EPA-600/2-77-028
January 1977
Environmental Protection Technology Series
RESIDENTIAL OIL FURNACE SYSTEM
OPTIMIZATION
Phase II
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161
-------
EPA-600/2-77-028
January 1977
RESIDENTIAL OIL FURNACE
SYSTEM OPTIMIZATION
PHASE II
by
L. P. Combs and A.S. Okuda
Rocketdyne Division/Rockwell International
6633 Canoga Avenue
Canoga Park, California 91304
Contract No. 68-02-1819
ROAPNo. 21BCC-027
Program Element No. 1AB014
EPA Project Officer: G. Blair Martin
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
The report describes the second phase of an investigation into
ways to improve the air pollutant emission and thermal effi-
ciency characteristics of residential oil furnaces. A proto-
type, low-emission, warm-air furnace, designed in the first phase
to embody a number of burner and combustor criteria for minimiz-
ing emissions compatible with high efficiency, was assembled and
tested. Design details were changed as necessary during labora-
tory testing to help achieve the objectives. Applicability of
the design criteria within current conventional oil-heat industry
practices was demonstrated. Compared with estimated average char-
acteristics of existing installed residential furnaces and boilers,
NO emissions were reduced by 65% or more, and steady-state effic-
iency was increased by a minimum of 10 percentage points. Experi-
mental results and component changes made in obtaining them were
incorporated into a preliminary design for an integrated low-emis-
sion furnace which should be commercially producible and cost-
competitive.
iii/iv
-------
CONTENTS
Section I: Conclusions 1
Section II: Recommendations 3
Section III: Introduction 4
Summary of Phase I 6
Section IV: Experimental Investigation of the Prototype
Low-Emission Furnace 12
Experimental Apparatus 14
Stock Furnace Prior to Modifications 14
Prototype Optimum Furnace 16
Test Facility and Instrumentation 25
Test Results 32
Stock Lennox Furnace 32
Prototype Optimum Furnace 39
Exploratory Modifications of th£ Prototype Optimum
Furnace Configuration 47
Discussion 67
Comparison With Other Residential Furnaces 67
Operational and Design Aspects 72
Section V: Integrated System Design 75
Burner Assembly 75
Combustion Air Control Device 75
Optimum Burner Head 79
Finned Air-Cooled Combustion Chamber 79
Heat Exchange Considerations 84
Vestibule Closure Panel 85
Section VI: References 86
Appendix A
Flue Gas Compositional Analysis 87
Appendix B
Data Tabulations: Stock Lennox Furnace and Prototype Optimum
Furnace Experiments 100
Appendix C
Data Tabulations: Experimental Furnace Tests With Heat
Exchanger Modifications 112
-------
ILLUSTRATIONS
1. Cutaway Drawing of the Stock Lennox Model 011-140 Furnace
(Reproduced from a Lennox brochure) 15
2. Layout Assembly Drawing, Prototype Optimum Warm-Air
Oil Furnace 17
3. Prototype Optimum Warm-Air Residential Oil Furance ... 18
4. Schematic of the Skewed-Lip "Quiet Stator" Extension
for the Optimum Burner Combustion Air Fan 20
5. Front View Photograph ot the Finned, Air-Cooled
Combustor for the Low Emission/High Efficiency Proto-
type Furnace System 22
6. Installation of the Finned Firbox and Compact Heat
Exchanger in the Prototype Optimum Furnace 26
7. Schematic of the Furnace Performance Evaluation
System 27
8. Flue Gas Gross Thermal Efficiency Losses as a Function
of Net Flue Gas Temperature and Composition 29
9. Cycle-Averaged Nitric Oxide Emissions from the Lennox
011-140 Furnace in its Stock Configuration 35
10. Thermal Efficiency Characteristics of a Stock Lennox
011-140 Warm-Air Oil Furnace Fired at 1.05 ml/s
in 4-Min-On/8-Min-Off Cycles 37
11. Comparison of Cycle-Averaged Nitric Oxide Emissions
of Various Oil Burners in the Prototype Furance and
in Their Respective Furnaces 40
12. Comparison of Thermal Efficiency Characteristics of
the Prototype Optimum Furnace With Various Burners
and System Changes at a Nominal Warm-Air Flowrate
of 0.566 m3/s 42
13. Schematic of the Finned-Combustor Williamson
Furnace Modification 49
14. Comparison of Cycle-Averaged Nitric Oxide Emission
Characteristics of Two Finned Combustion/Air-Cooled
Heat Exchanger Furnaces 50
VL
-------
J5. Schematic of the CJoi]-Cooled Modified Prototype
Furnace 52
16. Steady-State Nitric Oxide Emissions From the Coil-
Cooled Prototype Furnace With Various Coolant Media
and Configuration Changes 53
17. Schematic of the Prototype Optimum Furnace With an
Experimental Flue Gas Recirculation System 59
18. Prototype Optimum Furnace With Various Configurations
of Internal Water-Cooled Coil and Baffles 61
19. Steady-State Nitric Oxide Emission Concentrations
From the Prototype Optimum Furnace With an Additional
Water-Cooled Coil 62
20. Schematic of the 8-Tube, Low-Prcssure, Air-Cooled,
Supplemental Heat-Exchanger Installation in the
Prototype Optimum Furnace 66
21. Comparison of Cycle-Averaged NO Emissions From the
Prototype Optimum Furnace and Other Oil Furances ... 68
22. Comparison of Steady-State Thermal Efficiencies of the
Prototype Optimum Furnace and Other Oil Furnaces ... 71
23. Cutaway Perspective Drawing of Integrated, Low-Emission,
Warm-Air Furnace Design 76
24. Layout Assembly Drawing of the Integrated Furnace
Design 77
25. Drawing of the Combustion Air Control Device 80
•
26. Stamped and Folded Sheet Stainless-Steel Optimum Head . . 81
27. Cast-Metal-Finned, Air-Cooled Combustion Chamber .... 83
vLi
-------
TABLES
1. Furnace System Evaluation Test Matrix . 33
2. Comparison of Furnace Operating Conditions and
Cycle-Averaged Pollutant Emissions 70
C
v L i i
-------
SECTION I
CONCLUSIONS
1. Residential space heating systems can be designed in ways that re-
duce NO emissions substantially and that are also compatible with:
(1) low emissions of carbonaceous air pollutants, (2) high thermal
efficiencies, and (3) conventional oil heating industry practices
in manufacturing, marketing, installing, and servicing units.
2. Design criteria for low emission oil burners and combustion cham-
bers, derived from earlier research, have promising potential in
the development of efficient low-emission warm-air furnaces and
hydronic boilers.
3. Although appreciable reductions in NO emissions may be obtained
by partial application of the design criteria (e.g., by retrofit-
ting low-emission components into existing furnaces or boilers),
maximum benefits can be achieved only by considering design opti-
mization of the entire heating unit. By using the latter approach,
supplementary concepts which improve efficiency or otherwise reduce
residential fuel consumption also can be integrated into the total
design.
4. The prototype, low-emission, warm-air furnace, built and tested
to demonstrate proof-of-concept, showed the validity of these con-
clusions by achieving:65 to 70% reductions in NO emission levels;
X
acceptably low CO, UHC, and smoke emission levels; and steady-
state efficiencies approaching the 35% maximum for noncondensing
flue gas systems.
5. Cycle-Hveraged efficiencies, which are more difficult to measure
and less well documented for existing residential heating equip-
ment, were estimated to be a minimum of 10 percentage points higher
with the prototype furnace than the average of central residential
-------
oil-fueled furnaces. High steady-state efficiency contributed only
part of that increase; devices designed to reduce standby liedt
losses (e.g., a draft damper in the combustion air supply and n
sealed air system) accounted for the rest.
-------
SECTION II
RECOMMENDATIONS
The full value of *:he low-emission technology developed in this investi-
gation may be realized by applying it commercially. To further investi-
gate and demonstrate potential benefits of commercialization, it is
recommended that plans now being made to perform field testing of sev-
eral optimum low-emission furnaces in actual residences be carried out.
The planned program will be conducted in two phases. In the first
phase, further refinements of the optimum furnace design will be ef-
fected, partially to further optimize emissions and efficiency perform-
ance and partially to improve commercial producibility. Included will
be analytical and experimental investigations for: (1) simplifying
fabrication and reducing the mass of the firebox, (2) improving heat
exchanger effectivi'.y, (3) ensuring adequate performance at low ambient
temperatures, and (4) satisfying applicable codes and standards. In
the second phase, the finalized design will be used to construct ap-
proximately six low-emission furnaces for residential field testing
during the 1977-1978 heating season. Operation in different types of
residences and in at least two different climates is expected to yield
definitive data on the achievable levels and constancy with time of
air pollutant emissions and steady-state efficiency, as well as cycli-
cal and season-averaged efficiencies, general operability, and any un-
usual service requirements.
-------
SECTION Til
INTRODUCTION
This report documents the second phase of a research program to estab-
lish technology for optimization of residential oil heating systems with
respect to minimizing pollutant emissions and increasing heating system
thermal efficiency. General overall goals have been to reduce emis-
sions of oxides of nitrogen to less than 0.5 g NO/kg fuel burned, while
maintaining minimum emissions of CO, UHC, and smoke, and to increase
overall season-averaged furnace energy efficiencies by 10% or more above
those achieved by current conventional systems. Emphasis was also
placed on minimizing departures from existing heating industry manufac-
turing, distribution, installation, operation, and servicing practices
which would be required to implement the developed technology.
The current research program is one of a series whicli Rocketdyne has
carried out for the Environmental Protection Agency. The series began
with an intensive investigation of residential and commercial oil
burners (Ref. 1) whicli led to criteria for optimizing conventional
burner designs with respect to pollutant emissions. For high-pressure
atomizing, luminous-flame burners, it was found that:
1. Uniform mixing is beneficial. At a given overall burner
stoichiometric ratio, NO production is reduced by minimizing
local deviations from that overall ratio.
2. High-temperature adiabatic eddies embedded in the flame zone
should be avoided. Long gas residence times in such eddies
increase the production of NO .
3. liddy recirru l.ition nenr rool surfaces, on tlie other hand, may
be beneficidl. They help to reduce NO formation by supplying
partially cooled vitiated combustion Rases to the flame zone
and, by dilution, Lower flame temperatures .ippreci.ihly.
-------
ft. Uniform one-dimensiona] (plug) flow is preferable to strongly
swirling flow. Strong swirl promotes higher NO production
unless: (1) local stoichiometric ratios generally exceed 1.5
(i.e., greater than 50% excess air), (2) the foregoing cooled
recirculation is induced, or (3) combustion is completed so
rapidly that the combustor can be made very short (i.e.,
residence times at flame temperature are minimized).
Based on these criteria, minimum pollutant emissions were obtained with
burners having: (1) no flame-retention device, (2) choke diameter re-
lated quantitatively to the firing rate, and (3) oversized internal
peripheral swirler vanes which promoted reactant mixing but without
creating excessive turbulent recirculation. These burner design attri-
butes were all concerned with the burner "head", i.e., that portion of
the burner which admits prepared reactants into the combustion chamber.
For that reason, this development was referred to as the "optimum head."
In addition to minimizing formation of oxides of nitrogen, the optimum
head could be fired in the laboratory with considerably less excess
combustion air, without producing unacceptable levels of carbonaceous
pollutant emissions, than is conventional practice. Reducing excess
air decreases the sensible heat lost with the flue gases, so applica-
tion of the optimum head also has a potential for increasing overall
furnace fuel utilization efficiency.
A further aspect of the research reported in Ref. 1 was that some vari-
ations In the combustion chamber construction (chamber diameter and
relative orientations of the axes of the burner and chamber) affected
emission levels. The results clearly demonstrated that pollutant emis-
sions are sensitive to the design of each component comprising a resi-
dential heating combustion system and to design interactions among the
comnonents. It became obvious, therefore, that minimum emissions
-------
could ho achieved only by systematically optimizing the burner in con-
junction with the combustion chamber as well as with the furnace oper-
ating mode.
Nonetheless, it was recognized that the optimum burner head alone, as A
retrofit device for existing burners in existing furnaces, might be com-
mercialized more rapidly and less expensively than could an entire op-
timized furnace. Thus, in addition to the current program's investiga-
tions toward delineating requirements for optimizing the entire furnace,
a parallel study addressed the feasibility of direct commercialization
of the optimum head (Ref. 2). Two newly manufactured warm-air oil
furnaces were retrofitted with optimum heads made to simulate those
which might be produced by commercial stamping and bending of stain-
less-steel sheet. From the test results, it was estimated that wide-
spread retrofitting of old existing residential units could yield an
average increase of about 5% in the season-averaged thermal efficl-
encius, and average reduction of jbout 20% in the NO emissions for
X
those units retrofitted. Those estimated achievable improvements are
botli less than half the target gdins of the current research program.
Phase 1 of the residential oil furnace system optimization studies pro-
vided dn essential background to the studies delineated in this report.
It lias been documented in Ref. 3 and is summarized in the following
subsection.
SUMMARY OF PHASE I
The first phase of the research program comprised four distinct tasks:
I. Systems Analysis, in which current designs and practices em-
ployed in residential heating were reviewed and analyzed to
identify potential areas of improvement
-------
2. Convention.! 1 Burner/Combustor Matching Experiments, In which
the 1.0 ml/s (gph) optimum burner was tested in research com-
bustors having a variety of sizes, configurations, and con-
structions to broaden the design optimization
J. Recirculation Burner/Combustor Matching Experiments, in which
3.0 ml/s (gph) burners embodying forced recircuJation of
burned gases (to vitiate the burners' combustion air and lower
flame temperatures) were similarly tested in a variety of re-
search combustion chambers
It. Data Evaluation and Systems Analysis, in which the results of
prior tasks were synthesized to support preliminary conceptual
designs for two prototype, low-emission, improved-efficiency
residential heating units.
The first systems analysis task was concerned primarily with thermal
efficiency. The average steady-state efficiency, based on the fuel's
higher bed ting value, for all existing installed units is probably
between 72 and 75%, while mean season-averaged overall efficiencies
probably are between 60 and 65%. Heat convected up the flue accounts
for over 90% of the inefficiencies. Current residential heating tech-
nology is based on flue gas temperatures being high enough to ensure
an adequate draft in a furnace's firebox and to prevent moisture from
condensing in (and corroding) the furnace or flue. This concept limits
the maximum achievable steady-state efficiency to about 85%; the mini-
mum 155! decrement comprises: approximately 6 to 7% latent heat of
combustion generated moisture, 7 to 8% sensible heat of the flue gases,
and 1/2 to 1% cabinet or casing conduction losses. Tn practice, the
decrement usually exceeds 15% because flue gas temperatures exceed the
minimum to prevent condensation, because excess combustion air is not
minimized and, particularly for hydronic boilers, because casing losses
become greater than the minimum 1/2 to 1% range.
-------
Hent Losses are greater during cyclical furnace operation than during
steady state because heat continues to be conducted through the cabinet
and convected up the flue during standby periods when the burner is not
being fired. Cyclical casing losses for warm-air furnaces may be twice
their steady-state magnitudes. For hydronic boilers, because most
boiler components are at nearly the same temperature during standby as
during firing, cyclical casing losses may be three or more times those
during steady-state operation. Nonetheless, convection of heat up the
flue during standby usually accounds for most of the decrement between
steady-state and cycle-averaged efficiencies. When the burner is
turned off, a natural draft flow of air continues to pass through the
burner, into the firebox, etc., and up the flue. That draft air flow
cools furnace components between firings and can reduce cycle-averaged
efficiencies by as much as 15%, although the average is probably around
8 to 10%. Thus, season-averaged efficiencies are estimated to be about
10 to 15% Lower, on the average, than steady-state efficiencies.
Season-averaged thermal efficiencies of oil-fueled space heating equip-
ment can be increased by: (1) lowering the quantity of excess air
which dilutes the combustion product gases, (2) lowering the tempera-
ture of the gases admitted into the flue, (3) lowering or eliminating
the draft air flow through the combustion equipment during standby, and
(&•) increasing cabinet insulation. Additionally, fuel consumption can
be decreased significantly if: (5) the burner firing rate is properly
matched to the Local design temperature and to the residence's thermal
demand, and (6) outdoor air, rather than heated household air, is sup-
plied to the burner and to the unit's barometric control device.
Item A was considered not to be an appropriate area for study in this
program. Item 5 is not related to the design of the heat source,
per' se, but is related to how a heating unit fits in an overall resi-
dential heating system; therefore, this item was not studied either.
In a similar vein, many design aspects of a residence and its heating
system exert strong influences on thermal demand patterns and overalL
-------
fuel utilization efficiency, but neither influence directly the thermal
efficiency of the heat source nor are controlled by its design. Thus,
items 1, 2, 3, and 6 above were identified as major potential areas for
improving warm-air furnace and hydronic boiler performance.
In the second task, an optimum low-emission burner was laboratory-tested
at a fuel firing rate of 1.05 ml/s (1.00 gph)A in a variety of cylindri-
cal combustion chambers having different diameters and lengths, burner
orientations, and methods and degrees of wall cooling. It was found
that, to reduce NO emissions to a target level of 0.5 g NO/kg fuel at
low excess air levels (10 to 15%), the burner should be fired into a
combustor having the following design attributes:
1. The walls should be cooled so that approximately 20% of the
fuel's higher heating value is extracted from the flame zone.
The combustor wall temperature should be as uniform as pos-
sible during burner firing, and an elevated wall temperature
should be maintained during standby. These conditions were
achieved better with 90 C (194 F) water as the combustor cool-
ant than with warm air, but it was nearly possible to satisfy
them with the latter fluid.
2. The inside diameter of the combustor should be 0.28 m (11
inches) or greater for a 1.05 ml/s (1.00 gph) firing rate.
This parameter influences NO emissions strongly.
X
*Throughout this report, burner firing rates consistently are stated
to two significant figures to the right of the decimal point. They
-------
3. The effective combustion chamber length, from its end near the
burner to the location where the furnace heat exchanger begins
to quench the gas temperature rapidly, should be at least 0.5 m
(20 inches) and perhaps as long as 0.75 m (30 inches). The
shorter length is appropriate for minimum NO emissions, but
X
the longer length may be required to avoid excessive carbon-
aceous pollutant emissions at low excess air levels.
4. The combustor may be either side-fired or tunnel-fired, which-
ever is convenient for a particular furnace or boiler design.
Actually, lower NO emissions are produced by the tunnel-fired
X
configuration, but the other criteria have been stated such
that the more common side-fired configuration can meet the NO
target level.
Combustion gas recirculation (CGR) and flue gas recirculation (FOR) burn-
ers also were assessed in research combustor experiments. The CGR burner
extracted partially cooled gases from the combustion chamber and mixed
them with the combustion air upstream of the burner's air fan. This
burner was found to produce acceptably low NO emissions, but generally
produced unacceptably high CO and UHC emissions. Only very limited sets
of design and operating conditions were found where CO, smoke, NO, and
operability were all acceptable but, even then, UHC concentrations re-
mained high.
The FGR burner's combustion air was mixed with externally circulated
flue gases obtained downstream of the furnace heat exchanger. Being
cooler than combustion chamber gases, flue gas is a more effective
flame-zone diluent. Steady-state, low-excess-air, operating conditions
were found which had acceptably low emissions of all air pollutants
but, when tested in cyclical operation, burner startup spikes of exces-
sive emissions of carbonaceous pollutants were experienced. The ampli-
tude of the spikes wns lowered by reducing the amount of flue gas
10
-------
recirculated; however, NO production increased concurrently so that,
when operation and carbonaceous emissions were acceptable, about 0.6 g
NO/kg fuel was exhausted.
From the experimental results, it was concluded that the optimized con-
ventional burner had better potential for minimizing emissions of air
pollutants and maximizing efficiency than either the CGR or FGR burners.
Therefore, it was incorporated in preliminary designs for candidate
prototype, low-emission, residential heating units. One preliminary
design was developed for each of the two common cooling media, namely,
air and water. The warm-air furnace design was based on making ap-
propriate modifications to an existing warm-air furnace of contempo-
rary design. The hydronic boiler design, on the other hand, involved
all new construction. EAch of these prototype design concepts was dis-
cussed with engineering personnel of a manufacturer of that type of
residential heating equipment. Thereafter, the prototype warm-air
furnace was selected to be built and tested in Phase II. The design is
described in the next section.
11
-------
SECTION IV
EXPERIMENTAL INVESTIGATION OF THE
PROTOTYPE LOU-EMISSION FURNACE
An experimental, I ml/s (gph), prototype, low-emission, oil furnace was
constructed and tested. The objectives were to construct a prototype
system embodying the essential concepts and design features generated in
Phase I and to evaluate its capabilities to satisfy the emission control
and performance design goals. As reported in Ref. 3, a selection was
made to construct a warm air prototype furnace as opposed to a hydronic
type system.
The experimental prototype warm-air furnace was constructed using a
stock commercially available unit as the primary component base, with
modifications to provide the optimized burner and firebox components,
as well as other appropriate design features.
The designs of nonstandard components were such that they were amenable
to fabrication by conventional production techniques, even though those
techniques were not used on this one-of-a-kind unit. The furnace modi-
fications were designed such that further modifications could be in-
corporated as suggested by subsequent test results.
The general overall performance goals for the prototype furnace were:
1. To reduce air pollution emissions to or below the following
levels:
a. Oxides of nitrogen, 0.5 g NO (as NO)/kg fuel burned
b. Carbon monoxide, 1.0 g CO/kg fuel
c. Gaseous hydrocarbons, 0.1 g UHC/kg fuel
d. Smoke, No. 1 on the Bacharach scale
12
-------
In comparison with average emission levels reported from a
field survey of actual residential heating units (Ref. A),
these goals sought not to exceed the average CO and UHC emis-
sions from burners in their as-found conditions while reducing
smoke and NO emissions by 68% and 72%, respectively, from
their reported average levels.
2. To increase cycle-averaged thermal efficiency by 10% or more
above the mean achieved by existing installed residential heat-
ing units
3. To comply with all applicable safety codes and operational
standards
4. To the extent possible:
a. To remain cost competitive with currently manufactured
units
b. To decrease operating noise
c. To minimize unit volume
Emphasis was placed on developing advanced technology that can
be implemented by the U.S. heating industry with minimum de-
partures from current manufacturing, distrubution, operation,
and servicing practices.
Design options were evaluated on the basis of a newly manufactured pro-
duct line, with compromises involving retrofit versatility given a much
lower priority. However, product saleability certainly was of concern
and, therefore, unit costs and acceptability to manufacturers, service-
men, and customers were considered in the design selection process.
13
-------
EXPERIMENTAL APPARATUS
Stock Furnace Prior to Modifications
As indicated above, the experimental, prototype, low-emission furnace
was obtained by making appropriate modifications to a stock commercially
available warm-air furnace. The unit selected to fill this role was n
Lennox Model 011-140 furnace manufactured by Lennox Industries, Mar-
shaLltown, Iowa. Illustrated in Fig. 1, reproduced from a Lennox bro-
chure, the Lennox Oil series is an outstanding example of contemporary
residential oil furnace design. It is more compact than current models
offered by most manufacturers, primarily because of its unique-design
compact heat exchanger, and is capable of achieving quite high steady-
state efficiencies. The main design features of the stock Lennox 011-
140 upflow furnace are summarized in this subsection.
The stock furnace was equipped with a Lennox flame-retention-type burner
having radial slots in its choke plate and a short convergent enclosure
downstream of the choke plate. The recommended firing rate range is 0.89
to 1.05 ml/s (0.85 to 1.00 gph), and it is about the most compact unit
in its heating capacity range, with a 1.45 m high by 0.66 m wide by 0.61
m deep (57 by 26 by 24 inches) cabinet. The firebox is a refractory-
lined, side-fired-type enclosure. The firebox lining is a single piece
of molded refractory fiber material, approximately 0.25 m (9.8 inches)
inside diameter. It has a "corbel" top, with a 0.18 m (7.0 inches) exit
diameter.
The heat exchanger section is approximately 0.33 m (13 inches) high,
0.61 m (24 inches) wide and 0.444 m (17.5 inches) deep. Its primary
heat exchange section consists of an uninsulated 0.279 m (11 inches) di-
ameter central steel cylinder (an extension of the outer shell of the
firebox) with a rearward facing exit that channels the combustion gases
into a rear manifold. From there, the gases are distributed among six,
flat, heat exchanger panels with combustion gases inside moving toward
14
-------
(j_) Burner Assembly
©
®
©
©
©
©
Compact Heat
Exchanger
F i rebox
Chamber)
Warm-Ai r
Warm-Ai r
Furnace
(Combust ion
Blower
Fi Iter
Cabi net
Safety Controls
Figure 1. Cutaway Drawing of the Stock Lennox Model 011-140 Furnace
(Reproduced from a Lennox Industries Inc. brochure with
their permission)
15
-------
the front of the furnace, and coolant air outside flowing vertically
upward. Combustion gases are collected by a front manifold that dis-
charges them into the flue.
The firebox/heat exchanger enclosure utilizes the air-gap method of in-
sulation on three sides, with the warm-air baffling standing off from
the outer wall about 0.00137 m (0.5 inch). The common wall separating
the vestibule and the warm-air channel is insulated with a layer of
fiberglas matting.
Prototype Optimum Furnace
The overall design of the prototype optimum low-emission furnace is
shown in Fig. 2, an assembly layout drawing used in building it. Those
design features which differ substantially in concept from their count-
erparts in the stock Lennox furnace also are illustrated pictorially in
Fig. 3. As described in the following paragraphs, major changes were
made concerning the oil burner, the firebox, and the combustion air sup-
ply, while minor modifications were made in several other components.
Otherwise, the stock furnace external cabinet, warm-air blower and fil-
ter, compact heat exchanger, and all electrical circuits and controls
were retained without change.
Optimum Burner Unit - The oil burner utilized in the prototype furnace
design was a Beckett Model AF burner body fitted with an optimized non-
flame-retention burner head and a device to eliminate draft air heat
losses during standby.
The optimized burner head consisted of six air swirl vanes, canted 25
degrees from the blast tube centerline, and a firing-rate-dependent
choke diameter. The air swirl vanes were relatively large; they were
approximately 0.05 m (2 inches) long, and extended from approximately
0.03 m (1.2 inches) diameter out to the diameter of the blast tube.
16
-------
SECTION A -A
Figure 2. Layout Assembly Drawing, Prototype Optimum Warm-Air Oil Furnace
-------
00
WARM-AIR
OUTLET
SEALED COMBUSTION AIR SYSTEM
STANDBY DRAFT CONTROL
HEAT
EXCHANGER
BURNER
VESTIBULE
SEALED
COMBUSTION AIR FIL
QUIET PULSE FREE STATOR
AIR
BLOWER
OPTIMUM BURNER HEAD
WARM AIR INLET
AIR COOLED FINNED FIREBOX
Figure 3. Prototype Optimum Warm-Air Residential Oil Furnace
-------
The choke diameter was related to the specific installed firing rate
according to:
D ' Kc <*>'
-------
Quiet-Stator Lip
Extens ion
Skewed Leading
Edge
Original Parallel
Large-Gap Stator
Air Out
Air In
Figure 4. Schematic of the Skewed-Lip "Quiet Stator"
Extension for the Optimum Burner Combustion
Air Fan
20
-------
casings are made with a relatively large casing-to-fan gap is to avoid
high-frequency (~500 to 1000 Hz) burner noise problems resulting from
fan tip/casing interaction. To eliminate the production of discrete
pressure pulses (i.e., noise), the edge of the added lip extension was
oriented to interact with more than one blade at a time in a continu-
ous manner. In addition to increasing the burner fan output and making
the burner quieter (no tip whine), the pulse-free flow of air reduces
the likelihood of combustion instability being externally induced by
the air feed system, which would in turn result in combustion noise and
off-optimum (i.e., higher emissions) combustion.
Finned, Air-Cooled Firebox - The firebox for the prototype furnace was
a rather massive uninsulated steel assembly with external fins to in-
crease the effectiveness of the external warm-air flow in cooling the
component. The prototype firebox design is illustrated in Fig. 5. It
was constructed from a standard Schedule 40, 12 inch (0.305 m) diameter
pipe cap welded to a 12 to 10 inch (0.305 to 0.254 m) diameter eccen-
tric reducer, with twenty-four 0.0064 m (0.25 inch) thick fins welded
to the outside. The outer surface area was increased to approximately
six and one half times that of the unfinned shell. The eccentric re-
ducer was used, rather than a concentric one, so that the larger-diam-
eter firebox was shifted toward the rear of the furnace and did not
encroach upon the depth of the burner vestible. The rear-biased eccen-
tricity reduced the external cross-sectional area for coolant air flow
at the back of the firebox which, because the burner is directed toward
it, was expected to be the hottest protion of the firebox. To compen-
sate for that effect, extra fins were placed in that area.
The assembled prototype firebox mass was 57.2 kg (126 Ibm), an increase
of about 50 kg (110 Ibm) over the refractory-fiber-lined stock firebox
which is replaced. This massive construction was adopted intentionally
to accomplish two objectives in addition to extracting heat from the
flame zone; both are related to the heat sink nature of a massive com-
bustion chamber. First, a massive heat-sink chamber can more readily
21
-------
Heat Exchanger/Conbustor
Retainer Ring
5ZZ31-10/21/75-SID
Figure 5. Front View Photograph of the Finned, Air-Cooled Combustor
for the Low Emission/High Efficiency Prototype Furnace
System
22
-------
approach uniform inside surface temperatures than can a lightweight, un-
lined, metal firebox. This is important for avoiding excessive NO form-
ation and wall erosion (or possibly burnout) at "hot spots," and for
avoiding smoke and UHC formation at "cold spots." The second objective
is for the combustion chamber to retain most of its stored heat during
standby periods so that the firebox is considerably warmer at burner
startup than is the usual warm-air furnace practice. This objective
arose from the observation that start-transient spikes in the emissions
of CO, smoke, and UHC were considerably lower with water-cooled than
with uninsulated, air-cooled combustors (Ref. 3).
The mating of the heat exchanger to the prototype firebox was accomp-
lished by a bolted retainer ring compressing a pyroflex gasket between
the finned firebox and the 17-gage heat exchanger shell. This type of
firebox/heat exchanger attachment was selected for the experimental pro-
totype furnace to avoid problems of metal fracturing due to differential
thermal expansion between the two very different component thicknesses.
Combustion Air Supply - Two related modifications were concerned with
the combustion air supply. The first was the use of a sealed air sup-
ply system and the second was the provision of a separate filter for the
combustion air.
In the sealed air supply system, outdoor air is piped into the room
where the furnace is situated, and is supplied both to the burner for
combustion air and to the barometric control device for admixture with
flue gases. The main advantage of doing this is that residential fuel
consumption is reduced because heated (and perhaps humidified) air from
within the residence is not consumed by the furnace, so that the resi-
dence's thermal demand is lowered. Experimental data reported in Ref.
5, from which the system was adapted, showed that sealed air systems
can reduce fuel consumption by at least 5% and, in some installations,
by as much as 15%. An insulated galvanized air duct conducts outdoor
air to an air plenum constructed around the barometric control damper
23
-------
connection with the furnace's flue pipe.* The plenum serves as an
supply manifold where the air flow is divided, as required, between the
barometric flue pressure control device and the oil burner.
As described in Ref. 5, the combustion air was ducted from the plenum
directly to the burner inlet through a length of uninsulated flexible
plastic tubing. The installation for the prototype furnace was differ-
ent, in that the combustion air was ducted to a connection on the fur-
nace, rather than directly to the burner. This was done to facilitate
filtering the combustion air. A second air plenum was constructed with-
in the furnace's burner vestibule, and formed the supply side of a flat,
rectangular, fiberglas filter pane]. The burner vestibule was also
sealed so that the only source of combustion air was that which passed
through the filter.
Combustion air filtration was considered to be especially important for
the prototype optimum furnace because it was designed to operate with
close to minimum (10 to 15%) excess combustion air. Accumulations of
dust, lint, hair, etc., in the burner air passages may shift the opera-
ting point into a smoky and/or high CO condition more easily when the
burner is tuned for low excess air operation than conventional burners
operating with high excess air.
The combustion air filter was positioned so that, upon entering the
burner vestibule, air passed directly over the furnace electrical con-
trols, then completely across the vestibule to the burner's air entry.
This arrangement was meant to promote cooling of the electrical con-
trols, burner components, and furnace cabinetry, all of which were an-
ticipated as having potential overheating problems because the combus-
tion air flow was lower than the air flow through the stock furnace's
*The air supply duct was omitted from the installation of the prototype
furnace in Rocketdyne's outdoor test laboratory. The sealed air
plenum pictured in Fig. 3 was a prototype unit supplied by Lennox with
the stock furnace.
-------
vestibule before it was sealed. In addition to better component cool-
ing, this vestibule flow pattern was also expected to help temper the
outdoor air and at least partially offset a tendency for cold combus-
tion air to increase emissions of carbonaceous pollutants.
The enclosed burner vestibule, together with the rest of the sealed air
system,also should beneficially cut down on burner motor and fan noise,
combustion noise, and odors emitted into the furnace room and, presum-
ably, into other parts of the residence.
Other Modifications - Minor modifications were made to a few other fur-
nace components. The burner vestibule was sealed by covering louvers
and handholes in the vestibule closure panels. Some structural rein-
forcement was added inside the cabinet so that the extra weight of the
finned firebox could be supported evenly by the cabinet walls. Finally,
two baffles were installed, one on either side of the finned firebox,
to prevent a substantial fraction of the coolant air from bypassing the
finned firebox. The baffles are clearly visible in Fig. 6, which shows
two views of the firebox in the prototype optimum furnace during its
initial assembly.
Test Facility and Instrumentation
Performance of the prototype optimum furnace was evaluated in an out-
door laboratory facility having provisions for measurement of pollutant
emissions, operational characteristics, and thermal efficiency. Figure
7 is a schematic of the furnace evaluation system; it shows the instal-
lation of gas and air flow ducting and a variety of instrumentation.
Basic thermal performance measurement techniques conformed with re-
quirements of ANSI Z91.1-1972 (Ref. 6). Other instrumentation was
added to provide enlarged understanding of furnace behavior and data
for calculating cycle-averaged thermal efficiency.
25
-------
FLUE
CONNECTION
to
HEAT
EXCHANGER
BAFFLES
FINNED
FIREBOX
BURNER
PORT
WARM-AIR
INLET
OBSERVATION
PORT
•
SEALED
VESTIBULE
DOOR
Installation of the Finned Firebox and Compact
Heat Exchanger in the Prototype Optimum Furnace
-------
Sealed Air
System
3x3 Thermocouple
Matrix
Insulation
Front View
Adjustable
Louvers /
k/
Combustion and
Barometric Control
Air Inlet
Air
<=>
Inlet
Thermocouple, Free Air
•« Thermocouple, Attached
• Pressure Tap
Matrix of
6 Thermocouples
Flow Straightener
0.45m x 0.45m
Side Vjew
Figure 7. Schematic of the Furnace Performance Evaluation System
-------
Constituents in the flue gases were measured by continuously withdraw-
ing a gas sample from the center of the flue, at the location denoted
in Fig. 7, and passing it through an analysis train. The analytical
system provided for continuous analyses for CL, CCL, CO, NO, and UHC
species remaining in the dry gases after their passage through condens-
ible traps, filters, and dryers. Details on the setup and operation of
the train, instruments used and their ranges, data processing, etc.,
are given in Appendix A.
The furnace flue thermal losses were determined by making measurements
to support flue gas heat balances. Combustion gas mass flowrate was
back-calculated from measured fuel flowrate and stoichiometric ratio
(as determined from flue gas composition measurments). The flue gas
exhaust temperature was measured in an insulated flue pipe with an iron/
constantan thermocouple located 0.46 m (18 inches) above the centerline
of the heat exchanger exit. Flue draft, gas composition, and smoke
measurements were taken at successive 0.0317 m (1.25 inches) increments
downstream of the thermocouple, respectively.
Steady-state thermal efficiencies were derived from steady-state flue
gas temperature and C0~ concentration data according to a table of
values given in Kef. 6. The tabulated relationships are plotted in
Fig. 8 as a family of curves. During cyclical operation in which
steady-state was not reached, values for those parameters just prior to
burner cutoff were used in the same manner to get approximations of
steady-state efficiencies. Burner firing times of 10 minutes gave such
pseudo-steady-state efficiencies which were indistinguishable from
those derived from steady-state measurements; those calculated from 4-
minute burner firing time data were approximately 1/2 to 1% higher than
the steady-state efficiencies.
Determination of furnace thermal performance during cyclical operation
is more difficult than during steady-state operation. To avoid the
complications of measuring or estimating transient draft air and
28
-------
35
Nl
vo
in
in
U
a)
30
25
20
8 '5
O
10
ANSI 291.1-1972 (Ref. 6)
_ 1.0 1.1 1.2 1.3 I.* 1.5 1.6
Stoichiometric Ratio
15
1.7 1.8
III
13 12 11 10
Volume Percent CO,
65
70
75
80
u
*-»
U 0)
u- O
in —i
>. W
xi c
ID —
0) in
4J 10
in u
-o *e
-------
furnace cabinet heat losses, the method used to determine cycle-
averaged efficiency was to divide the warm-air furnace coolant net heat
gain by the gross heat input of the fuel burned in a cycle. This re-
quired measurements of oil flowrate, oil and combustion air temperatures
and, for the warm-air furnace coolant, flowrate and temperatures at the
inlet and outlet. Provisions were made for measuring all of those
parameters.
The inlet warm air was drawn into the furnace from the ambient outdoor
atmosphere through a 0.46 m (18 inches) square duct with an inlet flair
and internal "egg-crate" flow straightener. The volumetric air flow
was measured with a cumulative readout, gas-flow anemometer, i.e., it
integrated the total air flow admitted during each complete cycle. Am-
bient atmospheric pressure, temperature, and relative humidity were re-
corded continuously at a meteorological data station located approxi-
mately 15 meters from the furnace test stand. Furnace coolant air
temperatures were measured at the inlet aneometer location with a mer-
cury therometer and at the warm-air outlet as an average reading from
nine, ice-referenced, chromel/alumel thermocouples in a rectangular
grid array. In addition, an array of six more of the same type of
thermocouples (shielded from radiation) was installed in the warm-air
passageway between the firebox and the secondary heat exchanger. This
provided a measurement of the average air temperature between the fire-
box and heat exchanger, from which the heat being removed from the
finned firebox could be calculated.
A variety of combustion, heat transfer, and thermal cycling parameters
were also recorded. Six chromel/alumel thermocouples were attached to
the combustor and heat exchanger sections to monitor peak metal temper-
atures for an estimation of thermal stresses and service life. A dial
thermometer was used to measure temperatures in the sealed vestibule
area to determine the likelihood of falling outside of code specifica-
tions or of accelerated component degradation from anticipated higher
temperatures in the sealed compartment. Manometer taps in the
30
-------
combustor section and in the heat exchanger outlet manifold were used
to measure "over fire" draft and heat exchanger pressure drop. Calcu-
lated adiabatic flame conditions and the warm-air temperature rise
through the finned combustor section were used to estimate the combus-
tion gas temperature at the exit of the combustor.
Relative pressure measurements were taken at the warm-air blower outlet,
at a point between the combustor and heat exchanger, and at the warm-air
outlet. The outlet back pressure could be varied through a set of ad-
justable outlet louvers to simulate various installed ducting resist-
ances. The warm-air flowrate through the prototype furnace also could
be varied independently through use of a 3-speed fan control and a
variable-spacing (effective diameter) drive pulley.
The electrical consumption of the components could be measured indi-
vidually by a 1000-watt, alternating-current wattmeter, while the com-
ponents were operating at their respective design conditions. The
measured electrical power consumption then could be added to the fuel's
gross energy input to obtain total energy consumption figures.
31
-------
TEST RESULTS
The methodology used in the experimental testing was to proceed, more or
less sequentially, through the test matrix presented in Table 1 with con-
siderations for flexibility as problems or promising discoveries arose.
Basically, the test matrix was intended: (1) to establish baseline emis-
sions and thermal performance characteristics of the stock furnace prior
to its conversion to a prototype unit, and (2) to optimize the operation
of the prototype optimum furnace. Complete tabulations of the data ob-
tained are contained in Appendices B and C. Data were recorded for 401
runs. Due to alternating laboratory effort with another related study
(Ref. 2) and elimination of some checkout tests, there are some discon-
tinuities in the sequence of run numbers.
Stock Lennox Furnace
The Lennox Model 011-140 furnace, described earlier, was installed in
the laboratory test facility (Fig. 7), and tests were conducted to mea-
sure its air pollutant emissions and thermal efficiency characteristics.
Air Pollutant Emissions - The Lennox furnace was fired in its stock con-
figuration with the factory-supplied 0.85-70°-A oil nozzle*. Cycle-
nvernged pollutant emission data are given in Appendix B, Table B-l.
The stock furnace operated at a nominal stoichiometric ratio of —1.60,
which is fairly typical of new oil heating equipment presently being
produced. The cycle-averaged Bacharach smoke readings were No. 1 or
greater due to high readings obtained immediately after burner start.
The recovery to zero smoke on most of the runs occurred quickly, within
-Oil nozzle callouts designate the nominal fuel flowrate, the fuel
spray cone angle, and the general type of spray configuration. For
example, this "0.85-70°-A" callout denotes a firing rate of 0.85 gph
(0.89 m]/s) and a hollow-cone-type spray ("A") with a 70° cone angle.
Spray nozzles which produce solid-cone sprays carry a "B" designation.
32
-------
TABLE 1. FURNACE SYSTEM EVALUATION TEST MATRIX
Furnace
Unit
Stock
(Lennox)
Opt imized
Prototype
Fi r \rr}
Rate,
ml/s
0.89
i
f
1.05
,
1 0.89
j
1
Opt imum
Warm-Ai r
F 1 ow ,
m3/s
Manufacturer 's
Recommended
Value
\
0.708
0.637
0.566
Opt imum
\
0.
\
566
Opt imum
Stoichiometr ic
Ratio Range
1.40-1.60
i
1.10-1.20
f
Opt imum
•
i
1.10-1.20
i
i
Opt imum
Cycle Time
On/Off, minutes
Steady-State
4/8
8/4
10/20
Steady-State
4/8
8/4
10/20
Steady-State
4/C
Steady-State
4/8
8/4
10/20
Steady-State
4/8
8/4
10/20
Long-Term
Cycl ic
Operat ion
33
-------
the first half minute, and these runs are flagged by asterisks in
Table B-l. This was considered to be acceptable operation due to the
quick recovery and also because the averaging method gave excessive
weight to the start conditions. Taken once every minute, each smoke
reading was assumed to be a representative average for a full minute.
The initial reading, taken during the short high-smoke interval, was not
an accurate representation of the first minute average, and the magni-
tude of the error was emphazied by the short (4 minutes) burner firing
times.
Carbon monoxide and unburned hydrocarbon emissions were acceptably Inw
at all operating conditions tested. Nitric oxide emissions, plotted in
Fig. 9, generally exceeded 2.0 g NO/kg fuel burned*; they were higher
than anticipated for the 0.89 ml/s firing rate, based on experience wjth
other residential burners (Ref. 1). The flue gas temperatures were quite
low, on the order of 180 C (-360 F), consistent with the 0.89 ml/s
(0.85 gph) firing rate being the minimum recommended for this furnace.
The stork furnace was then fired with a 1.00-70°-A (1.05 ml/s) oil noz-
zle. It appeared to light-off better at this higher firing rate, seen
as dn improvement in the cycle-averaged smoke emissions, resulting in a
lower oper.iting stoichiometric- ratio range (~1 .40-1.50) than the 0.89
ml/s firing rate. The other pollutant concentrations remained essenti-
ally unchanged with the nitric oxide (NO) level remaining nominally
about 2.5 g NO/kg of fuel burned.
*0xides of nitrogen emissions (NO ) from residential heating systems
have been observed (Ref. 4 and 7; as comprising NO (nitric oxide)
and NO. (nitrogen dioxide) in volumetric proportions of about 9:1 to
10:1. Thus, on the average, NO accounts for more than 90 mole per-
cent of NOX emissions. For that reason, NO measurements were used
throughout this Investigation as a quantitative indicator of NO
emissions, and NO- emissions were not measured.
-------
3.0 _
« 2.5
3
en
^.
o
i/i
c
o
I/I
I/I
1
^Operation at SR's to the right of
marker, high smoke reading at start
only, recovery to zero within 30
seconds.
I
1.5
Stolchlometric Ratio
2.0
Figure 9. Cycle-Averaged Nitric Oxide Emissions from the
Lennox 011-140 Furnace in its Stock Configuration
35
-------
Thermal Efficiency - Cross thermal efficiencies (based on the fuel's
higher heating value) were derived by two methods. The first, referred
to as the warm-air method, involved calculation of the total heat trans-
ferred from the combustion products to the warm air, and was based upon
measured data for the warm air mass flowrate and temperature rise, and
the total fuel input. In cyclical tests, because furnace components are
often partially heated when the air is not flowing and partially cooled
when the burner is not being fired, these parameters were measured as
functions of time and averaged appropriately over the firing cvcle to
obt.iin cycle-averaged efficiencies. This method included all opera-
tional thermal losses (rasing, standby draft flow, flue gas, etc.).
The second method, referred to as the flue gas method, was adapted from
the method recommended in Ref. 6 for determining oil furnace efficien-
cies. The flue gas CO concentration and temperature were measured, and
an efficiency decrement corresponding to their values was read from
Fig. 8. Subtracting that decrement and a modest casing loss (2% or
less) for conduction and radiation to the surrounding from 100% gave the
estimated gross thermal efficiency. The ANSI method is for application
to steady-state; our adaptation was to assume that the last minute of
burner firing during cyclical operation is a good approximation of
steady state.
Estimated steady-state (flue gas method) and cycle-averaged (warm-air
method) efficiencies for the stock Lennox furnace are tabulated in Ap-
pendix B, Table B-2. Results of testing at 1.05 ml/s (1.00 gph) firing
rate (Runs 56 to 52) are plotted in Fig. 10 for both methods. Linear
legist-squares correlating lines fit to the data points indicate that
cycle-averaged efficiencies were approximately 5 to f>% lower than those
for steady state at comparable stoichiometric conditions. Nearly com-
patible 5% differences between steady-state and cyclical efficiencies
wi-re measured for two other worm-air furnaces tested earlier in this
36
-------
3U
Efficiency, %
g
.c
»—
« 70
in
60
iii i| ii 1111
\>x
^V
•Xv,
^v
>>x(t N, Steady-State
_ x>. Flue Gas Method
• \* (4th Minute)
* Warm A1r Method
(Cycle-averaqed)
™ ~
^^— Cycle-Averaged Smoke < 1
— — — Cycle-Averaged Smoke >1
i i i i 1 i i i i 1 i
1.0
1.5
Stoichiometric Ratio
2.0
Figure 10. Thermal Efficiency Characteristics of a Stock
Lennox 011-140 Warm-Air Oil Furnace Fired at
1.05 ml/s in 4-Min-On/S-Min-Off Cycles
37
-------
facility (Ref. 2). Those units were fired with longer 30-minute (10
minutes on/20 minutes off) cycles than the 12-minute cycles (4 minutes
on/8 minutes off) fired in the stock Lennox.
The data in Fig. 10 exhibit a characteristic observed in all'cyclical
furnace efficiency measurements made in this test facility, viz., the
scatter is very low in steady-state efficiencies obtained by the flue
gas method, and is quite large in cycle-averaged efficiencies obtained
by the warm-air method. The testing was delayed, and a substantial ef-
fort was expended toward understanding and eliminating causes for the
scatter. It was determined that much of the scatter resulted from test-
ing outdoors and could not be eliminated (see the Discussion section).
As a result, greater reliance was placed upon the steady-state effi-
ciency values for comparing one burner or one furnace with another while
recognizing that there was a larger undertainty in cycle-averaged effi-
ciencies for situations requiring comparison through them.
It is seen in Fig. 10 that the stock Lennox furnace can be tuned for
normal operation with 45 to 50% excess air where it achieves about 82 to
83% steady-state efficiency. This is well above the industry minimum
performance standard of 75% and very close to the maximum practical
limit of 85% (noncondensing flue gases). The compact heat exchanger ap-
pears to be very well designed and closely matched to the unit's burner
and firebox. Net flue gas temperatures are as low as should be designed
for, so the only way that steady-state efficiency could be improved is
by lowering the burner's excess air requirement. A potential gain of
only about 2 percentage points might be realized.
Optimum Burner in the Stock Furnace - The I ml/s (gph) optimum burner
unit was installed and fired in the stock Lennox furnace. Test results
(Runs 74-87, Tables B-] and B-2) indicated that this burner, as com-
pared with the stock Lennox burner, might offer modest improvements in
38
-------
both NO emissions and thermal efficiency: Measured NO emissions were
roduccd by about 5 to 7%, and smoke-free operation was possible with Less
than 27% excess air.
Prototype Optimum Furnace
Initial Shakedown Tests - A series of tests was made with the prototype
optimum furnace system, as designed, operating at its nominal 1.05 ml/s
(1.00 gph) firing rate conditions (Runs 149-154, Table B-3, Appendix B).
Operationally, the system behaved reasonably well for its first test
series. There were, however, some surprising results from the data
analysis. The burner was not able to operate smoke-free at excess air
levels below about 20%. Flue gas NO concentrations were 60 to 80%
higher than the target value of 0.50 g/kg, even though the temperature
measurements indicated that nearly 33% of the fuel's higher heating
value was extracted from the finned combustion chamber (versus the de-
sign target of 20%). Additionally, thermal efficiency was lower than
expected.
To investigate the cause of the higher excess air operational require-
ments and the higher NO emissions, the optimum burner was temporarily
removed and replaced, successively, by two different flame-retention
burners. The first was a Beckett Model AF burner, which was the stock
burner for a Williamson Model ]167-15 furnace tested extensively in the
Ref. 2 studies (and modified later in the current investigation). It
was fired in the prototype furnace in Runs 155 to 162 (Tables B-3 and
B-4). The second, the stock Lennox burner supplied with the stock
Lennox Model 011-140 furnace, was tested in Runs 163-170 (Tables B-3
and B-4). Results obtained with these burners are described in the fol-
lowing subsections together with results from using the optimum burner.
Pollutant Emissions - Cycle-averaged emissions data concerning flue
gas NO and the transition through No. 1 smoke are plotted in Fig. II for
39
-------
Q>
3
s
o
01
•o
X
o
0)
01
(D
l_
0)
3.0
2.5
2.0
.5
1.0
o 0.5
Cycle-Averaged Smoke <1
— — Cycle-Averaged Smoke^
1.0
/
/
Prototype Furnace/
Wi11iamson Burner
Stock Lennox
Prototype Furnace/
Lennox Burner
Prototype Furnace System
(Initial Configuration)
I
1.5
Stoichiometric Ratio
2.0
Figure 11. Comparison of Cycle-Averaged Nitric Oxide Emissions of
Various Oil Burners in the Prototype Furnace and in
Their Respective Furnaces
-------
the prototype optimum furnace tested with all three burners mentioned
above. Additionally, comparable data are shown for the two stock fur-
naces which were equipped originally with the flame-retention burners.
Comparison of the curves in Fig. 11 shows that: (1) NO emissions pro-
duced by the optimum furnace with the flame-retention burners were about
]5 and 25% lower, respectively, than those from the stock Williamson and
Lennox furnaces; (2) flame-retention burners produced approximately
twice as much NO when fired in the prototype furnace as did the optimum
burner; and (3) firing the flame-retention burners in the finned air-
cooled combustor, rather than in their stock furnaces' refractory-lined
combustors, did not change appreciably the excess air level at which
transition to No. 1 smoke occurred. Furthermore, there apparently was
no degradation of other operational characteristics (e.g., light-off,
transition to steady state, combustion noise) of the flame-retention
burners. The optimum burner, on the other hand, exhibited appreciable
low-frequency (rumbling) noise at the lower excess air levels tested.
It was thought that the operational behavior of the optimum burner
might have been degraded by extraction of more heat from the flame zone
than had been intended. Therefore, in an attempt to decrease combustion
zone heat extraction and increase combustor wall temperatures, the
3 3
warm-air flow was reduced from 0.57 m /s (1200 cfm) to 0.46 m /s (950
cfm) for a series of tests (Runs 171-176, Table B-3). Rather than
helping to broaden the operational envelope for the prototype furnace,
this change seemed to emphasize the tendency toward noisy combustion.
At the 22% excess air level, no smoke was formed but operation was mar-
ginal due to intermittent roughness; smooth operation was not attainable
at any lower excess air levels.
Thermal Efficiency - Steady-state thermal efficiencies for the pro-
totype furnace fired with the optimum burner and with the two flame-
retention burners from the stock Lennox and Williamson furnaces are tab-
ulated in Table B-4 and are plotted in Fig. 12. The steady-state effi-
ciency curve for the stock Lennox furnace (Fig. 10) also is reproduced
41
-------
I
OJ
ro
O
O
Ol
4-3
to
-t-J
I/I
I
0)
4->
uir Burner
—1-4 4 •
Optimum ,
0.45 m7s Harm-a|-fr)
I 1—• r
1.6 1.8
Stoichiometric Ratio
Figure 12. Comparison of Thermal Efficiency Characteristics of the
Prototype Optimum Furnace With Various Burners and System
Changes at a Nominal Warm-air Flowrate of 0.566 m-Vs
on this graph. It was noted above that about 33% of the fuel's higher
heating value was transferred to the warm-air furnace coolant through
the finned firebox walls. In spite of this substantial supplemental
heat transfer, the optimum furnace's steady-state efficiency was 4 to 5
percentage points lower than that of the stock Lennox furnace when they
were fired with the same Lennox burner and at equivalent stoichiometric
ratios. Its efficiency with the optimum burner at comparable operating
42
-------
conditions was even lower by an additional 1 percent, and was degraded
further by 1 percent or more when combustor cooling was reduced at the
lower warm-air furnace coolant flowrate.
The observed drop in thermal efficiency when the stock furnace was con-
verted to the prototype furnace was suspected initially to be caused
either by incomplete combustion in the firebox, resulting in some de-
layed heat release in the heat exchanger, or by the optimum burner some-
how producing very nonuniform, thermally striated flow through the heat
exchanger. If these phenomena existed, they should have been substanti-
ally eliminated by the flame-retention burners, since such burners usu-
ally mix the incoming reactants more vigorously and burn them more
quickly than do conventional burners. In other words, one or both re-
tention head burners should have restored most of any efficiency drop
caused by slow combustion. The fact that they did not do so was inter-
preted as meaning that the efficiency problem was probably caused by de-
sign changes in the warm-air side, rather than the combustion gas side,
of the furnace's heat transfer process.
Testing With Modified Components - It was evident from the results of
the shakedown tests described above that the emissions and performance
goals could not be met by the prototype furnace without some refinement
of its initial design. Achieving the emissions goals obviously would
require some improvements in the combustion equipment design. Achiev-
ing the efficiency goal, on the other hand, required improvements in the
heat transfer equipment, predominantly in the warm-air circuit. As dis-
cussed later in the description of the final design, it was believed
that heat absorption by the warm air could be corrected by redesigning
the baffles in its passages and, if necessary, could be supplemented
easily by increasing the secondary heat transfer area. As a result, the
subsequent testing was devoted almost exclusively to what was considered
to be the more difficult task of improving the combustion circuit to
achieve the emissions goals.
-------
The approach t.iken was to investigate the effects on acceptable operating
conditions and pollutant emissions of specific design differences between
the prototype furnace and the research combustor apparatus (Ref. 3) from
which the design criteria were derived. Because the optimum burner used
in the two systems was the same, its design was not varied during a
large portion of. the subsequent testing. Instead, attention was focused
on the finned air-cooled firebox, the transition between it and the main
heat exchanger, and the heat exchanger design and cooling media. The
results of testing configurational variations in these components were
very informative, but it appeared that applying them to achieve the
emissions goals would make the furnace appreciably more complicated and
expensive to manufacture. Attention was redirected, therefore, to de-
termine whether some further refinement of the burner design might be
beneficial. Experiments in that direction were performed with the pro-
totype furnace in its initial design configuration; for that reason, the
results are tabulated in Appendix B along with those from the initial
shakedown tests. Those results are described next, before describing
results from the intervening furnace modification tests.
Refined Burner Design Tests - The optimum low-emission burner was modi-
fied, as described on page 19, to reduce its potential for coupling with
and amplifying oscillatory or noisy combustion. A quiet stator plate
was installed on the discharge side of the combustion air fan, and a
large-diameter static disc was installed inside the blast tube. Addi-
tionally, some tests were run to determine the optimum oil spray angle.
The modified burner was tested in Runs 469 to 481 and 512 to 533 (some
of which also had an internal baffle added to the firebox or heat ex-
changer). Emissions and performance data are tabulated in Appendix B,
Tables B-7 and B-8, respectively.
With the 90-degree oil spray angle, the burner design changes dctu.illy
were detrimental; at comparable excess air conditions, flue gas temper.i-
ture and smoke
-------
Runs 452-457). With the smaller 60- and 70-degree spray angles, however,
these conditions were all improved, and the furnace could be operated
consistently at design goal conditions of only 15% excess air. Emis-
sions of NO were slightly lower with the 1.0-70°-A nozzle, so it was des-
ignated as the preferred nozzle for the refined design burner.
The NO emissions achieved by this refined optimum burner in the proto-
type furnace came close to but remained above the 0.5 g NO/kg goal. A
linear least-squares correlation of the data from Runs 469-477 ran from
0.51 g NO/kg at a smoky stoichiometric ratio (S.R. = 1.00)to 0.78 g/kg
at S.R. = 1.35; tuned to S.R. = 1.15 as a nominal design point, 0.63 g
NO/kg fuel burned would be produced.
Thermal efficiencies in that series of tests were slightly higher than
those measured in the initial shakedown tests. Referring to Fig. 12,
they were about midway between the lines for the Lennox and Williamson
burners, i.e., about 1% higher than with the optimum burner.
A series of four tests (Runs 478-481) was made in which the warm-air
blower was cut off immediately after burner cutoff to see if this would
further reduce the carbonaceous emissions. Although the CO emissions
were almost unchanged, both the UHC and smoke levels were increased.
Therefore, this change in furnace control method is not recommended.
In view of the improved operational capability and lower NO emissions
achieved by refining the burner design, an effort was made to further
reduce combustion noise by adding sound-absorbing material inside the
firebox and heat exchanger. A 0.01 m (3/8 inch) thick layer of pyro-
flex refractory fiber material was bonded to the dished bottom of the
finned firebox and another was bonded to the flat top of the central
dome portion of the heat exchanger. Test results (Runs 482-485, Table
B-7) showed that NO production was increased and that smoke levels were
unacceptably high over the entire stoichiometric range of interest.
45
-------
Obviously, the layer of insulation on the bottom of the firebox had al-
tered the flame zone recirculation and heat extraction characteristics
drastically, so it was removed for the next series of tests (Runs 486-
490). This restored the NO emissions to the former level. However, at
comparable excess air levels this configuration produced more smoke and
UHC than without any pyroflex lining.
Long-Duration Test - A simulated service test was conducted by run-
ning the prototype optimum furnace for a period exceeding 500 hours of
cyclical operation. The furnace was adjusted initially to operate at
approximately 15% excess air (design point), and set to fire automati-
cally on a 4-minute-on/8-minute-off duty cycle. Reduction of initial
performance data showed the actual excess air to be 17% (Run 529, Table
B-7). The furnace was left unattended for the entire duration of the
test with only general, external visual inspections made periodically.
At the end of the 500-hour test, inspection of the furnace and burner
revealed no signs of deterioration in any components. The excess air
level had not changed appreciably from the initial 17% excess air
setting.
Cycle-averaged flue gas concentrations of pollutant species were mea-
sured at the beginning and termination of the 500-hour test. There were
slight increases in i.he emission levels of carbonaceous pollutants:
smoke from 0 to an estimated 0.1", UHC from 0.035 to 0.041 g/kg fuel,
and CO from 0.35 to 0.54 g/kg fuel. Together with a slight increase in
flue gas CO^ concentration, these data indicate that the excess air
level probably decreased slightly, in contrast to the indicated con-
stancy from the measured C0_, 0,,, etc., data. The measured flue gas
concentration of NO corresponded to 0.64 g NO/kg fuel burned, agreeing
*The Bacharach smoke readings after 500 hours of service showed a very
faint, almost imperceptible shade of gray that was constant through
the firing period (not a start spike). As faint as this reading was,
the filter was not snow white as it was in the initial readings;
therefore, a nominal cycle-averaged value of 0.1 was assigned to the
smoke reading.
-------
precisely with the least-squares correlation cited earlier. There was
less than l/.i difference between the values measured Jt the beginning and
dt the end of the 500-hour test.
Exploratory Modifications of the Prototype
Optimum Furnace Configuration
Following the shakedown tests of the as-designed prototype optimum fur-
nace, in which it was found not to operate satisfactorily with less than
approximately 20% excess air and to produce substantially more than the
target NO emissions, a number of exploratory modifications of the fur-
nace configuration were investigated.
A close visual examination of the internal geometry of the firebox and
heat exchanger revealed that the large rectangular entrance to the rear
manifold was very close to the firebox. It appeared that the combustion
gas flow might enter the heat exchanger too quickly, thereby producing
smoke from premature quenching of combustion reactions. Therefore, a
large baffle was installed above the firebox that blocked half of the
central vertical cylinder, attached at the bottom of the rectangular
entrance to the heat exchanger and canted forward at a 45-degree angle.
The results (Runs 198-202, Table B-5) showed no significant improvement
in the pollutant emissions characteristics.
When that proved not to be beneficial, the optimum furnace's combustion
system was compared critically with the research comhustor apparatus
which had been used in the experimental derivation of the design cri-
teria forming the basis for the optimum furnace design (Ref. 3). De-
sign differences were identified which conceivably may have caused the
observed differences in their achievable operating conditions and emis-
sions. Furnace modifications made to assess the validity of hypothe-
sized causes and of potential solutions were in the general areas of
the firehox and the main heat exchanger.
-------
Firebox Design Differences - One hypothesis for explaining the higher
than expected NO results (0.85-0.90 g/kg) was that the eccentric: reduc-
tion in diameter (where the combustion chamber is joined to the air-
cooled heat exchanger) may have promoted greater recirculation in the
combustor and produced a significant increase in mean gas residence
time. To test that hypothesis, an available Williamson Model 1167-15
furnace was modified (see Fig. 13) by replacing its combustor with the
smaller-diameter (0.254 m inside diameter), straight-cylindrical, pre-
prototype, finned, air-cooled combustor tested earlier as a research
combustion chamber (Ref. 3, Appendix F). That chamber was shortened to
fit within the Williamson furnace cabinet, and was welded to the William-
son air-cooled heat exchanger. This modified furnace is referred to as
the "finned-romhustor Williamson" furnace. £t was tested with the opti-
mum 1 ml/s (gph) burner.
Flue gas NO emissions measured with the finned combustor Williamson fur-
nace (Appendix C, Table C-l) are presented graphically in Fig. 14 along
with earlier experimental results from tests of the finned-combustor in
the prototype optimum furnace and from tests of the two stock furnaces
from which these finned-rombustor furnaces were derived. The NO emis-
sion characteristics of the finned-combustor Williamson furnace are seen
to be more like those of the finned-combustor optimum furnace than of
the previous research setup rests of its finned combustor. In fact,
comparison of the several curves in Fig. 14 strongly suggests that the
larger-diameter (0.305 m) finned-combustor in the optimum furnace was
appreciably better with respect to NO and smoke emissions than the one
in the1 modified Williamson furnace. Therefore, the hypothesis that un-
cooled combustion gas recirculation (i.e., longer residence times) in
the larger firebox caused the higher-than-anticipated NO emissions from
the prototype furnace did not appear to be valid.
Heat Exchanger Design Differences - In each of these experimental fur-
naces, there was a closed-top cylindrical dome above the combustion
-------
WILLIAMSON 1167-15
HEAT EXCHANGER
PRE-PROTOTYPE
FINNED
COMBUSTOR
AIR SHROUD
WARM-AIR FAN
Figure 13. Schematic of the Finned-Combustor
Williamson Furnace Modification
-------
Cycle-Averaged Smoke <1
^^—Cycle-Averaged Smoke >1
2.5
O>
o>
2.0
01
•o
x 1.5
o
o
•r—
s_
1.0
01
>a
i.
>
I
O)
0.5
1.0
1.5
1.0
•Finned Combustor
Prototype Furnace
Stoichiometric Ratio
1.5
Stoichiometric Ratio
Figure 14. Comparison of Cycle-Averaged Nitric Oxide Emission CharacterLRCic«;
of Two Finned Combustor/Air-Cooled Heat Exc-han«er Furnaces
-------
chamber. The combustion gases flow up into the dome and out through a
rectangular opening in one side wall into the furnace's air-cooled heat
exchanger. In contrast, the research tests of the finned combustor
utilized a long, vertically disposed 0.25 m (10 inches) diameter pipe
above the finned combustor, with a water-cooled copper coil heat ex-
changer suspended in the combustion product gases. Thus, it appeared
that there were two significant design differences which might be re-
sponsible for the unexpectedly high NO and smoke conditions, viz, water
cooling versus air cooling and side discharge versus vertical updraft
discharge. These differences were tested sequentially.
To test water-cooling versus air-cooling effects, the prototype optimum
furnace was modified. The air-cooled heat exchanger was removed from
the furnace assembly and a 0.25 m (10 inches) diameter vertical pipe with
the water-cooled spiral coil was installed in its place (see Fig. 15).
This is referred to as the "coil-cooled" prototype furnace. Steady-
state NO emissions measured during several series of experiments (Runs
221-303, Table C-2) with this experimental furnace configuration are
plotted in Fig. 16; variations were made in the coil position and its
exposure to the flame zone and the coolant passed through the coil.
Notations assigned to the curves of Fig. 16 refer to: (1) heat ex-
changer position, 0.300.50 m (Ref. 3). The
bold-face L = 0.50 m, water-cooled curve in Fig. 16 is of great interest
because it duplicates exactly the NO and operational characteristics
that had been anticipated for the prototype combustor design. The NO
emissions are S. 0.50 g/kg at very low excess air (<12%>, and zero smoke
is produced at these conditions.
51
-------
WATER COOLANT
OR
AIR COOLANT
FLUE SAMPLING
SYSTEM
SPIRAL-COIL,
RESEARCH HEAT
EXCHANGER
PROTOTYPE
FINNED
COMBUSTOR
WARM-AIR
FAN
Figure 15. Schematic of the Coil-Cooled
Modified Prototype Furnace
52
-------
Smoke < 1
V/l
u>
— — — Smoke >1
1.1
en
O
o>
l/l
C
o
1/1
t/»
UJ
-------
As the water-coil heat exchanger was moved from L = 0.50 m to L = 0.30
m, an orderly family of curves was indicated, with successively lower
NO. However, when the heat exchanger was moved to L = 0.75 m, some very
unusual characteristics appeared. First, the NO curve dropped unexpect-
edly to~-0.35 g/kg. Both the instrumentation system and furnace opera-
tion were checked thoroughly, and no problem was found which would in-
dicate that the results were in error. The next day, the furnace was
started up without changing anything and, after it was warmed up, a
series of tests was begun to again recheck the NO versus S.R. curve
for L = 0.75 m. Initially, for three values of stoichiometric ratio,
the NO emissions were in the 0.6 to 0.7 g/kg range, i.e., where they'd
been expected to be, a priori. While running a fourth stoichiometric
ratio, the NO dropped abruptly to the 0.3 to 0.4 g/kg level and remained
there for the remainder of the test series. This behavior was not un-
derstood, but was suspected to be related to internal combustion gas
recirculation patterns of different sorts that happened to become
established.
To gain some insight as to whether the NO and smoke emissions, so dra-
matically improved by the change to the water-coil-cooled prototype
furnace, were altered more by the cooling medium or by the configura-
tion change, several retreats from full water cooling of the coil were
tested. First, the water was replaced by compressed air. At L = 0.50 m
(see Fig. 16), the NO emissions were increased by 30 to 40% above those
from the water-cooled case. Next, the cooling coil was removed entirely,
and it was observed that the emissions were only slightly different from
what they had been with the air-cooled coil. (Interestingly, these 0.7
to 0.8 g/kg NO emission levels are quite comparable with those obtained
in the same stoichiometric range with the prototype furnace, Fig. 11.
This suggested rather strongly that it was water cooling alone, and not
the configuration change, which influenced the NO emissions.) Finally,
water was again used as the coil coolant, but its flowrate was reduced
until the coil effluent was entirely gasified (steam). As expected, the
NO emission results (labeled "steam" in Fig. 16) were identical with
54
-------
chose measured with the air-cooled coil. Surprisingly, however, the
Hue gjs became smoky, producing significantly greater than No. 1 smoke
at conditions with less than 25% excess air. Neither the CO nor UHC
concentrations were unusually high during these smoky conditions. Again,
all systems were checked for indications of malfunctions and none were
found. In fact, it was observed that simply turning up the water flow-
rate to the coolant coil eliminated the smoky flue gas condition.
To place the foregoing data in perspective, it is appropriate to recall
that approximately 40% of the heat extraction from the flue gases was
effected by forced convective air cooling of the finned combustion cham-
fers. Thus, the mean temperature of the gases leaving one of these com-
bustors was considerably lower than the adiabatic flame temperature
which may be experienced at some portions of the flame zone. For ex-
ample, at 20% excess air and 85% overall thermal efficiency, 40% heat
removal would reduce the mean temperature from about 1830 C (3330 F) to
1080 C (1970 F). Such large temperature reductions should very effect-
ively quench the kinetically limited production of NO within the com-
bustion chamber. Therefore, it was considered that one or more of the
following hypothese must accout for (contribute to) the demonstrated
influence of the further-downstream, water-cooled, coil heat exchanger
on flue gas NO:
1. There is highly striated flow out of the combustor, with very
well-cooled gases near the chamber walls, and high-temperature,
NO-producing gases forming a central core flow. Penetration
of high-temperature striations into the heat exchanger tends
to distribute the quenching of the gases along the flowpath
length resulting in slower quenching, overall.
2. The flame zone, "seeing" the water-cooled coil, radiates suf-
ficient additional heat to it to reduce peak flame zone tem-
peratures appreciably, thereby lowering NO production rates.
55
-------
3. The copper tubes of the coiled heat exchanger act catalytically
to influence flue gas NO concentrations.
4. The heat exchanger influences NO production by inhibiting or
promoting vertical combustion gas recirculation patterns, with
stronger recirculation of cooler gases induced by the water-
cooled coil than by the various air-cooled heat exchangers.
Indications of highly striated flow were sought by measuring the radial
variations of flue gas temperature and NO concentration at the top of
the 0.25 m diameter combustion chamber extension pipe when the coil
heat exchanger was removed. Rather minimal radial variations (~10%)
were observed, so striated flow was thought to be an unlikely contribu-
tor to the heat exchangers' influence on flue gas NO concentrations.
Radiant heat transfer rates from a high-temperature gas to a cool sur-
face are proportional to the effectively seen surface area and to the
difference between the fourth powers of their absolute temperatures,
44 4
i.e., A(T ) = T - T ... The flame radiates to the combustor wall
gas surf
as well as to the heat exchanger. It is instructive to consider the
relative contributions of these two components to flame zone cooling.
If T = 2000 K, T , = 550 K, and T . = 300 K, the ACT^)
gas ' combustor ' coil '
driving potential for radiating heat to the "cold" coil is only about
1/2% higher than that for the combustor walls; i.e., to the flame, the
chamber is also quite "cold." Further, the inside surface area of the
combustor is at least four to five times the coil area that the flame
can possibly view. Thus, there is little likelihood that radiation to
a water-cooled coil cools the flame appreciably more than does radia-
tion to a somewhat warmer air-cooled coil. Nonetheless, some experi-
ments were conducted with a flat 0.18 m diameter stainless-steel "radia-
tion shield" installed 0.05 m upstream of the leading coil of the 0.15 m
diameter coil heat exchanger. In these tests, the radiation shield,
rather than the first coil, was spaced the distance, L, above the bottom
of the firebox. Three of the curves in Fig. 16 represent data obtained
with that shield in place. In the tests at L = 0.50 m, flue gas NO was
56
-------
about 10% higher with the radiation shield than without it for both
water and air as coil coolants. At least a portion of that increase is
believed to have resulted from the slight downstream displacement of
the coil. The NO level with the radiation-shielded, water-cooled coil
remained well below the air- and steam-cooled levels, so these data
were interpreted as refuting the hypothesis that the water-cooled coil
affects NO production by radiant cooling of the primary flame zone.
While the radiation shield was in place, the air-cooled coil was lowered
so that the radiation shield was located within what would be the short
0.25 m diameter cylindrical connection between the finned combustor and
commercial air-cooled heat exchanger. What was sought was an indication
of how the prototype furnace might perform if such a constricting plate
were installed to restrict vertical recirculation of gases from the
central closed-top dome back down into the combustion chamber. There
resulted a sharp increase in NO emissions to above 1 g NO/kg and exces-
sive smoke at 30% excess air and lower. This concept was an obviously
detrimental one.
If copper were a catalyst for some NO reduction reaction, its effective-
ness should be enhanced by moderate surface-temperature increases. Thus,
the observed'decrease in NO production with decreasing coil temperature
seemed like the wrong direction for copper to be catalyzing an NO con-
sumption reaction. As a final check, research combustor test data
reported in Ref. 3 were reviewed, and cases were found where NO emis-
sions obtained with the water-cooled copper coil were comparable with
those produced by a water-cooled steel coil (which the copper coil had
replaced).
These arguments left only combustion gas recirculation effects as the
most plausible way for the heat exchanger to influence NO production in
the flame zone. It was reasoned that the strength and direction of the
burner air jet, together with density gradients set up by the water-
cooled heat exchanger, induced a vertical recirculation pattern that was
57
-------
upward at the back combustor wall (opposite the burner) and downward at
the front wall. An experiment was devised, therefore, to see if the
beneficial effects of the water-cooled coil could be simulated by forc-
ing a flow of cooled, recirculated combustion gases toward the burner
side of the flame zone. For convenience in cooling and pumping, flue
gas was used as the recirculant, and it was put back into the combustor
through an existing peephole, as shown in Fig. 17. Also shown are two
additional gas-sampling locations ("B" and "C") that were installed to
attempt to track the formation of NO within the furnace. Not shown in
this schematic is a 0.11 m diameter pyrex x%rindow added to the top of
the central combustor cylinder to observe the flame during flue gas
recirculation.
Data resulting from experiments with this apparatus are listed in Table
C-3. The "Recir. Ratio" (listed in %) is defined as the ratio of re-
circulated stoichiometric burned gas to total "unburned" air. This
definition deducts excess air from the recirculated combustion gases
and combines it with fresh air injected by the burner so that the recir-
culation ratio is corrected for changes in overall stoichiometric ratio
conditions. Very large amounts of recirculated burned flue gases (40 to
50%) were required to lower the flue gas NO concentration below 0.6 g/kg
(Runs 304 to 329). An additional small water-cooling coil was added to
the FGR circuit to further cool the recLrculated gases (Runs 340 to 346).
The resultant 50 C reduction in temperature of the recirculated gas was
found to achieve an insignificant (—0.04 g NO/kg) further reduction in
NO emissions. It was believed to be unlikely that the large amounts of
flue gas recirculation that were required in these tests to reduce the
NO concentrations significantly could be induced by gravitational
effects on density gradients caused by the presence of the water-cooled
coil. It was thought a cold-coil-induced CGR flow pattern in the oppo-
site direction of rotation, i.e., rising on the front wall instead of
the rear wall, might be possible. However, no further mechanically
pumped FGR experiments were conducted.
58
-------
FGR
PUMPING
SYSTEM ~7
FGR
MEASUREMENT
SYSTEM
INSULATION
PLENUM
COMBUSTOR
COOLING
FINS
HARM AIR BLOWER
Figure 17.
Schematic of the Prototype Optimum Furnace With an
Experimental Flue Gas Recirculation System
59
-------
A series of experiments was conducted in an attempt to determine whether
NO was being formed in the heat exchanger. This involved the addition
of two sampling probes, labeled B and C in Fig. 17, within the heat ex-
changer. Location B extracted gas samples from directly above the com-
bustor and location C from the entry to the second section of the heat
exchanger. A comparison of flue samplings and internal gas samplings
(locations C and B) may be made from data gathered in several tests
(Runs 330 to 339, Table C-4). The data revealed some differences in the
completeness of combustion (CO and UHC emissions), but the NO concen-
trations showed no significant differences among the flue and the in-
ternal sample locations. This indicated that all of the NO was actually
being formed in the firebox combustion zone (as it was believed it must
be). However, the influences which downstream components can exert upon
the NO formation in the combustor have been demonstrated repeatedly.
This series of tests gave more credence to the combustion gas recircu-
lation concept than to the distributed-quench concept of NO production,
since the latter would have shown increasing NO concentrations as the
combustion gas progressed through a more slowly quenching, air-cooled
heat exchanger.
The experimental effort was then directed toward trying to improve the
emissions and operating characteristics of the prototype furnace, with
its commercial air-cooled heat exchanger, by inserting a small water-
cooled coil above the combustion chamber. A single water-cooled coil
2
(0.116 m surface area) was positioned at the L = 0.50 m chamber posi-
tion (see Fig. 18a).
The results of single water-cooled coil tests at both L = 0.50 m (Runs
347 to 356) and at L = 0.40 m (Runs 357 to 366) are tabulated in Appen-
dix C, Table C-5. All tests included in that table were with the 1.0-
60°-A oil nozzle, except for a few instances noted in the tabulation.
The single coil alone produced little .effect on the NO emissions (Fig.
19), so a baffle was added to the entrance to the rear manifold which
forced the combustion gases toward the coil before they could enter that
60
-------
cr>
(a) Single Coll
(b) Double Coil &
Rear Exit Baffle
(c) Double Coil & Extended
Rear Exit Baffle
Figure 18. Prototype Optimum Furnace With Various Configurations of Internal Water-Cooled
Coil and Baffles
-------
0>
«i
i/l
c
O
c/)
re
oo
Smoke > 1
L=0.50
1. Coil |&
Riar Baffle
0.50m
Coil &
Riar Baff
Figure 19.
1.2 1.3 1.4
Stoichiometric Ratio
Steady-State Nitric Oxide Emission Concentrations From
the Prototype Optimum Furnace With an Additional Water-
Cooled Coil
62
-------
manifold. The results from this configuration again showed very little
effect upon the NO concentrations (Runs 367 to 373). Another coil was
then added within the original coil to increase the cooling surface area
2
to 0.18 m (Fig. 18b). This configuration did have an effect upon the
combustion process (Runs 374 to 401). The NO concentrations dropped
from 0.80 to 0.55 g/kg, but the CO concentrations showed some sensitiv-
ity to this configuration.
To investigate whether the influence on the NO emissions stemmed from
combustion gas recirculation or radiation cooling of the flame zone,
an extension was added to the rear baffle (Fig. 18c). This extension
was canted to 45 degrees from vertical to induce a downward flow of
recirculant gases on the back wall. This is opposite to the direction
of the recirculation rotation that was imposed mechanically in the FGR
experiments described earlier in this section. A total of 24 firings
(Runs 402 to 425, Table C-5) was made with this extended rear baffle.
The results with water and steam cooling showed further substantial re-
ductions in NO emissions, adding support to the vertical combustion gas
recirculation hypothesis. The tests with no cooling coils (Runs 420 to
425) above the combustor showed no reduction in NO emissions, implying
that the recirculation is not a simple case of gas dynamics, but in-
volves the cooling coils to cool some of the combustion gases (i.e.,
increase gas density) to promote recirculation by gravitational effects.
To isolate this cooling-to-induce-recirculation from the simple rapid-
quench suppression of NO production, the existing double-cooling coil
was removed and a smaller cooling coil was installed. This new cooling
coil was only 0.05 m outside diameter, spiraled horizontally, and was
tucked under the overhang of the 45-degree extended baffle lip. This
coil was exposed to only a fraction of the combustion gas stream, and
it would be unlikely to influence the NO kinetics of the bulk of the
gases leaving the combustor. The fraction of the gas that it cooled
was expected (by the positioning of the coil under the baffle lip) to
be limited to the vertically recirculated combustion gases. The
63
-------
influence of this small, well-positioned cooling coil on the combustion
process was significant (Runs 426 to 433, Table C-5), showing a reduc-
tion in NO concentration from about 0.85 g/kg (no coil, Runs 420 to 425)
to about 0.65 g/kg. The CO concentrations showed a marked increase, an
additional indication that the flame zone was influenced by the small
coil.
Along with the increase in CO emissions came as associated Increase in
combustion oscillation. As a simple exploratory measure, a 0.15 m diam-
eter by 0.81 m long closed-end pipe was extended above the central com-
bustor/heat exchanger cylinder (0.25 m diameter by 0.69 m long) to
change the natural resonant frequencies of the system. This improved
the operational characteristics of the system somewhat (less combustion
oscillations) and, surprisingly, resulted in a further reduction ( -0.15
g/kg) in NO emissions (Runs 434 to 443, Table C-5). It was thought that
noisy combustion may cause transient departures of the flame zone from
its near-optimum mixing and burning conditions for producing minimum NO,
and that reducing combustion oscillations restored the flame to the
optimum combustion conditions. Evidence of this effect will be seen
again in a later discussion of the final optimization modifications to
the prototype furnace system.
The question as to whether the horizontal entry to the heat exchanger's
rear manifold could be contributing to the differences in operational
characteristics from those of the vertical combustor exhaust configura-
tion had not been fully resolved. Therefore, a set of experiments was
carried out with another heat exchanger configurational change. The
air-cooled heat exchanger was elevated by placing a 0.35 m long spool
of 0.25 m diameter pipe between it and the combustion chamber. The
double-coil exchanger was centered near the top of the spool. Combus-
tion gases were exhausted vertically upward, through a 0.18 m diameter
port at the top of the spool, after passing over the coil. A 90-degree
elbow then turned the flow into the air-cooled heat exchanger's rear
manifold. This configuration was tested with water coolant in the coils
64
-------
(Runs 444 to 447, Table C-5), and with reduced water flowrate to pro-
duce steam (Runs 448 to 451). Full water cooling produced exceptionally
low NO levels, but high CO and noisy combustion prevented operation at
acceptably low stoichiometric ratios. Reduced cooling of the coils
lowered the CO emissions somewhat, but also sharply increased the NO
production.
An attempt was made to reproduce the results obtained earlier with the
internal water-cooled coil configuration (Fig. 18c) using a low pres-
sure, air-cooled device. The low pressure requirement was added to
eliminate the need for an air compressor system within the furnace if
the results led to any working prototype. This air-cooled coil design
(Fig. 20) differed from the water-cooled coil design in that the spiral-
ing tubes formed eight parallel paths, i.e., lower pressure drop. The
air coolant was initially blown by a burner fan and was later augmented
by using two such fans in series. Air entered the top of the combustor
canister through a 0.051 m outside-diameter steel tube, which extended
down the center to the bottom of the coil heat exchanger assembly, then
was manifolded out radially into four 0.013 m outside-diameter and four
0.006 m outside-diameter copper tubes that spiraled back to near the
point of entry. The diameter of the outer coils was 0.146 m, and the
coil assembly was shrouded by a 0.152 m outside diameter by 0.229 m
long stainless-steel, sheet-metal cylinder. The shroud was added to
induce separate flow paths: external for upflowing hot gases and in-
ternal for downflowing cooled, recirculant gases. Test results are
presented in Table C-6, Runs 491 to 511, with minor changes in geometry
noted on the tabulation. The results showed that no benefits were
realized with the installation of this air-cooled coil. The minimum of
0..62 g/kg (CO < 1.0 g/kg, UHC < 0.1 g/kg, no smoke) at S.R. = 1.14
obtained with the basic prototype was not surpassed by any of the four
configurations of the low pressure, air-cooled coil system.
65
-------
Modified
Burner
B1 owe r
(Beckett)
r
Air F1ow
Control Valve
Air Outlet
Thermocouple
(Ambient Supply
or Add!t ional
Blower)
Removable
Rear Baffle
Figure 20. Schematic of the 8-Tube, Low-Pressure, Air-Cooled, Supple-
mental Heat-Exchanger Installation in the Prototype Optimum
Furnace
66
-------
DISCUSSION
The prototype optimum furnace, as designed and with the original opti-
mum burner from the Ref. 1 studies, fell somewhat short of achieving
its design goals. Of the many modifications which were tested experi-
mentally, that which came closest to satisfying the goals was retaining
the prototype furnace design unchanged and refining the burner design.
That system met all of the goals except the one for NO emissions: at
X
the nominal design point, cycle-averaged NO was below 0.65 g NO/kg fuel
burned, as compared with the target level of 0.5 g/kg.
NO emissions under that target were attained by some selected modified
X
configurations. Invariably, however, those configurations either ex-
hibited undesirable attributes concerning some other pollutant or in-
volved a "hybrid" heat transfer system (using both air and water cool-
ants) or both. It may be inferred that a hydronic boiler embodying the
low-emission burner and combustion chamber design criteria could readily
meet all the design goals. A warm-air furnace design with a portion of
its combustion gas cooling accomplished by a water-cooled heat exchanger,
on the other hand, would be at a competitive disadvantage because of
the additional complexity and cost of providing simultaneously for com-
bustion gas-to-air, combustion gas-to-water, and water-to-air heat
transfer. As a result, the remainder of this discussion is concerned
predominantly with the air-cooled prototype furnace with the refined
design optimum burner. In fact, from this point forward, the phrase
"optimum burner" will be applied to that refined design and the phrase
"prototype optimum furnace" will be used for the prototype unit in which
that burner was tested.
Comparison With Other Residential Furnaces
Pollutant Emissions - Flue gas concentrations of NO are plotted versus
stoichiometric ratio in Fig. 21. A shaded region near the middle of the
graph indicates that a large majority of existing residential oil
67
-------
3.0
=3
Li-
en
^
O
2.0
to
in
X
o
o
LU
13
1.0
CYCLE-AVERAGED
SMOKE > 1
CYCLE-AVERAGED
SMOKE < 1
ORIGINAL
STOCK FURNACE
APPROXIMATELY 80 PERCENT OF EXISTING
OIL FURNACES PRODUCE CYCLE-AVERAGE
• NOX EMISSION'S IN THIS RANGE(REF. V
PROTOTYPE FURNACE WITH WILLIAMSON
RETENTION HEAD BURNER
PROTOTYPE LOW-EMISSION FURNACE
TARGET LEVEL
I
1.0
1.5
STOICHIOMETRIC RATIO
2.0
Figure 21. Comparison of Cycle-Averaged NO Emissions From
the Prototype Optimum Furnace and Other Oil
Furnaces
68
-------
furnaces release between 1.3 and 2.2 g NO/kg fuel burned. An overall
average level of 1.8 g NO/kg fuel may be used Cor evaluating the poten-
tial impact of applying candidate NO reduction techniques (derived from
X
Ref. 4).
Measured NO emissions from the stock furnace fell on the high side and
above that typical range; at a nominal 50% excess air operating point,
it produced 2.2 g NO/kg fuel burned. Measured NO emissions from the
prototype optimum furnace were much lower. Tuned to the intended normal
operating condition with only 15% excess air, the unit produced 0.63
g NO/kg. That corresponds to reductions of about 72 and 65%, respec-
tively, from NO emissions produced by the stock furnace (at its nominal
x
operating point) and by the average estimated for all existing installed
units.
The Williamson flame retention burner, that was known to operate well in
the stoichiometric ratio range of interest, was fired in a series of
tests in the prototype optimum furnace. The data (Fig. 21) showed that
this combination produced intermediate-level NO emissions (1.5 to 1.7
X
g NO/kg fuel). This clearly shows that the optimized, finned, air-
cooled firebox is beneficial in its own right, since NO was reduced by
X
about 27% from the stock furnace's emission level, but is most effective
when combined with the optimum burner.
Carbonaceous emissions from the prototype furnace unit also were accept-
ably low at its nominal conditions, as indicated by the lower-than-No. 1
smoke. A comparison of values in Table 2 shows that CO and hydrocarbon
emission levels from the prototype furnace were somewhat higher than
those measured for the stock furnace, but were quite comparable with
averaged tuned values measured in the field survey of Ref. 4.
69
-------
Table 2. COMPARISON OF FURNACE OPERATING CONDITIONS AND
CYCLE-AVERAGED POLLUTANT EMISSIONS
Stoichiometr ic Ratio
Carbon Monoxide, g CO/kg Fuel
Unburned Hydrocarbons,
g UHC/kg Fuel
Smoke, Bacharach Number
Nitric Oxide, g NO/kg Fuel
Tuned
Averages
From Ref. k
1.85
0.6
0.07
1.3
1.8
Stock
Lennox
Furnace
1.50
0.2?
0.015
0
2.2
Prototype
Optimum
Furnace
1.15
0.55
0.055
0
0.63
Efficiencies - Steady-state efficiencies measured for the prototype op-
timum furnace are compared with those for its stock predecessor and
other residential units in Fig. 22 by superimposing the measured data
on the ANSI efficiency decrement plot of Fig. 8. Based on data from a
number of sources, it is estimated that a large majority (perhaps as
high as 80%) of existing installed residential oil heating systems oper-
ate in the shaded zone in the right-central portion of Fig. 22. Older
existing units tend to perform toward the upper and right-hand regions
of that zone, while newer equipment tends to congregate in the lower
and left-hand regions. Obviously, substantial numbers of units also
operate outside that shaded zone, and they are distributed around it on
all four sides. The average behavior of all United States oil-fueled
heating systems probably lies in the central crosshatched region of that
zone, with net flue temperatures in the neighborhood of 280 C (500 F),
COj concentrations of around 8%, and estimated average steady-state
efficiencies between 72 and 75%.
The performance curve for the original stock furnace fell well below
(i.e., at higher efficiencies) the shaded band representative of exist-
ing installed residential heating units. The stock unit could be tuned
to a moderately low 50% excess air nominal operating condition where its
70
-------
35
S 30
cc
2 25
00
in
o
20
15
10
APPROXIMATELY 80 PERCENT OF
EXISTING FURNACES OPERATE
IN THIS ZONE
SMOKE > NO. 1
1.0
1.2
I.'* 1.6
STOICHIOMETRIC RATIO
I I
1.8
2.0
2.2
13
12
1]
10
VOLUME PERCENT COj (DRY BASIS) IN FLUE GAS
65
70
75
00
O
80
85
90
Figure 22. Comparison of Steady-State Thermal Efficiencies of the Prototype Optimum
and Other Oil Furnaces
Furnace
-------
net flue gas temperature averaged only 180 C (325 F). The resultant
steady-state gross thermal efficiency was 82.5% (i.e., the stock furnace
was among the higher-performing units in the existing population).
Thermal efficiency levels achieved by the prototype optimum furnace were
qualitatively the same as those of the stock furnace. However, as is
evident in Fig. 22, flue gases leaving the optimum unit were 40 to 55 C
(75 to 100 F) hotter than those from the stock furnace, and the effi-
ciency decrement due to the higher net flue gas temperature was offset
by operating the prototype unit at substantially lower stoichiometric
ratio. The thermal behavior of the prototype optimum furnace was sur-
prising because the stock furnace's compact heat exchanger was retained
intact and was supplemented substantially by the finned firebox heat
exchanger. This apparently anomolous behavior probably was caused by
warm-air jets from the firebox region bypassing some of the main heat
exchanger. It should be relatively easy to correct this condition. It
can be estimated from Fig. 22 that if the prototype unit's net flue gas
temperature were the same as that of the stock unit, its efficiency
would be increased by about 2% to an overall steady-state gross thermal
efficiency of 84 to 85%.
The 82 to 83% steady-state thermal efficiency exhibited by the prototype
optimum furnace was close to the maximum achievable in noncondensing
flue gas residential systems. Taken alone, this is not unique, since
comparable efficiencies are attained by some current commercially avail-
able units (as exemplified by the stock furnace that was converted into
the prototype). What is unique and important about it is the demonstra-
tion that near-maximum, steady-state efficiency and near-minimum NO
X
emissions can be obtained simultaneously.
Operational and Design Aspects
The prototype optimum furnace test results confirmed the feasibility of
applying the several newly developed, low-emission, oil burner and
72
-------
firebox design criteria to residential space-heating equipment. The
experimental prototype unit came very close to satisfying all of the
pollutant emission and efficiency objectives for which it was designed.
Operationally, its behavior was quite comparable with current commer-
cially availabe furnaces. The 500-hour-duration test, equivalent to
about one-tenth of an average heating season, indicated that the unit
might serve through an entire no-maintenance winter heating season with-
out exhibiting appreciable shifts in operating conditions or pollutant
emission levels.
As delineated in the statement of the design criteria, even greater NO
X
reductions may be achieved by adopting tunnel-fired burner orientations
and/or water-cooled combustor walls in furnace designs. Some beneficial
reductions in NO emissions (up to about 25%) were achieved by invoking
X
either the burner design criteria alone or the firebox design alone,
but the NO emission goal could be approached only by using both sets
X
of criteria in combination. Furthermore, there was no assurance that
either set of criteria alone could minimize the excess combustion air
requirement.
The decrement between steady-state and cycle-averaged efficiencies of
the prototype, optimum furnace was smaller than the uncertainty in meas-
uring the latter value in the outdoor laboratory. As a result, quanti-
tative assessments were not obtained for features included in the pro-
tptype unit to cut down on cyclical heat losses. Elimination of draft
air heat losses was the most important of these. Although it was antic-
ipated that some burner or electrical control overheating during
standby might result from sealing the burner vestibule, absolutely no
indication of any problem was observed during the 500-hour test. Pre-
sumably, the metallic firebox was not hot enough to cause a radiation
problem, and conduction was acceptably low.
In view of its demonstrated steady-state efficiency and the apparent
effectiveness of its features for reducing standby heat losses, the
73
-------
prototype optimum furnace should achieve 75% or higher cycle-averaged
efficiency. This is perhaps 13% higher than the estimated mean season-
averaged performance of existing United States residential oil heat
sources that it might replace. Such replacement would be attended by
an average of about 17% reduction in fuel consumption. This estimated
lower fuel consumption may be combined with the 65% reduction in NO _
*\
(as normalized by the mass of fuel burned) to calculate the total effect
on mass emissions. The result is that, if a prototype optimum furnace
replaced an "average" existing unit and satisfied the same thermal
demand, the mass of NO emitted would be reduced by 71%.
Reductions in fuel consumption brought about by the sealed barometric
and combustion air supply system have not been included in the fore-
going discussion. This type of sealed air system has been shown in
laboratory testing (Ref. 5) to reduce residential heating oil consump-
tion by a minimum of 5% and up to 15%. An approximate average value
for its potential effect on optimum furnace fuel consumption is prob-
ably a little above the minimum, say 8%. Adding this to the estimated
17% due to higher unit efficiency yields an average anticipated fuel
savings of 25%. Corresponding to that is an estimated 74% overall
reduction in the mass of NO emissions.
x
-------
SECTION V
INTEGRATED SYSTEM DESIGN
It should be clear, from the foregoing description and discussion of
test results, that the prototype furnace closely approached an optimum
design with respect to air pollutant emissions and performance. Addi-
tional attention was given to some economic considerations, such as
unit weight, materials costs, and fabrication methods, and design de-
tails were derived for a candidate, integrated, low-emission, warm-air
furnace. The integrated system design is described in this section in
terms of those design features which differ from the prototype furnace
described in Section IV.
A cutaway perspective drawing of the integrated furnace is shown in
Fig. 23. This is supplemented by several views from an assembly drawing
in Fig. 24. Components which differ from the prototype furnace design
(Fig. 2) are the burner assembly, the finned combustor, the baffles in
the warm-air flow passage, and the burner vestibule closure panels.
Details are described in the following subsections.
BURNER ASSEMBLY
The basic burner assembly in the integrated furnace design is the opti-
mum burner as it was finally tested in the prototype furnace. That as-
sembly included the standby draft control device, the quiet stator on
the fan discharge, a static disc in the blast tube, and the research
optimum head. Two changes have been incorporated for the integrated
furnace design, as follows:
Combustion Air Control Device
A potential problem was cited in Ref. 3 in satisfying applicable safety
standards with the prototype standby draft control unit that relies upon
burner fan suction to open a flap valve held closed by gravity. Although
75
-------
FINNED
AIR-COOLED
COMBUSTOR
DRAFT CONTROL
ASSEMBLY
SEALED AIR
SYSTEM
COMBUSTION
AIR FILTER
OPTIMUM
BURNER
Figure 23.
Cutaway Perspective Drawing of Integrated, Low-
Emission, Warm-Air Furnace Design
76
-------
0.660m
DRAFT
CONTROL
VALVE
•* *»
FRONT VIEW
SEALED
COMBUSTION
AIR INLET
SEALED
VESTIBULE
(LOUVERS CLOSED)
COMBUSTION
AIR FILTER
48m
PULSE-FREE
STATOR
COMPACT HEAT
EXCHANGER
(LENNOX)
FINNED
AIR COOLED
COMBUSTOR
OPTIMUM
BURNER HEAD
WARM-AIR
FAN
SECTION A-A
(a) Front and Side Sectional Views
Figure 24. Layout Assembly Drawing of the Integrated Furnace Design
-------
COMBUSTION
AIR INLET
SEALED LOUVERS
FLUE OUTLET
TOP VIEW
COMPACT
HEAT EXCHANGER
(LENNOX)
o— ,<>- —o-—,-.G---O. -a
WARM-AIR BAFFLES
FINNED
AIR-COOLED
COM6USTOR
SECTION C-C
(b) Top, Rear and Bottom Sectional Views
Figure 24 (concluded). Layout Assembly Drawing of the Integrated
Furnace Design
78
-------
this device functioned well in the laboratory testing, it was antici-
pated that an interlock with a solenoid shutoff valve in the fuel supply
line, to ensure that the flap is open before fuel can flow to the
burner, might be required to satisfy the Underwriters' Laboratories
standard (Ref. 8). Therefore, a combustion air control device has been
designed in which the shutoff flap, upon opening, closes a normally
open microswitch. This is illustrated in Fig. 25. The microswitch is
mounted on the end of a threaded rod which also provides a positive stop
for the air flap. Adjustment of the rod's insertion depth controls the
distance that the flap can open, and so can be used to control the com-
bustion air flowrate.
Optimum Burner Head
The research version of the optimum burner head, fabricated by machin-
ing and welding stainless-steel plate (Ref. 1), was utilized throughout
the testing of the prototype furnace. Less expensive, commercially
practicable head fabrication methods were considered in the study re-
ported in Ref. 2. The preferred method was found to be stamping and
folding heads from stainless-steel sheet. Prototype heads made to sim-
ulate those which might be made commercially were tested and found to
reproduce quite well the performance of the research optimum head and,
potentially, to be durable and long-Jived. Therefore, the stamped
sheet metal optimum head is incorporated into the burner assembly for
the integrated furnace design. This head is illustrated in Fig. 26.
(reproduced from Ref. 2) as a composite plan view of the flat sheet-
metal stamping and a rear view of the optimum head after the six swirl
vanes have been folded twice into their final positions.
FINNCD ATR-COOLED COMBUSTION CHAMBER
The combustion chamber was redesigned with three principal objectives in
mind: (1) fabrication by a less expensive method than the machined and
79
-------
oo
o
•Draft Flap
Oil Pump
Bo 11 Ho 1 es
073
0 i1 Pump
Mount ing
Hole
Flap
Adjuster
Assembly
Microswi tch
NOTE: Dimensions in meters
0.036
Figure 25. Drawing of the Combustion Air Control Device
-------
00
6 VANES POSITIONED
AS SHOWN WITH
OVERLAP OF
CENTER HOLE
O 038 DIA
0.101 DIA
NOTE: DIMENSIONS ARE IN METERS
MATERIAL: TYPE 310 STAINLESS STEEL
SECTION A-A
I , r ,- 0.0015 R
I » | TYP 4 PLACES
C?Btx=—-
11 |J.«—0.038 DIA(REF)
'IP*— 0.048 DIA
(•*— 0.058 DIA
* 0.079 DIA
O.089 DIA
Figure 26. Stamped and Folded Sheet Stainless-Steel Optimum Head
-------
welded construction of the prototype unit, (2) reduction of weight, and
(3) provision for joining the firebox to the heat exchanger by welding.
The new firebox design is illustrated in Fig. 27. Metal casting was
selected as the best and cheapest way of making the relatively compli-
cated finned combustor assembly. The fins were changed from a predomi-
nantly radial orientation to a predominantly parallel orientation to
simplify and reduce the number of pieces needed for a casting mold.
The design provides a rolled carbon-steel ring, to be fitted in the mold
and integrated into the casting, that matches and is to be welded to the
central cylinder of the fabricated sheet metal heat exchanger. The ring
is perforated with a number of countersunk holes that, when filled with
casting metal, are intended to ensure permanency of the bond.
The design of the cast-metal firebox provides typical wall and fin
thicknesses of 0.0063 m (0.25 inch), which is close to the minimum for
reliable casting of a unit of this size and complexity. The reduced
metal thickness helps to lower firebox mass. Together with some short-
ening of the redistributed fins and the elimination of the bolted heat
exchanger attachment rings, the total mass has been reduced from the
61.2 kg (135 Ibm) of the prototype finned combustor into the 27 to 30 kg
(60 to 66 Ibm) range for a cast-iron unit. Although that would be on
the order of 50% reduction from the mass of the existing prototype unit,
it can hardly be said to be in economic competition with current com-
mercial construction.
Further appreciable weight reductions would probably require either re-
ducing the dimensions of the firebox or using a less dense construction
material. The firebox dimensions, having been derived from burner emis-
sion and performance optimization studies, are no longer considered to
he optional variables, but a material change may well be acceptable.
Cast ciluminum is the most likely candidate; its use would result in a
82
-------
•0.381 m
oo
CO
0.378
0.279
—O.OkQ
0.221 m
0.010 Dia
60° CSK
8 Places
Carbon Steel Ring
O.OA8-
•0.191
l~l V I
r»0. 098*^1
-0.273-
•0.3^6
•0.308-
Figure 27. Cast-Metal Finned,Air-Cooled Combustion Chamber
-------
firebox weight in the 10 to 12 kg (22 to 26 Ibm) range. Before choosing
this material, however, consideration should be given to its impact on
several areas other than weight. For example, cast aluminum will prob-
ably cost more than cast steel, so the tradeoCf between increased ini-
tial cost and lower shipping weight should be investigated. The dura-
bility and potential lifetimes of aluminum and steel fireboxes should
be estimated and compared; this should include fLime-impingement ero-
sion and long-term cycle fatiguing. Particular design attention must
be given to effects of differentia] thermal expansion where dissimilar
metals are joined. Prevention of excessive strain and yielding of one
or the other metal, leading to development of gas leaks where an alumi-
num firebox is bonded to a steel heat exchanger, is an area of major
concern. From d thermal or heat transfer standpoint, aluminum might be
preferable to steel because of its higher conductivity; the combustor
wall temperatures should be more uniform and slightly lower.
IIF.AT r.XCHANGE CONSIDERATIONS
The photographs of the partially assembled prototype furnace in Fig. h
show how the firebox and heat exchanger were installed and coupled.
The firebox was mounted directly above the warm-air blower so that
upward-flowing furnace coolant air flowed first over the finned hemi-
spherical bottom of the firebox, then vertically upward between vertical
fins on the firebox walls. Two sheet-metal baffles, one on either side
of the firebox, prevented the warm air from bypassing the finned fire-
box. Altogether, the cross-sectional area for air flow past the com-
bustor was reduced to about 58% of that in the stock Lennox furnace.
Combined with the effect of air heating in this section of the furnace,
the warm-air velocity past the top of the prototype firebox was in-
creased by about 80% above that in the predecessor furnace. Even though
the baffles flared out at that point, restoring the full nominal cross-
sectional flow area through the heat exchanger, the air jets between
baffles undoubtedly retained most of their elevated velocity and did
84
-------
not expand effectively to fill the available cross section. Thus, the
outermost heat exchanger panels, above the flared ends of the baffles,
were not cooled as effectively as were the inner panels. Additionally,
the flanged joint between the firebox and the heat exchanger protruded
about 3/4 inch into the air stream, tending to displace it away from the
central cylindrical dome section of the heat exchanger and to reduce
further the air-cooling effectiveness.
The design has been modified to help restore the effectiveness of the
heat exchanger. The baffles in the warm-air passage have been moved
further from the firebox, so that the flow cross section is less con-
stricted. The baffles also begin to taper out at about the midsection
of the firebox to permit the air flow to decelerate and expand to the
heat exchanger cross section more gradually. The less bulky joint be-
tween the combustion chamber and heat exchanger also should contribute
to smoother flow and effective heat exchange.
VESTIBULE CLOSURE PANEL
Only one modification to the exterior cabinet of the stock Lennox furn-
ace was required. The optimum 1 ml/s (gph) residential oil burner has
a longer blast tube than has the stock Lennox burner. As a result, the
burner protruded beyond the front of the burner vestibule and interfered
with the panel that closes the vestibule. For the integrated furnace
design, a vestibule closure panel has been provided with a bulge to ac-
commodate the burner.
85
-------
SECTION VI
REFERENCES
1. Dickerson, R. A., and A. S. Okuda: "Design of an Optimum Distillate
Oil Burner for Control of Pollutant Emissions," EPA-650/2-74-047,
Environmental Protection Agency, Research Triangle Park, North
Carolina, June 1974.
2. Combs, L. P., and A. S. Okuda: "Commercial Feasibility of an
Optimum Residential Oil Burner Head," EPA-600/2-76-256, Environ-
mental Protection Agency, Research Triangle Park, North Carolina,
September 1976.
3. Combs, L. P., and A. S. Okuda: "Residential Oil Furnace System
Optimization - Phase I," EPA-600/2-76-038, Environmental Protection
Agency, Research Triangle Park, North Carolina, February 1976.
4. Barret, R. E., S. E. Miller, and D. W. Locklin: "Field Investiga-
tions of Emissions from Combustion Equipment for Space Heating,"
EPA-R2-73-084a (also API Publ. 4180), Environmental Protection
Agency, Research Triangle Park, North Carolina, June 1973.
5. Peoples, C., "Sealed Oil Furnace Combustion System Reduces Fuel
Consumption," Addendum to the Proceedings, Conference on Improving
Efficiency in HVAC Equipment and Components for Residential and
Small Commercial Buildings, Purdue University, LaFayette, Indiana,
October 1974, pp A66-A72.
6. "American National Standard Performance Requirements of Oil-
Powered Central Furnaces," ANSI Z91.1-1972, American National
Standards Institute, Inc., New York, New York, June 1972.
7. Hall, R. E., J. H. Wasser, and E. E. Berkau: "A Study of Air Pollu-
tant Emissions from Residential Heating Systems," EPA 650/2-74-003,
Environmental Protection Agency, Research Triangle Park, North
Carolina, January 1974.
8. "Standard for Safety: Oil-Fired Central Furnaces," U.L. Standard
727, Underwriters' Laboratories, Inc., Chicago, Illinois, November
1973.
86
-------
APPENDIX A
FLUE GAS COMPOSITIONAL ANALYSIS
The sample flow train used for analyzing flue gas composition is illus-
trated in Fig. A-l. A 0.006 m (1/4 inch) diameter stainless-steel tub-
ing sample probe was inserted near the combustor or flue pipe center-
line, downstream of the heat exchanger. Flue gas aspirated into the
sample probe flowed through a line to an air-cooled condensibles trap
where particulates and heavy oils were separated out. Next, the gas
passed into an ice-cooled, stainless-steel condensibles trap where most
of the water and any condensible, low-volatility hydrocarbons were
removed. After the condenser, the gas passed into a Pyrex wool-filled
glass cylinder which served as a final separator for heavy oils and
particulates, and provided a visual indication of the cleanliness of
the gas being admitted to the analysis instruments. Table A-l gives
a summary of the gas analysis instruments used. The gas leaving the
glass-wool filter was split into three parallel paths. One path led
directly to the total hydrocarbon analyzer. A second path led through
a Drierite bed where water vapor was removed, then into the series-
plumbed CO, C0», and 0 analyzers. The third path passed through a
o
combined Drierite and 3 A molecular sieve bed for total water removal,
then into the nitric oxide analyzer. The gas was pumped through the
system by three diaphragm pumps located downstream of the nitric oxide
analyzer, total hydrocarbon analyzer, and the series of CO, CO and
0 analyzers.
When the analytical system shown in Fig. A-l is used to analyze gases
which may have been quenched before combustion was compled, there are
two factors that must be considered in reducing the data: (1) only
burned or partly pyrolyzed fuel is included in the analysis, since
minute quantities of liquid or vapor fuel may be removed by the cold
trap and (2) water formed from hydrogen and oxygen during the combus-
tion process is also removed from the analyzed sample by the cold trap.
87
-------
00
00
Smoke Sample
FlueGasN
Sample
Glass Wool
Filter
Air-Cooled
Condensibles
Traps
Ice-Cooled
3A Molecular
Sieve and
Drierite Bed
1
i
*.'•>
Drierite
Bed
Nitric Oxide
Analyzer
Total UHC
Analyzer
Carbon
Monoz i de
Analyzer
Carbon
Dioxide
Analyzer
Vent
Oxygen
Analyzer
Figure A-l. Analytical System for Fuel Oil Burner Emissions Analysis
-------
Table A-l. EXHAUST ANALYSIS INSTRUMENTS
oo
Type
Range
Sensitivity
Calibration
CO
MSA
Nondispersive IR
LIRA
Model 300
0 to 1SOO ppm
(nole)
30 ppm minimum
detectable
1000 ppn CO in
N. standard gas
co2
MSA
Nondispersive IR
LIRA
Model 300
0 to 20 mole \
0.25% minimum
detectable
141 C02 in N2
standard gas
NO
MSA
Nondispersive IR
LIRA
Model 200
0 to 500 ppn
(mole)
10 ppn minimum
detectable
0.821 C-H in
N used as
simulant for
4 10- ppm NO
standard
Total HC
MSA
H2 flame
lonization
detector
0.2 to 800 ppm
total HC by
volume as CH.
10 ppm minimum
detectable
31 CH4 in helium
used as a
standard
Oxygen
Beckman
polarographic
0 to lOOt
-o.n
Air - 21**
N? = 01
Snoke
Bacharach
(manual)
0 to 9
1
Ten spots of
monotonically
varying
darkness
-------
Values calculated from the measured flue gas compositional data included
the overall stoichiometric ratio, the weight of nitric oxide per unit
weight of burned fuel, and the weight of carbon monoxide per unit weight
of burned fuel. The method of calculation to obtain these values is
described below.
The calculations were based on air having the following nominal
composition:
Component Mole % Wt %
N2 78.08 75.63
02 20.95 23.19
Noble gases 0.94 1.13
(Ar, He and Ne)
C02 0.03 0.05
100.00 100.00
The composition of the fuel was assumed to be characterized by the
formula CH where, for the No. 2 fuel oil burned in this program,
X
x = 1.814. The following symbols were used in the calculations:
AIR = moles of air to produce 100 moles of dry flue gas
FUEL = moles of fuel to produce 100 moles of dry flue gas
CO = moles of carbon monoxide in 100 moles of dry flue gas
CO = moles of carbon dioxide in 100 moles of dry flue gas
NO = moles of nitric oxide in 100 moles of dry flue gas
Oj = moles of oxygen in 100 moles of dry flue gas
HC = moles of hydrocarbon, as CH,, in 100 moles of dry
flue gas
The values of CO, C0_, NO, 0~, and HC were obtained directly from the
analysis instruments. In the following, it is assumed that all hydro-
gen is oxidized to water and condensed out of the system at the cold
trap, prior to analysis.
90
-------
An oxygen balance yields:
0.2095 AIR = C02 - 0.0003 AIR + 0.5 CO + 0.25 x (C02 + CO (A-l)
- 0.0003 AIR) + 0.5 NO + 02
The left hand side of the above equation represents the total free
oxygen contributed by the air. The first two items on the right side
represent moles of oxygen tied up in C0_, less the amount of CO-
originally present in the air. The third term represents moles of
oxygen tied up as carbon monoxide. The fourth term represents oxygen
consumed to oxidize hydrogen, yielding the water condensed out in the
cold trap. The fifth term is the oxygen tied up in nitric oxide. The
sixth term is free oxygen remaining in the sample reaching the analysis
instruments. Equation A-l can be arranged to yield:
(1 + 7) CO. + (1/2 + £) CO + 1/2 NO + 0
ATD _ ^ *"
0.2095 + 0.0003 + 0.0003 x/4
A carbon balance can be used to calculate the moles of fuel burned per
100 moles of dry flue gas:
FUEL = CO- - 0.0003 AIR + CO (A-3)
The moles of air available per mole of burned fuel in the sample gas
can be obtained by taking the ratio of the values from Eq. A-2 and A-3.
AIR must be calculated first, before calculation of FUEL. If the com-
bustion were in stoichiometric proportions, the moles of air would be,
by an oxygen demand calculation:
AIR = (1 + */*> FUEL
A stoich 0.2095 U *'
The stoichiometric ratio of the locally sampled burned gases is a para-
meter frequently used in this report. It is defined as the ratio of
AIR to AIR , and Is designated SR.
91
-------
(A-5)
stoich
Combination of Eq. A-2 through A-5 yields a direct calculation of the
burned gas stoichiometric ratio in terms of the measured parameters:
(1 4- |) C02 4- (1/2 4- ^) CO + 1/2 NO 4- 02 (A_6)
0.2095 4- 0.003 + 0.0003 x/A
SR
(1 + f) f (14- £)CO, 4- (1/2 4- £) CO + 1/2 NO + oj
n 90Q. CO, 4- CO - 0.0003
0.2095 [^ 2 0.2095 4- 0.0003 4- 0.0003 x/A J
According to the above definition, when the sample contains just a
sufficient amount of air to oxidize all of the fuel in the sample to
COj plus condensed-out water, then SR =1. As a second example, if
there is twice the required amount of air for complete oxidation of the
fuel, then SR = 2. Note that the stoichiometric ratio, as calculated
from Eq. A-6 does not require that the products in the flue gas be in
chemical equilibrium.
Note that the accuracy of the stoichiometric ratio calculation would be
affected very little if all terms in Eq. A-6 containing the factors
0.0003 and NO were ignored. These factors represent the carbon dioxide
originally present in free air, and the oxygen tied up in nitric oxide,
respectively.
One partially questionable assumption made in the formulation of Eq.
A-6 was that all hydrogen originally present in the fuel becomes
oxidized to water and is removed in the cold trap. This was a neces-
sary assumption, since there was no instrument available to measure the
actual hydrogen content of the sample gas. The assumption is very good
under the combined conditions of air-rich stoichiometric ratios (SR > 1)
and chemical equilibrium. To test this assumption, a Rocketdyne thermb-
chemical computer code was used to calculate the species concentrations
under conditions of chemical equilibrium for stoichiometric ratios from
0.8 to 2.8. These calculations included the equilibrium presence of
free H_. The actual stoichiometric ratios of these combustion gases,
92
-------
compared to those calculated by Eq. A-6 (which does not recognize the
presence of H ) are given in Table A-2, where it can be seen that Eq.
A-6 is quite accurate except for SR < 1. Calculated equilibrium condi-
tions are tabulated in Tables A-3 and A-4.
Table A-2. VALIDITY OF STOICHIOMETRIC RATIO CONDITIONS
Actual Stoi ch iometric Ratio
0.800
1.000
1.200
1.600
2.000
2.*400
2.800
Stoichiometr ic Ratio Calculated
from Eq. B-6
0.8^
1 .003
1.197
1.600
2.002
2.1»0/4
2.80*4
The primary cause of the inaccuracy at SR < 1 is the unaccounted for
presence of H-. In nonequilibrium gases, there is likely to be H_
present even where none would be indicated from equilibrium calcula-
tions and, at fuel-rich conditions, there could be more or less than
indicated from equilibrium calculations. Because of this likelihood
of nonequilibrium, no attempt was made to correct the calculations
of Eq. A-6 by means of equilibrium calculations.
The concentration of C0_ (dry basis) in the flue gas in the parameter
most often used in the space heating industry as an indication of com-
bustion conditions. To illustrate the relationship of %CO? to the
stoichiometric ratio, equilibrium data from Table A-4 were used to
calculate the curve shown In Fig. A-2; a calculated
shown.
curve is also
A number of values of measured CO. concentrations in actual
93
-------
vO
(CH
1.814
Table A-3. EQUILIBRIUM COMBUSTION GAS PROPERTIES FOR NO. 2
DISTILLATE FUEL OIL BURNED WITH AIR
, 18,443 Btu/lb Net Heat of Combustion With Air at 14.67 psia)
Stoich.
Ratio*
0.8
1.0
1.2
1.4
1.6 Air
2.0 Rich
2.4 1
2.8 T
0.8
1.0
1.2
1.4
1.6
2.0
2.4
2.8
0.8
1.0
1.2
1.4
1.6
2.0
2.4
2.S
Oil » Air
Inlet Temp.,
F
0
70
200
Flame
Temperature,
F
3429
3614
3290
2940
2649
2209
1897
1663
3778
3649
3356
2991
2703
2765
1955
1722
3867
3709
3418
3035
2802
2369
2061
ISil
cp
Frozen,
Btu/lb-R
0.346
0.341
0.333
0.324
0.318
0.307
0.298
0.291
0.347
0.341
0.333
0.32S
0.318
0.308
0.299
0.193
0.347
0.342
0.334
0.326
0.520
0.309
0.501
0.211S
Y
Frozen
1.261
1.254
1.260
1.267
1.275
1.288
1.298
1.308
.261
.254
.259
.267
.274
.286
.297
.306
.260
.257
.259
.266
.273
.284
.294
.-505
Viscosity,
centipoise
0.0666
0.0687
0.0653
0.0615
0.0581
0.0527
0.0487
0.0456
0.0671
0.0691
0.0658
0.0621
0.0589
0.0535
0.0495
0.0464
0.0681
0.0698
0.0668
0.0652
0.0600
0.0548
0.0509
0.0-179
Thermal
Conductivity,
Btu/hr-ft-F
0.0702
0.0711
0.0661
0.0610
0.0567
0.0500
0.0452
0.0415
0.0709
0.0715
0.0667
0.0617
0.0574
0.0509
0.0461
0.0425
0.0720
0.0725
0.0678
0.0629
0.0588
0.0524
0.0477
0.0441
Prandtl
Number
0.7946
0.7984
0.7954
0.7915
0.7880
0.7820
0.7771
0.7750
0.7948
0.7984
0.7956
0.7918
0.7884
0.7825
0.7778
0.77?8
0.7951
0.7985
0.7958
0.7923
0.7890
0.7S34
0.7'90
o. •>-:,!
Molecular
Weight
27.73
28.80
29.00
29.03
29.03
29.02
29.01
29.00
27.72
28.77
29.00
29.03
29.03
29.02
29.01
29.00
27.71
28.73
28.98
29.02
29.02
29.02
29.01
29.00
"Stofchiometric ratio is unity at 14.'«9 masses of air per mass of fuel, and
proportionately greater than unity for increasing relative mass of air.
-------
Table A-4. CALCULATED EQUILIBRIUM COMBUSTION GAS COMPOSITION, VOLUME OR MOLE PERCENT
vo
in
Stoich.
Ratio
0.8
1.0
1.2
1.4
1.6
2.0
2.4
2.8
0.8
1.0
1.2
1.4
1.6
2.0
2.4
2.8
0.8
1.0
1.2
1.4
1.6
2.0
2.4
2.8
Oil + Air
Inlet Temp.. F
0
70
200
H
0.0630
0.0397
0.000
0.000
0.000
0.000
0.000
0.000
0.0737
0.04SS
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0964
0.0577
0.0000
0.0000
0.0000
0.0000
0.0000
o.ooon
0
0.0000
0.0313
0.0217
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0362
0.0261
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0468
0.0356
0.0000
0.0000
0.0000
0.0000
0.0000
Ar
0.821
0.866
0.882
0.890
0.895
0.902
0.907
0.910
0.821
0.866
0.882
0.890
0.895
0.902
0.907
0.910
0.821
0.864
0.882
0.890
0.89S
0.902
0.907
0.910
OH
0.0499
0.2816
0.1862
0.07S7
0.0790
0.000
0.000
0.000
0.0613
0.3072
0.2082
0.0885
0.0351
0.000
0.000
0.000
0.0878
0.3579
0.2533
0.1157
0.0493
0.0000
0.0000
0.0000
H2
2.016
0.250
0.030
0.000
0.000
O.OCO
0.000
0.000
1.996
0.269
0.036
0.000
0.000
0.000
0.000
0.000
1.964
0.304
0.048
0.000
0.000
0.000
0.000
0.000
H20
12.263
11.690
10.141
8.832
7.799
6.297
5.276
4.541
12.271
11.647
10.121
8.824
7.795
6.297
S.276
4.541
12.273
11.562
10.078
8.806
7.787
6.295
5.276
4.541
CO
7.243
1.393
0.161
0.0203
0.000
0.000
0.000
0.000
7.268
1.501
0.195
0.026
0.000
0.000
0.000
0.000
7.318
1.710
0.270
0.042
0.000
0.000
0.000
0.000
CO,
8.687
12.0S2
11.247
9.841
8.679
7.000
5.864
5.046
8. 659
11.934
11.210
9.835
8.678
7.000
5.863
5.046
S.604
11.705
11.127
9.816
8.672
7.000
S.S63
5.046
NO
0.000
0.253
0.390
0.2955
0.2080
0.0829
0.0339
0.000
0.017
0.272
0.404
0.322
0.223
0.096
0.041
0.018
0.027
0.310
0.451
0.373
0.268
0.125
0.059
0.028
N2
68.837
72.522
73.784
74.465
74.947
75.603
76.028
76.326
68.901
72.4S6
73.751
74.447
74.933
75.596
76.023
76.323
68. ^96
73.328
73.683
74.405
74.905
75.582
76.015
76.319
°2
o.coo
0.619
3.160
5.566
7.444
10.107
11.888
13.161
O.POO
0.666
3.159
5.553
7.432
10.100
11.884
13.159
O.OCO
0.754
3.162
5.526
7.406
10.085
11.876
15.154
-------
14
13
1C
e>
cu
n
CM
8
O)
10
61 I 1
1.0
CO- Data from Ref. A-1
C0» Data from Ref. A-2
Calculated Equilibrium; Ref. A-3
I I I 0
1.5 2.0
Stoichlometric Ratio
14
VI
12 3
0)
10 o
CM
o
O)
Ol
O.
Figure A-2. Flue Gas C02 and 02 Concentrations for No. 2 Fuel Oil
Burned in Ambient Air at 1 atm
96
-------
furnace flue gases are also plotted on Fig. A-2. The measured data
are seen to be very well correlated by the calculated equilibrium curve
at SR > 1.1 (the calculated maximum C0_ concentration as the stoichio-
metric condition is approached by reducing excess air is not normally
observed in furnace testing).
Other parameters of interest for the flue gases are the mass ratio of
nitric oxide to burned fuel, the mass ratio of carbon monoxide to burned
fuel, and the mass ratio of unburned hydrocarbons (as CH.) to burned
fuel. These ratios are generally expressed herein as grams of nitric
oxide per kilogram of burned fuel (g NO/kg fuel) , grams of methane
per kilogram of fuel (g UHC/kg fuel) , and grams of carbon monoxide per
kilogram of burned fuel (g CO/kg fuel) . These parameters are calcula-
ted by aid of Eq. A-2 and A-3 from the following relationships:
gNO (1000) (NO) (MWNQ)
^ '
'
kg fuel (C00 - 0.0003 AIR + CO) (MWJ (A~9)
i r
kg fuel
g CO
kg fuel
g UHC
(C02 - 0.0003 AIR + CO) (MWp)
(1000) (CO) MWCO
(C02 - 0.0003 AIR + CO) (MWp)
(1000) (HC) (MWCR )
where
MW = molecular weight of NO = 30.01
MW = molecular weight of fuel
= 12.01 + 1.008 x = 13.84
MW = molecular weight of CO = 28.01
dU
MW = molecular weight of methane = 16. 04
CH^
For calculation of the above quantities, the term 0.0003 AIR can bp
neglected without introducing more than about 0.1% error in the calcu-
lations, or AIR can be computed from Eq . A-3 and included in the
97
-------
calculation. The numbers given in this report include the effect of
the term. The experimental data were reduced, according to the above
equation, bv means of a remote terminal timeshare computer program.
In addition to the gaseous pollutants described above, the smoke
content of the mixed gases was also measured. The instrument utilized
for this purpose was a Bacharach smoke meter. (It is manufactured by
the Bacharach Instrument Company, Pittsburgh, Pennsylvania.) This is
a hand-held device which, when pumped, sucks flue gases from a 0.006 m
(1/4-inch) OD, uncooled sample probe through a piece of white filter
paper; 10 strokes of the pump, over a period of about 15 seconds,
3 2
causes the passage of 57.2 m of flue gas per m of filter paper
3 2
(2250 in. /in. ). The smoke particles deposit out on the filter paper.
A reading is taken by comparing the darkness of the smoke deposition
spot to a scale of 1-0 such calibrated spots provided with the instru-
ment. The readings vary from 0 to 9. A reading of zero corresponds
to no visually detectable deposit on the filter paper, while a reading
of 9 corresponds to a dark black deposit. Intermediate readings are
varying shades of black and gray, increasing in darkness with increas-
ing reading numbers. A reading of 1 is generally accepted by the
industry as a very acceptable degree of smoke. At the opposite extreme,
a reading of 9, which is totally unacceptable, still does not corres-
pond to sufficient smoke to be easily visible from observation of the
exhaust stack outlet.
98
-------
REFERENCES
A-l. Barrett, R. E., S. E. Miller and D. W. Locklin, "Field Investiga-
tion of Emissions from Combustion Equipment for Space Heating,"
EPA R2-73-084a (API Publ. 4180), Environmental Protection Agency,
Research Triangle Park, N. C. , June 1973.
A-2. Hall, R. E., J. H. Wasser, and E. E. Berkau, "A Study of Air
Pollutant Emissions from Residential Heating Systems," EPA-650/
2-74-003, Environmental Protection Agency, Research Triangle
Park, N. C., January 1974.
A-3. Dickerson, R. A., and A. S. Okuda, "Design of an Optimum Dis-
tillate Oil Burner for Control of Pollutant Emissions," EPA-650/
2-74-047, Environmental Protection Agency, Research Triangle
Park, N. C., June 1974.
99
-------
APPENDIX B
DATA TABULATIONS: STOCK LENNOX FURNACE AND PROTOTYPE
OPTIMUM FURNACE EXPERIMENTS
Experimental data are tabulated from laboratory tests of the stock
Lennox 011-140 warm-air furnace prior to its conversion to the prototype
optimum low-emission furnace and from tests of that latter unit in its
initial and subsequently modified configurations. Two adjacent tables
of data are given for each series of tests, one for flue gas concentra-
tion and air pollutant emissions data and the other for operational and
thermal efficiency data. Some of the data are from steady-state exper-
iments but most are from cyclical operation experiments. The tables
are self-explanatory in this regard. Pollutant emissions data from
cyclical testing were averaged over several (usually four) cycles while
performance data were usually averaged over and recorded for each indi-
vidual cycle.
100
-------
Table B-l. CYCLE-AVERAGED POLLUTANT EMISSION DATA: STOCK LENNOX
MODEL 011-140 FURNACE
(4-minute on/8-minute-off cycles)
STOCK LENNOX BURNER
RUN
T- 40
-U.B5 -70-A -
NOZZLE
& fe 4> b
A U fO —
, _ «
46
47
C «
,_l 49
1 5°
!z s.
L '
STB 1C.
RATia
1.49
1.37
1.81
1.62
1.62
1.38
1.51
1.62
1.67
1.37
1.41
1.57
1.68
SMBKE 0NLY
c»s
1
10.4
11.1
8.6
9.4
9.4
11.2
10.3
9.7
9.3
11.3
11.0
9.9
9.1
ae
X
7.4
6.0
10. 1
8.S
8.5
6.1
7.6
8.7
9.0
6.0
6.5
B.I
8.9
ON START.
C0
PPM
17
30
20
16
IS
IB
IS
12
IS
20
IS
IB
PPM
125
IS2
100
III
110
1 17
106
IIS
•114
106
103
113
UHC
PPM
5
4
3
3
3
3
2
2
2
2
2
2
ca
GM/KGM
0.36
0.55'
0.49
0.37
0.33
0.35
0.30
0.28
0.34
0.36
0.29
0.40
GM/KGM
2.672
P.980
P. 639
P. 602
2.577
2.315
2.305
2.692
8.737
2.079
P. 072
P. 557
15 109 P 0. 34 p. 647
IMMEDIATE kECBVERY TB ZERB
OPTIMUM
RUN
, 74
75
76
77
78
3
) 80
( 8'
I Be
1 "
\ 84
BS
86
•7
«•»
STOIC.
RAT1B
1.33
1.27
1.32
1.44
1.67
1.30
1.47
1.54
1.57
1.71
1.38
1.36
1.69
I.SS
C82
X
11.7
IP. 3
11.9
10.8
9.4
12.0
10.6
10.0
9.9
9. 1
11.4
11.4
9.1
10.0
02
X
S.S
4.8
5.5
6.9
9.1
5.3
7.2
7.9
8.2
9.4
6.3
5.9
9.1
8.0
ca
PPM
13
75
17
17
II
10
IS
11
II
II
IS
IS
II
10
NB
PPM
97
89
97
100
110
92
103
107
105
MO
95
103
112
114
IMC
PPM
2
3
2
e
4
1
1
2
1
3
e
e
3
2
UHC
GM/KOM
0.057
0.042
0.049
0.040
0.040
0.030
0.021
0.020
O.OP2
0.019
0. DPI
0.019
0.02.1
GACH.
SMBKF
I.P*
3. n
un*
O.R
n.9
4.0
1.0*
0.9
0.9
P.O
I.P*
I.P*
I.I*
TFG
C
160
147
185
168
168
147
164
185
196
155
I6P
183
194
BURNER
ca
QM/KGM
0.25
0.43
0.32
0.35
0.27
0.19
0.30
0.25
0.2S
O.P8
O.P8
0.27
Ok87
0.23
NB
GM/KGM
1.832
1.617
1.832
P.. 078
2. 6 SO
1.722
2. 165
P. 370
2.382
2.721
1.877
2.008
2.-73T-
2*545
UHC
GM/K8H
0.020
0.033
0.020
0.021
0.050
0.015
O.OIS
0.024
0.016
0.041
0.021
0.021
0*038
0.024-
BACH.
SM8KF
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
O.O
0.0
0.0
0.0
0.0
0.0
TFO
C
IBS
172
186
203
226
179
207
21 1
215
232
189
187
229
217
101
-------
Table B-2. CYCLE-AVERAGED AND PSEUDO-STEADY-STATE EFFICIENCY DATA:
MODEL 011-140 FURNACE
(Listed by individual 4-minute-on/8-minute-off cycles)
STOCK LENNOX
STOCK LENNOX BURNER
o
ro
0.85-70^A NOZZLE
RUN
us.
40
40
40
40
41
41
41
41
4?
4?
4P
4?
43
43
43
43
44
44
44
4b
43
45
45
,
STOIC
RATIB
1.45
1.45
1.45
I.4S
1.33
1.33
1.35
1.38
1.16
1.76
1.77
1.76
1.62
1.19
1.6?
1.50
1.56
1.60
1.60
1.36
1.35
1.33
1.33
GHBSS
EFF.
W.A.
X
RO. 11
HI. IS
BO. 31
78.62
HO. 05
HO. 77
76.79
BO. 43
77.30
78.82
76.65
76.76
DO. 09
74.74
78. O6
84. IP
81.97
HI. 68
77.70
81.37
73.64
73.75
75. 9R
75.68
-
77. 57
90.64
I?.?'
74.71)
81.36
GRBSi
EFF.
F.G.
X
R4.46
84.36
84.59
84.47
86. IZ
R6. 30
R6. 17
86.06
86.16
80.41
80.41
80.34
80.63
80.45
87.88
87.86
83. 13
83.05
82.98
83.20
83.07
83.07
83. 13
86.2O
86. 49
86.37
86. IP
86.28
BURN
TIME
SEC
857
739
940
943
2r>?
P57
P39
Pin
962
958
P4I
241
958
939
241
945
756
239
241
241
P44
?55
258
W.A.
TIME
SEC
391
403
37O
383
401
365
338
33?
344
3b9
317
331
344
334
343
341
464
434
44P
4tl
463
494
455
a
FUEL
« 1
8392
778?
7319
7917
8320
8177
7576
7 (.07
8615
8483
79?4
79P7
8275
7669
7736
7867
8763
8181
8750
849 1
8053
8153
8524
8619
0
U.A.
K J
6723
6315
6980
6224
6720
6P79
6093
5881
6603
651 1
6347
6741
6961
6287
6319
61 12
6453
6033
6268
ftftB 1
6246
7390
7OIO
6467
• ARM
Aid
M3/S
0.61 14
0.5739
0.6165
0.6121
0.6127
0.6147
0.6201
0.6144
0.6441
0.6539
O.610P
R. 6474
0.6434
0.6392
0.6484
0.6363
0.5038
n.5158
0.5191
0. 488 4
0.518?
0.5194
0.5113
0.5080
W.A.
DEL-T
C
P3.9
23.3
23.4
22.6
23.3
73. «
24.7
94.5
P5.3
24. 1
P5.4
P4.8
P6.7
95. 1
94.9
93.9
23.5
27.9
P3.3
73.6
PP.9
96. b
P3.6
P3.8
T(N>
F.G.
C
148
150
146
135
132
133
133
188
I8R
188
185
163
165
IM
161
160
160
160
1 57
132
129
131
135
T
AMP
C
sn
98
3n
30
3O
30
26
P6
P3
93
93
94
P4
P5
P6
26
30
30
30
3O
30
30
31
30
RUN
NO.
6
6
6
6
7
7
7
7
48
48
48
48
48
49
49
49
49
5O
50
5n
sn
51
51
51
51
52
59
5P
5?
STB 1C
RATIO
1.49
1.47
1.46
1.47
I.b7
l.»6
1.53
1.53
I.6P
1.64
1. 67
1 .67
1.6?
1.33
1.33
1.34
1.31
1.45
1.38
1*36
1.38
1.59
1.51
1.53
1.51
I.6S
I.6S
1.65
1.65
GROSS
EFF.
U.A.
X
80.53
HI. 77
80.89
82.05
81.79
70.75
75.76
72. 19
7R. 68
74. 34
75.19
77.05
73.97
77.06
73.37
75.45
76.67
80.95
76.97
77. SO
71.71
74.80
"78 1 *s
85.01
79.42
71. PP
75.69
78.61
76.36
77.22
75.95
74.66
71.83
72.70
73.79
Uj-7pg-A NOZZLE
GFBSS
EFF.
F.G.
X
83.87
33.65
83.91
83.81
83.81
81.53
81. 59
81.87
81.66
80.50
8P.73
80.77
80.60
80. »b
85. Pb
85.07
85. 13
85.30
85. 19
83.52
84.04
84. 1 1
83. R9
RI.4I
81.78
81.87
81.99
81.76
80. IB
80.55
80.35
80.59
30.40
BURN
IIME
SEC
941
245
963
P59
241
245
P63
957
949
264
759
241
74b
963
959
2 »0
26b
2b9
7 4O
240
744
263
258
739
246
269
2 SB
940
U.A.
TIME
SEC
287
3OI
373
308
247
245
300
332
771
317
305
27H
301
307
334
296
316
3O7
31 1
309
323
368
311
337
338
316
3OO
0
(•UEL
K 1
9122
9273
9954
9696
9311
9466
10161
999.9
9350
IOPP6
inoi 6
9320
9267
9948
9790
9015
9H73
9647
8959
9301
10037
9R46
9127
9489
10109
99b9
9258
«
W.A.
KJ
7345
7577
8051
7956
6588
7171
7335
78ie
7030
7354
7408
6715
*992
7677
7926
6981
7870
7PI6
699 6
7616
728O
1591
7740
6969
7807
7548
7149
6730
HARM
AIR
M3/S
0.8597
0.8635
0.8505
0.8545
0.8871
0.9b06
0.8179
0.7895
0.8453
0.3303
0.83O4
0.83O9
0.8319
0.8364
0.8193
0.8996
0.7372
0. 6985
0* 662P
0.7396
0.7330
0.72PI
0. 6460
0.7078
0.6934
O.7I64
0.7181
0.7355
U.A.
DEL-T
C
P5.S
94.8
24.9
95.1
26.1
86.1
Pb.b
2b.3
96.1
93.7
74.9
24. T
23.3
95.3
84.7
94.9
28. 8
89.3
PS • b
98. 1
27.9
P7.7
77.7
86.9
86.3
26.5
86.3
86.0
TCN>
F.G.
C
162
159
156
156
188
188
186
2O2
198
198
2O1
149
152
150
150
165
I6b
164
189
IBB
186
185
8O5
199
2OP
2OO
T
AMP
C
Pb
25
95
?5
23
P3
73
23
P3
25
25
25
24
24
93
23
17
16
19
20
20
9.1
21
P9
21
81
91
21
-------
Table B-2 (concluded). CYCLE-AVERAGED AND PSEUDO-STEADY-STATE EFFICIENCY DATA:
STOCK LENNOX MODEL 011-140 FURNACE
OPTIMUM BURNER
PUN
NB.
74
74
74
74
75
75
75
75
7«.
76
76
76
77
77
77
77
78
7S
78
78
79
79
79
79
80
80
80
80
STB 1C
PAT IB
I.P6
1.31
I.P9
1.3O
1.16
\ . P4
1.24
I.PS
I.P9
I.P6
1.29
I.P7
1.41
1.31
1.40
1.41
1.60
1.59
1.59
1.65
1.25
1.33
I.P7
1.25
1.42
1.43
1.42
1.40
GR8SS
EFF.
W.A.
X
86.33
88. 5P
80. 56
86. 19
85.40
R9.66
89.46
83.25
83.ee
86.40
83.9?
74.71
79.76
75.03
79.33
76.43*
73.70
75.00
74.37
74.87
66.37
73.98
71.74
77.86
71.24
75. PI
83.86
78.93
1H.fi
79.07
79.97
80.61
86.94
62.58
77.53
Grass
EFF.
F.G.
X
82.97
83. 17
8P.9B
83. 14
83.06
R4.17
84.53
84.56
84.71
84.64
83.36
83.53
83.33
83.24
83.36
81. 19
81. IR
RI.PI
81. 12
81.17
78.03
78. 19
77.94
77.99
78.04
83.53
83.97
83.87
R3.82
83.80
81.40
80.55
80. 19
BO. 43
80.64
BURN
TIMF
sec
263
P89
P82
259
263
P79
P74
P5b
PS6
2SS
235
235
260
261
241
P44
235
2S9
260
240
260
P40
238
246
260
260
241
237
U.A.
TIME
SFC
347
382
375
377
403
431
434
4PI
373
347
341
3I)P
308
293
26R
256
P36
P87
274
281
284
272
P63
283
32O
317
226
251
O
FUEL
K 1
9957
10942
10677
9815
9864
10465
10280
9567
9716
96RI
8916
8916
10168
10207
9422
9536
9450
10422
10456
9658
9788
8974
8899
9201
I025P
10256
9482
9324
a
U.A.
K 1
8597
9685
8601
8460
R844
9362
8558
79 6P
8144
762O
7IIP
6690
7771
7522
7O66
7092
627 P
7710
7501
7O37
7316
7525
7024
7201
8199
8267
8243
5835
HARM
AIR
H3/S
O.B945
0.8933
0.8640
0.8996
0.8869
0.8919
O.B843
0.8786
0.884R
0.8619
0.8600
0.8725
O.87P4
0.9027
O.9274
0.9444
0.9048
O.8815
0.8957
0.9225
O.RR38
O.9II6
0.8892
0.8838
0.909R
O.9434
1.2083
0.76SI
W.A.
DEL-T
C
23.5
24.2
PP.6
PI. 2
21.1
P0.7
19.0
18.3
21.0
21.7
20.6
PI. 6
P4.6
24.2
24. P
25.0
24.9
PS. 9
26.0
23.1
P4.8
25. 8
25.5
24.5
23. 9
93.5
25.7
25.8
1.00
TCN1
F.C.
C
190
IB3
IBB
185
170
167
166
163
IBP
191
182
195
208
21 1
208
209
244
243
247
240
182
170
175
177
2O3
PIT
224
221
_wo_A
T
AMR
C
26
26
26
27
28
28
29
P9
30
31
30
30
26
26
P6
PS
PI
22
21
22
24
23
23
23
22
23
18
18
any rt i
RUN
NO.
81
81
81
81
82
8P
82
82
83
83
83
83
84
84
84
84
85
85
85
85
86
86
86
86
B7
87
87
87
|
STB 1C
RATIO
1.51
1.51
I.SO
I.SO
1.53
1.53
I.SI
I.SI
1.65
1.65
1.65
1.65
1.38
1.31
I.P9
1.33
1.33
1.33
1.31
1.31
1.65
1.65
1.65
1.65
I.SO
1.47
I.S9
1.57
GRBSS
EFF.
U.A.
X
69.77
73.90
72.01
68.60
71.07
6H.S6
73.87
67.01
71.70
70.28
71.86
67.86
7O.75
61.40
67.97
73.91
72.08
69.93
71.23
71.79
79.71
75.62
79. P"?
72.71
76.83
71.31
73.28
74. 13
7O. 29
72.25
72.77
69.35
71.88
71.91
71.48
GRB
EF
F.
1
79.
79.
79.
79.
79.
79.
79.
79.
79.
79.
77.
76.
77.
77.
77.
81.
82.
82.
82.
8P.
82.
82.
82.
82.
82.
77.
77.
77.
77.
80.
80.
78.
78.
. *9.
SS
F.
G.
42
83
69
69
66
39
29
17
06
23
IS
84
19
22
10
67
12
56
17
13
83
48
61
82
69
71
SO
75
66
18
IS
47
56
34
BURN
TIME
SEC
258
259
239
237
760
P34
234
241
2S6
P37
234
243
256
236
233
243
240
23S
243
2S6
241
254
253
23S
240
253
2S3
233
U.A.
TIME
SEC
304
296
281
769
265
P33
217
226
25 1
236
254
280
PS 3
25O
268
263
259
250
274
286
272
272
P93
284
283
793
301
283
a
FUEL
KJ
10475
10508
9694
9613
IOPP.9
9197
9194
9476
IO586
9804
9680
10055
10065
9279
9161
9551
9264
9071
9382
9887
9803
10329
10295
9565
9766
10298
IOP95
9481
a
U.A.
K 1
7308
7766
6980
6S94
7013
6794
6161
6794
7607
66S2
6848
6174
7439
6688
64O6
6803
7384
6859
7437
7189
6991
7S69
7631
67P3
7107
7141
7399
6818
UARM
AIR
*3'S
9.8692
9.8779
9.8992
9.8892
9.8976
9.9490
9.929S
9.9653
D. 9802
D.94I6
9. 91 35
9.8484
1.8822
1.9100
1.9016
1.9135
.0782
.0920
.0741
.0476
.0550
.0870
.0738
.0551
.0641
.0696
.0599
.0587
U.A.
DEL-T
C
23.5
25.4
23.5
23.4
es. i
26. 1
26.0
26.5
26.3
25.4
25. 1
22.1
25.4
25.0
22.6
24. 1
22. S
21.4
21.5
2O. 4
20.7
21.7
20.6
19.1
20.1
19.4
19.7
19.4
TCNI
F.G.
C
231
224
227
227
P30
231
235
236
253
2SR
253
2S2
203
POO
I9S
198
187
193
192
IR9
245
248
244
219
PPI
P38
238
T
AMR
C
20
IB
IB
18
18
IT
16
17
17
17
17
18
17
17
17
17
P2
PP
22
73
23
P2
73
24
23
74
P3
23
-------
Table B-3. STEADY-STATE POLLUTANT EMISSION DATA: PROTOTYPE
OPTIMUM FURNACE WITH VARIOUS BURNERS
NOTE: 1.05 ml/s (1.00 gph) Firing Rate
0.57 m Is (1200 cfm) Warm-Air Flowrate (except as noted)
cc.
UJ
i
CO
1
H-4
t-
o.
o
cc.
Ul
1
CQ
5
l— «
•J
1
1
1
£
Of
>*.
p
UJ
a:
z: in
tome
z o
=> to
i. -»
S *""
i
i
RUM
NC.
149
ISO
151
IS2
153
154
ISS
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173'
1 74
175
176
STOIC.
RATIO
1 .20
1 .41
1 .48
1.15
1 .2?
1 .22
1 .10
1 .07
1 .20
1 -27
1 .32
1.39
1 .21
1 .14
1 .49
1 .45
1 .36
1 .29
1 .54
I .51
1 .47
1 .43
1 .28
1 .24
1 -22
1 .35
1 .32
1 .26
C02
*
13.0
11.0
10.4
13.4
12.7
12-6
14. 1
14.5
13-1
12*4
11.9
11.3
12.9
13.8
10.6
10.8
11.4
12. 0
10>2
10-3
10.6
10.9
12.?
12.6
12-8
11.5
11.7
12.4
02
Z
3.7
6.5
7.3
3.0
4.0
4.0
2.1
1 .4
3.7
4.8
5.5
6.4
4.0
2.7
7.5
7.0
6.0
5.1
8.0
7.7
7.2
6-8
4.9
4.3
4.0
5.9
5.5
4.6
CO
PPM
30
42
57
80
25
32
20
100
15
10
10
10
1 1
17
10
10
10
IS
7
7
10
10
20
20
20
18
20
20
NO
PPM
38
35
45
SI
46
48
93
85
97
93
88
84
96
95
86
90
92
92
85
86
87
90
59
57
55
59
59
56
UHC
PPM
5
17
21
21
4
S
1
2
1
2
2
1
2
2
1
0
0
O
0
0
6
0
0
0
0
q
0
0
CO
GM/KGM
0.48
0.81
1.15
1.22
0.42
0.53
0.29
1 .41
0.24
0.17
0.19
0.20
0.19
0.27
O-L'O
0.19
0. 18
0.26
0-1 7
0. 16
0.20
0. 19
0.36
0-33
0.34
0-34
0.35
0.33
NO
GM/KGM
0.653
0.703
0.968
0.846
0.804
0.847
1 .456
1 .288
1 .646
1 .679
1 .662
1 .674
1 .659
1 .531
1 -B32
1 .864
1 .798
1 .699
1 .886
1 .866
1 .841
1 .842
1 .080
1 .012
0.964
1 • 1 43
1 . 122
1 .013
UHC
GM/KGM
0.047
0-182
0.238
0.183
0.037
0.046
0.005
0.015
0.003
0.017
0-018
0.013
0-016
0-016
0-009
0.003
0.003
0.003
0.001
0.001
0.003
0.003
0.003
0.002
0.005
0-005
0-005
0-005
BACH.
SMOKE
2.1
0.2
o.o
0.7
0.5
0>6
0.5
2.5
0.0
o.o
o.o
o.o
o.o
0.0
o.o
0.0
1 -5
4.0
o.o
o.o
0*0
I .5
o.o
0-0
o.o
o.o
o.o
0-0
TFG
C
266
282
291
257
266
254
263
257
272
283
290
293
277
268
282
279
872
264
238
288
283
278
318
31 3
31 3
321
321
31 7
"F.6.
*
80.76
78.69'
77.64
81.76
80.88
81.19
81.95
82.66
80.75
79.95
79.40
78.65
80.60
81.74
78.47
78.92
79.71
80.40
77.69
77.81
78.44
79.10
78.20
78.71
78.85
77.42
77.65
78.44
104
-------
Table B-4. STEADY-STATE EFFICIENCY DATA: PROTOTYPE OPTIMUM FURNACE
WITH VARIOUS BURNERS
ce.
UJ
Of
co
13
Q.
O
ce.
UJ
a:
CD
o
in
i
_i
5
I
I
1
RUN
NC.
149
150
151
152
153
154
155
156
157
158
159
160
161
162
STOIC.
RATIO
1 .20
1 .41
1 .48
1.15
1 .22
1 .22
1 . 10
1 .07
1 .20
I .27
1 .32
1 .39
1 .21
1.14
C02
13
11
10
13
12
12
14
14
13
12
11
1 1
12
13
Z
• 0
• 0
.4
.4
.7
• 6
.1
.5
• 1
.4
.9
.3
.9
.8
nF.G.
80.
78.
77.
81.
80.
81.
81.
82.
80.
79.
79.
76
69
64
76
88
19
95
56
75
95
40
78.65
80.60
81.74
RUN
NO.
163
164
a 165
UJ
§• 166
o 16?
5 168
169
170
171
S ^ 172
z in
a: --N
OCO I *7 O
so E ' /3
Z: 0
g 3 174
»- o
& ~ 175
i
i
! 176
STOIC.
RATIO
I
1
1
1
1
1
1
1
1
1
1
I
1
1
.49
.45
.36
.29
.54
.51
.47
.43
.28
.24
.22
.35
• 32
.26
C02
Z
10.6
10.8
11 .4
12-0
10.2
10.3
10.6
10.9
12.2
12.6
12.8
11.5
11 .7
12.4
"F.G.
%
78.47
78.92
79.71
80.40
77.69
77.81
78.44
79.10
78.20
78.71
78.85
77.42
77.65
78.44
NOTES: 1.05 ml/s (1.00 gph) Firing Rates
0.57 m3/s (1200 cfm) Warm-Air Flowrate (except as noted)
105
-------
Table B-5. CYCLE-AVERAGED POLLUTANT EMISSION DATA: PROTOTYPE OPTIMUM
FURNACE WITH SOME MINOR MODIFICATIONS
RUN
NO.
177
178
< '7Q
a 180
3 m
I 18?
1
•
i '"
:
i
c
199
°i ?oo
W 201
1 20?
P03
PP4
ui
N
§ -206
<
3 ?08
POP
PIP
MB 1C.
HA'IIC
1 . 42
1.31
1 . VI
1.P1
1.16*
1.47
1.30
1.PP
1.16*
1.18*
1.27
1 . 39
1.17*
1. IP*
1 .09 *
1 . ?9
1.4?
I.P3
I.9P
1.13*
CO?
Z
11.1
IP.O
12.4
12.9
13.5
11.1
1P.O
12.6
13.3
13. 1
IP. 1
11.1
13. 1
13.8
14. 1
12.0
11.0
IP. 6
IP. 8
13. b
0?
2
fi.*
b.4
«.«
4.0
3. 1
7.b
b.P
4.0
3.0
3.5
4.7
6.3
3.2
P. 4
1.8
5.P
6.7
4.P
3.7
P. 6
Cf)
Hl'M
10
10
10
1 1
?P
1 1
17
17
20
18
11
13
17
PP
157
13
17
15
13
27
N0
PPM
b4
b4
5?
51
KO
w f\
b4
46
4b
45
46
46
45
46
45
4b
43
44
45
45
UML
PPM
0
0
0
0
1
0
1
0
0
0
0
p
,
1
45
1
1
1
1
1
C3
Rvi/xGM
0. 19
0. 17
0. 17
0.19
0.3b
0. 24
0.31
0.29
0.32
0.30
0.20
0.26
O.P8
0.34
P.P7
0.24
0.34
O.PA
0.??
0.4?
N0
GM/KG^I
1. 1 18
1.007
0.948
0.883
0.856
1. Ib4
1.015
0.805
0.747-
0.766
0.840
0.916
0.756
0.739
0.695
0.835
0.878
0.780
0.768
0.73?
UHC
.GM/KGM
0.001
0.001
0.001
0.001
0.009
0.001
0.01?
0.003
0.003
0.002
0.001
0.001
O.O05
0.013
0.370
0.010
0.01 1
0.01 1
0.009
0.013
BACH.
S^flKE
0.0
0. 0
0.0
0.0
0.0
O.P
0.0
O.O
1.5
0.0
0.0
0.0
0.0
O.P
0.0
0.0
o.n
0.0
0.0
0.0
"IFG
c
288
P79
P74
270
?63
-288
?78
868
263
?66
274
28?
288
285
?81
898
310
?99
296
889
* IN1EPMITTHO Cfl«1PU5TI0N RUMBLING
NOTE: Optimum Burner 1.05 ml/s (1.00 gph) Firing Rate
106
-------
Table B-6. PSEUDO-STEADY-STATE EFFICIENCY DATA: PROTOTYPE OPTIMUM FURNACE WITH
SOME MINOR MODIFICATIONS
RUN STOIC. C02 nF.G.
NO. RATIO * %
RUN STOIC. C02 nF.G.
NO. RATIO Z %
203 1.17 13.1
204 1.12 13.8
205 1.09 14.1
206 1.29 12.0
207 1.42 11.0
208 1.23 12.6
209 1.20 12.8
210 1. 13 13.5
177
178
179
<
8 180
4
V
III
k
L
B
!
181
182
183
EC
5
<
198
. 199
J 200
»•
201
202
1
1
1
1
1
1
1
1
1
1
1
1
.42
.31
.27
.21
• 16
.47
.30
.22
. 16
. 18
.27
.39
1
1
1
1
1
1
1
1
1
1
2
2
2
3
1
2
2
3
13
12
1
1
.1
.0
.4
• 9
.5
. 1
.0
.6
.3
. 1
. 1
. 1
78
79
80
80
81
78
79
.49
.62
.10
.63
.43
.49
.68
NOTE: Optimum Burner, 1.05 ml/s (1.00 gph) Firing Rate-
-------
Table B-7. CYCLE-AVERAGED EMISSION DATA: PROTOTYPE OPTIMUM FURNACE
WITH A REFINED DESIGN OPTIMUM LOW-EMISSION BURNER
C3 NO UHC BACH. TFG
PPI* GM/KGM GM/KGM G*!/KGM SMOKE C
0.68 0*678 0.115 0*1 232
0*70 0*682 0.066 0.0 227
0.68 0*643 0.090 0-0 229
0.76 0.622 0.122 0-0 227
4.48 0.598 0.689 O.I 216
0.70 0.713 0.117 Q.O 224
0.52 0.927 0.098 1.7 243
1*26 0*649 0*076 1*9 229
0.69 0.713 0.053 1.5 236
0*41 0*851 0.123 1*6 235
O.S6 0.634 0.076 0.0 238
0.45 0-647 0.078 0.0 241
0.50 0.679 0.179 0.0 243
0.66 0.710 0.246 0.0 243
0.50 0.689 0.096 0.0 243
0.37 0.671 0.064 0.0 243
34 2500 -22.13 0.512 19.759 2.3 210
0.38 0.684 0.067 0-3 227
1.09 0.599 0.058 0.0 227
0.68 0.621 0.065 0.0 227
0-81 0.676 0-046 0.2 227
0.51 0.655 0.064 0.0 235
0.48 0.774 0.068 0.2 246
0.74 0.721 0.101 1.4 235
0.53 0.834 0.105 0.0 246
0.72 0.599 0-108 0*0 257
RUN
NO.
452
«
" 453
N
O
<
1
°§i *"
° 456
«
457
458
c 462
-------
Table B-7 (concluded). CYCLE-AVERAGED EMISSION DATA: PROTOTYPE
OPTIMUM FURNACE WITH A REFINED DESIGN OPTIMUM LOW-EMISSION BURNER
KIJN)
STOIC-
C3?
39
CC
MS
•JO. HAT 13 • '- PP^ PP"
„ § 48?
1
Ib
1
I
1
L.
**
& 4dl
" a 4R4
485
4BA
487
1
1
1
1
1
1
.
•
.
•
.
t
96
09
19
32
O7
26
1?.
14.
13.
1 1 .
1 4.
12.
4
T
1
7
5
3
4.
1 .
3.
5.
1.
4.
6
9
6
5
4
A
31
42*
42
23
535
17
61
5}
54
57
3A
44
i K 48R
: 1% 489
I 0.1-
9
u
K
u
»•
U
n
a
u
J
1 «
•»
S «
t »
0 's
• u
),i
K
1
i
490
512
513
514
515
516
1 SI7
;*••
S 519
IA
| 520
521
522
523
524
. 525
1
1
'
1
,
1
1
1
|
1
1
1
1
1
1
1
.
-
•
•
.
•
lfE 526 1.
^
•^
"
i
i
j
L. i
"•R
to 5?7
SP8
g 529
t S»
MO
3 S3I
6 ^
533
1
1
1
1
1
1
1
.
-
.
•
.
•
•
19
13
38
19
25
37
24
15
06
40
28
22
22
09
33
14
25
16
12
36
1.7
17
37
32
28
ns
12.
13.
II •
12.
13-
11.
12.
13.
14.
II •
12.
12.
12.
8
3
'
7
8
3
2
3
3
1
1
7
6
14.1
II •
13
12.
13.
13.
II.
12.
13.
II.
II.
II.
14.
8
5
4
2
b
P
7
1
3
S
9
3
3.
a.
6.
3.
5.
6.
4.
3.
1.
6.
4.
4.
4.
1 .
5.
2-
4.
3.
e.
5.
3.
3.
6.
5.
4.
1.
6
5
1
5
0
0
2
0
4
4
9
0
0
8
6
7
4
0
4
n
i
3
1
3
R
1
20
37
45
IOO
00
32
45
91
1131
20
30
50
27
620
31
65
PO
35
no
an
20
3b
45
30
97
10*9
39
40
4R
<3
43
43
43
43
40
46
45
44
39
35
39
38
35
36
38
35
38
38
36
35
36
33
rr
IFC
I,
PPK W/rfCf GM/KGP CK/rfGK NK
a 0.17 I.10? 0-031 J.h 852
•>u 6.15 O. Ill 0. 141 3.3 Sa\
3 0.6b 0-927 C).0?7 3-0 9«9
5 0.4? 1.079 O.ObO 1.7 966
IPS 7.55 Q.55R 1.006 l.b ?35
10 0.30 0.790 0.095 1-0 257
b O*32 O.615 0*045 0*0 238
4 0.57 0.646 0.034 0.0 901
17 0-83 0.952 0.183 1-2 257
5 1.57 O.729 0-045 0-0 218
5 O.SO 0.768 0.047 0.0 238
10 0.60 0.842 0.104 0.0 241
4 0.74 0.760 0*042 0*0 232
4 1.41 0.708 0.035 0.0 254
125 15-93 0.609 1.006 1.5 250
0 0.37 0.918 0.005 0-0 271
0 O.5I 0.829 0.005 0.0 266
0 0.81 0.771 0.005 0-0 261
7 0.45 0.689 0.062 0.0 243
80 8.92 0.539 0.656 0.5 235
5 0.56 0.745 0.055 0.0 252
4 0.98 O.6I9 0.039 0.0 238
4 0.35 0.619 0.038 0.0 938
5 O.54 O.606 0.04? 0.0 229
7 1.19 O.604 0.059 0.0 PP9
9 0.54 0.696 0.093 0.0 244
4 O.3I O.643 0.035 0.0 254
S 0-54 0.639 0.041 0-1 243
8 O.H? 0.7IP O.O83 0.0 257
ft O.52 O.667 O.QSQ 0.0 2b7
A 0.41) 0.669 0.058 0.0 ?57
IPO 19.27 0.493 0.951 I.I
109
-------
Table B-8. CYCLE-AVERAGED AND PSEUDO-STEADY-STATE EFFICIENCY DATA: PROTOTYPE OPTIMUM FURNACE
WITH A REFINED DESIGN OPTIMUM LOW-EMISSION BURNER
RUM
MB.
• If
457
457
452
453
453
453
453
454
454
454
454
45S
455
455
455
496
456
4S6
456
437
457
457
457
STB 1C
RAT IB
1.25
I.f>5
l.?5
1.77
1.22
1.22
1.22
1.2?
1.16
1.16
1.16
i. in
1 . 1 7
1.13
1.13
1.13
1.03
1.04
1.04
1.01
1.25
1.26
I.2S
1.27
crass
EFF.
M.A.
I
go. ft
75.91
81.52
79.21
79.7?
47.95
87.06
81.83
87.95
89.70
83.7?
78.99
73.99
81.31
79.38
80.94
81.49
77.94
74. R8
7R.RI
83.57
83.17
84.39
80.91
93.01
80.27
83. Ol
87.71
88.90
84.97
GKflSS
EFF.
f.G.
I
81.66
81.66
81.84
81.56
81.68
82.30
87.30
82.30
82.30
82.30
9.7. 6S
82.97
87.64
82.49
R?.76
83. 1 6
83.10
83.04
82.94
13.06
84. 06
83.87
84.1 1
84.38
84. II
81.63
RI.92
81.98
8P.07
81.90
BURN
TIME
SEC
224
218
229
241
278
744
244
223
£23
220
230
243
219
227
243
243
223
240
228
220
244
241
273
216
U.A.
TIME
SEC
331
373
346
343
446
418
367
346
371
343
349
373
3S6
376
403
400
329
329
334
300
358
358
341
341
0
FUEL
K 1
8267
8O46
8457
8895
R327
8911
891 1
8144
8325
8213
8586
9071
8204
8477
9074
9O74
8745
9412
8706
8678
8988
8878
8215
7957
0
W.A.
K I
6633
6108
6R90
7tl40
8 lit
7758
7P92
7163
6978
6487
6353
7376
6908
7072
6795
7309
7R28
7347
6981
7215
7369
7205
7073
kAKM
AIR
13/S
0.5916
n.5616
0.5798
0.5748
0.59O4
0. 6008
0.607?
0.6436
0.5868
0.5887
0.5913
0.5X6?
O.S8O7
0.5ROI
0.5541
0.6005
O.606I
O. 6003
0.6033
0.6030
0.5876
O. 6035
0. 6062
h.A.
OEL-T
C
78.9
28.6
29.2
30.4
26.3
76.3
27.9
27.4
27. 1
77.4
26.2
7R.7
26.9
75.8
26. 1
31.4
33.'
31.2
32.9
28.5
29.8
29.8
29.1
1CNI
F.G.
C
717
21?
709
217
705
805
205
2K5
703
703
208
708
204
205
208
707
205
199
199
212
aoi
707
704
T
AMP
C
70
20
70
70
21
71
71
21
70
70
20
70
21
21
71
13
13
13
13
1 6
16
16
16
46? 1.29 75.06 BO.45 225 269 927? 6959 0. 62O7 35.4 232 II
75.06 BO. 45
463 I.OB 10.75 82.87 Ell 274 8695 7021 O.63I5 34.5 718 II
80.75 82.B7
464 1.14 71.51 HI.96 730 765 9475 6776 0.6248 34.8 725 10
71.51 RI.96
465 1.73 74.28 81.09 243 296 10010 7435 0.6744 34.2 224 10
74.28 81.09
466 I.10 81.37 82.30 220 375 9520 7741 0.6106 33.7 775 I?
81.37 87.30
467 1.13 83.7*2 81.91 718 314 9428 7893 O.6204 34.4 229 II
83.72 BI.9I
468 1.17 83.12 81.44 223 329 9650 8021 0.6122 33.9 731 12
83.12 81.44
KUN
NP.
469
470
471
477
473
4*4
475
476
476
476
476
477
477
477
47R
478
47R
47R
479
479
479
479
480
480
4RO
4RO
4RI
481
481
481
SIBIL
I«A1 l«
1.78
1.74
1.70
1.07
1.16
1.07
1. 13
1.09
1.09
1.09
1.08
1. 19
1. 19
1. 19
1. 19
1. 18
1. 19
1.19
I.OR
1.08
1.07
1.06
1.74
1.73
1.71
1.24
1.31
1.30
1.37
1.32
UMISb
t»f.
w. A.
78.34
71. 14
7H. 36
/d. 3b
14. 6J
H4. A3
7«. 47
79.47
K3.34
13.34
17. 59
«!£. 39
77.91
77.91
87. 89
16.73
16. 17
79.71
85. 17
73.43
76. 60
11.96
HI .89
73.70
70.81
75. 13
76.93
74. 14
74.91
76.74
70. II
71.75
73.38
75.29
73.8?
76.57
7O. 99
74. 17
79.77
77.63
81.72
77.97
79.77
GRObl
tff.
t.G.
80.46
10.4(1
R0.73
HO. 13
HI. 05
HI. 05
13.16
K3.1t
17. 4O
87.40
83.16
H3. 1 A
X2.69
HP. 69
83.30
R1. 1 1
R3. 17
83.01
13. 17
81. R4
81.7?
RI.90
81.79
HI. 21
81.29
81. IB
Rl. IR
11.72
82.60
8P.46
87.60
82.76
82.61
80.94
80.9?
80.99
81.77
RO.9I
79.99
80. 16
80.04
80. 16
80.09
PUhN
TIME
?3P
741
74?
773
7IR
771
776
214
77P
229
P4O
7? 4
719
716
719
??R
243
242
718
779
743
741
216
731
243
740
776
741
744
225
w.o.
IIFIE
774
790
313
Tin
79?
79ft
784
317
370
317
332
331
304
300
199
730
227
/37
P09
277
P13
741
233
741
261
756
PS4
77?
3OO
760
O
FUEL
9097
95J5
9574
1171
1638
901?
8953
1669
1970
»876
9719
8982
1779
1653
1664
9070
9626
9516
8716
916?
9716
9636
8661
9263
9741
9624
89R4
9581
97OO
8944
a
H. A.
7176
7471
RI03
7015
7199
7459
6974
76IK
7736
H058
7699
6596
6724
7697
6385
63H7
7737
7375
6530
7031
6812
6913
6521
6838
74b8
M32
7 167
7437
79B7
6974
WAKM
AIR
0.6146
0.62^8
0.6164
0. 6709
0.6179
0. 6091
O. 61 69
0.6139
n. 6277
0.6350
0. 6037
0.5171
0. 6054
n. 6477
0. 6667
0.6471
0.5639
0.6792
O.6667
0.6659
0.6710
0. 6226
0. 61 4B
0.6197
0.6295
0.6086
0.5960
0. 6257
0.6146
0.5938
0.605?
U.A.
DEL-T
36.0
3S.O
35.7
34.3
17.9
.15.7
33.1
33.3
37.7
34.0
37.6
78.3
31.1
33.1
47.2
41.8
39.8
40. 6
39. R
19.2
J9.9
39.7
31.3
38. 3
38.9
38. 1
38.3
37.8
37.9
37.6
TCM1
F.F.
733
73!!
73?
191
713
713
713
707
71 1
710
713
218
711
?n
232
73?
233
733
773
726
775
775
777
230
P3P
231
241
23R
731
735
T
AHP
10
10
10
1 1
13
If
17
16
17
16
15
16
Ib
1 4
10
10
12
12
II
12
1 1
II
16
16
15
16
18
11
IB
IH
NOTES: Burner fired at 1.05 ml/s (1.00 gph) in 4-minute-on/8-minute-off cycles
-------
Table B-8 (concluded). CYCLE-AVERAGED AND PSEUDO-STEADY-STATE EFFICIENCY DATA: PROTOTYPE
OPTIMUM FURNACE WITH A REFINED DESIGN OPTIMUM LOW-EMISSION BURNER
PUN
NO.
41?
4R3
4R4
486
487
488
489
490
5,7
513
514
515
571
571
571
571
522
52?
57?
52?
573
523
573
523
STOIC
PATIO
1.21
,.„?.
1. 14
1.01
1.8?
1. 17
1. 12
1.35
1. 18
1.25
1.33
'•"
1. 19
1.19
1.19
1.19
I.OP
1.0?
1.07
1.06
1.78
I.8R
1.33
crass
tFF.
U.A.
I
77.65
77.65
74. 69
74.69
84.33
84.33
73. 17
67.57
67.57
69.46
69.46
74.69
74.69
74.9?
74.9?
65.61
65. M
14.31
84.31
78.31
7S.3I
R4.03
R4.03
R3. 7R
83.28
74.68
85.58
76.4?
14.93
80.40
80.05
80.47
R0.95
90.6?
R3. 15
79.92
HO. 3?
77.83
79. H?
79.47
GhOSS
FfF.
t
10. R?
RO.R2
83.00
83.00
RI.52
81.52
79.70
83.48
83.4?
80.44
80.44
81.73
81.73
82. OR
87.08
79. 5R
79.51
83.02
13.07
81. 17
81. 17
80.70
80.70
RI.R6
81. HO
81.49
81.49
RI.60
RI.49
81.5?
83.66
S3. 5?
83.84
13.30
83.43
80.57
80.59
RO. 34
10.48
RO. 49
BURN
TI1E
SEC
743
774
7?7
219
244
74,
224
230
2IR
229
244
2P?
245
219
219
775
746
84?
723
740
723
219
U.A.
1IHF
51 1
754
709
714
20?
24?
269
229
717
254
236
280
750
844
235
763
298
290
775
7R3
778
295
769
765
II
FUEL
KJ
9738
8968
9094
R67?
9865
9747
9056
9296
R487
8910
9500
8658
9479
8478
8428
8659
9492
933i
8602
8379
9434
9359
R696
8540
0
U.A.
KJ
7O74
6698
7668
58 6O
685?
7279
6785
6099
7156
6971
79D?
7205
7O4I
7713
6441
7354
7598
755R
6963
7548
7540
1517
6768
6817
UARM
AIR
13/S
0.60O4
0. 6636
0.691 1
O. 6??7
0. 6344
0.6280
0.6159
0. 629 1
0.6131
0.6162
O. 6345
0.6153
O. 6399
0.613?
0. A34R
0. 5929
n. 5906
0.5971
P. 6044
0. 5R14
0.6313
0. 6181
O. 5946
0.5873
0. 6024
h.A.
DEL-1
C
39.4
41.0
44. 1
38.9
38. 3
37.3
40.1
39. P
3R.R
39.6
39.4
3R.3
39. 1
39.6
39.3
40.7
36.3
36.6
36.6
36.0
37.3
36.5
36.4
36.3
T(N)
F.G.
C
236
7.77
834
25 1
272
244
774
727
245
199
??„
721
71 1
276
?? 6
771
776
213
216
713
714
230
730
730
732
T
AMP
C
15
13
14
12
12
13
IP
IP
18
17
,,
?n
17
17
17
17
21
71
71
70
21
21
71
21
RUN
ND.
524
S24
524
584
525
525
525
575
526
576
576
576
527
527
527
527
528
528
528
<2B
579
Sf9
5?9
Si"9
530
531
53?
S33
STB 1C
KA110
1. 10
1. 18
1.09
1.09
1.21
1.21
1.71
i.ai
1.13
1.13
1.13
1.13
1.10
1.10
1.10
1.10
1.38
1.35
1.33
1.33
1. 14
1.14
1. 14
1. 13
1.33
1.31
1.77
1.0?
GPCSS
EFF.
W.A.
I
17.42
83.71
78.48
69. RO
78.59
75.43
94. II
85.08
82.98
84.40
84.43
RI.23
83.82
85.42
83.72
86.27
85.18
88.99
81.92
85.59
70.19
79.21
77.57
76.40
75.84
7R. 60
qo. 17
BO. 91
76.73
7«». IP
16.16
76. 16
75. 6R
75. 6R
75.91
75.91
79.41
79.41
GROSS
FIF.
F.G.
X
87.73
R7.54
87.87
8S.9R
12. 7R
81.45
11.65
81. RR
8I.9R
81.74
83.19
83.34
83. 19
83.32
83.26
83.42
93.78
83.B9
B4.0P
83.76
80. R7
80.78
81.05
80.87
8O.B9
R7. O1
RI.94
R7.PR
82.04
1?. 0?
10. PI
no. 01
80. 12
80. 1?
RO. 37
80.37
BURN
1IME
*EC
240
741
725
219
227
740
242
224
2Z7
241
240
223
224
843
243
BI9
24?
244
823
818
779
241
74?
775
771
740
243
723
k.A.
line
SEC
298
310
269
263
193
232
232
209
254
750
769
257
212
302
364
355
217
212
185
112
775
?R4
313
7R6
778
7RI
772
XK7
0
FUEL
K 1
9365
9401
8711
8543
8162
9266
9350
8657
8784
9386
9281
8631
8616
9429
9432
8512
9268
9344
8540
8348
HO |0
9319
9434
R17I
BR80
9344
9467
R6S?
e
W. A.
KJ
7719
7870
6882
5963
6609
R12I
7955
7 183
7416
7575
7784
1318
1485
R03I
8393
6973
6505
1401
6624
6378
7010
75P7
7633
67.10
(•763
7012
1IR6
6894
HARM
AIR
M3/&
0.5968
0. 608 1
0.6151
O. 6O90
0.8388
O.B8?4
0.8894
O.H17O
0.8343
O. 8392
0.8260
0.8315
0.8746
0.8351
0.1719
0.7600
0.8154
0.9624
1.0305
1.0614
0.6111
n. (717
0.6C91
0.5913
0.6173
0. 6252
O. 6377
0.6561
W.A.
OEL-l
C
36.9
35.5
35.3
31.6
34.7
36.7
32. H
33.3
29.7
30.1
29.8
29.1
2B.3
27. 1
25.4
2?.0
29.1
30.8
29.5
29.5
35.4
36.7
34. 1
33.9
33.5
34.2
35.7
31.1
1CNI
F.G.
C
217
218
716
214
721
EI7
213
21?
702
200
PO?
201
203
197
194
191
216
218
216
718
773
776
27?
775
737
738
237
T
AME
C
28
2^
7?
2?
7?
28
73
24
76
86
76
78
28
31
38
35
21
21
21
77
19
20
71
71
19
18
20
18
-------
APPENDIX C
DATA TABULATIONS: EXPERIMENTAL FURNACE TESTS
WITH HEAT EXCHANGER MODIFICATIONS
Experimental data are tabulated from several series of tests in which
the prototype optimum and Williamson furnaces' heat exchangers were mod-
ified substantially. Emphasis in these experiments was on pollutant
emissions reduction so only emissions-related data were measured and
recorded. Both steady-state and cyclical tests were performed as de-
noted in the table titles.
One series of tests (Table C-3) involved forced recirculation of the
flue gases back into the combustion chamber, so there is a table column
labeled "Recirc. Ratio, %." That parameter was calculated from the
formula given below the table title where SR was estimated from cal-
mix
culated 0^, CQ~, and CO concentrations in a hypothetical mixture of the
recirculated gases and the fresh reactants supplied to the burner. The
recirculation ratio is defined as the relative mass of recirculated
burned gases (at stoichiometric conditions) to the mass of unburned air
(including unburned air in the recirculated gases); it indicates the
relative dilution of combustion air with inert gaseous diluents.
112
-------
Table C-l. CYCLE-AVERAGED POLLUTANT EMISSION DATA: FINNED-COMBUSTOR
WILLIAMSON FURNACE CONFIGURATION
RUN
M0.
184
185
186
| 187
Jj 188
«
1
189
190
21 1
212
213
ft
I 214
215
216
ST0IC.
RATI0
1.31
1 .26
1 .21
1.13
1 . 46
1 .42
1 .23
1.08
1.15
1.23
1 .42
1.36
1. 14
C02
Z
11.9
12.4
12.7
13.5
10.6
10.8
12.6
14. 1
13.4
12.5
10.9
1 1 .3
13.4
02
X
5.4
4.6
3.9
2.6
7. 0
6.6
4. 1
1 . 7
3.0
4.2
6.6
5.9
2.7
C0
PPM
20
20
20
22
20
20
20
40
20
20
20
20
25
N0
PPM
67
68
68
69
63
63
67
50
50
49
45
45
52
UHC
PPM
1
0
0
0
0
0
0
0
0
0
0
0
0
C0
Gi1/KGM
0.35
0.33
0.32
0.34
0.39
0.38
0.32
0.53
0.31
0.33
0.38
0.36
0.38
N0
GM/KGM
1.263
1.220
1. 182
1.111
1.331
1.299
1 . 174
0.772
0.823
0.864
0.924
0.884
0.849
UHC
GM/KGM
0.006
0.005
0.004
0.003
0.001
0.002
0.003
0.001
0.001
0.001
0.001
0.001
0.001
BACH .
SMOKE
0.0
0.0
0.0
1.5
0.0
0.0
0.0
4.0
2.5
2.0
1. 0
2.0
3.0
TFG
C
263
257
254
248
271
270
258
238
249
257
268
274
257
NOTES: Optimum Buriwr Find at 1.0S rnt/i (1.00 gph) in 4 min. on/8 min. off cycta.
-------
Table C-2. STEADY-STATE POLLUTANT EMISSIONS DATA: COIL-COOLED PROTOTYPE
FURNACE WITH VARIOUS COMBUSTOR LENGTHS AND COOLING FLUIDS
RUN STOIC.
NO- RATIO
I
1
«3
|
e
i
i
i
1
"j
i
K
£
1
1
1
H
1
E
ID
^
221
?22
223
224
225
326
227
220
22V
229
329
230
231
732
!>33
234
235
236
237
?3B
239
240
241
1 .1 1
1.09
1 .07
1 .05
1 .23
1 .90
1-13
1 -Ib
1 .10
1 .12
1.13
1 .03
1 -06
1 .16
1 .24
1 .12
1 -25
1 • 16
1 .08
1 .03
1 .09
1.13
1.19
car.
X
13.7
13.9
14.1
1 4.4
12.4
12.6
1 J.5
13.3
13-d
13-3
13.4
14. R
I 4.4
13-3
13.9
11. R
12.?
13-3
13.9
14. »
1 4.9
13.3
12.7
02
X
2.1
1 .8
1 .5
1 .0
4.2
3.7
?-4
2.V
2.0
?.4
j>.6
0.6
1 .3
3.0
4.R
?.3
4.3
3.1
1 .*
0.7
0.4
2.5
3-5
CO
PPM
25
30
47
1 10
20
20
31
*a
10
10
1 1
no
>0
1 1
\9
"
60
30
30
12*
-I6QO
20
16
PPM
32
31
31
30
31
30
30
31
46
46
45
4S
45
43
46
45
45
43
44
43
41
46
43
UHC
PPM
0
0
1
6
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
2no
IJ
0
CO
Gr/xGM
0.37
0.43
0.68
1.32
0-33
0.32
0.33
0-31
0-16
0.15
0.11
1 .09
0.28
0.18
0-?l
0-18
1.00
0-46
0-43
1 .71
-9\ .36
0.30
0.30
NO
CM/KGC
0.50?
0.437
0.473
0.434
0-S45
0.526
0.491
0.504
0-726
0.739
0.733
0-658
0.687
0.747
O.RI9
0.721
O.BI3
0-70D
0.683
0.635
0.604
0.736
0-771
UHC
GM/KGK
0.001
0.003
0.008
0.048
0.001
0.002
0.003
0.003
o.oo?
O.flO?
0-OOP
0-016
0.00?
0.00?
O.OOI
n.oo"
-.000
-.000
-.000
-.000
1.540
-.000
-.000
BACH.
SPOKE
0.0
0.0
0.0
0-0
0.0
0.0
0.0
0.0
0.0
0.0
0-0
2.0
q.O
O.O
0.0
O.O
0-0
0.0
0.0
0.0
3.0
0.0
q.O
TFG
C
204
202
202
199
21 3
210
207
203
621
•g
j
£
&
|
«s
621 *
?5
621
621
621
621
338
377
377
382
377
404
399
* Ti
il
31
11
> BE
I
A f
il
J-i
"l
< K
RUN
NO.
242
243
244
245
246
?47
248
249
250
251
752
253
254
255
256
257
25R
259
260
261
263
264
265
STOIC.
RATIO
1 .05
1 -08
1 .10
1.13
1.19
1 .29
1.23
i .or
1.09
I.3S
1 .19
1.31
1 .26
1 • ?0
1.14
1 .07
1.15
1.16
1.10
1.06
1.32
1-27
1.13
C02
Z
14.5
14.1
13.9
13.6
12.9
12.0
12-4
I4.|
13-9
1 1 .4
12.9
11.5
12.0
12.5
13.3
13.9
1 4.6
13. 1
13.8
14.1
1 1-5
12.0
13.3
02
X
1.0
1.7
2.1
2.3
3.5
5.0
4.2
1.7
1.9
5.8
3.5
5.1
4.6
3.7
2.6
1.5
3.?
3.0
1.9
1.2
5.4
4.7
2-5
CO
PPK
177
40
30
2O
18
15
II
41)
50
20
2O
20
2O
?O
30
1 10
20
s>\
50
139
17
20
II
NO
PPM
44
43
45
4S
43
48
49
41
31
31
31
37
36
n
33
31
35
47
30
49
50
50
50
UHC
PPM
6
1
0
0
0
0
0
1
,
0
0
1
1
1
,
3
1
0
1
4
0
0
o
CO NO UHC
GP/KGM GH/KGM GM/KGK
2-46 0-663 0.048
O.S7
0.44
0.30
0.30
O.P7
0.30
0.59
0.72
0-38
n.3l
0.35
0-34
0.32
0.45
,.56
0.32
0.34
0-72
1 .93
0.32
0.34
O.IR
0-671
0-7O6
0-721
0.736
0.886
0.865
0.634
0.482
O.59B
0.523
0.705
0.664
0.615
0.540
0.473
0-571
O.737
O.780
O.74I
O.948
O.908
0-807
0.005
0.003
0-002
0.001
0.001
0.001
0.007
O.OOB
0.001
O-001
0.006
0.007
O.OOd
0.010
0.02B
0-009
0.001
0.005
0.029
0.002
0.002
O.O04
BACH.
SPOKE
3.O
2.5
1 .5
1 .3
1 .5
0-0
1 .0
2-5
0.0
0.0
c.o
0.0
O.O
0.0
0-0
0.0
(1.0
0.0
1 *3
3.0
0-0
0.0
0.0
TFG
C
254
257
26O
264
260
?S7
234
f.3
?4I
23B
243
?Ud
302
3IO
306
29V
31 1
3»6
3R8
391
410
416
416
NOTE: Optimum burner fired at 1.05 ml/s (1.00 gph)
-------
Table C-2 (concluded). STEADY-STATE POLLUTANT EMISSIONS DATA: COIL-COOLED PROTOTYPE
FURNACE WITH VARIOUS COMBUSTOR LENGTHS AND COOLING FLUIDS
VSl
f.
i]
•I
ftfll
• 1
?
J
•
«J
1
*
f1
§
i
i
*
i
I
i
•
\
i»UN
N3.
266
268
869
270
271
272
813
274
275
876
277
273
279
88C
231
888
383
STOIC.
RAT 1C
I.IS
0.28
1 .28
1.34
1.26
1.32
1 .36
1 >2b
1 .22
1 .18
1 .14
1 .09
1 .10
1.83
1-19
I.IS
1.32
1.29
C02
5
13-3
18.6
ie.o
1 1*4
18.0
11.7
1 1.3
12-8
l?.4
13.0
13.3
13.8
13.6
12. 4
IB. 7
14.7
II. S
11.7
38
X
8.9
4.0
4.6
S.7
4.6
5.4
5.9
4.S
4.0
3.3
8.7
1.6
2*0
4. 1
3-5
3-1
S.4
S.O
CO
PPf
45
30
21
30
21
25
31
87
33
41
80
6?0
4S6
61
65
90
67
60
MS
PPK
59
59
59
60
59
31
31
30
30
30
88
21
81
8S
86
23
83
81
UHC
PP»
O
0
0
o
1
1
0
o
o
8
54
3O
,
1
2
8
2
CO
GK/KGf
0.68
0.43
0.37
0.54
0.37
0*44
0-59
0.47
0.49
0.6S
1 .21
3.93
6.A4
1.01
1.02
1.36
1.80
1.03
MO
GP/KQP
0-970
1 .038
1.037
1 .157
1.069
0.593
0.6O4
0-545
0.52?
0.510
0.462
0-324
0.327
0-450
0-440
0-3H2
0. 444
0-397
UHC
GK/KGK
O.OOA
0.003
0-004
0.001
0.001
0.010
0.010
o.oos
0.004
n.nq?
0.018
0. 444
0.?49
0.014
O.OI3
0-021
0.024
o.oie
RACH.
SKQKE
3.0
8.0
1 .0
o.o
o.o
o.o
0.0
0.0
o.o
o.o
o.o
o.o
o.o
o.o
o.o
o.o
o.o
o.o
TFG
C
391
407
41 3
421
260 1
2!>2 1
<
271 |
1
I
5
i
f
3
26B £ 6
-'" | §
ID =
259 $ §
T
359 J
271
274
'74
974
2S7
271
*
I
5
n
RUM
.MC.
284
88 S
286
887
288
289
290
291
292
293
294
295
896
297
298
299
300
301
302
303
"STOIC
RATIO
I-O5
l-ll
1.23
1.88
1.31
1.17
1 .18
1 .1?
1.06
1 -03
1.93
1-13
1.30
1 .12
1 .13
1 .30
1.16
1 .10
I.IO
1.34
• C08
X
14.3
13.4
12.2
11.8
11.5
12.7
13.1
13.6
14.]
14. R
13.9
13.5
II. 8
13.8
13. S
11.7
14.7
13.0
13-8
11.3
0?
m
0.9
4.0
4.8
5.2
3-1
3.4
2.4
1 .3
0-6
4.6
2-5
5.1
8.3
2.5
5.0
3.4
2.0
2-0
5.6
CO
PPK
198
20
IR
17
17
16
._*
—
—
—
—
—
—
-
—
~~
-
120
40
.113
PPr
21
IR
17
80
19
SO
40
40
35
25
21
19
21
19
19
22
21
19
24
25
UHC CO NO UHC
PPK GP/KGM Gr/KGM GK/MGr
25 2.74 0.311
1 0.31 0.298
1 0.31 0.318
1 0.31 n.374
0 0.31 0.373
0 0.26 0*339
1 - 0.683
1 — O.A49
8 ~ 0.53S
80 — 0.366
1 - 0.367
? — 0.119
2 — 0-388
2 - 0-316
1 — 0.319
0 — 0.40*
0 — 0.34S
6 ~ 0.312
3 1.74 0.375
1 0.71 0.47V
O.I9K
O.OOS
0.007
O.OOA
O.OOS
0.004
0.007
0.007
O'OI 4
0.622
0.006
0.018
0.018
0.020
0.01 1
0-OO3
0-OO3
0.050
0.025
O.OIQ
•«CH. TFG
SKCKF r
1 *0
O.5
0.0
O.O
0-0
0.0
0*0
0- 0
0. 0
1 .0
0-3
0.0
0.0
o.n
0.0
0.0
0.0
0.0
o.o
o.o
1 63
171
179
1 70
1 77
177
?07
?Q4
P02
?02
?07
? 1 0
20?
?07
904
196
202
199
IBB
207
•'•CO meter was inoperative during
runs 290 through 301.
-------
Table C-3. STEADY-STATE POLLUTANT EMISSION DATA: OPTIMUM FURNACE
WITH FLUE GAS RECIRCULATION
I00«(14 49»1.0) /kgi of OurntdiJJ "\
MM
304
309
306
307
308
309
310
•311
312
313
314
319
316
317
318
319
320
321
329
323
324
329
326
327
328
329
RUN
NO.
340
341
342
343
344
345
346
STOIC.
RATIO
1.30
1.41
1.10
1.21
l.ll
1.08
1.10
1.32
1.16
1.36
1.28
1.22
l.ll
1.22
1.09
1.21
1.25
1.21
1.29
1.32
1.29
1.29
1.29
l.ll
I.IT
1.22
STOIC.
RATIO
1.20
I. fO
1.19
1.19
1.19
1.19
1.21
cat
S
11.9
II. 0
11.9
12.6
13.7
14.1
13.8
14.3
13.1
11.3
12.0
12.4
13.8
12. S
13.8
12.6
12.1
12.6
11.9
11.6
11.9
II. 8
11.9
13.6
12.8
12.4
C02
I
12.7
12.7
12.7
12-8
12. R
12-7
12.4
HKtl
02
*
S.I
6.S
S.I
3.8
2.2
1.6
2.1
6.7
3.0
S.9
4.9
4.0
2.2
4.0
1.9
3.9
4.4
3.8
S.O
S.I
S.O
S.O
S.O
2.1
3.2
3.9
92
X
3.6
3.6
3.S
3.4
3.S
3.S
3.8
ir* iwi
CO
Pftl
17
120
60
SO
38
2S4
130
20
IS
IB
20
60
SO
21
693
70
81
IS
IS
18
20
45
80
40
20
18
CO
PPM
IT
14
20
35
7S
80
30
(s
N8
pra
41
28
31
35
45
43
35
46
SO
54
AS
41
52
52
40
40
38
SO
S4
S4
49
44
40
SO
SO
SO
NO
PPM
51
47
45
41
16
IS
38
w
UIC
PPM
1
7
1
2
1
30
II
1
0
0
0
0
1
0
120
0
1
1
a
4
90
0
0
UMC
PPK
0
0
1
1
4
6
1
.0)14.49
C6
CI/KCH
0.31
2.2S
I.OS
0.80
O.S7
3.62
1.90
0.35
0.2S
0.14
0.16
0.97
0.73
0.36
10.02
1.13
1.36
0.24
0.26
0.33
0.34
0.77
1.37
0.59
0.31
0.31
CO
r,M/KGn
o.?e
0.10
0.31
0.55
l.lft
1.26
0.48
1 (19 of Unturned Air 1
NO
C.1/UCM
0.762
O.S6S
O.S73
0.60I
0.7 IB
0.667
O.S63
0.869
0.826
I. 047
0.83I
0.7I7
0.820
0.904
0.632
0.704
0.682
0.863
I. 009
I. 029
0.90S
0.822
0.7SO
0.796
0.844
0.879
1C
G»/KCM
O.H7T
0.816
0.170
O.69S
0.4M
O.A07
0*661
UIC
G.1/KC1
0.006
0.030
0.030
0.022
0.013
0.244
0.092
0.006
0.001
0.001
0.001
0.003
0.013
0.001
0.991
0.001
0.010
0.010
0.024
0.035
0.752
UMC
CH/KCM
O.OOI
0.001
0.005
O.OI 1
0.040
0.054
0.014
BACH.
S.IOKE
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.s
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
I.S
0.0
0.0
HACK,
SHOKC
0.0
0.0
0.0
0.0
0.0
0.0
0.0
TFO
c
293
338
338
329
277
274
274
279
263
302
321
329
271
307
302
327
2S2
291
311
321
335
333
271
277
TFC
C
?T1
316
318
321
38 4
321
321
nscinc
TATIJ
t
0.0
11.1
38.0
45. S
0.0
0.0
56.7
0.0
0.0
18.8
11. 1
40. S
0.0
32.9
50.1
12.8
39.6
0.0
0.0
21.9
31.4
36.7
37.9
0.0
0.0
0.0
RCCIRC
RATIO
t
0.0
34.2
36.1
44.4
69.6
45. 8
39.8
T
RFC
C
171
171
171
168
60
I3S
177
52
IS7
IS7
171
91
IS2
132
199
T
RFC
C
I3S
74
74
74
141
146
116
-------
Table C-4. STEADY-STATE COMBUSTION GAS COMPOSITION DATA
AT TOO LOCATIONS IN THE PROTOTYPE FURNACE
LOCATION C = ENTRANCE TO SECONDARY PORTION OF
THE MAIN HEAT EXCHANGER
LOCATION B = TOP OF CENTRAL CYLINDRICAL DOME
CO NO UHC BACH. TFG
PPM GM/KGM GM/KGM GM/KGM SMOKE C
0*18 0.867 0.003 0*0 277
0.19 0.869 0.003
0.58 0.787 0.008 1.0 271
3.81 0.775 0.080
0*30 0*827 0.001 0*0 274
0.88 0.795 0.008
0.21 0.999 0.001 0.0 277
0.25 0.979 0.001
0.29 1.121 0.001 0.0 291
0.27 1.074 0.001
0.28 0.845 O.OOS 1.0 277
0*60 0*827 0.005
RUN STOIC.
NO. RATIO
330
C
331
C
332
C
333
C
334
C
335
C
1.20
1.21
1.09
1*06
1.14
l.ll
1.29
1.25
1.43
1.37
1.16
1.13
C02
Z
12.7
12.7
13.9
14.3
13*3
13.7
11.9
12.2
10.8
11. 1
13.1
13.4
02
Z
3*7
3.8
1.9
1.2
2.7
2.1
5.0
4.5
6.7
6.0
3.1
2.5
CO
PPM
10
II
40
272
20
60
II
15
15
IS
17
40
NO
PPM
SO
50
50
51
51
SO
54
54
54
54
51
51
UHC
PPM
0
0
1
10
0
1
0
0
0
0
1
1
336
B
337
B
338
B
339
B
1
1
1
1
1
1
t
1
• 16
• 14
.11
.07
.07
• 03
.43
.38
13.
13*
13.
14.
1
3
7
0
14.1
14.
10.
II.
5
B
1
3*
2*
2.
1.
1.
o.
6.
6.
1
6
1
5
4
7
7
1
18
18
28
50
620
711
IS
IS
SO
50
51
49
50
46
SI
SO
1
1
1
1
50
62
0
0
I 0.29 0*837 O.OOS 0.0 268
0.29 0.809 O.OOS
0.42 0*803 0.007 0.0 271
0.71 0.751 0.008
8.76 0.760 0.403 l>5 270
9.72 0*681 0.484
0.29 1.046 0*002 0.0 288
0.29 0.998 0.002
117
-------
Table C-5. STEADY-STATE POLLUTANT EMISSION DATA: PROTOTYPE FURNACE
WITH VARIOUS CONFIGURATIONS OF INTERNAL COOLING COILS AND BAFFLES
RUN
HO.
347
348
349
350
351
392
353
394
395
356
397
358
399
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
stoic-
BSIIU
1.40
1.25
l.l«
1 .13
1 >l 1
1.31
1 .3?
1 -21
1.18
1.46
1.48
1.26
I-IB
1.37
1.43
1.33
1.24
1.26
1.43
1.29
1.38
1.30
1.23
1.17
1.30
1.38
1.17
1-34
1-19
1.15
1.29
1.23
1.23
1.13
1.16
1.35
C32
1 1 .n
12-2
12.7
13.4
13.5
1 1.7
11.6
12.6
13.6
10.4
10.4
12.2
13-1
11.3
10.8
11.6
12.4
12.0
10.8
11.9
ll.l
11.6
12.2
12.9
11.7
II. 0
12.7
11.5
12.7
13.3
11.9
12.4
12*3
13.3
12.9
11-3
02
I
4.4
4.5
3.9
2.6
2.1
5.3
5.4
3.8
2.4
7.0
1.2
4.6
3.4
6.0
6.7
9.9
4.4
4.6
6.7
9.0
6.1
5.1
4.1
3.1
S.I
6.0
3.1
5.6
3.5
2.8
4.9
4.1
4.1
2.8
3.4
5.8
C3
PPM
20
20
20
50
130
17
20
20
120
20
21
30
75
21
30
30
40
35
20
30
15
IS
20
120
20
20
27
20
75
157
20
30
30
120
55
20
NO
PPH
46
43
45
43
41
40
43
41
41
45
49
50
47
47
40
42
45
45
46
47
46
46
46
45
50
50
49
33
31
31
38
33
31
31
32
33
UHC
PPM
0
0
1
2
7
0
0
0
4
0
0
0
3
1
0
0
0
0
0
0
0
0
0
21
0
2
1
0
6
7
0
0
0
6
1
0
CO
0.37
0.33
0.32
0.75
1.90
0.31
0.35
0.34
1.78
0.41
0.43
0.50
1.17
0.40
0.57
0.53
0.66
0.59
0.40
0.51
0.28
0.26
0.31
1.85
0.35
0.37
0.41
0.37
1.18
2.40
0.34
0.49
0.49
1-82
0.86
0.38
10
GN/KGN
0.932
0.779
0.771
0.702
0.648
0-763
0.814
0.717
0.656
0.955
1.043
0.901
0.802
0.934
0.814
0.808
0.807
0.81*
0.950
0.879
0.91 7
0.864
0.806
0.746
0.932
0.998
O.BIB
0*638
0.524
0.50S
0.702
0.585
0.543
0.512
0.537
0.638
UMC
Gn/KGM
0.001
0.003
0.005
0.021
0.059
0.001
0.001
0.003
0.038
0.001
0.001
0-003
0.024
0-006
0.003
0.003
0.003
0.003
0.001
0.003
0.001
0.001
0.001
0.185
0.003
0.021
0.005
0.001
0.054
0.065
0.001
0.003
0.003
0.052
0.009
0-001
B«CH.
SKOKC
0.0
0.0
0.0
0.0
I.S
0.0
0.0
0.0
1.0
0.0
0.0
0.0
1 .5
0.0
1.5
2.5
3.0
2.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.5
1.0
0.0
TFG
c
274
263
257
252
249
260
271
263
256
279
266
249
241
259
267
260
252
249
264
252
260
254
246
241
257
264
246
214
207
200
2O 6
204
207
204
207
221
NOTE: Optimum burner fired at 1.05 ml/s (1.00 gph)
118
-------
Table C-5 (continued). STEADY-STATE POLLUTANT EMISSION DATA: PROTOTYPE
FURNACE WITH VARIOUS CONFIGURATIONS OF INTERNAL
COOLING COILS AND BAFFLES
RUN
NO.
383
UJ
§ 385
o 386
-i
d
^
387
5 388
J h-
£ •" 389
1 390
I 3"
u; 392
| 393
£ 394
Eg 3"
S I 396
d J
i i» 397
1 398
3 399
UJ
*° 400
401
STOIC.
RATIO
1.37
1.30
1.23
1.19
1.19
1.30
1.24
1.35
1.40
1.30
1.22
1.16
1.10
1.16
1.21
1.27
1.30
1.22
1.15
C02
S
II. 1
11.7
12.2
12.7
12.7
11.7
12.3
11.3
II. 0
11.7
12.4
13.1
13.6
11.5
12.6
11.9
11.7
12.4
13.1
02
t
A.O
S.I
4.1
3.5
3.5
5.0
4.2
5.8
6.4
5.0
4.0
3.0
2.0
2.6
3.8
4.7
5.0
4.0
2.9
CO
PPM
20
20
38
80
55
30
35
28
30
20
35
100
tliOO
100
80
67
20
35
115
NO
PPM
39
40
38
35
36
38
40
40
35
35
31
30
27
33
31
31
31
34
31
UHC
PPM
0
0
0
3
2
0
0
0
0
0
0
7
300
6
4
3
0
0
6
CO
G1/KGM
0.38
0.36
0.64
1.26
0.87
0.53
0.57
0.52
0.56
0.36
0.57
1.53
123.18
1.53
1.28
1.15
0.36
0.57
1.75
NO
GM/KGM
0.780
0.747
0.672
0.607
0.624
0.717
0.717
0.788
0.721
0.645
0.548
0.502
0.421
0*549
0.541
0.562
0.582
0.599
0.508
UHC
GM/KGM
0.001
0.001
0.003
0.027
0.018
0.001
0.001
0.001
0.001
0.001
0.001
0.066
2.484
0.052
0.041
0.029
0.001
0.001
0.052
BACH.
SMOKE
0.0
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
o.o
o.o
1.5
o.o
o.o
o.o
1.0
o.o
o.o
TFG
c
241
238
229
227
229
241
238
248
23S
227
224
218
211
216
221
227
232
229
224
119
-------
Table C-5 (continued). STEADY-STATE POLLUTANT EMISSION DATA: PROTOTYPE
FURNACE WITH VARIOUS CONFIGURATIONS OF INTERNAL
COOLING COILS AND BAFFLES
RUN
NO.
„ 402
a
ii_
i
o 4Q3
o 404
c
j
i
406
407
408
409
UJ
A.
U.
S
410
53 «»'
AJ UJ
2 S *lfi
3
Z
UJ
_
413
ftj
i
o
u
UJ
3 a
414
415
416
' 5 417
418
419
420
421
J
o 423
424
425
STOIC.
RATIO
1.44
1.28
1 .24
1.19
1.26
1.19
1 .35
1.38
1.38
1.26
1.21
1.14
1.37
1.28
1.22
1.18
1.13
1.36
1.33
I. 25
1.19
1-16
1 .13
1.33
C02
X
10.6
11.9
12.1
12.6
12.1
12.6
II. 1
II. 0
11.0
11.9
12.4
13.0
11.2
11.9
12.5
12.9
13.4
11.3
11.5
12.2
12.7
13.1
13.4
11.5
02
X
6.7
4.9
4.3
3*5
4.5
3*5
5-6
6.0
6.0
4.S
3.7
2.7
6.0
4.9
4*0
3.3
2.5
5.9
5. 5
4.4
3.6
3.0
2.4
5.5
C3
PPM
40
60
90
231
SO
157
37
30
40
SO
80
456
60
70
80
110
711
25
17
20
30
45
100
20
NO
PPM
30
23
21
18
23
21
25
25
23
20
21
19
38
33
33
32
31
40
52
49
50
49
SO
54
UHC
PPN
0
4
6
37
1
IS
0
0
0
1
3
45
7
5
4
4
42
0
0
0
1
1
5
0
CO
GM/KGM
0.77
1*04
1.48
3.64
0.83
2*49
0.68
O.SS
0.73
0.84
1.28
6.90
1.09
1.19
1.29
1.71
10.60
0.45
0.32
0.33
0.47
0.69
1.49
0.35
NO
GM/KGN
0.627
0.430
0.371
0.320
0*420
0*355
0.481
0.493
0.463
0.368
0.360
0.323
0.750
0.613
0.581
0.535
0.496
0.792
0.997
0.886
0.861
0.820
0.802
1.023
UHC
Gn'KGN
0.001
0.035
0.057
0.333
0.014
0.135
0.003
0.001
0.001
0.010
0.027
0.388
0.078
0.049
0.042
0.040
0.357
-.000
0.001
0.003
0-005
0.013
0.043
0.005
BACH.
SMOKE
0.0
0.0
0.0
1 .0
0.0
0.0
0.0
o.o
o.o'
0*0
o.o
o.o
0*0
0*0
0*0
0.0
o.o
0.0
0.0
o.o
o.o
0*0
0*0
0*0
TFG
c
216
207
204
199
214
211
227
232
239
229
224
218
243
243
241
239
237
254
268
261
2S4
252
249
277
120
-------
Table C-5 (concluded). STEADY-STATE POLLUTANT EMISSION DATA: PROTOTYPE
FURNACE WITH VARIOUS CONFIGURATIONS OF INTERNAL
COOLING COILS AND BAFFLES
G
1
fr
i
:
i
|>
li
O
Ul
2 5 ,
0 H 1
1
G ce «
o :
in >-
® z
0 3
Ul
ce
E
00
o
X
o :
6 j
U\ 1
^
Jj U
ft 3
G
ft
s
*.'
1U
I! 3
O u
Ci »-
i/
RUN
NO.
436
427
e
u
J428
429
430
431
s
- 432
ft
433
434
435
j
c 436
437
438
439
440
E
- 441
n
442
443
444
445
446
447
448
449
' 450
451
STOIC. C02 02 CO Mfl UHC C3 NO UHC BACH. TFG
RATIO * S PPM PPM PPM GM/MGM GH/KGM GM/KGM SMOKE C
|.2ri 11.9 4.° 70 40 2 1.21 0.738 0-018 0.0 238
1.19 12.7 3.5 177 35 IS 2.80 0.599 0.135 0.0 227
1.32 11.6 5.3 25 44 1 0.44 0.836 0.010 0.0 241
1.19 12.7 3.5 157 36 11 2.49 0.624 0.099 0.0 227
1.23 12-2 4.1 60 40 2 0.98 0.714 0.019 0.0 231
1.23 12.2 4.1 SO 40 1 0.82 0.714 0.009 0.0 235
1.31 II. 5 5.2 20 44 0 0.35 0.825 0.001 0.0 243
1.21 12.7 3.9 231 36 IS 3>71 0.635 0.137 0.0 218
1.37 11.3 6.0 20 31 1 0.36 0.60S 0.010 0.0 229
1.33 11.6 5.5 IS 31 0 0.28 0.587 0.001 0.0 229
1.22 12.7 4.0 Ml 28 10 1.80 0.493 0.092 0.0 217
1.17 13.1 3.2 426 22 35 6.59 0.365 0.309 0.0 213
1.35 II. 5 5.9 17 31 0 0.32 0.600 0.001 0.0 229
1.34 II. 5 5.7 IS 36 0 0-27 0.707 0.001 0.0 235
1.25 12.3 4.5 35 35 1 0.58 0.623 0.009 0.0 227
1.19 12.9 3.6 148 30 9 2.34 0.509 0.081 0.0 221
1.18 13.4 2*5 1259 22 ISO 18.72 0.359 1.274 I.S 214
1.33 11.6 5.5 15 35 0 0.26 0.680 0*001 0*0 231
1.34 11.2 5.6 30 17 0 0.54 0.343 0*001 0.0 174
1.26 12.0 4.5 50 14 0 0.85 0.265 0.001 0.0 174
1.21 12.4 3.9 51 400 16 3000 &2S.64 0.290 27.471 1.0 166
1.40 11.0 6.4 38 IS 0.73 0.318 0.0 IBS
1*35 11-3 5.7 20 43 8 0.36 0.830 0.082 0.0 191
1.25 12.3 4.4 80 35 8 1.32 0.637 0.076 0.0 IBS
1.19 12.7 3.5 315 23 20 4.97 0.397 0*180 0.0 IB2
1.27 12*6 5.0 40 31 2 0.68 0*562 0.019 0.0 188
121
-------
Table C-6. CYCLE-AVERAGED, FLUE GAS POLLUTANT EMISSION CONCENTRATIONS
FOR THE MODIFIED, 1.0 ml/s PROTOTYPE FURNACE SYSTEM WITH AN 8-TUBE,
LOW-PRESSURE AIR-COOLED COIL, SUPPLEMENTARY
HEAT-EXCHANGER INSTALLED
11
in in
V (9
Ol
.* 19
S §
3 J=
o in
0 E
I'
o.o
0.0
TFG
c
222
219
217
221
229
?35
229
238
210
227
237
257
21 3
207
210
207
213
241
231
24S
249
122
-------
TECHNICAL REPORT DATA
(Please read Inuructiom on the reverse before completing)
1 REPORT NO.
EPA-600/2-77-028
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUB TITLE
Residential Oil Furnace System Optimization--Phase II
5. REPORT DATE
January 1977
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
L.P. Combs and A.S. Okuda
8. PERFORMING ORGANIZATION REPORT NO.
R76-105
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Rocketdyne Division/Rockwell International
6633 Canoga Avenue
Canoga Park, California 91304
10. PROGRAM ELEMENT NO.
1AB014; ROAP 21BCC-027
11. CONTRACT/GRANT NO
68-02-1819
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 8/75-9/76
14. SPONSORING AGENCY CODE
EPA-ORD
is SUPPLEMENTARY NOTES JERL-RTP project officer for this report is G.B. Martin, Mail
Drop 65, 919/549-8411 Ext 2235.
i6. ABSTRACT
report describes the second of a two-phase investigation into ways to
improve the air pollutant emission and thermal efficiency characteristics of residen-
tial oil furnaces. A prototype, low-emission, warm-air furnace (designed in Phase
I to embody a number of burner and combustor criteria for minimizing emissions
compatible with high efficiency) was assembled and tested. Design details were chan-
ged as necessary during laboratory testing to help achieve the objectives. Applicabil-
ity of the design criteria was demonstrated within current conventional oil- heat indus-
try practices. Compared with estimated average characteristics of existing installed
residential furnaces and boilers, nitrogen oxides emissions were reduced by 65% or
more , and steady-state efficiency was increased by a minimum of 10 percentage
points. Experimental results and component changes made in obtaining them were
incorporated into a preliminary design for an integrated low-emission furnace which
should be commercially producible and cost-competitive.
17.
KEY WORDS AND DOCUMENT ANALYSI1
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Croup
Air Pollution
Furnaces
Fuel Oil
Residential Buildings
Tests
Thermal- Efficiency
Nitrogen Oxides
Warm Air
Heating
Air Pollution Control
Stationary Sources
Oil Furnaces
Residential Heating
Emission Control
13B
13A
21D
13M
14B
20M
07B
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (Tha Report)
Unclassified
21. NO. OF PAGES
132
2O. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
123/124
------- |