5. REPORT DATE
April 1976
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
6.R. Offen, J.P. Kesselring, K. Lees G. Poe, K.J. Wolfe
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
Aerotherm Divlslon/Acurex Corporation
485 Clyde Avenue
Mountain View, CA 94040
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-1318
2. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
5. SUPPLEMENTARY NOTES
6. ABSTRACT
This report identifies possible control measures which Federal, State, or
local control agencies might use to reduce particulate emissions in areas where
the combustion of oil has a significant impact on air quality. T,. arrive at viable,
effective controls, a comprehensive survey was conducted into such cechniques
as burner design changes, fuel restrictions, additives, mandatory periodic inspection/
maintenance programs and particulate collection devices. Emerging technology that
may assist boiler and furnaces operators to achieve lower particulate emissions
was also identified. The experiences gained by agencies with active particulate
control programs were evaluated, and their procedures included among the list of
possible control measures where appropriate. These findings are summarized along
with tabulations of possible control measures according to their stringency and
applicability to residential, commercial, industrial, and utility boilers and
space heating furnaces.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Oil Burners, Boilers and Furnaces
Burner Design Changes
Use of Additives
Mandatory Inspection/Maintenance
Particulate Collection Devices
Fuel Oil Utilization
Particulate Control
Regulations
Emerging Technology
13. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)'
Unclassified
21. NO. OF PAGES
288
Unlimited
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
F-1
-------
-------