EPA
U.S. Environmental Protection Agency Industrial Environmental Research
Office of Research and Development Laboratory
Research Triangle Park, North Carolina 27711
EPA-600/7-77-108
September 1977
STUDY OF A THERMAL
AEROSOL OIL BURNER
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into seven series. These seven broad categories
were established to facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously planned to foster
technology transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the effort
funded under the 17-agency Federal Energy/Environment Research and Development
Program. These studies relate to EPA's mission to protect the public health and welfare
from adverse effects of pollutants associated with energy systems. The goal of the
Program is to assure the rapid development of domestic energy supplies in an environ-
mentally-compatible manner by providing the necessary environmental data and
control technology. Investigations include analyses of the transport of energy-related
pollutants and their health and ecological effects; assessments of, and development
of, control technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect the
views and policies of the Government, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information
Service, Springfield, Virginia 22161.
-------
EPA-600/7-77-108
September 1977
STUDY OF A THERMAL
AEROSOL OIL BURNER
by
J.E. Janssen, J.J. Glatzel,
E.R. Wabasha, and U. Bonne
Honeywell, Inc.
10701 Lyndale Avenue, South
Bloomington, Minnesota 55420
Contract No. 68-02-2194
Program Element No. EHE624
EPA Project Officer: Robert E. Hall
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, N.C. 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
-------
FOREWORD
The present need to use fossil fuel more efficiently and at the same
time reduce pollution from combustion processes is placing new demands
on residential heating systems. Residential heating systems are one of
the major consumers of refined heating oil and gas. They also contribute
to atmospheric pollution. The Industiral Environmental Research Labora-
tory of the U. S. Environmental Protection Agency is concerned with both
• Reducing pollution, and
• Improving efficiency of combustion systems.
This report presents the results of a research project undertaken to
explore the idea of using a thermal aerosol generator as an oil burner.
It was postulated that the improved atomization would produce clean effi-
cient combustion at relatively low firing rates.
111
-------
PREFACE
Energy conservation has become a major national concern. At the
same time we are committed to improvement of the environment. Ways
must be found to reduce waste and pollution in the combustion of fossil
fuel for home heating.
It is well known that residential oil burners frequently do not perform
at the best efficiency with the lowest possible emission of noxious products.
Poor fuel atomization contributes to this problem. The thermal aerosol
generator has been used since before World War n to produce smoke from
fuel oil. In this device, oil is first vaporized. The vapor then condenses
in droplets of around 0.1 micrometer diameter. Droplets of this size pro-
duce a dense white fog.
W. L. Tenney (consultant to this project) proposed using a similar
principle to atomize fuel oil in an oil burner and obtained a U. S. Patent
(4, 013, 396). His idea was to heat the oil under some pressure and then
allow it to be atomized through a nozzle. Some of the oil would flash to
vapor and in the process would help break up the liquid phase. The pre-
sence of vapor phase in the nozzle also would reduce the flow rate.
Reduced firing rate is desirable since most residential oil burners are
grossly oversized and this leads to reduced seasonal system efficiency.
This project was undertaken to explore the feasibility of this idea.
Since the project was quite modest, it was not possible to study each of the
effects in depth. Rather, a survey of conditions was made to assess the
promise of this principle and to define areas needing further study.
IV.
-------
CONTENTS
FOREWORD iii
PREFACE iv
FIGURES vi
TABLES vii
ACKNOWLEDGEMENT viii
1. Introduction 1
2. Conclusions 3
3. Recommendations 4
4. Experimental Procedures 5
Apparatus 5
Burner Design 5
Fuel Handling 8
System 8
Instrumentation 8
Test Procedure 10
5. Results and Discussion 13
Aerosol Size Distribution 13
Firing Rate 16
Efficiency 16
Emissions 23
Transient Performance 29
Effect of Swirl 30
-------
FIGURES
Number Page
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Experimental Burner
Fuel Handling System
Instrumentation
Spray Size Distribution
Spray Mass Size Distribution
Effect of Fuel Temperature and Pressure on Flow; ....
0. 89 cm3/s Nozzle, No. 1 Oil
Effect of Fuel Temperature and Pressure on Flow; ....
0. 89 cm3/s Nozzle, No. 2 Oil
Effect of Fuel Temperature and Pressure on Flow; ....
1. 05 cm3/s Nozzle, No. 1 Oil
Effect of Fuel Temperature and Pressure on Flow; ....
1. 05 cm3/s Nozzle, No. 2 Oil
Effect of Fuel Temperature and Pressure on Flow; ....
1. 31 cm3/s Nozzle, No. 1 Oil
Effect of Fuel Temperature and Pressure on Flow; ....
1. 31 cm3/s Nozzle, No. 2 Oil
Correlation of Flue Parameters
Gross Steady State Efficiency
Effect of Operating Parameters on NO ;
0. 89 cm3/s Nozzle, No. 2 Oil
Effect of Operating Parameters on NO ;
1. 25 cm3/s Nozzle, No. 2 Oil x
Flame with Swirl
6
7
9
14
15
17
18
19
20
21
22
24
25
27
28
32
33
VI
-------
TABLES
Number Page
1 Fuel Properties , 10
2 Firing Conditions for No. 1 Fuel Oil H
3 Firing Conditions for No. 2 Fuel Oil 12
4 Contribution of Fuel Nitrogen to NO 26
A
5 Effect of Fuel Temperature on Transient Performance . . 29
6 Effect of Fuel Temperature on Smoke 31
vu
-------
ACKNOWLEDGMENTS
We are grateful to W. L. Tenney for his cooperation and stimulating
discussions in the application of thermal aerosol technology. As a result
of initial discussions, R. H. Torborg, and A. E. Johsnon, under the direc-
tion of U. Bonne, carried out an experiment which demonstrated that the
thermal aerosol principle was attractive for low capacity residential oil
burners. This experiment led to the proposal for this project. We are
grateful to EPA for supporting the study.
via
-------
SECTION 1
INTRODUCTION
Recent studies (!_, 2, 3) have shown that excess capacity in resi-
dential heating systems~leads to excessive stack losses and reduced
seasonal system efficiency. A study of 26 oil burners in the Boston
area (4) revealed an average excess capacity of 147% (i. e. 247% of
design load). The capacity of these burners was then reduced an
average of 27. l%by installing smaller nozzles. It was found, however,
that excess air had to be increased an average of 36. 9% to prevent
smoke. The increased losses associated with increased excess air
largely cancelled the improvement to be expected from reduced capacity.
Many conventional oil burners are too large to heat a residence
efficiently. Greater use of insulation compounds the oversize problem.
There appears to be a real need for a burner with a capacity of 0. 4 to
0. 6 gph (0. 42 to 0. 53 cmvs). Pressure atomizing nozzles of this size
have such small orifices and passages that they are prone to becoming
plugged after unacceptably short operating times.
It is well known that improved atomization, i. e. smaller drops,
permits better mixing of the air and fuel and reduces the tendency to
form soot. It would appear, therefore, that a burner that could achieve
good atomization without resorting to very small orifices in the nozzle
for a flow rate of about 0. 4 cm3/s should be very attractive.
A droplet of fluid is held together by surface tension. The surface
to volume ratio increases when a given volume of fluid is divided into a
larger number of smaller drops. Energy input is needed to overcome the
surface tension force when the surface area is increased. It is shown in
the Appendix that the radius of a drop is given by the following equation:
=
r
AE/V
where: o = surface tension
AE = energy input
V = fluid volume
-------
Appendix equation (6) gives the average drop size consistent with
energy consideration. Any spray, however, will have drops with a wide
range of'sizes. Thus while it may not be possible to compute actual drop
size from equation (6), it does show the benefits of heated fuel. Heating
the fuel both reduces the surface tension and increases the energy content.
W. Tenney patented a "Fuel Aerosolization Apparatus" based on a thermal
principle (5). In this device fuel oil under pressure is heated and then
allowed to expand through a nozzle. The hot oil flashes to a vapor as the
pressure drops in passing through the nozzle. Upon cooling after passing
through the nozzle, the vapor condenses into very small (less than 1 micro-
meter) drops which produce a dense white smoke. The U. S. Navy used
this device to produce smoke screens in World War II. We proposed this
principle as a means of atomizing fuel oil in a residential oil burner. This
contract was awarded as a result of that proposal.
The objective of this study was to show that a thermal aerosol generator
could be used as an oil burner with a capacity of about 0. 4 cmv s (0. 39 gph)
and to define the operating parameters needed for clean, efficient combustion.
-------
SECTION 2
CONCLUSIONS
The use of heated fuel oil to produce a thermal aerosol makes possible
the atomization of fuel oil at lower pressures and reduced firing rates.
Heating the fuel oil, both No. 1 and No. 2, to 150 C greatly increases the
number of droplets in the 0.1 micrometer range.
The formation of bubbles as fuel flows through the swirl passages in
the nozzle restricts flow. This permitted standard 0. 89 to 1. 31 cm3/s
(0. 85 to 1. 25 gph) hollow cone spray nozzles to be fired at only 0. 42 cm3/s
(0. 4 gph). Atomization achieved in this manner reduced smoke, hydro-
carbons and carbon monoxide.
When combined with air swirl (swirl parameter as defined in the
Appendix) equal to 4. 5, non-luminous flames were achieved at excess
oxygen levels of 2. 0% (10% excess air). The Bacharach smoke was zero
and there was no measurable hydrocarbon or carbon mono:xide. Luminous
flames were present without swirl but the thermal aerosol was beneficial
in reducing emissions.
The thermal aerosol combined with swirl reduced NO formation in
most cases when clean burning was achieved. This effect^ended to dis-
appear when the excess oxygen was reduced below 2% (10% excess air).
The use of heajted fuel improves combustion during burner start-up.
Fuel heated to 150 C achieved zero smoke in less than 1. 5 minutes. Fuel
at 80 C produced a No. 5 smoke at 1. 5 minutes but achieved a zero smoke
within 5 minutes.
The flow rate of a 1. 05 cm3/s (1. 0 gph) nozzle was varied from 0. 2 to
0. 6 cmvs by varying fuel temperature. Clean combustion with low excess
oxygen was achieved at all firing rates.
Thus the combination of the thermal aerosol generator with substantial
air swirl gives a non-luminous flame with low emissions at an excess oxygen
level of 2%, If the flue temperature rise is held to 300 C this results in a
combustion efficiency of 81%.
-------
SECTION 3
RECOMMENDATIONS
The results of these experiments show promise for a residential oil
burner of 0. 41 to 0. 52 cm3/s (0. 4 to 0. 5 gph) rated capacity. A prototype
burner of this size should be built. The burner should have a blower with
sufficient pressure rise to supply the vortex mixing chamber. Provision
should be made for regeneratively heating the fuel, but a small electric
powered heater should be provided for initial heating of the fuel. Excess
air should be held in the 5 to 10% range. Also, recirculation should be
investigated as a means of further reducing NO •
-------
SECTION 4
EXPERIMENTAL PROCEDURES
APPROACH
Our approach was to design a burner that would permit us to control
each of the operating variables separately. These were fuel temperature,
fuel pressure, fuel flow rate and air flow rate. Since fuel flow is a function
of fuel pressure and temperature at the higher fuel temperatures, our first
approach was to design a variable area nozzle which would permit independent
control of fuel flow rate. It became clear, after two attempts, that the design
of a satisfactory variable area nozzle would consume too much time and effort.
Instead, we substituted three standard oil burner nozzles. These were rated
at 0. 89, 1. 05 and 1. 31 cm3/s (0. 85, 1. 0 and 1. 25 gph). Using the highest
fuel pressures of interest with the small nozzle and the lower pressures with
the large nozzle, it was possible to stay within the flow range of interest.
Combustion efficiency was computed from measurement of the oxygen
content and temperature of the flue gases. The heat exchanger coupled to
the combustion chamber was not designed for efficient energy absorption but
rather to simulate the temperature quenching that normally occurs. There-
fore, efficiency calculations are based on an assumed flue gas temperature
rise of 300 C (540 F).
The flue gas analysis was made with instruments discussed under the
section on instrumentation.
Burner Design
The burner consisted of a steel pipe 178 mm (7 in. ) i. d. by 508 mm
(20 in. ) long. The first half was insulated with 12. 6 mm (0. 5 in. ) thick
moldable ceramic fiber insulation. * The remainder of the chamber including
the inlet end was insulated with 6. 35 mm (0. 25 in. ) of the same material.
These dimensions gave a volumetric heat release rate of 1,928 kj/s-m^
(186, 000 BTU/hr-ft^). This was consistent with standard practice.
WRP Ceramic fiber insulation, Refractory Products Co. , 12W. Main
St. P.O. Box 309, Carpentersville, IL 60110.
-------
FLUE
FUEL VALVE
INSULATION
HEAT EXCHANGER
AIR INPUT
TUBES
COMBUSTION CHAMBER
MIXING CHAMBER WATER COOLING COIL
VIEWING WINDOW
FIGURE 1. EXPERIMENTAL BURNER
-------
TO NOZZLE
HEATERHOUSING
HEATER CARTRIDGE
PRESSURE CONNECTION
POWER CORD
DELIVERY TUBE
STRAIN GAUAGE
WEIGH BEAM
PRESSURE VESSEL
FIGURE 2. FUEL HANDLING SYSTEM
-------
The nozzle was screwed into a fitting attached to the fuel valve. The
nozzle in turn sprayed fuel into a vortex mixing chamber attached to the
end of the combustion chamber. This v.ortex chamber was 76 mm (3 in. )
dia. by 25 mm (1 in. ) long. A 38 mm (1. 5 in. ) orifice separated the vor-
tex chamber from the combustion chamber.
Air was admitted tangentially to the vortex chamber through four 13 mm
(0. 5 in. ) steel tubes spaced 90 apart around the vortex mixing chamber.
The air was supplied through a single orifice meter to a header. The lay-
out of the combustion chamber is shown in Figure 1.
Fuel Handling System
Fuel was supplied from a weigh tank mounted in a pressurized chamber.
The weigh tank was mounted on a cantilevered beam equipped with strain
gauges. The chamber was pressurized with compressed nitrogen which
supplied the atomizing pressure. Fuel flow rate was determined by ob-
serving the change in weight of the weigh tank over a measured time interval.
The weigh tank had a capacity of 700 cm3 (0. 185 gal) which gave a running
time of 20 to 30 minutes before refueling was needed.
The fuel was piped from the weigh tank chamber to the electric powered
heater located directly above the weigh tank. This was a 1690 watt heater
mounted in a section of 38 mm (1. 5 in. ) pipe. The high-powered heater was
needed to keep the heat flux density at the heater surface to the recommended
level of 13,951 W/m2 (9 W/in. 2). The fuel valve and burner nozzle were
mounted directly on top of the fuel heater. Layout of the fuel measuring and
heating system are shown in Figure 2.
Instrumentation
The instrumentation layout is shown in Figure 3. Instruments used
were as follows:
Strain Gauge on Fuel Weigh Tank - Balwin Lima Hamilton
SR4 Strain Indicator, Type N.
Oxygen Measurement - Westinghouse, Hagan Oxygen Monitor
Nitrogen Oxides - Aero Chem Chemiluminescence Monitor
for NO, NO .
li
Hydrocarbons - Beckman Model 400 Hydrocarbon Analyzer
Carbon Monoxide, Carbon Dioxide, Methane - Honeywell
Non-dispersive Infrared Analyzer
Particle Size Measurements - Thermo Systems Inc. Model
3030 Electrical Aerosol Size Analyzer for 0. 01 to 1. 0
micrometer range and ROYCO Particle Analyzer for the
0. 5 to 10 micrometer range
-------
FURNACE
u
ICE BATH
WATER
TRAP
FILTER
•+T+
VENT
SAMPLE
PUMP
FLOW
+-r#
AERO-CHEM
NO, NOX
BEG KM AN
400
FID
HYDROCARBONS!
HONEYWELL
MULTICHANNEL
NDIR
CO,C02,GH4
WESTINGHOUSE
OXYGEN
METER
FIGURE 3. INSTRUMENTATION
-------
Fuel Temperature - Copper vs. Constantan thermocouple
and Honeywell Class 19 Recorder.
Fuel Pressure - Heise precision pressure gauge, 0-50
PSI range.
Air Flow - 0. 5 in. dia. orifice and water filled "U"
tube manometer.
Smoke Measurements - Bacharach Smoke Tester
These instruments all had meters which were read and manually re-
corded. Thus, measurements had to be made during steady state operation
or during slowly changing transients.
TEST PROCEDURE
The test procedure was to fill the fuel weigh tank with No. 1 or No. 2
fuel oil. Properties of the fuel oil are presented in Table 1.
TABLE 1. FUEL PROPERTIES
(Measured by Twin City Testing Laboratory)
Property
Value
No. 1 Oil No. 2 Oil
API Gravity, 15. 5°C (60°F)
Heating Value
(J/m3)
(BTU/gal)
Carbon (%)
Hydrogen (%)
C/H
Sulfur (%)
Nitrogen (%)
Nickel, ppm
Vanadium, ppm
Lead, ppm
42.0
538,862
134,924
86. 26
13. 58
6. 35
0.19
0. 0064
26
23
28
34.8
545,661
136, 627
87. 15
12. 68
6.87
0. 37
0.012
32
24
42
The desired pressure on the fuel would be set. The fuel valve was
opened slightly to bleed air from the system and fill the fuel heater. The
valve was then closed and power to the heater was adjusted to give the
desired fuel temperature in the heater. Since the nozzle tended to pick up
considerable heat from the flame, it was necessary to compensate for this
in adjusting the fuel heater temperature. When the desired temperature
was reached, the air supply was turned on, fuel was turned on and ignited
with a propane torch. After about 2 or 3 minutes the burner insulation was
10
-------
hot and the flame was stabilized. The air flow was then adjusted to give
the desired excess oxygen level. The various instruments were then read.
Following this the air flow or fuel pressure was adjusted to a new value
and a new set of readings was taken. Usually about four sets of readings
were taken before it was necessary to refuel.
The large number of permutations of the variables made it impractical
to test all possible combinations. Instead, operating parameters were
selected to cover a wide operating range and additional detail when results
of special interest were observed. Tables 2 and 3 show the combinations
of firing conditions studied. A brief test of the effect of swirl and a tran-
sient test were carried out in addition to the steady state tests.
TABLE 2. FIRING CONDITIONS FOR NO. 1 FUEL OIL
Nozzle
cmvs
(GPH)
0. 89
(0.85)
1. 05
(1.00)
1. 31
(1.25)
Fuel
Temp
20
20
20
150
150
20
20
20
20
20
20
150
150
Fuel
Pressure
kPa gage
207
276
345
207
276
138
207
276
103
138
207
103
138
207
Excess
Oxygen
%
2; 4; 8
2; 4; 8
<1; 2; 4; 8
2; 4; 8
2; 4; 8
2; 4; 8
2; 4; 8
4; 8
2; 8
2; 8
2; 8
2; 8
2; 8
2: 8
11
-------
TABLE 3. FIRING CONDITIONS FOR NO. 2 FUEL OIL
Nozzle
cm3/s
(GPH)
0.89
(0.85)
1.05
(1.00)
1. 31
(1.25)
Fuel
T^mp
\-s
20
20
20
90
90
90
150
150
150
20
20
20
20
20
150
150
150
Fuel
Pressure
kPa gage
207
276
345
207
276
345
207
276
345
138
207
276
103
138
103
138
207
Excess
Oxygen
%
2; 4; 6; 8
<1; 2; 4; 8
<1; 2; 4; 8
2; 4; 8
2; 4; 8
<1; 2; 4; 8
2; 4; 6; 8
2; 4; 8
<1; 2; 8
2; 4; 8
2; 4; 8
<1; 2; 4; 8
2; 4; 8
8
2; 4; 8
2; 4; 8
<1; 2; 4; 8
12
-------
SECTION 5
RESULTS AND DISCUSSION
The experimental results are presented in Figures 4 through 17 and
Tables 4 and 5.
AEROSOL SIZE DISTRIBUTION
Figure 4 shows how the aerosol drop size distribution is affected by
fuel temperature and pressure. Figure 5 shows the relative volume dis-
tribution as a function of drop size. Fuel temperature of 200 C and above
produced substantial numbers of drops in the 0. 1 micrometer range. How-
ever, the fuel temperature had to be 250 C or higher to produce a large
volume concentration.
The data in Figure 5 cannot be used to estimate air/fuel ratio because
the measurements ignore droplets bigger than about 5 micrometers. The
larger droplets tended to settle on the walls of the sampling tube and sample
chamber. While the number of larger drops was small compared with
smaller drops, the mass of fluid in the larger drops probably was substantially
greater. Figure 5 is useful for comparing the relative volume of fuel in the
0. 03 to 3. 0 micrometer range.
As expected, increased pressure and lighter weight (No. 1) fuel in-
creased the number of small drops. When converted to a volume or mass
distribution the distribution peaks at around 3 micrometers. The apparent
discontinuity at around 1. 0 micrometer was at least partially due to the
use of two different measuring instruments. A Thermo-Systems Inc. model
3030 Electrical Aerosol Size Analyzer was used in the range 0. 01 to 1. 0
micrometers. This instrument is based on a measurement of the charged
particle current carried by the collected particles after they have been
charged. A ROYCO Particle Analyzer was used to measure the concentra-
tion of drops in a 0. 5 to 10 micrometer range. This instrument is based
on an optical principle.
These instruments required a sample of the fuel spray to be fed into a
separate sampling chamber. An aerosol sample was then fed into the instru-
ment from this chamber. Only the smaller air-borne drops would stay in
suspension through this sampling system. Hence the larger drops in the fuel
spray were ignored.
13
-------
CD
LJ
_)
O
h-
a:
a:
LJ
GO
LL)
>
UJ
a:
4
10
,0°
260°C,207kPa,NO.I FUEL
0-89 cm3/S (0-85GPH) NOZZLE
200°G,207kPa,
NO. I FUEL
0.89cn?/S(0.85GPH)
NOZZLE
20°C,207kPa,NO.I
FUEL, 0.89cm3/S
(0-85GPH) NOZZLE
I50°C, 207kPa
NO.I FUEL, 089cm3/S
(085GPH)NOZZLE
_L
300°G, 138kPa,
NO. 2 FUEL VARIABLE
REA NOZZLE 0-23
cm3/S (0-22 GPH)
j_
.01 -0316 .1 .316 1.0 3.16
DROPLET DIAMETER, MICROMETERS
FIG. 4 SPRAY SIZE DISTRIBUTION
10
14
-------
10
(O
cr
UJ
h-
UJ
5
O
or
o
:t
O
UJ
2
O
>
UJ
UJ
(T
260°0,207 kPa,No. I FUEL
0.89cm7S( 0.85GPH) NOZZLE y
10
200°Q. 207 kPo NO.
0.89crr#S (0.85 GPH )NOZZL
10
,o2
10
0
10
207 kPa,
No I. FUEL
300°C,I38
No. 2 VARIABLE
IL AREA NOZZLE
(0.22 GPH)
0-89 crrT/S(0-85 GPH) NOZZLE-
I I I I
.01
.0316
.316
1.0
3.16
10
DROPLET DIAMETER, MICROMETERS
FIGURE 5. SPRAY MASS SIZE DISTRIBUTION
15
-------
FIRING RATE
Figures 6 through 11 show the effect of fuel temperature and pressure
on flow for each of the three nozzles and each fuel. Our objective was to
explore conditions in the flow range around 0. 42 cmvs (0. 4 gph). To do
this the pressure range was lowered from 207 to 345 kPa (30-50 psi) with
the 0. 89 cm3/s (0. 85 pgh) nozzle to 103-207 kPa (15-30 psi) the 1. 31 cm3/s
(1. 25 gph) nozzle. Fuel pressure was very stable and was controlled by
the regulated nitrogen pressure in the fuel chamber.
Fuel flow measurements were made with the burner operating. The
nozzle picked up considerable heat from the flame. Consequently the
nozzle temperature was substantially higher than the fuel heater tempera-
ture. The nozzle temperature was measured at the fitting into which the
nozzle was screwed. The nozzle orifice temperature probably was a few
degrees higher. Because of these complications the nozzle temperature
and therefore the fuel temperature varied somewhat as a function of flow
rate. We endeavored to compensate for this, but a 10 C uncertainty in
the temperature of fuel entering the nozzle probably occurred. Since we
were looking for the effect of a 100 C, or higher, fuel temperature change,
the uncertainty was not serious.
The flow decreased with decreasing pressure as expected. We expected
flow to increase slightly with increasing temperature due to lower viscosity.
When the boiling range was reached the flow was expected to decrease rapidly
due to vapor formation in the nozzle orifice. However, in all cases except
that with No. 1 oil in the 0. 89 cm3/s (0. 85 gph) nozzle, flow decreased with
increasing temperature. We believe this was due to the fact that fuel oil
boils over a rather wide temperature range. The light ends probably pro-
duce some small bubbles in the nozzle orifice at fuel temperatures as low
as 100 C. The restriction produced by bubble formation increases with
increasing fuel temperature.
The one case where the 0. 89 cm3/s (0. 85 gph) nozzle was operated at
205-215 C and 276 kPa (40 psi) with No. 1 oil (Figure 6) did show a sub-
stantial flow restriction. The No. 1 oil probably produced a substantial
amount of vapor under these conditions. No. 2 oil (Figure 7) did not show
any break, but probably would have shown a break in the curve at some
higher temperature.
In general we found we could operate reliably at 150 C and lower, but
coking became a problem during shut down at higher fuel temperatures.
Apparently, when the fuel was shut off, heat from the combustion chamber
raised the nozzle temperature even higher than its operating temperature.
EFFICIENCY
Combustion efficiency was calculated from excess oxygen concentration
and an assumed flue temperature rise. The actual flue temperature rise in
any real furnace would be a function of heat exchanger design. We, there-
fore, assumed that for any given set of combustion conditions, a heat ex-
changer could be designed to achieve the assumed flue temperature rise.
A flue temperature rise of 300 C (540 F) is a reasonable value.
16
-------
0.6-
- 0-6
0.5 -
ro
o
UJ
tr
o
0.4-
0.3 -
0.2-
0.1
0.4 gph
- 03
0.89cm/5(0-85 GPH) NOZZLE,
+ 207 kPo (30psi)
A 276 kPo (40psi)
* 345 kPa (50psi)
\
A\
25
50 75 100 125 150
NOZZLE TEMPERATURE, *C
175
200
FIG.6 EFFECT OF FUEL TEMPERATURE AND PRESSURE ON FLOW WITH NO.I OIL
-------
ro
u
r*
UJ
o
- 0.6
0-6 -
en 0.5 -
-0.5
0.4
_-0.4gph
0.3 -
0.2 -
0-1
\
XX
-0.3
0.89 Cm3/S (0.85 GPH) NOZZLE
+ 207 kPo ( 30psi)
A 276 kPo ( 40psi)
x 345 kPa ( 50psi)
0
25
50
75
100
125
150
175
200
NOZZLE TEMPERATURE, °C
FIG.7 EFFECT OF FUEL TEMPERATURE AND PRESSURE ON FLOW WITH NO.2 OIL
-------
CO
o
_l
u_
0.6-
05 -
E
o
r»
aj ~A
\~ 0.4
at:
0.3-
0.2 -J
0-1
-0-6
- 0.4 gph
-0.3
1.05 crr/B( 100GPH ) NOZZLE
o 138 kPo (20psi)
+ 207kPa ( 30psi)
A 276 kPo (40psi)
X345 kPo (50psi)
25
50
150
175
75 100 125
NOZZLE TEMPERATURE ,°C
FIG.8 EFFECT OF FUEL TEMPERATURE AND PRESSURE ON FLOW WITH NO-I OIL
200
-------
0.6 J
0.5 -
to
o
0.4 -
0.3 -
0-2 -
O.I -J
- 0.6
- 0.5
*- 0.4 gph
- 0.3
1.05 cnrT/S (1-00 GPH ) NOZZLE
o 138 kPo (20psi)
A 276 kPo (30psi)
+ 207 kPo(40psi)
25
50
175
200
75 100 125 150
NOZZLE TEMPERATURE,* C
FIG 9 EFFECT OF FUEL TEMPERATURE AND PRESSURE ON FLOW WITH N0.20IL
-------
0.8 J
hO.8
0.7-4
10
E
u
JS K 0.64
cc
O
054
0.4
0.3-1
0,\
0-6 gph
-0.5
J-0.4
207 kPo ( 30psi)
1.31 cm/S( L2 5 GPH) NOZZLE
• 103 kPa (ISpsi)
e !38kPo (20psi)
•h 207 kPo (30psi)
A 276 kPa (40psi)
25
50
75
100
125
150
175
200
NOZZLE TEMPERATURE, °C
FIG. 10 EFFECT OF FUEL TEMPERATURE AND PRESSURE ON FLUE WITH NO 2 OIL
-------
0.6-
05-
10
o
0.4-
0-3-
0.2-
O.H
^0.6
rO.5
0.4 gph
-0.3
25
50
1.31 cm^S (1.25 GPH) NOZZLE
• 103 kPo (ISpsi)
0138 kPo (20psi)
-t-207 kPa (30psi)
175
75 100 125 150
NOZZLE TEMPERATURE,°C
FIG.M EFFECT OF FUEL TEMFERTURE AND PRESSURE ON FLOW WITH NO.I OIL
200
-------
Figure 12 presents a correlation among excess oxygen, flue carbon
dioxide and excess air for the two fuels studied. A 0 C (32°F) dew point
was used in preparing Figure 12 because we used an ice bath water trap
to remove moisture from the flue sample.
Figure 13 gives the gross steady state efficiency (based on higher
heating value of fuel) as a function of flue temperature and oxygen content.
Swirl in combustion chambers has been defined by different authors in
somewhat different ways (6). We have used a simplified form of Beer's
definition which is derived~in the Appendix. The swirl parameter so de-
fined for this combustion chamber was 4. 5. It probably would be considered
moderately high.
This swirl produced clean combustion under almost all firing condi-
tions. It was feasible, therefore, to assume a burner of this type could
be operated at 2% excess oxygen (10% excess air). This would give a
steady state efficiency of slightly over 81 % with a 300 C flue temperature
rise. A heat exchanger designed to achieve a lower flue temperature rise
obviously would raise the efficiency.
EMtSSIONS - STEADY STATE
Nitrogen Oxides
Although the nitrogen oxides are reported as NO , a cold trap used in
the sampling line (Figure 3) probably removed most of the NO? and some
hydrocarbons. It was necessary to add this cold trap to protect the NDIR
Instrument and Oxygen Analyzer from any condensed water droplets. The
experimental set-up dictated the location at the point shown in Figure 3.
Under most operating conditions, the concentration of condensed species
was probably quite small.
The response of the gas analyzing instruments and sampling system
was quite slow. Thus, only steady state or slowly changing conditions
could be observed. With the exception of one test, the data reported are
steady state values.
The oxidation of nitrogen in a flame depends on time and temperature.
Higher temperatures increase the rate of formation of NO , but shorter
residence times reduces the production of NO • Figures ?4 and 15 show
both of these effects for the 0. 89 cm3/s (0. 85xgph) and 1. 31 cm3/s (1. 25
gph) nozzles operating with No. 2 fuel oil.
The flue NOX measured, increased as the excess oxygen decreased due
to the higher flame temperature associated with more nearly stiochiometric
combustion. Incomplete mixing limited this effect at an excess oxygen level
of about 0. 3% with the 0. 89 cmvs (0. 85 gph) nozzle when operated at 344 kPa
(50 psi) with 50 C, No. 2 oil. The flow rate under these conditions was
0. 49 cm3/s (0. 47 gph) or 55% of rated capacity. The 1. 31 cm3/s (1. 25 gph)
nozzle operated at 207 kPa (30 psi) with No. 2 oil at 150 C had a flow rate
of only 0. 45 cm3/s (0. 43 gph) or 34% of capacity. It showed the effect of
incomplete mixing (maxima in the NOX curve) at about 1% excess oxygen.
23
-------
15 -
14
13
12
8
10% EXCESS AIR
0.2 FUEL 01LTH/C= 1.7467
X
o
FLUE CARBONDI
5 -
9
No. 1 FUEL 02 L .
H/C= 1.8898
-
80
90
100
2468 10
EXCESS OXYGEN, %
FIGURE !2- CORRELATION OF FLUE PARAMETERS
24
-------
O
z
UJ
O
li.
UJ
80
82
80
78
^ 76
74
72
70
68
FLUE TEMP RISE
NO. 2 FUEL OIL
8
10
EXCESS OXYGEN, %
FIGURE 13. GROSS STEADY STATE EFFICIENCY
25
-------
In most, but not all, cases the NOX decreased with increasing fuel
temperature and pressure. Improved atomization achieved with increased
fuel temperature and pressure would permit faster mixing in the turbulent
air stream. This, in turn, would lead to smaller flame volume and less
residence time in the high temperature region. Measurements and calcula-
tions by other investigators (6>, 1) also have shown that increased swirl tends
to reduce NOX formation.
Thus the use of high turbulence and a thermally augmented atomizer
appears to be beneficial in reducing NOx formation. Provision for recircula-
tion could enchance this situation.
The fuel nitrogen was measured as shown in Table 1. If this was all
converted to NOX, the contribution of fuel nitrogen to the NOX concentration
is given in Table 4.
TABLE 4. CONTRIBUTION OF FUEL NITROGEN TO NOV
Fuel
No.
1
2
Fuel
Nitrogen
%
. 0064
. 0120
NOx Normalized
to 3% Flue 02
PPM
8. 1
15.5
Figures 14 and 15 show that at 8% excess oxygen the NOX can be attri-
buted mainly to fuel nitrogen. However, at lower flue oxygen levels, more
of the NOX must come from atmospheric nitrogen.
Other Emissions
Smoke, total hydrocarbons and carbon monoxide were essentially un-
detectable for all firing conditions except when excess oxygen levels were
well below 1 °]o.
The total hydrocarbons analyzer had two ranges, 100 ppm and 1000 ppm
full scale. The minimum detectability was about 1 ppm. The carbon monox-
ide sensor was calibrated for 1000 ppm full scale with a resolution of about
10 ppm. Methane could be detected at a level of about 5 ppm.
The high degree of swirl achieved complete combustion of the fuel with
almost no excess air. During one run with theQ0. 89 cm3/s (0. 85 gph) nozzle
firing No. 2 oil at a nozzle temperature of 165 C, the excess oxygen was ad-
justed to less than 1 % with 0 to 1 smoke number. Total hydrocarbons, carbon
monoxide, and methane were immeasurable. The NOX concentration was
then 55 ppm. The inlet air pressure was then reduced slightly to yield a
0. 15% decrease in excess oxygen (0. 2% increase in flue CO2). Carbon-
monoxide increased to more than 1000 ppm and NOX decreased to 51 ppm.
Total hydrocarbons and methane were still unmeasurable, however. Thus
the point of incomplete combustion was very pronounced and occurred at
about 0. 8% excess oxygen.
26
-------
100
ro
O
O
UJ
M
QL
O
80
60
E
o. 40
Q.
20
0
0.89cm3/S(0.85)GPH NOZZLE, NO. 2 OIL
344kPoT 50°C
344kPotl50°C
x L —
207kPat50°C
276kPQ]5O°C
276kPa,l5Qt>C
20 7k Pa
I 234567
EXCESS OXYGEN,%
FIGURE 14. EFFECT OF OPERATING PARAMETERS ON NOX
8
-------
100
to
CO
ro
O
O
liJ
M
a:
o
80
60
40
o.
•»
o*
UJ
_l
U.
20
1.31 crri7S( 1.25 GPH)NOZZLE No.2 OIL
138 kPo. 178°C
207 kPot 172 °C
103 kPo.85°C
103 kPo. I50°C
207 kPo. I50°C
138 kPo. 150° C
138 kPo. IOO°C
345
EXCESS OXYGEN, %
8
FIGURE 15. EFFECT OF OPERATING PARAMETERS ON NOX
-------
TRANSIENT PERFORMANCE
Some smoke, hydrocarbons, carbon monoxide and even me thane were
usually observed right after ignition. It seemed that these disappeared
more quickly when hot fuel was used. This initial burst of emissions tended
to saturate the sampling system and measuring instruments. Since the gas
analysis system was slow to recover, we made a practice of removing the
sampling tube from the combustion chamber during ignition. Although the
instrumentation was not well suited to transient measurements, one tran-
sient experiment was conducted. The 0. 85 gph nozzle was used with No. 2
oil at a pressure of 345 kPa(50 psi). Two runs were made; the first was
with unheated oil and the second was with oil heated to yield a 150 C nozzle
temperature. The results are presented in Table 5.
TABLE 5. EFFECT OF FUEL TEMPERATURE ON TRANSIENT PERFORM.
0. 89 cm /s (0. 85 GPH) NOZZLE, 345 kPa, No. 2 OIL
Time After Nozzle Flue
Ignition Temp. 09 Smoke NO
Min. °C To No. PPM
0 35
1.5 80 --- 5
5 88 --- 0
10 100 3.1 0 40
15 100 3.0 6 40
20 100 3. 0 5 41
0
1.5
5
10
15
20
23
71
121
150
150
150
150
150
5.0
4.7
4.5
4.5
4.5
0
1
0
0
3
1
___
26
28
28
29
29
29
-------
After 1. 5 minutes of operation the unheated oil gave a smoke No. 5
whereas the heated oil was already down to zero smoke. The nozzle with
unheated oil rose to 80 C in 1. 5 minutes whereas the nozzle temperature
with heated oil reached 121 C in 1. 5 minutes. Within 5 minutes the smoke
No. was down to 0-1 in both tests. The excess oxygen was about 3% for
the unheated oil case, whereas it was about 5% with heated oil. This dif-
ference in excess oxygen had no effect during steady state runs since smoke
was immeasurable unless excess oxygen was reduced well below 1%. For
some unexplained reason, smoke increased again after 15 to 20 minutes.
This phenomenon was not observed in any other tests which were run for
as long as 30 minutes.
The NO level was higher with unheated oil. The incre ase was un-
doubtedly diie to the lower excess oxygen. A knowledge of carbon monoxide
and total hydrocarbon levels during ignition and shut down is needed to fully
assess the effect of thermal aerosol atomization. However, response of the
sampling train and measuring instruments was so slow compared to the com-
bustion transients that these measurements were unreliable. Thus, the tests
were not conclusive, but they did support the qualitative observation that
heated fuel produced a shorter transient.
EFFECT OF SWIRL
The high swirl seemed to mask any effect of improved atomization with
hot fuel. We therefore modified the burner to eliminate swirl. The objective
was to try and answer the original hypothesis that improved atomization
achieved through the use of hot fuel would reduce emissions and improve
efficiency. Four additional air tubes were installed in the end of the mixing
chamber to admit air parallel to the fuel jet axis. The tangential air j eis
were plugged. The results of a comparison using No. 1 oil and 1. 05 cm /s
(1 gph) nozzle are presented in Table 6. When the swirl was eliminated
there was a correlation between smoke and fuel temperature. A nozzle
fuel temperature of 114 C produced No. 9 smoke, 118 ppm CO, 5 ppm CH4,
31 ppm total hydrocarbons and 11 ppm NO even with a excess oxygen of
4. 9%. Increasing the fuel temperature to 179 C with essentially the same
excess oxygen reduced all emissions except NO. . A further increase in
fuel temperature to 199 C and a decrease in exdess oxygen to 4.1% reduced
emissions still further. All emissions except NO. disappeared when swirl
was restored even though fuel temperature was varied from 121 to 178 G
and excess oxygen was varied from 8.1 to 2.1%. Elimination of swirl appears
to reduce NO slightly; however, this is achieved at increased levels of smoke,
carbon monoxide and total hydrocarbons.
With swirl present the flame was very compact and non-luminous.
Eliminating swirl produced a much larger volume luminous flame. The
pictures in Figures 16 and 17 show this difference.
3
The total pressure of the air supplied when operating at 0. 42 cm /s
(0. 4 gph) and 2% excess oxygen was about 1. 2 kPa (5 in. water). This
undoubtedly could be reduced by increasing the area of the air inlet passages
and by reducing the swirl parameter to perhaps 2. Swirl does, however,
add to the pressure requirement of the blower.
30
-------
TABLE 6. EFFECT OF FUEL TEMPERATURE AND EXCESS OXYGEN
ON SMOKE - No. 1 OIL, 1. 05 cm3/s (1 GPH) NOZZLE
Without Swirl:
Fuel
Toemp
114
179
199
121
163
159
178
Fuel
Press.
kPa
138
138
138
With Swirl:
138
138
138
138
Excess
Oxygen
%
4.9
4.8
4.1
8.1
5.1
5.4
2.1
Smoke
No.
9
3
4
0
1
0
0
CO
PPM
118
45
10
0
0
0
0
CH4
PPM
5
0
0
0
0
0
0
THC
PPM
31
5
3
0
0
0
0
NOX
11
23
19
18
25
29
63
31
-------
FIGURE 16, NON-LUMINOUS FLAME WITH SWIRL, DARK CENTER SPOT IS
INLET ORIFICE WITH MIXING CHAMBER IN BACKGROUND.
-------
FIGURE 17. LUMINOUS FLAME WITHOUT SWIRL
33
-------
REFERENCES
1. Bonne, U. and A. Johnson. Thermal Efficiency in Non-Modulating
Combustion Systems. Conference on Improving Efficiency in HVAC
Equipment and Components in Residential and Small Commercial
Buildings, Purdue University, October 1974.
2. Bonne, U., J. E. Janssen, R. H. Torborg, and A. Johnson. Digital
Simulation of the Seasonal Efficiency of Combustion Heating Systems.
Seminar on Efficient Fuel Utilization in Heating Systems, ASHRAE,
June 1975.
3. Bonne, U. , J. E. Janssen and R. H. Torborg. Efficiency and
Relative Operating Cost of Central Combustion Heating Systems,
IV. Oil Fired Residential Systems. Presented at ASHRAE 1977
Semi-Annual Meeting, Chicago, Illinois, February 1977.
4. Bonne, U., L. Katzman, and G. E. Kelly. Effect of Reducing Excess
Firing Rate on the Seasonal Efficiency of 26 Boston Oil-Fired Heating
Systems. Conference on Efficiency of HVAC Equipment and Compon-
ents II, Purdue University, Indiana, April 12-15, 1975, Proceedings,
p. 81.
5. Tenney, W. Fuel Aerosolization Apparatus. U. S. Patent No.
4, 013, 396, 22 March 1977.
6. Baldwin, J. D. C. and C. H. Long. Effect of Swirl on NOX Emissions
from a Gas-Fired Bruner. ASME paper 74-WA/FV-3, December, 1974.
7. Bonne, U. Source Emissions Control - The Key to Clean Air. Honey-
well Computer Journal, 7(3):199, 1973.
34
-------
APPENDIX
Drop Size
Surface tension is defined as the work required to displace a unit
area of fluid surface.
a = WK/A (1)
For a spherical droplet,
WK = aA =
-------
Air enters the vortex chamber tangential to the axis and at a radial
distance, r^. The air is accelerated due to the conservation of angular
momentum as it goes through the mixing orifice which has a radius, ro.
Thus the tangential velocity at the orifice is:
Vt = (8)
where: Q = air flow rate
Ai = area of air inlet tubes
i-j. = radial distance of tubes from axis
ro = radius of orifice
The axial velocity, Va is
va • -2- (9)
Then
(10).
Four the four air inlet tubes
Aj = 4rrri2
= rt r0
NS = 1I_4£ (ID
= 1. 5 in.
r0 = . 75 in.
rj = . 25 in.
= (1.5) (75)
4 (. 25>
36
-------
TECHNICAL REPORT DATA
(Please read luuruelions un llic reverse be fore completing)
. REPORT NO.
EPA-600/7-77-108
2.
3. RECIPIENT'S ACCESSION NO.
. TITLE AND SUBTITLE
tudy of a Thermal Aerosol Oil Burner
5. REPORT DATE
September 1977
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
.E. Janssen, J.J. Glatzel, E.R. Wabasha, and
U. Bonne
. PERFORMING ORGANIZATION NAME AND ADDRESS
Honeywell, Inc.
0701 Lyndale Avenue, South
Bloomington, Minnesota 55420
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
68-02-2194
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 9/76-5/77
14. SPONSORING AGENCY CODE
EPA/600/13
.SUPPLEMENTARY NOTES jERL-RTP project officer for this report is Robert E. Hall, Mail
Drop 65, 919/541-2477.
ABSTRACT Tne repOrt gives results of a. study of a thermal aerosol oil burner, aimed
at counteracting the poor atomization and excess burner capacity that are known to
reduce seasonal efficiency and contribute to excess emissions in residential oil
burners. Generation of a thermal aerosol of the fuel was shown to improve combustion
in terms of: (1) increased quantity and volume of fuel droplets 1 micrometer and
smaller; (2) permitted firing rate reduction in standard nozzles of 50 to 70%; (3) when
combined with swirl (swirl parameter = 4. 5), gave increased combustion efficiency
3y permitting operation at 2% flue oxygen with nonluminous flame, zero Bacharach
smoke No. , no hydrocarbons, and no detectable CO; (4) reduced NOx formation in
most cases; (5) reduced emissions during burner start-up; and (6) permitted modula-
tion of firing rate without affecting combustion adversely.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Combustion
Fuel Oil
Burners
Residential Buildings
Aerosols
b.IDENTIFIERS/OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Thermal Aerosols
COSATI rickl/Group
13B
21B
21D
ISA
13M
07D
13. DIS ri-tldUTION STATEMENT
Unlimited
19. SECURITY CLASS f'lliis Kcport)
Unclassified
21. NO. OK I'AGLS
45
20. SECURITY CLASS (Thi
Unclassified
22. I'RICE
EPA Form 221'0-1 (9-73)
37
------- |