EPA-600/2-76-2 56
September 1976
Environmental Protection Technology Series
COMMERCIAL FEASIBILITY OF AN
OPTIMUM RESIDENTIAL OIL BURNER HEAD
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
c 3!egories were established to facilitate further development and application of .
elvironrnental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
T 18 five series are:
1.
2.
3.
4.
5.
Environmental Health Effects Research
Environmental Protection Technology
Ecological Research
Environmental Monitoring
Socioeconomic Environmental Studies
T ,is report has been assigned to the ENVIRONMENTAL PROTECTION
TI::CHNOLOGY series. This series describes research performed to develop and
d,~monstrate instrumentation, equipment. and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
wJrk provides the new or improved technology required for the control and
trl~atment 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
vj.ews 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 Technicallnforma-
tion Service. Springfield. Virginia 22161.
-------
EPA-600/2-76-2 56
September 1976
COMMERCIAL FEASIBILITY
OF AN OPTIMUM
RESIDENTIAL OIL BURNER HEAD
by
L. P. Combs and A. S. Okuda
Rocketdyne Division
Rockwell International Corporation
6633 Canoga Avenue
Canoga Park, California 91304
Contract No. 68-02-1888
ROAPNo. 21BCC-058
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 feasibility of commercial application of an optimum head for distil-
late oil l,urners to effect simultaneous reductions in emission of air
pollutanb and consumption of fuel by residential space heating equip-
ment has been investigated. The optimum head technology was developed
under an f:arlier EPA-sponsored study and was shown to minimize emission
of oxides of nitrogen from a variety of research combustors while main-
taining beth high efficiency and low emissions of other pollutants.
The current study was concentrated on selecting the best commercially
practiced fabrication method for making optimum heads, on determining
that prototype heads made to simulate such production units effectively
reproduce the research head's beneficial results, and on extending the
data base by testing the prototype heads in two commercial residential
furnaces.
Sheet metal stamping was selected as being the best fabrication method.
A one-piece stamped and folded optimum head design was evolved, and
prototype commercial optimum heads were fabricated. These were shown
in researC1 combustion chamber tests to be equivalent to the earlier
research h~ad. Tested as retrofit replacements of the stock burner
heads in t',110 new warm-air oil furnaces, the prototype optimum heads
were found to be operationally satisfactory and potentially durable and
long-lived, Measured retrofit effects on furnace thermal efficiency
and NO emi:;sions were compared with estimated average characteristics
of existin!~ installed residential heating units to project estimates
that widespread retrofitting of old existing residential units could
increase moan season-averaged thermal efficiency (averaged over those
units retrofitted) by about 5 percentage points and simultaneously
reduce NO emissions from these sources by about 20%. Several issues
x
are noted as being unresolved, including logistics of a retrofit pro-
gram, service personnel training needs and requirements to ensure
meeting codes and standards.
ii
-------
CONT~NTS
Section I
Conclusions
Section II ,
Recommendations
Section III
Introduction
Section IV ,
Analytical Investigations
Oil Furnace Operations Analysis
Selection of Commercial Fabrication Methods
Design of Com~ercia1 Prototype Optimum Heads,
Section V
ExPerimental Investigations
Fabrication of Prototype Commercial Heads
1
1
"
3
3
4
4
7
"
7
7
24
28
33
33'
33
performance of Prototype Commercial Optimum Heads in
Research Combustion Chambers
36
Performance of Prototype Commercial Optimum Heads in
Residential Furnaces
Discussion
Section VI
References,
APpendix A
Flue Gas Compositional Analysis
References
Appendix B
Data Tabulation:
Research Combustor Experiments
Appendix C
Data Tabulations:
Warm-A~r Furnace Experiments
Hi
49
67
77
77
79
91'
92
96
-------
ILLUSTRATIONS
1.
1 n.l/s (gph) optimum low-emission residential oil burner
(Rd. 5)
2.
Flue gas gross thermal efficiency losses as a function of net
flue gas temperature and composition
3.
Reduction in fuel consumption when a furnace is replaced by
a ITore efficient unit
4.
5.
Calculated effect of burner on-time upon furnace efficiency.
Calculated effect of firing rate upon furnace efficiency
6.
Typical smoke and gaseous emission characteristics for a
residential unit in the tuned condition (Ref. 2)
Distribution of smoke, CO, and HC emission for residential
7.
8.
units (Ref. 2)
Layout drawing of the stamped-formed, sheet metal prototype
9.
optimum burner head
Sec~nd prototype optimum burner head design with a modified
10.
cho~e plate configuration
Photograph of two initial design sheet metal prototype
11.
comnercial optimum oil burner heads constructed from
typl~ 430 stainless steel
Pho':ograph of modified prototype commercial head showing the
12.
0.0020 m deep reinforcement channel
Exp(~rimental combustion chamber and heat exchanger
13.
arrangement.
Schl~matic of oil burner and research combustion chamber
test installation.
14.
Photograph of the initial 21 gage sheet metal prototype
comrlercial head installed on the Williamson burner body
15.
Photographs of the initial sheet metal design, prototype
commercial heads after approximately 15 hours of cyclic
service in a 0.22 m I.D. insulated side-fired combustor
iv
6
11
15
19
19
21
22
29
32
34
37
38
40
42
43
-------
25.
16.
Photograph of the 21 gage type,430 stainless steel prototype
commercial optimum head with a 0.0020 m 300 s~ries stainless
steel reinforcement plate after appro~imately 15 hours of
service
17.
Effect of oil nozzle spray cone angle upon flue gas nitric
oxide concentrations using prototype commercial heads in the
0.22 m I.D. insulated cylindrical research combustion
chamber
18.
Comparison of flue gas nitric oxide concentrations between
the experimental optimum burner head and the commercial
19.
20.
prototype optimum burner head in a 0.22 m I.D. insulated
cylindrical combustor
Schematic of the furnace performance evaluation system
Gross thermal efficiency characteristics of furnace~ tested
in their stock configurations.
n.
Cycle-averaged flue gas nitric oxide concentrations for
furnaces in their stock configurations
Photographs of the modified sheet-metal prototype commercial
optimum oil burner heads after 500 hours of cyclic service
22.
23.' Comparisons of pseudo-steady-state thermal efficiencies from
the Williamson 1167-15 and the Carrier 58HV-l56 furnaces
u~ing their stock burner heads and prototype commercial
optimum burner heads.
24.
Effect of the commercial prototype optimum head upon cycle-
averaged nitric oxide emissions from the Williamson 1167~l5
furnace
Effect of the commercial prototype optimum head upon cycle-
averaged nitric oxide emissions from the Carrier 58HV-156
furnace
26.
Steady-state thermal efficiency relationships for r~sidential
furnaces
v
45
47
48
54
, 58
61
63
65
66
66
69
-------
TABLES
L
Thern.al and Service Efficiencies of Two Residential Hot
Water Boilers (Ref. 9)
13
2.
Steady-State Absorption Efficiency of Oil Heating Units
(Ref. 10)
13
3.
Summary of "WAFURN" Computer Program Results of a Refractory-
Lined Combustor, Warm-Air Furnace Model
18
4.
Comparison of Commercial Fabrication Methods for Optimum
Burner Heads.
26
35
5.
Composition of Various Types of Stainless Steels.
vi
-------
SECTION I
CONCLUSIONS
1.
The concept of undertaking commercial production, distribution, and
marketing of an optimum distillate oil burner head for retrofit
application to residential oil heating equipment was found to be
technically feasible.
2.
The optimum head technology should be applicable with some benefit
to an estimated 50% or more of existing residential oil-fired warm-
air furnaces and hydronic boilers in the United States. Inapplica-
bility to some units would result from equipment incompatibilities,
operational problems, and/or degradation rather than benefit of
efficiency, pollutant emissions, or both.
3.
Benefits from retrofitting an individual heating unit would be a
modest increase in thermal efficiency and a significant reduction
in emissions of oxides of nitrogen (NO) air pollutants. Estimated
x
increases in a given furnace's efficiency range from zero (or less)
for units already operating with low excess air to 10% or more for
units now operating with very high excess air. Judicious applica-
tion of retrofitting to large numbers of appropriate units should
achieve increases in mean steady-state and overall season-averaged
thermal efficiencies (averaged over those units retrofitted) of at
least 5 percentage points. Simultaneously, NO emissions :fromthose
x
same units should be reduced by about 20%. Retrofitting also might
reduce significantly the average smoke emissions but would not
change appreciably the already low emissions of carbon monoxide and
unburned hydrocarbons.
4.
Sheet metal stamping was assessed as being the best commercial fab-
rication method for making optimum heads. Single piece heads with
integral swirl vanes, choke plate, and attachment provisions can
be made and can satisfy all technical requirements. Sheet metal
thicknesses in the range of 0.00087 m to 0.00127 m (0.0344 to
1
-------
0.0500 inch) are satisfactory, but a relatively refractory stainless
stee;. must be used. While tested for only 500 hours of simulated
servi.ce, Type 310 stainless steel showed excellent resistance to
degrulation and should have an indefinitely long lifetime. The
less expensive Type 304 stainless steel may also prove to be
adequate, but its test exposure duration was not long enough to
give a valid indication of potential durability.
5.
In vclume production, it was estimated that stamped stainless steel
sheet metal optimum heads should cost the manufacturer on the order
of $1.50 each.
2
-------
;,
SECTION II
RECOMMENDATIONS
1.
If application of the optimUm burner head technology is to be pur-
sued. it would be prudent and logical to conduct comparative before
and after performance testing of a representative sample of old
existing units in actual residential use. Incremental changes in
efficiency and NO emissions should be measured and correlated to
x
initial burner type. and the requirements for ensuring that retro-
fitted burners meet applicable codes and standards should be
determined.
. 2.
A market survey should be conducted to define as fully as possible
the current makeup of the existing oil furnace and boiler popula- .
tion. purner types. firing rates. blast tube diameters. ignition
electrode diameters and spacing. etc., and the potential burner
retr~fit or replacement rate as a function of capital cost to the
homeowner and potential annual fuel savings. These data should be
combined to define the minimum number of different head designs
which can satisfy the market requirements.
3.
A manufacturing cost analysis should be performed to project capital
apd operating expenses for making and distributing the different
head designs. The effect of production rate on cost per head. of
distribution and markup costs on price to the consumer and of price
on replacement rate should be included in an economic analysis lead~
ing to estimated production rates for each head design.
4.
Because the manufacturing cost per head is closely tied to material
costs, ways of minimizing material weight per head should be inves-
tigated carefully. Clever interlacing of adjacent stampings should
be examined. Also. because the perimeter of each stamping is con-
trolled largely by the size of the swirl vanes. some additional
~~perimental furnace testing might be devoted to investigating
the minimum required vane size.
3
-------
SECTION II I.
INTRODUCTION
The Envi:~onmenta1 Protection Agency has sponsored studies over the past
few year:; to document the emission of air pollutants from existing resi-
dential and commercial oil-fired space heating units (Ref. 1 and 2).
Concurrently, the EPA has also supported applied research programs to
determine the effects of "controllable" parameters on emission levels
and to devise strategies for minimizing pollutant emissions (e.g., Ref.
3 through 51. These and related studies have shown that substantial
reductioIlS in total emissions can be effected by combustion modifica-
tions such as advanced burner designs, flue gas recirculation and two-
stage combustion.
In particular, an intensive Rocketdyne investigation of residential and
commercial oil burners (Ref. 5) led to criteria for optimizing conven-
tional burner designs with respect to pollutant emissions. For high-
pressure atomizing, luminous-flame burners fired into refractory-lined
combustion chambers, 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 oriented at 25 degrees relative to the blast
tube axis. This swirler vane angle gave the best compromise between
smoke emissions and nitric oxide emissions, while the optimized choke
diameter 'Jroduced minimum nitric oxide emissions.
Those burner design
attributes were all concerned with the burner "head," Le., the portion
of the bUTner that admits prepared reactants into the combustion chamber.
For that :~eason, this development was referred to as the "optimum head."
In additi
-------
and smoke), with considerably less excess air than is usual in residen-
tial furnace installing. Reducing excess air decreases the sensible
heat losses in the flue gase and, therefore, results in a net increase
in thermal efficiency.
Because the optimum burner's distinguishing design features were con-
fined to the burner head, the optimum head was recognized as present-
ing a very attractive possibility for helping to reduce simultaneously
both pollutant emissions and fuel consumption with a low-cost retrofit
device for existing burners in existing furnaces. Previous experience
with optimum heads, however, was limited to two research heads (one
each in two sizes) fabricated by machining and welding stainless-steel
plate and tested predominantly in research combustors (Fig. 1). Addi-
tional experience, including testing as a retrofit device in commer-
cially available furnaces, was obviously needed to establish the feasi-
bility of applying the optimum head commercially.
The investigation described in this report was undertaken, therefore,
with the objective of evaluating and establishing the technical feasi-
bility of commercial application of the optimum distillate oil burner
head to residential furnaces. Several aspects of commercialization
were explored in ~ntegrated studies described and discussed in the fol-
lowing sections of the report. Predominantly analytical aspects are
reported in Section IV. Those include: the analysis of current resi-
dential furnace operation and performance to determine design and oper-
ating variables affecting fuel economy and the proper firing rate ranges
for optimum head experimentation; consideration of commercial fabrica-
tion techniques selected as being most applicable for economic commer-
cial production of optimum heads; and the design of prototype heads
producible by the selected fabrication technique(s). Aspects that are
predominantly experimental are reported in Section V, Experimental
Investigation. Included are the fabrication of prototype optimum heads
to simulate those that might eventually be made commercially, and the
laboratory testing of those heads: (1) in research combustors to
5
-------
establish correspondence with earlier preprototype research head results,
and (2) in typical commercial warm-aIr furnaces to evaluate performance
under realistic cyclical operation.
a
,~L
i
,;~~
,
D
.
00
"
1__-
5DZ2l-S/6/73-S1B
EXTERNAL VIEW
-.-......,~
5DZ2l-s/6/73-S1A
OPTIMUM HEAD
Figure 1.
1 ml/s (gph) optimum low-emission residential oil
burner (Ref. 5)
6
-------
SECTION IV
ANALYTICAL INVESTIGATIONS
Brief background studies were made in two different areas preparatory
to designing commercially producible prototype optimum heads. The first
was concerned with estimating effects conversion to the optimum head
might have on typical furnaces' fuel economy and pollutant emissions,
and with selecting an appropriate burner firing rate for subsequent
prototype head testing. The second was concerned with studying commer-
cial fabrication methods that might be used for producing optimum heads
and selecting the one or two judged to be best for high-volume production
of low-cost, durable units. Because a fabrication method's applicability
depends on having a workable head design, a preliminary design effort was
conducted concurrently with .that study. The results led naturally to
finalized designs for protot.ype optimum heads.
OIL. FURNACE OPERATIONS ANALYSIS
In 1970, approximately 14% of the U.S. energy consumption was used for
residential space heating and domestic water heating (Ref. 6). Essen-
tialty, all of that energy was derived from fossil fuel combustion,
either directly with combustion equipment on the premises or indirectly
through the use of electrical resistance heating. That quantity of en-
ergy was derived from various energy sources in approximately the follow-
ing distribution: 65.7% from gaseous fuels, 24.3% from distillate fuel
oils, 6.4% from utility electricity, and 3.6% from coal, wood, solar,
etc.
At that time, nearly one-half of U.S. residences were heated by forced-
draft warm-air furnaces, of which approximately 72% were gas-fired, 22%
oil-fired and 5% electrical resistance heated. Almost one-quarter of
the U.S. homes were heated by steam or hot water from hydronic boilers,
with about 40% gas-fired and 54% oil-fired. The remaining homes (about
7
-------
one-third of the total) were either unheated or employed a broad range
of equipITent ranging from fireplaces to floor or wall-mounted direct-
heaters.
The total population of oil-fired residential furnaces and hydronic boil-
ers in 1970 exceeded 13 million units. Since then, annual sales of such
units have averaged about 525,000; additionally, an average of about
154,000 conversion burners have been sold annually. From these figures,
an oil furnace replacement rate of about 4 to 5% per year can be esti-
mated; it is evident that the average makeup of the installed population
will change only rather slowly over the years. Thus, even though units
sold today may offer significantly better fuel economy than the vast
majority )f older furnaces, their sales rate increases the overall aver-
age efficiency only slightly from year to year.
There hav~ been strong market incentives (increased oil prices and uncer-
tain suppLies) since late 1973 to replace old inefficient residential
heating equipment. Nonetheless, a homeowner who has a unit that is in-
stalled, '.\'orking, and paid for should be expected to move slowly and
reluctantLy toward deciding to spend several hundred dollars now to re-
duce futu:~e heating costs. A retrofit burner head offering efficiencies
comparabll~ with (or even higher than) current new equipment would have
considerably greater consumer appeal than would replacement of an entire
boiler, ftlrnace, or even burner, particularly if it were easily installed
at low co~;t. Retrofit should, therefore, proceed at a considerably
faster pace and contribute substantially to raising the national average
efficiency of utilizing heating oil. The magnitude of the consumer
appeal would obviously depend on how large an efficiency gain an individ-
ual homeo\mer could realistically expect.
Thermal Efficiencies
Oil furnac.es and hydronic boilers are manufactured in conformity to na-
tional sta.ndards. Current testing and rating standards are ANSI 291.1-
8
-------
1972 for oil-fired warm-air furnaces (Ref. 7) and the Hydronics Insti-
tute 1975 standard for oil-fired hydronic boilers (Ref. 8). Both of
these standards specify that a new unit's steady-state thermal effici-
ency, based on the fuel's higher heating value, shall equal or exceed
75%. At steady-state, most of the heat that is not recovered is con-
vected up the flue as latent heat of the water vapor formed in the com-
bustion process and as sensible heat of the flue gases, with only minor
losses (typically 1/2 to 1-1/2%) conducted through the cabinet, etc., and
radiated or convected to the surroundings.
The latent heat of combustion-formed water vapor represents between 6
and 7% of fuel oil's higher heating value. Condensation in the f~ue sys-
tem is intentionally avoided in conventional furnace and boiler technol-
ogy, so losses of this magnitude form an una~oidab1e baseline. The way
condensation is prevented is. by maintaining the flue gases above their
dew point everwhere in the system; in practice, this is ensured by de-
signing for net unit exhaust gas temperatures* of about 220 C (400 F) or
greater. As a direct result, there are substantial thermal losses in
the form of flue gas sensible heat. Their minimum value is on the order
of 8% when combustion is carried out at stoichiometric conditions (i.e.,
no excess air) and the flue gases are cooled to 220 C (400 F) net stack
temperature. Adding these approximate latent and sensible heat losses
suggests that conventional oil furnaces and boilers might have steady-
state efficiencies as high as 85%. Usually, however, sensible heat
losses are greater than 8% for two reasons:
(1) some excess air is re-
quired to avoid formation of excessive CO, UHC, and/or smoke, ~nd (2)
average net flue gas temperatures are higher than the minimum needed to
avoid condensation.
Tabular values of steady-state flue gas thermal efficiency decrements
are given in Ref. 7 as functions of net flue gas temperature and flue
*Net flue gas temperature is the difference between actual flue gas
. temperature and mean heated room temperature.
9
-------
gas con\:entration of C02 (which is the parameter measured by heating
industr:r personnel, rather than excess air level). These data are
plotted in Fig. 2 as a family of curves, with efficiency decrements
along the left hand ordinate. The decrements are converted to estima-
ted steady-state efficiencies along the right-hand ordinate by sub-
tracting them from 100% and assuming that 2% of the fuel's higher
heating value is transferred ("lost") to the surrounding's through the
furnace cabinet. A supplemental scale is given relating exhaust gas
stoichiometric ratio (SR) to its C02 concentration. Defined as the
actual o.ir-to-fuel weight ratio divided by the theoretical stoichio-
metric M'eight ratio, SR is related directly to the excess air level.
For exanple, SR = 1.50 corresponds to 150% stoichiometric air and
this is equivalent to 50% excess air.
Residential furnaces and hydronic boilers are typically operated in an
on-off cyclical manner. The excess air level and flue gas temperature
must be controlled to avoid smoke formation and condensation over a
range of cycle conditions, and this generally forces them to be higher
than would be appropriate for steady-state operation only. This is one
of the m,ljor reasons for operating at less than optimum efficiency con-
ditions. During cyclical operation, efficiency is further degraded by
some transient contributions to a unit's heat losses. When the burner
is not bl~:ing fired (standby), .a natural draft flow of air through the
burner, ::irebox, etc., cools furnace components and continues to convect
heat up the flue. Typically, this loss may cause cycle-averaged effi-
ciencies to be 3 to 5% lower than steady-state although, in some situa-
tions, the decrement may be as large as 15%. External heat losses from
the cabinet also continue during standby. With warm air furnaces, addi-
tional cyclical cabinet losses are moderately small (-1/2 to 1%) because
furnace components are cooled considerably before the warm air flow is
turned off. With hydronic boilers, however, they may be substantially
larger (~'1-1/2 to 3%) because most boiler components are at nearly the
same temperatures during standby as during firing.
10
-------
~
..
~
~
o
-J
>.
U
C
0)
.....
u
.....
Ct-
Ct-
.....
to-' ~
to-' S-
O)
.c
.....
~
~
o
S-
c:;
Figure 2.
(",.....--.
35
. . ~ t . :
. --. --..........-.--. ""'" _u ....-.-- .. ...-., --
. . ." ,
--:-"..._--- _...~:_- :----! "':----:-___h"':' -:---t-.....
. . .
. " . t
... - -. '"
. . . .
ANSI Z91.1-1972 (Ref.i 7) ;
.---..-.---..-
..- -. ...., _u.
30
.-.
... .-...
25
20
. ~ &rQ()' f)
~~'L C. .-- ".... -.
. .: 300 f) ...:.
,61 C. {., :
. j...-.-..
... .... . ..
15
.... .-..
.. . --., 1.- .... .-: . .. .-,.
. .."..
10
'" _... -.,.-"
15
12 11 10
Volume Percent C02.
Flue gas gross thermal efficiency losses
temperature and composition
14
13
9
8
65
~
..
70 t'
c:
4)
.-,....
~.
u"'O
.-4)
Ct- E
Ct- ::J
ILl ~
CII
75 4) ~
4-J
"' CII
4-J CII
CII 0
I -J
>-
"'0 0\
IV c:
4).-
4-J CII
80 V1 tV
u
"'0
4)~
4-J N
IV-
E
4-J
CII
ILl
85
90
as a function of net flue gas
-------
The decr,~ments between steady-state efficiencies and cycle-averaged
(service') efficiencies also depend on the mean cycle timing, growing
larger a ~ standby time increases and vanishing as. steady-state operation
is approached. This is exemplified by laboratory data for two hydronic
boilers :~eported in Ref. 9 and reproduced in Table 1; general agreement
with the ranges stated above is evident.
Over a long period of time, such as an entire heating season, there will
be a wido distribution of thermal demand conditions ranging from nearly
continuous standby to essentially continuous operation. Season-averaged
efficiencies are rarely (if ever) measured, but several investigators
have estimated values from shorter term testing.
A recent example is
given in Ref. 10. Steady-state (absorption) efficiencies were measured
in the ":;,5 found" and subsequent "tuned" conditions for a representative
sample of residential heating systems in northern New Jersey. The re-
sults art:' reproduced in Table 2 and show little influence of unit type
or burner tuning. Subsequently, 11 units were instrumented to record
performarce data, including cycle timing for a long enough period of
time (1 to 3 weeks) to correlate cycle behavior and oil consumption to
outdoor ambient conditions typically encountered over a heating season.
An overall average efficiency (pseudo-season average) of 60% was ob-
tained, indicating that standby losses (predominantly component cooling
by draft air flow through the combustor) averaged 15%. It was also
sh~Wn that these rather large standby losses could be approximately
" halved by reducing the units' firing rates (by an average of 25%) so
that they would fire continuously when the ambient temperature dropped
to the local design temperature. Many other investigators' estimates
of season-average efficiencies could be discussed but there are so many
uncertainties associated with each that this won't be pursued here.
Rather, a 60 to 65% range will be assumed to be reasonably valid.
The field survey data reported in Ref. 2 for an entirely different, but
similar s:Lze, sample of residential oil furnaces and boilers was con-
cerned wi":h emissions rather than thermal performance. Nonetheless,
12
-------
Table 1. THER~~L AND SERVICE EFFICIENCIES OF TWO RESIDENTIAL
HOT WATER BOILERS (REF. 9)
Thermal Service Efficiency,%
Efficiency, Standby Loss, (en/off time, minutes)
Boi ler and Cond i t ion % % of Input (20/10) (15/15) (10/20)
Boi ler A
New 76.2 2.02 75.50 74.80 71.;30
After 6 months 74.8 -- 74.10 73.30 70.00
After 10.5 months 73.43 -- 72.87 ]2.06 68.78
Boi ler. B
New 72.60 2.40 71.50 70.00 67.10
After 6 months 71.50 -- 70.30 68.80 66.00
NOTES: Electric consumption of accessories included in input.
Boi ler A differs from most contemporary oil-fired equipment
Table 2. STEADY-STATE ABSORPTION EFFICIENCY OF
OIL HEATING UNITS (REF. 10)
Type of No. of Absorption Efficiency, %
Heat Units As Found Tuned
Warm Air 12 72.6 :t:3 75.4:t:5
Hot Water 11 74.4 f:5 72.3:t:7
Steam 15 75.7f:3 76.7 f:3
Total 38 75.4 74.6
(average)
13
-------
steady-stc.te
minute stc.ck
the averag:es
efficiencies can be derived from the reported C02 and tenth
temperature data. Assuming a uniform 2% casing heat loss,
of 33 units' estimated efficiencies were: 71.0%(+10, -21)
as found, and 72.1% (+7, -14) tuned. 1t is possible that there was a
consistent bias in the excess air levels between these two surveys, but
it seems ITore likely that any bias was in the instrumentation and meas-
urement methods. The scatter is great enough, particularly in the Ref.
2 data, that the extreme values influence the averages significantly.
While this suggests that larger samples would be desirable, for our pur-
poses, the two surveys can be averaged to obtain an approximate charac-
terization of the entire U.S. residential oil furnace and hydronic boiler
population:
.
Average steady-state conditions somewhere in the ranges:
90 :tlO% excess air; 8 .:to.3% C02 } 72 to 75% gross
. 500 :t60 F net flue gas temperature thermal efficiency
.
Se,lson-averaged gross thermal efficiency in the range of
60 to 65%
Now we can turn to the question of how much fuel can be saved by replac-
ing or ret:~ofitting an old inefficient unit with equipment amenable to
. higher efLciency. The range of possibilities is illustrated in Fig. 3.
As an example, if a unit averaging 60% thermal efficiency were replaced
(or retrofitted) to increase the average efficiency to 85% (a 25% gain),
its fuel consumption would be reduced by 29%. It has been seen earlier
that 85% r(~resents an approximate upper limit within the conventional
furnace anc. boiler technology.
Therefore, a dotted line has been drawn
through the locus of points where the old efficiency and the efficiency
gain sum tc 85%. Portions of the efficiency gain curves to the right of
that dotted line are dashed to indicate the impracticality of operating
in that region with present-day equipment.
14
'.'
-------
c
o
...,
a.
E
:J
. UI
C
o
U
t,
40
~
,~Condensation in
, Flue Likely
, .
,
,
\
30
r'
, ........
\ -,-, 25
I ,...............
'.." ............... -.. 20
~I- --- - I
'\ r
,- -
20
CI)
:J
.....
C
c
o
...,
U
:J
"0
CI)
. ex:
10
o
40
60 80
Mean Season~Averaged Efficiency of
Furnace Being Replaced, %
Figure 3.
~
<1 ~,IC'
1? ~.
~ <'
'/ ".
Q) ~.,
~ <'
--- 1 5 ~".! '..J-
~.., ~
(4 ())....
..,~
()'
-:.....::- - 1 0
I 5
I'~ -,--
100
Reduction in fuel consumption when a furnace is replaced
by a more efficient unit
15
-------
If the entire population of residential oil heating equipment were re-
paired, retrofitted, or replaced with comparable equipment capable of
I)
raising the average steady-state efficiency from, say, 74% to 82%, the
8% efficiency gain would reduce national consumption of oil for these
purposes by about 10%. If the firing rates were simultaneously de-
creased to eliminate overfiring, thereby reducing standby losses by
another 7%, total estimated reduction in residential oil consumption
would be approximately 20%. The 82% steady-state level can be attained
by lowering' net flue gas temperatures to practical minimum values and
lowering excess air to the 20 to 25% range.
Analysis )f a Specific Furnace - To obtain more confidence in the valid-
ity of the foregoing discussion, an analytical computer model of a warm-
air furna:e was used to calculate thermal efficiency behavior of a par-
ticular fJrnace as if it were retrofitted with an optimum head and run
with l5%~xcess air instead of earlier representative conditions of 50
and 85% excess air.
The existing WAFURN (warm air furnace) computer program is a transient
heat tran.;fer analysis with the capability of accounting for the effects
of typicaL furnace operating variables on efficiency (Ref. 11). A WAF URN
program evaluation results in a calculated net furnace cycle thermal
efficienc:r.
The cyclic analysis is conducted through iterative calcu-
late/balm~e loops to ensure cycle-to-cycle continuity. The detailed
thermal analysis allows variation of many parameters such as: (1) oil
input (fL~ing rate and temperature); (2) stoichiometric ratio; (3) cycle
length and profile; (4) flue gas temperature (heat exchanger capability);
(5) fuel quality (heat of combustion and water emulsion); and (6) input
combustion air conditions (indoor or outdoor supply).
The operation of a warm-air furnace with a refractory-lined combustion
chamber wc.s modeled. Parameters varied were: (1) cycle timing, (2)
cycle len~:th, (3) firing rate, (4) combustion air temperature, and (5)
stoichiomE!tric ratio. A summary of calculated results is presented in
16
-------
Table 3. The thermal efficiencies (nf ) listed include the draft air
urn
heat losses but do not include the relatively constant loss through the
furnace casing (-1 to 2%). The SR = 1.15 data represent .the expected
operating condition of the optimum head design while the SR = 1.85 data
represent the season-averaged operating condition of existing units in
the field. The warm-air flowrate for the 10 runs was fixed at 0.566
m3/s (1200 cfm) and, except for Case 6, combustion air was brought in
from outdoors at 0 C (32 F).
The data from Table 3 are plotted in Fig. 4 and 5. Figure 4 shows the
effect of burner on-time on the furnace efficiency. An interesting
result is the leveling off of the efficiency curves at about 67% burner
on-time; the profiles in Fig. 4 compare well to the behavior of the data
in Table 1. Typical burner on-times are more on the order of 33%, show-
ing some room for about a 2%. efficiency gain by changing the furnace
operational profile. This, of course, would be dependent on the geo-
graphical location of the installation and the margin required by the
local weather characteristics.
Figure 5 shows the effects of changing the firing rate in a fixed fur-
nace configuration. The graph also shows a comparison of the optimum
head against a typical burner/furnace installation. Given a typical
furnace operating at 1.0 ml/s with a calculated 77.5% efficiency, re-
placing the burner head with the optimum head was calculated to increase
the furnace efficiency to 84.4%.
However, to maintain the same rate of
heat output, the firing rate could also be reduced to 0.92 ml/s (i.e.,
: ?7.5/84~4) resulting in an overall anticipated increase in efficiency
of up to 7.6% (85.1 minus 77.5%) with the installation of an optimum
burner head unit.
Cases 1 and 6 in Table 3 provide an efficiency comparison between an out-
door (T = 273 K) and an indoor (T = 293 K) furnace installation (9r" com-
bustion air supply).
The heat transfer analysis shows an improvement of
17
-------
Table 3.
SUMMARY OF "WAFUR.N" COMPUTER PROGRAM RESULTS OF A REFRACTORY-LINED
COMBUSTOR, WARM-AIR FURNACE MODEL
......
00
I I , J .
Stoichiometric Ratio
Cycle 1.15(Optimum Head) 1. 50 1. 85(Ref. 2 Avg.)
Firing Timing, Draft Ai r Draft Ai r Draft Air
Case Rate, minutes T)furn, Heat Loss, T)furn, Heat Loss, T1furn, Heat Loss,
. No. ml/s on/off % % % % % % Remarks
8 1.0 12/0 86.07 0.0 82.75 0.0 79.77 0.0 Steady-state
9 2/10 81. 72 6.23 77 .97 8.02 74.17 9.78 1 7% on, 12-minute cycle
1 4/8 84.40 .2.62 81.01* 3.37 77 .53 4.11 33% on, 12-minute cycle
6 4/8 85.38 2.17 82.24 2.80 79.17 3.43 Indoor Air (68 F), 12-minute cycle
2 8/4 86.04 0.77 82.76 0.99 79.72 1.20 67% on, 12-minute cycle
10 10/2 86.11 83% on, 12-minute cycle
3 10/20 84.73 2.12 81. 43 2.65 78.18 3.14 33% on, 3D-minute cycle
4 0.75 5.33/6.67 86.29 2.12 84.09 2.67 80.41 3.27 Lower firing rate, same heat input
as Case 1
7 0.75 5.06/6.94 85.56* 2.31 83.38 2.91 80.01 3.58 --
5 1.25 3.2/8.8 82.22* 3.06 78.58 3.91 75.03 4.73 Higher firing rate, same heat input
as Case 1
*Cases at different firing rates but having equiva1ent heat outputs.
-------
85
""
~ 80
'"
c
...
:::>
~
c:
....
\D
75
Figure 4.
o
SR = 1.15
(Optimum Head)
SR = 1.85
(Ref. 2 Average)
1 m1/s Firin~ Rate
12 minute cyc1S
T outdoor = 273 K
50
Burner On Time, %
100
Calculated effect of burner on-time
upon furnace efficiency
85
Constant Heat Input
(4 on/8 off @ 1 m1/s)
""
CIJ
U
'"
c.
...
:::>
.....
c:
80
SR = 1.85
(Ref. 2 Average)
75
0.75
1.00
Firing Rate, rn1fs
Figure 5.
Calculated effect of firing rate
upon furnace efficiency
,,.-1
-------
about 1 to 1-1/2% in furnace efficiency by utilizing the warmer indoor
air supp:.y. However, the computer model does not account for the source
of the heated (and humidified) air and, therefore, it does not show the
additional 3 to 4% heat output required to heat the consumed air.
The
net effect of using living space air for combustion air is more on the
order of negative 2-1/2%
Pollutant Emissions
Several definitive studies of pollutant emissions from residential oil
heat systems have been conducted previously (e.g., Ref. 2, 3, and 5).
Typical characteristics are indicated in Fig. 6, reproduced from Ref. 2.
It is seen that there is an operating range over which the emissions of
incomplete combustion products (smoke, unburned hydrocarbons, and carbon
monoxide) are low. Generally, the .width of this region is constrained
by production of excessive smoke as excess air is reduced (increasing
C02) and by generation of excessive CO as excess air is increased (de-
creasing :02)' Examining Fig. 6, it is apparent why existing oil heat-
ing equip'nent is adjusted to conditions that produce, on the average,
about 8% ,:02 flue gases.
Emissions differ from unit to unit. The distributions of levels of pol-
lutants ~nitted from 33 residential oil heating units reported in Ref. 2
were pres,mted graphically in that report. To ensure their accessibil-
ity to in':erested readers, they are reproduced here as Fig. 7. It is
seen that tuning* a burner has a substantial effect on smoke, small but
observable effects on CO and HC, artd practically no effects on C02' NOx'
and filterable particulates.
*Tuning refers to the burner and heating system service procedure of
cleaning, adjusting and/or replacing burner components (electrodes,
blower wheel, blast tube, oil filter, oil nozzle), finding and sealing
easily corrected air leaks, adjusting firebox draft, and setting the
combustic'n air level for maximum C02 with minimum smoke from a stable
flame. .
20
-------
9
8 32 160
7 \ 28 140
,
... \ Operating Range for
Q) 6 Low Carbona<;eous 24 120
.0
E Emissions
:J . E
.2 \ a.
Q) 5 E 20 100 a.
.:r&. )(
o c.
E a. 0
Cf) . 2
4. \ NOx,>'" -,s:- ~ 16 80"0
.c c
u 0
o ".
... ". 0
o
.c 3 12 60 u
u pp[
0
£D
2 8
.
\..-co 4
0 0
5 7 10 II 12
CO 2 I percent
Figure 6.
Typical smoke and gaseous emission
characteristics for a residential
unit in the tuned condition (Ref. 2)
21
-------
.L Smoke 1 3001 Carbon Monoxide 11 ::1 HC Emissions .{15
IIOJ- 1T
'V~-
Phase I II 10
70 As found . .
8 . Tuned.o 6 80
~ 5 10
7 . a 30 60
~ CI '0
- 0 0>
E
j 0 24 .-- 0 .-
z 6 ... Q cu 0 40 cu
...... ::J Q 5 ::J
QI :f! 3"'- ...... ....
~ 0> .D 6 0>
o
E ,,; 20 ~ """
5 - ,,; 20 -
II) c: 0 u
o u c: ::J::
-iii 0 .2
.c: II> 0> 0~3 DI
u 'E 16 II>
0 rf '"
4 - 'E. 2,0
0 IJJ 2
.c: Mean Values . '. W
N u 0 '.
0 .
N m As Found u 12 . u
.... :I: 1.5 ..
. 0 0.2
..
000
66
00
.00
20
100-
.8 cf>
1.0
Mean Values ....
-~ 000
As Found . oOC
.....~o
0,5 Tun~~..,~~6
....~~
i~oO .
00
0.1
Tuned.
.... &)(X)()O()()
6
~066
00
60 80 100
Percent of Units Having
(Mean'values
o
100
Emissions Less Than or Equal to Stated Value
exclude units in need of replacement)
Figure 7.
Distribution of smoke, CO, and HC emission for residential units (Ref. 2)
-------
9
C
QJ
U
& 8
N
o
u
N
W
13
Carbon Dioxide
12
II
. Phase .1 1i
As found 8 .
Tuned 0 6
10
o
~I»
I»M"
~
Mean Values ~~
Tuned a.Oo,;8A.
As Found 6 .
....
i~
6~ .
.
7
6
o
5
08
.48
30
. ;
40
NDx Emissions
36
32
o 28
0>
8 24
"-
,a'
. ~
,,*0
Mean Values ..~~
Tuned oeP6P:P
As FounJodJ...8.
00.--
oIfjf/'8~
100
Filterable Particulate'
16
8.
o 14
o
Q
"-
,a 12
GJ
~ 10'
~
.~
o
Q. 8.
GJ'
:c
o
~ 6
~
4 .
Meon Values
Tuned . 8"""
2 As Found 8 ~ ~6
~~~80A!.~
o . .
o 20 80
Percent of Units Having Emissions Less Than or Equal to Stated Value
. CMean va1ues exc1ude units' in need ofrep1acement)
Figure 7 (Continued).
100
00
Distribution of C02' NO , and filterable particulate emissions for
residential units (Re~ 2)
J#.
~
..
2.5
2.0
8
-
CII
:II
IL.
1.5 ~
~
-
~
1.0
0.5
o
100
~/
"-",
-------
Flue gas C02 concentrations for about 80% of those units field tested
in Ref. 2 were linearly distributed between about 6-1/2 and 10% C02'
which con'espond to 130 and 50% excess air, respectively. The research
optimum hE!ad (Ref. 5) was capable of operating smoke-free in the neigh-
borhood of 10% excess air (13-1/2% C02) when fired in research combus-
tQrs. Allowing a small margin for seasonal degradation, it was antici-
pated that burners fitted with commercially produced optimum heads could
be tuned to operate with as little as 15% excess air (13% C02)' It was
also expected that emissions of CO and UHC would be acceptably low (i.e.,
below the "as found" mean values of approximately 1 g CO/kg fuel and
0.1 g UHC/kg fuel noted in Fig. 7) at the optimum burner's tuned
condition.
In addition to a higher C02 level, it was anticipated that use of the
optimum head would effect signifi~ant reductions in NOx emissions. When
fired continuously at 10 to 20% excess air conditions (13-1/2 to 12-1/4%
C02) in research combustors, the optimum burner produced approximately
35 to 40% less NO than the average of several stock commercial burners
(Ref. 5). More data were taken, in that study, with the tunnel-fired
burner ori~ntation than with the side-fired orientation, 50 the stated
NO reduction is weighted toward the former configuration.
Side-fired
NO emission levels tended to be proportionately increased (by 50 to 100%)
over the tunnel-fired levels, although there was wider diversity among
the stock hurnersand their, average NO emissions may have increased
somewhat less than did the optimum burner's. This means that the r,educ-
tion in NO emissions to be expected by retrofitting predominantly side-
fired existing burners with optimum heads is relatively uncertain but
might be a:i much as 35% on the average.
SELECTION (IF COMMERCIAL FABRICATION METHODS
The task of selecting candidate methods, for fabricating the commercial
prototype Yeads commenced with a comprehensive 'assessment of current oil
24
-------
burner head manufacturing techniques. The evaluation included consider~
ations of unit costs (high and low production rates), saleability, manu-
facturability, and design compromises or advantages, held within the
restrictions of a retrofit application to existing burner/furnace sys-
tems. It became apparent early in the evaluation that any multiple-
piece assembly or tooling would increase the mass production unit cost
significantly, and the assessment soon narrowed to considering seriously
only the one-piece design options.
The available options were reduced to what seemed to be the best three
methods. A summary of these three methods is presented in tabular form
in Table,4. The evaluation summarized there is based on a I-year service
life and includes considerations for both the prototype units and the
mass production units. The selection criteria were weighed primarily'
on 100,000 units/year production rate with the higher 1,000,000 un~ts/
year figures included as reference values.
The final column of Table 4 ranks the three fabrication methods in a
numerical order of preference for commercial production of optimum heads.
Recommended order of preference is: (1) stamp forming of sheet metal,
(2) cast forming, and (3) injection molding. The stamp-formed, sheet-
metal method of fabrication was selected as the first choice primarily
on the basis of its design versatility in both fabrication and applica-
tion. This versatility enhances saleability of stamped sheet-metal heads
with respect to those made by the other two methods. In the fabrication
phase, the stamp-form tooling will allow changes in the material type
and also the material thickness with only minor readjustment of the basic
tooling. The design can also be made to incorporate some options that
will enable it to: (1) fit a number of blast tube sizes (nominally 0.1 m
diameter), and (2) accommodate a wide range of firing rates (0.5 to 3.0
ml/s). In addition, the sheet-metal method was ranked either best or
next to best in most all other categories listed in Table 4. These design
features are discussed in detail in a following section describing the
candidate commercial prototype optimum head.
25
-------
Table 4.
CUJ."ll'AK1~UN Ul" CUMMt;KC1AL l'AJJK1CAT1UN METHODS FOR OPTIMUM BURNER HEADS
Fabrication Method
Compar i son Items Iron Casting Sheet Metal Stamping Injection Molding
Material Heat resistant cast iron 430 stainless steel Alumina Ceramic
Ease of Changing Hater ial .Once pattern is made a large Can be made of any ductile Too I i n(j (jood for on Iv one
Functional Considerations
----~--~ -
Geometry Compromises
N
0\
I
I
I Ease of Varying Geometry
I .
j Fabricatio~'r;;)erances
Ease of Changing Center
Hole Diameter in
Production
--
variety of cast iron materials
can be used.
I. High temperature scaling
resistance may be prob-
lem wltn regular grey
iron.
May require heat resistant
alloy.
Will run cooler than sheet
J11!'-~~-'----
I. Spiral vanes to simplify
production process.
2. Shorten vanes and remove
from center hole to make
more castable.
3. Increase thicknesses and
provide fillet radii to
allow cast ing.
4. I ncrease vane angle to 30°.
I. Can machine off vanes or
enlarge center hole.
2. May be able to weld or
braze additional mate-
rial on vanes or hole.
I. Production tolerances
~0.3 mm (0.03 inch)
2. Vane angle +0.5°
3. May require-machining
0.0. to obtain index-
ing step with sharp
corner radius.
Requires minor tooling cost,
~$I ODO/di ameter change
sheet material.
1. Holes around edge allow
some air leakage.
£. ~ome warpage may occur
during heating.
I. Reduction in vane length
to minimize amount of
material used.
2. Can be made to original
configuration at ~$O. 15
extra/part.
3. I ncrease vane ang I e to 30°.
I. Vanes can be bent to differ-
ent angles if required.
2. Material can be welded to
vanes or center hole.
3. Vanes or center hole can
be trimmed.
- .-----
I I. Center hole ~0.25 mm
I (0.010 inch)
2. Vane angle ~0.5°
May be able to use knockout
ring to adjust on installation
~ $2000 add it iona I tool i ng or
can change hole size in tooling
for minor additional cost
~ $1 000/ change
material due to shrinkage of
part after molding.
1. Parts wi II crack if thermal
shock is too great; must be
I veri tied during test.
2. Brittle- will break if drop-
ped on hard surface.
3. No problem with oxidation
at operating temperature.
I. Reduction in vane length
to allow molding.
2. Thicker sections and fi Ilet
radii to allow molding.
3. I ncrease vane angl e to 30' .
I. Part cannot be varied once
made.
I. Center hole ~O. 13 mm
(0.005 inch)
2. Vane angle ~0.5°.
May sag during sintering.
Might require special
support blocks.
Can change hole size in tooling
for minor,cost if optional hole
is planned for.
-------
Table 4.
(continued) .
COMPARISON OF COMMERCIAL FABRICATION METHODS FOR OPTIMUM BURNER HEADS
N
o"J
Fabrlratlon Method
Campa r I son I terns Iron' Casting Sheet Metal Stamping Injection Holding
Installation Method Hatch drill holes In tube Hatch drill holes in tube and Drill hole In tube to match
and head and attach with head and attach with sheet hole In head, secure with sheet
. drive screw. metal screw. metal screw through tube.
Expected Li fe 10 year goal, verify during 10 year goal, ver I fy dur I ng 10 year goal, verify during
test. test. test.
Determined by oxidation Detennl ned by oxidation Determined by thermal shock
resistance. res I stance. resistance.
Sales Features I. Looks rugged I. Good appearance, light and I. Good looking white part
2. Rough surface fin Ish easy to carry. with good surface finish.
2. Easy to Install.
3. Knockout center hole to
adjust air velocity.
Estimated Costs
8 Two Prototypes .. 8 $2057 8 $400 8 $2600
8 4 weeks delivery 8 4 weeks delivery 8 4 weeks del Ivery
8 100,OOOlYear $1.50 each, including $1.29 each, Including $1.65 each, Including
$10,000 tool ing $27,000 tooling $90,000 tool ing
8 I,OOO,OOOIYear $1.40 each, Including $1. I 2 each, I nc I ud I ng $0.72 each, including
$10,000 tooling $27,000 tooling $245,000 tooling
Conrnents 1. Poor tolerance on 1.0. 1. Not as easy to Install
and 0.0. without costly 2. Not fully developed
machining fabrication process,
more risk in meeting
schedule and more uncer-
tainty in production cost.
3. No existing production
facility. Must set up
related buildings; etc.,
to house equipment.
RecOlJl1!endations Number 2 choi ce because of Number I choice because of Number 3 choice becauie of
reasonable production cost, cost, lightweight, saleability, risk of part cracking, lack
and tooling cost. deve lopment versat.lll tv., In - of development versatility,
".1""00 ..,.. ',,';!!.,' lack of production facility,
low tooling cost, poss'bility large capital Investment
of using knockout ring In cen- requ I red.
ter hole to reduce In entory
requ I rements.
-------
The casting method of fabrication was selected as the second best candi-
date for a retrofit, commercial prototype head. Its ruggedness and sim-
plicity of::er very saleable features to both the serviceman and the cus-
tomer. It is a ~'ell-proved and accepted manufacturing method in the oil
burner industry. At a production rate of 100,000 units/year, its esti-
mated $1. 50 cost is competitive. However, due to the large amount of
material TE:quirec (-0.5 kg/head) and additional labor costs (minor ma-
chining), jts estimated cost at a much higher rate of 1,000,000 units/
year shows only a slight decrease of $0.10. Additional distribution
costs will also be experienced if the heavier cast heads must be shipped
over long cistances.
The injection molding method of construction was selected as third best
candidate for fabrication of commercial optimum heads. It has several
advantages, of which the very low estimated unit cost of $0.72/head would
be a omajor consideration for high output production. Many of its features
are comparable to the cast-formed head. However, there certainly would
be some developme~t effort required to produce a satisfactory final pro-
duct. This, coupled with the high initial capital investment required
for tooling, make; it the least attractive of the three fabrication
options.
DESIGN OF C)MMERCTAL PROTOTYPE OPTIMUM HEADS
Initial Stamped Sheet-Metal Heads
For the preferred sheet-metal stamping fabrication method, an optimum
head design concept was selected so that the entire head can be stamped
and formed ::rom Hat sheet-metal stock. Figure 8 is a layout drawing
illustrating the design concept. The right-hand view is a.composite
showing the plan \Oiew of the initial flat stamping before the six swirl
vanes are folded l!p and a rear view of the prototype head after folding
the vanes. This design incorporates "sprung" vanes, folded to 83 degrees
rather than a full 90 degrees, so that the 00 of the vanes' outer edges
28
-------
,
,
\.
\
,
I ~\\, [0.0089
\ \ \
~0.047 \ \ 'v >< ',)-- '\
'~ \ /' - ° 0300
0.0127 -"\\~5;>
~-~ \
0.02~ /\ ~
./ 'V0508
N
\0
SIDE VIEW
Figure 8.
,.
'-'-
, 1
0.0381DIA. -:---" j :
'1,
L.
. FOLD AT FOOT OF VANES
r TO BE 830t 2° FROM FACE
PLANE - TYP. 6 PLACES
. I 6 VANES POSITIONED AS SHOWN
I WITH OVERLAP OF CENTER HOLE
I
//--
'~
60° ;t 2°
6 PLA-cES .
.> ~ 0.00635 \
't~LACE\
I
0.1067DtA
0.0965
PLAN VI EW
NOT~: DIMENSIONS ARE IN METERS -
MATERIAL: TYPE 430 STAINLESS STEEL
Layout drawing of the stamped-formed, sheet metal
prototype optimum burner head
-------
IS slightly larger than the 1D of a burner's blast tube. This allows
snug fitting and self-centering of the head in a variety of commonly
found blast tube ,:liameters (0.102 to 0.108 mOD). . Fold tabs were added
to the outer perimeter of the choke plate for screw attachment to the
larger diameter blast tubes.
An attracti.ie fea~:ure of this design is that prototype heads can be made
that duplic,lte the essential features of the machined and welded research
optimum head testc~d before (Ref. 5). Thus, the design of Fig. 8 has the
same number of sw:.rl vanes having the same length, width, and orientation
as the research h(~ad had. Similarly, the choke plate and its simple cen-
tral circular opening simulate those of the research heads very well. The
only basic discrepancy between this design and the earlier head is the
less-than-complet(~ closure of the j oint between the head and the blast
tubej perhaps as nluch as 15% of the. combustion air could leak through the
small openings where the swirl vanes are folded away from the choke plate.
They were IHft as large as they are to accommodate making the first few
prototype hE,ads by manual shearing and folding simulations of stamping
operations. For actual commercial production, it was believed that care-
ful attenticn to stress considerations and tolerances would allow substan-
tial reduction of these openings. Similarly, the outer edge of the choke
plate was recognized as being ~ather jagged and unattractive; in a com-
mercial stamping operation, thus undoubtedly could be finished in a way
.
that would both strengthen and beautify the head as well as provide for
attachment to the blast tube.
Production design:ould also incorporate a series of partially cut con-
centric rings arou:ld a minimum size center hole, allowing the serviceman
to "knock out" rin:~s to adapt the head to any firing rate from 0.5 to 3.0
mIls, requiring stocking of only one "universal" size head in his inven-
tory for resLdential heating units.
30
-------
Revised Stamped Sheet-Metal Heads
After testing prototype heads of the foregoing design (Section V), design
modifications were made so that the heads would be less susceptible to
metal scaling and dimensional distortion caused by exposure to intense
thermal loads and temperature gradients. One principal design modifica-
tion was provision of a recessed channel section in the previously flat
choke plate (Fig. 9). The strengthening channel design was selected be-
cause it offered a minimum of compromises over the goals of the original
prototype head design. The channel design provided rigidity at both the
perimeter and near the center of the choke plate. The required tooling
was simple and amenable to mass production.
31
-------
FOLD AT FOOT OF VANES
TO BE 83°:t 2° FROM FACE
PLANE - TYP. 6 PLACES
6 VANES POSITION.ED
AS SHOWN WITH
OVERLAP OF
CENTER HOLE
....
",
,
'''",
~-_.- -...,. . -..- ...._, .-. -~- ..
/;~-,
0.10701A
'9---- .
~..
-_.~_.. _h_.
-"""---' .
Hu_t
- 0.097
/-
NOTE: DIMENSIONS ARE IN METERS
MATERIAL: TYPE 310 STAINLESS STEEL
SECTION A.J'"
,
. i 0.0015 R
:0. 20: I TYP.4 PLACES
. n- ~ ~ ~~Oi8 DIA (REF)
l 0.048DIA
0.058DIA
O.079'DIA
... 0.089 DIA
Figure 9.
Second prototype optimum burner head design with
a modified choke plate configuration
32
-------
SECTION V
EXPERIMENTAL INVESTIGATIONS
FABRICATION OF PROTOTYPE COMMERCIAL HEADS
Initial Design
A fabrication bid package for prototype sheet metal optimum heads was
submitted to several local commercial shops having sheet metal fabrica-
tion capabilities. The low bidder was selected to fabricate two pro-
totype optimum heads using commercial shop practices with minimum tool-
ing to simulate the product which would eventually result from volume
stamping operations.
The initial prototype design of Fig. 8 was used. One head was made of
18 gage [0.00127 m (0.050 inch)] and another of 21 gage [0.00087 m
(0.034 inch)] Type 430 stainless steel sheet. The thicker (18 gage)
prototype optimum head was the primary design choice and was expected
to endure the testing schedule with little or no degradation. The
second (21 gage) unit was built to explore the effects (and limits) of
thinner (i.e., lower cost) stock material construction. A photograph
of these two initial heads is shown in Fig. 10.
Revised Design
As described and discussed in the next subsection, the initial sheet
metal prototype optimum heads experienced substantial thermal distor-
tion and exhibited inadequate resistance to scaling of the metal.
Therefore, the design was modified to strengthen the choke plate, and
a more refractory grade of stainless steel was selected.
33
-------
t
~~
IJ,)
.f:'o
;/<
\C,
~ III
III
-- -- ---
\
,
AI
;'II
~~"
'(i; ,~~1 tii
'III
"1
~
.
~
\Ii
,
~~:'-~
~,
.... ~",,;;;
w;"1W~ .. ~ "'1'IiI ',...",
",.
I
f$'".. ..""~~~
1IIIi~~.
t
w 21 ga (O.0008m)
18 ga
-- -
50P44~9/2/75-S1B
Figure 10.
Photograph of two initial design sheet metal prototype commercial
optimum oil burner heads constructed from type 430 stainless steel
-------
The research optimum head was made of Type 321 stainless' steel.
The
composition and some other characteristics of Types 430, 321 and some
other candidate stainless steels are listed in Table 5.
Table S.
COMPOSITION OF VARIOUS TYPES OF STAINLESS STEELS*
AISI TYPE Approximate
Stainless Cr, Ni, C, Other, Scaling Price
Steels % % % % 0 $/kg
Temp~rature, C
304 19 10 0.08 max - 900 2.20
310 25 20 0.25 max - 1125 3.30
321 18 10 0.08 max -0. 4 Ti 900 2.20
430 16 - O. 12 max - 800 1. 75
*Base metal - iron
Type 304 stainless steel has composition and scaling resistance very
close to those of Type 321, so it may be a good candidate for stamped
sheet metal heads. However, the 18 gage (0.00127 m) sheet metal pro-
totype head's choke plate is only about half as thick as that of the
research head (0.0025 m) so the warping characteristics may be in-
adequate. Therefore, it was decided to use Type 310 stainless steel
for the revised design prototype heads, even though this material costs
about 50% more than Type 304.
Two heads, one made of 18 gage and one of 21 gage Type 310 stainless
steel sheet according to the revised design of Fig. 9, were procured
subsequently from the same commercial shop which had made the initial
prototype heads. Also, because the cost of additional test units was
very low once the vendor's patterns and jigs were established, a com-
parable pair of heads was made from Type 304 stainless steel. The
Type 310 heads were considered to be the primary set. The Type 304
heads were kept for backup and, if the Type 310 heads were found to be
satisfactory, were to be exposed to cyclical furnace firing at some
35
-------
convenient time to gain at least a
the less costly Type 304 stainless
Figure 11 is a fire-side face-view
design prctotype commercial heads.
preliminary assessment as to whether
steel might also be satisfactory.
photograph of one of these revised
PERFORMANCE OF PROTOTYPE COMMERCIAL OPTIMUM
HEADS IN RESEARCH COMBUSTION CHAMBERS
The first experimental tests of the prototype commercial optimum heads
were carried out in laboratory research combustion chambers, rather
than in residential furnaces. An early comparison was desired between
their pollutant emission performance and that of the prior research
optimum hea.d. Most of the prior experience with the research head had
been in research combustion chambers, so that was the most appropriate
vehicle for such a comparison.
Experimenta.~aratus
The most common combustion chamber design in residential oil heating
units is an uncooled, refractory-lined cylindrical chamber, approxima-
tely 8 to 10 inches inside diameter, with a vertically disposed axis
and a horizontally disposed, side-fired burner orientation. Therefore,
the testing was begun with the 1 mIls (gph} optimum burner side-fired
in an uncooled O.22m (8.75 inch) inside diameter cylindrical chamber
lined with O. 03n (1. 2 inch) thick refractory fibre (Pyroflex) insula-
tion. Later, tests were conducted with the burner tunnel-fired in the
same chamber.
Figure 12 depicts schematically the tunnel-fired combustion chamber
arrangement, with. a fibre refractory liner in one end of the chamber
and a movable, water-cooled heat exchanger inserted in the other end.
The side-fired configuration was achieved simply by turning the com-
bustor end-for-e:1d, with the refractory liner, heat exchanger, blank
flange, and burn~r-port flange appropriately relocated.
36
-------
,*
~
Jj
IW ... r8 I
iii
1£~
m,
"
.,
-.. .. .. - - ----
IJ ~
:'M
~
...
.!II/iI
~.
lIP
" .4IiIfj
Ii
"
g X\ W
~b..
~
.~
III
,..
"
.,
'.ilN
'0:0""'1,,/
. ~ .
~
.,
11(" W
1
.
'II',
~.\
i 1w
11!
I
I
~
~ & .'.I'!i
!!r"
Of
"
;.<.<
~
J
II'
~
:'t'
~
"
#~
~
f'
'h
'r"
I>
ill
III
SOP34-ll/24/7S-Sl
Figure 11.
Photograph of modified prototype commercial head showing
the 0.0020 m deep reinforcement channel (18 gage, type
310 stainless steel)
37
-------
~
00
Sample
"'-
~
Heat Exchanger
t Water
A I
I ~
Refractory Insert
0.03m Pyroflex liner
(Double Thickness)
"-
"'-- Movable Heat
Exchanger
0.28 m Diameter
Combustion Chamber
SIDE-FIRED
BURNER PORT
~
Figure 12.
. Tunnel-fired Burner
Por;t End Fl ange
Experimental c~mbustion chamber and heat exchanger arrangement
-------
The water-cooled heat exchanger was used as a convenient means of
rapidly quenching the combustion product temperature and was made mov~
able so that heat exchanger position (i.e., firebox length) could be
readily varied to observe its effects on pollutant emissions. It con-
sisted of a nested double coil of 0.013 m (0.50 inch) copper tubing and
had outside dimensions of approximately 0.15 m (6 inch) diameter and
0.76 m (30 inch) length. Several semicircular baffle plates were cut
from 21 gage stainless steel sheet and were slipped between coils at
regular intervals, from alternate sides, to ensure that combustion
products passed repetitively over the coils and did not bypass around
the outside of the coils.
The research combustor was tested at an outdoor facility depicted
schematically in Fig. 13. The principal components were attached to
a waist-high steel table as shown. Not shown is a Unistrut super-
structure at the right-hand end of that table to support the vertically
mounted combustion chamber and to suspend the spiral-wound heat
ex~hanger within it. The facility was organized for rapid and easy
changing of combustion chambers, burner orientation, and heat exchanger
position. Minimum protection from inclement weather was provided by a
simple sheet metal roof over the test apparatus.
Experimental data requirements were primarily concerned with flue gas
pollutant concentrations. Concentrations of most pollutant species
were measured by conducting a continuous flue gas sample to a train of
analysis instruments located indoors in a nearby laboratory. Flue gas
smoke content was measured intermittently at the flue with a manual
smoke meter. The instruments used, analyses performed, and types of
data obtained are described and discussed in Appendix A. In addition,
the firing rate was monitored regularly by measuring the fuel oil flow-
rate, the flue gas temperature was indicated by inserting a thermocouple
downstream of the heat exchanger, and the temperature rise of the heat
exchanger coolant water was measured. Miscellaneous data taken less
39
-------
HEAT EXCHANGER
WATER COOLANT
, t
\.\~~
\~~
s'U~'\
~~,,~
C;,~
~
/" .
/'
TO C1AS ANALYSIS
INSTRUMENTS
~~
~ ,COMBUSTOR
J:CE BATH
WATEH VAPOR TRAP
BURN.ER ON-OFF
SWITCH
.-'
Figure 13.
Schematic of oil burner and research combustion
chamber test installation
40
-------
regularly were firebox draft conditions, firebox shell metal tempera-
tures, and combustion air fan characteristics.
Experimental Results
The cycle-averaged pollutant emission results are tabulated in Appendix
B by run number. The operational results are described in this sub-
section, along with a discussion of both types of results.
The first series of tests (runs 463 to 469) was made with the initial
21 gage sheet metal prototype optimum head on a burner (Fig. 14) side-
fired in the refractory-lined research combustor. The emission results
were entirely satisfactory, but the head did not stand up very well to
the thermal load to which it was exposed. The photograph in Fig. 15 (a)
shows the condition of th~ head after approximately 15 hours of
simulated furnace operation.* The view is from the bottom side of the
burner's blast tube, where the scaling and warpage were the most severe.
The heat-induced scaling showed a vertically oriented pattern with the
greatest scaling at the bottom of.the burner head. Maximum warpage of
the 21 gage head was approximately 0.0064 m (0.25 inch) from the
original face plane. The distortion was apparently caused by the flame
during the burner-on period rather than by overheating during the
standby period, since there was very little evidence of a matching high
temperature discoloration pattern on the back side of the choke plate.
Although the warpage was estimated to increase the air leakage around
the head's perimeter by about 12% of the total air flow, the nitric
oxide emission results were quite comparable with the approximately
2 g NO/kg fuel burned observed in earlier tests of the research optimum
head (Ref. 11).
Because of the thermal distortion experienced, the 21 gage head was
replaced with the 18 gage head. Also, in an attempt to relieve the
. , I
*The 21 gage prototype optimum head was fired for about 8 hours prior
to run 463 to cure a new refractory fiber lining in the combustion
chamber and that time is included in the stated 15 hours.
41
-------
~~';"*'
n
III
"
~
,.
'"
,JII/l
~
I
1
.
.
.?1
pi <'"
t'~~~.....
., . ~ I
." ,
... '~.'
",' ,~ ,11~
m
Stock
Williamson
Burner Head
w
'I!
1
~
. ,T
4*'n.
<:!~~~~.'.~. '*1
II'!
II!
"
iF~
~
-'*"~ .1
5Z231-9/5/75-51
Figure 14.
Photograph of the initial 21 gage sheet metal prototype
commercial head installed on the Williamson burner body
42
-------
'II
:/,
;~.
..
(b)
5ZZ36-9/15/75-S1A
18 gage (0.00127 .), type 430 stainless steel head
- -
~-
Q- ~IOII
~
.
~~
,~ -
It
I
I""
I
II
.
, I
I
.,
-------
thermal load on the head somewhat, the 60 degree spray angle nozzle was
rep1acec. with a 30 degree nozzle. This combination was fired in runs
470 thrcugh 481 with two different heat exchanger positions. Again,
the polJutant emissio~ results were quite comparable with earlier
experier;ce, but the heavier 18 gage head also suffered metal scaling
and distortion. Figure 15 (b) is a photograph of this 18 gage head
after aIproximately 15 hours of service and it shows somewhat more
scaling, but less distortion (-0.003 m) than the 21 gage head.
Also evident in Fig. 15 is substantial scaling of the burner blast
tubes to which the prototype sheet metal heads were attached. This
phenomenon was not studied to establish whether or not it was related
to the head distortion and scaling. However, circumstantial evidence
suggests that it was: the blast tubes had not shown evidence of scal-
ing in earlier tests of the burners' stock heads, and further scaling
was not observed in subsequent testing after the head geometry was
stabilized.
To continue with the proof-of-concept firings, a 0.0020 m (0.080 inch)
300 series stainless steel reinforcement plate was added to the 21 gage
head to combat the scaling and warping problem. The plate was sized
so as nct to change any air flow characteristics of the original sheet
metal design, especially the peripheral air leakage. Testing was then
resumed with frequent inspections between firings. Figure 16 is a
photograph of the reinforced 21 gage head after approximately 15 hours
of hot-fire service. showing only slight oxidation and no distortion
of the flat choke plate. Therefore, testing in the research combustor
was carried through to completion using this reinforced prototype head.
The nature of the test matrix conducted is shown in Appendix B.
~ariations were made, for both side-fired and tunnel-fired burner
orientations, in burner firing level, oil nozzle spray angle, and opera-
ting stoichiometric ratio. Two heat exchanger positions were used,
Several
44
-------
\
\,
, 'f~
. '. ":\
' ,~~~,_.
'vll!l~
.\. .
( ~)
-...-; ,..
'."'" 'It'
,,~"~"
~
. r
J
...,..
,....
~.
'I
..
!II
. ~ I
r;
-
5ZZ3l-9/1S/75-Sl
Figure 16.
Photograph of the 21 gage type 430 stainless steel prototype
cemmercial optimum head with a 0.0020 m 300 series stainless
steel reinforcement plate after approximately 15 heurs of
service
45
-------
with a greater number of tests conducted at the 0.75 m position. Addi-
tionally, short test series were made to investigate the effects of
turning O.e spark igniter off immediately after ignition and of shor-
tening the cycle time from 30 minutes to 12 minutes.
Operationllly, the burner with the prototype optimum head behaved the
same in t:1is combustion chamber as it had earlier with the research
optimum h~ad. Smooth burning was experienced over the entire operating
ranges wht~re emissions were satisfactory, although some noisy combusti,on
did occur on startup when the tunnel-fired burner was fired at low-
exc ess ai:r conditions. In agreement wi tho the earlier results, many
como)inatilms of design and operating variables permitted operation with
as low as 10% excess air without producing smoke exceeding No.1 on the
Bac:1arach scale.
The emiss:Lons of nitric oxide conformed to those from earlier research
opt:Lmum head testing in several ways. Variations of NO with oil spray
conl~ angle and stoichiometric ratio are plotted in Fig. 17 for both
.
sidl~-fired and tunnel-fired arrangements. As before, the 60-degree
spray ang:le gave the best overall results (1. e., operation at low
stoichiomHtric ratio with low NO and low smoke) for both burner orienta-
o .
tiolls. D:.rect comparisons of NO emissions for the two heads in the
two burne]~ orientations are presented in Fig. 18 and show some differ-
encl~s in the tunnel-fired configuration, but a very close correspond-
enC4~ betwE!en the results with the research and prototype optimum heads
in the side-fired configuration.
The prototype head demonstrated good firing rate flexibility in the 3/4
to :l-I/4 1111/s (gph) range (runs 502 to 512 and 542 to 550). Also, the
intc~rruptHd igniter tests, in which the spark was turned off immediately
aftHr ignition, showed that only slightly lower cycle-averaged NO
embsions (0 to 14 ppm) might result from adopting this change for the
burrler firing sequence in the slde-fired orientation.
This corresponds
46
-------
"0
C1I
£:
"-
::J
co
2.5
,....
C1I
::J
...
~
o
en
:::: 2.0
en
..
C1I
"0
.,....
)(
o
u
.....
"-
of-)
.....
z:
1.51.0
"0
C1I
E 2.0
::J
co
,....
C1I
::J
...
~
o
en
~
.......
en 1.5
..
C1I
"0
.,....
)(
o
u
.,....
"-
of-)
.....
z: 1.0
1.0
Figure 17.
-
1.0-30oA *
1.0-60oA
/
/
'- 1.0-90oA
J SIDE-FIRED I
--
--
...-
,...
1.1
1.2
1.3
1.4
1..5
< No.1 Smoke
- - - ~No. 1 Smoke
-
--
-
'-
1.0-60oA
1.0-30oA
--
---
, TUNNEL-FIRED I
1.1
1.2
1.3
1.4
1.5
Stoichiometric Ratio
*Spray nozzl~ callouts designate Firing Rate (gph) and Spray Angle
(degrees). The "A" signifies a hollow-cone spray.
Effect of oil nozzle spray cone angle upon flue gas nitric
oxide concentrations using prototype commercial heads in
the 0.22 m 1.0. insulated cylindrical research combustion
chamber
47
-------
2.0
'1:J
CII
.c
...
::>
lEI
Qj. ~:~
::>
.....
It-
o
D.
~
......
cr.
I - Srroke < 1
- - - SMOke ~ 1
0.75 m
. -- ..-
~ 1.0
-
)(
o
u
1:
...
Z
.Q-o
I TUNNEL FIRED
0.51. 0('
2.5 .
'1:J £:]
CII
E
::> a.
co
2.0
Gi . - 0 "-o.som
::::I ::-0 _.0 -. .
..... ...-.
It- .£:] .
0
01
~.
......
01
CII 1.5
'1:J Research Head Prototype
K (Ref. 11) Head (App.B)
o
u 0 0.75m . O. 75m
1: I SIDE-FIRED f 0 o. SOm . o. SOm
...
....
z
1.0
1.10 1.20 1.30 1.40 1.50
Stoichiometric Ratio
Figure H:.
Conparison of flue gas nitric oxide concentrations between
the experimental optimum burner head and the commercial
prctotype optimum burner head in a 0.22 m 1.0. insulated
cylindrical cembustor
48
-------
with the conclusion from earlier research head testing that an inter-
rupted spark would have no appreciable effect on NO emission levels.
It was concluded that the operational and emissions characteristics of
the initial prototype sheet metal head were close enough to those of
the research head to support proceeding with its evaluation in com-
mercial residential furnaces.
However, it was decided to do that with
; the design revised to eliminate the untenable metal ,scaling and thermal
distortion experienced with the initial single-piece design. Rather
than re-testing the revised design heads in the research combustion
chamber, however, it was decided to conduct some preliminary furnace
evaluations with the strengthened initial-design prototype head (Fig.
16) to provide data for a comparison basis.
PERFORMANCE OF PROTOTYPE COMMERCIAL OPTIMUM
HEADS IN RESIDENTIAL FURNACES
Suitability of the prototype commercial optimum heads as retrofit
devices for existing residential furnaces was investigated experimen-
t~lly. Two commercially available furnaces were selected as being
representative of a large fraction of the designs in the existing
population of residential space heating systems. New units were acquir-
ed and were tested in a laboratory simulation of field operation. The
stock furnaces' thermal and pollutant emission performances were charac-
terized before their burners were retrofitted with the prototype
optimum heads. This provided a comparison basis for evaluating the
subsequent data on the furnaces' behavior with the prototype heads.
Selection of Furnaces
A variety of information was considered in attempting to select just
two furnaces as being reasonably representative of the breadth of oil
furnace and boiler types, manufacturers, sizes, ages, burners, etc.,
existing in the United States.
It was decided that both units should
49
-------
be af the \farm-air type.
Far the mast part, the cambustars in hydranic
bailers ar!! simiJar (refractary-lined, side-fired) to. thase in a majar-
ity af warn-air furnaces, so. the additianal cast of a hydranicunit and
greater camplexity af installing and instrumenting it in the labaratary
were unwarranted far this investigatian.
The mast prevalent basic cambustar design, used in perhaps 75 to. 80% af
existing ill,its, is a refractary-lined cylindrical steel shell with the
side-fired burner arientatian.
The refractaries used in current can-
structian aTe primarily light-weight manalithic refractary-fiber
structures, but thase in existing units are still predaminantly hard
castables and firebrick.
To. match this "mastcamman" firebo.x design,
the Willian san Madel 1167-15 with a hard cast-refractary, side-fired
cambustian chamber was selected. This furnace also. has, a relatively
camman type af heat exchanger: a central, cylindrical, unlined-steel
extensian abave the firebax fram which the gases flaw aut ane side to.
the inside af an annular, dauble-walled, welded-steel heat exchanger.
Its burner, a Beckett Madel AF with a flame-retentian head, is nat so.
camman, hawever, since abaut 80% af existing burners are af nan-flame-
retention types.
The secand :nost p:revalent cambustar design in warm air furnaces is that
manufacture:! by Duquesne and supplied to. many name-brand furnace manu-
facturers.
On thl~ arder af 15% af existing warm-air ail furnaces have
these combu;tars, which are characterized by two. cancentric harizantally
disposed un insula':ed metal chambers. A pseuda-tunnel-fired burner
orientatian is used; it is fired alang the axis of the inner cylindrical
chamber. Combust:lan praducts are typically discharged through a narrow
slot alang ":he length af the tap af the, inner chamber into. the secand
(outer) chamber, \lhich is integral with the welded steel heat exchanger.
The inner chamber usually is made af a high-temperature stainless steel.
It is caaled, partially by canvection to. a small fractian af the cam-
bustian air which is bypassed around its autside, but principally by
radiation to the ~;urraunding auter ,chamber. The warm-air furnace
50
-------
coolant passes over the outside of the outer chamber before being
admitted to the main heat exchanger. Thus, the combustion chamber.
accomplishes an early part of the furnace's heat exchange. The second
f~rnace selected, a Carrier Model 53HV-156, has this type of firebox.
It is fitted with a Wayne burner having a conventional burner head.
Taken together, the combustors and heat exchangers in these two furnaces
are believed to be characteristic of well over half of the warm-air
furnace configurations found in U.S. residences. There are, of course,
a number of furnace and installation variables which make it improbable
, that a sample size of two could truly represent a population that
exceeds 13 million units. For comparable furnace designs, the burners
and their firing levels, discussed in later paragraphs, are undoubtedly
more influential variables than are combustors and heat exchangers..
Concerning furnace components, effects of aging (e.g., deterioration
of refractories, scaling of heat transfer surfaces and development
of air leaks) are probably the most distinguishing differences between
units in the field and those tested in this study. Conclusions based
upon comparison of experimental results obtained before and after
retrofitting the burner heads should not be negated by such differences.
Any given model of furnace will exhibit some variations in performance
and pollutant emissions as a result of installation and operational
differences among many residences. Probably the most influential
variable is the firebox pressure or draft. For operational safety,
nearly all residential furnaces and boilers are designed to operate
with a slight negative pressure over the fire, typically in the range
of 5 to 10 Pa (0.02 to 0.04 inch of water). Excessive firebox draft
may degrade combustion efficiency and increase emission levels of car-
bonaceous air pollutants, presumably by drawing the flame out of the
firebox and into the heat exchanger where combustion reactions are
quenched a bit too soon. Draft dampers and barometric control devices
should be adjusted to minimize the effects of installation differences,
51
-------
but they <:an not be eliminated entirely.
Transient effects associated
with starting and stopping a unit, prevailing and gusting wind condi-
tions, and rapidy changing barometric pressure are almost impossible
to normaLze ame,ng different installations. Ultimately, these dif-
ferences (Lre anticipated to some degree when an experienced serviceman
tunes an oil furnace's burner (Ref. 12). Presumably ,then, special
condition~, applicable to a burner in an existing installation would
also be applicable if it were retrofitted with an optimum low-emission
head. Thjs reasoning is probably valid for a majority of burners, but,
because seme burners are less sensitive than others to variations in
firebox draft, it is not universally applicable.
A burner \\hich is retrofitted with an optimum head becomes a "conven-
tional" type of high-pressure atomizing oil burner. There are two other
principal types of high-pressure atomizing burners in use in the United
States: (1) shell-head burners, and (2) flame-retention-head burners.
These types have been developed more recently than the conventional
head burners and, although some manufacturers' designs do not perform
significantly better than do many conventional burners, they have the
general reputation of achieving higher efficiencies and of being more
forgiving of operational peculiarities than do conventional burners.
Retrofitting rel~tively new burners, particularly shell-head or flarne-
retention-head d~signs, may not be justifiable. As discussed in
Section IV, to t:,e extent that new burners approach the limits of the
current technol0,~y, it will be difficult to demonstrate advantages to
retrofitting tha': equipment with the optimum head. Thus, comparison
of the res'llts which follow represents a severe test of the optimum
head techn)logy.
52
-------
Acquisition of Furnaces - The selected furnaces were ordered directly
from the manufacturers. In each instance, questions were voiced by
the manufacturers concerning suitability of. the ordered furnaces for
use in southern California, availability of the ordered and alternate
equipment, etc. Thereupon, the intended use for the furnaces was di-
vulged to and discussed with the suppliers. The engineering and dis-
tribution personnel of both Williamson and Carrier were very inter-
ested., cooperative, and helpful in ensuring qmely delivery of pre-
cisely the units selected. In fact, the Williamson Company cooperated
to the extent of supplying their unit cost-free to Rocketdyne for this
investigation, requesting only that they be informed of the published
results.
The firing rates for about 2/3 of the existing oil furnaces fall in the
range of 0.79<:w .1~1.42 mIls (0.75
-------
,
Jl
1 t 1 SMOKE
~GAS SAMPLE
I ,
- DRAFT
+---
----- FLUE TEMP.
t
'-
U1
~
. "
r-'lIi--r ~1
;:,:,~J 0': J
---
. -.- -- --
. ,
! ti "
! I i
I . -~- ~r i
. ,-----, I
j. R"""
~ I,
I - '-r,l:. I
i :' r j
, '.'. I
. L- --"-- -.J
Figure 19.
INSULATION
1- IrSH"""
I n~
O~5 r t
1--, n..
v.o. 6'
m $,
IT T:'
lor
, 't ,
I I, II
:! I I I J
I:~~", r.::, ::
", !;. I 1
1s1'\' ,----T
I I I I
I ;..--_.J I
i.~J1Lj .
.. -.., ,I.
(.
$
t
r 3 x 3 THERMOCOUPLE
MATRIX .
',. " '. ,', .~.". '1<~.'" ",:/~g~~ETR~BLE
'", --
: ~
...: .
,-. .-.... -'.,. ,- .- .. .
+ THERMOCOUPLE, FREE AIR
e PRESSURE TAP
DJ ANEMOMETRIC FLOWRATE
DI MEASUREMENT
ANEMOMETER
! (WARM-AIR FLOW RATE I
I
~
-
jI-
FLOW STRAIGHTENER
Schematic of the furnace performance evaluation system
-------
of instrumentation.
Basic thermal performance measurement techniques
conformed with requirements of ANSI 291.1-1972 (Ref. 7). Other instru-
mentation were added to provide enlarged understanding of furnace be-
havior and data for calculating cycle-averaged thermal efficiency.
The furnace flue thermal losses were determined by making measurements
to support flue gas heat balances. The combustion gas mass flowrate
was backcalculated from measured fuel flowrate and stoichiometric ratio
(as determined from flue gas composition measurements). The flue gas
exhaust temperature was measured in an insulated flue pipe with a
thermocouple located 0.46 m (18 inch) above the centerline of the heat
exchanger flue exit, as per ANSI 291.1-1972. Flue draft, gas composi-
tion, and smoke measurements were taken at successive 0.0317 m (1.25
inch) increments downstream of the thermocouple, respectively.
Steady-state thermal efficiency can be calculated, according to the
ANSI 291.1-1972 recommended procedure, from the steady-state flue gas
temperature and C02 concentration (see Fig. 2). During cyclical op-
eration in which steady state was not reached, values for those param-
eters 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 in-
distinguishable from those derived from steady-state measurements;
those calculated from 4-minute burner firing time data were approxi-
mately 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 fur-
nace cabinet heat losses, the meth6d* used to calcuiate cycle-averaged
efficiency was to measure the net heat gained by the warm-air furnace
*Difficu1ties experienced in actual application of this method led to
large uncertainties in the data, as discussed in the next subsection.
55
-------
coolant ar.d divide it by the gross heat input with the fuel burned in
a cycle.
This ITethod required measurements of oil flowrate, oil and
combustior. air temperatures and, for the warm-air. furnace coolant,
flowrate End temperatures at the inlet and outlet. The inlet warm air
was drawn into the furnace from the ambient outdoor atmosphere through
a 0.46 m (18 inch) square duct with an inlet flair and internal egg-
crate flo, straightener.
The volumetric air flow was measured with a
cumulativE readout, gas flow anemometer (!l%), i.e., it integrated the
total furf'ace-coolant air flow admitted during each complete cycle.
Ambient atmospheric pressure, temperature, and relative humidity were
recorded continuously at a meteorological data station located approx-
imately 15 meters from the. furnace test stand. Furnace coolant air
temperatures were measured with a thermometer at the inlet anemometer
location and at the warm-air outlet as an average of nine thermo-
couples in a rectangular grid array. The outlet ducting was wrapped
with 0.025 m (1 inch) thick fiberglass matting for thermal insulation.
The outlet back pressure was varied by means of a set of adjustable
outlet louvers to simulate various installed ducting loads.
Stock Furnace Characterizations
The Williamson MJdel 1165-15 and Carrier Model 58HV-156 furnaces were
tested in their stock configurations to characterize their thermal ef-
ficiency and emissions performance. Nearly all the firings were cycli-
cal tests, with the burner fired for 1/3 of the cycle time. Cycle
times of 12 minutes were used, primarily. Cycle-averaged data were.
obtained by: (1) firing the furnace for approximately 15 minutes to
warm it up, (2) initiating cyclical operation, (3) waiting for the
third cycle before commencing measurements, (4) collecting detailed
data during four successive cycles, and (5) taking appropriate arith-
metic averages of the resultant data. Data from these tests are tabu-
lated in Appendi< C.
56
-------
Efficiency - The measured gross thermal efficiencies for the two stock
furnaces are plotted in Fig. 20. Those in Fig. 20(a) are pseudo-steady-
state efficiencies derived from Fig. 2 as functions of flue gas temper-
ature and C02 concentration just before burner cutoff, and so are in-
dicated as being "flue gas" derived. Those in Fig. 20(b) are cycle-
averaged efficiencies derived from calculation of the cycle-averaged
heat transferred to the warm-air furnace coolant stream, and so are
indicated as being "warm-air" derived. Also shown in Fig. 20(b) is a
shaded band representing the range of efficiencies reported in Ref. 13
for tests of six different burner heads in an earlier model Williamson
furnace. (Net efficiencies reported in Ref. 13 were multiplied by the
ratio of the lower to the higher heating values of No.2 fuel oil for
the purposes of this graph.)
The correlating lines drawn through the data in Fig. 20 are all least-
squares fits. It is evident that there is considerably greater scatter
among the "warm air" data than among the "flue gas" data. This is un-
doubtedly due, in part, to greater uncertainties and experimental
errors in measuring the air flowrate and its rather modest «50 C)
temperature rise. The air stream is more voluminous than the flue gas
stream, and there are more opportunities for its flow to become striated
in both the furnace inlet and outlet measuring sections. Nonetheless,
the magnitude of the scatter seen in Fig. 20(b) is surprisingly large;
e.g., the five Williamson data points at about 1.4 stoichiometric ratio
range from 69 to 83%. It was also found that there were unexplained
shifts of 10% or more in the indicated efficiency when one furnace was
removed from the facility and another installed. A portion of this
(up to -4%) was found to be related to thermally striated flow in the
exit metering section; it was improved by placing a flat baffle up-
stream of the thermocouple matrix, but the best position and orienta-
tion of the baffle had to be determined experimentally for each
furnace's tests.
57
-------
~
VI
It!
CJ
Q)
:J
r-
1.1...
........
>,
u
c
Q)
or-
U
or-
\J1 Cf-
00 Cf-
I.LJ
~
I-
Q)
..r.
~
VI
VI
o
l-
t!)
75
70
65
60
1.0
Smoke < 1
Smoke> 1
I:
---
Williamson 1167-15
Carrier 58HV-156
60
1.4 1.6 1.8 2.0 1.0 1.2 1.4 1.6
Stoichiometric Ratio Stoichiometric Ratio
(a) Pseudo-Steady-State Efficiency (b) Cycle-Averaged Efficiency
Figure 20. Gross thermal efficiency characteristics of furnaces tested
in their stock configurations
~~
80
..-..
l-
e:(
E 80
I-
10
3
>.
u
c
~ 75
.U .
..-
~
~
I.LJ
r-
~ 70
Q)
..r.
~
VI
VI
o
I-
CJ 6S
i
. ;
I
1.2
85
.
I
I ~ i
I ~ I
I '
I'
I
.,.
1.8
2.0
-------
Some of the data scatter was believed to result from testing the furn-
aces in an outdoor facility.
Both the combustion air and furnace
coolant air supply temperature and humidity varied more than if the
unit has been tested indoors. The furnace cabinet was exposed .to out-
door air currents and winds as well as variations in solar isolation.
Resultant variations in heat losses through the cabinet undoubtedly
contributed to the data scatter, although they were estimated to be
small.
Another factor which contributed to scatter in the cycle-averaged ef-
ficiency was variation in the cycle timing. The burner-on, burner-off,
and warm-air blower-on times were controlled by a mechanical timer.
The firing interval was observed to vary by about ~S%. Cut off of the
warm-air blower was effected by a thermoswitch in the warm-air dis-
charge, which was nominally set at 46 C (115 F). For a constant ambient'
temperature, the burner-off blower-off time was fairly consistent
(~lO%). However, as the ambient temperature of the outdoor facility
went up, that time interval also increased~ To keep variations of that
interval in reasonable bounds, the cutoff temperature was manually ad-
justed to higher temperatures--up to SS C (131 F)--as ambient tempera-
tures rose. Even so, the duration of blower operation following burner
cutoff varied between approximately two and three minutes. While this
may have been a substantial contributor to the scatter seen in Fig.
20(b), it had little effect on the variation of efficiency from cycle-
to-cycle, which also exhibited quite large scatters.
The remainder of the inconsistencies were presumed to arise from meter-
ing the air flow. The anemometer was recalibrated repeatedly, replaced
once, and its application in the inlet duct was checked on occasion by
probing the inlet section with a small hot-wire anemometer. It became
apparent that some uncontrolled phenomenon was interfering with the
air-metering measurement. The scatter suggests that the degree of
influence varied from cycle to cycle;'perhaps it was caused by different
59
-------
vortex patterns lt the inlet to the warm air blower propagating up...,
stream and altering the flow pattern in the inlet duct.
In short, a substantial amount of effort was expended in attaining the
data in Fig. 20(:)) and no clear resolution of the apparent instrumental
problems was in 5ight. Comparison of Fig. 20(a) and (b) shows that, for
both furna:es, t:le mean cycle-averaged efficiencies are about 5% lower
than the pseudo-5teady-state efficiencies at low stoichiometric ratios
and about S-1/2% lower at high stoichiometric ratios. Differences of
these magnitudes apparently are characteristic
burner-firing-time/cyc1e-time ratio of 1/3, so
only the p 5eudo-:;teady-state efficiency as the
subsequent testing.
of the furnaces at the
it was decided to use
comparison basis for
Emissions ,. Cycle-averaged flue gas NO emissions from the stock
Williamson and Carrier furnaces are shown as functions of operating
stoichiome':ric ratio in Fig. 21. The NO emissions from the Williamson
furnace we:~e comparable with the average values from existing furnaces.
Those from the Carrier were about 25% higher than had been expected
from its radiation-cooled wall, modified tunnel-fired combustor.
Both Fig. :W and 21 also show that the stock Williamson furnace could
be operated with as little as 10% excess air before its smoke emissions
exceeded Bachara(~ No.1. Since its burner can already be tuned for
normal ope:~ation at excess air levels in the target range for burners
retrofitted with optimum heads, little or no gain in thermal effi-
ciency should be expected to result from retrofitting the burner head
supplie,d w:.th this furnace. The Carrier furnace, on the other hand,
produced greater than No.1 smoke at excess air settings below about
35%.
I f r(~trofi tting its burner with an optimum head were to allow
tuning for 15% e):cess air, a modest 3% increase in thermal efficiency
would be e)~ecte~,.
60
-------
Smoke < 1
- - - Smoke> 1
3.0
C') 2.5
~
......
C')
..
QJ
"0
.....
X
0
u 2.0
.....
~
~ /~iamson
.....
z:
QJ
C')
10
~
QJ
> 1.5
c:( 1167-15
1.0
1.5
Stoichiometric Ratio
2.0
Figure 21.
Cycle-averaged flue gas nitric oxide concentrations
for furnaces in their stock configurations
61
-------
Performance of FJrnaces With Prototype
Optimum Heads
After the performance of the Williamson furnace in its stock configu-
ration had been ,::haracterized, its burner head was replaced by the
reinforced proto.~ype optimum head of the initial design (Fig. 16) and
tests were made .:0 provide a comparison basis for .subsequent tests
with the s~cond prototype head design. The data obtained are tabu-
lated in kJpendi:( C, Runs 100-A, B, and C*.
Upon recei;,t of the Type 310 stainless-steel prototype heads of the
revised de:;ign (Fig. 11), they were installed on the furnaces'
burners, with tho 18-gage head in the Williamson furnace and the 21-
gage head :Ln the Carrier unit. The furnaces with the prototype opti-
mum heads were checked out, and then each was tested for several days
to measure its efficiency and emissions performance. Data acquired
are tabulal:ed in Appendix C, Tables C-I through C-3 for the Williamson
and Tables C-4 through C-6 for the Carrier furnaces, respectively.
Thereafter; moderately longer-term simulated service testing (4 minutes
on/8 minutes off cycles continuously for about 3 weeks) was undertaken,
and total test times of approximately 500 hours were accumulated with
each head. The modified design Type 310 stainless steel heads were in
excellent condition when their testing was completed. Neither the 18
gage nor d,e 21 gage material showed any signs of either metal scaling
or distortion, Fig. 22. **
*Compariscn of performance and emissions data from these runs with
those from Runs 101 through 117 and 135 through 137, made with the
revised design prototype head, shows that the two heads behaved
essentially the same in the Williamson furnace.
**The revised design prototype commercial head made from 18-gage
Type 304 stainless steel was subsequently tested for a total of
20 hours of 4 minutes on/8 minutes off cyclical operation in the
Carrier furnace. No indications of any scaling or warping problems
were evident after that exposure time.
62 ~
-------
~~
- 3fI' ""Ir- """'-- ~.-
'ill I !II . . "'...............
~ ~ ~~
I'J
r.~
>.i
'iI
.'~ 'M"
.~l'$
50P37-l/l2/~6-S1B
(a) 18 gage (0.00127 m), Type 310 Stainless Steel Head
~
~iiIiIiiiiI'...rtfIII 'I
....~. ..d I'I! "'
., t£
~
I!;,
.r III
'"
"""..
J III.
.')
:\
ij11"~~
,,
-------
Efficiencies - Pseudo-steady-sta~e efficiencies, measured for both
furnaces retrofitted with prototype optimum heaas, are plotted in
Fig. 23 togethEr with those obtained with their stock burners. De-
tailed d&ta arE listed in Tables C-2 and C-3 for the Williamson furn-
ace and Table (-5 and C-6 for the Carrier unit.
The efficiency performance of the Carrier furnace with the prototype
optimum head was essentially identical to that with its stock burner
head. The limit of smoke-free operation occurred at about 35% excess
air with both heads, indicating that neither an efficiency gain nor
loss wou]d be Experienced by retrofitting this furnace with an optimum
head.
The efficiency performance curve for the Williamson furnace with the
prototypE! optintUm head was about 1% below that for the stock furnace.
The drop in efficiency level.was attended by an increase of about 17 C/
30 F in average' of about 17 C/30 F in average. Presumably, this re-
suI ted from the. burner having been converted from a flame retention
burner to'a cOLventional type of burner when its head was replaced by
the optimum he;;d. Moreover, the retrofit prototype optimum head pro-
duced gn:ater than No.1 smoke when operated with less than about 30%
excess air. Cc,mbined, these two effects would force the efficiency of
this furnace with a tuned retrofit optimum head to be about 3% lower
than that with a tuned stock head.
Emission~~ ~ Cycle-averaged NO emissions from the Williamson furnace
with the prototype optimum and stock burner heads (Table C-1) are
plotted in Fig. 24 and similar results for the Carrier furnace (Table
C-4) are shown in Fig. 25. Both furnaces with the optimum heads pro-
duced about 1. ~ to 1. 6 g NO/kg fuel burned, which is substantially
below the apprc,ximately 2 g NO/kg fuel level experienced in the re-
search c(lmbustc,r experiments (Fig. 17). As a result, it is seen that
the prototype c'ptimum heads reduced NO emissions by 15 to 20% from the
Williamson furr..ace and by 20 to 25% from the Carrier furnace.
64
-------
---- SMOK~ ~ 1
OWllllAMSON (STOCK)
. WilliAMSON/OPTIMUM HEAD
SMOKE < 1
V' CARRI ER (STOCK)
~ CARR I ER/OPT I MUM HEAD
85
~ ~ 0
WilliAMSON/STOCK
V)
-
u
Z
LIJ
U
LA..
LA..
LIJ
...J 75
~
ex:
LIJ
::I:
~
V)
V)
0
ex:
CJ
70
65
1.0
Figure 23.
1.2
1.4 1.6
STOICHIOMETRIC RATIO
2.0
1.8
Comparisons of pseudo-steady-state thermal efficiencies from
the Williamson 1167-15 and the Carrier 58HV-156 furnaces
using their stock burner heads and prototype commercial
optimum burner heads .
65
-------
II 3.0 I n II II
I .
I Smoke < 1
II l ~-~- Smoke> 1 II
"0
Cl.I
c 2.5 I
~ ! I
c:.o I I I i
I .
~ Carrier Burner Head;
QI
:3 I
u.. I I
'+-
0
1;;, 2.0
.>0::
........
cr.
~
Cl.I
:3
u..
I
~ ----H-1 ; ~
lilli'
Williamson Burner Head
Smoke < 1
"U .
Cl.I
E 2.5
~
~: 2.0
........
0>
2.0
~
Cl.I
"0 .
......
X
o
u
''-
~
.....
/
I
'1 I
CommerCla Proto-
type Optimum Head
I I i I
1.5
C1\
C1\
''-
z:
1.0
1.0
1.5
Stoichiometric Ratio
Figure 24.
Effect of the commercial prototype
optimum head upon cycle-averaged
nitric oxide emissions from the
Williamson 1167-15 furnace
.
Cl.I
-c
x
o
u 1.5
-r- '.
'~ .
.....
''-
z:
Figure 25.
1.0
1.0
1.5
2.0
Stoichiometric Ratio
Effect of the commercial prototype
optimum head upon cycle-averaged
nitric oxide emissions from the
Carrier 58HV-156 furnace
-------
Cycle-averaged emissions of CO and UHC are also listed in Tables C-l
and C-4 for the Williamson and Carrier furnaces, respectively. Com-
parison of the data for the stock units with those from the corres-
ponding optimum head retrofitted units reveals that these emissions
were increased somewhat by retrofitting the Williamson unit while they
were essentially unchanged by retrofitting the Carrier unit. Broader
comparison with the Ref. 2 field survey data, summarized in Fig. 7,
shows that the stock Williamson emissions of these pollutants were
exceptionally low, while those from both retrofitted furnaces at 35%
excess air were lower than the tuned condition averages of those
surveyed in the field.
DISCUSSION
The foregoing results from tests of the prototype optimum head as a re-
trofit device for existing residential oil furnaces are discussed in
this section in terms of potential impact on thermal efficiency and air
pollutant emissions of existing installed space heating units.
Efficiency
A convenient method of comparing furnace efficiencies is to superimpose.
general furnace population behavior and individual furnace operating
lines on the efficiency decrement curves of Fig. 2. This is done in
Fig. 26. The general behavior of a large percentage of existing oil-
fired residential heating units (estimated to. be 80%, from data in Ref.
2 and 10) is indicated as a shaded zone. The average of all existing
units is estimated to be in a smaller crosshatched zone imbedded in the
shaded zone.
Boundaries of the crosshatched zone conform to the esti-
mated average operating conditions for all existing residential fur-
naces, discussed in Section IV. Old oil-fueled equipment, including
67
-------
units convertnd from coal, tend to operate toward the upper and right-
. .
hand regions of the shaded zone, while newer equipment tends to perform
toward ':he Imler and left-hand portions of that zone. Obviously, a
great many units operate outside of the shaded zone, and they are dis-
tributed around it on all sides.
Operating curves for the test furnaces are also shown on Fig. 26. (Be-
cause o:f the different plotting basis, the efficiencies versus stoichi-
ometric ratio indicated by these curves differ slightly from the cor-
relating 1ine~; in Fig. 23.) As-might be expected with new furnaces
conform:.ng to contemporary design practices, the burners in both stock
furnace:i could be tuned for normal operation (e.g.. the point corre-
sponding to a No.1 cycle-averaged smoke reading) at significantly lower
excess air le\'els than can most existing residential oil furnaces.
The WiLiamsor. unit was especially impressive in that regar~. being cap-
able of operating satisfactorily with as little as 15% excess air (13%
C02)' This capability is undoubtedly attributable to its flame reten-
tion hec:.d burner and a good match between the burner and firebox. Fur-
ther, tt.e net temperatures of the stock Williamson's flue gases was on
the low-side of the shaded band in Fig. 26, so that the unit could
achieve an estimated steady-state efficiency of nearly 84%. Obviously,
the performance capability of this stock furnace .left no margin for
efficier.cy improvement via retrofitting with an optimum head.
Indeed, retrofitting the Williamson furnace with the prototype optimum
head re::ulted in both a significantly higher excess air requirement and
somewhat higher. flue gas temperatures, so that achievable steady-state
efficiency was lowered by about 3 percentage points. Approximately 2/3
of that decrement was caused by the higher excess air requirement and
the other 1/3 by the increased exhaust temperature. Both of those com-
ponents of the total effect upon ef~iciency
replacing an effective, well-designed flame
the combustor with a conventional type head
design.
were undoubtedly caused by
retention head matched to
of universal application
68
-------
'"
\C
35
30
"*
.
>-
u
z
w
u 25
u..
u..
w
-l
«
~
a::
w
:I:
t-
In
In
0
a::
c.::!
:z
t-
:z
w
~
w
a::
u
w
0
10
15 14
I I .
, . I
1.8
I . I . I
2.2
1.4 1.6
STOICHIOMETRIC RATJO
8
7
13
12 11 . 10 9
VOLUME % C02 (DRY BASIS) IN FLUE GAS
Figure 26.
65
70 0'<
.
>-.
u
:z
w..--
-0
u UJ
-~
u.. ::>
u.. In
75 UJ ~
w
t- In
«In
t-o
In -l
I
:- '.::
<-
UJ en
80 t;; j
0"*
w N
t--
«
~
t-
In
UJ
I
85
90
6
Steady-state thermal efficiency relationships for residential furnaces
-------
It is informative to review the data reported in Ref, 3, wherein
six
different burne,r heads were tested on a single burner in another model
of Willi(~son furnace. The stock burner head was.of the conventional
type, as were four of the other heads. Their measured cycle-averaged
efficiencies ranged from 70.5 to 76.6% and averaged 73.9%. Concur-
rently, their average operating stoichiometric ratio, set by tuning for
No.1 ter.th-minute smoke, was 1. 6 (:to. 20) . By contrast, the sixth head
(a flame- retention type) could be tuned to operate at 1.19 stoichiometric
ratio, wt.ere it achieved 83.0% cycle-averaged efficiency. That reten-
tion head appears to have capabilities comparable with those of the
stock flame-retention head Williamson burner tested in the current pro-
gram. The 9% decrement between it and the average of five conventional
heads was three times as large as that between the stock Williamson
burner ar:d the retrofitted prototype optimum head. Two conclusions may
be drawn from this. First, a decrease in performance definitely should
be expected if one of the better flame-retention heads is replaced by a
conventicnal burner head. Second, the magnitude of that efficiency de-
crease may be substantially smaller if such a retention head is replaced
by an optimum low-emission head than if it were replaced by any of those
five conventional heads. By inference, a corollary to the latter con-
clusion is that the efficiency of a furnace which now has a conven-
tional head might be increased by retrofitting it with an optimum head
and a simultaneous reduction in NO emissions achieved.
x
How large such 3fficiency gains from retrofitting conventional burner
heads might be is the next question to address. There are few quanti-
tative data to consider. With the Carrier furnace, essentially identi-
cal performance was observed with the stock and optimized low-emission
heads (Fig. 26). Steady-state efficiency with each head was approxi-
mately 77% when the burner was tuned to operate at 1.35 stoichiometric
ratio. That op'~rating condition coincides with the achievable stoichi-
ometric ratio fi)r the optimum low-emission head in the Williamson fur-
nace. In the e:ulier development of the optimum head, as a part of an
optimum blrner, it could be tuned to stoichiometric ratios of 1.15 or
70
-------
lower. What can be achieved in t~is regar~ also depends upon the de-
signs of the furnace's firebox, heat exchanger, and the transition be-
tween them, as well as upon the burner firing level and firebox draft
condition. Thus, it should be expected 'that the optimum head could be
tuned to a range of excess air levels corresponding to variations among
many residential installations. In light of its known tunability to 10
to 15% excess air in some cases and 35% in others, a conservative esti-
mate is that the 35% excess air level represents a reasonable average
retrofit optimum low-emission head operating condition.
A second crosshatched zone is plotted on Fig. 26 to designate the prob-
able location of the average operating conditions for a large number of
retrofitted existing units. This zone represents a projection of the
preretrofit crosshatched zone from an average of 90% excess air to an
average of 35% excess air by following the slope treads of individual
furnace operating lines, rather than following the flue gas isotherms.
The steady-state efficiency level of this second crosshatched zone is
about 6 percentage points higher than that of the first, existing fur-
nace population zone.
Air Pollutant Emissions
Smoke emission data for the test furnaces, tubulated in Appendix C and
indicated in Fig. 23 through 26 as being less than (solid curves) or
greater than (dashed curves) No. I on the Bacharach scale, are all
cycle-averaged values. It is known (Ref. 3) that burners tuned to a No.
I smoke reading at steady-state according to recommended practice (Ref.
12), typically have cycle-averaged smoke readi~gs between No.2 and No.
3. Since cycle-averaged rather than steady-state, smoke readings of
No. I or less were used above to select 35% excess air as an average
condition to which furnaces retrofitted with optimum burner heads could
be tuned, this choice is conservatively high. Cycle-averaged emission
levels for the other carbonaceous air pollutants, from both test
71
-------
tJ
furnace tuned.ta that canditian, were less than the tuned average
levels reparted fram the field survey .of Ref. 2.
The emis sian h~vels .of NO fram the test furnaces, tuned ta 1.35 staichi-
ametric ratias, were reduced an average .of appraximately 20% when they
were retrafitted with .optimum heads. That is .only abaut .one-half .of the
reductian anti:ipated from the earlier tests .of the research .optimum
head in resear:h cambustors (Ref. 5). It is, nanetheless, an appreci-
able reduction which, together with patential efficiency gains, makes
cammercializatian .of the .optimum head attractive as a retrafit device.
Patential Applicability
The majar paint in favar .of developing the optimum head far retrafitting
existing burneTs is its patential far increasing thermal efficiency and
lawering fuel cansumption. There are undoubtedly ather existing burner
heads, particularly some .of the better flame-retentian heads that cauld
alsa be used a~j efficiency-impraving retrafi t devices. Hawever, nane .of
them is blawn ta offer the other potential benefit .of simultaneausly
lawering the emissions of .oxides .of nitragen. Thus, the .optimum head
investigated he:re is singularly unique as a candidate retrafit device
far simultaneat.sly reducing fuel cansumption and air pollution.
The incentive far a particular hameawner ta retrafit his ail heating
system's burneI' with an .optimum head will be manetary, i. e., the sav-
ings which can be realized because fuel cansumptian is reduced. An
average ~;ystem, operating at 62% seasan-averaged efficiency, might burn
1300 gal] ans .of Na. 2 ail in a season at a cast .of nearly $600. If re-
trofittirg were to increase the unit's efficiency by 5 percentage paint~.
fuel con~umptian wauld be reduced by 7% (Fig. 3), carrespanding ta an
annual savings of abaut $42. Thus, a retrafit head cast and installa-
tian expense tatalling as much as $38 cauld be recovered in a single
heating seasan by the "average" hameawner (a 10% "cost .of maney" charge
has been deducted).
72
-------
The same unit could recover a $38 total installation cost in three heat-
ing seasons if the efficiency increase due to retrofitting were as small
as 1.6%. A lower installation cost could be justified by even lower
efficiency gains. Ifowever, it seems unlikely that a homeowner would be
satisfied that it was a good investment if the payback were prolonged so
that it was obscured by year-to-year climatic variations. It is prob-
ably inadvisable to retrofit any burner unless an efficiency increase of
1-1/2 to 2 percentage points or more can be assured. Referring to the
furnace operating lines on Fig. 26, a 1-1/2% gain is indicated, on the
average, by reducing the excess air level from 45% to 35%. Thus, any
burner capable of being tuned to a stoichiometric ratio of 1.45 or lower
(~lO-I/4% C02) probably should not be retrofitted.
Most existing burners capable of being tuned to stoichiometric ratios
~l.45 probably have flame retention type burner heads. However, not all.
flame retention heads can be tuned so low*, so some are candidates for
being retrofitted.
On the other hand, not all burners with conven-
tional heads should automatically be considered to be retrofit candi-
dates. In particular, as exemplified by the Carrier furnace's stock
burner, those used in current construction and in relatively new units
may be exempted by their performance capabilities. Retrofitting of
any burner less than 5 years old probably should be approached with
caution.
Additionally, it is anticipated that it will not be possible to retro~
fit some residential oil burners because of basic equipment incompatabil-
ities. Low pressure atomizing burners and rotary burners are in this
category. Also, some high-pressure atomizing burners will probably ex-
hibit poor flame patterns, noisy combustion, and/or an inability to
tune for low smoke, etc., at a low enough stoichiometric ratio to be
beneficial.
As an example, it was attempted to retrofit the prototype
*The average tuned stoichiometric ratio was 1.42 for 10 flame-retention
heads and burners tested in Ref. 3.
73
-------
optimum h9ad to a Lennox Model 011-050-321-4 oil burner acquired as
the stock burne:(' in a Lennox 011-140 warm-air oil furnace. That burner
has an unJsuall:r short blast tube, a slower (1725 rpm fan, and a flame-
retention heed. \fuen tested in the laboratory, combustion in the oil
furnace with th,~ retrofitted burner was noisy and excessively smoky.
SatisfactJrY op~rating conditions could not be found, so the tests
were terminated without any data being recorded. From data on burner
types in Ref. 2, it may be estimated that as many as 25% of the exist-
ing installed r,~sidential burners may fall in this category. Combining
this with an estimated 5% as high-performing retention head burners and
another 10% as :ligh-performing conventional head burners leaves a bal-
ance of approxi:nately 60% of existing residential oil burners ,...hich
might appropriately be retrofitted with optimum low-emission heads.
In summary, it 5hould be beneficial to retrofit 50% or more of existing
U.S. resi.:iential space heating oil burners with optimum low-emission
heads. The pri:lcipal benefit would be modest increases in steady-state
thermal efficie:lcies, and these should translate directly to equivalent
increases in seison-averaged efficiencies. If the distribution of ac-
tual initial efficiencies among the units retrofitted were identical to
the distributiol among all existing units, an average of about 5 effic-
iency points sh)uld be gained by retrofitting. There are many existing
burners, however, for which retrofitting would not improve efficiency
appreciably. If these were identified and omitted from the retrofitting
program, then the average initial efficiency for those which are modi-
fied would be lower than the overall average, and this would provide a
margin fer achieving greater than a 5% average efficiency gain.
Because cf similarities in combustion chamber designs and burner orien-
tations, the oftimum head technology is believed to be equally as appli-
cable to hydronic boilers as to warm-air furnaces.
74
-------
Unresolved Issues
There are several subject areas related to successful commercialization
of optimum low-emission burner heads as retrofit devices which have not
been considered in this research program. The first is that retrofit-
ting a residential oil burner with an optimum head makes it into a dif-
ferent burner and this may obviate whatever certification it may have
had concerning conformance with national, state, or local building, fir~
and safety codes and standards. The magnitude and potential solutions
for this. problem need to be defined as an early part of any serious com-
mercialization effort.
A number of allusions have been made in preceding subsections to re-
strictions on the applicablity of the optimum heat technology. It can-
not, in fact, be applied indiscriminately to all residential oil burners."
A corollary is that heating industry service personnel, i.e., those who
would actually effect retrofitting of existing furnaces, must be able to
discriminate between those burners which should and should not be mod-
ified. They will need to be more sophisticated than the average service
man now is in utilizing the adjustment guidelines, such as Ref. 12, in
determining whether a sufficient potential for higher efficiency exists
to justify changing to the optimum head, and in tuning modified burners
for minimum pollutant emissions and best efficiency. A commercial man-
ufacturer of optimum heads would need to assemble and supply to the oil
heating service industry a range of background information such as
reco~nended retrofit procedures, guidelines concerning burners built by "
many manufacturers, and guidance in selecting and using adequate instru-
mentation. Success of a retrofit program might even depend upon provid-
ing formal personal training of service personnel.
75
-------
Conunercialization of optimum burner heads will require serious consider-
ation of a numher of logistics problems. They range from determining
the minirmm number of optimum head designs needed for modifying many
manufacturer's burners with various firing rates, and determining the
appropriate production rate for each design to establishing distribution
and mark(~ting ~;ystems. Some of these have been included in the recom-
mendations, although they are not discussed further here.
76
-------
. SECTION VI
REFERENCES
1.
HaIl, R. E., J. H. Wasser, and E. E. Berkau, "NAPCA Combustion Re-
search Programs to Control Pollutant Emissions From Domestic and
Commercial Heating Systems," New and Improved Oil Burner Equipment
Workshop, NOFI Tech. Publ. 108 ED, National Oil Fuel Institute, Inc.,
New York, New York, September 1970, pp 83-93.
2.
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~ Re-
search Triangle Park, N.C., June 1973.
3.
Hall, R. E., J. H. Wasser, and E. E. Berkau, "A Study of Air Pollut-
ant Emissions From Residential Heating Systems," EPA-650j2-74-003,
Environmental Protection Agency, Research Triangle Park, N.C.,
January 1974.
4.
Martin, G. B., and E. E. Berkau, "Evaluation of Various Combustion
Modification Techniques for Control of Thermal and Fuel-Related
Nitrogen Oxide Emissions," presented at the Fourteenth Symposium
(International) on Combustion, Pennsylvania State Uni versi ty,
August 1972.
5.
Dickerson, R. A., and A. S. Okuda, "Design of an Optimum Distillate
Oil Burner for Control of Pollutant Emissions," EPA-650j2-74-047,
Environmental Protection Agency, Research Triangle Park, N.C., June
1974.
6.
Stanford Research Institute, "Patterns of Energy Consumption in the
United States," Office of Science and Technology, Executive Office
of the President, Washington, D.C., January 1972.
('~"
(i
7.
"American National Standard Performance Requirements for Oil-Powered
Central Furnaces," ANSI Z91.1-1972, American National Standards
Institute, Inc., New York, N.Y., June 1972.
77
-------
1')
'.J
8.
Hydronics Institute, "Testing and, Rating Standard for Cast Iron and
Ste.el Heating Boilers," Hydronics Institute, Berkeley Heights, N.J.,
March 197".
9.
DeWerth, D. W., and J. F. Dicaprio, "Performance Tests of Residen-
tial Hot Water Boilers; Gas vs Oil," Catalog No. H00220, American
Gas Association, Cleveland, Ohio, January 1966.
10.
Tur1er, D. W., et al., "Efficiency Factors for Domestic Oil Heating
Units," Addendum to the Proceedings of the Conference on Improving
Efficiencf in HVAC Equipment and Components for Residential and
Small Co~~ercial Buildings, Purdue University, Lafayette, Ind.,
Oct,)ber 1974.
11.
Com')s, L. P., and A. S. Okuda, "Residential Oil Furnace Optimiza-
tio:1-Phasl~ I," EPA-600j2-76-038, Environmental Protection Agency,
Res~arch Triangle Park, N.C." February 1976.
12.
"Guidelinns for Residential Oil Burner Adjustments," EPA-600j2-75-
069-a, Environmental Protection Agency, Research Triangle Park,
N.C., October 1975.
13.
Howf~kamp, D. p., and M. H. Hooper, "Effects of Combustion Improving
Devices on Air Pollutant Emissions From Residential Oil-Fired Fur-
naces," Paper No. APCA 70-75, Annual Meeting of the Air Pollution
Control A:;sociation, St. Louis, Mo., June 1970.
78
-------
APPENDIX A
FLUE GAS COMPOSITIONAL ANALYSIS
The sample flow train used for analyzing flue gas composition is illus-
trated in Fig. A-I. 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-I gives
a su~nary of the gas analysis instruments used. The gas leaving the
glass-wool filter was
directly to the total
a Drierite bed where
split into three parallel paths. One path led
hydrocarbon analyzer. A second path led through
water vapor was removed, then into the series-
plumbed CO, C02' and 02 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, C02' and
02 analyzers.
When the analytical system shown in Fig. A-I is used to analyze gases
which may have been quenched before combustion was completed, 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.
79
-------
~
~ ---......
Flue Gas
Samp1e
Smoke Sample
Glass Wool
Filter
00
o
I
Condensib1es
Traps
~
Figure A-I.
Ice-Cooled
Nitric Oxide
Ana1yzer
3A Mo1ecular
4. ~ c:::-...- ---'
I f~~iD~i;;i;~"Bed
~~ .
Total UHC
Analyzer
Vent
Carbon
Monozide
Ana1yzer
Oxygen
Analyzer
Carbon
Dioxide
Ana1yzer
Analytical system for fuel oil burner emissions analysis
-------
Table A-I.
EXHAUST ANALYSIS INSTRUMENTS
co
......
CO C02 NO Total HC Oxygen Sr.\oke
Type MSA MSA ~fSA ~iSA Beckman Bacharach
Nondispersive IR Nondispersive IR Nondispersive IR H2 flame polarographic (manual)
LIRA LIRA LIRA ionization
~!odel 300 ~Iode 1 300 ~del 200 detector
Range 0 to 1500 ppm 0 to 20 mole \ 0 to 500 ppm 0.2 to 800 ppm 0 to 100~. 0 to 9
fnole) (mole) total HC by
volume as Cf!4
Sensitivity 30 ppm minimum 0.25% minimum 10 ppm minimum 10 ppm minimum -0.1 % 1
detectable detectable detectable detectable
Calibration 1000 ppm CO in 14% C02 in N2 0.82\ C2H4 in 3\ CH4 in helium Ai r - 21',; Ten spots of
N2 standard gas standard gas "'2 used as used as a monotonically
slmulant for standard N = 0'. varying
4l0-ppm.NO 2 darkness
standard
-------
Values calculated from the measured flue.gas .compositional data included:
the overall stoichiometric ratio, the weight of nitric oxide per unit
weight 0:: burned fuel, and the weight of carbon monoxide per unit weight
of burnell fuel 0 The method of calculation to obtain these values is
describe oxidized to water and condensed out of the system at the cold
trap, prier to analysis.
82
-------
An oxygen balance yields:
0.2095 AIR = C02 - 0.0003 AIR + 0.5 CO + 0.25 x (C02 + CO
- 0.0003 AIR) + 0.5 NO + 02
(A-I)
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 C02' less the amount of C02
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-I can be arranged to yield:
AIR
=
x . x
(1 + 4 ) C02 + (1/2 + 4 ) co + 1/2 NO + 02
0.2095 + 0.0003 + 0.0003 x/4
(A- 2)
A carbon balance can be used to calculate the moles of fuel burned per
100 moles of dry flue gas:
FUEL
=
C02 - 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 combustion were in stoichiometric proportions, the moles of air
would be, by an oxygen demand calculation:
AIR t . h
S OlC
=
(1 + x/4) FUEL
0.2095
(A-4)
The stoichiometric ratio of the locally sampled burned gases is a param-
eter frequently used in this report. It is defined as the ratio of AIR
to AIR t . h:
s OlC
SR
=
AIR
AIR. h
St01C
(A-5)
83
-------
Combination of Eq. A-2 through A-5 yields a direct calculation of the
burned gas stoichiometric ratio in terms of the measured parameters:
SR =
(1 + ~) C02 + (1/2 + ~) CO + 1/2 NO + 02
0.2095 + 0.0003 + 0.0003 x/4
(1 +~) [ " (1 +~) C02 + 0/2 +~) CO + 1/2 NO + °2 ]
0.2095 C02 + CO - 0.0003 0.2095 + 0.0003 + 0.0003 x/4
(A-6)
According to the above definition, when the sample contains just a
sufficie"lt amount of air to oxidize all of the fuel in the sample to
C02 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 calcu-
lated fn)m Eq. A-6 does not require that the products in the flue gas
be in ch\~mica.l equilibrium.
Note tha": the accuracy of the stoichiometric ratio calculation would
be affec":ed very little if all terms in Eq. A-6 containing the facto"rs
0.0003 and NO \lrere ignored. These factors represent the carbon dioxide
originaEy pre~;ent in free air, and the oxygen tied up in nitric oxide,
respectively.
One parLally cuestionable assumption made in the formulation of Eq.
A-6 was that a]l hydrogen originally present in the fuel becomes
oxidized to water and is removed in the cold trap. This was a neces-
sary asstlmptior., since there was no instrument available to measure the
actual h)'clroger. content of the sample gas. The assumption is very good
under the combined conditions of air-rich stoichiometric ratios (SR > 1)
and chemj.ca1 ecuilibrium. To test this assumption, a Rocketdyne thermo-
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 H2' The actual stoichiometric ratios of these combustion gases,
84
-------
compared to those calculated by Eq. A-6 (which does not recognize the
presence of Hz) are given in Table A-Z, where it can be seen that Eq.
A-6 is quite aCC1ITate except for SR < 1. Calculated equilibrium condi-
tions are tabulated in Tables A-3 and A-4.
TABLE A-Z.
VALIDITY OF STOICHIOMETRIC RATIO CONDITIONS
Stoichiometric Ratio Calculated
Actual Stoichiometric Ratio from EQ. B-6
0.800 0.844
1.000 1.003
1.200 1.197
1.400 1.400
1.600 1.600
2.000 . 2.002
2.400 2.404
2.800 2.804
The primary cause of the inaccuracy at SR < 1 is the unaccounted for
presence of HZ' In nonequilibrium gases, there is likely to be Hz
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 COZ (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 %COZ to the
stoichiometric ratio, equilibriumdaha from Table A-4 were used to
calculate the curve shown in Fig. A-Z;
shown.
A number of values of measured
a calculated%OZ curve is also
COz concentrations in actual
Fig. A-Z. The measured data
furnace flue gases are also plotted on
85
-------
Table A- 3.
EQUILIBRIUM COMBUSTION GAS PROPERTIES FOR NO.2
DISTILLATE FUEL OIL BURNED WITH AIR
(CH1.814' 18,443 Btu/1b Net Heat of. Combustion With Air at 14.67 psia)
ex>
0\
Oil + Air Flame Cp Thermal
Stoith. Inlet Temp., Temperature, Frozen, y Viscosity, Conductivity, Prandt1 Molecular
l\a1.10- t t I:!tu{lb-R Frozen centipoise Btu/hr-ft-F Number Weight
0.8 0 3429 0.346 1.261 0.0666 0.0702 0.7946 27.73
1.0 3614 0.341 1. 254 0.0687 0.0711 0.7984 28.80
1.2 I 3290 0.3:;3 1.260 0.0653 0.0661 0.7954 29.00
1.4 2940 0.324 1. 267 0.0615 0.0610 0.7915 29.03
1.6 Air 2649 0.318 1. 275 0.0581 0.0567 0.7880 29.03
2.0 Rich 2209 0.307 1. 288 0.0527 0.0500 0.7820 29.02
2.4 t 1897 0.298 1.298 0.0487 0.0452 O. 7771 29.01
2.8 1663 0.291 1.303 0.0456 0.0415 0.7730 29.00
0.8 70 3778 0.347 1.261 0 . 0671 0.0709 0.7948 27.72
1.0 3649 0.3<11 1.254 0.0691 0.0715 0.7984 28.77
1.2 3336 0.333 1. 259 0.0658 0.0667 0.7956 29.00
1.4 2991 0.325 1.267 0.0621 0.0617 0.7918 29.03
1.6 2703 0.318 1.274 0.0589 0.0574 0.7884 29.03
2.0 2765 0.308 1. 286 0.0535 0.0509 0.7825 29.02
2.4 1955 0.299 1.297 0.0495 0.0461 0.7778 29.01
2.8 1-"" 0.193 1. 306 0.0464 0.0425 0.7738 29.00
1_-
0.8 200 3867 0.347 1.260 0.0681 0.0720 0.7951 27 .71
1.0 3709 0.342 1. 257 0.0698 0.0723 0.7983 28.73
1.2 3418 0.334 1.2S9 0.0668 0.0678 0.7958 28.98
1.4 3035 0.326 ' 1. 266 0.0632 0.0629 0.7923 29.02
1.6 2802 0.320 1.273 0.0600 0.0588 0.7890 29.02
2.0 2369 0.309 1.28.1 0.0548 0.0524 O. '7834 29.02
2... 2061 0.301 1.294 0.0509 0.0.177 0.7790 29.01
2.8 1331 0.29:; 1.303 0.0.179 0.0441 0."':-51 29.00
*Stoichiometric ratio is unity at 14.49 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
.00
........
Stoich. Oil + Air H 0 Ar 011 H2 H20 CO C02 NO N 02
2
Ratio Inlet Terr.p.. F
0.8 0 0.0630 0.0000 0.821 0.0499 2.016 12.263 7.243 8.687 0.000 68.837 0.000
1.0 0.0397 0.0313 0.866 0.2816 0.250 11 .690 1.393 12 .052 0.253 72.522 0.619
1.2 0.000 0.0217 0.882 0.1862 0.030 10.141 0.161 11.247 0.390 73.784 3.160
1.4 0.000 0.0000 0.890 0.0757 0.000 8.832 0.0203 9.841 0.2955 74.465 5.566
1.6 0.000 0.0000 0.895 0.0790 0.000 7.799 0.000 8.679 0.2080 74.947 7.444
2.0 0.000 . O. 0000 0.902 0.000 O.OGO 6.297 0.000 7.000 0.0829 75.603 10.107
2.4 0.000 0.0000 0.907 0.000 0.000 5.276 0.000 5.864 0.0339 76.028 11. 888
2.8 0.000 0.0000 0.910 0.000 0.000 4.541 0.000 5.046 0.000 76.326 13.161
0.8 70 0.0737 0.0000 0.821 0.0613 1.996 12.271 7.268 8..659 0.017 68.901 0.000
1.0 0.0455 0.0362 0.866 0.3072 0.269 11.647 1.501 11 .934 0.272 72 .456 0.(66
1.2 0.0000 0.0261 0.882 0.2082 0.036 10.121 0.195 11.210 0.404 73.751 3.1 S9
1.4 0.0000 0.0000 0.890 0.0885 0.000 8.824 0.026 9.835 0.322 74.447 5.553
1.6 0.0000 0.0000 0.895 0.0351 0.000 7.795 0.000 8.6:'8 0.223 74.933 7.432
2.0 0.0000 0.0000 0.902 0.000 0.000 6.297 0.000 7.000 0.096 75.596 10. 100
2.4 0.0000 0.0000 0.907 0.000 0.000 5.276 0.000 5.863 0.041 76.023 11 .8S4
2.8. 0.0000 0.0000 0.910 0.000 0.000 4.541 0.000 5.046 0.018 76.323 13.159
0.8 200 0.0964 0.0000 0.821 0.0878 1.964 12.273 7.318 8.604 0.027 68.796 o.oeo
1.0 0.0577 0.0468 0.864 0.3579 0.304 11. 562 1.710 11.705 0.310 73.328 0.754
1.2 0.0000 0.0356 0.882 0.2533 0.048 10.078 0.270 11 .1;;:7 0.451 73.683 3.162
1.4 0.0000 0.0000 0.890 0.1157 0.000 8.806 0.042 9.816 0.373 74.405 5.526
1.6 0.0000 0.0000 0.895 0;0493 0.000 7.787 0.000 8.672 0.268 74.905 7.406 .
2.0 0.0000 0.0000 0.902 0.0000 0.000 6.295 0.000 7.000 0.125 75.582 10.085
2.4 0.0000 0;0000 0.907 0.0000 0.000 5.276 0.00(1 5.863 0.059 76.015 11.876
2.8 0.0000 0.0000 0.910 0.0000 0.000 4.541 0.000 5.046 0.028 76.319 13. 154
-------
15
14
. - C02 Data from Ref. A-l
8- C02 Data from Ref. A-2
----- Calculated Equilibrium;
Ref. A-3
14
13
en
/0 en
C) 12 12 /0
C)
cu
:J cu
,... :J
LL ,...
LL
t' ~
c 11 10
c
C
or- C
or-
C\.!
0 C\.!
U 0
~ 10 8 ~
C
cu cu
u u
s.. s..
cu cu
0.. 0..
cu cu
E 6 ~
:J 9
e-
o 0
> >
8
7
6
1.0
_I
1.5
Stoichiometric Ratio
2.0
Figure A-2.
Flue gas C02 and 02 concentrations for no. 2 fuel oil
burned in ambient air at 1 atm
88
4
2
o
-------
are seen to be very well correlated by the calculated equilibrium curve
at SR > 1.1 (the calculcated maximum C02 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 CH4) 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:
g NO (1000) (NO) (MWNO)
=
kg fuel (CO - 0.0003 AIR + CO) (MWF)
2
g CO (1000) (CO) MWCO
=
kg fuel (CO - 0.0003 AIR + CO) (MWF)
2
(1000) (HC) (MWCH4)
g UHC = (CO - 0.0003 AIR + CO) (MWF)
kg fuel 2
(A-7)
(A-8)
(A-g)
where
MWNO
MWF
=
molecular weight of NO = 30.01
=
molecular weight of fuel
12.01 + 1.008 x = 13.84
=
MWCO =
MWCH =
4
molecular weight of CO = 28.01
molecular weight of methane = 16.04
For calculation of the above quantities, the term 0.0003 AIR can be
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
89
-------
calculation. The numbers given in this report inlcude the effect of
the term. The experimental data were reduced, according to the above
equation, by m~ans of a remote terminal timeshare computer program.
In addition to the gaseous pollutants described above, the smoke
content Df the mixed gases was also measured.
The instrument utilized
for this purp05e was a Bacharach smoke meter. (It is manufactured by
the Bacharach Instrument Company, Pittsburgh, Pennsylvania.) This is
a hand-held de1ice which, when pumped, sucks flue gases from a 0.006 m
Clj4-inch) OD, 'lncooled sample probe through a piec~ of white filter
paper; 10 stro:
-------
REFERENCES
A-I.
A-2.
A-3.
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.
Hall, R. E., J. H. Wasser, and E. E. Berkau, A Study of Air
Pollutant Emissions from Residential Heating Systems, EPA-6S0/
2-74-003, Environmental Protection Agency, Research Triangle
Park, N. C., January 1974.
Dickerson, R. A., and A. S. Okuda, Design of an Optimum Distil-
late Oil Burner for Control of Pollutant Emissions, EPA-6S0/
2-74-047, Environmental Protection Agency, Research Triangle
Park, N. C., June 1974. ~
91
-------
APPENDI X B
D~rA TABULATION:
RESEARCH COMBUSTOR EXPERIMENTS
Cycle-averaged flue gas composition data are tabulated for tests of the
I mIls (gph) optimum burner, fitted with various prototype sheet metal
optimum 'leads :md fired in a 0.222 m (8.75 inch) ID cylindrical
refractory-lin,~d, research combustion chamber. Notations are given in
the table to ddineate burner orientation with respect to the combustor,
combusti,)n chamber length upstream of the water-cooled copper-coil heat
exchanger, burner firing level, and spray angle. The latter two pieces
of infor:nation are contained in the coded designation for a spray
nozzle, (~.g., "1.0-60o-A" denotes a firing rate of 1.0 gph (1.05 mIls)
and a ho llow-cone (A) spray angle of 60 degrees. Cycle timing for all
tests wa:5 10 m:Lnutes on/20 minutes off except for Runs 516 to 518 which
were 4-mlnutes on/S-minutes off cycles.
92
-------
SIDE-FIRED
21-GAGE, TYPE 430 STAINLESS-STEEL, PROTOTYPE
OPTIMUM HEAD (INITIAL DESIGN) .
RII:>j STOIC. CO:? 02 CO :>j\) UHC CO :>jC UHC BACH. TF'G
NJ. RAllO : : PPM PPM ['PM GM/K c;~, G~1I K l"1 CM/K GK S~IOKE C
T: 1.25 1203 4.5 21 9::1 0.37 1 .65~ 0.008 0.2 366
464 1.46 10. I> 7.0 ::>2 93 0.45 1.9(0 0.013 0.2 429
..: E 465 1.39 11.1 6.3 22 96 0 0.43 1.915 0.005 0.3 410
I
. .,
0 ...
'" 466 10\2 13.8 2.3 3C 116 0 0.44 1.1133 0.002 0.9 332
I 0
<> .
..J
467 1.07 14" 1.5 26 110 0.38 1.6£10 0.008 0.7 332
466 1.20 12;' 3.6 25 117 0 0.40 1.999 0.000 0.0 416
469 1.16 13" 3.0 25 121 0 0.38 1.996 0.001 0.3 413
I8-GAGE, TYPE 430 STAINLESS-STEEL, PROTOTYPE
OPTIMUM HEAD (INITIAL DESIGN
RUN ST:HC. C02 02 CO NO UHC CO NO UHC BACH. TF'G
NO. RA TI 0 :t :: PPM PPM PPM G~/KGM "GM/KCI1 GM/KCM SM OK£ C
470 10\4 13.5 2.7 10 121 0 0015 1.961 0.004 0.1 318
471 1.12 13.6 2.4 10 114 0 0015 1.612 0.004 0.2 341
472 1.09 14"0 1.9 21 103 2 0.32 1.596 0.017 0.7 324
E
~ 473 1.05 14.5 1.0 139 136 0 1.93 2.026 0.002 2.0 362
0
n 474 1.44 10.6 6.7 II 129 0 0.23 2.646 0.002 0.0 435
..: ..J
. I
0 475 1.37 11.1 5.9 10 134 0 00\6 2.620 0.001 0.0 429
'"
,
~ 5.3 10 136 0 0018 2.560 0.001 0.0 432
476 1.32 \1.5 ".
471 1.22 12.7 4.0 10 136 I) 0016 2.368 0.001 0.0 393
478 1.33 11.6 ~.6 10 129 0 0.20 2.451 0.004 0.0 321
~
~ 479 1.16 13.3 3.1 10 121 0 0.17 2.093 0.002 0.2 291
~
. 4'>0 1.05 14.5 I" 130 121 3 1.61 1.814 0.026 4.5 277
..J
4<11 1.45 10.6 601' 10 1~5 0 0.19 2.596 0.003 0.0 332
93
-------
SIDE-FIRED
21-GAGE INITIAL DESIGN PROTOTYPE OPTIMUM HEAD
WITH 0.00020 m REINFORCEMENT PLATE
RUN STOIC. CJ;? ;)2 C;) :'>J" 0JHC f.:l NO UtiC F"\c;H. TrG
~,\. RI\1" I J :: :: Pt'M pp" pp~ GMhGM GM IK G~I V:/KGM 9101o:[ C
4~2 ,.~S \:>..4 4.5 II 106 O..~O I. ,<9 1 (1.006 0.3 .?B5
(' 483 1.~1 12.6 J.9 15 109 0 0.24 1.891 0.00'1 O.iI 279
"
" 4134 10I~ 12.7 3.4 17 110 0 0.28 1.856 0.004 0.4 279
,
.J
1 4105 1.23 12.1 401 15 106 0 0.~6 I.B57 0.004 0.3 279
0466 1.0(, \4.5 1 .3 167 130 10 2.35 1.969 0.080 2.0 338
oJ>
0 4SS 1.29 12.0 5.0 10 118 0017 2.164 0.006 0.0 J7i1
o
,;;. 489 1.25 12.2 4.5 10 121 0 0018 20167 0.005 0.0 . 368
"
'~ 490 1.20 12.8 3.7 10 121 0 0.16 2.069 0.004 0.0 366
,
.J
I 491 1.37 II.) 6.0 15 121 2 0.21 2.318 0.021 0.0 J8J
r 492 1.44 10.6 6.8 15 114 4 0.31 2.354 0.049 0.0 391
493 1.62 9.5 8.5 6 85 3 0.15 1.980 0.043 0.8 371
« 494 1.45 10.6 6.9 8 95 2 0.17 1.911 0.020 0.5 366
.' E
011\ 495 1.29 11.9 5.0 12 104 0 0.22 1.927 0.001 1.0 346
"'.....
I
00
.. 496 1.18 12.8 3.4 20 110 0 0.31 1.841 0.004 1.2 32<1
-...I
+- 491 1012 13.6 2.3 40 110 6 0.59 1.748 0.051 201 307
498 1.19 12.8 3.6 20 131 0.32 2.235 0.010 0.0 388
« 499 1.10 13.9 2.1 218 140 300 4.06 2.190 2.500 0.8 382
t-J\
-'T"'"
I 500 1 .27 12.2 <1.8 20 136 0.35 2.<166 0.01<1 0.0 401
00
. a
-...I
t- SOl 1.33 11.5 5.5 16 129 2 0.30 2.444 0.018 0.0 401
502 10\6 12.7 0.0 20 126 0 0.33 20114 0.004 0.0 421
«
.. E 503 1.25 12.4 4.4 16 121 0.28 2.151 0.011 0.0 435
011\
oJ>"'"
. .
00 504 1.13 13.6 2.6 11 129 0.21 2.069 0.006 0.2 421
. .
-...I
t 505 1.06 14.3 1.2 25 132 0.36 1.990 0.001 0.5 404
506 1.66 9.3 8.9 13 85 0.31 2.033 0.019 0.0 346
507 1.01 15.0 0.2 620 103 20 8.30 1.415 00153 5.7 251
508 1..<16 10.9 7.2 10 93 0.21 1.942 0.008 0.0 318
«
'&~ 509 1.24 12.7 '4.4 10 95 0 0.16 '1.674 0.005 0.2 291
.o.....
0
11\0 510 10\6 13.6 301 15 103 0 0.23 1.103 0.004 0.2 277
.....
....1
0
t- '" 1.09 14.5 1.8 30 lOB 0.43 1.671 0.006 0.8 266
512 1.41 11.0 6.5 2 97 2 0.06 1.953 0.021 0.0 310
~ ~I J 1.25 12.3 4.5 20 liB 0.33 2.111 0.008 0.2 302
:: 514 1012 13.6 2.4 ;;0 117 0 0.30 1.874 0.003 0.<1 3<11
.. 515 1.37 1101 6.0 17 114 2 0.31 2.233 0.024 0.0 374
o' r::
C'1l.,"".
oJ>""
~ Q ~ 516 1.37 1101 6.0 1<1 112 2 Q.J5 20\97 0.026 0.0 35~
.. "
-...J ,
~ 517 1.05 1<1.5 1.0 315 110 2<1
-------
TUNNEL-FIRED
21-GAGE INITIAL DESIGN PROTOTYPE OPTIMUM HEAD
WITH 0.0020 m REINFORCEMENT PLATE
RU:II ~J OJ C. C(12 ('I:~ CJ ~J u~c C:J NO UlIC BIICH. TfG
~J. "liT! 11 t ~ t"P~' PP'M t>,.,~ G~/K Ct,,, :;~ IK G~~ G~/t<(j~ S"OKE C
519 I .2~ II. ~ 04.9 II 81 0.20 1.498 (1.006 0.0 351
520 I. 'J1 14.1 1.4 25 8\ 0 0.~5 1.23.0 0.004 0.6 3.01
52\ \ .00 15.0 0.1 ~1600 86 \1 ~2\ .24 1.22.0 0.083 5.5 329
52:? I. ,35 11.3 5.1 10 81 0 0.20 1.551 0.00.0 0.0 35.0
E' 523 \.41 \0.2 1.0 \5 8\ 0.29 \.108 0.009 0.0 311
'"
r-.
C 524 1018 13. I 3.5 1":? 8\ 0 0.20 \.370 0.004 0.0 341
<: ~
'b 525 \012 \3.8 2.4 16 !15 0.25 '.352 0.005 0.0 332
",
, -
0
.- 526 loll 13.9 2.3 456 76 25 6.12 1.212 0.210 0.4 329
....
0
C 527 1.43 10.1 6.6 \6 81 0 0.32 1.669 0.004 0.0 352
en
- 528 \ .09 \3.4 1.8 1\79 75 5 16.98 \ 01 67 0.04\ 0.\ 338
529 1.18 \3.1 3.5 .06 19 0.74 1.328 0.009 2.3 257
e;
0 530 \ .40 \1.0, 6.4 60 76 \.\2 \.540 0.016 0.0 285
'"
0
a 531 \ .32 \1.1 5.4 62 72 2 \.\0 \.357 0.020 0.0 268
~
532 1.25 \2.2 4.5 41 10 4 0'.80 \.257 0.040 0.6 260
533 \ .57 9.9 8.1 20 73 0.42 \.652 0.0\4 0.0 341
:Ji- 534 \.11 \3.9 2.3 2\ 93 0 0.32 \.47\ 0.00\ \.0 288
or-.
-------
APPENDIX C
rATA TABULATIONS:
WARM-AIR FURNACE EXPERIMENTS
Cycle-averaged flue gas composition data and thermal efficiency data
are tabu.lated for tests of two warm air furnaces: a Williamson Model
1167-15 and a Carrier Model 58HV-156.
Each furnace was tested in the
laboratcry with its stock burner, then with the stock head replaced by
one or !Tore prototype optimum burner heads. For the Williamson furnace,
fired at a nominal rate of 1.0 gph (1.05 ml/s),emissions data are
given in Table C-I, and efficiency data are listed in Tables C-2 and
C-3. Similarly, Table C-4 lists emissions data, and Tables C-S and
C-6 show efficiency data for the Carrier furnace. The stock Carrier
furnace was tested at a nominal 1.1 gph (1.16 ml/s) firing rate; with
the optimum he~d, the firing rate was 1.0 gph (1.05 ml/s). In the
tables of effi~iency data, values are given for each of four cycles
at each test c,)ndition; these are followed by effi.ciency averages for
the test.
Emissions data tables are shorter because only the averages
have beell given.
96
-------
Tab Ie C-l. CYCLE-AVERAGED FLUE GAS COMPOSITION DATA FROM TESTS OF THE WILLIAMSON MODE L 1167-15 FURNACE
PROTOTYPE OPTIMUM HEAD
(21-gage. type 430 stainless s tee I. I nit ia I design
wi th 0.0020 m reinforcement plate)
~'::; 7':'dIC. C...,~ ~~ co! tl:~ .'llt.: L" :.jJ "'It l.;,,"';.'. 7F:;
,e... !'="TI...' .. ~ rr"'.1 P!\1 ,,~. :';."I/. llHC PIIC", TF(; (18-gage. 310 stainless-steel. design)
>:3. J;ATIC : : pp,: Po'" PPP< '''''KG'' G*":/o(G~ ~~n(C;~ S..-Co([ C type revised
5J 1.41 11.0 6.5 " 84 0.34 1.706 0.010 0.0 243 101 1.37 11.4 6.0 20 74 2 O. J8 1.450 0.026 0.0 2.6
5. 1.32 11.11 5.. IS 91 '0 0.28 I .71 7 0.00.. 0.0 232 102 1.59 Ol.!I 8.3 48 foOl 15 ,.0. 1.579 0 oJ 82 0.0 271
55 1.26 1:'.2 4.6 15 93 0 0.25 1.6~3 0.005 0.0 221 103 1..0 10.. 7.3 20 73 . 0.57 1.561 0.044 0.0 271
56 \.28 12.2 S.O 10 109 0 0." 2.003 0.003 0.0 24\ 10. 1.15 13.3 2.9 3S 80 O.~3 1.309 0.007 3.0 232
S7 1.20 12.9 3.8 10 110 0 001 6 \ .887 O.COS 0.0 221 105 1.28 12.0 4.8 I:' 81 0.27 1.486 0.309 1.7 2.9
58 '013 13.S 2.6 20 107 0.30 1.726 0.0' 1 0.1 ~16 106 1.38 11.3 6.3. 30 76' 2 0.55 \..SI9 0.02\ 0.3 254
\0
--.J S9 \ .06 14.2 1.2 78 97 1.11 I. .S6 0.009 3.5 20. 107 \.5S 10.0 8.0 4\ 12 9 0.67 I. ~07 00107 0.0 268
60 1.11 12.0 1.9 25 106 0.37 1.666 0.008 3.0 20. 108 \.5. 1001 7.9 35 73 6 0.72 1.614 0.07\ 0.'0 271
61 1.42 11.0 6.6 10 93 C.:!1 1.900 0.008 0.0 243 109 \.39 11.2 6.. 20 83 0.37 1.654 0.012 0.0 254
62 1.49 10.3 7.3. 10 86 0.22 1.8.9 0.011 0.0 246 110 1.35 11.3 5.7 13 78 0.2S 1.501 0.007 I.S 252
63 1013 13.. 2.5 \8 106 0 0.28 \.707 0.003 2.0 210 111 I.S8 9.8 8.2 38 76 6 0.82 1.723 0.072 0.0 211
64 1.32 11.8 5.4 15 107 0 0.26 2.015 0.003 0.0 229 112 1.35 11.3 5.8 25' 77 0.05 1.500 ci.~\ 0 001 2.9
65 1.47 10.6 .7.2 12 9. 0 0.25 1.993 0.006 0.0 2.9 113 1.30 11.8 5.2 23 79 0.42 1.483 0.0\0 2.0 246
66. 1..3 10.9 6.7 12 f!~ 0 0.25 \.809 0.005 0.0 246 II. 1.42 11.0 6.7 30 80 0.59 ,.639 0.012 0.0 257
67 1.26 1201 ..6 15 102 0 0.25 1.842 0.0:13 0.0 227 115 1.41 10.. 7.. 36 74 0.74 1.595 0.014 0.0 260
68 1.20 12.8 3.8 11 106 0 0.29 1.82. 0.004 0." 218 116 1.54. \0.1 7.9 .1 12 3 0.87 1.597 0.034 0.0 268
69 1.08 .14.1 1.7 21 96 0.32 I. .78 0.006 3.0 199 111 1.62 9.6 8.6 6' 64 18 1.35 I.SII 0.?24 0.0 277
70 1.09 14.1 1.8 20 103 0.30 1.603 O.OOf. 2.5 199 135 1.33' 11.6 5.6 21 81 0.39 1.~40 0.010 1.1 260
71 1..'" II).e ".8 10 93 0 0019 1.911 0.00. 0.0 23e 136 I.SO 10.. 7.5 26 17 2 0.5. 1.1.66 0.018 0.0 268
72 1.30 12.0 5.3 14 107 0 0.25 1.990 0.003 0.0 2..d 137 1.40 11.1 6.S 22 19 0.43 ,.591 0.009 1.2 257
73 1.11 14.0 2.3 18 108 0.28 , .71. 0."1)6 "01 2111 *Runs 100A. B. and C were steady-state experiments.
-------
Table C-2.
CYCLE-AVERAGED THERMAL EFFICIENCY DATA FROM TESTS OF
MODEL 1167-15 FURNACE WITH ITS STOCK BURNER HEAD
THE WILLIAMSON
GRISS GROSS G118SS ;a8SS
RUN ST"IC £,.r. E". BURN V.A. Q 41 ....." W-A. Tun T RUN STOIC [". E'''. 8URN W.A. ....." If.a. 1(N) T
ND. .aT!'" ".1:. "I.... ....... rll'" "D "'" _8 or.~. .-- - .~~. -. ......- ,.";; I&'U. 'ULL ..". ",. ..... T '.6.""
SEC SEC !<.I 1<.1 "3/$ C C C ~ I SEe SEC Kj Kj "2/5 C C C
53 ..38 67.93 80.13 2'0 3" '50..... 6.n 0.5185 21.1 811 19 63 1.1. 13-$. 83.19 265 001 1051 I 7726 0.""" 28.. .n 15
53 ..38 '9.10 80.16 205 380 974' 613. 0.5'70 26.1 811 eo 63 1-.3 75.25 83.55 260 388 10316 7"2 0.5155 .... 197 IS
53 ..38 69.'2 80.43 263 388 10462 13.5 0.5121 ea.o 823 80 63 1.0' ".e! 84.09 e" 311 '562 7:tOl o. '800 88.. 19. IS
53 ..38 72.85 80.52 n8 311 10256 7411 0.5806 2..0 e81 .8 63 1.11 7'.4' 8.0 383 9525 7572 0.5801 88.9 U
--'. - --~-. ,.." ..,..,
5. ..28 76.25 SI..5 201 311 9581 1305 0.5779 2'.0 813 ,. 60 1.29 69.83 81.12 e65 3'2 10640 he9 0.5'9" 88.3 e08 18
5. 1.29 10.29 81... 205 356 9143 6808 0.5796 28.2 812 19 6. 1.29 71.0. 81.58 260 '19 10442 '.18 0.5611 26.6 elo 18
5. "28 72.65 81.45 265 383 10535 7653 0.5915 28.1 eu II 6' t .27 13.74 801 385 9679 7131 0.5816 21.1 18
5. t.21 14.22 258 3'2 10253 1610 0.5925 21." 8 6' 1.28 14.29 81.62 800 000 9138 723. O. '65' .,.. elo 18
13.35 8'.4. '2.6 11.6.
55 1.23 '..01 82.06 23' 311 9302 6885 0.5133 21.5 808 .. os 1.02 72.82 ".90 865 001 10152 1829 0.5801 28.11 230 18
55 ..23 71.34 82.03 200 386 9338 6662 0.5631 26.1 208 11 65 1..5 69.10 19.15 260 38. 1054' 1281 0.5688 88.0 230 18
55 1.23 ".39 2.. 316 9.'4 7347 O.!t869 28.3 11 65 "43 75.02 19.78 2'0 3'2 9131 7300 0.5808 27.3 2:1' 11
55 1.23 15. J3 82.03 26. 386 10272 1138 0.5853. .'.1 e08 IT 65 1.41 14.46 80.01 200 31' 9636 7175 0.581' 27.' 1!I!1 II
14.52 82.0. 12.85 19.88
56 t.22 83.15 81.26 23' 3.8 928i 1117 0.5996 21.5 223 ,. 66 t.38 11.62 80.03 262 "6 10.02 801. 0.590. 89.0 832 16
"6 1.23 79.. 41 81.06 200 .391 9320 1001 0..5821 21.1 886 .. 66 1.40 19.13 2.0 001 9532 7543 0.5915 86.8 11
56 ..23 81.06 81.48 2" 003 9415 1680 0.5814 21.6 811 .. 66 t. 39 11.78 80.28 202 3'2 '61. 10'711 0.5906 81.5 ees 11
56 1.29 16.95 80.69 260 02' 10252 1888 0.5180 27.4 826 .. .6 1.40 1S.07 80.10 2.6 00' 917) 1630 0.SS6:! 21.. 288 11
80.14 81.12 18.15 80.14
51 1.18 71.72 82..68 2" 315 '463 1355 0.5.25 28.8 805 IS 61 1.25 18.48 81.9. 261 '28 1026' 8059 0.5823 81.5 1101 18
51 1.15 82. J.. 83.11 2" '51 '410 1191 0..001 24.2 80' 16 61 1.23 83.81 200 .., '43' 1911 0.5896 2'.1 18
51 1.16 16.43 83.01 2.5 .25 9621 1358 0.5899 25.0 801 16 61 1.24 81.61 82.09 2" "0 "88 7'.3 0.6001 e4.' 206 2.
\0 51 ..16 16.51 83.01 263 '00 10331 190' 0.5810 26.3 eOI 11 61 1.25 18.98 82.08 2'6 '60 9691 7654 0.5870 24.1 805 8.
18.25 82.96 80.12 82.04
00
58 1.01 19.20 83.8" 258 "8 10256 8123 0.5195 26.6 19. 18 68 1.16 80.63 83.22 2" 3'5 '093 1332 0.6106 25.9 198 2.
58 1.10 81.6.. 83.91 200 .5. 95..1 7189 0.585. 24.9 193 ,. 68 1.17 88.16 83.09 202 028 9124 804. 0.6013 26.3 199 18
58 1.10 16.'2 8<4.32 201 036 9581 1369 0.51'19 25.2 .n .8 68 1.18 81.67 83.08 2.5 031 9243 15" 0.5850 25.5 198 2'
'8 1.16 16.38 83.46 2.. "3 '700 1008 0.5168 24.1 193 .8 68 "11 81'.55 83,'2 260 .6. 9951 8117 0.5818 21.2 198 19
18.S.. 83.89 84.50 83..3
59 1.05 80.28 8<4.19 258 ... 10151 8'" 0.51'85 25.6 182 18 .. 1.04 13.4' 85.19 2'2 001 9533 7003 0.5806 25.6 11. 81
5. ..02 80.31 8<4.11 200 .52 ..08 1588 0.5760 2<4.8 .8T 80 .. 1.04 15.55 85.01 203 ..6 9513 '232 O. '108 24.6 111 81
5. 1.03 8"42 8..95 U. '02 9.88 1125 0.51"4 25.1 18. 20 6. . .04 80.31' 84.95 206 '06 9615 7176 0.$910 21.6 "' 18
59 1.04 19.28 205 '63 96<45 1641 0.5188 24.3 eo .. 1.04 15.12 8..80 265 '0' 10<41' 7821 0.H62 21.9 '83 11
80.32 a4.a4 16.12 n.oo
60 1.25 68.16 83.40 25' 316 10303 1022 0.5845 21.2 180 80 70 1.03 79.07 84.'0 20' 310 9601 1592 0.8172 es.5 182 16
60 1.25 11.53 83.33 200 3" 954. 1391 0.5836 21.4 In 18 10 1.03 11.13 84.91 266 335 10350 80<45 0.1953 25.1' 18. 11
60 1.26 18.42 83.21 200 391 9535 1411 0.581. 28.0 186 IT 70 1.04 82.01 8...92 260 300 10116 .8302 0.8.80 25.6 III 11
60 ..27 18.88 a3.11 205 3'2 9133 1611 0.5808 ea.1 186 11 10 "0. 84.59 8..92 2.0 335 9338 '7119. 0.802' 85.0 III 11
15.15 83.28 80.87 8..93
61 ,.00 66.81 80-16 26' 35. .0181 7201 0.5"2 28.6 8e6 13 1, 1.40 85.8<4 80.60 2.' 330 9611 8255 0.8865 es.5 218 18
.. 1.38 65.'4 80..13 260 303 10621 7003 0.5859 2'.7 230 13 11 ..00 80.39 265 356 10324 8299 0.8008 2..6 10
61 1.38 12.15 10.26 200 303 9804 1013 (). 60 I 1 8'.2 821 II 11 1.<40 80.69 80.70 260 3'6 10133 81'J5 o. "'" es.3 8., 20
61 1.:39 10.61 80.10 2'! 32' 9848 6954 0.5943 30.1 22' II 1, ..00 84.54 80..0 201 301 93.5 1903 0.831!O 23.4 213 II
68.88 80.16 82.86 80.13
62 1.45 70.34 19.51 260 311 10680 '1513 0.5806 e..e U5 II 12 1.2. 14.84 600 752 2391" "898 0.1692 26.3 13
62 1.46 10.31 19.63 260 3'2 10522 1398 0..158 .,., 231 I. 12 1.21 12.21 80.81 600 101 23899 11256 0.11'83 26.1 223 22
62 I..' 69.15 19.48 2., 306 9150 6800 0.5T23 2..2 e32 13 12 1.25 83.02: 80.92 600 806 21'''2 18215 0.1700 25.0 22' 23
62 1.45 68.81 tol 315 9153 6111 0.5131 26.6 ,. 12 1.25 83.54 81.14 600 881 22'02 '9166 0.1634 ...1 litO I.
6'.81 19.54 18.00 80..8
13 1.05 81.3B '<4.35 600 '05 22951 18682 0.1.21 23.0 192 25
13 1.06 1".66 8..28 600 ... 22' 51 182B8 0.1625 2..6 192 25
13 1.06 1'8.10 84.13 600 151 23.06 181180 0.1586 21.' ..6 t2
13 1.06 16.88 84.13 600 131 23658 18183 0.19.. 26.. 196 t2
'79.00 '4.22
-------
Table C-3.
CYCLE-AVERAGED THERMAL EFFICIENCY DATA. FROM TESTS OF THE WILLIAMSON MODEL
1167-15 FURNACE RETROFIT WITH A PROTOTYPE OPTIr~ BURNER HEAD
GAUS GltUS GAISS GIIISS
RUM srllC r.,.,. r.,.,.. BURN W.A. 0 0 va." W-A. TCN) T AUN STIIC E". I". 8URN iii... 0 0 "'AN V.A. TCN) T
Me. IItATI. v... '.G. Tllllt TIME 'UE!. v..- An OEL-T '.0. A" NIl. RATIO V.A. '.G. TIME TIIIIE 'un. ".A. AlA DI1.-T '.6. Afl.
I I SEC SEC KJ KJ "3'5 C C C I I SEC SEC KJ KJ 1113'5 C C C
101 1.32 10.13 80.18 241 316 10029 7033 0.1'0, 84.0 235 II III 1.~3 1!h43 11..85 235 .61 961. 7255 0..11'78 26.9 25. 16
10' 1.33 '7.98 80.06 25' 302 '06Q~ 'JIO? Q."8..Q 25.' 236 . .n 1.~6 18.33 "..III e03 53. 995. 1800 0.50.. 1..'5 252 ..
'01 1.32 66.71 80.04 26. 3.3 11191 '.'3 0."12 ,... 1138 10 III '.56 75.90 1'1.'. 260 SO. 10329 11140 0.5081 25.8 253 IT
101 1.33 10.55 ".82 26. 355 1.092 7825 O. "52 ...2 1102 12 1\1 ,.,. 10.95 18.22 200 502 .n3 6778 0.5081 112.6 24' 21
68.16 80.03 75.15 "."
'02 ..53 68.13 ".15 2.' 33' 1018S 693' 0.'650 23.3 250 18 112 t .33 73.57 151.' 5 230 ... 9215 "80 0.4'55 24.8 2.. 20
102 1.5. 62.09 ".66 251 3\1 105'7 6580 0."" a:.... 256 12 112 1.33 11.55 80.66 2.5 ." 9652 6905 0.4933 2411.' 1122 2\
102 1.56 67.80 '7.32 210 352 11291 1655 0.7687 24.1 260 1\ 112 1.33 '5.01 80.B4 261 so. 10282 1713 0.5053 25.5 219 8\
'02 1.56 68.89 17.64 247 326 10212 1035 o. '7'6 23.5 255 13 112 '.33 19.45 80.21 260 SO. 10236 8133 0.5031 2'7.0 231 80
66.13 71.59 14.90 80.13
'03 1..3 62.71 18.17 22S 245 8991 5638 0.7681 25.4 250 10 113 1.28 13.24 80.61 233 311 '265 6186 0.4"0 3D.8 22' 19
.03 1.45 68.28 18.68 231 2'8 '245 6312 0.7723 23.3 250 \8 113 1.28 11.00 80.70 245 001 "4110 1596 0.4'96 31.8 2n 18
103 1.46 63.10 78.02 241 25. '636 6138 0.8182 25.4 260 II 113 1.28 74.56 80.10 25. ... 10296 7617 0.500' 3D.8 2n 18
103 1.41 71.09 78.45 24. 215 9756 6936 0.854111 85~ I 260 II 113 ,.a8 76.42 80.10 25. 027 10296 7868 0.5012 ]1.3 22T 18
66.45 78.48 75.56 80.67
10. 1.13 81.25 82.43 233 3" 9614 1811 o. '552 25.6 218 13 II. 1.38 81.93 79.57 258 03' 10355 8... !h51'411 31.1 8" IT
10. 1.10 73.12 82.90 23' 35. 9804 "68 0.861 3 19.7 21. IT 110 1.38 81.16 19.10 23. .33 9472 7688 0.5050 29.' 23. IT
100 1.13 69.72 82." 2.' 001 10228 7131 0.664113 112.. 216 16 II. 1.38 19.95 79.98 234 0.. 9395 7511 0.5015 28.0 233 18
10. 1.13 64.65 82.46 260 3.6 10131 6'38 o. '622 22.4 218 13 II. 1.38 81.91 79.98 2.. .50 991 5 8127 0.5186 29.1 233 18
12.19 82.5' 81.25 79.81
105 1.25 6'.41 80.51 200 31T 10386 7208 O.Tl23 8..1 236 12 115 1.45 80.10 19.08 231 .3. .585 "3D 0.5212 28.1 842 20
105 1.25 14.39 80.61 22. 3., 9694 7211 0.7512 83.1 233 '8 115 1.45 81.28 19.08 233 .36 9364 1611 0.5292 28.' 242 20
105 1.25 70.64 80.39 245 322 10599 1411T 0.7734 25.6 ..0 1\ 115 1.45 82.76 79.11 23' ..2 9528 7885 0.5331 28.5 242 20
'05 1.23 69.40 80.75 258 3.. 11161 17.6 0.1518 25.5 23. II liS 1.45 82.71 19.11 258 ..2 10266 8491 0.5132 2'.2 "2 20
10.96 80.51 81.11 ".0'
\C 106 1.33 '5." 7'.8' 265 38. 101'5 8201 0.62'3 88.5 8.., 8 116 1.50 711-01 78." 24' .S, 9921 7749 0.515. 28.. 2.. 20
\C .06 '.33 61.28 ".13 266 361 10846 12'7 0.6310 .'.8 ... '0 116 1.50 78.10 18.5' 233 422 '3'. 736' 0.5185 28.6 245 80
106 "33 65.87 7'.51 2.6 326 1002. 6603 0.62.. 21.6 8" 8 116 1.51 80.81 78.36 242 .3. 9729 7862 0.523. 29.1 24' 20
106 1.33 71.56 79.68 240 361 9877 7068 0.'28. e'.1 US . 116 1.50 10.50 18.62 265 .15 10654 7510 0..'63 28.3 US 20
70.17 79.12 71.02 78.50
10' 1.50 65.SI 71.63 266 392 10954 720' 0.6323 1!4.1 261 10 117 1.59 18.99 71.59 23. ." 9506 150' 0.5200 21.. 252 21
'01 1.49 65.96 77.84 265 391 10923 1205 0.0250 2...1 25' II 111 1.57 12.4' 77.68 233 .22 9466 6854 0.5137 26.. 252 21
101 1.51 62.10 17.10 266 368 10954 OIl" 0.6221 25.5 258 10 111 le59 "..9 11.35 243 425 '866 '842 0.536. 2'.3 256 20
'01 1..9 0.00 18.07 20' 361 10172 0 0.0000 n.5 255 10 "' 1.59 87.04 71.42 256 .SI 10593 '220 0.5.4110 ».0 us \8
'4.83 11.82 ".. 11.51
108 1.50 71.96 17.73 25. ... IQ551 7596 0.5380 28.8 25. 8 135 1.29 69.66 ".87 2.7 256 9983 6954 0.732. 31.6 20' 12
.08 1.4' 10.48 71.70 26J .19 10935 7101 0.5.'. 28.6 261 . .35 1.2' 74.92 79.85 241 28. 9983 147' 0.'89 31.0 2.' 12
108 1.50 73.89 11.63 2.. .33 10812 8033 0.5528 28.6 261 10 .35 "31 74.39 ".76 227 24. 9180 682. 0.7100 3..3 206 13
.08 ..51 73.40 17.5' P.44 00' 10055 1380 0.5538 21.8 260 II 135 1.31 67.16 ".16 225 238 9100 6111 0.120' 30.3 206 .3
72.43 17.66 7t .53 19.81
.0. 1035 81.65 SO.14 u. .5\ '948 8122 0.5911 25.' 232 .. 136 1..5 13.41 'S.31 23. 2.. 94'0 6951 0.8113 29.' 256 I.
.0. "35 15.31 79.92 261 .6. 10.72 1881 0.51.0 25.2 231 IT '36 1.43 68.08 78.11 201 30. '996 6805 0.6691 28.. 251 ,.
10' 1.33 7..22 80.03 261 .15 10686 "31 0.587' 2..2 23' IT .36 1.45 69.31 18.4' 208 3'0 10039 696. 0.68.' 27.' 253 IS
10. 1.34 '3.60 80.'2 20' .52 9870 1265 0.5109 20.0 833 IT '36 ...5 6'.31 1S.46 221 212 9186 63'7 0.68'5 28.8 253 \.
16.20 80.05 70.0. '8..'
110 1.31 11.99 80.24 260 391 1064 1661 0.521' 31.5 235 .. 131 1.35 08.20 79.68 208 289 10036 68.. 0."2' 29.' 242 I.
110 1.33 13.91 80.01 261 00' 10683 189S 0.5231 31.5 23' .. 13' ..35 13.55 79.68 208 311 10036 '38' 0.6868 29.. 242 ..
110 1.33 '0.89 80.08 241 35. '759 6918 0.5\51 31.8 236 15 13' 1.35 72.03 1'9.68 22' 286 9261 6615 0.6888 28.8 242 I.
110 1.33 12.32 SO.34 23. 385 9.'9 6855 0.515. .9.4 230 16 131 1.35 70.90 79.68 225 21. tlOS '.55 0.69.' 28.' 242 ..
.2.28 80.16 'U .17 79.6S
-------
Table C-4. CYCLE-AVERAGED FLUE GAS COMPOSITION DATA FROM TESTS
OF THE CARRIER MODE L 58HV...156 FURNACE
i (21 PROTOTYPE OPTIMUM HEAD
STOCK BURNER HEAD '
gage. ~ype 310 stain1ess-stee1. revised design)
RU'I STOIC. CC2 02 CO :1) 10.5 1.6 23 96 0,"8 2.067 0.013 0..0 327 125 1.33 1103 6.3 23 72 2 O."~. I. ~27 0.019 1.9 299
o
0 96 1.57 10.1 8.~ 23 91 2 0.51 2.070 0.027 0.0 329 126 1.27 12.4 ~.9 :>a 76 0.49 1.392 0.01~ 2.8 221
97 1.10 9.2 9.~ 35 16 ~ 0.80 1.886 0.055 0.0 332 127 1,'0 11.3 6.5 21 79 O. ~I 1.591 0.016 1.3 302
98 1.18 E.B 9.9 "0 72 8 0.96 1.1166 1).105 0.0 303 128 1.54 10.2 8.0 35 17 6 0.72 1.719 0.071 0.2 307
129 1.07 .10.6 7.2 25 81 ~ 0.51 I. 70~ 0.045 0.3 307
1(1) I "'S 12.2 7.9 35 16 1 0.63 1.532 0.071 O.~ 318
131 I ..~ 10.8 6.9 21 73 2 o. 4~ 1.515 0.018 0'. 310
132 1."9 10.. 1.~ 21 81 5 0.4., 1.739 0.057 0.0 301
133 1.38 11.1 6.2 25 77 ~ 0.46 1.527 0.002 1.1 302
13~ I."''' 10.8 6.9 20 78 2 0.40 1.611 0.027 1..
-------
Table C-5.
CYCLE-AVERAGED THERMAL EFFICIENCY DATA FROM TESTS OF THE
MODEL 58 HV-l56 FURNACE WITH ITS STOCK BURNER HEAD
CARRIER
GR,US GROS! GROSS GRDSS
RUN STDIC [y,. E"". BURN W.". . WAIIM w.... Tun T QUN STalC [Fr. En. BUliN W.oII. . . IOU" V.A. TC N) T
NO. RATIO W.A. ".G. TIME TIME .un. V.A. AIR DEL-' '.G. AM8 NO. RATIO W.A. , .G. TIM[ TIME 'UEL till-A. AI. DEL-r '.G. AMS
. . SEC SEC KJ KJ "3/5 C C C . . SEC SEC KJ KJ "3'$ C C C
88 I.S'? 64.08 '..32 245 281 994" 63'4 0.5542 34.8 312 20 95 1..5 72..9 7$.8 i 206 311 9391 ...11 0.5774 32.2 305 2\
88 1.51 6'7..' ,..4' 258 332 104'5 7070 0.5,43'7 33.3 30' 80 95 1.43 71.58 1~.9~ 2.7 313 9838 7002 0.5516 3..3 304 22
88 1.57 '4.60 7...43 238 296 9659 6240 0.5421 33.0 310 " 95 t .42 70.18 16.07 2" 358 10316 7301 0.. 5356 32.. 30' 22
88 1.53 65..9 15.00 235 280 9538 624' 0.5432 3..4 30T " " 1.42 17.18 76.24 2'8 379 9856 '606 0.5401 31.6 30\ 22
~'S.~ "..,.5 1J.01 76-02
89 t .56 63.02 1..13 238 266 9'.' 6270 0.5722 35.0 318 '8 96 I.. 47 10.15 75.'4 236 371 9.18 6663 0.5.... ...2 303 25
89 t .56 66- 36 ,...27 235 28' 9824 6519 O.53~1 3'.9 3.. IS '6 1..9 71.01 75.10 245 ... 9783 6'.6 0.52'5 27.0 300 26
8' 1.56 65.07 "4.16 245 302 102.5 6666 0.5388 341.8 318 19 .. 1.53 71.15 75... 259 '\8 10339 7356 0.5370 27.9 297 26
8. 1.57 65.35 73.71 259 292 10618 6939 0.5381 37.5 323 " 96 1.50 74.:!8 15.43 259 '36 10339 7690 0.559' 26.9 303 26
'..95 74.07 11.82 15.55
'0 1.30 12.96 18.41 234 308 '407 "'63 0.5.68 3..6 28' 20 97 1.59 67.60 74.35 232 281 9166 6195 0.5618 33.3 302 26
90 1.31 13.21 18.20 243 317 9172 7160 0.'521 341.8 28J 81 97 1.64 66.22 74.11 2'0 300 9..85 6281 0.5..' 32.8 305 26
'0 1.33 11.84 17. ,. 257 3.6 10339 7027 0.5411 33.8 285 2\ .7 1.604 69.8~ 14.08 25. 3\ 3 10035 7006 0.5428 35.1 306 26
90 1.33 71.45 77.91 25. 338 10635 75" 0.5.63 35.0 285 2. 97 1.69 13.19 73.69 256 319 9900 1310 0.5319 36.3 305 26
12.)8 78.12 69.36 74.06
91 1-16 75.." 80.24 25' 359 10.30 7868 0.5468 3..1 260 22 98 1 .12 '!Io6.0J 72.75 233 271 9394 6203 0.5538 3!h2 31> 25
91 1.18 15.1. 80.13 232 3u 9526 7159 0.5.., 32.5 260 22 '8 1.72 65.59 12.53 240 271 9613 63.. 0.5531 36.0 318 25
91 1.23 11.07 19.88 231 332 '.85 6141 0.5396 32.0 257 22 '8 1.72 63.26 72.51 255 283 1027. 6.99 0.546. 35.8 318 2'
91 ! .20 73.17 79.95 240 338 9855 7211 0.55.. 32.7 260 22 '8 1.74 67.10 12.32 253 283 10298 6909 0.5789 35.9 318 2'
t-' 73.71 80.05 6!h.9 72.5.
0 92 1.91 61.48 70.36 257 28' 107'4 6611 0.5512 3'.4 326 22
t-' 92 1.88 60.26 70.58 237 280 9929 5983 0.5511 32.6 326 22
.2 1.88 63.02 70.5. 232 277 9717 6123 0.'6.' 33.3 326 22
92 1.88 58.82 70.54 242 ::e. 10235 6020 0.550' 32.7 32~ 22
60.89 10.51
93 1.62 58.09 13.21 .33 230 9823 5106 0.5.94 38.) 323 16
'3 1.62 63.23 13.11 .40 241 10124 64;01 0.59'6 31.9 325 17
'3 1.63 65.2~ 73.11 256 272 10806 105. 0.5810 31.9 324 18
.3 1.62 68.04 73.21 257 287 10852 138. 0.51., 38.1 323 19
63.66 13'-1 Eo
.. 1.36 69.61 71.01 236 301 '6S4 6741 O.5~0' 34.6 297 2\
9. I. 3S 10..tO 71.22 2.. 316 10009 1046 O.'.IS 3S.1 295 20
.. 1.3S 11.80 71.10 258 3IT 10581 1601 0.5523 36.9 297 21
.. 1.35 16.17 77.01 258 30T 10.'8 198: 0.5659 3..6 2" 20
12.01 71.10
-------
Table C-6. CYCLE-AVERAGED THERMAL EFFICIENCY DATA FROM TESTS OF THE CARRIER
MODEL 58 HV-l56 FURNACE RETROFIT WITH A PROTOTYPE OPTIMUM BURNER HEAD
GRUS 611155
RUN STIIC E'''. Err. BURN w.". Q WAR" w.a. TCN) T GIIISS GRISS
NB. RATle w.". '.G. TI ME TlMt run. W.A. "IR on.T r.G. atIB RUN STeHC En. £,.,. 8URN ".A. " .. V"... W.A. TeN' T
. . SEt SEC .~ IW "315 t C C ND. RATiG "'.A. r.G. TIME TIME ru£l. W.A. "IR DEL-T r.G. "...
2 2 SEC SEe K~ K~ 1113.15 t C t
ou . .~. ...-.... ....-. --- --- ....- .-- -- - . -. ---- ... ..
..8 . .50 72.54 75.81 230 313 8949 64'2 0.5.:!8 32.5 2U " In '.20 11.3,3 19.26 230 256 9114 6501 0.6IH 35.e 217 13
... I-50 12.94 75.76 238 317 925. 6750 0.5.:!5 33.4 ." 16 \26 1.21 76.26 19.22 226 260 8955 &829 0.62" 35.4 217 13
... 1.50 '76.29 15.91 253 353 9S4, 7501 0.$34' 33.1 2.2 " 126 1.21 66.81 ".35 230 281 9111 6091 0.580' 31-1 213 13
14."" 75.84 126 1.21 19.31
71.41 19.30
"9 1.29 78.36 18.95 231 40' 8118 6831 0.$626 25.4 1170 eo
,,9 1.30 15..19 78.95 229 40\ 8"" 6502 0.5683 e~.. 3 268 21 \21 1.33 69.4. 11.88 225 283 8651 6005 0.571'3 31.3 286 \ 5
"9 1.29 16.33 ,.9.03 239 .,. 9026 ".9 0.5633 24.8 U8 121 121 1.33 12.20 1'7.82 236 2" 9068 6547 o. 60.' 32.3 28' 10
"9 1.29 16.49 78.98 252 006 9111 '7.33 0.S1U! 2'.3 270 eo \27 1.34 11.'0 ''7.79 2.7 302 9901 Tlt8 0.617. 32.4 286 IS
76.'9 78.98 121 1.3!» 1'2.69 1"7.91 207 317 9908 7202 0.6060 31.' 282 16
11.55 ''7.85
120 1.59 68.21 75.30 232 3.S 90"7 6111 0.514.. 23.6 2.' 2\
120 "58 65.0'7 '75.26 221 33S 88"" 51$4 0.5891 24.6 1192 eo 12. 1.4' 68.29 76.11 225 28\ .''" 5ue 0.6040 29.9 293 16
120 1.51 72.22 '75.62 236 33. .,88 6636 0.5852 28.. n6 18 128 1.49 13.48 '6.12 234 2S7 9096 6683 0.6071 32.6 129\ 16
120 1.53 '2.2.. 15."9 249 36S 9899 1151 0.58411 2..3 296 18 128 1.50 10.12 16.02 2.6 313 9565 610" 0.$911 30.8 290 16
69."4 '7'.42 \2. 1.49 13.41' 15.99 2.6 290 9562 '025 0.'25" 32.' 293 16
71.3" 16.06
121 1-35 '9.'5 ".88 241 292 9353 7418 0.6051 36.0 1282 13 129
12\ 1.34 16.63 17.16 252 31. 9789 1501 0.5''73 34.0 286 IS . .41 15.32 76.86 229 268 8995 6115 0.62'1 34.2 291 16
\2\ 1.34 15."" 11.19 253 329 9831 1...,. 0.5828 32.' 286 IS 129 1 .42 68.28 16.11 22. 2S6 8196 6005 0.5'80 30.9 291 \5
\2\ 1.34 14.13 ".'9 233 31. 9054 6'711 0.5187 3(.4 286 IS 129 1.02 68.69 16.7. 234 210 9189 63te 0.6162 31.8 291 IS
'6.5.. "7.80 129 1.40 '" 63 71.10 2.6 301 9660 6919 0.6070 32.2 1288 IS
I-' 10.98 16.8S
0 122 I. 56 69.29 15.00 23. 296 9254 ...12 0.5845 31.5 301 16 130
N \22 I. 53 10.80 15.18 25. 316 9883 699' 0.5184 32.6 303 17 ...., 68.33 15.61 250 251 9192 6691 0.626' 36.1 30. II
122 1.54 '0. l' 1!u13 253 3\9 .832 6961 0.5155 32.3 301 16 130 I..' 65.5t 15.48 250 263 9192 6423 0.585. 35.4 30' II
122 "51 12.1' 14.84 231 1292 8885 6412 0.5184 32.3 302 U 130 1.41 64.18 1'5.53 2JO 2.\ 901. 5839 0.5995 3... 306 12
10.16 15.0. 130 1..7 63.70 15.56 228 242 884' 563S 0.5821 3".0 305 12
65.60 15.55
123 1.47 11.66 15.93 239 311 9103 65e. 0.5126 3'" an 17 131
123 1."'1 13.14 15.'5 252 329 9602 1023 0.5..\ 3\.3 U7 U 1.38 67.23 n..ts 2.9 2., 9711 656' "'''.5889 33.7 21' ..
123 '.47 13.5' 15.93 253 341 9631 7090 0.510. 3"0 an .. 131 1.40' 61.56 16.82 209 266 9762 6595 0.5992 35.1 294 12
123 1..4'1 10.95 16.1. 233 311 8692 6161 0.5761 119.3 In .. 131 1-.40 65.21 '6.70 229 248 8981 5.56 0.5840- 3".3 296 13
12.33 15.99 131 \.38 6'7.51 16.'" 225 241 8132 5895 0.5936 3".2 296 13
66..' 16."
12. \.02 14.8. 11.25 248 362 9360 1005 0.$7" 28.. 281 20 \32
\24 1.4' '1.89 ".12 253 365 9536 6855 0.5763 17.' 286 18 1.42 6h90 76.11 234 283 919. 6151 0.6106 30.3 290 16
\2. 1.42 16.14 11.13 251 373 94'0 1211 0.5'60 28.6 28' 20 132 1. ~5 ".1 I 16.S2 2.6 301 9666 ..81 0.5781 31.1 290 16
120 1.02 18.28 11.02 23\ 356 SM8 1036 0.'732 29.3 .86 .. 132 1..46 11.00 16.3'7 206 292 9663 6..0 0.6255 32.0 291 16
15.19 71.13 132 1.45 6'.70 16.49 228 28\ 9142 6006 0.5931 30.6 291 16
".61 16.$..
125 ..33 13.'4 78...0 239 371 9213 6U2 1).57'" 27.0 n. 110 133
125 1.3S 16.12 18.22 252 36. 911" 1394 0.'885 2..0 n. 110 1.35 66.69 1'.61 223 2.7 8762 58.3 0.5804 29.8 2C> 16
125 ..33 11.04 18.21 250 331 9631 6.. 0.55" 31.5 180 .. 133 1.35 63.'" 1'7.16 235 2.. 923. 5860 0.51"" 30.5 285 16
125 1.33 61.80 11.88 230 265 '"!6 6181 0.58" 34.1 28' \5 133 1.35 66.99 '7.'6 2.6 305 9666 6..15 0.5'73 30.2 285 16
11.22 18.18 133 . .:15 69.14 71.73 .2.6 311 9864 6819 0.5841 32.1 285 16
66.72 1'1.11
13. 1...0 '6.11 1'7.1:1 213 251 8367 SS81 0.625. 30.2 en 16
134 1.40 68.3" 11.10 225 263 iU5 6031 0.634' 30.' n8 IS
13. \.40 6S.46 71.25 23. 269 9189 6015 0.5895 32.2 286 15
134 1."0 16.52 ".22 2... 270 9657 1389 0.6320 36.3. 286 IS
69.26 11.18
-------
TECHNICAL REPORT DATA
(Please read blUTIINiolls on the rel'erse before completing)
1. REPORT NO. 12. 3. RECIPIENT'S ACCESSION-NO.
EPA-600/2-76-256
4. TITLE AND SUBTITLE 5. REPORT DATE
COMMERCJrAL FEASXB][L][TY OF AN OPTIMUM September 1976
RESIDENT][AL OXL BURNER HEAD 6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S) 8. PERFORMING ORGANIZATION REPORT NO.
L. P. Combs and A. S. Okuda R-76-103
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT NO.
Rocketdyne Division lAB014; ROAP 21BCC-058
Rockwell International Corporation 11. CONTRACT/GRANT NO.,
. 6633 Canoga Avenue 68-02-1888
,.. O'~ P~V'k (" ~. A . 01 <10A
12. SPONSORING AGENCY NAME AND ADDRESS 13. TYPE OF REPORT AND PERIOD COVERED
EPA, Office of Research and Development Final: 6/75-2/76
JIndustrial Environmental Research Laboratory 14. SPONSORING AGENCY CODE
Research Triangle Park, NC 27711 EPA-ORD
15. SUPPLEMENTARY NOTES IERL-RTP project officer for this report is G. B. Martin, Mail
Drop 65, 919/549-8411, Ext 2235. . . . .
16. ABSTRACT The report gives results of a study of the feasibility of commercializing
optimum oil burner head technology developed earlier for EPA. The study included:
selecting the best commercial method for fabricating optimum heads; determining
that prototype simulated-production heads could reproduce an earlier research head's
beneficial results; and testing prototype heads as retrofit devices in two commercial
residential furnaces. Sheetmetal stamping was selected as the best fabrication method.
A one-piece stamped and folded design was evolved and prototype commercial heads
were fabricated. Research combustion chamber tests showed these to be equivalent.
to the earlier research head. Tested as retrofit replacements for stock burner heads
in two new warm-air oil furnaces, the prototype heads were found to be operationally
satisfactory and potentially durable and long-lived. It was estimated that widespread
retrofitting of old residential units could increase mean season-averaged thermal
efficiency (averaged over those units retrofitted) by about 5% and simultaneously
reduce NOx emissions from these sources by about 20%. Logistics of a retrofit
program, training for service personnei, and requirements to ensure meeting c.odes
and standards were not resolved.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b.IDENTIFIERS/OPEN ENDED TERMS C. COSA TI Field/Group
Air Pollution Air Pollution Control 13B
Combustion Stationary Sources 2lB
Oil Burners Res idential Burners 13A
Nitrogen Oxides Burner Heads 07B
Thermal Efficiency Commercialization 20M
18. DISTRIBUTION STATEMENT 19. SECURITY CLASS (Tllis Report) 21. NO. OF PAGES
Unclassified l1o.
Unlimited 20. SECURITY CLASS (This page) 22. PRICE
Unclassified
EPA Form 2220-1 (9:73)
103/104
------- |