INCINERATOR OVERFIRE MIXING STUDY
a report to
CONTROL SYSTEMS DIVISION
OFFICE OF AIR PROGRAMS
ENVIRONMENTAL PROTECTION AGENCY
FEBRUARY 1972
Arthur D Little, Inc.
-------
INCINERATOR OVERFIRE MIXING STUDY
A Report to
CONTROL SYSTEMS DIVISION
OFFICE OF AIR PROGRAMS
ENVIRONMENTAL PROTECTION AGENCY
UNDER CONTRACT EHSD 71-6
February 1972
By
Arthur D. Little, Inc.
Cambridge, Massachusetts
Walter R. Niessen - Principal Investigator
Principal Authors
Dr. C. Michael Mohr
Raymond W. Moore
Dr. Adel F. Sarofim(l)
Anne N. Dimitriou
Contributors
Dr. Richard Kronauer(2) Dr. Frank B. Tatom(3)
James I. Stevens
(1)
Assistant Professor, Chemical Engineering, Massachusetts Institute
of Technology, Cambridge, Massachusetts
(2)
(3)
Professor, Harvard University, Cambridge, Massachusetts
Georgia Institute of Technology, Atlanta, Georgia
ADL Reference No. 72940
Arthur D Little, Inc
-------
ABSTRACT
Incineration, as an increasingly important tool in municipal solid
waste management, can be a significant source of air pollution in urban
areas. Combustible air pollutants (carbon monoxide, soot or char, and
hydrocarbons) exist in the gases leaving incinerator furnaces as a con-
sequence of poor combustion. Since these pollutants are removed from the
exit gases with difficulty, improving the combustion environment is an
attractive approach for improved pollutant emission control.
The processes occurring in a burning refuse bed are analyzed to yield
estimates for the rate and quantity of combustibles emitted from the bed.
It is shown that gases of over 14% carbon monoxide content can be released.
Methods are then developed to analyze the flow through the furnace. The
analysis shows that stratification of the flow and incomplete mixing of
fuel and sufficient oxygen-containing gases can occur. Design equations
are presented describing the behavior of overfire air or steam jets for
mixing, tempering and/or bringing oxygen to the combustible gases. Refuse
incinerator overfire air system design methods developed from the above
analyses are presented and contrasted with similar design methods for
solid fuel bed coal-firing furnaces.
i
Arthur D Little, Inc
I'
-------
ACKNOWLEDGEMENT
This effort was supported by the Control Systems Division, Office of
Air Programs of the Environmental Protection Agency under Contract EHSD
71-6. We wish to express our appreciation to Mr. R. C. Lorentz of CSD for
his many contributions and suggestions. Our thanks also go to Mr. Willard
S. Pratt and others of the City of Newton, Massachusetts for their coopera-
tion and assistance in developing a municipal incinerator overfire mixing
test plan. ..
ii
Arthur D Little, Inc
-------
.
--.--
TABLE OF CONTENTS
Page
i
ABSTRACT
ACKNOWLEDGEMENT
H
vi
LIST OF TABLES
LIST OF FIGURES
vii
LIST OF SYMBOLS
CHAPTER I - INTRODUCTION AND SUMMARY
x
A. BACKGROUND
B. OBJECTIVE & SCOPE
C. APPROACH
D. SUMMARY
E. REFERENCES
CHAPTER II - BED BURNING MODELS
I-I
I-I
1-3
1-4
1-5
1-9
A.
INTRODUCTION
PREVIOUS WORK
BED COMBUSTION MODEL
II-I
II-I
II-2
B.
II-16
II-20
C.
D.
CALCULATIONS USING BED COMBUSTION MODEL
PRACTICAL IMPLICATIONS
II-23
II-24
II-25
E.
F.
SUMMARY
REFERENCES
G.
A.
B.
INTRODUCTION
PREVIOUS WORK
III-l
III-l
CHAPTER III - FURNACE FLUID FLOW
C.
ANALYSIS OF FURNACE FLOW
CASE STUDIES
III-3
1II-5
III-I0
D.
1.
2.
III-I0
1II-14
III-17
OVERFlRE REGION
CHANNEL FLOW REGION
E.
APPLICATION TO DESIGN AND PERFORMANCE EVALUATION
SUMMARY
III-19
1II-20
F.
G.
REFERENCES
Hi
Arthur D Little, Inc
-------
TABLE OF CONTENTS (Con t. )
Page
A.
B.
INTRODUCTION
THE USE OF JETS FOR COMBUSTION CONTROL
1. JET DESIGN FOR INCINERATORS--A STATEMENT OF
THE PROBLEM
2. EXPERIENCE IN JET APPLICATION FOR COAL-BURNING
SYSTEMS
IV-l
IV-l
IV-5
CHAPTER IV - OVERFlRE JETS
IV-5
IV-6
C. REVIEW OF THE PRIOR ART
1. ROUND ISOTHERMAL JETS
2. BUOYANCY EFFECTS
3. CROSSFLOW EFFECTS
4. BUOYANCY AND CROSS FLOW
5. DESIGN METHODS
D. CONTRIBUTIONS TO THE ART
IV-12
IV-13
IV-16
IV-19
IV-29
1.
MATHEMATICAL MODELING OF COMBINED BUOYANT AND
CROSSFLOW EFFECTS
QUANTITATIVE MODEL EXPERIMENTS
COMBUSTION EFFECTS
IV-31
IV-37
IV-37
2.
3.
4.
TENTATIVE INCINERATOR OVERFIRE AIR JET DESIGN METHOD
IV-41
IV-45
IV-51
IV-55
IV-56
E.
F.
SUMMARY
REFERENCES
CHAPTER V - DESIGN METHODS FOR INCINERATOR OVERFIRE AIR SYSTEMS
A. STATEMENT OF THE PROBLEM
B.
EVALUATION OF SYSTEM OPERATING CHARACTERISTICS
V-I
V-I
V-2
JET BEHAVIOR
V-2
V-7
V-8
l.
2.
3.
BED PROCESSES
FURNACE ENVIRONMENT
iv
Arthur D Little, Inc
-------
TABLE OF CONTENTS (Cont.)
CHAPTER VI - TEST PLAN
A.
INTRODUCTION
TECHNICAL DISCUSSION
B.
1.
2.
3.
4.
5.
6.
CONCEPTUAL DESIGN
PREPARATION FOR TESTS
UNCONTROLLED EMISSION RATES
INCINERATOR BEHAVIOR
MIXING-EXPERIMENTS
DESIGN & OPERATING GUIDELINES
C. REFERENCES
CHAPTER VII - BIBLIOGRAPHY
A.
BIBLIOGRAPHY
1. SMOKE. ABATEMENT, APPLICATION OF OVERFIRE AIR
2. BASIC JET AND MIXING THEORY
3. OVERFIRE JET DESIGN
4. SOOT FORMATION AND BURNOUT
5. COMBUSTION ON FUEL BEDS
6. FURNACE DESIGN
7. FURNACE GAS FLOW
8. MISCELLANEOUS
ABSTRACTS OF FOREIGN ARTICLES
B.
APPENDIX A - EQUATIONS FOR INVISCID NON-CONDUCTING FLOW IN
TWO ZONES
APPENDIX B - INSTABILITY IN STRATIFIED FLOW
APPENDIX C - DECAY OF SHEAR VELOCITY IN A CHANNEL
APPENDIX D - EXPERIMENTAL APPARATUS FOR JET-IN-CROSSFLOW
STUDIES
APPENDIX E - DERIVATION OF COMBINED EFFECT MODEL
APPENDIX F - RESULTS OF FLUE SAMPLING TESTS AT MUNICIPAL
INCINERATOR, NEWTON, MASSACHUSETTS
v
Page
VI-l
VI-l
VI-3
VI-3
VI-9
VI-9
VI -11
VI-13
VI-IS
VI-17
VII -1
VII - 3
VII-3
VII - 7
VI 1-11
VII-13
VII-14
VII-2l
VII-26
VII-28
VII-29
A-I
B-1
C-l
D-l
E-l
F-l
Arthur D Little, Inc
-------
Table No.
I-I
II-I
II-2
II-3
II-4
III-1
III-2
III-3
IV-l
IV-2
V-I
V-2
V-3
V-4
V-S
VI-1
LIST OF TABLES
ESTIMATED INCINERATOR EMISSIONS
SUMMARY OF OCEANSIDE DATA
PERCENTAGE OF UNREACTED AIR BY-PASSING
WATER-GAS SHIFT EQUILIBRIUM TEMPERATURE
GASIFICATION PRODUCTS AS A FUNCTION OF MOISTURE CON-
TENT, FRACTION OF CARBON GASIFIED, AND HEAT LOSS
BASES FOR GASES ANALYZED
Page
1-2
II-4
II-6
II-7
II-21
III-ll
RESULTS OF CALCULATIONS 111-13
PERCENT AIR ADDITION TO FUEL-RICH HOT GAS FOR COMPLETE 111-16
COMBUSTION
CHEMICAL CHARACTERISTICS OF COAL AND REFUSE
CONDITIONS FOR BUOYANCY-CROSSFLOW MODEL TESTS
TOTAL UNDERGRATE AIR FLOW RATES
FRACTION OF TOTAL UNDERGRATE AIR BYPASSING BED
THEORETICAL BED COMBUSTION CHARACTERISTICS
THEORETICAL CHAR REGION OFF-GAS CHARACTERISTICS
THEORETICAL GAS VELOCITIES ABOVE REFUSE BEDS
SET POINTS FOR GRATE AIR FLOW STUDIES
vi
IV-9
IV-42
V-3
V-3
V-S
V-6
V-9
VI-10
Arthur D Little, Inc
-------
Figure No.
II-I
11-2
11-3
II-4
II-5
II-6
II-7
II-a
II-9
III-l
IV-l
IV-2
IV-3
IV-4
IV-5
IV-6
IV-7
LIST OF FIGURES
SCHEMATIC OF OCEANSIDE INCINERATOR SHOWING LOCATION
OF GAS SAMPLE POSITIONS
ESTIMATED ENERGY RELEASE RATES FOR OCEANSIDE INCINERA-
TOR
TRAVEL OF EVAPORATION AND BURNING FRONTS AND TOTAL
WEIGHT LOSS AS FUNCTIONS OF TEST TIME IN A TYPICAL
BED OF REFUSE HAVING A MOISTURE CONTENT OF 50 PERCENT
AND A BED DEPTH OF 18 INCHES IN BU MINES (19-inch
diameter test incinerator)
SCHEMATIC DIAGRAMS OF BED BURNING MODELS
UNDERFEED BURNING, ACTION THROUGH COAL BED EXPRESSED AS
WEIGHT OF FUEL PRODUCTS CARRIED PER POUND OF DRY AIR
SUPPLIED, 3/4 - 1 INCH - ILLINOIS COAL
UNDERFEED BURNING, HIGH-TEMPERATURE COKE: RATE OF IG-
NITION AND RATE OF BURNING WITH RATE OF PRIMARY AIR &
SIZE OF COKE AS VARIABLES
VOLATILE MATTER AND ASH CONTENTS OF COAL BED LAYERS,
LOW AIR RATE
ISO-VOL. AND ISO-ASH CONTENTS THROUGHOUT THE COAL BED
SCHEMATIC OF CROSS-FEED BED BURNING PROCESS (Assuming
Combustion Process Raw+Dry~Volatilize~Char+Ash)
FURNACE FLOW MODEL FOR ANALYSIS
REGIONS IN JET FLOW
RELATION BETWEEN COAL CHARACTERISTICS AND THE SIZE OF
COMBUSTION SPACE REQUIRED IN USMB TEST FURNACT AT COM-
BUSTION RATES OF 50 LBS/HR FT2 AND 50% EXCESS AIR
LINES OF EQUAL HEATING VALUE (KG-CAL PER STD CU M) OF
FLUE GAS FIRING LOW-VOLATILE BITUMINOUS COAL AT A FUEL
RATE OF 28 LB PER SQ FT PER HOUR
COMPARISON OF OBSERVED FLAME CONTOURS AND CALCULATED
TRAJECTORIES OF OVERFlRE AIR JETS
SCHEMATIC OF JET FLOW
COMPARISON OF THE PREDICTIONS OF ABRAMOVICH AND FIELD
ET. AL. WITH THE DATA OF SYRKIN AND LYAKHOUSKY ON
BUOYANT JET BEHAVIOR
JET CROSS-SECTION AND CIRCULATION PATTERNS FOR ROUND
JETS IN CROSSFLOW
vii
Page
II-3
II-5
II-a
II-I0
II-II
II-12
II-13
11-15
II-17
1II-6
IV-4
IV-7
IV-8
IV-II
IV-17
IV-18
IV-20
Arthur D Little, Inc
-------
Figure No.
IV-8
IV-9
IV-10
IV-ll
IV-12
IV-13
IV-14
IV-15
IV-16
IV-17
IV-18
IV-19
IV-20
IV-21
IV-22
VI-1
VI-2a
VI-2b
VI-3
B-1
B-2
C-1
D-1
LIST OF FIGURES (Cant.)
COORDINATE SYSTEM FOR ROUND JET IN CROSSFLOW
TRAJECTORY OF CONCENTRATION AND VELOCITY AXES FOR JETS
IN CROSSFLOW (DATA OF PATRICK) M = 0.05
COMPARISON OF TRAJECTORIES AT M = 0.001 (uo/u1 = 31.6)
FOR JETS IN CROSSFLOW
COMPARISON OF TRAJECTORIES AT M = 0.01 (uo/ul = 10) FOR
JETS IN CROSS FLOW
Page
IV-22
IV-24
IV-26
IV-27
EFFECT OF JET SPACING ON TRAJECTORY FOR JETS IN CROSSFLOW IV-28
(AFTER IVANOV)
COMPARISON OF IVANOV'S TRAJECTORY CORRELATION (Equation IV-30
IV-13) WITH JET PENETRATION CORRELATION
RELATIVE IMPORTANCE OF BUOYANCY AND DRAG
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 1
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 2
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 3
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 4
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 5
TEMPERATURE VS RADIAL DISTANCE (Jet Diameter = 1. 5")
TEMPERATURE VS RADIAL DISTANCE (Jet Diameter = 2")
TEMPERATURE VS RADIAL DISTANCE (Jet Diameter = 4")
SCHEMATIC DIAGRAM:
FURNACE FLOW
OPPOSED JETS
TWO-FLUID MODEL OF INCINERATOR
INTERLACING JETS
CROSS-FURNACE STATIONARY (CFS) PROBE
SCHEMATIC OF STRATIFIED FLOW CONDITION
GROWTH RATE OF DISTURBANCES IN NON-ISOTHERMAL PARALLEL
FLOW
DIAGRAM ILLUSTRATING MOMENTUM EXCHANGE IN A SHEAR LAYER
LOW VELOCITY WIND TUNNEL
viii.
IV-39
IV-42
IV-43
IV-43
IV-44
IV-44
IV-47
IV-48
IV-49
VI-5
VI-7
VI-7
VI-16
B-2
B-5
C-2
D-2
Arthur D Little, Inc
-------
Figure No.
D-2
D-3
E-l
LIST OF FIGURES (Cont.)
APPARATUS TO PRODUCE COLD NITROGEN-VAPOR JET
PHOTOGRAPH OF TEST APPARATUS
TRAJECTORY OF DEFLECTED JET
ix
Page
D-3
D-S
E-2
Arthur D Little, Inc
-------
A
a
a
1,2.. .
B
Bl
b
C
n
c
c
c
m
Co
cp
D
d
do
E
e
F
FxB
FxD
f
G
g
LIST OF SYMBOLS
2
Duct cross-sectional area (ft )
Moles CO generated in bed per mole of oxygen (Equation II-I)
Constants (dimensionless)
Buoyant force on jet in crossflow (lbf)
Constant (ft2/sec2)
Half-width of shear layer (Appendix C) (ft)
Effective drag coefficient of crossflow on jet (dimensionless)
Rate of spread of jet width in cross flow (dimensionless)
Time averaged concentration of jet fluid along radius (lb/lb of
mixture)
Time averaged jet centerline concentration of jet fluid (lb/lb
of mixture)
Time averaged concentration of jet fluid at nozzle (lb/lb of
mixture)
Heat capacity of ambient fluid (Btu/lb OF)
Drag force on jet in crossflow (lbf)
Moles H20 generated in bed per mole of oxygen (Equation II-I)
Nozzle diameter (ft)
Constant (Equation 111-5) (dimensionless)
Moles C generated in bed per mole of oxygen (Equation II-I)
Force ratio on buoyant jet in cross flow (dimensionless)
Buoyant force in x direction (lbf)
Drag force in x direction (lbf)
Wall friction factor (dimensionless)
Solutions to characteristic equation (Equation B-8)
2
Acceleration due to gravity (32.2 ft/sec )
x
Arthur D Little, Inc
-------
H
H
c
h
hfg
i
j
K
k
k
c
k
w
L
Lj
t
M
M
x
m
mo
m
x
N
NR
NFr
NGr
n
P
LIST OF SYMBOLS (Cont.)
Total head (feet water gauge)
Heat of combustion of furnace gases (Btu/lb oxygen consumed)
Jet width in crossflow (feet)
Latent heat of vaporization (Appendix D) (Btu/lb m)
Moles H2 generated in bed per mole of oxygen (Equation II-I)
Moles C(H20)n gasified per mole of oxygen (Equation II-I)
Wavenumber of disturbance of interface (Appendix B) (ft-l)
Coefficient - Equation IV-18 (dimensionless)
-1
CO-C02 equlibrium constant (atm. )
Water gas shift equilibrium constant (dimensionless)
Length of physical system (feet)
The jet penetration distance - Equation IV-18 (feet)
Path length of axis of curved jet (feet)
Momentum flux ratio of external to jet flow (dimensionless)
Rate of change of jet momentum in x direction (lbf)
Constant
Jet mass flow at nozzle (lb /see)
m
Jet mass flow at distance x (lb /sec)
m
Number of jets in row - Equation IV-23 (dimensionless)
Reynolds number (dimensionless)
Froude number (dimensionless)
Grashof number (dimensionless)
Moles H20 per mole
Pressure (lbf/ft2)
of carbon in refuse
xi
Arthur D Little, Inc
-------
!-u
,
p
QL
Qp
QR
QT
Qu
q
qH
r
S
S
n
S
no
s
T
Tl
TC
TF
TH
TR
T
c
Tj
T
m
To
LIST OF SYMBOLS (Cont.)
Moles C02 generated in bed per mole of oxygen (Equation II-I)
Energy loss rate by radiation from bed (Btu/lb mole of oxygen)
Energy loss rate by convection from bed (Btu/lb mole of oxygen)
Energy release rate in bed (Btu/lb mole of oxygen)
Total quantity of overfire air (cfm)
Energy release rate above bed (Btu/lb mole of oxygen)
2 2
Term defined in Equation A-6 (lb m/ft sec )
Heater input power (Appendix D) (Btu/sec)
Radial distance from jet centerline (ft)
Dimensionless jet spacing (s/do)
Cross-sectional area of jet in cross flow (ft2)
2
Cross-sectional area of jet at nozzle (ft )
Spacing between jets (feet)
Gas Temperature (degrees absolute)
Temperature of furnace gases (OR)
Temperature of cold gases (OR)
Minimum ignition temperature of furnace gases (OR)
Temperature of hot gases (OR)
Reference temperature for enthalpy (OR)
Temperature of mixture (after combustion) of nozzle fluid and
ambient (OR)
Average temperature of nozzle and entrained fluid in jet (OR)
Temperature of mixture (uncombusted) of nozzle fluid and
ambient (OR)
Temperature of jet fluid at nozzle (OR)
xii
Arthur D Little, Inc
-------
t
u
ul
u'
Uj
u
u
m
Uo
v'
w
x
x
y
y
z
z
a
0.0
8
80
y
o
LIST OF SYMBOLS (Cont.)
Time (see)
Gas velocity (feet/see)
Velocity of crossflow fluid (feet/see)
r.m.s. fluctuating velocity component in axial direction
(feet/see)
Average velocity of nozzle and entrained fluid in jet (feet/
see)
Time averaged jet velocity along radius (feet/see)
Time averaged jet centerline velocity (feet/see)
Time averaged jet nozzle velocity (feet/see)
r. m. s.
see)
fluctuating velocity component in radial direction (feetl
Mass flow per unit area (lb /ft2)
m
Dimensionless distance from jet entry plane
Distance from reference point or nozzle (feet)
Dimensionless distance above jet entry plant
Vertical distance above reference point or nozzle (feet)
Height of physical system (feet)
Height above reference plane (feet)
Angle between jet centerline and the horizontal for up-flowing
ero5sflows (degrees)
Injection angle of jet above horizontal (degrees)
Angle between jet and vertical (degrees)
Injection angle between jet and vertical (degrees)
Secant 6 (dimensionless)
Temperature coefficient of expansion (reciprocal degrees
absolute)
xiii
Arthur D Little, Inc
-------
e:
r;;
II
jJ
v
E:
E;o
P
Pa
Pc
Ph
Pj
Po
C1
(p(x,y)
IjJ
~
LIST OF SYMBOLS (Cont.)
Exponent in Equation IV-46 (dimensionless)
Vertical displacement of disturbed interface (Appendix B) (feet)
Oxygen demand of furnace gases (lb 02/lb furnace gas)
Viscosity (lb 1ft see)
m
Characteristic length (feet)
Plane jet half-angle of spread in a channel (radians)
Plane jet half-angle of spread in unconfined space (radians)
Gas density (lb Ift3)
m
Ambient fluid density at nozzle (lb Ift3)
m
Cold gas density (lb Ift3)
m
Hot gas density (lb Ift3)
m
Average density of nozzle and enttained fluid in jet (lb Ift3)
m
3
Jet fluid density at nozzle (lb 1ft)
m
-1
Growth exponent (Appendix B) (see)
Potential function (Appendix B)
Characteristic thickness of shear layer (Appendix B) (feet)
Concentration of a component of nozzle fluid in the jet relative
to the nozzle concentration (dimensionless)
xiv
Arthur D Little, Inc
-------
CHAPTER I
INTRODUCTION AND SUMMARY
Arthur D Little, Inc
-------
CHAPTER I
INTRODUCTION AND SUMMARY
A.
BACKGROUND
Solid waste disposal by incineration is now and will continue to be
an important part of our national solid waste management program. Be-
cause of this reality and in the light of increasingly stringent environ-
mental constraints and standards, reliable and effective incinerator de-
sign methodologies are needed. As described in a previous Arthur D. Little,
Inc. (ADL) report titled "Systems Study of Air Pollution from Municipal
Incineration", the technology supporting municipal incinerator design
has evolved slowly and is yet immature. Indeed, there are few design
parameters which have been well-characterized and proven useful in prac-
tice. Most often, the only measures of design adequacy have been such
pragmatic "variables" as plant availability or the lack of maintenance
and operating headaches. In the areas of combustion efficiency and air
pollution performance, there is almost no guidance for the designer; only
the documented facts of unacceptably high emissions.
1
The estimates presented in Table I indicate the magnitude of the
incinerator air pollution problem both now and in the future. These
emission projections show clearly the important contribution to total
emissions of combustible gaseous and particulate pollutants. Previous
work has shown that the very existence of these pollutants in incinerator
flue gas is a consequence of inadeqaute mixing within the furnace and
flues of the incinerator. Specifically, theoretical and experimental
studies indicate that at the temperature and air-fuel conditions which
are obtainable in well-mixed incinerators, the majority of combustible
pollutants are destroyed within a small fraction of the typical gas and
particle residence times. Thus, if means can be found to assure that com-
plete mixing occurs witiin incinerator furnaces, the emission levels of
most combustible pollutants can almost be eliminated.
The design of most existing municipal incinerators is seriously de-
ficient in provision for overfire mixing. Practical engineering knowledge
of design and operating parameters required to optimize mixing and conse-
quent burnout in incinerator combustion chambers is also lacking. A
study of combustion chamber mixing factors leading to development of de-
sign principles would offer one of the most readily applicable and least
expensive means for upgrading operation and reducing combustible emissions
from municipal incinerators.
1-1
Arthur D Little, Inc
-------
TABLE I
ESTIMATED INCINERATOR EMISSIONS
(thousands of tons per year)
1968 Emissions Estimate 2000 Emissions Estimate
Furnace Stack % Furnace Stack %
Combustible Particulate 38 32 7.1 131 49 3.3
Carbon Monoxide 280 280 62.2 829 829 56.0
Hydrocarbons 22 22 4.9 64 64 4.3
Polynuclear Hydrocarbons 0.01 0.005 0 0.03' 0.0009 0
Subtotal (Combustibles) 340 334 74.2 1,024 942 63.6
Mineral Particulate 90 56 12.4 708 118 8.0
Sulfur Dioxide 32 32 7.1 161 160 10.8
Nitrogen Oxides 26 22 4.9 147 114 7.7
Hydrogen Chloride 8 6 1.3 219 147 9.9
Volatile Metals 0.3 0.3 1.0 0.055 0.025 0
TOTAL 496 450 100.0 2,259 1,481 100.0
Reference 1
1-2
Arthur 0 Little, Inc
-------
B.
OBJECTIVE & SCOPE
Because of the lack of reliable design guidelines for combustible
pollutant control, the Office of Air Programs (OAP) of the Environmental
Protection Agency initiated studies aimed at providing this technology to
the incinerator design industry. This report documents the results of the
first steps in these studies:
.
A review of the literature and other available sources of
information relating fuel bed combustion and combustion chamber
mixing to overall furnace combustion performance and when
available emission characteristics.
.
Development of theoretical and analytical models which describe
basic elements of the combustion or mixing processes within
the refuse incinerator for use as design guidelines.
.
Conduct of laboratory experiments to confirm or amplify the
developed theoretical mixing models.
.
Development of a plan for testing in an existing municipal in-
cinerator the effectiveness of the overair mixing guidelines
established during the program.
1-3
Arthur D Little.lnc
-------
C.
APPROACH
In order to avoid duplication of past work and to incorporate the
experience gained in incinerators and other stoker-fired combustors, a
review of the literature was carried out (Chapter VII). Also, discussions
were held with firms now engaged in the design of incinerator and fossil-
fuel-fired boilers.
After analysis of this information, it was evident that the objec-
tives of the program could best be met by limiting the study to the effects
of overfire air and steam jet mixing and of the bed processes giving rise
to the combustibles. The effects of baffle systems on mixing were not
extensively considered.
The approach in developing design guidelines was to carry out analyses
to:
1.
Suggest, based on coal-burning experience, step-by-step
methods to specify incinerator overfire air mixing system
parameters; and
2.
Analyze pollutant generation and destruction processes and
jet dynamics to contribute to an understanding of the be-
havior of the pertinent components of the furnace system:
the burning bed, the enclosure and its influence on flow and
mixing; and jet b~havior in furnace environments.
Recognizing the limitations of analysis, an experimental evaluation
of operation and emission variables in a state-of-the-art incinerator was
seen as the next logical step in providing the fundamental engineering
data needed to validate the design guidelines. For this purpose, an in-
cinerator of modern design was selected and a test program was prepared
which included, in addition to emission rate characterization, an evalua-
tion of the engineering configuration (flow rates, heat release rates and
so forth) of the unit during the tests.
1-4
Arthur D Little, Inc
-------
D.
SUMMARY
A review of the prior art (documented in the Bibliography--Chapter
VII) in combustion design technology yielded nothing oriented specifically
toward overfire mixing in incineration systems. Overfire air jet correla-
tions were found which had been developed for application to coal-burning
systems. Although these approaches to jet design (Section D of Chapter IV)
could be interpreted as directed at realization of complete overfire com-
bustion, their experimental basis shows them to be directed at2smoothing
the temperature distributions in the gases rising from the bed or the even more
pragmatic goal ~f assuring penetration of the jet flow across the full width
of the furnace. As a consequence, their effectiveness in meeting the ob-
jective of combustible pollutant control in incinerators is unknown.
Design technology for passive (baffle) systems was found to be largely
an empirical art. Consideration of furnace baffles as mixing devices, how-
ever, showed them to be relatively ineffective in generating intense tur-
bulence. Baffles are well suited to meeting specific gas flow control needs
for a given furnace design but they are too inflexible to effect any reliable
level of control of the constantly varying combustion process. ' Since they
have little effect on the needs and design parameters of jet mixing systems,
little effort was expended on this aspect of the design art.
The results of our study of the bed burning process (Chapter II), com-
bustion chamber flow (Chapter III), and jet behavior (Chapter IV) yielded
useful contributions to better understanding of the incineration process
and suggested techniques for design evaluation and equipment specification.
1.
THE BED BURNING PROCESS
In order to provide a tractable analysis problem, the refuse pyrolysis
and ~asification processes were evaluated on an overall basis. Based on
data which suggested the off-gases were approximately in equilibrium accord-
ing to the reaction:
+
C02 + H2 + CO + H20
energy and material balances were ,written
bustible pollutant loading (CO and H2) as
moisture content, and underfire air rate.
the following result:
which allow estimates of the com-
functions of the refuse composition,
Using this model of bed combustion,
a.
Combustible pollutants will ~lways b~ emitted from the bed
into the overfire volume and, except when large fractions
of the undergrate air bypass the bed (channeling), the off-
gases always present an oxygen demand.
1-5
Arthur D Little, Inc
-------
2.
b.
The oxygen demand can be calculated allowing estimation of
the minimum overfire air demand as a function of pertinent
operating and refuse variables.
c.
The refuse pyrolysis mass-addition rate to the undergrate
air flow can be estimated to provide input data to analysis
of the flow in the furnace enclosure.
d.
Calculations of bed behavior for different assumptions re-
garding refuse and underfire air parameters provide estimates
of the flexibility required of jet systems, fans and the like.
FURNACE FLOW BEHAVIOR
Complex heat and mass transport phenomena, compounded by combustion
effects, make rigorous analysis of furnace pattenns impossible
without further simplifying assumptions. In this study, the
bed processes were considered as a generator of an input stream to the
overfire volume. Combustion effects in the gas phase were ignored and
the gas flow was assumed to move, without significant mixing, at velo-
cities corresponding to potential flow in the presence of buoyant forces.
As a consequence of the temperature and mass flow distribution along the
grate, this resulted in the acceleration of the hot gases entering the
enclosure in the pyrolysis region. Methods were developed to relate gas
velocity to the input temperature and mass-flow distributions calculated
from the bed-burning model and to the physical dimensions and configuration
of the furnace enclosure. Methods were also developed to allow simulation
of turbulent mixing to develop estimates of the velocity, temperature, and
composition (degree of mixing) nonuniformity that could be expected at the
discharge plane of the physical system in question.
3.
Using this model of furnace dynamics, the following results:
a.
The velocity field in the incinerator furnace can be
estimated by multi-zone treatment of the system.
b.
Calculated furnace gas velocities can be used in subse-
quent analyses to determine the magnitude of cross flow
effects on overfire air' or steam jet trajectory. .
c.
Calculated velocities can be used to anticipate erosion
problems. This can be of considerable importance if boiler
tube passes are located in the outlet flue of the furnace.
d.
Estimates of the adequacy of alternatives in furnace design
to effect n~eded turbulent mixing without jets can be made.
JET BEHAVIOR
At present, most jet behavioral analyses used in incinerator design
are based on relationships which describe the behavior of jets discharging
1-6
Arthur D Little, Inc
-------
into a fluid at rest under relatively isothermal, non-reacting conditions.
In an incinerator, however, the jet is cold and the ambient is hot, sug-
gesting the importance of buoyancy effects on jet trajectory. Of equal or
greater importance, the jet traverses a cross flowing stream of combustible
gases. Thus, in contrast to the bed and flow analyses, the effort on jets
sought complexity as it was needed to provide reliable predictions of jet
behavior in a complex flow environment.
The effects of crossf1ow and buoyancy, acting alone, appeared adequately
described by existing correlations and experiment. Their combination, how-
ever, in the "competitive" situation where buoyancy tends to drop the jet
trajectory and crossflow to raise it, was not well characterized.
While some workers have indicated that buoyancy effects could be ne-
glected, our analysis showed that under many conditions of incinerator
design and operation, buoyancy could'dominate the flow. We confirmed
the qualitative va11~ity of our analysis by modeling experiments. A
method was developed to determine the conditions where buoyancy dominates
under given furnace environment conditions (calculated using the bed and
flow analyses) to allow avoidance of designs where the jet would sink and
disturb the bed (entraining fly ash or overheating the grates). Also, re-
lationships were developed permitting a rough estimate of the effect of
combustion on jet temperatures as they may effect overheating of surfaces
on which they impinge.
The following summarize the results of the jet behavior analysis:
a.
Correlations were developed tQ, compute jet trajectory
under combined crossf1ow and buoyancy conditions, using
inputs from the bed and flow analyses which support more
confident estimation of jet behavior than in the past.
b.
The effect of buoyancy was shown to be important under
some design and operating conditions and means were pro-
vided to allow evaluation of the trajectory under these
conditions.
c.
A calculational method was developed for rough evaluation
of combustion effects on jet temperature distributions.
4.
DESIGN GUIDELINES
The design guidelines reported or developed here fall into two cate-
gories: a rule-of-thumb method derived from coal-burning technology and an
analytical method to predict the behavior of alternatives in system design.
Two generalized design approaches based on coal-~urning. experience are
presented. The first approach is that of Ivanov who based his method on
the results of experiments in a laboratory flow apparatus and confirming
tests in full-size boilers. His method (Section C-5a and D-4 of Chapter IV)
is directed toward smoothing composition and temperature profiles. To the
extent that adequate destruction of combustible pollutants will occur under
such conditions, the method appears sound.
1-7
Arthur D Little, Inc
-------
A second generalized design approach arises from work done at
Bituminous Coal Research, Inc. for the National Coal Association. 3 This
method (Section C-5b of Chapter IV) appears based on isothermal jet be-
havioral relationships and has the objective of assuring jet penetration.
It does not explicitly address the problem of obtaining mixing or allow
compensation for the effects of cross-flow. With these apparent weaknesses,
however, it should be noted that these correlations have proven successful
in visible smoke abatement in grate bed coal-fired boilers.
The design guidelines based on analysis methods developed under this
program are described above. Their application in a design situation is
described in Chapter V. There, the analysis is shown to provide estimates
of air requirements, burn-out levels, flue gas velocities and temperatures
and so forth to give the designer perspective and quantitative estimates
of system behavior under varying conditions. It should be recognized that
a number of reasonable but largely untested assumptions are required in the
course of the analysis. The engineering data to confirm the assumptions
are lacking but, to an extent, will be sought in the test program to follow.
5.
TEST PROGRAM
The characteristics of the proposed test program are described in
Chapter VI. The tests should be divided into three groups. The first test
series would provide a reference baseline as to the emissions in the ab-
sence of overfire air mixing. The second test series would include experi-
ments to study emission characteristics and chamber flow patterns as they
relate to system variables by measurement of gas compositions, velocities
and temperatures throughout the furnace. The third test series would show
the effects on emission rate of various operating configurations of over-
fire air jet systems. It is reasonable to expect that with a relatively
small number of tests, the effectiveness (if not the reason for the effective-
ness) of the overfire air jet system could be demonstrated.
1-8
Arthur D Little, Inc
-------
E.
REFERENCES
1.
W. R. Niessen et. a1., "Systems Study of Air Pollution From Municipal
Incineration", report to NAPCA under Contract CPA 22-69-23 by Arthur
D. Little, Inc., 3 Volumes (1970)--Available from Clearinghouse for
Federal Scientific and Technical Information, Springfield, Virginia
(PB-192-378, PB-192-379, PB-192-380).
2.
Y. V. Ivanov, "Effective Combustion of Overfire Fuel Gases in Furnaces",
Astonian State Pub. House, Tallen (1959).
3.
"Layout and Application of Overfire Jets for Smoke Control in Coal-
Fired Furnaces", National Coal Association, Washington, D.C., Section
F-3, Fuel_Engineering Data, December 1962.
4.
E. R. Kaiser, personal communication to Walter R. Niessen.
1-9
Arthur D Little, Inc
-------
CHAPTER II
BED BURNING MODELS
Arthur D Little, Inc
-------
"
CHAPTER II
BED BURNING MODELS
A.
INTRODUCTION
The processes occurring in a refuse bed are extraordinarily complex,
involving the interaction of heat transfer, mass transfer, chemical ki-
netics and fluid flow in a heterogeneous material with a chemical compo-
sition and physical configuration which varies in both space and time.
Exact analysis of such processes is clearly out of the question. But the
processes occurring in the fuel bed generate the volatile matter that
must be burned in the overfire regime, and qualitative and rough quantita-
tive models of the fuel bed are needed in order to define the design of
the overfire air supply.
To provide a means to estimate the approximate composition and flow
rate of combustibles into the overfire volume, a method will be presented
to enable calculation of these quanti ties. The analysis draws .from coal
and refuse combustion theory and data.
II-I
Arthur D Little, Inc
-------
B.
PREVIOUS WORK
Guidance for the development of bed burning models is provided by the
experimental studies on the Oceanside Incinerator by E. Kaiserl, on simulated
refuse beds by the Bureau of Mines2, and on coal, coke and lignite beds by
a number of investigators.3,4,5,6
The pertinent results of Kaiser inc1~de measurements of the gas con-
centration at a number of positions above a refuse bed, and calculations,
based on the gas analyses, of the distribution of heat release rate both
within and above the fuel bed.
The schematic elevation and plan views of the Oceanside Incinerator,
Figure 1-1, show the positions at which gas samples were obtained.
The operating conditions of the unit corresponded to a refuse feed
rate of 300 TPD and a total air flow of 150-200 percent excess air, 60
to 80 percent of which was supplied under the grate through three of the
four available windboxes at windbox pressures as shown in Figure 11-1.
Additional observations on the operation of the Oceanside Incinerator by
Kaiser are that:
1.
The ignition plane intersected the grate at approximately
the midpoint of the second windbox from the feed chute for
regular refuse and toward the end of the second windbox for
refuse with 43 percent moisture content.
2.
Approximately half the undergrate air was supplied by the
second windbox, three-eighths in the third, and one-eighth
in the fourth. No forced undergrate air was supplied to the
first windbox.
3.
Very little flame was observed over the fourth windbox.
4.
Approximately 9 percent of the total feed, containing 3 to
4 percent of the combustible, was lost as siftings through
the openings (20 percent of the total area) in the grate.
5.
Two percent of the feed was carried off as fly ash containing
50 percent carbon and 16 percent, containing 4 percent com-
bustible, was discharged off the end of the grate.
A summary of the gas compositions and temperatures reported by Kaiser are
given in Table 11-1. The total carbon in the gases was obtained from the
sum of the carbon contents of the C02' CO, CH4 and of the particulate and
liquid collected in the sample train, with the exception of two runs in
which only the gas contributions were included.
From consideration of the above data and observations, Kaiser computed the
distribution summarized in Figure 11-2 of the rate of energy release with-
in and above the fuel bed. These results by Kaiser are the only subst"antive
figures for incinerators, and merit special consideration.
II-2
Arthur D Little.lnc
-------
I
4'
H
J
K
-0.3" (Draft)
A,B
X
---
---
--- --- C,D
I --- X
~ 11 Feet---...\ ---- ----
------ E F
Air at 1.3" ............... """'-- X .
Air at 0.7" ~
5'
No Undergrade
Air
Elevation
Air at O. 1"
G
~
t
10'
B
D
F
G
A
C
E
Plan
FIGURE 11-1 SCHEMATIC OF OCEANSIDE INCINERATOR SHOWING
LOCATION OF GAS SAMPLE POSITIONS
II-3
Arthur 0 Little.lnc
-------
TABLE 11-1
SUMMARY OF OCEANSIDE DATA 1
% C Burned to Steam
Location COz 0z CO HZ .94 NZ T (OF)* C02 CO Unburned M 1bs/hr Remarks
A 12.53 6.38 0.75 0.45 0.20 79.69 1418 92.90 5.58 1.52 45.0
B 14.30 1.83 4.47 3.38 0.13 75.89 1693 75.61 23.70 0.69 69.2 Gas Only
B 15.37 0.83 6.43 3.07 0.77 73.53 1789 68.07 28.51 3.42 72.0
C 8.40 10.40 1. 73 1.36 0.30 71. 81 1316 76.41 15.54 8.05 40.3
D 13.43 2.47 7.91 4.71 0.41 71. 07 1634 58.32 34.73 6.95 55.3
H
H E 0.17 20.67 0.00 0.00 0.20 78.96 1679 45.79 0.0 54.21 58.1
I
~
F 0.27 20.46 0.00 0.07 0.32 78.88 1695 40.68 0.0 59.32 62.1 Gas Only
G 0.0 20.60 0.00 0.07 0.00 79.33 1510 0.0 0.0 0.0 59.3
H 9.25 10.38 0.0 0.04 0.18 80.15 1287 97.20 0.0 2.80 29.9
H 12. 78' 5.85 0.0 0.01 0.20 81.16 1605 97.98 0.0 2.02 64.8
J 3.53 16.53 0.0 0.14 0.15 79.65 1460 95.28 0.0 4.72 28.7
K 0.95 19.38 0.0 0.02 0.23 79.42 1236 99.75 0.0 20.25 48.0
:t>
..., L 5.20 14.78 0.0 0.02 0.20 79.80 1322 95.15 0.0 4.85 39.1
-
:::r
c:
..., L 5.80 13.83 0.0 0.02 0.15 80.20 1563 97.41 0.0 2.59 62.5
0
C
-
- * Thermocouple near sidewall at J
{~
:J
0
-------
~
<'I,;: 60
-;:-
:: 56
::J
...
CD
"b 52
....
gj 48
...
E' 20
Q)
c:
w
b 16
Q)
...
i:. 12
FIGURE 11-2
72
68
64
Total (In Bed Plus Above Bed)
8
4
4
8
12
16
20
24
28 32
36
40
Distance from Feed Chute (Ft)
44
48
ESTIMATED ENERGY RELEASE RATES FOR OCEANSIDE INCINERATOR 1
II-S
Arthur D Little.lnc
-------
One can make the following further observations:
1.
In the gasification and burning regimes of interest, most
of the carbon in the refuse is oxidized to either C02 or
CO. An apparent exception is at locations E and F where
the composition is close to that of air. This resulted
from the fact that at these grate positions burnout of the
refuse was essentially complete and only trace amounts of
combustion and gasification products were present.
2.
Some of the air does not react within the fuel bed as a
consequence of by-passing the combustible through blow-
holes or through an area on the grate covered by non-
combustibles. From the data in Table II-I, the amount of
unreacted air by-passing can be estimated. The values
shown in Table 11-2 were obtained by assuming that the com-
bustible content of refuse could be approximated by C(H20) .
n
Clearly, the amount of air by-passing the refuse will vary
with refuse type and loading conditions. The limited avail-
able data suggest that on the average about 25 percent of
the air by-passes the fuel.
TABLE II-2
PERCENTAGE OF UNREACTED AIR BY-PASSING
Position
B
D
C
A
B
Percentage
32.8
10
4.2
53
12.4
3.
The products of combustion appear to refl~ct equilibrium of
the water-gas shift reaction: (C02 + H2 + CO + H 0). To
test this postulate, the water vapor content of tfie gases
leaving the bed must be estimated by assuming the HIC ratio
is in the range 1.5 to 2.0, bracketing data for municipal
waste. 7 Table 11-3 pre~ents the values of the water-gas
shift constant and the corresponding equilibrium tempera-
ture for the two postulated H/C ratios. The calculated
temperatures are within the range l50Q-2QQQoF of tempera-
tures commonly found in burning fuel beds. These results
suggest that the water-gas equilibrium is approached within
the fuel bed. This behavior is very similar gO that observed
in the gasification of coal, lignite and wood ,9 for which
equilibrium is also approached when the temperatures are in
the range l50Q-2000°F.
II-6
Arthur D Little, Inc
-------
TABLE 11-3
WATER-GAS-SHIFT EQUILIBRIUM TEMPERATURE
Position A B B e D
PH Peo
Hie = 2.0 2 2 0.595 0.7 0.405 0.73 0.464
PH oP eo
2
T(OF) 1810 1690 2190 1635 2020
PH Peo
Hie = 1.5 2 2 0.813 1.03 0.595 1. 053 0.685
PH oPeo
2
T(OF) 1603 1500 1810 1490 1710
The additional observation made by Kaiser that an ignition wave propagates
through the fuel bed is supported by results on a U.s. Bureau of Mines batch
fed 19-inch I.D. test incinerator. Thermocouples placed within a simulated
refuse bed at 6-inchintervals showed the propagation of a drYing plane and
an ignition pla~e through the bed. Figure 11-3, taken from the U.S. Bureau
of Mines Report, shows the rates of propagation of the evaporation and
burning front. Additional unpublished results provided to ADL by the
U.S.B.M. support the conclusions drawn from Figure 11-3. Unfortunately,
the burning rates obtained in the U.S.B.M. tests were obtained mostly
with little or no underfire and therefore are not directly applicable
to municipal incinerators.
The results by Kaiser and U.S.B.M., though limited, illustrate that
burning refuse beds have characteristics similar to those observed in. the
burning of other solid fuels. The extensive literature on the burning and
gasification of coal in beds is, therefore, very pertinent to the present
study.
II-7
Arthur D Little, Inc
-------
~u
r-r
Travel of evaporation front
.5 15
o
LIJ
III
I.L.
o
Q.
o.
~ 10-
o
0:::
I.L.
LIJ
()
Z
~
a 5
. Source. Reference 2
/ .
FIGURE 11-3
TRAVEL OF EVAPORATION AND BURNING FRONTS AND TOTAL WEIGHT
LOSS AS FUNCTIONS OF TEST TIME IN A TYPICAL BED OF REFUSE HAVING
A MOISTURE CONTENT OF 50 PERCENT AND A BED DEPTH OF 18 INCHES
IN BU MINES (19-lnch Diameter Test Incinerator)
II-8
10
-E.
en
VI
o
~
I-
:x:
C>
Ui
~
5
o
Arthur D Little, Inc
-------
The classical experimental and theoretical studies of the processes
occurring in a burninr fuel bed are those of Kreisinger et. al.,3,5
Nicholls,4 and Mayer. 0 In the discussion to follow, distinctions will
be made between overfeed, underfeed, and channel burning, which roughly
approximate the processes occurring at various positions on a traveling-
grate stoker. As shown schematically in Figure 11-4, air and fuel flow
co currently in an underfeed stoker, and countercurrently in an overfeed
stoker. The former corresponds roughly to the processes occurring at the
front end of a traveling-grate stoker where the ignition plane is traveling
towards the grate and the latter to the char burnout region where the
carbon burns preferentially at the ash char interface. Channel burning
refers to air flowing through a fuel channel and is presented here as an
approximate model of the conditions in a blowhole in a refuse bed. It is
important to note the distinction between the propagation of the ignition
plane, corresponding to the "plane" at which the fuel ignites, and of the
burning plane, corresponding to the plane at which the fuel has been com-
pletely burnt.
Concentration profiles obtained by Nicholls for underfeed operation
with an Illinois Coal (35% volatile) are shown in Figure 11-5. The
oxygen is rapidly consumed above the ignition plane, with C02 as the main
product. The C02 subsequently reacts with the coked coal particles to
form some CO. Small amounts of soot, tar and methane are formed primarily
in a pyrolysis zone near the ignition plane. The concentration of the
products correspond to a water-gas-equilibrium temperature of about l800oF,
but is subject to uncertainty resulting from probable errors in the estimates
of the low H20 concentrations. The results are in agreement with the con-
clusions drawn from Kaiser's results on the Oceanside Incinerator that
most of the carbon is converted to C02 or CO and that the water-gas reaction
is close to equilibrium. For the runs on the Illinois coal approximately
one-third of the heat of combustion is released within the bed and two-
thirds above the bed, in contrast to Kaiser's estimates shown in Figure
11-2 which suggest that approximately two thirds of the heat is released
in the bed of the Oceanside Incinerator.
Additional information of value derived from Nicholls' results is the
effect of underfire air rate on the rates of ignition and burning. The
values for a high-temperature coke are shown in Figure 11-6. At low air
velocities, the rate of ignition. is greater than that of burning, and,
therefore, the depth of the burning fuel zone increases with time until
the ignition plane intersects the grate. The rate of ignition imposes an
upper limit on the rate of burning. As the air velocity is increased,
the burning rate increases faster than the ignition wave up to the point
that the two velocities become equal. The change in structure of a coal-
burning bed6as a consequence of the above trends has been shown by Marksell
and Miller. Figure 11-7, taken from their paper, shows the volatile and
ash content of six slices within their bed with the bottom slice numbered 1.
As the ignition wave propagates through the bed, the volatile content fall
from the original value of 36 percent to zero with the topmost layers being
devolatized first (Figures II-7a and 7c). At low air rates, the burning
rate is relatively slow and the residual char combustion is not completed
,
II-9
Arthur D Little, Inc
-------
t
Overfeed Fuel
Ash
~
o
Ignition Plane
Burning Fuel Bed
Ignition Plane
Burning Fuel Bed
Fuel
##
##
##
##
##
##
t
Air
t
Air
~
(a) Underfeed
(b) Overfeed
Ignition Plane
sh .......
-.--.--.-- -- ------.--
# t t t t t t I
.... Air Air Air AiL Air Air #
~.-_.__._-_._._._-------
.....
Fuel
(e) Traveling Grate
Fuel
Fuel
Fuel
t
Air Air
(d) Channel-Burning
t
FIGURE 11-4
SCHEMATIC DIAGRAMS OF BED-BURNING MODELS
11-10
Arthur D Little, Inc
-------
0:: t' 0 0 0' ;
~ E ~"~. ~.
.02
.01
-g 0
'"
o
Il. .02
...
.01
o
o
....
o
"0
c:
'"
o
Il.
~ .12
c.
...
~
C>
'0;
;: .10 -
.08
.06
.04
.02
o
-2
FIGURE 11-5
'n j' 0 'j
-1 - i
~ ":'":0 '
o
c
:
Soot
.
. .
I
, ~ Tar
'~ "o-.o--o~
CH4
8
I C in CO - i- A
. 6. --- I
i ~ ~./"~;- A A:
: g ~/ !
I I !~
c: I' , CD
~ / I~
f',' ,"~ i '\' ~
c:' I-
&:. '4 :
! , - /...o.,,"'().. 0 C in C02 0 I
i I i ~-o --- ------(T~-g- --
! :' . . . Free H2
I ---- - - --.-8
, A
.
o
6
8
10
12
2
4
Height above Plane of Ignition, Inches
UNDERFEED BURNING, ACTION THROUGH COAL BED
EXPRESSED AS WEIGHT OF FUEL PRODUCTS CARRIED PER
POUND OF DRY AIR SUPPLIED, 3/4. 1 INCH -ILLINOIS COAL
II-ll
Arthur D Little, Inc
-------
lo1
f~
~)
8
A\r~ tLFi~lLJ~CY. r;~:":'~:'. .:\r
!.() 2r> 0
--mLL! _/ -.I j L~
. .. I J I'
-.-- --1'- tl: - --- /-
--" I
-J.Z'. - --.~. .-.. .-[;..-1-. L - r::-
,.. / (~
~' [J
: . .\ I r;o~. 0( ~::(i-'-I ~
t.; _..~. to, -_...~ e
F / 1~' 1\ /-<
:.:. -. -- 7 -r-q- g<
II .. 1.\0]1." . l'~d
~~"~-l~---l'--''''>.o..).~.-=- ~~l-' . tj
: I //J,.: ~ I.
p{ :;;- '-1-7 --/N ~17-. \1 .
[; ".. r-ttti I L ".- --- /
~~ L '. ~'. : I I!~ I v
. I ""'Ir.. I I J " 1
r~ , ..~~~;;~.. ".
.- - '1'-' "" - '.... - '<~..
~ -- I - -.i..l1.' J tl-;j - )\.
\..:: / t:j, . l~ tAl -:--/ .
fw . I ',' - I '-' "
~.'l.l f~1-.t7~'I)~rf-- -
!~ - -'"--/-~---I-' ._- -'-
D ;11' ~to'2r'_\___,_.- \..-'
0-'. \, J
o .
u - _7 - - --\1'
} : /
'" ... '. 'f~.r =. - -~C=~OU1rl1l"t if,I)i:c .
tf.~~~= -~~.:-_--l:-. .
CI )(~, :::,.J :i'<) ..;,) :,,'J ~)
PHil!..;';:Y j,1i.. PEP. SQ. VI', pr:n !lOUR, POUIW~
I
'.1..
,,:.,..
(0
FIGURE 11-6 UNDERFEED BURNING, HIGH.TEMPERATURE COKE; RATE OF
IGNITION AND RATE OF BURNING WITH RATE OR PRIMARY
AIR AND SIZE OF COKE AS VARIABLES
11-12
Arthur D Little, Inc
-------
Yo VOl81,le\
20
:0
a
4r. M,nulet
,8 '9 10
o
"/. II,,,
60
60
40
20
'0
",,2 '3 -4 ,5 .6
b
,7 . ,6 .9 10
Fraction of Combustion Time
FIGURE 11-7
. .1.~18Ii'e5
100 'YoA~"
30
20
c
.10
o
,9 10
, ~
23 Minute.
60
60
I d'
<40
20
o
.2 '3 -4 ,5 -6
,7 -8 .g 1:0
Fraction of Combustion Time
VOLATILE MATTER AND ASH CONTENTS
, OF COAL BED LAYERS, LOW AIR RATE
II-l~
Arthur 0 Little.lnc
-------
until the ignition wave reaches the grate, at which stage the bed burns
in an overfeed mode with the ash content buIlding up to 100 percent in
the bottom layer first and the top layer last (Figure II-7b). At high
air rates, the rates of ignition and burning are equal and devolatization
and ashing take place simultaneously (Figures II-7c and 7d). The iso-
volatile and iso-ash counters for the conditions of Figure 11-7 are
presented in Figure 11-8. Plots (a) and (b) show devolatilization pre-
ceding burnout; (c) and (d) show simultaneous devolatilization and burn-
out.
The high air rate case has been presented here for completeness.
For refuse, excessive particle entrainment from the fuel bed is expected
at velocities lower than that at which the propagation of the ignition
wave becomes limiting. In such cases, the rate of burning within the
bed is expected to be proportional to the underfire air supply.
Studies5 on lignite with high moisture (40%) and low fixed carbon
content (25%) showed results very similar to those for coal. It is ex-
pected, therefore, that the same principles apply in the different fuel
beds and that consequently the above conclusions can be used to develop
qualitative and quantitative models for refuse beds.
II-14
Arthur D Little, Inc
-------
8
6
~ 4
c:
- 2
o
o
~
8
9 10
~
2
3 4 5 6
45 Minutes
7
(a) Fuel Bed lso-Vols
8
6
X; 4
.t:
U
.:
2
o
o
2
3 4
5 6
7
8
9 10
Fraction of Combustion Time
(b) Fuel Bed Iso-Ash-Contents
8
6
:G 4
.t:
U
.: 2
o
o
..
3 4 5 6
23 Minutes
7 8
10
.-
9
2
(c) Fuel Bed Iso-Vols
8
6
II>
'" 4
.t:
U
.:
2
o
o
2
8 9
3
4
5
6
7
10
Fraction of Combustion Time
(d) Fuel Bed lso-Ash-Contents
FIGURE 11-8
ISO.VOL. AND ISO-ASH CONTENTS THROUGHOUT COAL BED (a)
AND (bl.lOW AIR RATE, (c) AND (d) AT HIGH AIR RATE
II-15
Arthur D Little, Inc
-------
C.
BED COMBUSTION MODEL
A qualitative model of a refuse bed is shown in Figure 11-9. Drying
and ignition waves propagate through the refuse, followed by a devolatiza-
tion zone. The gases from the pyrolysi~ section pass through the char bed from the
top of which a mixture of mainly low-molecular weight gases emerge with
small amounts of soot and tars. The char that is formed follows an over-
feed burning mechanism. Order of magnitude estimates of the dimensions
of the different zones may be obtained from the discussions in the pre-
ceding sections. The rate of propagation of the drying and ignition waves
is determined by the rate at which energy is transferred ahead of the propa-
gating front. The rate has been shown to be a function of under grate
air supply (Figure 11-6), particlz size (Figure 11-6), air preheat,4
moisture content,2 and fuel type. For refuse, the rate estimated by
Kaiser for tests on the Oceanside Incinerator vary from 0.3 feet per minute
for wet refuse to 0.5 feet per minute for average refuse. The U.S.B.M.
tests indicate that the distance of separation of the ignition and drying
wave is of the order to 0.5 feet, but their ignition rates measured with
little underfire air were lower than those observed by Kaiser.l
The total burning rate will be determined by the rate of supply of
undergrate air. These can be calculated by drawing on the results from
pyrolysis studies 11, showing that the amount of char produced is ex-
pected to be in the range of 0.10 to 0.2 pounds per pound of refuse. The
oxygen requirement for the 0.9 to 0.8 pounds of refuse gasified will be
determined by the water-gas shift and enthalpy requirements as will be
discussed below. Once the refuse has been completely devolatized, the
rate of burnout of the char will be determined by the rate of oxygen
supply, with the combustion first yielding C02' which then reacts with
more carbon to yield CO. The thickness of the regimes of thermal pyrolysis
and of burning char observed in Nicholls' experiments on coal are of the
order of a few inches. For refuse, these will be significantly larger,
since the larger refuse elements will present major diffusional resis-
tances to heat transfer for the pyrolysis reactions and to oxygen and CO
transfer in the char burning regions. Determinations of the depth of 2
these layers would require much more detailed modeling of the kinetic
processes occurring within a bed. The development of such models has been
pioneered by MayerslO who was able to predict with remarkable accuracy
the gas concentration profiles measured by Nicholls in coke beds. A
modification of the Mayers' model which is simpler to apply has been
proposed by Stewart13, but both Mayers and Stewart do not make allowance
in their models for the diffusional resistance within pyrolyzing and burning
particles.
The above sections indicate that the gross behavior of burning fuel
beds are very strongly tied to the air distribution below the bed. In-
crease in air rates results in proportional increases in overall gasifica-
tion or char-burning rates. These increased air rates often will result
in increases in the widths of the gasification (or volatization) and
char-burning zones. In the present study, the thickness of the differ-
ent zones is of secondary importance and, therefore, emphasis will be
focused on the gasification rates.
II-16
Arthur D Little, Inc
-------
Bed
Height
H
H
I
......
"-J
»
~
=r
c
..,
u
c
...
...
{t
::I
o
\
\
\
RAW
~
GRATE LINE
~mission
CO Emission
FLUE GAS COMPOSITION
Top of Refuse Bed
.L
\
\
,
"
"
"
REGION OF NON-BURNING CHAR v~~",,"
~CJ "
~~ "
~o~ " "
d:- "
~o "
C:>~ "
o~ /
o /
~ /
"" ".
~"'/
/
"
\
\
\
\
~\
~ \
~\
~ \
~, -----
~ \-
~
l1ll1wlNC
REGION OF THERMAL
PYROLYSIS
(VOLATILIZATION)
~~
v
~~
~~
~~
o
-------
A quantitative model will be developed first for complete gasifica-
tion) corresponding) in terms of Nicholls' experiments) to the regime
where the rates of ignition and burning are equal. For purposes of
simplifying the calculations, the organic content of refuse will be
modeled by cellulose [C6(H20)] . A synthetic refuse with 22 percent
inert, 25 percent moisture) an~ ~3 percent cellulose would have proper-
ties very similar to an average municipal refuse with a higher heat co~
bustion of 4300 Btu/lb. .
The gasification reactions occurring within the bed produce CO) H2'
H20, C02) and small amounts of CH4' tar, and soot from the refuse. As a
good approximation) as may be seen from both Kaiser's and Nicholls' results)
the CH4' tar and soot may be neglected. The gasification products can then
be calculated from consideration of the stoichiometric relationship be-
tween products and reactants) the water-gas shift equilibrium) and an overall
energy balance. For purposes of computation ease) the composition of the
inert-free content of moist refuse can be simulated by C(H20) with n having
a value of (5/6) for a dry cellulose) 1.55 for a refuse with 23% inerts and
25% moisture, and 2.0 for a refuse with 23% inerts and 34% moisture. Desig-
nating the number of units of C(H20)n gasified by one mole of oxygen by..x,
one can formulate the gasification reaction as follows:
02 + j C(H20)n + 3.76 N2 + aCO + p C02 + i HZ + d H20 + e C + 3.76 N2
(11-1)
Equation (II-I) contains six unknowns; "e") the amount of char residue
produced per mole of oxygen during gasification) has been shown in the
discussions above to be a function of the rates of ignition and burning.
At high air rates) when the burning and ignition rates are equal, e is
equal to zero. At lower air rates) the value of e will depend on the rate
of gasification, but it has an upper limit imposed by the energy requirements
of the overall gasification reaction. In the numerical illustrations to
follow) the values of 1 and 0.5 will be assumed as the fiaction of carbon
gasified (equal to the quantity [l-e/~) to cover cases of high and inter-
mediate air rates.
With the relationship between e and x assumed) the residual unknowns
in equation (II-I) number five) and solutions can be obtained from con-
sideration of carbon, hydrogen, and oxygen balances) the water-gas shift
equilibrium, and an overall energy balance. The element balances yield:
Carbon a+p+e = j (11-2)
Hydrogen i + d = nx (II-3)
Oxygen p + (a + d)/2 = 1 + nj/2 (II -4 )
II-18
Arthur D Little, Inc
-------
For the water-gas equilibrium,
[C02] [H2] - (p)(i) -
1H20](COf - (d) (a) - kw
(II-5)
The basis for the energy balance is the requirement that the heat generated
by the gasification reactions provide the sensible energy of the gasification
products and the heat loss by the section of the fuel bed in question to
other parts of the fuel bed or to the furnace enclosure and grate. The amount
of energy release QR,in Btu per pound mole of oxygen, is given by the differ-
ence of the heats or formation of the products and reactants:
QR = 47,560a + l69,290p + 104,240d - 320,940(t) - 104,240 (n-5/6)j
(II-6)
The first three terms represent the CO, C02' and H20 contributions; the
fourth term, the heat of formation of cellulose; "and the last term is a sub-
tractive term for the moisture content of the refuse exclusive of the (5/6)
mole per carbon atom that is chemically bound in the cellulose. The heat
requirements are those for the vaporization of the (n-5/6)x moles of water
vapor and the heating of all the gasification products to the temperature
at which the gases leave the fuel bed. Although the temperature of the leav-
ing gases depends on the conditions within the bed, the variation of tempera-
ture with conditions is not large, and will be neglected here. A tempera-
ture of 2000°F is selected since this is representative of Nicholls' results
on coke and coal beds and is also consistent with the water-gas-shift
equilibrium temperatures calculated from Kaiser's concentration measure-
ments above the fuel bed in the Oceanside Incinerator. The average heat
capacity in Btu/(mole) (OF) between 60°F and 2000°F for the gasification pro-
ducts are 13.8 for C02' 11.0 for H 0, 8.3 for CO, 7.6 for H2' and 8.2 for
N2. The energy convected out of t5e bed by the gasification products is
tfien given by:
Qp = (8.3a + l3.8p + 7.6i + lId + 30.8)(1940) + (18;700)(n-5/6)j
(II-7)
where the units of Q are Btu per pound mole of oxygen
tion reaction. If the rate of heat loss from the bed,
of oxygen, is QL'
used in the gasifica-
again in Btu per mole
QR = Qp + QL
(II-8)
II-19
Arthur D Little, Inc
-------
D.
CALCULATIONS USING BED COMBUSTION MODEL
Equations (11-2) through (11-8) provide the relations necessary for the
calculation of the composition of the gasification products. These are pre-
sented in Table 11-4 for a limited number of combinations of moisture con-
tent, heat loss or gain by the bed, and fraction of carbon gasified. Case 1
is that of the complete gasification of a dry refuse with no net heat loss
or gain. For every mole of oxygen introduced with the undergrate air, 2.22
moles CO, 0.62 moles C02' 1.47 moles H2' and 0.91 moles of H20 are generated
within the bed. The moles of C(H20)5 gasified per mole of oxygen equal
2.85. For the dry refuse, the combusf~on products are rich in CO and hydrogen
as shown in the first column of Table 11-4. The energy release Q above
u
the bed can be calculated from the energies released on the completion of
combustion of the carbon monoxide and hydrogen. In Btu per mole of oxygen,
it is given by:
Q = 121,730a + 104,240i
u
For the conditions of Case 1, it is predicted that 74 percent of the energy
will be released above the bed.
Comparison of Cases 1 and 2 in Table 11-4 show the effect of increasing
the moisture content of the simulated refuse from 0 to 33 percent. The
amount gasified is reduced to 58 percent of its previous value; the combustible
content of the gases generated is greatly reduced; and the predicted per-
centage of heat released above the bed is reduced to 44 percent from 74 per-
cent. The explanation for these trends is evident when gasification is
thought to consist of the following sequential steps:
02 + 3.76 N2 + C
+ C02 + 3.76 N2' Q = 169,290 Btu/mole
02 + 3.76 N2 + C(H20)n + C02 + nH20 + 3.76 N2' Q = 54,950 - 22,200 (n-5/6)
C(H20)n
+ C + nH20, Q = 8,600 - 22,200 (n-5/6)
H20 + C
+ CO.+ H2' Q = -36,620
C02 + C
+ 2CO, Q = -44,200
The first step is the exothermic oxidation of carbon or cellulose to C02'
as shown in the first two reactions. The heat evolved in these steps
provides the energy for the endothermic C02-C and H O-C reations. The
energy release given after each reaction includes t~at required to heat
the gases up to 2000°F. The decomposition of refuse to char shown in the
third reaction may be exothermic or endothermic depending on the moisture
content. For moist refuse the excess energy generated by the first three
reactions is reduced and therefore the last two gasification reactions can-
II-20
Arthur 0 Little, Inc
-------
I -~-~
TABLE II-4
GASIFICATION PRODUCTS AS A FUNCTION OF MOISTURE CONTENTt
FRACTION OF CARBON GASIFIEDt AND HEAT LOSS
Case 1 Case 2 Case 3 Case 4 Case 5 Case 6
A. ASSUMED CONDITIONS
Moisturet moles per C atom
:: (n - 5/6) ° 1.0 1/6 1/6 1/6 1/6
Fraction of carbon gasified
:: 1-e/j 1.0 1.0 0.5 0.5 0.5 0.5
Heat Gain (Loss) Btu/mo1e
02 :: QL 0 0 182tOOO (21tOOO)(56,000)(94,000)
B. CALCULATED VALUES
Mole CO/mole 02 :: a 2.22 0.67 5.0 1.49 0.82 0.13
Mole C02/mo1e 02 :: p 0.62 0.98 2.1 1.26 1.12 1.00
Mole H2/mo1e 02 :: i 1. 47 0.63 7.3 2.01 1.05 0.13
Mole H20/mole 02 :: d 0.91 2.39 7.0 3.49 2.82 2.13
Mole C(H20)n gasified/mole 02 :: x 2.85 1.65 14.3 5.5 3.88 2.27
Gas Composition, Dry Basis,
%CO 27.5 11.1 27.8 17.5 12.2 2.6
%C02 7.7 16.2 11. 6 14.7 16.6 19.8
%H2 18.2 10.4 39.9 23.6 15.5 2.6
%N2 46.6 62.3 20.7 44.1 55.7 75.0
% of energy released above bed 74.0 44.0 81.0 60.0 46.0 11.0
II-21
Arthur D Little, Inc
-------
not proceed to any significant extent. The importance of supplying energy
for the gasification reactions is underlined by the results for cases 3
to 6, calculated for a postulated fixed low moisture content (9%) and a
fixed percentage of carbon gasified (50%), but for a varying rate of
energy addition or loss. When energy is added to the bed (Case 3), for
example, by intense radiation from an overhead flame in the early portion
of the grate (before an insulating layer builds up above the gasification
zone), the refuse gasified per mole of oxygen and the combustible content
of the product gases increase. When energy is withdrawn, (Cases 4 to 6)
for example, by radiation to cooled walls (as in a boiler) or to the
grate, these quantities decrease.
The gas compositions and the amount of energy released above the bed
cover a wide range which encompasses the results obtained by Kaiserl and
summarized in Table 1. Good matches with the measured values could be
obtained by suitable adjustment of the moisture content and energy loss.
For Cases 3 to 6 in which part of the carbon in the refuse is left as a
char during gasification, the gasification reactions will be followed by
a carbon burnout zone. The rates of reaction in this zone are again
limited by the rate of oxygen supply. The product gases are mostly CO,
sometimes with a little C02. The composition of the gases leaving the bed
is often determined by the C02 + C ~ 2CO equilibrium at a temperature close
to 2000°F and is predominantly CO. The total length of the char burnout
section may then be calculated from the length required to supply oxygen
to convert the carbon residue, mostly to co.
The above analyses are based on the assumption that the gasification
and burnout processes are limited by the rate at which oxygen diffuses
to the reactants. In refuse, however, objects that are oversized or
that have a very high moisture content, e.g., telephone books or melons,
will react at a much slower rate than predicted from the rate of oxygen
supply. For the watermelon case, the water will vaporize at a rate de-
termined by the heat transfer rate to the watermelon until the moisture
content is significantly reduced. For the telephone books, long times
are needed for the pyrolysis and ignition waves to propagate to the
center. Either case will lead to combustible evolution in zones where
most other elements are completely reacted. The prediction of the
occurence of these isolated sources of combustibles requires a statistical
analysis of the distribution of such items in refuse. The design of the
overfire air system should make allowance for the presence of these items.
II-22
Arthur D Little, Inc
-------
E.
PRACTICAL IMPLICATIONS
The model developed in the preceding section, although admittedly
approximate, has the capabilities of generating much information of prac-
tical significance.
The first major conclusion is that the refuse bed behaves as a gasi-
fier and that therefore air must be provided above the grate, in amounts
that will depend on fuel conposition, undergrate air flow and the rate of
heat loss. Most of the combustibles above the bed are expected to be CO
and HZ but some tars, soot and CH4 will also be present.
The analysis further shows how the underfire air requirements increase
with increasing moisture content, a characteristic well known to incinerator
operators. (It should be here emphasized that the increased air rate is
to support an increased energy release rate to vaporize water and not to
act as a moisture carrier.) The amount of air required above the bed
correspondingly decreases with increases in moisture content. The pro-
nounced effect of cooling on the rate of gasification has significance in
the start-up period when refractory walls are cold or for waterwa1l units.
Although some of the above conclusions were known to incinerator
operators, they had not previously been known quantitatively. It is felt
that it will now be possible to anticipate the effect of changes in both
refuse composition and incinerator design on the underfire and overfire
air requirements.
This information and the aerodynamic models described in Chapter III
provide the basis for the design and regulation of incinerator air-supply
systems.
II-Z3
Arthur D Little, Inc
-------
F.
SUMMARY
One of the most important characteristics of continuous-feed incinera-
tion systems is the combustion process occurring in the refuse bed. This
region is the source of combustible pollutants. A description of its be-
havior as a function of operating conditions and refuse composition thus
provides the designer with a powerful tool in the selection and placement
of overfire air jets. To provide this tool, an overall pyrolysis-combustion
model has been developed which avoids consideration of the diversity in
refuse shape and composition by using data which suggests approximate equi-
librium is realized in the off-gases for the reaction:
-+
CO + HZO + COZ + HZ.
Using this assumption and energy and material balances, a set of
simultaneous equations are developed which allow calculation of the off-
gas composition as a function of refuse composition and undergrate air
flow. Perturbation of the independent variables gives the designer per-
spective as to the range of gas compositions, velocities and temperatures
which can be expected. Also, information on expected refuse burnout and
heat release distributions can be developed.
II-Z4
Arthur D Little Inc
-------
10.
G.
REFERENCES
1.
E. R. Kaiser, Personal Communication to W. R. Niessen, C. M. Mohr,
and A. F. Sarofim, 1970.
2.
M. Weintraub, A. A. Orning, and C. H. Schwartz, "Experimental Studies
of Incineration in a Cylindrical Combustion Chamber", u.S. Bureau
of Mines, RI6908, 1967.
3.
H. Kreisinger, C. E. Augustine, and F. K. Ovitz, "Combustion of Coal
and Design of Furnaces", U.S. Bureau of Mines, Bulletin 135, 1917.
4.
P. Nicholls, "Underfeed Combustion, Effect of Preheat, and Distribu-
tion of Ash in Fuel Beds", U.S. Bureau of Mines, Bulletin 378, 1934.
5.
H. Kreisinger, C. E. Augustine, and W. C. Harpster, "Combustion Ex-
periments with North Dakota Lignite", U.s. Bureau of Mines, Technical
Paper 207, 1919.
6.
W. C. Marskell and J. M. Miller, "Mode of Combustion of Coal on a
Chain Grate Stoker", Fuel, 25 (1),4-12 (1946).
7.
W. R. Niessen et. a1., "Systems Study of Air Pollution From Municipal
Incineration", Report to NAPCA under Contract CPA 22-69-23, Arthur
D. Little, Inc., Cambridge, Mass. (1970).
8.
w. Gumz, "Gas Producers and Blast Furnaces", John Wiley and Sons,
New York, 1950.
9.
C. G. Von Fredersdorff and M. A. Elliot, "Coal Gasification", Chapter 20
in H. H. Lowry's "Chemistry of Coal Utilization". Supplemental Volume,
John Wiley and Sons, New York, 19
M. A. Mayers, "Temperature and Combustion Rates in Fuel Beds", Trans
A.S.M.E., 59, 279-288 (1937).
11.
D. A. Hoffman and R. A. Fitz, "Batch Retort Pyrolysis of Solid Munici-
pal Wastes", Environmental Science and Technology, 1, 11, November
1968, pp. 1023-1026.
12.
1. M. Stewart and J. Saville, "A Simplified Heat Transfer Equation
for Sinter Beds", Mechanical and Chemical Engineering Transations,
the Institution of Engineers, Australia, November 1968, pp. 135-143.
II-25
Arthur D Little, Inc
-------
CHAPTER III
FURNACE FLUID FLOW
Arthur D Little.lnc
-------
CHAPTER I I I
FURNACE FLUID FLOW
A.
INTRODUCTION
The flow of fluid in an incinerator furnace may be thought to
originate in the vicinity of the grates. Typically, under fire air is
blown up through the grates by fans, passes through the refuse bed and
is heated as it engages in the pyrolysis and combustion processes
occurring within the bed. The gas mixture passes upward and into the
volume above the grates where, in many cases, it is further mixed with
air injected over the bed, i.e., over fire air. The hot gases pass into
the main volume of the furnace where further mixing and combustion
occur. Finally, the gases pass out of the furnace and into downstream
units where, in some installations, heat is extracted by an array of
boiler tubes, or the gas may simply be quenched and cleaned by a
scrubber or other gas cleanup equipment before passing into the stack.
The still-warm gas in the stack, being lighter than ambient air,
creates a draft that supplies some of the motive force for moving gases
through the furnace. The pressure in the furnace is normally main-
tained at somewhat below atmospheric so that flow through cracks and
openings in the walls will be air leakage into the furance rather than
hot gas leakage out of the furnace. In many incinerators, such air
inleakage represents a considerable portion of the total air flowing
out through the stack. For the furnace to operate below atmospheric
pressure, either stack draft or induced draft fans are needed.
to provide the motive force for exhausting the gases. Thus, the amount
of underfire or overfire air that can be blown into the furnace by
fans, while still maintaining the furnace pressure below ambient pressure,
can be limited by the exhaust system.
In the following sections, we will focus our attention on the flow
from the burning refuse bed through the high-temperature volume of the
furnace where, ideally, combustion should be completed. The fluid flow
in this region produces mixing and combustion that may have a strong
influence on the level of combustible pollutants produced by the incinera-
tor. The nature of the flow in the furnace will be determined by the
interaction of pressure, gravity, viscous and inertia forces on the fluid.
Combustion above the bed will tend to raise the gas temperature, lower
its density, and increase the volume flow. It will also create significant
density gradients which, through buoyant effects, can exert considerable
influence on the flow fluid.
The flow field can be roughly described by the spacial distributions
of velocity (u), temperature (T), density (p), and viscosity (~), and by
one or more characteristic dimensions of the furnace [e.g., diameter (D),
height(Z), or length (L)]. Local gas temperatures range from outside air
III-I
Arthur D Little, Inc.
-------
temperatures to 3500°F; velocities from 1 to 40 feet per second; and
characteristic dimensions from 2 to 50 feet. The relative importance
of the various forces in affecting the flow field is indicated by the
usual non-dimensional parameters. The Reynold's number, given by
N = P u L
R ~
is a measure of the ratio of inertia to viscous forces in the fluid.
The Froude number, given by
2
u
NFr = gZ
(where "g" is the acceleration due to gravity) is a measure of inertia
to gravity forces in the fluid. The Grashof number, given by
L3 2 0
N = P g. 6T
Gr 2
~
where 0 is the temperature coefficient of expansion) is a combined measure
of the ratio of buoyant to viscous forces times the ratio of inertia to
viscous forces. Considering the ranges of velocities, temperature and
characteristic lengths noted above and assuming that the properties of
the gases are similar to those of air, we can expect the Reynold's number
will be in the range 104 to 105, the Froude number in the range 0.1 to
3 and the Grashof number in the range 109 to 1011. Clearly, the gross
flow field will be turbulent, and turbulent mixing will dominate the viscous
effects. The effects of gravity, i.e., buoyant forces, will also be im-
portant in determining certain aspects of the flow field.
The fluid flow in the furnace, accompanied by combustion, is an ex-
tremely complex process. Analytical determination of the detailed nature
of the flow is beyond the present state of the art. Moreover, only very
limited data on furnace flow are available. Therefore, we have directed
our efforts to obtaining a rough description of the flow field, including
such information as levels of velocities, approximate temperature distribu-
tion, approximate streamline patterns and estimates of degree of mixing
between fuel-rich and air-rich portions of the flow before the gases exit
from the furnace. The concepts developed provide some certain insights
about operation and performance.
111-2
Arthur D Little, Inc
-------
B.
PREVIOUS WORK
As noted above, the furnace flow is turbulent, and buoyant effects
are important. A considerable amount of background has been developed
in flows where buoyant forces alone cause fluid motion, i.e., in free-
convection flows.l,2,3 Of necessity, most analytical work on free-
convection deals with laminar flow. Various configurations have been
analyzed, and in all but the very simplest of cases, numerical solution
of the differential equations via high-speed comuuter is required.
Turbulent flows are less amenable to rigorous analysis because the re-
lationship between shear stresses in the fluid and other flow parameters
can only be approximated. Prandtl's mixing-length theory for turbulent
flows is perhaps the most widely used of the approximations.4 It has
been applied with considerable success to flows in which a characteristic
dimension that represents the scale of turbulence can be readily identified,
such as the flow of a turbulent jet, where the width of the jet is a
good measure of the scale of turbulent eddies. It is doubtful that the
flow in a furnace can be characterized throughout by a single charac-
teristic dimension for turbulence; nor can the distribution of eddy size
throughout the furnace flow be readily determined. The presence of
density gradients, buoyant effects, convective heat transfer and com-
bustion complicate the picture further.
It is perhaps not surprising that our literature survey (Chapter VII)
has not revealed any attempts to make a detailed analysis of the tur-
bulent flow in furnaces. Most of the work on turbulent free-convection
has been experimental in nature and directed toward obtaining useful
correlations for heat and mass transfer for relatively simple geometries,
i.e.! flow over a flat plate, around cylinders, between f~at plates,
etc. ,2 Most analytical treatments of furnace dynamics5, have been
limited to open-hearth furnaces or comparable systems where the charac-
teristics of the flow are dominated by the action of jets, i.e., pressure,
inertia and viscous forces are controlling. Most physical modeling studies
are also carried out with flows that are primarily governed by these forces.
Almost none of these studies have dealt with the effects of buoyancy or
combustion heat release on the flow.
The long-time interest in gas flow in furnaces had led to a number
of highly simplified treatments. In a very old work, Groume-Grjimail07
likened the buoyant flow of hot gases in furnaces to an inverted flow of
water in open channels. He presented design equations for configurating
reverberatory furnaces that were derived from analysis of the flow of water
over weirs. HarrisS investigated the validity of Groume-Grjimailo's
equation experimentally in a small apparatus, and found that the form of
the equation was correct but that the constants were somewhat in error.
These old works clearly demonstrate the importance of buoyant effects in
furnace flows and draw attention to the commonly observed phenomenon
of the hotter gases rising to the upper section of the furnace and flowing
along the upper surfaces.
III-3
Arthur D Little, Inc
-------
Experimental measurements of velocity distribution at the furnace
outlets have been made for two incinerators as part of an evaluation of
their emission characteristics.9 The measurements in one of the incinera-
tors (at Newton, Massachusetts) show evidence that even near the furnace
outlet, hotter gases tend to flow at higher velocities in the upper sec-
tions of the furnace.
Gas velocity measurements at several sections in a grate-rotary
kiln incinerator 10 show that gas velocities are not everywhere uniform.
Near the furnace outlet, velocities in the upper part of the duct are
nearly three times those near the bottom.
III-4
Arthur D Little, Inc
-------
C.
ANALYSIS OF FURNACE FLOW
1.
OVERFIRE REGION
We begin by considering the flow in a continuous feed incinerator
furnace. It is clear that the hottest gases are produced near the input
end of the burning grate, while the gases at the discharge end are cooler
because of the higher percent excess air. If, as is frequently the case,
the flow of overfire gases is in the same direction as the fuel bed, the
gases will stratify with the hot, fuel-rich components, forming a zone
near the roof of the furnace. The hot gases will be accelerated as they
flow upward in the pressure field produced by the denser, cool gases.
The nature of the flow field out through the furnace will depend on the
furnace configuration. Figure 111-1 shows one common configuration. There,
the flow will be turned as it approaches the roof, and will enter the
breeching area of the furnace with a predominantly horizontal direction.
In some incinerators, the flow from the furnace is directed upward through
an array of boiler tubes.
Because of the relative uniformity of refuse and air distribution
across the grate and the fact that often the width of the furnace is rela-
tively small compared to vertical and horizontal dimensions, it seems
reasonable to consider the flow field as being two-dimensional. Referring
again to Figure III-I, we will imagine that, in the region above the bed
where vertical flow and turning into the horizontal direction occurs,
buoyant effects are most important and that viscous and heat transfer
effects are secondary. Considering this flow to be that of an inviscid,
non-conducting fluid without eddies, in which fluid mixing and heat
transfer effects are not present, should yield useful information on its
behavior. For a flow of this nature, a Bernoulli-type equation is applicable
along each streamline, i.e., the total head, H, is constant along a stream-
line. H is defined by:
Pg 2
H=---.£+~+z
p 2g
c
(III-I)
where:
P = pressure
p = gas density
g = acceleration of gravity
u = gas velocity
z = elevation above some reference plane
H may vary from streamline to streamline. Now, referring to Figure
III-I, we will make the further simplications that the flow originates at
a surface above the refuse bed, Section a, and is divided into two
distinct zones with fluids of different densities, Zone 1 and
1II-5
Arthur D Little, Inc
-------
H
H
H
I
'"
»
...,
...
:r
c
...,
o
C'
...
...
in
:J
o
Re fuse
Feed
Section 'a'
Section 'b'
ulb b ul .
Interface
Zone CD
A2b A
u2b u2 .
Zone@
Yb
u2a
Ala
A2a
FIGURE III-l
FURNACE FLOW MODEL FOR ANALYSIS
t
To Scrubber
and Stack
-------
Zone 2. If the pressure is constant across Section a, and the mass flow
per unit area is uniformly distributed over Section a in each zone, the
total head, H, for all streamlines in each zone is the same, and known,
and one can readily compute velocities at certain points in the flow
field. For example, at Section b, where the streamlines are essentially
parallel, the pressure gradient perpendicular to the streamlines will be
due only to the hydrostatic pressure gradient in the fluid, and the velocity
in each zone will be constant, but different in the two zones. There-
fore, calculation of conditions at Section b given those of Section a is
straightforward. The equations relating the conditions at Section b to
those at Section a, based on the above assumptions, are developed in Appen-
dix A.' Given the width and velocity at Section a for each zone, and the
density of the fluid in each zone, the widths and velocities for each zone
at Section b can be calculated. The approximate velocities in the region
between Section a and Section b can then be inferred from a general knowledge
of this type of curved potential flow field. Case studies based on this
analysis are described subsequently.
The analytical model described above can also be applied to incinera-
tors where the flow out of the furnace is directed upward or at angle to
the horizontal. If a horizontal plane through the vertical flow channel,
where the streamlines were more or less parallel, were defined as Section b,
the equations of Appendix A could be used to calculate conditions there,
given those just over the refuse bed, at Section a. The magnitude of
velocities in the flow field between the sections could then be inferred.
2.
INSTABILITY OF HORIZONTAL FLOW
In an incinerator like that shown in Figure III-I, the flow from Section b
to the furnace outlet is essentially horizontal with warm, lower-density
gases tending to remain at the top and to flow with higher velocity as a re-
sult of acceleration in the vertical flow from the refuse bed. It is de~irable
that, in this part of the furnace, turbulent fluid mixing will enable completion
of combustion. Hence, it is of interest to determine whether the stable
density distribution will significantly inhibit turbulent mixing of the hot,
fuel-rich gas in the upper regions with the cooler, air-rich gases below. An
analysis of the stability of this type of flow is presented in Appendix B.
It is concluded that, for common incinerator configurations, turbulent mixing
will not be suppressed by the thermal stratification of the gases. Hence,
the horizontal flow downstream from Section b will be dominated by turbulent
mixing and velocity, temperature or composition gradients at Section b will
tend to be dissipated.
3.
TURBULENT MIXING IN CHANNEL FLOW
In an incinerator configuration like that shown in Figure III-I, the
flow out of the furnace, i.e., downstream of Section b, can be likened to
two-dimensional, turbulent flow in a duct of constant width. As shown in
the previous section, thermal stratification does not suppress turbulent
mixing. A turbulent shear region characterized by a velocity gradient
between the higher velocity, hot gas near the top of the duct and the low
velocity, cooler gas below will develop.
1II-7
Arthur D Little, Inc
-------
When the shear layer is narrow compared to the total height of the
flow channel, the exchange of mass and momentum effected by the turbulence
will act to increase the width of the shear layer, but leave the bulk of
the flow unaffected. Once the shear layer has spread to include most of
the flow, the effect of turbulent exchange is to reduce the difference in
velocities on either side. In a configuration such as Figure III-I, con-
ditions at Section b are probably best approximted by considering that the
shear layer includes the entire flow.
While there may be gradients of other variables in the flow, i.e.,
temperature and chemical composition, it is the gradient of velocity
which drives the turbulence and, thereby, controls the rate of change
of all other variables. Therefore, the appropriate calculation pro-
cedure is to deal with the velocity field first, and then calculate the
effect on other variables in proportion to the velocity exchange.
The decay of a velocity differential in a channel filled with a
shear layer can be related to the half-angle of growth of an unconfined
shear layer by the assumption that the momentum exchange across the shear
layer is the same in both cases--only its manifestation is different.
It is shown in Appendix C that this assumption leads to the result:
I d(ul - uZ) - - 3~
(ul - uZ) dx - ZA
(III-2)
where ~ is the half-angle of spread in the unconfined case.
of spread can be related to the half-angle of spread of the
jet into quiescent fluid. ~ . by a Galilean transformation.
o
The half-angle
edge of a plane
Thus.
ul - u2
~ = ~
o ul + u2
(UI-3)
where ~ is a constant equal to about .049.
o
d(ul - uZ)
dx
3 ~o (ul-
= - ZA . (ul +
into (111-2) gives
2
u2)
u2)
(UI-4)
Substituting this result
Note that the rate of decrease of the shear velocity (ul - u2) with distance.
is proportional to the square of the shear velocity. whIch is to be expected
since the turbulent shear stress is proportional to the square of the shear
velocity. This equation is easily integrated, assuming ul + u2 to be
approximately constant, yielding the result
ul - u2
=[
ul + u2
3~ x
~A + E]
-1
(UI-5)
111-8
Arthur D Little, Inc
-------
Estimates of the effects of mixing for typical cases, based on Equation
(111-5), are presented in Section D.
Two other'effects which have been neglected in the discussion above
arise from the turbulence that would exist in the absence of internal
shear--natural channel turbulence. One effect of this turbulence is to
retard the high-velocity portion of the stream more than the low-velocity
portion by surface shear against the channel walls. The other is the
internal momentum transfer due to this turbulence. A comparison of the
shear stress associated with momentum exchange depicted in Figure C-I(b)
to the shear stress at the wall of pipe shows that the two are equal
when
ul - u2 J~
+ = ---2~ ~ 0.22
ul u2 ,,"0
(III-6)
where f is the wall friction factor.
There is another consideration of importance in evaluating the
effectiveness of turbulent mixing. The scale of the dominant turbulence
is related to the width of the shear region, which in the present case
is the entire channel dimension. The reduction of shear velocity, and
the associated reductions of temperature and composition differentials,
is due primarily to mixing of fuel-rich and fuel-lean constituents on the
gross scale of the dominant turbulence. That is, at any station in the
flow field, a time sequence of samples will show gas compositions alternating
between rich and lean and passing by in large, relatively distinct chunks.
The completion of chemical reactions requires mixing on a scale comparable
to the diffusive processes which ultimately control combustion, and times
and flow lengths much longer than are realized in existing incinerators.
Thus, even though a significant amount of gross mixing may be predicted
by the analysis outlined here, the effect on gas composition (burnedness)
may be much less within the flow length available. The most effective
means of promoting combustion is to generate high turbulence levels at a
very small scale.
III-9
Arthur 0 Little, Inc
-------
D.
CASE STUDIES
1.
OVERFlRE REGION
Using the analytical model described in Section C-l, we have analyzed
a number of cases for various conditions of incinerator operation. The
bases for the cases analyzed are summarized in Table III-I. In addition,
the following assumptions are made in all cases:
a.
The furnace configuration is that shown in Figure III-I, with
Yb = 3.7 feet and A = 12 feet.
Both the hot and cool gases have a molecular weight of 29.
b.
c.
The air input under the grates is uniformly distributed
along the length of the grates.
d.
The average refuse burning rate per unit of grate area
equals 60 lb/hr-ft2.
e.
The absolute pressure level is approximately 1 atmosphere
everywhere.
f.
The majority of refuse gasification (mass addition to the
under grate air flow) takes place in Zone 1.
Other assumptions implicit in the 2-zone flow model have been described
in Section C-1.
Cases A-I and A-2 differ in that the refuse is moist in A-I and
dry in A-2. Case A-3 is similar to A-2, except that air bypass through
voids in the refuse bed (channeling) is assumed in A-3, in an amount
equal to that taking part in the gasification reactions.
Cases A-4, A-5, and A-6 are identical to A-I, A-2, and A-3, respec-
tively, except that the hot gas temperatures above the gasification
section of the bed differ. In Cases A-I through A-3, this temperature
is assumed to be 2000°F, a value thought to be representative of average
conditions across the gasifier section. In Cases A-4 through A-6, hot
gas temperatures corresponding to complete combustion are assumed. The
latter represent an upper limit that might be achieved well above the
bed where mixing with additional air might enable completion of combustion.
Thus, an actual case where further combustion occurs well above the bed,
but not to completion, might lie somewhere between Cases A-I and A-4,
for example.
Cases B-1, B-2, and B-3 have only 125 percent of stoichiometric
air flow, but are otherwise similar to Cases A-I, A-2, and A-3, respec-
tively, which have 150 percent of stoichiometric air flow.
111-10
Arthur D Little, Inc
-------
TABLE 111-1
BASES FOR CASES ANALYZED
H
H
H
I
I-'
I-'
Hot Zone Airflow
Air input to Air By-Pass Hot Gas
through Hot Zone Cool Zone Assumed Assumed
Refuse* Total ** Gasification Voids in Total Total Temperature Hot Gas Cool Gas
Case Reaction for Complete
Characteristic Airflow in bed, Refuse Gasf1ow, Gasf1ow, Combustion, Temp. Temp.
1b/hr (Channeling) , 1b/hr 1b/hr of of of
1b/hr
A-1 Moist 150 37,515 0 66,420 54,735 2,860 2,000 1,000
A-2 Dry 150 21,525 0 39,975 70,725 3,400 2,000 1,000
A-3 Dry 150 21,525 21,525 61,500 49,200 3,400 2,000. 1,000
A-4 Moist 150 37,515 0 66,420 54,735 2,860 2,860 1,000
A-5 Dry 150 21,525 0 39,975 70,725 3,400 3,400 1,000
A-6 Dry 150 21,525 21,525 61,500 49,200 3,400 3,400 1,000
B-1 Moist 125 37,515 0 66,420 39,360 2,860 2,000 1,000
B-2 Dry 125 21,525 0 39,975 55,350 3,400 2,000 1,000
B-3 Dry 125 21,525 21,525 61,500 33,825 3,400 2,000 1,000
B-4 Moist 125 37,515 0 66,420 39,360 2,860 2,000 1,500
B-5 Dry 125 21,525 0 39,975 55,350 3,400 2,000 1,500
B-6 Dry 125 21,525 21,525 61,500 33,825 3,400 2,000 1,500
»
.,
....
~
C
.,
o
C'
....
....
{5""
:J
o
*
"Moist" is equivalent to a moisture content of 35 percent by weight and "dry" to 0 percent moisture.
**
Airflow quantities are expressed as percentages of equivalent stochiometric air flow.
-------
Cases B-4, B-5, and B-6 are identical to B-1, B-2, and B-3, respec-
tively, except that the cool gas temperatures differ. The higher cool
gas temperature of l500°F for Cases B-4 through B-6 will reduce the
density difference between hot and cool gases. Hence, these cases will
show reduced buoyant effects.
The results of calculations for the various cases are summarized in
Table III-2.
The gas in both Zones 1 and 2 is accelerated between Sections a and
b. The average velocity must increase because the flow area decreases.
However, the hot gas in Zone 1 experiences a larger velocity increase due
to its buoyancy relative to the cooler, denser gas in Zone 2. In effect,
a vertical pressure gradient due to the gravity field, and equal to
(PI - P2)g, is imposed on the gas in Zone 1, in addition to any gradient
associated with flow area change. It produces the greater acceleration
in that zone.
As would be expected, the velocity increase of the hot gas is most
pronounced in Cases A-4 through A-6 where the temperature difference
between hot and cool gases and, hence, the density difference, is largest.
The velocity increase of the hot gas is least in Cases B-4 through B-6
where the temperature difference is smallest. Cases A-2, A-5, B-2, and
B-5 represent conditions with dry refuse and no air bypass in the gasifica-
tion section. The air requirements for gasification are least with dry
refuse, and with no bypass the flow rate of hot gas is minimal. In these
cases, the hot gas at Section B occupies only about one-quarter of the
height of the duct. Case A-5, in which the temperature of the hot gas
is that corresponding to complete combustion, produces the highest hot
gas velocity at Section B of any of the cases. This case represents
an extreme in terms of the hot gas being confined to a narrow, high-
velocity zone near the roof of the furnace. Cases B-4 and B-6 indicate
the opposite situation, i.e., the velocity difference between hot and cool
gases is minimal and the fraction of the flow area occupied by hot and
cold gases is nearly the same. A vertical velocity profile in the breeching
area corresponding to these cases would indicate minimum velocity gradients.
The latter conditions seem to agree best with the limited observations
that have been made in the incinerator at Newton, Massachusetts. Finally,
the differences in conditions at Section b brought about by the reduction
of air flow from 150 percent of stoichiometric to 125 percent, i.e.,
comparing Cases A-I through A-3 to B-1 through B-3, do not appear sig-
nificant.
Table 111-2 indicates conditions only at Sections a and b. The
velocities near the lower edge of the hot zone will range between the
values at Sections a and b, generally increasing as the flow proceeds
to Section b. A pressure gradient perpendicular to the stream lines
will be associated with their curvature, i.e., the turning of the flow.
The pressure in the upper left-hand region of Zone 1 will be somewhat
higher than near the interface between Zones 1 and 2. Hence, the
velocities in the upper left-hand region will be lower.
III-12
Arthur D Little, Inc.
-------
TABLE III-2
*
RESULTS OF CALCULATIONS
Case Ala A2a u1a u2a A1b A2b u1b u2b
ft ft ft/sec ft/sec ft ft ft/sec ft/sec
A-I 16.3 23.7 8.8 3.0 5.38 6.61 26.5 10.5
A-2 9.3 30.7 9.2 2.9 3.04 8.95 28.2 10.0
A-3 18.7 21.3 7.1 3.0 . 5.22 6.77 25.3 9.3
A-4 16.3 23.7 11.8 3.0 5.52 6.47 34.9 10.8
.
A-5 9.3 30.7 14.5 2.9 3.19 8.80 42.2 10.2
A-6 18.7 21. 3 11.1 3.0 5.49 6.50 37.7 9.7
B-1 19.5 20.5 7.3 2.5 5.88 6.11 24.2 8.2
B-2 11.2 28.8 7.7 2.5 3.23 8.76 26.6 8.1
B-3 22.4 17.6 5.9 2.5 5.67 6.33 23.3 6.8
B-4 19.5 20.5 7.3 3.3 7.04 4.96 20.3 13.6
B-5. 11. 2 28.8 7.7 3.3 4.20 7.80 20.4 12.2
B-6 22.4 17.6 5.9 3.3 7.17 4.83 18.4 12.0
*
The physical significance of the noted quantities (Ala' u1b' etc.) may
be seen by reference to Figure III-I.
III-13
Arthur D Little, Inc
-------
Since we do not have an accurate picture of the actual flow in an
incinerator furnace, we can only speculate on the differences between
an actual flow and the highly simplified model described above. To
begin with, air flow may not be introduced uniformly along the bed.
Operation with lower pressures in the wind boxes under the grates near
the discharge end is not uncommon. Overfire air may also be introduced
at various places over the bed. The air flow just above the refuse bed
will not be sharply divided into two zones. Instead, continuous velocity
and temperature distributions will exist as a result of the combustion
and local mixing processes. As the warmer gas flows upward and is
accelerated, mixing between it and the cooler gases would induce momentum
exchange, tending to slow down the hot gas and accerate the cool gas in
the mixing region, and further smooth out any sharp velocity gradients.
The cool gas tends to flow at much lower velocities, particularly when
air flow is reduced toward the end of the bed. The momentum exchange
in the mixing zone adjacent to the hotter gases will tend to produce
higher, cool gas velocities in that region and lower velocities in the
region away from the hot gas. In effect, a kind of circulation can be
induced in the cool gas. Under certain conditions, it might even be
possible to have cool gas flowing downward in the region farthest away
from the hot gas. By the time the flow reaches Section b, we would
expect a smoothed out velocity distribution of the Gaussian-type, with
higher velocities at the top and lower nearer the bottom. The maximum
and minimum velocities are expected to be of the magnitudes indicated
in Table 111-2, depending' on the operating.conditions.
The discussion thus far has dealt with furnace configurations like
that shown in Figure III-I. As previously noted, the analysis of Section
C-l could also be applied to different configurations, such as to those
where flow out of the overfire region ~ere' directed upward or at at angle
to the horizontal. If, in such configurations, the vertical distance
from just above the refuse bed to a section across the outflow duct is
larger than the height (y + A2 ) in Figure III-I, the gravity pressure
gradient acting on the ho~ gas ~P2 - PI)g, could produce larger velocity
increases than for the cases of Tables 111-1 and 111-2. Thus, the
tendency for hot gases to flow in a narrow, high-velocity zone in the
duct would be more pronounced. Application of the equations of
Appendix A to such configurations could provide estimates of gas
velocities and zone widths like those in Table 111-2.
2.
CHANNEL FLOW REGION
The flow from Section b to the furnace outlet will be modified by
turbulent mixing. Velocity, temperature and composition gradients will
tend to be reduced. If we assume that the shear layer at Section b
occupies the entire duct, Equation (111-5) can be used to estimate the
reduction in velocity differential. Taking x = 0 at Section b, and
using the representative velocities (per Table 111-2) of
III-l4
Arthur D Little, Inc
-------
u1b = 30 FPS
u2b = 10 FPS
we find that
u1b - u2b
= 0.5
u1b + u2b
and from Equation (111-5), C = 2. Assuming that the flow length to the
furnace outlet is approximately three times the height of the breeching
(x ~ 3A), we find that at the outlet:
3Cto x 9
2A = 2 Cto = .221
and
u1 - u2 1
= [2.221]- = 0.45
ul + u2
Thus, the differential velocity (u1 - u2) would be reduced by
x 100 = 10%. To achieve a 50 percent reduction would require
3Ct x
o
2A = 2, or x ~ 30A.
shear velocity decays, natural. channel turbulence becomes relatively more
important and will augment the decay.
(0.5 - .45)
0.5
The latter estimate is conservative because as the
A 10 percent decrease in differential velocity coressponds to only a 5
percent change in both ul and u , i.e., ul would decrease by 5 percent
and u2 increase by 5 percent. ~xchanges of energy (or temperature) and
composition between the high and low velocity regions would be of the
same magnitude. Thus, it appears that for conditions represented by the
ranges of parameters in Tables 111-1 and 111-2, a rather limited degree
of mixing can be realized downstream of Section b.
The percentages of air addition to the fuel-rich gases required to
enable completion of combustion for the cases of Tables 111-1 and 111-2
are shown in Table 111-3 and are based on the results of the bed analysis
presented in Chapter II. Comparison of these values to the magnitude of
composition mixing indicated by the example above suggests that mixing of
the fuel-rich and air-rich portions of the gas flow downstream of Section b
will not be effective in completing combustion.
III-IS
Arthur D Little, Inc
-------
TABLE III-3
PERCENT AIR ADDITION TO FUEL-RICH
HOT GAS FOR COMPLETE COMBUSTION
Cases
Volume
Percen t
Air Addition
A-I, A-4, B-1, B-4
A-2, A-S, B-2, B-S
A-3, A-6, B-3, B-6
36
100
30
As noted in Section C-3, mixing on a scale comparable to the diffusion
processes that limit the combustion rate is required to assure burnout of
combustible pollutants. Hence, the gross mixing indicated by the discussions
above must be considered as an upper limit on the changes in chemical co~
position on a scale appropriate to chemical reactions.
ul - Uz
The dependence of mixing on the shear velocity ratio ( )
ul + Uz
and the observations above suggest that ideal mixing, .in terms of com-
bustion reactions, might be accomplished by small-diameter, high-velocity
jets flowing in the opposite direction to the main stream, i.e., with
negative Uz either in the furnace or in the breeching. The effect would
be to maximize shear velocity, minimize the average velocity of the shear
layer which carries the fluid downstream, and to produce turbulence at the
appropriate scale. Though perhaps difficult to achieve in practice, the
concept may serve as a useful goal to approach.
I
I '
111-16
Arthur D Little, Inc
-------
E.
APPLICATION TO DESIGN AND PERFORMANCE EVALUATION
The preceding sections have outlined techniques and reasoning pro-
cesses for establishing the general features of the flow field in the
furnace. Using the approaches discussed, the approximate levels of
velocities for the hotter and cooler gases can be estimated at a cross-
section where the flow leaves the overfire region and the flow strea~
lines are essentially parallel. The width of the hot gas and cool gas
zones can also be estimated and used to establish a rough velocity pro-
file across this section. Velocities between the section just above
the refuse bed and the outlet from the over fire region can then be roughly
inferred. To make these estimates, knowledge of the gas flow from the
refuse bed is required. Such information can be developed from the estimates
of the undergrate air flow distributions and the concepts described in
Chapter II.
An approximate description of the general flow field is useful in
defining the flow environment for the injection of air or stream jets
to promote mixing in the furnace. Knowledge of the flow field can
suggest locations of the jets, the direction in which they should point
and their velocity level. It can also provide an indication of where
the hottest gases will flow, and the maximum velocities they can achieve.
Such information can be helpful in arriving at furnace configurations
that avoid the deleterious effects that might be caused by hot gases
impinging on certain surfaces (such as boiler tubes or bridgewalls)
with relatively high velocities.
The need for the addition of air to the hot, fuel-rich gases is
evident from the discussion of the bed combustion process in Chapter II.
Even if the total underfire air flow were greater than 100 percent of
stoichiometric for complete combustion, the tendency for thermal
stratification of gases in the overfire region and for limited mixing
between hot, fuel-rich gases and the cooler, air-rich gases below, as
the flow proceeds out of the furnace, indicates the need for auxiliary
means of mixing the air-rich and fuel-rich gases. The addition of over fire
air to foster complete combustion can also be effective, provided it is
introduced in a way that promotes turbulent mixing with the fuel-rich
gases on the relatively small-scale that promotes chemical reactions.
The potential effectiveness of jets flowing counter to the hot gas has
been noted. In short, our findings about the refuse bed combustion
process and the general nature of flow in the furnace provides a rationale
and certain criteria for the use of overfire air and/or steam jets to
promote turbulent mixing in the furnace.
The fact that turbulence with small characteristic dimensions is
required to effectively complete combustion also has a bearing on the
design of baffles, grids or other passive, stationary mixing devices for
furnaces. Large baffles with dimensions of the same magnitude as the
height of the flow passage can redirect the gross flow and produce tur-
bulent mixing as a result of flow separations and eddies downstream.
However, this turbulence will be on the' scale of the baffle dimension
III-17
Arthur D Little, Inc
-------
and, in effect, will at best only promote gross mixing. Though such
mixing is useful, it is but one step in the process leading to complete
combustion.
For the remainder of the process, the use of baffles with charac-
teristic dimensions measured in inches, or even fractions of an inch,
would appear to be superior to those with dimensions measured in feet.
One can speculate that, for a furnace which is ideal in terms of achiev-
ing complete combustion, gross mixing should be accomplished early in
the flow process, i.e., in the overfire region, by the use of air or
steam jets, or stationary mixing devices. Further mixing on a small
scale should then be promoted by stationary mixing devices with small
dimensions, at, say, the inlet to the breeching area, so that completion
of reactions can be accomplished before the flow exits from the furnace.
In practice, attention must be given to the problems of draft loss and
plugging with such closely spaced baffles. Experience with convective
boiler surfaces, however, which provide such a mixing action, suggests
that these problems are not insurmountable.
Finally, calculations performed for various ~perating conditions
such as those of Table 111-1, or for other appropriate conditions, can
indicate how operating conditions will affect flow conditions and
turbulent mixing in the furnace.
II1-18
Arthur 0 Little, Inc
-------
F.
SUMMARY
The combustible pollutants arising from the burning refuse bed flow
through the furnace and flues in a non-uniform manner. Temperature, gas
composition and velocity in the flow are different as a consequence of the
geometry of the system, the spatial distribution of their entry points and
the inadequacy of self-mixing. Understanding the flow field and the effects
of design and operating variables on the flow can assist in the furnace
enclosure design; indicate roughly the degree of self-mixing which occurs;
support the design of baffles and other flow-control surfaces; importantly
affect decisions on the placement and size of convection boiler tubes; and
indicate the flow environment which interacts with overfire air or steam
jets.
As for the bed processes, analysis of the flow field required simplifi-
cation. The approach taken was to assume a two-dimensional flow pattern
and to break the flow into several regions. In some portions of the fur-
nace, an approximation to potential flow occurs and, after including buoyancy
effects in the equations, estimates were made as to gas velocity as a function
of position. In other regions of the furnace, shear layers are developed
and consideration must be given to turbulent mixing. Here, concepts drawn
from plane jet theory were applied to yield estimates of mixing levels and
degrees.
From these results, flow patterns and velocity distributions were de-
veloped which provide the designer with new insights into system behavior.
These results, both quantitatively and qualitativel~ are particularly help-
ful in establishing the positions and flow environments of air and steam
jets for combustible pollutant control.
III-l9
Arthur D Little, Inc
-------
G.
REFERENCES
1.
W. H. McAdams, "Heat Transmission", McGraw Hill Book Company, Inc.,
1954.
2.
W. M. Rohsenow and H. Y. Choy, "Heat Mass and Momentum Transfer",
prentise Hall, Inc., 1961.
3.
E. R. G. Eckert, E. M. Sparrow, W. E. Ibele, R. J. Goldstein and
C. J. Scott, "Heat Transfer--A Review of Current Literature",
International Journal of Heat and Mass Transfer, 1968.
4.
H. Schlichting, "Boundary Layer Theory", McGraw Hill Book Company,
Inc., 1955.
5.'
W. E. Groume-Grjimailo, "The Flow of Gases in Furnaces", John
Wiley & Sons, INc., 1923.
6.
E. L. Harris, "The Flow of Gases in Furnaces", 1925.
7.
Private communication from Metcalf and Eddy, Inc., Boston, Massa-
chusetts.
8.
P. H. Woodruff and G. P. Larson, "Combustion Profile of a Grate-
Rotary Kiln Incinerator", Proceedings of 1968 National Incinerator
Conference, ASME (1968).
111-20
Arthur 0 Little.lnc
-------
CHAPTER IV
OVERFIRE JETS
Arthur D Little, Inc
-------
CHAPTER IV
OVERFIRE JETS
A.
INTRODUCTION
The preceding chapters on the behavior of refuse beds and on the flow
dynamics of furnace gases indicate that situations exist with incinera-
tor furnaces where jet systems could be of assistance in realizing better
burnout of combustible pollutants and in controlling furnace temperature
distributions. A review of the design and operating characteristics of
existing incineration systemsl and discussions with incinerator designers
suggest the need for better correlations supporting the design of these
jet systems. As discussed below (Section B), the fluctuating conditions
of gas movement and composition within incinerator furnaces present a
considerable challenge to the detailed analysis of any device which inter-
acts with the flow and combustion processes. As a consequence, our analysis
was necessarily somewhat simplified. We feel, however, that our results
will support the design of practical systems which will perform effectively
and in accord with th~ expectations of the designers.
This chapter consolidates existing overfire jet design correlations.
Building on existing theory and incorporating the results of original ex-
periment and analysis carried out as part of this effort, improved design
methods are then suggested for application to refuse incineration systems.
1.
USE OF JETS
Jets have been utilized for many years as an integral part of furnaces,
boilers and other combustion systems. In boilers fired with pulverized
coal, for example, air jets are used to convey the fuel into the combustion
chamber, to control the heat release patterns and to supply secondary air
for complete combustion. In processes employing- a burning fuel bed, properly
placed air jets supply secondary air where needed above the fuel bed to
complete combustion. Also, jets of air and/or steam are used to induce
turbulence and to control temperature by dilution of furnace gases.
The important characteristics of jets which underlie all of these uses
are:
.
The controlled addition of mass to contribute to the oxidation
process (air jets) or to serve as a thermal sink to maintain
gas temperatures below levels where slagging, corrosion, or
materials degradation may occur (air or-steam jets); and
.
The controlled addition of momentum to promote mixing of the
jet-conveyed gas with gases in the combustion chamber or to
promote mixing of gases from different parts of the combustion
IV-l
Arthur 0 Little, Inc
-------
chamber. In the latter case, high-pressure steam jets are
often used to provide high momentum fluxes with a minimum
introduction of mass. .
The basic challenge to the combustion system designer is to employ
these characteristics to maximum advantage in meeting his overall design
goals. How jet capabilities relate to the design goals of the incinera-
tor designer is discussed in Section B of this chapter.
2.
GENERAL CHARACTERISTICS OF JETS
Because of the long-standing practical interest in the use of jets,
a large body of. literature has been developed which quantitatively
characterizes the nature of jet flow. Jets are conveniently categorized,
according to flow regime (laminar or tubulent, supersonic or subsonic)
and geometry (round or plane). Laminar jets occur only at very low jet
velocities and are of no interest here. Supersonic jets, of interest
for describing high-pressure steam flows, are of potential interest but
are not considered here. Plane jets, which issue from a slot finite in
one dimension and effectively infinite in the other, are primarily of
academic interest. We will, therefore, focus in this discussion on
round, low-subsonic, tubulent jets and return later to the fact that a
row of closely-spaced round jets behaves, to a degree, like a plane jet.
Other important parameters characterizing the jet flow behavior in-
clude the relative densities of the jet and ambient fluids, the velocity
of the ambient fluid relative to the jet velocity, and the degree to which
the space into which the jet issues is confined by walls. Also, in situa-
tions where the combustion can occur (jets of fuel into air as in burners
and jets of air into fuel vapors--the so-called "inverted flame"), the initial
temperature and combustible content of the jet and ambient fluid are of
interest. All of these factors are important in the application of jets
to incinerators, and their effects, singly and in combination, on jet
characteristics are discussed in Sections C and D. To set the stage for
this discussion, we consider here the basic characteristics of jets issuing
into an infinite atmosphere of quiescent fluid of the same density as the
jet fluid.
The round, isothermal turbulent jet shows three characteristic regions
(Figure IV-l). Immediately adjacent to the nozzle mouth is the mixing
region. Fluid leaves the nozzle with an essentially flat velocity profile.
The large velocity gradients between this "potential core" and the ambient
fluid induce turbulence which causes ambient fluid to mix into the jet.
The mixing results in momentum transfer between the jet and ambient fluids
and progressively destroys the flat velocity profile. In a distance of about
4.5 jet nozzle diameters downstream" and turbulent diffusion has wOI~ed its
way to the centerline of the jet and eliminated the potential core.
It is important to note that the "nozzle diameter" characterizing jet
flow is not necessarily the physical dimension of the orifice from which
the jet issues. If, for example, the jet issues from a sheet-metal plenum,
a flow contraction to about 60% of the open discharge area (the area of the
vena contracta) characteristic of the flow past a sharp-edged orifice, will
IV-2
Arthur D Little.lnc
-------
define the effective nozzle diameter and the location
discharge plane will be displaced about two-thirds of
of the orifice (the location of the vena, contracta).
diameters) constant area section lies upstream of the
nozzle diameter may be taken as the orifice diameter.
given, therefore, to the geometry of the entire nozzle
tem in determination of the flow rate of jet fluid.
of the effective jet
a diameter downstream
If a relatively long (2-3
discharge plane, the
Attention should be
fluid delivery sys-
In the region from 4 1/2 to about 8 diameters downstream, the tran-
sition of the flat entrance velocity profile to a fully developed profile
is completed. Beyond this transition zone, the velocity profile remains
more or less of constant shape relative to the velocity on the axis of the
jet and is referred to as "self-preserving."
Important jet characteristics include:
a.
The centerline velocity and concentration changes with axial
distance from the nozzle mouth;
b.
The shape of the velocity and concentration radial profiles
in the fully developed region.
c.
The intensity of turbulence in the jet; and
d.
The rate of entrainment of ambient fluid into the jet.
These characteristics are all interrelated; turbulence generated by
high velocity gradients induces entrainment which causes momentum and
mass transfer between the jet and the ambient fluid.
These characteristics are important in practice because they determine
the quantititive effect of firing a jet into a combustion chamber. The
axial decay of velocity establishes. how far the jet effectively penetrates
into the chamber. The radial velocity distributions determine how large
a volume is affected by the jet. The entrainment rates determine how
effectively furnace gases are mixed along the jet path.
IV-3
Arthur D Little, Inc
-------
r- ---- -
--
H
<
I
~
Mixing
Region
Constant Axial
Velocity
»
~
g-
..,
o
-
,
;::;:
-
{5"
R
I
FIGURE IV-'
Transition
Region
Decaying Axial
Velocity
REGIONS IN JET FLOW
Fully Developed
Region
-------
B.
THE USE OF JETS FOR COMBUSTION CONTROL
As described in Chapter I, municipal incineration is an important
source of combustible pollutants. The analytical developments in Chapters
II and III showed that these pollutants would necessarily arise in the
pyrolysis zone of the grate and could possibly arise in the discharge
zone. Further, it was shown that the turbulent mixing processes naturally
occurring in the flow through the furnace could be inadequate to supply
the stoichiometric oxygen to the fuel-rich gases. Even when this oxygen
is admixed in the latter stages of furnace flow, however, it is entirely
possible that the residence time remaining may be too short for complex
organic species or for thick-sectioned char particles to completely burn.
These observations lead to the conclusion that systems are needed to
provide air near the pyrolysis zone and/or to induce high-intensity turbu-
lence at strategic locations within the incinerator furnace. Although
passive mixing systems, such as baffles or checkerwork, may have some.value
in the inducement of turbulence, they clearly are unuseful in supplying air
and are not alterable to cope with changes in the distribution of combustible
pollutant release along the bed and throughout the chamber as refuse compo-
sition and burning characteristics change. As a consequence, incinerators,
as one of a family of grate-burning systems, have turned to the use of
overfire jets of either steam or air as a low-cost and effective control
technique.
1.
JET DESIGN FOR INCINERATORS--A STATEMENT OF THE PROBLEM
Combustible pollutants appear to be generated along the full length of
the incinerator grate, although their discharge rate into the overfire
volume is relatively low in the drying and ignition zones prior to the
introduction of underfire air. From the standpoint of total pounds per
hour per square foot release rate, the pyrolysis zone probably qualifies
as the single most important source.of carbon monoxide, soot and hydro-
carbons (Figure II-~ of Chapter II). Carbon monoxide and coked ash material
will be evolved in the region between the pyrolysis zone and the burnout
region. Also, data developed in the course of this program (see Chapter II)
indicate that combustibles may be evolved in the discharge grate section.
Our analyses, data and speculations indicate that overfire air is definitely
required in the region of pyrolysis and char burnout; reQu~ed undergrate air
flows and turbulence inducement is re9uired in the area ov~r the discharge
grate; and some means may be required to increase the general level of
turbulen~e throughout the upper regions of the incinerator furnace.
The specification of jets for incinerator applications meeting the
requirements listed above places great demands upon the designer. It is
clear that the jet behavior should be known in a flow field where combus-
tion ,cross-flow and buoyancy effects are all potentially important; and,
for some systems, the jets must operate over long distances. This latter
characteristic arises from the shape of most continuou~-feed incinerators
which tend to be long and narrow, thus devices acting over the discharge
grate region which are expected to carry bed off-gases back towards the
IV-5
Arthur D Little, Inc
-------
pyrolysis region must act over distances of 10-30 feet (20 to 100 or more
jet diameters). The location, number, and flow parameters appropriate to
these jets should be consistent with the overall furnace geometry, be easily
maintained and operated, and should be controllable to the extent demanded
by the fluctuations in refuse composition and burning characteristics. Par-
ticularly in the case of air addition, the jet design should add sufficient
air to meet the oxygen requirement of the rising fuel vapors yet not pro-
vide so much air as to overly cool the gases, thus quenching combustion.
Also, the draft capabilities of the furnace must be considered in deter~Dlin-
ing the amounts of air introduced.
2.
EXPERIENCE IN JET APPLICATION FOR COAL-BURNING SYSTEMS
Overfire air systems have been used for over ninety years in coal-
burning practice. In some respects, the combustion characteristics of coal
burning on a grate are similar to those of refuse. Typically, however,
coal ignites more readily (partly due to its lower moisture content), .burns
with more regularity and predictability and, for overfeed or cross-feed
situations, is typically burned in furnaces with grates which are short
relative to those used in many continuous-feed refuse-burning incinerators.
Therefore, although the probems are not identical, it is of value to review
experience in coal-burning practice as an indication of the potential of
jet systems for combustion control.
The use of controlled overfire air in industrial solid fuel combustion
systems was stimulated by the desire to improve boiler efficiency through
complete combustion of soot and carbon monoxide and to reduce smoke emissions.
Although the historical patte~ of technological development of over fire
air systems is unclear, Stern mentions that patents and active marketing
of steam-air jets, primarily for smoke control, began in 1880. Quantita-
tive appreciation of the benefits of smokeless combustion on overall fuel
economy was widely argued until documented by Switzer3 in 1910. Switzer's
work, carried out at the University of Tennessee, involved measurements of
jet system steam consumption, smoke intensity and boiler efficiency on a
hand-fired return--tubular boiler fired with bituminous coal. The results
of his tests showed an increase in thermal efficiency from 52.6% to 62.1%,
an increase in the effective range of the boiler from 80% to 105% of its
rated capacity before smoking occurred, and a steam consumption for the over-
fire jet system of only 4.6% of the total steam raised.
Recognition of the importance of overfire air and mixing stimulated
considerable research in the first decade of this century. Some of the more
completely documented and detailed laboratory and field data was produced
by the Bureau of Mines who were conducting "investigations to determine how
fuels belonging to or for the use of the United States Government can be
utilized with greater efficiency." Kreisinger et. a1. 4 studied the combus-
tion behavior of several coals in a special research furnace under a variety
of combustion air and firing rate conditions. Their work showed a strong
relationship between the burnedness of the flue gas, the properties of the
coal, and the size of the combustion space (Figure IV-2). Their results were
interpreted in agreement with prior suggestions of Breckenridge5, to result
IV-6
Arthur 0 Little, Inc
-------
»
~
::r
!::
""'I
o
C
......
......
jb
5'"
o
~
I
.....
'- '- 1111. . I. I II r I
molS coo I II I II V
.
I '"
...1 ~I ... V
II
I
'- J I ;...
Pittsburgh - ~~
I II ~,
0)
I ,
I ~
I I ~
Pocahontas I / I,?
~
~
Q
-l11inois coal II V ,; 0,-6
1.1 -
~ V q.)'. \1.;' -'- b:::i-
'- l"j -j
'- -Pittsburgh V -I"" v
-t7 L.; ./
I I I I ~b ~ ,~v ...-
'- I-Pocnhont,:is - I !...-
'I
'Il\inois coal 1 I'
I '.I " \/
'? II 1/ /
J V
f- Pittsburgh l1 r/ II
I I I J /
- .1 II "/
~0C2honlas
I I I I
I I I I
I I I
60
I I I I I I
_l-lIIinois cool I
I II J
~ '!I / I......
0 / 0./
I 1 .1 ...
/
Pittsbu! z:!I.
J
I
I I
I
I ,
I
'- Pocahontas
I
. C
.
~
I, ~\' ~@,
Illinois coal } ,} ',:; !o .
/ ' ~
.,}I '.. 1/ ~
~ '10/ \. ./
/ 1/ "[; ~
I- - Pi'ttsburgb J J /
I II
I
V
I /
I Ii,
I- - Pocahontas
~ Z300
g:)toJtoJ
~=g
::>~p::
I:QjQ
:<: ° ~I
8 ;> taJ 200
taJro-...J
...Jop:)
-0<
,j~~
'o~>.
,> toJ <
ro- :I: 0100
o (04 F'
tiE3 z' .
::>~g
Q~~
0. <
f V 0
~~A
ti <.J"
-a::
OtaJ
::g/l.
Z.J
-<
~~30
-
f:=
j~4
OI:Q
>:)
tz.-
0<
0>
~'~'2
10
2 4 .6 . 8
RATIO OF COMBUSTION SPACE TO GRATE AREA,
CUBIC FEET PER SQUARE FOOT OF GRATE AREA
'0
10
2 . 4 6 8
: RATIO OF COMBUSTION SPACE TO GRATE AREA.
. CUBIC FEET .E'ER SQUARE FOOT OF GRATE AREA
FIGURE IV-2
RELATION BETWEEN COAL CHARACTERISTICS AND THE SIZE OF
COMBUSTION SPACE REQUIRED IN USBM TEST FURNACE AT
COMBUSTION RATES OF 50 LBS/HR FT2 AND 50% EXCESS AIR
-------
from differences in the emission of volatiles between coal varieties and
the rate-limiting effects of inadequate mixing. Their correlating param-
eter, which they called the "undeveloped heat of combustible gases", repre-
sented the heat of combustion of the flue gases relative to the heat of
combustion of the coal burned. In view of the trends shown in Figure IV-2
and the composition data in Table IV-I, it would appear that the volume
requirements for complete burnout in refuse incineration may be considerably
in excess of those acceptable in coal-fired combustors. Quantitative extrapola-
tion from their data to incinerators, however, would be highly speculative.
Unfortunately, within the scope of the U.S.B.M. experimental program,
generalized design guides for the flow rate and locations appropriate for
overfire air systems were not developed.
Some of the earliest test work directly aimed at finding the benefits
of overfire air in utility combustion systems was conducted in 1926 by
Grunert6 on forced draft chain grate stokers at the Commonwealth Edison
Company in Chicago. Grunert's data showed that overfire jets discharging
over the ignition zone could reduce the carbon monoxide levels at the entrance
to the first pass of boiler tubes from an average value of 1% to essentially
zero. Also, the gas temperature and composition profile could be made con-
siderably more uniform. Of importance to fuel economy, it was found that
although additional air was introduced through the over fire air jets, the
, total combustion air was susceptible to reduction. Similar work in Milwaukee,
reported by Drewry7 also showed performance improvement (an increase of 7.2%
in boiler efficiency), smoke elimination, and complete burnout of combustibles
within the firebox. Once again, however, design correlations generally
applicable to the coal-burning industry were not presented.
Major contributions to the overfire jet design art were published in
the mid-1930's. Of particular importance were reports on a number of meticu-
lous test prggrams carried out in Germany; perhaps typified by the work of
A. R. Mayer. Although still not ,proyiding generalized design criteria,
Mayer maie gas composition traverses (45 points) within a traveling grate
stoker furnace firing low~volatile bituminous coal. His results are shown
in Figure IV-3.
FIGURE IV-3
LINES OF EaUAL HEATING VALUE (KG-CAL PER STD CU M)
OF FLUE GAS FIRING LOW-VOLATILE BITUMINO~ COAL
AT A FUEL RATE OF 28 LB PER SO FT PER HOUR
a
No OV8rfir8 Ail
Overfire -Air Pressure, 1 inches
IV-8
c
Over'ire - Air Pressure, 10 inches
Arthur D Little.lnc
-------
TABLE IV-1
CHEMICAL CHARACTERISTICS OF COAL4 AND REFUSE1
Pocahontas Pittsburgh Illinois
Item Characteristics Coal Coal Coal Refuse
1 Volatile Mattera 18.05 34.77 46.52 88.02
2 Fixed Carbona 81.95 65.23 53.48 11. 99
3 Total Carbona 90.50 85.7 79.7 50.22
4 a 8.55 20.47 26.22 38.22
Volatile Carbon (Item 3 -
Item 4)
5 Available Hydrogen a 3.96 4.70 3.96 1. 57
6 a 2.16 4.35 6.60 24.34
Ratio Vol. C to Avail. H2
7 Oxygen a 3.32 5.59 10.93 41. 60
8 Nitrogen a 1.19 1. 73 1. 70 1. 27
9 ~ercentage of Moisture 2.53 2.88 22.07 55.19
Accompanying 100 Percent
of M & AF coal or refuse
10 Product of Items 1 and 6 39 151 307 2142
11 Ratio of Oxygen to Total 0.0367 0.0652 0.137 0.828
Carbona
12 Total Moisture in Furnace Per 0.409 0.501 0.700 1.161
Pound of Coal or Refuse Reduced
to M & AF Basis (Pounds)
a.
Percent on moisture and ash-free basis (M & AF)
IY-9
Arthur D Little, Inc
-------
As a measure of the completeness of combustion in the over fire space, Mayer
determined the heating value of the gases (Btu/cu ft) as calculated from the
complete gas analysis. Without overfire air, as seen in Figure IV-3(a), strata
of combustible gases rise into the combustion chamber and persist as the
gases enter the first boiler pass. This is indicated by the zero heating
value curve whicb is not closed. Figure IV-3(b) shows the effect of medium-
pressure overfire air jets. It can be seen that combustion is improved,
yet some fraction of tbe combustible still enters the boiler passes.
Figure IV-3(c) shows the effect of further increases in the overfire air
plenum pressure. Under these latter conditions, combustion is complete
within the furnace volume. For these tests, Mayer employed jets directed
towards the bed just beyond the ignition arch.
Also in the 1930's, developments in fluid mechanics by Prandtl and
others provided mathematical and experimental correlation on the behavior
of jets. Application of this understanding to furnace situations was pre-
sented in some detail by Davis.9 His correlations, although based on greatly
simplified assumptions, were of considerable interest to the furnace design-
ers in that time. As an example of the applicability of his work, Davis
explored the trajectories anticipated for jets discharging over coal fires
and compared his calculated trajectories with data by Robey and Har10w10
on flame shape in a furnace at various levels of overfire air. The results
. of this comparison are shown in Figure IV-4. Although general agreement is
shown between the jet trajectory and the flame .patterns, correlation of the
meaning of these parameters with completeness of combustion is unclear and
not supported by Robey and Harlow's data. The results do give confidence,
however, that jet trajectories can be calculated under a variety of furnace
conditions to produce reasonable estimates of behavior and thus permit
avoidance of impingement of the jet in the bed.
Although work on the applications and advantages of overfire air con-
tinued through the war years (particularly with reference to avoidance of
smoke in naval vessels to preclude some easy identification and submarine
attack2), the wide introduction of pulverized coal-firing in electric
utility boilers and the rapid encroachment of oil and gas into the domestic,
commercial, an4 industrial fuel markets rapidly decreased the incentive for
continued research into overfire jet systems for application in stoker-
fired combustors. Indeed, the number of literature references on this topic
falloff rapidly after 1945.
In summary, during the seventy or more years during which overfire air
jet application to stoker-fired systems was of significant importance, no
generalized design criteria of broad applicability had been developed. An
art had arisen regarding the use of overfire jets, typically over the ignition
arch and in the sidewalls of traveling grate stokers burning bituminous coal.
Sufficient jet design technology had been developed to allow specification
of jets which would penetrate adequately the upf10w of gases arising from the
bed and which served to smooth the temperature and gas composition profiles
at the entrance to the boiler passes above. Even in 1951, however, the
comment was made by Gumz11, a well-recognized contributor to combustion
technology, that "the number of nozzles, their location and direction are
the most disputed factors in the use of overfire air jets." .
IV-10
Arthur D Little.lnc
-------
FIGURE IV-4
COMPARISON OF OBSERVED FLAME CONTOURS AND
CALCULATED TRAJECTORIES OF OVER FIRE AIR JETS 9
Percentage of overfire air at the following points; A, 5.65;
B, 10.5; C, 16.6; D, 20.0; E, 21.4; F, 22.8; G, 26.8; H, 28.8.
IV-II
Arthur 0 Little, Inc.
-------
c.
REVIEW OF THE PRIOR ART
In 1880, the first patents were issued for steam-air jet devices to
supply overfire air and induce turbulence in hand-fired furnaces burning
bituminous coal. The development and marketing of these proprietary jet
systems reflected a need for improvements in combustion efficiency and
for means to reduce smoke emissions. Since that time, a number of re-
finements in the physical arrangements and design characteristics of over-
fire: jets have been offered to the technical community. In the literature
surveyed in the course of this effort, however, few instances were found
where comprehensive design correlations were presented. In the great
majority of cases (e.g., references 12, 13, and 14), the technical content
of the papers was limited to documentation of improvements in performance,
particularly with reference to smoke abatement, resulting from the use of
spec.ific arrays of overfire jets in a specific comb us tor. Mos t design
information dealt with such topics as the pumping efficiency or the estima-
tion of steam consumption in steam ejectors. We found few instances where
attempts were made to couple an analysis of jet behavior to an analysis of
furnace behavior.
To some extent, the tendency of early workers to report only emptrical
results reflected the limitations of theoretical understanding or math-
. ematical treatment techniques of their time. Also, the complexity of
furnace dynamics presents a considerable challenge to the analyst and thus
generalization is difficult. It is noteworthy, for example, that the rigor-
ous mathematical treatment of the behavior of two-dimensional plane jets
has only recently been solved in detail.lS Solution of this problem re-
quired the use of high-speed computers and complex numerical techniques.
Efforts at a similar analysis of the axi-symmetrical round jet are now
in process but, at the present time, trial solutions exceed reasonable core
storage and computation time on the fastest, modern computers.
Ideally, the design basis used in overfire air jet designs for incinera-
tor applications should recognize the effects of buoyancy, cross-flow and
combustion as they are experienced in real incineration systems. With two
exceptionsl6,17, our review of the existing art showed no correlations able
to cope with this full specf7um of potential interactions. These workers
however (Davisl6 and Ivanov), were concerned with coal-fired systems, ,
and application of their correlations to incinerators must be approached
with caution. It is of value, however, to review the prior art as it
reflects the basis of our analysis. Much of the prior work, although
supported by analytical studies, rests heavily on experimental results,
particularly to provide the constants used in the equations. Indeed, in
view of the complexity of even modest deviations from simple isothermal
free jet behavior, our work showed the need for empirical results as a check
on analytical predictions.
r 1\1.-12
Arthur D Little, Inc
-------
1.
ROUND ISOTHERMAL JETS
The behavior of circular jets discharging into a quiescent, non-
reacting environment at a temperature similar to that of the jet fluid
provides the starting point in any review of jet dynamics. Indeed, the
behavior of jets under such conditions has often been the primary guide-
line in the design of overfire jets for incinerator app1ications.1
Because of the relatively simple nature of jet structure under such con-
dit:lons, this configuration is perhaps the most studied, both analytically
and experimentally, and good correlations are available describing jet
trajectory, velocity, entrainment, turbulence levels, and the like.
The correlation of data taken by many experimenters leads to the
following expressions for the axial decay of centerline velocity and con-
centration.19 .
u p 1/2 d
moo
- = 6.3 (p-) (x + 0.6 d )
u a 0
o
(IV-1)
c p 1/2 d
--2!! == 5.0 8).
o
20
Abramovich gives the axial velocity and concentration decay to be:
u . do
--2!! = 0.48 (a x)
u . 1
o
(IV-2a)
c do
--2!! = 0.35 (ax)
c 1
o
(IV-2b)
IV-13
Arthur D Little, Inc
-------
where the value of a depends on the velocity profile at the nozzle mouth.
For a flat profile, al= 0.066. For an equilibruum turbulent velocity pro-
file, al= 0.076. Substituting the latter value into Equations (IV-2a&b) yield
u d
m 0
- = 6.3 (-)
x
(IV-2c)
u
o
c d.
m 0
- = 4.6 (-)
x
(IV-2d)
c
o
which agree well with Equations (IV-I) and (IV-2).
The experimentally measured radial velocity and concentration profiles
in the fully developed region can be represented by i~ther Gaussian or
cosine functions. The Gaussian representations are:
u
-=
r 2]
exp [-96 (-)
x
(IV-3)
u
m
c
-=
r 2
exp [-57.5 (-) ]
x
(IV-4)
c'
m
where u and c are the time-averaged velocity and concentration at distance
x downstream and distance r from the jet centerline.
Abramovich20(p. 89-97) applies Taylors physical mode] of turbulence
and shows that this theory predicts that
- 0.5
c u
- = (-)
(IV-4a)
c
m
u
m
The relationship between Equations (IV-3) and (IV-4)
- e:
c u
- = (-)
(IV-4b)
c
m
u
m
IV-14
Arthur D Little.lnc
-------
o'r
r
exp [-57.5 (r)2] = [exp [-96 (r) 2]]
x x
(IV-4c)
so that
L -57.5 - 0 6
= -96 - .
which agrees quite well with the value of 0.5 derived by Abramovich.
The spread of the jet is defined in terms of the half-angle to the half-
velocity point (i.e., the angle subtended by the jet centerline and the line
from the centerline at the nozzle mouth to the point where the velocity is
one-half of the centerline velocity. This angle is independent of distance
from the nozzle mouth in the fully developed region, a consequence of the
self-preserving nature of the velocity profile. The half-angle of the half-
velocity point is 4.85°; based on concentration in the same way, the half-
angle is 6.2°.
The turbulent intensity of the jet is defined in terms of u' and v',
the r.m.s. fluctuating velocity components in axial and radial directions,
respectively. Data of CorrsiJ1 show that the intensity ratio u' /um and
V"/urn depend on the ratio r/x. At x/do = 20, each velocity ratio varies
from about 27 percent at the centerline to about 5-7 percent at r/x = 0.16.
Ricou and Spaul~ing22 measured entrainment rates and determined that
the mass flow rate (m ) in the jet is linearly related to x according to
x
.
m P 1/2
~ = 0.32 (....!) (~)
Po do
mo
(IV-5)
A similar relationship can be computed from Equations (IV-I) and (IV-3) for the
uniform density case. Defining
m = f pu2'T1'rdr
x
.
00
(IV-6)
o
and
do/2
- 2
pUo'Tl'do
.
mo = f
o
puo2'T1'rdr
=
4
(IV-7)
Taking the ratio of Equations (IV-6) and (IV-7), substituting equations
(IV-I) and (IV-3) to give us a function of x and r, arid'
IV-IS
Arthur D Little, Inc
-------
o
m
x
0-=
mo
x
0.26 (-d) .
o
(IV-8)
which agrees well with Equation (IV-5).
These relationships describing the behavior of isothermal jets entering
a quiescent fluid are well-documented and form the basis for our design cri-
teria relative to the use of jets in incinerators. In this application,
however, the effects of crossflow of the ambient fluid and density differ-
ences between the jet and ambient fluids are important. These effects are
less well-documented, and their inclusion in the design criteria poses some
difficult problems. These matters are dealt with below.
2.
BUOYANCY EFFECTS
When the jet and ambient fluids are of different density, the buoyant
forces acting on the jet can cause deflections of the jet trajectory. This
effect is potentially important in incinerator applications since the air
itltroduced by the jets will be much colder than the furnace gases and hence
of higher density. From an incinerator design and operating standpoint,
this could be critical: jets could "sink" from an anticipated flow trajec-
tory passing above the bed to one causing entrainment of particulate from
the bed or causing overheating of the grates with a "blowpipe" effect.
Relatively little experimental or theoretical work has been done to
characterize jet performance under these conditions. Figure IV-5 shows
the geometry of the system considered and defines symbols used in the dis-
cussion. A jet of density Po issues at a velocity of Uo from a circular
nozzle of diameter do. The ambient fluid is at rest and of density Pl'
Abramovich20analyzed the trajectory of a heated jet issuing into a cold
ambient fluid and compared his theoretical result with the data of Syrkin
and Lyakhovskiy.23rhe resulting expression is:
I,
I
d P - Po 3
(Zd) = O. 052 (~) ( ,a ) (x)
.. 2 Po d
Uo
(IV-9)
'Fi.gure IV-~;shows a comparison of this expression with experimental
which the ratio (p a - Po)/Po was varied in the approximate range of
0.8. Equation (IV-9) generally underestimates the buoyancy induced
of: the jet.
data in
0.2 to
deflection
, 19
Field et. a1. also considered the behavior of a buoyant jet and obtained
the expression:
IV-16
Arthur D Little, Inc
-------
y
H
<:
I
......
"
..:~J)If~!lltl~ljt?
Jet Properties:
Nozzle Velocity Uo
Density Po
Ambient Fluid Properties
At Rest
Density P
a
»
:4
:r
c
.,
o
C'
.....
.....
ib
S-
f)
FIGURE IV-5
SCHEMATIC OF JET FLOW
jet NOZz\e ~)(.is
- ----
x
9
-------
H
<:
I
.....
00
~
~I
-eo 0
Q. Q.
~
~.300
o I Q.
Q.'
CO
~
,--....
~:3' 1"& 200
""'--"'"
>1"0
»
:4
::r
c:
~
o
r-
;:;
-
in
R
600
600
Q!I
o
/
/.
If
I II
,
"lIP /
V IIiI
V Field et. al. Theory
~ Eqn (IV-1O)
I I I
500
500
400
400
~I~
Q. , .
co
Q..
~300
""'--"'"
>1"0
e
200
100
100
, Abramovich Theory ,
Eqn (lV-9)
o
o
o
16 18 20
o
2
4
6
8
10 12
x/d
14 16
18
20
2
8
4
6
10 12
x/d
14
FIGURE IV-6
r
COMPARISON OF THE PREDICTIONS OF ABRAM_OVICH20 _AND FIELD ET. AL.19
WITH THE DATA OF SYRKIN AND LVAKHOUSKy23ON BUOYANT JET BEHAVIOR
-------
() P - Po P 1/2 3
(Y...d) = (.!d) tan a 0 + . 0.047 (.&L) ( a ) (~) (.!d)
cos ao. 2 Po Po
uo
(IV-lO)
For a jet injected normal to the gravity field (ao = 0), Equation (IV-IO)
reduces to
d P - P Q P a 1/2 3
(Y...d) = O. 047 (~2) ( a ) (-) (x d)
Po po
uo
(IV-lOa)
which differs from Equation (IV-9) in the value
(0.047 as opposed to 0.052) and the presence of
of the leading constant
the term (p /po)1/2. In
a
incinerator applications, where the jet and ambient temperatures are
approximately 100°F (5600R) and 2500°F (3060°F), respectively, this term
has the value of
1/2
(p ip 0)
1/2
= (To/Tl)
1/2
= (560/3060)
= 0.43
The deflections predicted by the two equations will differ by a factor of
two.
of Equation (IV-lOa) with the data of
data, the term (p /Po)l/~ varies from
the term (p /Po)l?2 results in better
a
agreement with the data, particularly at the larger values of x/do.
Figure IV-6 shows a comparison
23
Syrkin and Lyakhovskiy. In these
about 1.1 to 1.4. The inclusion of
3.
CROSSFLOW EFFECTS
The need to understand the behavior of a jet issuing into a crossflow
normal to the jet axis arises in the analysis of furnaces, plume dispersion
from chimneys and elsewhere. The deflection of the jet by the crossflow
has been studied extensively, both experimentally and analytically, although
most workers have lim:lted their work to descriptions of centerline trajectory
and gross entrainment rates. To our knowledge, no analysis has been carried
out, either experimentally or theoretically, which characterizes in detail
the radial distribution of velocity or concentration.
With cross flow, the interaction of the flows deflects the jet and alters
the cross-sectional shape of the jet. Figure IV-7 is a diagram of the jet
cross-section several nozzle diameters along the flow path. The originally
IV-l9
Arthur [) Little.lnc
-------
j----
---~
---~
FIGURE IV-7
JET CROSS SECTION AND CIRCULATION
PATTERNS FOR ROUND JETS IN CROSS FLOW
IV-20
Arthur D Little, Inc
-------
-- - ~--- -"'- .~~ --~- -----"~-~---
circular cross-section has been distorted into a horseshoe shape by the
shearing action of the external flow around the jet, and internal patterns
of circulation have been set up. Measurements in the external fiow around
the jet show a decreased pressure downstream of the jet, recirculation of
the external fluid, and a process leading to the periodic shedding of
vortices into the wake of the jet. These phenomena are similar to those
obsE~rved in the wake of a solid cylinder exposed to cross flow.
Dimensional analysis considerations suggest that the coordinates of
the jet axis (x/do, y/do) should depZnd on the ratio of momentum fluxes in
P ul P ouodo
the external and jet flows M = (a 2) and the Reynold~ number Re = ( ) .
IJo
PoUo
4
For turbulent jets in the Reynold's number range above 10 , correlation
of experimental data suggest that the Reynold's number effect is negligi.ble
and the momentum ratio is the predominant variable characterizing the fl~~.
In terms of the geometry illustrated in Figure IV-8, the following
expressions for computing the axial trajectory of a single jet have been
repClrted. The jet axis is taken to be the locus of maximum velocity.
(t) = 1.0 (M)1.12 (~o)2.64
Patrick24 (1967)
o < M < .023, ~o = 0
(IV-ll)
. Reference:
2.55
(1-) = M x
do d
x
+ (1 + M) [tan ooHd::-]
.046 < M < 0.5
(IV-12)
Reference:
Shandorov25 as cited in Abramovich20
3
(1-) = (M)1.3 (x) +
do d
x
[tan oO][d:-]
17 20
Ivanov as cited by Abramovich
.001 < M < 0.8
(IV-B)
Reference:
+ + 2.175
(y~~ ) = 5. 5 ~. 175 (x~: )
0.01 ~ M ~ 0.028, ~o = 0
(IV-14)
where y+/d and x+/d denote the end of a zone of establishment.
vary somewhat with M but are of the order of one or less.
Values
Reference:
26 .
Keffer and Barnes as cited by Field et. a1.
IV-21
Arthur D Little, Inc
-------
",-
",
,.,..
Jet Flow
-
Velocity::; u 0
Density::; P 0
~~
T ",- CXo
,,+ - - ..L
do
-*-
t
t
~
~
--
~
FIGURE IV-8 COORDINATE SYSTEM FO~ ROUND JET IN CROSSFlOW
t
Ambient Flow
Velocity = ~
Density = Po
IV-22
r
~ x
g
Arthur D Little, Inc
-------
Abramovich20 derived an analytical relation of the form:
1 1/2
(~) = 14.4 MC) log [1 + 0.1 (t) (1 +
do x
do
1 + 20 -]
y
(IV-15)
where C is an effective drag coefficient relating drag on the jet to the
moment~ flux in the external flow.
A simplified treatment of ~et behavior in crossflow with temperature
effects was presented by Davisl in 1937. The crossflow effect was intro-
duced by the assumption that tha jet rapidly acquired a velocity component
equal to the crossflow velocity. The jet was then seen to follow a path
corresponding to vector addition of the crossflow velocity to the jet center-
line velocity (the laf~er being calculated using a simplified velocity
decay law by Tollmien ). Temperature effects were introduced as being
reflected in increases in jet velocity due to expansion of the cold nozzle
flu:ld (initially at T ) after mixing with the hot furnace gases (at T ).
o s
Davis' final equation for the deflection (y) is given by:
y =
ulx(x + 4docos ~o)
2
2aldouo cos 0.0
To 1/3
[~] + tan 0.0
s
(IV-l6)
where "a?" is a constant depending on nozzle geometry (1.68 for round jets
and 3.15 ~for long, narrow plane jets). The many rough assumptions in Dav:ls'
analysis (some of which have been shown to be in error) would indicate that
its use should be discouraged. Comparison of calculated trajectories witl1 ob-
served flame contours, however, (Figure IV-4) suggests it may have some general
value. Interpretation of the meaning of the general agreement between calculated
jet trajectory and flame contour as shown in Figure IV-4 is difficult, however, and
use of the Davis equation in incinerator applications is uncertain.
24
Patrick reported the trajectory of the jet axis (defined as the maximum
concentration) to be
(L = L 0 Ml. 25 (~ ) 2 .94
do do
(IV-l7)
Figure IV-9 shows a plot of the velocity axis [Equation (IV-12)] and the
concentration axis [Equation (IV-17)] for a value of M = 20. The concen-
tration axis shows a larger deflection than does the velocity axis. This
is probably due in part to the assymetry of the. external flow around the
partially deflected jet. Also, recent calculations by Tatoml5 for plane
jets suggest that under crossflow conditions, streamlines of ambient fluid
can be expected to cross the jet velocity axis.
IV-23
Arthur D Little.lnc
-------
o
~ 100
>-
200
/
/
concentratioj
Axis
/ /
I /
/
// / Velocity
/ Axis
h V
/ V
-----
~
180
160
140
120
80
60
40
20
o
o
2
8
4
6
10
x/do
12
14
16
18
20
FIGURE IV-9
TRAJECTORY OF CONCENTRATION AND VELOCITY AXES FOR
JETS IN CROSS FLOW (-DATA OF PATRICK24_) M = 0.05
IV-24
Arthur D Little.lnc
-------
For our purposes, we are interested in jets which penetrate reasonably
far into the crossflow (i.e., those which have a relatively high velocity
relative to the crossflow). The empirical equations of Patrick24 [Equation
(IV-II)] and Ivanov17 [Equation (IV-13)] were developed from data which
satisfy this condition. Figures IV-lO and IV-II show comparisons of these
two equations for values of M of 0.001 (uo/ul = 30) and M = 0.01 (uo/ul = 10).
Ivanov's expression predicts higher deflections at large (x/do, particularly
at M I:: 0.01.
17
Ivanov also investigated the effect of the spacing between jets in
a linear array on jet trajectory. He measured the trajectories of jets
under conditions where M = 0.01 at spacings of 16, 8 and 4 jet diameters.
His results are shown in Figure IV-12 along with the trajectory of a single
jet (infinite spacing). The data show that reducing the spacing between
jets causes greater deflection of the jets. As spacing is reduced, the
jets tend to merge into a curtain. The blocking effect of the curtain im-
pedes the flow of external fluid around the jets and increases the effective
deflecting force of the external fluid. The increase in deflection is most
notable as s/do is reduced from 16 to 8. Above a/do = 16, the merging of
the jets apparently occurs sufficiently far trom the nozzle mouth to have
little effect on the external flow. At s/do = 8, the jet merger apparently
takes place sufficiently close to the nozzle mouth that further reduction in
has little added effect.
spacing
20
Earlier data (Abramovich ) on water jets colored with dye issuing into
a con'fined, cross-flowing stream was correlated in terms of jet penetra-
tion distance. The penetration distance, Lj' was defined as the distance
between the axis of the jet moving parallel to the flow and the plane con-
taining the nozzle mouth. The axis was defined as being equidistant froDl
the visible boundaries of the dyed jet. The resulting correlations was
~I:: k~
do ul
where k is a coefficient depending on the angle of attack and the shape of
the nozzle. Defining the angle of attack (S) as the angle between the jet
and the crossflow velocity vectors, ([90 - ao] in the terminology shown in
Figure IV-8) , the recommended values of k are:
(IV-18)
For a
=
90°, for round and square nozzles; k = 1.5
For a =
90°, for rectangular nozzles; k = 1.8
For a = 120°, for all nozzles; k I:: 1.85
Figure IV-13 shows a comparison of Ivanov's correlation [Equation (IV-13)]
with the jet penetration correlation [Equation (IV-18)], for M = 0.001 and
M = 0.01. Equation (IV-18) 'predicts a smaller jet penetration than does
Equation (IV-13). There are two possible explanations for this discrepancy.
IV-25
Arthur D Little, Inc
-------
o
. ~ 2
>
4
:-1
o
2
--
4.
6
8
10
14
18
20
12
16
x/do
oPatrick24 Eqn IV-11
Olvanov 17 Eqn IV-13
22
28
30
24
26
FIGURE IV-10
COMPARISON OF TRAJECTORIES AT M= 0.001
. (UO/UI = 31.6) FOR JETS IN CROSSFLOW
IV-26
Arthur D Little.lnc
-------
o
~
70
/
v
I
/
/ 1/
t /
/
I
/ I
V
/
~ v
~
~ V
~ 0 Patrick24 Eqn (lV-11)
--~ P A Ivanov17 Eqn (lV-13)
,- I I I I
60
50
40
30
20
10
o
o
2
4
6
8
10 12
14 16 18
(x/do)
20 22 24 26 28 30
FIGURE IV-11
COMPARISON OF :TRAJECTORIES AT M = 0.01
(uolu1 = 10) FOR JETS IN CROSS FLOW
IV-27
Arthur 0 Little.lnc.
-------
30
20
o
~
~
10
o
o
FIGURE IV-12
48
s/do
16 00
M = 0.01
10
20
(x/do)
EFFECT OF JET SPACING ON TRAJECTORY
FOR JETS IN CROSS FLOW (AFTER IVANOV 17)
IV-28
30
Arthur D Little.lnc
-------
a.
The data on which Equation (IV-13) was based do not extend
to large values of x/d.,and extrapolation of the data may be
in error.
b.
The data on which Equation (IV-18) was based were taken in
a confined cross flow in which the lateral dimension (normal
to both the jet axis and the cross flow) was sufficiently
small to interfere with normal jet spreading. The jet
effectively filled the cross-section in the lateral dimension,
behaving like a series of jets at low spacing.
Ivanovts data (Figure IV-12) show that penetration is reduced at lower
spacing. The penetration given in Equation (IV-13) for M = 0.01 was reduced
by 25% (see the dotted trajectory marked s/do = 4, M = 0.01 in Figure IV-13).
Agreement between this adjusted trajectory and the penetration given by
Equation (IV-18) is better.
4.
BUOYANCY AND CROSSFLOW
When a cold air jet is introduced tnto a crossflowing combustion cham-
ber, both buoyancy and crossflow forces act simultaneously on the jet.
20
Abramovich reports the results of experiments conducted by inject-
ing cold jets into a hot cross flow. Temperature ratios of as much as 3
to 1 were used (corresponding to the jet fluid having a density three
timeS that of the cross flowing fluid), with the values of M in the range
of 0.045 to 0.5. The normal crossflow trajectory equation correlated the
data when the value of M was computed using actual fluid densities. From
theBe data, Abramovich concluded that buoyancy effects could be neglected,
other than as density differences were incorporated into the crossflow parame-
ter M.
17
The same conclusion was drawn by Ivanov who injected hot jets into
a c()ld crossflow. The ratio of temperature (and density) between jet and
ambient fluids was 1.9 and M ranged from 0.005 to 0.02. The geometry of
the tests was not clearly stated by either Abramovich or Ivanov. It appears
that the buoyancy force acted in the same direction as the crossflow force
in Ivanov's tests (i.e., a hot jet discharging into an upflow).
Application of these conclusions to incinerator design practice, however,
1s Elubject to question because of the large geometrical scale-up involved.
Physical reasoning suggests that the ratio of buoyant force to drag force
act:l.ng on a non-isothermal jet in crossflow depends on scale. The buoyant
for(~e (B) is a body force and is, therefore, proportional to jet volume.
The drag force (D) exerted by the crossflow has the characteristics of a
surface force and is, therefore, proportional to the effective cylindrical
areEl of the jet. Therefore, the ratio of buoyant force to drag force is
proportional to jet diameter. A simple analysis, discussed in Section D,
gives:
IV-29
Arthur 0 Little, Inc
-------
50
10
'M = 0.001 "
1
40
30
o M = 0.01
~
20
o
o
10
20
30
40
50
60
70
, (x/do)
FIGURE IV-13
COMPARISON OF IVANOV'S TRAJECTORY CORRELATION17 (Eg" IV-13)
WITH JET PENETRATION CORRELATION (Eg" IV-18)
IV-30
Arthur D little.lnc
-------
Po
~ = 2~ (&) (1 - -)
x U12 P 1
(IV-l9)
where C is the effective drag coefficient.
The value of the effective drag coefficient (C ) is believed
to be in the range of 1 to 4; analysis given in Secfion D suggests that 4
is the better value.
17
Typical values of the physical parameters in Ivanov's experiments are:
do = 5 to 20 mm (0.0164 to 0.063 ft.)
Uo/u1 = 10 to 20
To /T 1 = pip 0 = 2
- 2 2
Pouo /p aUl = 100 to 200
u1 = 3.68 to 4.16 m/sec (12 to 13.6 ft/sec)
The maximum value of B/D results from the maximum value of nozzle
diameter d and the minimum value of crossf1ow velocity u. Substitution
of these values into Equation (IV-19) yields a force ratto of 0.0026 which
indicates that the crossflow effect completely dominated the buoyant effect
on the small jets used in Ivanov's tests. Therefore, we speculate that his
results (no buoyancy effect) could be anticipated under his test conditions.
In incinerator applications, jet diameters in the neighborhood of 4
inches are contemplated, along with crossf1ow velocities in the order of
2 to 5 feet per second. For a 4 inch jet in a 2 feet per second crossflow
with the same 1.91 density ratio as in Ivanov's tests,
! - (3.14)
D - (2) (4)
(32.2)(0.033)
(2)2
1
(1 - 1.91)
= 0.50
which suggests that the buoyant and crossf1ow forces are of the same order
of magnitude.
5.
DESIGN METHODS
The correlations given above provide the basic tools for the analysis
of jet behavior in real furnace environments. In general, the correlations
are based in theory and corroborated with data. Translation of behavioral
relationships into designs can be approached in two ways:
IV-3l
Arthur 0 Little, Ine
-------
Detailed analysis of the actions and interactions of each
component of the system t.mder design, "building" an t.mder-
standing of system behavior from an t.mderstanding of its
parts; and
.
Assembly of generalized correlations into "rules-of-thumb"
and the like which show applicability to a number of systems
similar to the device in question.
.
The jet design correlations above, when coupled with the bed burning
and chamber flow analyses, are supportive of the first approach. As such,
they are broadly applicable but their use makes demands upon the designer
for data and t.mderstanding which he may not possess.
Approaches to a "rule-of'7thumb", generalized method for overfire air
have been proposed by Ivanovl and by Bituminous Coal Research, Inc.28
Although these design guides were developed for coal-fired boilers, they
are presented here as an indication of approaches successful in other
applications. Their applicability to incineration systems, however, has
not been shown.
a.
jet design
The Method of Ivanov--Ivanov conducted a number of experiments
in a non-combusting model furnace to determine the effects of
various jet configurations on the temperature profiles above
a burning coal bed. He concluded that:
.
It is preferable to position overfire jets in the front
wall of the furnace rather than in the rear wall.
.
Close spacing of the jets is desirable
jets form an effective curtain. Above
rotary motion of the gases is induced,
greatly to the mixing process.
in order that the
this curtain, a
which contributes
.
If maximum temperatures occur near the center of the grate,
rather than near the front, the design depth of penetration
of the jet should be increased by 5%.
.
Slightly better mixing is obtained if jets are fired from
one wall, rather than if the same flow is divided between
jets on opposite walls. This applies whether the opposing
sets of jets are directly opposed, staggered but on the same
horizontal level, or on different horizontal levels.
.
A given level of mixing is achieved at lower power cost and
with introduction of less air if small diameter jets are
used rather than large ones.
IV-32
Arthur D Little, Inc
-------
For conditions which gave good mixing, Ivanov computed
the jet penetrations from Equation (IV-20).
Uo
Lj = dok (-)
u1
(IV-20)
and normalized the values obtained with respect to LT' the
axial length of the model furnace.
Using the values of k given on page IV-25 to compute L.
he correlated hiLT with the relative jet spacings "s/d~ J
(where s is the center-to-center jet spacing and do is the jet
diameter) and obtained the following values:
(s/do)
4
5
6
Front Arch
Furnace
~lbrL
0.90
0.95
1.10
Rectangular
Furnace
~j1br-
0.80
0.90
1.0
This correlation is the basis for his design method.
Ivanov's design method is as follows:
(1)
Nozzles should be located not less than 3 nor more
than 6 1/2 feet above the fuel bed.
(2)
The angle of inclination of the jets is determined
by aiming the jets at a point on the grate 4-6 1/2 feet
from its far end. Jets fired from the underside
of a front arch may be angled downward as much as
500 from the horizontal, if the fuel bed is not
disturbed by the resulting jet.
(3)
The relative jet spacing should be in the range of
sid equals 4 to 5.
(4)
The velocity of gases in the furnace at the cross-
section where the jets are located is computed
from known overall air rates and grate areas and
corrected for temperature. This velocity is the
crossflow velocity u1 and the density is P a.
The jet velocity is set by the capability of the
overfire fan, but should always be 200 feet per
second for cold jets and 230 feet p~r second for
heated jets. He assumes a fan outlet pressure of
about 14 in w.g. and computes the jet velocity from:
(5)
IV-33
Arthur 0 uttle, Inc
-------
- - J 2gP
Uo - ( 1. 2) P 0
(IV-2l)
where Po is the density of the nozzle fluid.
(6)
The required jet diameter is computed from Equation
(IV-22):
do =
Lj
\10 [P:
k-J~
ul Pa
(IV-22)
where k is 1.6 for s/do = 4-5, and L is taken to be
a factor times LT' the axia~ length df the furnace.
the factors were given on page IV-33 and range from
0.8 to 1.10.
(7)
The number of nozzles in the row (N) is then calculated
from the furnace width (B) and the jet spacing (s)
according to:
B-4s
N=-
s
(IV-23)
(8)
The required fan capacity is then computed from:
2
1Tdo -
Q = N ~ Uo
(IV-24)
b.
The Bituminous Coal Research (BCR) Method--The National Coal
Association has published a design handbook for "Layout and
Applica~ion of Overfire Jets for Smoke Control", based on work
by BCR. 8
The NCA recommends:
.
Side wall placement;
.
Location of nozzles about 18 inches above the fuel bed
in modern furnaces and from 9 to 12 inches above the bed
in older, small furnaces;.
.
Introduction of from 10% to 30% theoretical air via jets
depending on whether the smoke formed is "light" or "heavy."
IV-34
Arthur D Little,lnc
-------
The design method is as follows:
(1)
Read the required volume of air (cfm per lb coal
burned per hour) from a table, given the heating
value of the coal and whether the smoke is light,
moderate or heavy. Compute the air requirement
in cfm.
(2)
Decide where nozzles will be located (front, side,
or back wall).
(3)
Read the number of nozzles required from a table,
given the dimension of the wall on which the jets
are to be located and the penetration distance
(equal to the axial dimension of the furnace).
(4)
Compute the air requirement per nozzle by dividing
the result of Step 1 by the result of Step 3.
(5)
Read nozzle diameter and required fan pressure from
graphs, given the air requirement per nozzle (Step 4)
and the penetration distance (Step 3).
(6)
Determine duct size from a nomograph given total air
requirement (Step 1).
The design criteria on which this method is based are not
readily apparent. Examination of the tables and graphs included
in the references shows the following relationships.
.
.
The number of jets is approximately proportional to the length
of the furnace wall where the jets are installed and approxi-
mately inversely proportional to the penetration distance.
The penetration distance appears to be defined as that dis-
tance required to reduce the velocity of a jet, issuing into
a quiescent chamber, to 8 feet per second.
Several qualitative statements can be made. First, the cross-
flow velocity does not enter explicitly into the design method.
Second, working out several examples shows that relative spacings
(sId) of up to ten or more result. This is at odds with Ivanov's
finding that spacings of 4 to 5 jet diameters are optimal.
General Discussion
The design methods cited above apply to furnaces burning coal or shale,
and are generally used in boiler design. These applications are characterized
by:
IV-35
Arthur D Little, Inc
-------
.
A uniform and predictable fuel supply which burns in a regular
and repeatable pattern along the grate;
.
Use of high heating value fuel (in the range of 10,000 to 15,000
Btu/lb), with low moisture and ash content;
.
The desirability of minimizing excess air so that high combustion
temperatures and high heat recovery efficiency can be obtained;
.
Relatively low combustion volume per Btu/hr capacity.
In contrast, incinerators are characterized by:
.
A variable and generally unpredictable fuel
sition and moisture content very seasonally
predictable manner and hourly (as fired) in
manner;
supply; the compo-
in a somewhat
an unpredictable
.
Use of low heating value fuel (4450 Btu/lb average as fired),
with relatively high ash (20%) and moisture (28%) contentl;
.
No general requirement for high combustion gas temperature,
except in heat recovering incinerators;
.
Relatively large combustion volumes per Btu/hr capacity.
In both cases, complete fuel burnout is desirable and combustion gas
temperatures must be kept below the point where slagging or damage to the
refractory occurs. Both types of units have fly ash problems, although
the incinerator problem is more severe since relatively large pieces of
unburned paper can be lifted into the combustion volume.
The differences in characteristics place different requirements on
the overfire jets. Jet systems in incinerators must contend with:
.
A shifting combustion profile caused by variations in the up-
flow gas temperature, composition and velocity, and in the moisture
content and composition of the fired refuse;
.
Large pieces of partially burned refuse in the combustion volume;
.
Large combustion volumes per Btu/hr. which increases difficulty
of mixing the combustion gases.
In meeting these conditions, minimization of excess air introduced in
the jets is not as important as in heat recovering boilers. The principal
factors which mitigate for low excess air in incinerators are draft limi-
tations, higher costs of air pollution control, fan and stack equipment,
power costs and the general requirement that the overfire air not quench the
combustion reaction. Although these factors are important, realization of
complete combustion of pollutants,materials survival and inhibition of
slagging are predominant concerns.
IV-36
Arthur D Little.lnc
-------
D.
CONTRIBUTIONS TO THE ART
The published methods for designing overfire jet systems are directed
toward coal-fired boilers which fire a well-defined fuel and are generally
more square in cross-section than are incinerators. In addition, the de-
sign methods ignore the buoyancy effects associated with the introduction
of cold jets into the combustion space. The buoyancy effect was neglected,
however, on the basis of laboratory data, where the jet diameter was
sufficiently small that buoyancy effects would be expected to be negligible.
In larger scale systems, this buoyancy effect can be important~
The variable nature of refuse and the bed-burning processes on the
temperature and concentration profiles within the incinerator suggests that
overfire jet systems in incinerators should have sufficient flexibility
to meet the moment-to-moment variations in the location of combustion vol-
ume segments which require overfire oxygen or turbulence.
The high aspect ratio of incinerators (large grate length to width
ratio) takes the design of end wall jet systems into a region of jet be-
havior where relatively little data exists. Few measurements have been
made of jet behavior at large (several hundred) values of normalized axial
distance (x/d).
In building on what is known in the design of overfire air jet systems
for incinerators, we are cognizant of the required system flexibility.
Our studies of axial temperature and concentration profiles above the fuel
bed and the nature of the flow patterns in the combustion space have enabled
us to estimate the degree of variability to be expected.
Accumulation of additional experimental data to quantify the behavior
of jets at large distances from the nozzle mouth and under conditions where
buoyant effects might be significant was beyond the scope of this study.
We have attacked the buoyancy problem analytically in an attempt to define
those sets of conditions where buoyant effects become important and have
run some simple qualitative experiments to shed further light on the prob-
lem. The results of these studies are discussed in the following sections.
1.
MATHEMATICAL MODELING OF COMBINED BUOYANT AND CROSSFLOW EFFECTS
Figure IV-8 defines the geometry on which our evaluation is based. This
first step in analysis is concerned only with jet behavior very near the nozzle
mouth. The objective is to establish the relative magnitude of the buoyant
and drag forces acting on the jet.
The mathematical model is based on a segment of the jet dx in length.
The buoyant force (B) on this differential volume element is:
2-
dB = g (p - p ) Crdo) dx
a 0 4
(IV-25)
IV-37
Arthur D Little, Inc
-------
The drag force (D) is computed by analogy with the way in which the
drag on a solid cylinder is computed,
1 2
dD = (zPaul ) Cx (do)(dx)
(IV-26)
where C is the drag coefficient.
x
The ratio of these forces F is:
dB Po - P
F - - - (21TC ) (~) ( a
- dD - x 2 Pa
ul
(IV-27)
The significance of this relationship is that:
If F > 1, buoyant forces predominate (the cold jet sinks).
If F < 1, drag forces predominate (the jet is "blown away").
Comparison of this type of drag force model with experimentally measured
deflections of jets in crossflow suggests that the value of Cx lies in the
range of 3 to S. Using Abramovich's correlation of the deflection data20
yields Cx = 4.75. Therefore, Equation (IV-3) becomes:
F = 0.33
gdo
(-)
2
ul
Po - Pa
(
Pa
(IV-28)
Equation (IV-28), with F = 1 defines the functional relationship between
the crossflow velocity (ul) and jet diameter (do) for given furnace gases
and jet fluid properties for which buoyant and drag forces are equal. A
plot of this function (Figure IV-14) divides the do-ul space into two areas
of buoyant and drag domination. In Figure IV-14, the temperatures of the
gas and jet fluids are taken to be 2000°F and 100°F, respectively. Differ-
ences in molecular weights are neglected.
Figure IV-14 shows, for example, that with a jet diameter of 3 inches,
drag forces predominate if the cross flow velocity is greater than 3 feet
per second. This plot serves as a rough design tool. The region to the
left and above the line should be recognized as a design regime where the
predominance of buoyant forces may cause the jet to "fall into" the bed
and disturb the fuel distribution or create hot spots. As one moves away
from the line down and to the right, drag forces become increasingly pre-
dominant and the design assumptions of Ivanov more applicable.
In an attempt to compute the trajectory of a jet subjected to both
drag and buoyant forces, we expanded the simple model to include the effects
of spreading and change in direction which the jet experiences in cross flow.
1V-38
Arthur D Little, Inc
-------
14
13 Fluid Temperatures - Jet - 100°F
Furnace Gas - 20000F
12
11
10
en 9
Q)
.l:
U
c:
:£ 8 ~'"'
~
... ~
Q)
... 7 ~'
Q)
E Buoyant Forces ~c"
.!!! Predominate ~
0 6 "
...
-------
This mathematical model is based on conservation of momentum in the x and
y directions. In the axial (x) direction, this amounts to simply the con-
servation of the momentum in the jet at the nozzle mouth. In the cross-
flow (y) direction, the effects of both buoyant and cross flow forces on
the initial momentum are included.
The complete derivation of the model is given in Appendix E; the re-
sulting trajectory of the jet centerline is given by:
2.25 C
(t) = (~o) tan ao + ( 1TCOS :0)
2
P u
, a 1 )
" - 2
PoUo
2
(~o) + [( 3~ ~o:ao)
2
P u
( a I )
- 2
PoUo
(IV-29)
3 gdo Po - P 3
- (~) (- 2) ( Po a)] (~o)
Uo
where:
C is the nominal drag coefficient relating crossflow drag on the
x
jet to systems parameters.
c is the rate of spread of the jet in the lateral dimension (h):
where h = 2.25 d + c~.
~ is the path length along the jet axis.
y is equal to sec S, where S is the angle between the jet center-
line and the vertical; y is equal to csc ao at the nozzle mouth
and increases without limit as the jet becomes parallel to the
crossflow.
For the case where ao = 0 (the jet is introduced normal to the cross flowing
stream), Equation (IV-29) reduces to:
2.25 C 2 2 c C y 2 3 gdo
P u P u
(1.) = ( x) ( a 1 ) (~) + [( x ) ( a I) (f9:2) (-) (IV-30a)
d 1T -2 do 31T -2 -2
PoUo P oUo Uo
Po - Pa 3
(Po )] (~)
do
In ()rder to evaluate Equation (IV-30a) for a particular case, average
values for y, c and C are required. We are first concerned with behavior
x
when the jet is close to the horizontal, in order to establish whether the
drag or buoyant forces predominate. In this region, a '" ao, and y = l.
=
IV-40
Arthur D Little,lnc
-------
For a rectangular jet introduced into a quiescent
Recent data29,24 suggest that the spreading rate for a
greater than for a straight jet. Based on these data,
0.32.
volume, c = 0.22.
deflected jet is
we estimate c =
As before, we assume that C = 4.75.
x
Substitution of these values into Equation (IV-30a) yields:
(Y) - Mf3.4X2 +
p - P 1
[(0.16) - 0.052 (gd2) (0 a)] x3J
u Pa
1
(IV-3l)
where:
Y = y/do
X = x/do 2
P u
M=(al)
- 2
Pouo
This model does not allow for s~mple comparison of drag and buoyant
forces due to the presence of both X and X3 terms. The coefficient of
X2 is always positive (since it arises from the crossflow drag term). The
coefficient of X3 may be either positive or negative depending on the value
gdo P - P
o a
of the group (~) ( ) which has the same form as occurred in the
u Po
1 3
simple model. If the coefficient of X is negative, the slope of the
trajectory will become negative at sufficiently large values of X; the
mort~ negative the coefficient, the smaller the value of X at which this
occurs.
The efficiency of the model is difficult to evaluate because of the
assumptions inherent in the derivation and the empirical parameter re-
lationships which are used. The drag force concept is somewhat artificial
at best in this application, and is increasingly suspect as the jet is
deflected and distorted in the crossflow. The empirical relation be-
tween jet spread and axial distance from the jet mouth is only applicable
for some 6 or more jet diameters from the jet mouth. Experimental data
is needed to test the model, and to serve as a basis for development of
reliable design correlations.
2.
QUANTITATIVE MODEL EXPERIMENTS
In order to shed light on the behavior of buoyant jets in crossflow,
we conducted experiments using the apparatus shown in Figure D-l in
Appendix D. A jet of cold nitrogen gas at essentially the boiling point
of nitrogen (139°R) was injected horizontally into the test section. Air
at room temperature was blown upward in the test section so that the buoyant
and cross flow forces on the jet were in opposition. Water vapor condensed
IV-4l
Arthur 0 Little, Inc
-------
when the cold jet mixed with the humid
to be observed. A grid of strings was
measurement of jet deflection. A more
equipment is given in Appendix D.
room air, allowing the jet path
arranged on 6" centers to allow
detailed discussion of the test
Runs were made using two nozzle diameters at several combinations
of jet and cross flow velocities. Conditions for the runs are shown in
Table IV-2, along with the calculated values of the buoyancy/drag force
ratio [Equation (IV-28)].
TABLE IV-2
CONDITIONS FOR BUOYANCY-CROSSFLOW MODEL TESTS
Nozzle Crossflow Jet
Run No. Diam. (in) Velocity (fps) Velocity (fps)
1 0.469 1. 31 40
2 0.469 1.31 8.1
3 1.063 1.31 2.5
4 1.063 1. 31 1.3
5 1. 063 2.10 2.5
F*
0.66
0.66
1.49
1.49
0.58
* Force ratio as defined in Equation (IV-4)
Figures IV-IS to IV-19 show the jet trajectory for each of the runs.'
In runs 1, 2, and 5, in which the buoyancy/drag ratio was less than unity,
the jet shows no tendency to drop. In runs 3 and 4, where the ratio was
about 1.5, buoyancy forces cause the jet to drop. While these tests are
not sufficiently comprehensive to prove the validity of the force ratio
parameter in determining when buoyancy effects are important, the data
do support the contention that F ~ 1 is a valid criterion to identify con-
ditions when jet sinking can be anticipated.
IV-41a
Arthur D Little. Inc.
-------
FIGURE IV-15
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 1
IV-42
Arthur D Little.lnc
-------
H
<:
I
~
W
»
..,
~
::r
c:
..,
o
c
~
~
.ro
::J
o
FIGURE IV-16
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 2
i
I .
i
. .-- 1
FIGURE IV-17
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 3
-------
»
...,
.....
::r
t:
...,
o
c
.....
.....
,(b
:J
(I
H
<:
I
.p.
.p.
FIGURE IV-18
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 4
FIGURE IV-19
PHOTOGRAPH SHOWING OBSERVATIONS FOR TEST 5
-------
3.
COMBUSTION EFFECTS
The analysis in Chapter II indicated that under all conditions the
flow of air through a refuse or char bed will produce gases containing
unburned combustible. Although bypassing (channeling) of air through the
bed can provide some of the needed oxidant to the fuel-rich gases, the
data by Kaiser referenced in Chapter II indicate that the gases, on the
average, remain fuel-rich. As a consequence, it should be expected that
some combustion will occur as the oxygen-bearing overfire air jet penetrates
the hot gas flow rising from the bed. This results in a so-called "in-
vert(~d flame" where a jet of oxidant discharges and burns in a fuel-rich
environment.
This phenomenon has been observed in both coa~- and refuse-burning
practice where impingement of a jet on the opposite sidewall has resulted
in refractory overheating and slagging contributing to premature wall
failure. In another case, a jet of air moving beneath a long arch over the
discharge grate of an incinerator furnace yielded temperatures of over
2500('F in the brickwork. Because of the potential importance of this com-
busUon effect, a simplified mathematical model of jet behavior under these
conditions was developed and the effect of the pertinent variables was ex-
plorE!d.
The analysis makes use_of jet concentration correlations describing
the axial (c ) and radial (c) weight concentration of nozzle fluid as func-
m
tions of the distance from the nozzle plane (x) and the radial dimension
(r). The concentration at the nozzle is Co and the nozzle diameter is do.
The ambient (Tl) , nozzle fluid (To), and mixture (Tm) temperatures are
those prior to combustion. For non-combusting jets, in the absence of
crossflow and buoyancy, these variables are related by:
T 1/2 do
c = 5 Co (..1.) (-)
m To x
2
c = c exp (-57.5 [E] )
m x
(IV-32)
(IV-33)
Assuming equal and constant specific
fluid, an energy balance yields:
T 1/2
T = T + 5 (..1.)
m 1 To
heats for the nozzle and ambient
do 2
(To - Tl) X- exp (-57.5 [~] )
(IV-34)
Using Equation (IV-34), we can calculate the mixture temperature. Then,
by co:mparison with an assumed minimum ignition temperature (TF) (say 1100°F),
it caa be determined whether or not combustion will occur. (Note that this
shows "quenching" of combustion in the cold core of the jet.)
IV-45
Arthur D Little, Inc
-------
From the analysis method described in Chapter II, the oxygen demand (A
pounds of oxygen per pound of ambient fluid) and the heat of combustion
(H Btu per pound of oxygen reacting) of the furnace gases can be estimated.
De~ining 0 as the concentration of nozzle fluid in the mixture relative
to therozzle concentration (c/co), we find that if T >T and if A (1-0)
-0 co~O, co~bustion will occur to the extent of the ~~aIlable oxygen re-
leasing HoOco/A Btu per pound of mixture. The resulting gas temperature
(T ) is given by:
c
Tc ::: TR + ~p [(1 - 0) (Tl - TR) cp + Hyco/A]
(IV-35)
Where TR = the reference temperature for enthalpy (say 60°F) and cp = the
average specific heat of the gases between the reference temperature and T .
c
For the oxygen-rich case, if T ~T and if A(l - Q)- Qco
-------
'""'
LL
-
~
W
o
'-'
H W
'f ~
.J:- :J
" .-
<
~
W
0...
~
W
~
»
..,
....
:::r
c:
..,
o
c
....
....
~
:J
(")
3000-
2500-
2000-
1.500-
1.000-
500..
!
0..
-4..
-3.
Distance From
Nozzle Plane (ft.)
8.0,-
10.5. .~
5.5, .
3.0
-2..
-1...
I
I
I
I
I
I
I
I
I
I
I
I
I
I Jet Axis
I
I
I
I
I
0..
2..
3-
1..
RADIAL DISTANCECFEET)
."
c;)
C
::D
m
<
I
II.)
o
-t
m
~
'"0
m
::D
»
-t
c
::D
m
<
m
::D
en
C
en
::D
»
c
»
,...
c
en
-t
»
2
n
m
c::
CD
..
c
ii)'
3
CD
..
CD
...
II
...
~
4-
-------
3000. r
Distance From
Nozzle Plane (ft.)
2500. 8.0
lO.~ "T1
5.5 I C)
I
3. I C
I :D
I m
I
I <:
J I
2000. I N
I ooA
u.. I
I
I -t
. I m
H ~ I s:
I
<: W I "a
J m
~ 0 I :D
J
OJ '-' I »
W 1.500. I -t
I C
~ I :D
:J I m
I <
I- I m
< I :D
I en
eJ I C
I en
I
0... I :D
~ 1.000. I »
I C
W I »
I- I
I ~
I C
I en
I -t
I »
I Z
I n
500. I m
I c:
I
I a
I C
I
:> I iij"
I 3
~ I CD
::r I Jet Axis ..
CD
C ...
.., I II
o O. I ~
C'
.... -4.. -3. -2.. -1... O. 1.. 2. 3. 4.
....
.ro
:J
0 RADIAL DI5TANCE(FEET)
-------
H
<
I
~
\0
»
...,
....
:r
c
...,
o
r
;:;:
....
1D
5'"
o
"""'
lL.
a
~
W
o
"'-'
w
(k:
:::J
I-
ffi
~
I-
3000-
2500-
2000-
1.500 -
1.000-
500-
r
o.
-4.
-3.
Distance From
Nozzle Plane (ft.)
1
-2.
1..
2.
-1.-
0-
RADIAL DISTANCE(FEET)
3.
'T1
...
w.
c:
::D
m
<
I
N
N
-I
m
s:
~
m
::D
»
-I
c:
:D
m
<
m
:D
en
c:
en
:D
»
o
i=
r-
o
en
-I
»
Z
(")
m
~
. CD
~
o
Di"
3
~
CD
...
II
~
4.
-------
The above provides confirmation of the occurrence of "blow torch"
effects from air jets and shows that temperatures near 2500°F such as
have been experienced with jets could be anticipated. Clearly, however,
the utility of the above is only to provide perspective as to the nature
of air jet behavior in incinerators. No allowance is made, for example,
of crossf10w effects which are known to increase jet entrainment rates
and thus "shorten" the inverted flame described by the analysis. The
method, therefore, can be expected to produce a conservative result.
24
The correlations of axial concentration by Patrick provide a means
to estimate the effect of cross flow in shortening the distance to the
point of completion of the combustion reactions. From Equation (IV-32)
an analysis readily shows the distance from the nozzle plane to the point
where the gas on the axis is at a stoichiometric ratio to be given by:
T1 1/2 Co
~ = 5 (-) (- + 1)
do To A
(IV-37)
Patrick found the centerline concentration to vary along the jet path
length "t" in crossf1ow according to:
Co
[(~) exp (7.8 Mt/2 - 1.856)]1.18
do
(IV-38)
c
m
and, for no crossflow according to
Co
1.18
0.112 (~)
do
(IV-39)
-=
c
m
Equating (IV-38) and (IV-39) establishes the relationship between
the centerline distance "x" for non-crossf10w which corresponds to the same
concentration ratio as for a jet in crossf10w which has traveled over a
path length "s."
x tl/2
do = 1.42 (d;) exp (7.8 ML - 1.856)
(IV-40)
The path length can be easily calculated by numerical integration of
Equation (IV-41) which also can be derived from Patrick's trajectory re-
1aUonships:
x/do 2 55 4 0.5
(~) = J [9 M' (~) + 1] d (~)
do do do
o
(IV-4l)
IV-50
Arthur D Little, Inc
-------
Therefore, to find the distance from the nozzle plane to the point
where such peak (stoichiometric) temperatures will be obtained on the
cent,erline, the non-crossflow distance is calculated from Equation (IV-37);
the resulting value is substituted into Equation (IV-40) to yield the
crossflow path length at an equivalent degree of mixing; and the integra-
tion given in Equation (IV-4l) is carried out to define the dimensionless
distance x/d integration limit which causes the integral to assume the value
of the calculated path length. This latter x/d value corresponds to the
horizontal distance from the nozzle to the plane where peak temperatures
exist. The vertical displacement of the jet in this plane may then be
calculated by substitution into Equation (IV-42).
3
(L) ::: Ml. 28 (~)
d. d.
The analysis shows that jet temperatures can be considerably elevated
by combustion effects. Therefore, when jet operation is desired in regimes
where buoyancy analysis (neglecting combustion) suggest jet drop would be
important, these effects could provide counterbalancing jet temperature in-
creases. The complete analysis suggesting the degree to which the buoyancy/
drag criteria could be slackened by consideration of the combined effects of
buoyancy and crossflow with combustion, however, was beyond the scope of this
analysis effort.
4. TENTATIVE INCINERATOR OVERFlRE AIR JET DESIGN METHOD
(IV-42)
Although little data is available to give specific guidance for in-
cinerator overfire air jet design, the prior art and the studies carried
out under this contract provide the basis for tentative guidelines and a
design methodology. Experiments anticipated in subsequent phases of this
program (Chapter VI) will be helpful in strengthening these arguments.
The basic parameters to be selected in design of an overfire air jet
system are:
.
The diameter (do) and number (N) of the jets to be used:
.
The placement of the jets;
.
The quantity of air to be overfired (QT)
Rela.ted but not independent variables are the jet velocity (u ), and the
a
head. requirements for the overfire air fan (P).
The tentative design method is based on that of Ivanov17 which was
disc.ussed in Section C-5. It is important in usin~ this method to comnare
the values of do and u obtained with Figure Iv-i4 to determine if buoyant
forces might be importlnt. Values should fall in the "drag forces predominate"
regi.on.
IV-51
Arthur D Little.lnc.
-------
The basic equations on which the design method is based are as follows:
Air Flow Relation
Q = 47 Nd~ Uo
T
(IV-43)
where:
QT is the overfire air rate (CFM)
N is the number of jets
do is the jet diameter (feet)
Uo is the jet velocity (fps)
Ivanov's Penetration Equation
L u. fio
~ = 1.6 (-) -
do ul PI
where: Lj is the desired jet penetration (feet)
(IV-44)
Ul is the estimated crossflow velocity in the incinerator
(fps) calculated by the methods of Chapters II and III
Po and PI are the jet and crossflow densities, respectively.
Jet Spacing Equation
L + SNdo
(IV-45)
where:
L is the length of furnace wall on which the jets are to
be placed (feet), and
S is the desired value of jet spacing measured in jet diameters.
Inherent in Eq.uation (IV-45) is the assumption that the jets are placed
in a single line. Ivanov recommends that S be in the range of 4 to 5, al-
though values as low as 3 can probably be used without invalidating the
penetration equation [Equation (IV-44)].
The form of these equations sheds some light on the options open to
the systems designer. For a given set of furnace conditions, Equation
(IV-44) can be rearranged to give:
d.uo = constant = BI
(IV-46)
The product of jet diameter and velocity is fixed by furnace conditions.
IV-52
Arthur D Little, Inc
-------
.
,--
Substitution of Equations (IV-45) and (IV-46) into Equation (IV-43)
yields:
0.326Bl
QT/L = S
(IV-47)
The air flow per length of wall is fixed by furnace conditions, except in-
asmuc:h as S can vary from 3 to 5.
The design method can best be illustrated by example.
incinerator with the following properties:
Consider an
Capacity - 250 TPD
Total Stoichiometric Air Req. - 15,000 cfm
Length of Wall for Jet Placement (L) - 15 ft
Desired Depth of Penetration (Li) - 8 ft
Crossflow Velocity (ul) - 4 fp~ .
Jet and furnace temperatures of 100°F (5600R) and 2000°F (2460oR),
respectively, so that:
----'
J Po = [T 1 - 12460 - 2 10
p ~ T ~ 560 .
a 0
Step 1.
Comp~te the product douo from Equation (IV-44).
1 Pa
douo -= hUl (1. 6) ~
(IV-48)
= (96) (4)
(1.6)(2.10) = 171.4
,Step 2.
Select value of do.
Figure IV-l4 shows that for ul = 4 fps, any value of do below about 5
inches will allow drag forces to predominate over buoyant forces. Select
do = 3 inches.
,Step 3.
Compute Uo from Equation (IV-48).
--
Uo =
171. 4
3 = 57.1 fps
The pressure requirement for the overfire air fan depends mainly on
uo. Equation (IV-49) may be used to calculate the required velocity head
(H) as a function of Uo
IV-53
Arthur D Little, Inc
-------
L
-2
pouo
H=1.2 [ 2
(IV-49)
For Uo = 57.1 fps, the required head is about 1 inch of water plus ducting
and nozzle losses.
Step 4.
Compute N from Equation (IV-45).
Setting S= 4, Equation (IV-45) yields:
180
N = (4) (3) = 15
Step 5.
Compute QT from Equation (IV-43).
QT = (0.327)(15)(3)3(57.1) = 2520 cfm
Step 6.
Compare QT with the theoretical air requirement.
The theoretical air requirement is 15,000 cfm, so that the overfire
air requirement is [(2520/15,000)](100) = 17% of theoretical. The sound-
ness of this value can be checked by comparison with the air requirements
defined by the bed burning process (Chapter II) and by reference to ex-
perience. It is worthy to note, however, that few data exist to allow
confident valuation of the performance of existing plants with respect to
combustible pollutant emissions and the design and operating parameters
of the overfire air systems.
The amount of overfire air can be increased or decreased within limits
without seriously affecting the performance of the jets by changing the
jet spacing parameter (liS II in Equation (IV-45) within the range of 3 to 5;
by modification of the assumed value for ilL 11; or by using opposed jet
placement which prevents impingement of jets on the opposite wall.
IV-54
Arthur D Little, Inc
-------
E.
SUMMARY
The behavior of cold, oxygen-rich jets in a crossflow of hot, fuel-
rich furnace gases is not perfectly understood. Basic experimental and
theoretical work showing the interactions of buoyant and crossflow effects
in the presence of combustion is lacking. However, our experimental and
analytical work, together with previous work reported in the literature,
allows a semi-quantitative analysis of these effects which provides a basis
for the design of overfire jet systems.
Specifically, the prior art and the original work described above support
the following design objectives:
1.
Determination of the conditions where buoyancy
tant relative to crossflow effects: Using the
oped, jets can be designed to pass close above
without danger of disturbing the bed.
effects are impor-
criterion devel-
the refuse bed
2.
Approximation of the temperature profiles in a combusting jet
as a function of jet parameters and axial distance from the
nozzle mouth: These profiles show how jet design influences the
probability of impinging the opposite wall with a hot jet.
3.
Presentation of mathematical correlations and behavioral descrip-
tions of jet characteristics under conditions to be expected in
incinerators (drawing on the results presented in Chapters II and
III to describe the furnace environment): With this understanding,
the jet designer can gain insight and anticipate interactions
between the jet and the furnace.
4.
Presentation of a procedure for designing overfire jet systems:
This procedure incorporates the criteria listed under 1. and 2.
above, both of which mitigate toward small diameter jets, with
the general procedure of Ivanov.
The analysis methods for jet behavior presented in this chapter, when
coupled with the results of bed and flow analysis, give the incinerator
designer a new and powerful tool for the control of combustible pollutant
emissions. He can now estimate how much air is needed, where, and the
extent to which the incinerator flow will perturb the jet behavior. To be
sure, the correlation between realized performance (reduced combustible
emissions) and optimum placement/design/operation of jets is not known.
It is reasonable to assume, however, that some measure of improvement must
be associated with more strategic jet operation.
Beyond the correlations and design approaches presented here, it can
be seen that these analytical tools provide the basis for control systems
for incinerators. Knowledge of the relationships between air demand, refuse
properties and burnout is the cornerstone of control logic. Also, under-
standing of the furnace dynamics assists in the placement of instrumentation
and control sensors and in the interpretation of their output.
IV-55
Arthur D Little, Inc
-------
10.
11.
12.
13.
14.
15.
F.
REFERENCES
1.
W. R. Niessen et. a1., "Systems Study of Air Pollution From Municipal
Incineration", report to NAPCA under Contract CPA-22-69-23 by Arthur
D. Little, Inc., 3 Volumes (1970)--Avai1ab1e from Clearinghouse for
Federal Scientific and Technical Information, Springfield, Virginia
(PB-192-378, PB-192-379, PB-192-380).
2.
A. C. Stern, "Abating the Smoke Nuisance", Mechanical Engineering~,
1932, pp. 267-8.
3.
J. A. Switzer, "The Economy of Smoke Prevention", Engineering Magazine,
Dec. 1910, pp. 406-412.
4.
H. Kreisinger, C. E. Augustine, and F. K. Ovitz, "Combustion of Coal
and Design of Furnaces", U.S. Bureau of Mines Bulletin 135, 1917.
5.
L. P. Breckenridge, "Study of Four Hundred Steaming Tests", U.S.
Geo1. Survey Bulletin 325, 1907, pp. 171-178.
6.
A. E. Grunert, "Increasing the Oxygen Supply Over the Fire", Power,
Jan. 25, 1927, pp. 130-131.
7.
M. K. Drewry, "Overfire Air Injection With Underfeed Stokers", Power,
Sept. 21, 1926, pp. 446-447.
8.
A. R. Mayer, "Die Wirkung der Zwert1uft in der Wanderros1feuerung"
("Effect of Secondary Air in the Traveling Grate Stoker Furnace"),
Thesis, Braunschweig, Germany, 1938, Z. Bayr. Rev. Verein, Vol. 42.
9.
R. F. Davis, "The Mechanics of Flame and Air Jets", Proc. Institution
of Mech. Eng., Vol. 137, 1937, pp. 11-72.
E. W. Robey and W. F. Harlow, "Heat Liberation and Transmission in
Large Steam-Generating Plant", Proc. Institution of Mech. Eng., 125,
1933, pp. 201. ---
W. Gumz, "Overfire Air Jets in European Practice", Combustion, Q,
April 1951, pp. 39-48.
"The Reduction of Smoke From Merchant Ships", Fuel Research Technical
Paper No. 54 issued 1947 by the Department of Scientific and Industrial
Research, London, England.
J. A. Switzer, "Smoke Prevention With Steam Jets", Power, January 1912,
pp. 75-78.
L. N. Rowley and J. C. McCabe, "Cut Smoke by Proper Jet Application",
Power, November 1948, pp. 70-73.
F. B. Tatom, ScD Thesis, Georgia Institute of Technology (1971).
IV-56
Arthur 0 Little, Inc
-------
I____._~-~L--_- ----------
I
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
R. F. Davis, "The Mechanics of Flame and Air Jets", Proc. Institution
of Mech. Eng., 137, 1937, pp. ll-7~.
Y. V. Ivanov, "Effective Combustion of Overfire Fuel Gases in Furnaces",
Astonian State Pub. House, Tallin (1959).
E. R. Kaiserand J. B. McCaffery, "Overfire Air Jets for Incinerator
Smoke Control", Paper 69-225 presented at Annual Meeting APCA, June
26, 1969.
M. A. Field, D. W. Gill, B. B. Morgan and P. G. W. Hawksley, "Combustion
of Pulverized Coal", BCURA, Leatherhead, Surry, England (1967).
G. N. Abramovich, "The Theory of Turbulent Jets", M.LT. Press, Cambridge,
Mass.
S. Corrsin, NACA, Wartime Reports No. ACR 3L23.
F. P. Ricou and D. B. Spaulding, 1961, Inl. Fluid Mech., 11, pp. 21-32.
A. N. Syrkin and D. N. Lyakhovskiy, "Aerodynamics of an Elementary Flame",
Sooshch. Isentr. Nauchn-Issled. Kotloturvinnyi Inst. (1936).
M. A. Patrick, "Experimental Investigation of the Mixing and Penetration
of a Round Turbulent Jet Injected Perpendicularly into a Transverse
Stream", Trans. Inst. Chem. Eng., 45 (1967), pp. T-16 to T-31.
G. S. Shandorov, "Flow From a Channel into Stationary and Moving Media",
Zh. Tekhn. Fiz., 12,1 (1957).
J. F. Keffer and W. D. Barnes, Inl. Inst. Fuel, 15, 1963, pp. 481-496.
W. Tollmien, "Berechnung Turbulenter Ausbreitungsvorange", Z Fur
Angewandte Mathematik and Mechanik, ~ (1926) pp. 468.
"Layout and Application of Overfire Jets for Smoke Control in Coal-Fired
Furnaces", National Coal Association, Washington, D.C., Section F-3,
Fuel Engineering Data, Dec. 1962.
J. D. McAllister, "A Momentum Theory for the Effects of Crossflow on
Incompressible Turbulent Jets", ScD Thesis, University of Tennessee
(1968).
IV-57
Arthur D Little, Inc.
-------
CHAPTER V
DESIGN METHODS FOR INCINERATOR OVERFIRE AIR SYSTEMS
Arthur D Little.lnc.
-------
CHAPTER V
DESIGN METHODS FOR INCINERATOR OVERFlRE AIR SYSTEMS
In the previous chapter, it was cautioned (Section IV-C-5) that use
of the "ru1e-of-thumb" approaches presented there for the design of in-
cinerator overfire air jet systems could result in the installation of in-
effective devices. Uncertainty in this matter arises primarily from the
unknown risk arising from the use in refuse-burning applications of gen-
eralizations developed from coal-burning experience. Differences between
refuse and coal (as reflected in heat release distributions, typical furnace
shapE~s used, fuel chemistry, bed uniformity and so forth) could be expected
to influence the desired design characteristics of the overfire air systems.
This chapter presents an illustration of a more tailored design approach
which draws heavily upon the analysis methodology presented in Chapters II,
III, and IV. The assumptions explicitly and implicitly incorporated into
these analysis methods are admittedly speculative and they require experi-
mental verification. It would appear, however, that development of a design
approach along lines which allow input of parameters unique to the system
and refuse in question would provide the vehicle for better new plant de-
sign optimization; for more rapid resolution of operating problems; and for
the evaluation of alternatives in plant upgrading.
A.
STATEMENT OF THE PROBLEM
It is required to evaluate the design requirements of the overfire air
system for a triple traveling grate boiler-type incinerator furnace with a
capacity of 250 tons p~r 24-hour day. Each grate is approximately 15-feet
long. The furnace is 8-feet wide. Its general configuration is shown in
Figure 111-1. Because of the use of silicon carbide sidewall construction
along the grate line, the air will be introduced 3 feet above the top of the
refuse bed. Experience has shown that an average refuse residence time of
45 minutes will be required under most circumstances, resulting in an average
grate speed of 60 feet per hour (1 foot per minute) and an average initial
grate loading of 40 pounds of refuse per square foot.
The refuse to be burned in the unit ranges in free-moisture content
between 0 and 30%. ("n" varies from 5/6 to 1. 75 in the "refuse compound"
C(H20)n discussed in Chapter II.)
V-1
Arthur D Little, Inc
-------
L-
B.
EVALUATION OF SYSTEM OPERATING CHARACTERISTICS
1.
BED PROCESSES
In analyzing the bed processes according to the methods presented in
Chapter II, it is necessary to specify the carbonization characteristics of
the refuse (the fraction gasified), estimate the temperature of the gases
leaving the bed, and estimate the heat losses or gains experienced across
the plane at the top of the bed plane. Based on the observations of workers
in the field and some pyrolysis data, let us assume that 80% of the carbon
is gasified. Further, let us assume that the gases leave the bed at 2000°F
and that the average heat loss by the bed to the waterwalls is 56,000 Btu/
mole of oxygen passing through the grate. (It should be noted that, although
these values were assumed for the trial calculations below, a broader range
of values should be tested in actual system designs to explore more fully
the spectrum of possible situations.)
Using the technique of analysis presented in Chapter II, the gas compo-
sitions leaving the bed can be calculated. Using an assumed distribution of
unde'rgrate air flow rates (e.g., Table V-I), the flow rate of each gaseous
compound (CO, CO , 02' N , H 0) entering the furnace volume can then be
calculated for t~e gasifIcatIon and char burnout regions. Further, if some
estilnated fraction of the total undergrate air flow is assumed to bypass the
bed due to channeling affects (Table V-2) , the effect of this secondary
air I)n the combustion gas composition and temperature can be calculated.
These composition estimates may then be used to identify the oxygen (overfire
air) requirement to effect gas-phase burnout.
For these calculations, the equilibrium constants for the water-gas shift
and the carbon-CO-C02 reaction at a temperature T (OR) may be calculated from
the following relationships:
a.
Water-Gas Shift Equilibrium
[C02] [H2]
kw=
[H20] [CO]
3000
10glOkw = ~ - 1.587
b.
CO-C02 Equilibrium
[C02]
kc =
[CO]2
-1
atm
15169
10glOkc = T - 8.821
The calculation technique also allows estimates to be made of the length
of the gasification and char burnout zones. This is accomplished by considera-
tion of the initial quantity of refuse on the grate and the mass loss rate per
V-2
Arthur D Little, Inc
-------
TABLE V-I
TOTAL UNDERGRATE AIR FLOW RATES
(sefm/square foot of grate)
Grate
1 2 ...l....
Case
A 10 50 25
B 10 70 25
C 10 100 25
TABLE V-2
FRACTION OF TOTAL UNDERGRATE AIR BYPASSING BED
Grate
% BvPassing
1
2
3
10
15
20
V-3
Arthur D Little, Inc
-------
foot of grate travel due to gasification of char arising from the
undergrate air flow. [Note that the results illustrated in Table 11-4 are
in quantities (Btu, moles, etc.) per mole of oxygen passing through the
grate.] Such calculations will show the effects of refuse composition
(specifically inerts and moisture) and undergrate air flow on residue burn-
out. Alternatively, by holding burnout constant, the variation of system
capacity (tons per day) with these parameters could be estimated.
The results of a series of calculations using the parameters given in
Tables V-I and V-2 are shown in Table v-3. The composition of the gases
over the char region on grates 2 and 3 are given in Table V-4.
The strong relationship between gas composition, overfire air needs
along the grate, refuse characteristics and undergrate air flow can be seen
readily. Also, since gas temperatures, especially in the char region, can
be c;onsiderably elevated above those normally considered "safe" to prevent
slagging, it may be desirable to estimate the quantity of overfire air
needed to meet both the combustion requirements of the off-gases and also
to (:001 the gas mixture.
Although it must be emphasized that results of the type
Tables V-3 and v-4 are not rigorous, they will be helpful to
ditions to be expected and the system response to refuse and
practice changes. For example:
shown in
identify con-
operating
.
Dry refuse yields more CO in the pyrolysis region than wet
refuse;
.
More CO is emitted in the char-burning zone than elsewhere in
the furnace and will present a large overfire air demand; and
.
The underfire air requirement to achieve burnout increases as
the refuse moisture content increases.
Changes in the heat loss or gain term in the analysis, based on
estimates of the radiative heat transfer within the system, would allow
evaluation of the effect of uncoo1ed combustion chambers (refractory sys-
tems). Also, the effects of differential grate speeds and other alterna-
tives in undergrate air flow should be readily explored. It should be
noted that although the computations are straightforward, they rapidly
become tedious and a simple computer program was prepared to produce the
valuc~s shown.
The calculation results of interest in overfire air jet design are
pri~~rily the air requirements shown for the gases leaving each of the
gratt~s in the gasification and char burnout zones. The air requirements
in Table V-3 are given in standard (70°F, 1 atm) cubic feet per minute per
square foot of grate area. Therefore, the total overfire air requirements
can be calculated in view of the 8-foot furnace width and the calculated
zone length. In trial 2, for example, a minimum of 28 to 36 scfm are required
over the last half of the first grate, from 160 to 240 scfm over the full
length of the second grate, and 100 to 150 sefm over the first half of the
third grate.
V-4
Arthur D Little, Inc
-------
'<
I
V1
»
..,
-
:r
c
..,
o
r-
~
-
in
~
TABLE V-3
THEORETICAL BED COMBUSTION CHARACTERISTICS
Gasification Zones Grate 1 - Gasification Zone
Grate Burning Zone Lengths (a,b) Gas Composition (Vol. %) Gas Composition (Vol. %) Final Overfire Air R~q'd,
.Refuse Air flow Gasification Char Excluding Bypass Including Bypass (c) Temgerature (d) SCFM/ft I
Trisl Moisture Case From To From To CO ~2 .!!2 ~2 ~~ fQ C02 !!z !iz Q2 !!zQ F(e) w/o Bypass, w/Bypassl
-
I
1 0.0% A 1-8 3-8 3-9 9.8% 14.6 10.5 9.748.7 16.4 12.4 11.4 7.9 51.4 0.0 16.9 2324 8.4 7.4
2 15.0'. B 1-8 2-15 2-15 3-8 7.7 12.7 5.9 51.0 22.7 5.8 13.4 4.2 53.6 0.0 22.8 2327 4.5 3.5
3 21.4'. A 1-8 3-13 3-14 14.0% 5.6 13.2 4.7 51. 7 24. 9 3.8 14.0 2.9 54.4 0.0 24.9 2329 3.3 2.3
4 21.4% B 1-8 2-15 2-15 3-8
5 25.0'. A 1-8 3-15 16.7% 1.9 14.1 1. 7 53.2 29.2 0.2 14.8 0.2 55.8 0.0 29.0 2332 1.1 0.1
6 25.0% B 1-8 3-3 3-4 3-11
7 25.0'. C 1-8 2-12 2-13 2-14
8 29.2% B 1-8 3-4 3-5 3-10 0.1 14.4 0.1 53.9 31.5 0.0 13.4 0.0 55.8 1.429.3 2012 0.0 0.0
Grate 2 - Gasification Zone
Grate 3 - Gasification Zone
Char Zone Overfire 2
Air Req'd (d) SCFM/ft
Grate 2 Grate 3
w/o BP w/BP w/o BP w/BP
~
Gas Composition (Vol. %)
Inc1uding Bypass (c)
fQ C02 !!2 !i2 Q2 !!2,Q,
Gas Composition (Vol. %)
Including Bypass (c)
fQ C02 !!2 !i2 Q2 !!2,Q,
Final Overfire Ai~ R~q'd
Temgerature (d) SCFM/ft
F (e) w/o Bypass w/Bypass
Final Overfire Air Re2'd
Temgerature (d) SCFM/ft
F (e) wlo Bypass w/Bypass
1 11.2 11.9 6.9 52.8 0.0 17.1 2499 39.8 32.3 9.9 12.4 5.9 54.3 0.0 17.4 2682 18.7 13.7 N/A I N/A 18.2 13.2
2 4.8 13.9 3.4 55.0 0.0 22.9 2503 29.9 19.4 54.1 43.7 18.2 13.2
3 2.8 14.5 2.1 55.8 0.0 24.9 2505 15.7 8.1 1.7 15.0 1.2 57.2 0.0 24.9 2691 7.4 2.4 N/A N/A 18.2 13.2
4 21.9 11.4 54.1 43.7 18.2 13.2
5 0.0 14.7 0.0 57.6 0.0 28.4 2360 5.3 0.0 0.0 13.8 0.0 57.8 1.5 26.8 2347 2.5 0.0 N/A N/A N/A N/A
6 7.4 0.0 2.5 0.0 N/A N/A 18.2 13.2
7 10.6 0.0 77.4 62.5 N/A N/A
8 0,0 12.9 0.0 56.7 2.2 28.2 2012 0.3 0.0 0.0 12.3 0.0 57.7 3.0 27.0 2012 0.1 0.0 N/A N/A 18.2 13.2
~:
. <.a)
(b)
(c)
Assumes combustion with available oxygen and re-adjustment
of equilibrium.
Units are SCFM of air per foot of grate per foot of furnace
width for stoichiometric combustion. .
Gas temperature after combustion with bypass gases.
Assumes no burning on first 9 feet; all distances in feet.
From 2-7 to 3-5 should be interpreted as: from 7 feet down Grate 2
down Grate 3; percent (if given) denotes % unburned carbon in total
when burnout is incomplete.
to 5 feet
residue
(d)
(e)
-------
TABLE V-4
THEORETICAL CHAR REGION OFF-GAS CHARACTERISTICS
Gas Composition (Vol. %)
Excluding Bypass
Gas Composition (Vol. %)
Including Bypass (a)
Grate 2 (b) Grate 3 (c)
Component
CO 32.1 21.0 17.5
C02 1.6 6.3 7.8
02 0.0 0.0 0.0
K 66.3 72.0 74.7
2
Temperature (a)
2700°F
2900°F
(a) Assumes complete burning with available oxygen.
(b) Assumes 15% of the total undergrate air bypasses the bed.
(c) Assumes 20% of the total undergrate air bypasses the bed.
V-6
Arthur D Little, Inc
-------
The range of air requirements mentioned in the preceding paragraph
arises from consideration of the effect of bypassed undergrate air on the
air requirement of the gases leaving the grate. The validity of the bypass
or channel-burning concept rests on the data of Kaiser mentioned in Chapter
II and on observations of incinerator burning where cracks may be clearly
seen from time to time in the burning mass. The fact, however, that by-
passing occurs on a random basis suggests that there may be times when the
full undergrate air supply is provided to the bed. Under such circumstances,
a greater quantity of overfire air would be required (calculated by dividing
100-% Bypassed
the indicated "without bypass" air requirement by the quantity: 100 ).
Whatever the relevance of the bypass question when applied to the gasifica-
tion zone, it is clear that under almost all circumstances, substantial
amounts of carbon monoxide will be formed and released to the over fire volume
in the char zone. Since the assumptions used in calculating the gas compo-
sit:lons assumed only carbon was present in the bed, it can be noted that the
air requirement is related only to the underfire air rate and thus can be
madf~ smaller (but never zero since some air '£low is needed for grate cooling)
in direct proportion to the underfire air rate. Since the char zone is almost
invariably located on the burnout grate, this emphasizes the desirability of
minimizing the undergrate air flow in this area consistent with obtaining
complete burnout. Studies on particulate emission also suggest the desirability
of maintaining low underfire air rates in this area to minimize ash entrain-
ment. It is clear, however, (Case A), that reducing the underfire air rate
too much will result in incomplete burnout.
2.
FURNACE ENVIRONMENT
Based on the calculation method described above, the composition, tempera-
ture and mass flow rate pf gases entering the furnace enclosure can be estimated.
The next step in analysis of the system is to evaluate flow patterns and mixing.
The procedures described in Chapter III are directly applicable. It may be
advcmtageous, however, to consider a three zone flow analysis (gasification,
char burnout and ash cooling) and to carry out, using numerical techniques, an
analysis of the (curving) streamline patterns throughout the furnace enclosure.
FrODi such a more detailed evaluation, the mixing of oxygen-containing gases
frODi (under some conditions) the gasification zone and from the ash-cooling
zones could be estimated. Such an analysis would suggest the minimum re-
quirements for overfire air systems, would identify velocity distributions,
and would affect the convection boiler design (regarding tube spacing and
protection from erosion), and would provide a basis for estimation of overfire
air jet crossflow effects.
Of particular importance is estimation of the velocity environment with
which overfire air jets will interact. As suggested in Chapter III, the hot
gases rising from the gasification and char burnout zones are accelerated in
the hydrostatic pressure field created by the slow-moving cold gases in the
discharge grate region. The velocity of the gases may be estimated by the
following rationale.
V-7
Arthur D Little, Inc
-------
Assuming that the behavior of the bed off-gases of density Ph in rising
from the top of the bed at an elevation zl to another elevation z2 and ex-
periencing a pressure change from Pl to P2 may be described by the Bernoulli
equation:
Pg 2
Total Head = Constant = ~ + ~ + Z
Phg 2g
(V-l)
where g is the acceleration of gravity.
Therefore, from elevation 1 to 2, the hot gas behavior may be described by:
2 2
u2 ul (Pl - PZ)gc +
2g = 2g + Phg
(Zl - z2)
(V-2)
In the cold gases of density P , it is assumed that velocity changes
c
are small, such that:
Plg Pzgc
~ + zl = - + z2
P cg P cg
(V-3)
Combining these equations and in recognition of the reciprocal rela-
tionship between temperature and density, the following results:
2
u2
=
2 ~ TH
Ul + (z2 - Z ) (--
gc 1 TC
- 1)
(V-4)
where TH and TC are the temperatures (OR) of the hot and cold gases, re-
spec ti vely .
Using Equation (V-4) and substituting the temperatures calculated above,
hot gas velocity may be estimated as a function of the height over the bed.
The results of such calculations are shown in Table V-5. It is of interest
to note the rapid increase in velocity over the first few feet of rise. This
has particular significance in jet trajectory calculations where it can be
seen that the velocity environment traversed by the jet is strongly related
to the elevation and is quite different from that calculated without considera-
tion of this effort.
3.
JET BEHAVIOR
The evaluation of overfire air jet behavior can be readily carried out
using the correlations presented in Chapter IV. With these correlations,
trajectory estimates can be prepared which incorporate crossflow and buoyancy
effects (using values such as those in Tables V-3, V-4 and V-5). The co~
bustion processes associated with the jet can also be evaluated using the gas
composition data. In establishing the flow rate requirements, attention
will be required concerning the length of grate needing air and the effect
of overfire air in escalating gas temperatures and the possible need for
excess air to provide coolin~.
V-8
Arthur D Little, Inc
-------
»
...,
-
~
c
...,
1""""'\
~
C
-
-
{D
::1
o
TABLE V-S
THEORETICAL GAS VELOCITIES ABOV~ REfUSE BEDS (ft/sec)
Gasification Zone Gasification Zone
Grate 1 Grate 2
Feet Above Bed 0 2 4 6 .Q 1. !l. 6
Trial
1 1.3 10.7 15.1 18.5 7.0 13.3 17.5 20.8
2 1.3 10.9 15.4 18.8 9.4 14.8 18.8 22.0
3*
4 1.3 10.7 15.1 18.5 9.2 14.6 18.5 21. 7
5*
6 1.2 10.7 15.1 18.5 8.5 13.8 17.5 20.6
7 1.2 10.7 15.1 18.5 12.3 16.4 19.6 22.4
8 1.1 9.4 13.2 16.2 7.6 12.0 15.2 17.8
"f
\C
Char Zone Char Zone
Grate 2 Grate 3
Feet Above Bed 0 2 4 E. 0 2 !i E.
Trial
1 3.3 13.4 18.7 22.7
2 8.7 15.0 19.4 22.9 3.3 13.4 18.7 22.7
3*
4 3.3 13.4 18.7 22.7
5*
6 3.3 13.4 18.7 22.7
7 3.3 13.4 18.7 22.7
8 3.3 13.4 18.7 22.7
* Negligible cold gas zone in t hes e trials. Analytical model does not apply.
o
Gasification Zone
Grate 3
1. !i
3.6
17.3
12.5
.6.
21.0
3.0 11.1 15.5 18.8
3.0 11.1 15.5 18.8
2.7 9.7 13.4 16.3
-------
CHAPTER VI
TEST PLAN
Arthur D Little, Inc
-------
A.
CHAPTER VI
TEST PLAN
INTRODUCTION
Because of the complexity of the incineration process, design approaches
based on analytical predictions of behavior require experimental demonstra-
tion. This chapter presents a test plan for demonstration of the effective-
ness of design methods using the analysis methods presented in Chapters II,
III and IV. We recommend that such tests be performed in full-scale plants
rather than in pilot plant or laboratory units. In view of the complexity
of t.he process, scaling laws for extrapolation to full-scale are uncertain.
The objectives of the study are to:
.
Determine the effect of operating and design parameters on
combustible pollutant emissions by quantitative assessment
of the performance of a selected incinerator;
.
Evaluate the effectiveness of a number of alternatives in
jet mixing systems to effect burnout of these combustible
pollutants within the primary combustion chamber; and
.
Develop guidelines for incinerator design and operation
based on gathered data which identifies promising combustible
pollutant control techniques.
These objectives would be met by efforts including the following
steps:
a.
Conceptual Design--Prior to the implementation of a test pro-
gram at an incinerator, a review of current combustion theory
is necessary to delineate the areas where improved design can
be effected. Such information is available from this report.
The pertinent analyses and conclusions are presented below.
b.
Preparation for Tests--The first step in, implementing such a
study is to negotiate with an appropriate municipality to make
an incinerator facility available. Then, final detailed con-
struction plans for jet mixing systems can be prepared and
appropriate equipment purchased and installed at the incinerator
for use in the test program. Tests should then be carried out
at the incinerator site to define the flow of air and refuse
to the system and install measuring instruments to monitor
these flows during subsequent tests.
VI-I
Arthur D Little, Inc
-------
c.
Uncontrolled Emission Rates--The second step is to carry out a
series of tests approximating seven (7) test days to determine
the combustible pollutant emissions (both gaseous and particulate)
typical of the selected incinerator as they relate to feed
rate and undergrate air flow and distribution. These data will
be analyzed to suggest the sources of combustible pollutants
within the system; to identify any relationships between the
emission rates of the various pollutants one to another and to
incinerator operating conditions; and to identify through pro-
longed testing the test duration necessary to obtain meaningful
average values of measured quantities.
d.
Incinerator Behavior--The third step is to conduct a series of
experiments to characterize the behavior of the burning refuse
bed and furnace flow dynamics. Sampling of bed off-gas as a
function of position would identify the need and location for
overfire air. Also, the data would be useful in corroborating
the bed and flow behavioral analyses. Ten (10) test days and
some laboratory flow modeling studies is suggested as a target
for these tests.
e.
Mixing Experiments--The fourth step is to conduct a series of
tests using steam and air injected into the overfire combustion
volume and determine the consequent effects on combustible
pollutant emissions. The effect of injection angle, discharge
velocity and mass flow will be determined for both air and
steam injection. In the course of this program, a cross-furnace
stationary probe will be evaluated. A target of fourteen (14)
test days is suggested for these tests.
f.
Design and Operating Guidelines--The final step is to analyze
the data collected in the test program above. From this analysis
would result:
.
Guidelines for incinerator operations to minimize combustible
pollutant emissions for existing plants; and
.
Design guidelines for modification of existing plants and
for new construction to enable the design of systems with
minimum combustible pollutant emissions.
A detailed discussion of these steps follows,including necessary implementa-
tion plans and technical considerations.
VI-2
Arthur D Little, Inc
-------
B.
TECHNICAL DISCUSSION
1.
CONCEPTUAL DESIGN
From this report, several conclusions can be drawn as to the type of
inci,nerator to be used, the location and characterization of the jets, and
data acquisition scope and methods.
a.
The Incinerator
There are more than 500 incinerator furnaces now operating in the
United States. Even within one plant, strict duplication is seldom found.
This fact emphasizes the need for generality in the results of any design
correlation development effort, and also highlights the difficulty in
finding a truly representative incineration system. It is our feeling that
an incinerator for testing should meet the following criteria: it should
be of relatively recent construction, laid out such as to facilitate modifi-
cati,on and testing, and should be representative of the type of incinerator
presently enjoying the greatest "popularity." Although a few batch-feed
systems are being built throughout the nation, the majority of new furnaces
are of the continuous-feed variety, utilizing multiple mechanical grates
to move the refuse through the primary furnace enclosure.
For a given incinerator, there are a limited number of variables which
may be controlled. These include the grate speed (refuse feed rate), the
undergrate air supply (total mass flow and its distribution to the various
undergrate plenums), and the flow dynamics of the overfire air system.
The refuse composition and heating value is largely uncontrollable, reflect-
ing the daily receipts, but could be modified to reduce heating value by
mete.red addition of water, either to the pit (preferred) or to the charging
chute.
The under grate air supply and distribution is important from two stand-
points: the undergrate air flow largely defines the burning rate within the
syst.em and is also instrumental in controlling grate surface temperatures.
Undergrate air rate is also important as it affects the formation of residue
clinkers by forcing temperatures within the bed to ranges where the residue
mate,rials exceed their sticking temperatures. Typically, incinerators are
operated with total undergrate air flows corresponding to an overall 50%
exce.ss air. Often, however, the distribution of the air is not such that
its introduction corresponds with the air demand of the bed. In most cases,
much of the air passes up through the discharge grate where it is relatively
ineffective for combustion.
b.
The Jet System
Prior to the actual design of the overfire air and steam injection
systems, two initial design details must be determined: the location of
the jets and the proposed jet variables.
VI-3
Arthur D Little, Inc
-------
Two conceptual pictures have been suggested to identify the location
within an incinerator furnace where combustible materials are formed.
.
In the pyrolysis region of the furnace (the first few feet of
the second grate) the combustion air demand exceeds the air
supply and incomplete combustion occurs. The gas flows are such
that the combustibles would be expected to be found at the top
of the primary furnace and breeching. Combustion of the fuel-rich
gases could be effected by overfire air injected at high velo-
cities directly over the pyrolysis region or by overfire air
directed from elsewhere in the furnace which is caused to enter
the rising stream of pyrolysis gases. These approaches are
represented in the sidewall and roof jets suggested for these
tests and illustrated in Figure VI-I.
.
In the colder regions of the furnace (near the discharge end of
the third grate) burning occurs with such an excess of air that
combustion is quenched. In this case, we would expect to find
the combustible pollutants in the cooler gases at the bottom of
the outlet breeching of the primary chamber. If this region is,
indeed, the source of a sizable fraction of the combustible pollu-
tants emitted (probably more associated with carbon monoxide and
hydrocarbon emission than soot), means are needed to reduce the
flow rate of cooling air and to effect mixing of the remaining
off-gases with the hot gases generated elsewhere in the furnace.
The former aspect of combustible pollutant control can be effected
directly by reduction of the undergrate air flow to the minimum
required for grate cooling. The second aspect could involve the
use of steam jets (injecting mixing energies rather than addi-
tional air), directed in such a manner as to prohibit escape of
the cool gases along the bottom of the furnace and flues and to
encourage their mixture with the hot gas stream. This is suggested
and is also indicated in Figure VI-I.
In order to provide a rough estimate of the relative importance of the
pyrolysis region and discharge region hypotheses for combustion pollutant
generation, a limited number of tests (reported in Appendix F) were carried
out. These tests show higher carbon monoxide concentrations in the cooler
(lower) regions of the furnace breeching and almost none in the hotter (upper)
regions, suggesting the dominant importance of the discharge region as a
source of combustible pollutants. However, the fact that at the time of the
tests the furnace was being operated with a dry commercial refuse (offed.ng
near optimal combustion characteristics) biases the situation so that, for
the moment, the hypothesis that the pyrolysis zone can be an important
source of these pollutants should be retained. Indeed, smoking in the py-
rolysis region is common when the refuse is wet. Also, no tests were made
of hydrocarbons or combustible particulate loadings, and thus the source of
these pollutants (anticipated to be the pyrolysis region) was not established.
VI-4
Arthur D Little, Inc
-------
NDTC. s:
V: VE./..OCITY
p' PR.E"~Uf(£ Rt:.LATNE TO ATMOsPHERE
T' T£MPCRATURE
ANI:;LE/)
i!ocF JEr$
O SAMPLIN~
"'Po,,,,,
Top OF 1= URNA C£
<:
H
I
VI
......- -----
/'
/
I
/
- -~@,b. I
->~...,.,.c "'/ ~- ----~~ '51bE-~1 ~ll
e--- __Au~ :Ti
@ .
~ ~ 1'--
SrA6NAAJT
FLUID
L"
s-
rDAlrS
-
~
-,
.. -'-------..- - - --
-'- ----k
-~
GRATE "'3
»
..,
-
:r
s::
..,
o
r
;::;:
-
is"''
5'"
o
"0
4-'
FIGURE VI- 1
SCHEMATIC DIAGRAM: TWO-FLUID MODEL
OF INCINERATOR FURNACE FLOW
.---...,
..........
"
"-
\
I
I
/
-- +---
ScRuB8IN~
CflAHtSER.
-------
Following selection of the jet location, consideration must be given
to the dynamic flow parameters which should be associated with the jets:
discharge velocity, mass flow rate and injection direction. With respect
to the velocity, care must be taken for the sidewall jet system that im-
pingement on the opposite wall does not occur. Because overfire combustion
is anticipated for air jets discharging into the pyrolysis region, tempera-
tures within the jet could exceed 2500° F.: Impingement of such high-temperature
flows directly on refractory can lead to slagging, fluxing and rapid refrac-
tory degradation. As a consequence, one approach to jet operation should
include opposed jets (Figure VI-2a) where high jet discharge velocities
can be used, but the jets located on the opposite side of the furnace act
to prohibit direct wall impingement. One might anticipate problems in
exact opposition of these systems; and, as a consequence, some jet deflec-
tion into the bed may occur. As this could cause the entrainment of fly
ash material, thus disadvantageously increasing the particulate loading in
the flue gases, this effect must be monitored.
An alternative to opposed jets involves an interlacing of the jets
(Figure VI-2b). .. For this approach, care must be given to the selection of
jet operating characteristics to avoid penetration distances greater than
the: furnace width.
The velocity and mass flow rate for the jets should be selected on
the: basis of the anticipated air requirement and the expected jet penetra-
tion and flow path. Since some uncertainty exists in these flow paths
(see Chapter IV), careful observation of the system in operation and pro-
vision of a wide range of possible operating characteristics (as reflected
in the fan pressure and volume flow capability) is necessary.
c.
Data Acquisition Scope and Methods
In order to provide a meaningful indication as to the need for and
effectiveness of jet mixing systems, a location must be found which pro-
vides a valid measure of the effect of combustion chamber mixing on com-
bustible pollutant emission rate.
As a minimum, sampling ports should be installed in the outlet breech-
ing of the primary chamber. A disadvantage of measurements limited to the
brE!eching area is that the conditions of gas composition and temperature and
velocity found in the breeching represent the integrated effect of a number
of interacting processes and forces within the furnace which are not ex-
plicitly revealed in point measurements taken in the breeching. To par-
tially offset this limitation, we recommend that the majority of tests
USE~ a profile method of measurement where the vertical gradients of tempera-
ture, composition and velocity of the gas flow would be measured. Although
thE! relationship between the fluid sampled at, say, the upper region of the
brE~eching and a specific region within the furnace cannot be established with
high assurance, some inferences can be drawn. It is also possibie that the
data from a cross-furnace stationary (CFS) probe may make it possible to
better relate breeching profiles to incinerator furnace enclosure flow
behavior. ,
VI-6
Arthur D Little, Inc
-------
-----
FIGURE VI-2a OPPOSED JETS
FIGURE VI-2b INTERLACING JETS
VI-7
Arthur D Little, Inc
-------
The variables of interest include gas temperature, velocity and compo-
sition. Detailed data of this sort will enable calculation of energy and
material balances about the primary furnace, suggest the sources (pyrolysis
or discharge grate areas) of combustible and mineral pollutants, and permit
the estimation of furnace emission rates of air pollutant species. Initially,
measurements will be made of:
.
Combustion Gases (CO, COz' 0z' NZ' HZ' HZO)
Hydrocarbons
.
.
Mineral Particulate
.
Combustible Particulate
In regions of high CO concentration, the CO/C02/NZ/OZ measurements can
be made on a batch basis with a manual or electric Orsat. In regions of low
CO concentration, non-dispersive infrared methods are necessary. The
latter measurement method (which we expect to be necessary in the breeching
area) will give continuous CO concentrations for one probe or, by use of a
sequenced solenoid valving arrangement, will give CO readings at multiple
elevations at frequent intervals.
Water vapor determinations can be made by metering condensate and gas
flow rate in areas of high moisture content and by wet/dry bulb means for
lower moisture.
Hydrogen determinations (important in studying the water-gas reaction
equ.ilibria) can be made on a batch basis using a gas chromatograph.
Hydrocarbon determinations can be made using flame ionization methods.
Particulate matter determinations (dry dust only) can be made with
filters alone (excluding the impingers associated with the full EPA sampling
train) .
Probes for the test program should be water-cooled to enhance their
survival in the high-temperature environment and, importantly, to quickly
quench combustion reactions of the gases, aerosols and particulate.
Ide.ally, the probe should also incorporate a thermocouple junction, shielded
from radiation and immersed in the in-flowing gas stream. Also, and es-
pec.ially for tests in the breeching area, the probe should be equipped to
determine the impact and static pressure (velocity) of the gases in the
sampling region.
In addition to the sample gases, data acquired during the tests should
include a variety of quantitative and qualitative data regarding the sys-
tem operating characteristics. These would include:
VI-8
Arthur D Little, Inc
-------
8 All forced airflows in to the furnace (underfire air under each
grate, overfire air or steam jet flows);
8 Grate speed (as it indicates refuse feed rate);
8 Readings of standard furnace instruments (draft gauges, thermo-
couples, etc.);
8
Commentary on the appearance of the feed refuse (dry
fluffed or compact; unusually high concentrations of
plastics, metal; etc.);
or wet;
leaves,
8
Commentary on the appearance of the main flame (shape, end
point along grate, length);
8
Commentary on point along grate where burn-out of refuse is
complete and qualitative evaluations of the combustible con-
tent of the residue;
8
Commentary on the appearance (penetration, trajectory, shape)
of observable jet flows.
2.
PREPARATION FOR TESTS
The initial groundwork for the tests includes the selection of an
adequate incinerator and the preparation of the final shop drawings of the
overfire air and steam injection systems. Following the formalization of
the arrangement between the contractor, EPA and the incinerator authorities,
the contracting for the installation of air and steam-handling equipment and
for refractory modification is possible. .
The pre-test effort would also include the installation and calibra-
tion of appropriate flow-measuring devices to enable monitoring of all forced
air flows both above and below the grate. In addition, the sampling probes
would be designed and analytical techniques wIll be tested appropriate to
the special needs of the program.
3.
UNCONTROLLED EMISSION RATES
It will be necessary to provide a reference emission baseline prior to
the conduct of tests on the effectiveness of overfire air systems. For these,
and in subsequent tests, the majority of the emission determinations could
be made in the breeching leading out of the primary combustion chamber (Figure
VI-i).
In this phase of the program, the effects on emissions of two operating
variables would be determined: the underfire air flow rate and distribution
and the grate speed (refuse feed rate). The underfire air distributions ten-
tatively suggested for tests on a typical triple-grate continuous feed incinera-
tor configuration are shown in Table VI-I.
VI-9
Arthur D Little, Inc
-------
TABLE VI-l
SET POINTS FOR GRATE AIR FLOW STUDIES
Grate
Air Flow (percent of stoichiometric)
Condition I II III
IV
--
Total Undergrate Air
o
75
25
100
o
50
50
100
o
113
37
150
o
75
75
150
1
2
3
In order to avoid the possibility of overheating the grate, experi-
ments should not be carried out at underfire air rates less than that
equ.ivalent to stoichiometric rate. We would suggest that the total under-
gra.te air flow should be varied from stoichiometric to 50% excess air.
Also, the distribution of undergrate air should be modified. In order to
assure a continuing refuse disposal capacity, the refuse feed rate (grate
speed) should not be reduced below 75% of the rated furnace capacity.
This range of parameters should give a strong indication of the effects
of variation of these parameters on pollutant emission rates. In tests on
the, effects of overfire air on combustible pollutant emissions, a subset
of three of the conditions shown in Table VI-I and the two suggested grate
speeds should be used. These conditions would be chosen in part to reflect
"normal operating practice", best operating practice and "undesirable" con-
ditions in that combustible pollutant emission rates may be high, but other
system characteristics (e.g., feed rate) may be desirable. Measurements
to be taken in the breeching would include profiles of the following variables:
.
Gas velocity;
.
Gas temperature;
.
Carbon monoxide, hydrocarbon and hydrogen concentrations;
.
C02' 02 and moisture concentrations; and
Particulate emission rate (reported separately as combustible
and mineral particulate emission rates).
.
VI-IO
Arthur D Little, Inc
-------
The results sought during the seven (7) test days which we feel approp-
riate for this series would include data illustrating the vertical profile
of pollutants and their response to changes in firing rate and undergrate
air flow rate and distribution and data illustrating the timewise variation
of pollutant emissions. From these data, the following would be sought:
.
Postulates regarding the sources of combustible pollutants
within the furnace system (based on the distribution of the
pollutants in the different gas temperature and flow regions);
.
Relationships between the emission rates of the various com-
bustible pollutants to simplify subsequent testing and analytical
procedures;
.
Perspectives which will identify the minimum test duration
needed to obtain valid average values of the fluctuating pollu-
tant emission rates;
.
Reference points in system performance against which to judge
the effectiveness of mixing systems; and
.
Relationships between the emission rate of the various pollutants
and the operating characteristics of the incinerator.
4.
INCINERATOR BEHAVIOR
It is appropriate to study incinerator behavior both for verification
of the analytical tools developed in Chapters II, III and IV and for in-
formation of the jet locations, air rates and injection velocities proposed
for the demonstration tests. Specifically, we recommend sampling of the
space over and along the bed to provide composition and temperature data
to test the bed burning model.
In addition, it may be of interest to study the flow patterns in the
furnace. The data from these tests would be useful in checking the analysis
giv'en in Chapter III and in interpreting the data on incinerator behavior.
Paralleling the study of incinerator gas flow, modeling studies (in the
laboratory) could be of value in providing a tool for application in future
incinerator design efforts.
a.
Confirmation of the 'Bed Burning Model
The bed burning model verification requires data on gas compositions,
estimates of heat flux across the top of the bed and temperature measure-
ments of the leaving gases. The gases rising from the bed should be sampled
to determine:
VI-ll
Arthur D Little, Inc
-------
.
Temperature;
.
Concentration of C02' 02' N2' H2' H20 and CO; and
Particulate loadings.
.
The sampling should be done at various positions along the length of the
bed and the results correlated with data on the gross refuse composition
[C(E20)n] as measured by analysis of the flue gases at the breeching to the
secondary chamber.
Particulate loading measurements would be determined as a function of
posi.tion and undergrate air flow.
Flux to the top of the bed could be estimated by pyrometric evaluation
of refractory wall temperature and subsequent radiative heat transfer analysis.
An approximate analytical method should be developed for estimation of this
quantity in the general case. This method could be checked by the measured
wall temperatures.
This experimental program and analysis effort would seek to provide the
following:
.
Data to show the source of particulate and gaseous combustible
emissions along the grate to suggest means to reduce emission
rate.
.
Confirmation of the water-gas shift equilibrium hypothesis.
.
Development of a method for estimating heat flux to the refuse
surface.
8
Confirmation of the ability of the bed-burning model to pre-
dict gas compositions and burnout.
.
A test of the theories proposed in the past as to the relation-
ship between underfire air rate and particulate emissions.
b.
Measurement of Gas Velocities
There is little data on furnace gas flow. The analytical models that
are discussed in Chapter III implied certain velocity levels and profiles
in the overfire region. To assess the validity of these models, we suggest
measurements of gas flow velocity !n the incinerator. Photographic
techniques, as described by Lavrov3, could be used for this purpose. The
technique uses high-speed photographs of the hot gas flow field, to determine
velocities by analysis of the trajectories of incandescent particles. In
the case of an incinerator, it may be possible to observe particles rising
from the refuse bed. Alternatively, it may be desirable to inject fine
powders of magnesium or other material to obtain a more controlled source
of particles. The observations would be made through holes in the sidewalls
or ceiling of the furnace at points where other measurements are planned.
VI-l2
Arthur D Little, Inc
-------
c.
Scale Modeling of Incinerator Flow
A second way to assess the validity of the analyses of the gas flow
is to compare its results to observations of flow in a laboratory scale
model. In the course of the experimental study of jets in a crossflow de-
scribed in Chapter IV, we developed techniques for simulating flow fields
with large temperature (hence, density) gradients. Specifically, room
temperature air can be utilized to simulate the hottest gases in an in-
cinerator if nitrogen vapor near its boiling point (-320°F) is used to
simulate the coolest gases. Water vapor in the air condenses at points
in the simulated flow where the temperature is below the dew point so that
the mixing region between the cold nitrogen vapor and the room air is
clearly visible and can be photographed.
We suggest construction of a scale model of the incinerator
(desk-top size) and simulation of the furnace flow using this technique. Two
or more zones of flow should be utilized. This laboratory approach has the
advantage that input velocities and flow rates can be accurately determined
and alternate geometries can be studied. The flow field can be photographed
and relatively simple temperature measurements can be made to obtain an
indication of the degree of mixing. The model could be used to assess the
validity of the analysis, for example, by comparing observed widths of
zones of flow in the breeching area with those calculated. This work could
also provide a new approach for simulating furnace flow that could be applied
to various types of configurations.
As a result of these efforts, the following would be sought:
.
A bringing together of an analytical technique, a low-cost
laboratory technique and data on the full-scale device to
enable evaluation of incinerator flow dynamics.
.
Confirmation of the theoretical analysis of gas acceleration
and mixing in incinerators.
.
Development of a method for study of alternate chamber
geometries which allows simulation of the important buoyancy
effects.
5.
MIXING EXPERIMENTS
The experimental tests outlined above would document a baseline of
combustible pollutant emissions against which to judge the effectiveness
of the mixing tests described here. As shown in Figure VI-2, the incinera-
tor 'Jould be equipped with three systems for overfire volume mixing; a roof-
entry air jet system, a sidewall-entry air jet system, and a discharge
end-'o1all steam jet system. In the fourteen (14) test days envisioned for
this part of the program, these jet systems would be operated and their net
effel:t on combustible pollutant emission rates determined by profile measure-
ments in the breeching. Also, a cross-furnace probe could be evaluated
for use in the follow-on test program described below.
VI-13
Arthur D Little, Inc
-------
The test matrix presented for the baseline series of measurements
(Table VI-I) indicates eight combinations of grate underfire air rate and
distribution and grate speed which are of interest. From among those
eight conditions and in light of the baseline tests, two or three should
be i;elected for use in the study of jet effects. The overall conditions
within the refuse bed will likely be controlled by the underfire air and
grate speed parameters. Since the bed takes long times to reach an equi-
lib:rium state, the plant should be operated at a constant underfire air
ratl= and distribution and grate speed over an entire test day. We anticipate,
however, that the overfire jets will not strongly affect the bed processes
but will rapidly (within seconds) shift the pollutant emission characteristics
to 'reflect new values corresponding to equilibrium in the overfire combustion
and mixing processes. As a consequence, we speculate that a large number of
jet effect tests can be run in a given day.
a.
Sidewall Jet Tests--The experiments with sidewall jets would
fall into two categories. As shown in Figure VI-2, the side-
wall jet system would be comprised of a total of ten jets
arranged in two banks of five on either side of the furnace.
The jets would be located directly opposite one another.
Three jets on either side can then be operated in an opposed
jet arrangement and a three-and-two combination can be used
for interlacing jets (Figure VI-3). The sidewall jets will
be operating in a region of the furnace where we would an-
ticipate high release rates of pyrolysis products and thus the
jets would be expected to have a great effect on the produc-
tion of soot (combustible particulate) within the furnace.
b.
Roof Jet Tests--The roof jets would operate using a feed-end
roof jet bank modified to inject the air at" an angle toward
the discharge end of the furnace. As for the sidewall jets,
roof ,jets would add air to the stream of pyrolysis products
rising from the bed and have the advantage in comparison to
the sidewall systems that hot spots arising from impingement
on the walls should not be experienced.
c.
Steam Jet Tests--The steam jets operating from the discharge
end-wall of the incinerator would serve to induce mixing and
control flow patterns in the cooler, air-rich gas arising
from the grate near the discharge end of ~he furnace. It
has been found that under some conditions of operation, a
substantial fraction of the carbon monoxide appears to arise
in this region of the furnace. Thus, by inducing mixing of
this cold gas with the hotter gases generated towards the
feed end of the furnace, higher temperatures and CO burnout
can be anticipated.
VI-14
Arthur D Little, Inc
-------
In addition to the tests of the jet systems as determined by measure-
ment. of the furnace output at the breeching, we would also suggest use of
a cross-furnace stationary (CFS) probe (Figure VI-3). Determinations of
the temperature and gas composition distribution within the furnace would
be of great interest and value. If, as described below, it appears valuable
to study the furnace environment in more detail, we suggest that use of a
grid. pattern of CFS probes, such as the one to be tested, would be a de-
sirable approach. Therefore, we suggest use of a prototype of the CFS
probe to gather data during the jet evaluations. The probe will provide
for the measurement of temperature, for gas sampling (non-isokinetic) in
the furnace and may be suitable for obtaining gas velocity data.
6.
DESIGN AND OPERATING GUIDELINES
The data developed above would provide estimates of the effectiveness
of overfire mixing in the reduction of combustible pollutant emissions.
, In addition, information would be derived on the correlation between the
emission rates of combustible and mineral particulate as they relate to
incinerator operating parameters (feed rate and undergrate air flow). To
the extent justified by the data, generalizable conclusions and recom-
mendations can be drawn from the experimental results which will con-
stitute guidelines for incinerator operations and design to minimize com-
bustible pollutant emissions. These guidelines would find use in the
operation of existing plants, would suggest approaches to plant modifica-
tion and would be helpful in the development of designs for new plants.
It should be recognized, however, that the data taken at the outlet
breeching of the furnace represent the integrated result of interactions
of the fuel bed, furnace enclosure flows and jet flows. To an extent,
it may be possible to infer the combustion chamber dynamics from the pro-
files determined in the outlet beeeching. The development of definitive
and highly credible design correlations, however, would rest on data
taken throughout the furnace enclosure. These tests could be part of a
subsequent program, although they would make extensive use of the cross-
furnace stationary probe and cantilever probe sampling techniques de-
veloped in the course of this effort.
VI-IS
Arthur D Little, Inc
-------
<:
H
I
I-'
0\
»
~
:r
c::
..,
o
C'
,....
,....
{t)
:J
()
Water Inlet
o . -
Water Outlet
..
...
.. .. .
..... .
... -. .
Metal
Gas
,.,-
Sampling Tubes
, ---....
... ... ., ....
-o. ..
.. .. -. -. ..
,. .... .....
-. ~ . .,.
/
FIGURE VI-J
CROSS-FURNACE STATIONARY (CFS) PROBE
.. . - .
. -. . .
. .
..
o .
..
.
~
..
-
..
..
.
..
.
..
.
.",
.
Ceramic Thermocouple Well
Thermocouple
-------
C.
REFERENCES
1.
Niessen, W. R. et.al., Systems Study of Air Pollution from Municipal
Incineration, report to NAPCA under Contract CPA 22-69-23 by Arthur
D. Little, Inc., 1970.
2.
Private Communication with Elmer R. Kaiser, New York University, re-
garding tests at the Babylon and Oceanside incinerators.
3.
Lavrov, P. 1., "Some Problems in the Technique of Approximate Modeling
of Furnace Hydrodynamics", Institute of Heat Energy of the Academy
of Sciences of the Ukranian Soviet Socialist Republic, Kiev, U.S.S.R.
VI-17
Arthur D Little, Inc
-------
CHAPTER VII
BIBLIOGRAPHY
Arthur D Little, Inc
-------
CHAPTER VII
BIBLIOGRAPHY
Chapter VII is made up of two parts: (A.) a biblio-
graphy prepared as part of the Arthur D. Little, Inc. study of
incineration overfire mixing; and (B.) abstracts of the foreign
articles appearing in the bibliography. The abstracts were,
for the most part, prepared by a member of the ADL project team.
In cases where they have been taken from CHEMICAL ABSTRACTS or
ENGINEERING INDEX, the source has been noted. Reference 90
has not been abstracted as it was a major source for this report
and has been cited extensively.
The bibliography is a compilation of selected refer-
ences from specific subject areas. These subject areas include:
Subject
Smoke Abatement, Application of
Overfire Air
References
1 to 51
Page
VII - 3
Basic Jet and Mixing Theory
52 to 81
VII - 7
Overfire Jet Design
82 to 102
VII -11
Soot Formation and Burnout
103 to 118
VII-13
Combustion on Fuel Beds
119 to 191
VII-14
Furnace Design
192 to 251
VII-21
Furnace Gas Flow
252 to 265
VII-26
Miscellaneous
266 to 268
VII - 2 8
The comprehensive literature search conducted for this
project covered the years 1901 through 1956. A check in more
recent literature indicated that the technology we sought infor-
mation on was not covered after the early 50's. Domestic techni-
cal and government sources were reviewed as well as foreign
technical literature. The most pertinent sources of information
were found to be ENGINEERING INDEX and a listing of Bureau of
Mines publications.
VII -1
Arthur D little, Inc
-------
The following sources were covered:
CHEMICAL ABSTRACTS
ENGINEERING INDEX
1907 thorugh 1956
1901 through 1953
1934 through 1954
INDUSTRIAL ARTS INDEX
Additional sources were "International Symposium on Combustion,"
numbers 1 through 12 and "List of Publications Issued by the
Bureau of Mines, July 1, 1910 to January 1, 1960." Bibliographies
of all articles reviewed were scanned for pertinent references
and extensive use was made of the ADL and MIT library holdings.
VII-2
Arthur D Little, Inc
-------
A.
1.
BIBLIOGRAPHY
Smoke Abatement, Application of Overfire Air
1.
NEEDLE JETS OF SUPERHEATED STEAM TO PREVENT SMOKE:
LUCKENBACH BOILER FURNACE
Anon
Eng. News, pp. 713-4 (29 December 1910)
THE
2.
RE-CYCLING OF FLUE GASES IN BOILER FIRING
Anon
Fuel Economist 11 (125), 203-5 (February 1936)
3.
MIT WELCHEM UNTERDRUCK WIRD DIE ZWEITLUFT DEM FEUERRAUM
AM ZWECKMAESSIGSTEN ZUGEFUEHRT? (At What Pressure
Should Secondary Air Be Supplied to a FurnaceJ)
Anon
Warme 61 (5), 97 (29 January 1938)
4.
SMOKE ELIMINATING APPARATUS
Anon
Steam Engr. ~ (108), 427 (September 1940)
5.
DATA ON OVERFlRE JETS FOR FUEL SAVING AND SMOKE REDUCTION
Anon
Heating and Ventilating ~ (10), H & V's Reference Data-279
and 280 (1944)
6.
THE REDUCTION OF SMOKE FROM MERCHANT SHIPS
Anon
Fuel Research Tech. Paper No. 54, Department
Industrial Research, London, 1947, 39pp
of Scientific and
7.
OVERFlRE JETS IN ACTION FOR SMOKE ABATEMENT
Anon
Bituminous Coal Research, Inc., Brochure, 1948
8.
THE EFFECT OF CERTAIN FACTORS ON THE EFFICIENCY OF A HAND-FIRED,
NATURAL-DRAUGHT, LANCASHIRE BOILER
Anon
Fuel Research Tech. Paper No. 55, Department of Scientific and
Industrial Research, London, 1949, 35pp
9.
TEST EFFECT OF OVERFlRE AIR JETS ON BOILER EFFICIENCY
Editors of Indus. and Power (abstract of report by Engdahl and Stang)
Indus. and Power 51 (3), 66-9, 98 (September 1946)
10.
SMOKE ABATEMENT PROGRESS
Victor J. Azbe
Power Plant Eng., pp. 1102-3 (1 October 1930)
VII-3
Arthur D Little, Inc
-------
1.
Smoke Abatement, Application of Overfire Air (cont.)
11.
MODERN APPLICATIONS OF OVERFIRE AIR
H.C. Carroll
Trans. Am. Soc. Mech. Eng. 65 (2), 73-86 (1943)
12.
OVERFIRE AIR PERFORMANCE APPLIED TO STATIONARY PLANTS
H.C. Carroll
Power Plant Eng. il (11), 81-3 (1943)
DOWN JET COMBUSTION
V.E. Chancellor
Gas Times 66 (399-401 (1951)
13.
14.
UBER DIE WIRKUNGSWEISE VON RUSSBLASERN (On the Way in Which
Soot Blowers Work)
K. Cleve and R. Muller
Arch. Warmewirtsch. u. Dampfkesse1w. 21 (1), 17-9 (1940)
15.
AN INVESTIGATION OF COMBUSTION AIR FOR REFUSE BURNING
L.J. Cohan and R.C. Sherrill
Presented at National Incinerator Conference, ASME, New York,
May 18-20, 1964
16.
AIR CONTROL FOR BITUMINOUS UNDERFEED STOKERS
Noel Cunningham
Heating and Ventilating 33, 50-3 (September 1936)
17.
APPLICATION OF OVERFIRE AIR SYSTEMS IN SMOKE CONTROL
Edward T. Douglass, Jr.
Southern Power and Ind. 68 (2), 72-3, 77 (1950)
18.
OVERFIRE AIR INJECTION WITH UNDERFEED STOKERS
M.K. Drewry
Power 64 (12), 446-7 (21 September 1926)
EFFECT OF OVERFIRE AIR ON THE EFFICIENCY OF A
BOILER
Richard B. Engdahl and John H. Stang
Nat1. Engr. 51, 322-7 (May 1947)
SMALL INDUSTRIAL
19.
20.
DIE TECHNIK DER ZUFUHRUNG VON VERBRENNUNGSLUFT (The Technique of
Supplying Combustion Air)
R. Fehling
Arch. Warmewirtsch. u. Dampfkesse1w. 11 (4), 119-23 (1930)
21.
USE OF THE DOWN-DRAFT COKING METHOD FOR SMOKELESS COMBUSTION
Julian R. Fellows and John C. Miles
Heating, Piping, Air Conditioning 15 (8), 431-5 (1943)
VII -4
Arthur D Little, Inc
-------
1.
Smoke Abatement. Application of Overfire Air (cont.)
22.
FUNDAMENTALS OF SMOKE ABATEMENT
Joseph P. Flynn
Natl. Engr. ~ (7). 378-9 (1948)
STEAM JET BLOWERS AND FANS ON STEAM BOILERS AT GAS-WORKS
R.1. Greaves
Gas J. 250. 288-93 (7 May 1947)
23.
24.
(Secondary Air for Grate
25.
ZWEITLUFTZUFUHRUNG BEl ROSTFEUERUNGEN
Firing)
Wilhelm Gumz
Feuerungstech. ~ (11), 123-4 (1935)
ZWEITLUFTZUFUHRUNG. DIE DUSENANORDNUNG UNTER BESONDERER
BERUCKSICHTIGUNG DES SYSTEMS BADER (Secondary Air Supply.
Arrangement with Special Regard to the Bader System)
Wilhelm Gumz
Feuerungstech. 30 (2), 32-6 (15 February 1942)
Jet
26.
A JOINT MEETING WITH THE NATIONAL SMOKE ABATEMENT SOCIETY.
I. RECENT ADVANCES IN SMOKE ABATEMENT, BASED ON THE WORK OF THE
FUEL RESEARCH STATION
T.F. Hurley
J. Inst. Fuel 20 (115). 189-94 (August 1947)
27.
VERSUDHE AN FLAMMROHR-INNENFEUERUNGEN MIT ZWEITLUFTZUFUHRUNG
(Experiments with Secondary Air Supply in a Flame Tube Internal
Furnace)
H. Janissen
Warme ~ (12). 205-7 (25 March 1939)
28.
THE USE OF SECONDARY AIR IN BOILER FURNACES
N.Y. Kirov
Bull. Brit. Coal Utilization Research Assoc. 12 (6), 205-13 (1948)
29.
THEORETISCHES UBER ZWEITLUFTZUFUHR BEl ROSTFEUERUNGEN (Theory of
Secondary Air Supply for Grate Firing)
P. Koessler
Arch. Warmewirtsch. u. Dampfkesselw. 19 (6). 153-6 (1938)
30.
ZWECKMABIGE ZWEITLUFTZUFUHR BEl FEUERUNGEN- STAND DER FORSCHUNG
UND ZUKUNFTIGE AUFGABEN (Appropriate Secondary Air Supply in Firing.
State of Research and Future Problems)
P. Koessler
Arch. Warmewirtsch. u. Dampfkesselw. 19 (7), 169-73 (1938)
31.
BLOWER INSTALLATIONS AND AIR DUCTS
R.A. Langworthy
Practical Engr., pp. 1078-80 (1 December 1915)
VII-5
Arthur [) Little, Inc
-------
1.
Smoke Abatement, Application of Overfire Air (cont.)
32.
OVERFIRE JETS FOR SMOKE ABATEMENT
William S. Maj or
Southern Power and Ind. 65 (4), 44-7, 121-2 (1947)
33.
FURNACE PERFORMANCE WITH OVERFIRE JETS
W.S. Major
ASME - Advance Paper #49-S-18 for meeting May 2-4, 1949
34.
ABATEMENT OF SMOKE AND CINDERS
HOUSE TERMINALS
H.E. May
Proc. Smoke Prevention Assoc.
FROM LOCOMOTIVES AND AT ROUND-
Amer. 42, 50-7 (1949)
35.
USE OF JETS TO PRODUCE TURBULENCE IN SPREADER-STOKER FIRING
H.G. Meissner and M.O. Funk
Combustion 16 (3), 42-6 (September 1944)
36.
ZWEITLUFTZUFUHRUNG ZUM EINHALTEN DES WIRTSCHAFTLICHEN CO -GEHALTS
DER RAUDHGASE (Secondary Air Supply to Keep an Economic C02 Content
of the Off-Gases)
Obering. Mortensen
Warme 60 (49), 799-801 (4 December 1937)
37.
DEVICE FOR PREVENTING SMOKE
W.H. Odell
Power, pp. 66-7 (10 January 1911)
38.
SMOKELESS COMBUSTION OF COAL IN BOILER FURNACES
D.T. Randall and H.W. Weeks
U.S. Bureau of Mines Bull. No. 40, 1912, 188pp
39.
SULLE POSSIBILITA DI MIGLIORARE IL RENDlMENTO DELLE CALDAIE
INSUFFLANDO ARIA SUPPLEMENTARE SOPRA IL COMBUSTIBILE (On the
Possibility of Improving Efficiency of the Boiler by Blowing
Supplementary Air Over the Fuel)
Antonio Rasi
Energia termica I, 202-6 (1939)
40.
WANT YOUR OVERFEED STOKER TO BE SMOKEFREE?
L.N. Rowley and J.C. McCabe
Power 92 (3), 158-61 (1948)
41.
CUT SMOKE BY PROPER JET APPLICATION
L.N. Rowley and J.C. McCabe
Power ~ (11), 676-9 (1948)
42.
GASBLANDNINGENS INFLYTANDE PA FORBRANNINGSHASTIGHETIN I FLAMMOR
(The Influence of Gas Mixing on Combustion Velocity in Flames)
John Rydberg
Feuerungstech. 30 (11), 257-9 (15 November 1942)
VII -6
Arthur D Little, Inc
-------
1.
2.
Smoke Abatement, Application of Overfire Air (cont.)
43.
WIRBELLUFTZUFUHRUNG (Traveling Grate with
WANDERROSTFEUERUNG MIT
Turbulent Air Supply)
W. Schultes
Arch. Warmewirtsch. u.
Dampfkesse1w. 16 (5), 117-8 (1935)
44.
HOW OVERFIRE AIR ELIMINATES SMOKE FORMATION
Herbert A. Scruggs
Nat1. Engr. 54 (9), 16-17 (1950)
45.
SOME NOTES ON SECONDARY AIR
J.F. Springer
Steam Engr. ~ (67), 294-6 (April 1937)
SOME NOTES ON SECONDARY AIR
J.F. Springer
Steam Engr. ~ (71), 458-61 (August 1937)
46.
47.
ABATING THE SMOKE NUISANCE
A.C. Stern
Mech. Eng. 54 267-8 (1932)
THE ECONOMY OF SMOKE PREVENTION
J .A. Switzer
Engineering Mag., pp. 406-12 (December 1910)
48.
49.
SMOKE PREVENTION WITH STEAM JETS
J.A. Switzer
Power, pp. 75-8 (16 January 1912)
50.
OVERFIRE AIR SYSTEM ELIMINATES SMOKE
George H. Watson
Heating and Ventilating 42 (2), 93-4 (1945)
51.
SMOKELESS COMBUSTION OF WOOD WASTE
B.H. Whitehouse
Power Plant Eng. 48 (12), 96-7 (1944)
See also:
References 85, 88, 91, 149~ 162
Basic Jet and Mixing Theory
52.
AIR MACHINERY -- AN INVESTIGATION OF THE PRINCIPLES OF THE AIR
INJECTOR
A. Bailey
Mech. Eng. 55, 762 (December 1933)
VII-7
Arthur 0 Little, Inc
-------
2.
Basic Jet and Mixing Theory (cant.)
53.
CHARACTERISTICS OF AXISYMMETRIC AND TWO-DIMENSIONAL ISOENERGETIC
JET MIXING ZONES
R.C. Bauer
Arnold Engineering Development Center, Air Force Systems Command,
U.S. Air Force, Tech. Documentary Rept. No. AEDC-TDR-63 253
(December 1963)
54.
MIXING AND FLOW IN DUCTED TURBULENT JETS
H.A. Becker, H.C. Hottel, and G.C. Williams
Ninth Symposium (International) on Combustion,
Academic Press, 1963, pp. 7-20
Cornell Univ., 1962,
55.
TRAJECTORY AND SPREADING OF A TURBULENT JET IN THE PRESENCE OF A
CROSSFLOW OF ARBITRARY VELOCITY DISTRIBUTION
W.W. Bowley and J. Sucec
ASME Paper 69-GT-33, presented at the Gas Turbine Conference and
Products Show, Cleveland, Ohio, March 9-13, 1969
56.
A GENERAL CORRELATION OF TEMPERATURE PROFILES DOWNSTREAM OF A
HEATED-AIR JET DIRECTED PERPENDICULARLY TO AN AIR STREAM
Edmund E. Callaghan and Robert S. Ruggeri
National Advisory Committee for Aeronautics, Tech. Note 2466,
September 1951
57.
DIE WIRKUNGSWEISE VON WIRBELLUFTDUSEN (How Jets Promote Turbulence)
Karl Cleve
Feuerungstech. ~ (11), 317-22, 362 (1937)
THE MECHANICS OF FLAME AND AIR JETS
R.F. Davis
Proc. Inst. Mech. Eng. 137, 11-72 (1937)
58.
59.
HEAT TRANSFER BETWEEN A FLAT PLATE AND JETS OF AIR IMPINGING ON IT
Robert Gardon and John Cobonpue
"International Developments in Heat Transfer," Amer. Soc. Mech.
Engr., New York, 1962, pp. 454-60
60.
USE OF MIXING PATTERNS TO PREDICT GAS TEMPERATURE PROFILES IN
BOILER FURNACES
P.M. Griffin
M.E. Thesis, Massachusetts Institute of Technology, 1956, 20pp
61.
TRANSFER OF HEAT AND MATTER IN THE TURBULENT MIXING
AXIALLY SYMMETRICAL JET
J.O. Hinze and B.G. Van Der Hegge Zijnen
Appl. Sci. Research (Hague) AI, 435-61 (1949)
ZONE OF AN
VII-8
Arthur D Little, Inc
-------
2.
. 67.
Basic Jet and Mixing Theory (cont.)
62.
COMBUSTION OF A TURBULENT JET IN BURNERS WITH PRELIMINARY MIXING
V.N. Ievlev
Sixth Symposium (International) on Combustion, Yale Univ., 1956,
Reinhold Publ. Corp., 1957, pp. 317-25
63.
INVESTIGATIONS OF THE TURBULENT MIXING REGIONS FORMED BY JETS
Arnold M. Kuethe
J. Appl. Mechanics 22, A87-A95 (1935)
64.
THE RATIOANLE OF AIR DISTRIBUTION AND GRILLE PERFORMANCE
C.O. Mackey
Refrig. Eng. 35, 417-19+ (June 1938)
65.
COAXIAL TURBULENT JETS
B.R. Morton
Intern. J. Heat Mass Transfer 2, 955-65 (1962)
66.
K VOPROSU 0 RASCHETE OSTROGO DUT'YA (On Overfire Air Jet Calculations)
I.K. Nairmark
Sovet. Kotloturbostroenie 2, 253-7 (1939)
BERICHT UBER UNTERSUCHUNGEN ZUR AUSGEBILDETIN TURBULENZ
(Report of Investigation of Developed Turbulence)
L. Prandtl
Z. angew. Math. u. Mech. 2, 136-9 (1925)
68.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN. I. THEORETISCHE VORBEMERKUNGEN
(The Influence of the Mixing Process on the Combustion of Gases,
Fuel and Air in Furnaces. I. Introductory Theoretical Remarks)
Kurt Runune1
Arch. Eisenhuttenw. 10 (11), 505-10 (May 1937)
69.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN TElL. II. VERSUCHE AN DER BRENNERSTRECKE
(The Influence of the Mixing Process on the Combustion of Gases,
Fuel. and Air in Furnaces. II. Experiments with the Firing)
Von Kurt Rummel
Arch. Eisenhuttenw. 11 (1), 19-30 (July 1937)
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN TElL. III. MODELLVERSUCHE UBER DIE MIS CHUNG
VON GAS- UND LUFTSTRAHLEN (The Influence of the Mixing Process
on the Combustion of Gases, Fuel and Air in Furnaces. III. Model
Attempt for the Mixing of Gas and Air Jets)
Von Kurt Runune1
Arch. Eisenhuttenw. 11 (2), 67-80 (August); (3), 113-23 (September);
(4), 163-81 (October:l937)
70.
VII-9
Arthur D Little, Inc
-------
2.
~asic Jet and Mixing Theory (cont.)
n.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN TElL. IV. NACHPRUFUNG DER ERGEBNISSE DER
MODELLVERSUCHE (The Influence of the Mixing Process on the
Combustion of Gases, Fuel and Air in Furnaces. IV. Summary of
the Results of the Furnace Investigations)
Von Kurt Rummel
Arch. Eisenhuttenw. 11 (5), 215-24 (November 1937)
LAMINARE STRAHLAUSBREITUNG (Laminar Jet Expansion)
H. Schlichting
Z. angew. Math. u. Mech. 13, 260-3 (1933)
72.
73.
TRANSVERSE JET EXPERIMENTS AND THEORIES - A SURVEY OF THE
LITERATURE
Donald J. Spring, Troy A. Street, and James L. Amick
Aerodynamic Branch, Advanced Systems Lab., R & D Directorate,
U.S. Army Missile Command, Redstone Arsenal, Rept. #RD-TR-67-4
74.
MIXING AND COMBUSTION
RECIRCULATION
P.D. Sunavala
J. Sci. Ind. Research
IN COAXIAL STREAMS:
PART I -- THEORIES OF
(India) 20B, 246-56 (June 1961)
75.
MIXING AND COMBUSTION IN FREE AND ENCLOSED TURBULENT JET DIFFUSION
FLAMES
P.D. Sunavala, C. Hulse, and M.W. Thring
Combustion and Flame (London) 1, 179-93 (1957)
76.
THE ADDITION OF AIR IN STAGES TO A PERFECTLY STIRRED REACTOR
M.W. Thring and E.G. Masdin
Combustion and Flame (London) 1, 125-30 (1959)
77.
BERECHNUNG TURBULENTER AUSBREITUNGSVORGANGE (Calculation of
Turbulent Expansion Processes)
Walter To11mien
Z. angew. Math. u. Mech. ~ (6), 468-78 (1926)
78.
ENTRAINMENT AND JET-PUMP ACTION OF AIR STREAMS
G.L. Tuve, G.B. Priester, and D.K. Wright, Jr.
Heating, Piping, Air Conditioning 13, 708-15 (November 1941)
79.
SURFACE PRESSURE DISTRIBUTIONS INDUCED ON A FLAT PLATE BY A COLD
AIR JET ISSUING PERPENDICULARLY FROM THE PLATE AND NORMAL TO A
LOW-SPEED FREE-STREAM FLOW
Raymond D. Vogler
NASA Tech. Note D-1629 (March 1963)
80.
DIFFUSION IN A LAMINAR CONFINED JET
Benjamin H. Wood, Jr.
D. Sc. Thesis, Massachusetts Institute of Technology, August 28,
1964, 12pp. + figs.
VII -1 0
Arthur D Little, Inc
-------
3.
Basic Jet and Mixing Theory (cont.)
81.
UBER DIE STROMUNGSVORGANGE 1M FREIEN LUFTSTRAHL (On the Flow
Processes of Free Air Jets)
W. Zimm
VDI-Forsch. Gebiete Ingenieurw. 11 (234), 5-35 (1921)
See also:
Reference 33
Overfire Jet Design
82.
HOW TO CONSTRUCT OVERFlRE JETS
Anon
Power Generation ~ (4),64+ (1948)
83.
IGNITION JET SYSTEM FOR BOILERS
Anon
Engineer 185 (4815), 452-3 (7 May 1948)
84.
EXACT DATA ON THE RUNNING OF STEAM BOILER PLANTS
No.3. THE AMOUNT OF STEAM USED BY STEAM JETS
D. Brownlie
Engineering 109 (2820), 71-4 (16 January 1920)
85.
THE DESIGN AND STUDY OF STEAM JETS FOR SMOKE ABATEMENT
John Du P~row and Edmund B. Bossart
B.S.M.E. Thesis, Case School of Applied Science, 1927
86.
DESIGN DATA FOR OVERFlRE JETS
R. B. Engdahl
Combustion 15, 47-51 (1944)
87.
OVERFlRE AIR JETS
R.B. Engdahl and W. C. Holton
Trans. Am. Soc. Mech. Engrs. ~ (7), 741-54 (October 1943)
88.
APPLICATION OF OVERFlRE JETS TO PREVENT SMOKE FROM STATIONARY PLANTS
R.B. Engdahl and W. S. Major
Bituminous Coa~ Research Inc., Tech. Paper No. VII, 1957
89.
OVERFlRE AIR JETS IN EUROPEAN PRACTICE
Wilhelm Gumz
Combustion~, 39-48 (April 1951)
90.
EFFEKTIVNOE SZHIGANIE NADSLOINYKH GORIUCHIKH GAZOV V TOPKAKH
(Effective Combustion of Stratified Fuel Gases in Furnaces)
Iu. V. Ivanov
Estgosizdat, Tallin, 1959
91.
OVERFlRE AIR JETS FOR INCINERATOR SMOKE CONTROL
Elmer R. Kaiser and Joseph B. McCaffery
Paper 69-225 presented at the Annual Meeting Air
Assoc., June 26, 1969
Pollution Control
VII -11
Arthur [) Little, Inc
-------
3.
100.
101.
102.
Overfire Jet Design (cont.)
92.
A SIMPLE AIR EJECTOR
J.H. Keenan and E.P. Neumann
J. Appl. Mech. 64, A75-A8l (1942)
93.
TESTS TO DETERMINE MOST PRACTICAL TYPE OF OVER-FIRE STEAM JETS
H.K. Kugel
Power ~, 638-9 (10 April 1928)
94.
WHERE TO PUT OVERFIRE JETS
William S. Major
Power Generation ~ (3), 95-7 (1948)
95.
WE TAILORED OUR SMOKE CONTROL TO FIT
J. Richard Manier
Power Generation 53 (7), 56-8 (1949)
96.
COMPARATIVE TESTS OF PITOT-STATIC TUBES
Kenneth G. Merriam and Ellis R. Spaulding
National Advisory Committee for Aeronautics, Technical Note No. 546
97.
EFFICIENCIES AND ENTRAINMENT-RATIOS OF AIR JET PUMPS
George Peter Miller
M.S.M.E. Thesis, Case School of Applied Science, 1940, 77pp
98.
A STUDY OF THE INDUCTION TUBE AND ITS EFFECT UPON COMBUSTION
H. Misostow
Power lQ, 484-5 (1929)
99.
STEAM JETS FOR PREVENTING SMOKE
R.H. Palmer
Am. Machinist, pp. 936-7 (2 July 1903)
CRITICAL PRESSURE RATIOS FOR STEAM NOZZLES
J.T. Rettaliata
No.2 of a Series of Engineering Bulletins published by Allis-
Chalmers, Milwaukee, Wisconsin. Leaflet 2357
IMPROVEMENT IN COMBUSTION BY THE USE OF STEAM JETS
C.D. Zimmerman
Power ~ (5), 159 (February 1927)
GRAPHISCHES BERECHNUNGSVERFAHREN FUR MEHRSTUFIGE DAMPFSTRAHLAPPARATE
(Graphical Method for Computing Multiple Steam-Jet Installations)
Ladislaus Zimmermann
Chem.-Ing.-Tech. ~ (11), 665-71 (1953)
See also:
References 4, 9, 19, 32, 47, 57, 197
VII-12
Arthur D Little, Inc
-------
4.
Soot Formation and Burnout
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
COMBUSTION TRIANGLE AND SOOT FORMATION
G. Ackermann
Mech. Eng. 56, 618-21 (October 1934)
FORMATION AND DEPOSITION OF
LOW RANK COALS
J.R. Arthur and G. Durand
Fuel 35, 514-15 (1956)
SOOT DURING THE IGNITION OF SOME
CARBONACEOUS DEPOSITS FROM HYDROCARBON DIFFUSION FLAMES
J.R. Arthur, P.K. Kapur, and D.H. Napier
Nature 169 (4296), 372-3 (1952)
EVOLUTION AND COMBUSTION OF VOLATILE MATTER FROM COALS.
PART III. COMBUSTION AND RELATED FACTORS
J.R. Arthur and D.H. Napier
Bull. Brit. Coal Utilization Research Assoc. 16 (7), 309-19 (1952)
BASIC CONSIDERATIONS IN THE COMBUSTION OF HYDROCARBON FUELS WITH AIR
H.C. Barnett and R.R. Hibbard
National Advisory Committee for Aeronautics Rept. 1300, 1957
EVOLUTION AND COMBUSTION OF VOLATILE MATTER FROM COALS.
PART II. THERMAL DECOMPOSITION OF COALS
A.H. Billington, I.G.C. Dryden. and D.H. Napier
Bull. Brit. Coal Utilization Research Assoc. 16 (6), 256-70 (1952)
FACTORS INFLUENCING THE PRODUCTION OF SMOKE FROM COAL
D.J. Bradbury and R.A. Mott
Fuel 20 (5), 100-5 (1941)
SOME COAL RESEARCH PROBLEMS AND THEIR INDUSTRIAL IMPLICATIONS
R.L. Brown
J. Inst. Fuel 29, 218-36 (May 1956)
THE TENDENCY TO SMOKE OF ORGANIC SUBSTANCES ON BURNING. PART I.
A.E. Clarke, T.G. Hunter, and F.H. Garner
J. Inst. Petroleum 32, 627-42 (1946)
SMOKE
W.T. Cosby
Bull. Brit. Coal Utilization Research Assoc. 13 (7), 225-31 (1949)
TENDENCY TO SMOKE OF ORGANIC SUBSTANCES ON BURNING. PART II.
PRODUCTION AND BURNING CHARACTERISTICS OF HYDROCARBON GELS
F.H. Garner, T.G. Hunter, and A.E. Clarke
J. Inst. Petroleum 32, 643-55 (1946)
SMOKE
VII-13
Arthur [) Little, Inc
-------
4.
Soot Formation and Burnout (cont.)
114.
IN HYDROCARBON FLAMES
115.
116.
117.
118.
LUMINOSITY AND SOOT FORMATION
D.W. Gill
Bull. Brit. Coal Utilization
(November/December 1958)
Research Assoc. ~ (12), 487-506
EVOLUTION AND COMBUSTION OF VOLATILE MATTER FROM COALS.
PART I. THERMAL DECOMPOSITION OF ORGANIC SUBSTANCES
P.H. Given
Bull. Brit. Coal Utilization Research Assoc. 16 (6), 245-55 (1952)
THE RATE OF GROWTH OF SOOT IN TURBULENT FLOW WITH COMBUSTION
PRODUCTS AND METHANE
K.S. Narasimhan and P.J. Foster
Tenth Symposium (International) on Combustion, University of
Cambridge. 1964, The Combustion Institute. 1965, pp. 253-7
UNDERFEED COMBUSTION,
ASH IN FUEL BEDS
P. Nicholls
U.S. Bureau of Mines
EFFECT OF PREHEAT AND DISTRIBUTION OF
Bull. No. 378. 1934. 76pp
COMBUSTION RATE OF CARBON
C.M. Tu. H. Davis, and H.C. Hottel
Ind. Eng. Chern. ~, 749-57 (July 1934)
See also:
Reference 175
119.
Combustion on Fuel Beds
5.
120.
121.
122.
(Operating
BETRIEBSERFAHRUNGEN MIT DEM NEUEN TURBINENROST
Experiences with New Turbine Grate)
Anon
Arch. Warmewirtsch. ~ (11). 299-301 (1925)
COMBUSTION IN FUEL BEDS
J.R. Arthur, D.H. Bangham, and M.W. Thring
J. Soc. Chern. Ind. 68 (1). 1-6 (1949)
EFFECTS OF FURNACE GAS STRATIFICATION ON OVERFEED STOKER EFFICIENCY
H.H. Baumgartner
Nat1. Engr. 31 (6). 263-6 (1927)
FEUERUNGSTECHNIK DES WANDERROSTES,EINFLUSSE VON BRENN STOFF ,
LUFT UND ROSTBAUART (The Firing Technique of the Traveling Grate.
The Impact of Fuel, Air and Grate Design)
Karl Beck
Arch. Warmewirtsch. u. Dampfkesse1w. 20 (4), 93-8 (1939)
VII-14
Arthur D Little, Inc
-------
5.
Combustion on Fuel Beds (cant.)
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
PHYSICAL PROCESSES IN A BED OF FUEL
J.G. Bennett
Inst. Fuel (London) 14, 47-62 (December 1940)
GAS FLOW IN FUEL BEDS
J.G. Bennett and R.L. Brown
Inst. Fuel (London) 13, 232-46 (1940)
BARK AND REFUSE FUEL BURNING
E.J. Calnan and J.N. Franklin
Pulp & Paper Mag. Can. 42, 111-16 (1941)
EFFECT OF EXCESS AIR ON CHAIN-GRATE STOKER OPERATION
E.C. Cawrse
Power 11, 1042-4 (1930)
ROSTWIDERSTAND VERSCHIEDENER KOHLENSORTEN (Combustion Resistance
of Various Types of Coal)
Roman Dawidowski
Z. Obersch1es. berg-u. huttenmann. Ver. Katowice 65 (10), 660-7;
(11), 728-34 (1926)
IMPROVING BOILER ROOM OPERATION
Otto de Lorenzi
Combustion 18, 198-202 (October 1928)
FACTORS AFFECTING THE TEMPERATURES OF TRAVELING-GRATE STOKERS
A.C. Dunningham and E.S. Grume11
Fuel 12, 327-34 (1938)
COMBUSTION OF FUEL ON A TRAVELLING GRATE
A.C. Dunningham and E.S. Grume11
Inst. Fuel (London) 11, 87-95 (December 1938)
PHYSlKALISCHE THEORIE DER VERBRENNUNG (Physical Theory of
Combustion)
W.H. Fritsch
Warme 60, 749-57 (13 November); 768-73 (20 November 1937)
RAPID COMBUSTION IN BOILER FURNACES
W.H. Fritsch
Steam Engr. 2, 1~3-5 (1938)
FLOW OF GASES THROUGH BEDS OF BROKEN SOLIDS
C.C. Furnas
U.S. Bureau of Mines Bull. No. 307, 1929, 144pp
THEORY OF FUEL COMBUSTION IN BOILER FURNACES
W.E. Garner
Gas World ~, 120-1 (1923)
VII-IS
Arthur D Little, Inc
-------
,--
5.
Combustion on Fuel Beds (cont.)
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
THE MECHANISM OF BURNING COAL ON A CHAIN-GRATE STOKER
E.S. Grume11
Mech. World 91, 440-2 (6 May 1932)
THE EVALUATION OF FUEL FROM THE CONSUMERS' VIEWPOINT
E.S. Grume11
Inst. Fuel (London) i, 361-74 (August 1932)
(The Process
DER VERBRENNUNGSVORGANG IN DER WANDERROSTFEUERUNG
of Combustion with Chain Grates)
W. Gumz
Feuerungstech. 24, 10-2 (1936)
ZONENEINTEILUNG UND ZWEITLUFTZUFUHRUNG BEl WANDERROSTEN (Division
into Zones and Secondary Air Supply for Chair Grate Firings)
Wilhelm Gumz
Feuerungstech. 30 (11), 256-7 (1942)
ULTIMATE BOILER CAPACITY LIMITED BY STOKER CONDITIONS
Joseph Harrington
E1ec. Rev. ~ (12), 451-3 (19 March 1921)
THE BURNING OF BITUMINOUS COAL ON LARGE UNDERFEED STOKERS
Bert Houghton
Intern. Conf. Bituminous Coal ~, 276-93 (1931)
TESTS ON A RECENT TYPE OF CHAIN GRATE STOKER AND NEW METHOD OF
BAFFLING STIRLING BOILERS
John A. Hunter
Engrs'. Soc. West. Penna., Proc., pp. 23-55 (February 1951)
FIRING WITH MULTIPLE-RETORT UNDERFEED STOKERS
George P. Jackson
Combustion 13, 30-3 (November 1941)
DE LA COMBUSTION
Materials)
Victor Kammerer
Bull. Soc. indus.
DES MATIERES VOLATILES (Combustion of Volatile
Mu1house 92 (2), 111-33 (1926)
ZUR AERODYNAMIK DER BRENNSTOFFSCHUTTUNG IN ROSTFEUERUNGEN
(On the Aerodynamics of Fuel Charging in Grate Firing)
H. G. Kayser .
Forsch. Gebiete Ingenieurw. B ~ (2), 89-100 (March-April
1935)
POWER-PLANT ENGINEERING -:"'"' AERODYNAMICS OF FUEL MOTION IN FURNACES
H.G. Kayser
Mech. Eng. ~, 315-6 (May 1936)
VII-16
Arthur D Little, Inc
-------
5.
Combustion on Fuel Beds (cont.)
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
VERSUCHE AN WANDERROSTFEUERUNGEN (Experiments on Chain-Grate
Furnaces)
W. Koeniger
Arch. Warmewirtsch. 10, 243-8 (1929)
COMBUSTION IN THE FUEL BED OF HAND-FIRED FURNACES
Henry Kreisinger, F.K. Ovitz, and C.E. Augustine
U.S. Bureau of Mines Tech. Paper 137, 1917, 76pp
COMBUSTION IN THE FUEL BED OF HAND-FIRED FURNACES
Henry Kriesinger, F.K. Ovitz, and C.E. Augustine
Fue114 (9), 271-6 (September); (10), 296-9 (October);
(11),~31-7 (November); (12), 364-70 (December 1935)
THE PROPAGATION OF AN IGNITION ZONE IN A SOLID FUEL BED AGAINST A
HIGH VELOCITY AIR STREAM
Donald C. Lea
M.S.C.E. Thesis, Massachusetts Institute of Technology, 1950, 33pp
CHEMISTRY OF COMBUSTION IN COAL-FIRED FURNACES
W.K. Lewis
Ind. Eng. Chern. 15, 502-3 (1923)
DIE SCHUTTHOHE EINER ROSTFEUERUNG (The Height of the Bed in
Grate Firing)
A.R. Leye
Brennstoff- u. Warmewirtsch. !I (2), 15-21 (1935)
THE PERFORMANCE OF SEVERAL TYPES OF BITUMINOUS COAL ON SMALL
UNDERFEED STOKERS
H.R. Limbacher and Ralph A. Sherman
Bituminous Coal Research, Inc., September 1938, 27pp.
VERBRENNUNGSVERLAUF VON STEINKOHLE AN EINER WANDERROSTFEUERUNG
(The Combustion of Bituminous Coal on a Traveling Grate)
Rud Loewenstein
Warme 2I, 97-101 (17 February); 121-5 (24 February 1934)
BEITRAGE ZUR FEUERUNGSTEDHNIK VON STEINKOHLEN AUF DEM WANDERROST
(Contribution to Firing Practice of Bituminous Coal on a
Traveling Grate)
Marcard
Warme 55, 397-401 (11 June 1932)
BEITRAGE ZUR FEUERUNGSTEDHNIK VON STEINKOHLEN AUF DEM WANDERROST
(Contribution to Firing Practice of Bituminous Coal on a
Traveling Grate)
Marcard
Warme 55, 417-22 (18 June 1932)
VII-17
Arthur D Little, Inc
-------
5.
Combustion on Fuel Beds (cont.)
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
DIE VERBRENNUNG ALS STROMUNGSVORGANG (Combustion as a Flow Process)
W. Marcard
Warme 60, 257-66 (24 April 1937)
COMBUSTION ON TRAVELLING-GRATE STOKERS
W.G. Marske11
Inst. Fuel (London) lQ, 100-7, 116 (1947)
MODE OF COMBUSTION OF COAL ON A CHAIN-GRATE STOKER. I.
OF RATE OF COMBUSTION ON THE COMPOSITION OF THE FUEL BED
W.G. Marske11 and J.M. Miller
Fuel 25, 4-12 (1946)
THE EFFECT
MODE OF COMBUSTION OF COAL ON A CHAIN GRATE
III. STOKER LINK TEMPERATURES
W.G. Marske11, J.M. Miller, and M.R. Webb
Fue1~, 78-85 (1946)
DIE WIRKUNG DER ZWEITLUFT IN DER WANDERROSTFEUERUNG (D84)
(The Effect of Secondary Air on Traveling Grate Firing)
Albert R. Mayer
Z. bayer. Revisions-Ver. ~ (4), 31-3 (February 1938)
STOKER.
DIE VORGANGE 1M FEUERRAUM EINES KESSELS MIT WANDERROSTFEUERUNG UND
IHRE ANDERUNG DURCH ZWEITLUFTZUFUHR (The Reactions in a Boiler
Furnace and Their Alteration with Twin Air Supply)
Albert R. Mayer
Feuerungstech. ~ (5), 148-50 (1938)
UNTERSUCHUNGEN UBER ZWEITLUFTZUFUHR IN WANDERROSTFEUERUNGEN
(Investigation of Secondary Air Supply in Traveling Grate Firings)
Albert R. Mayer
Feuerungstech. ~ (7), 201-10 (1938)
SOME FACTORS AFFECTING COMBUSTION IN FUEL BEDS
Martin A. Mayers
Am. Inst. Mining Met. Engrs. Tech. Pub. No. 771, 1937, 18pp
TEMPERATURE AND COMBUSTION RATES IN FUEL BEDS
Martin A. Mayers
Trans. Am. Soc. Mech. Engrs. ~, FSP, 279-88 (May 1937)
THE FUEL-BED TESTS AT HELL GATE GENERATING STATION, 1937-1938
M.A. Mayers, W.H. Dargan, Joseph Gershberg et a1
Trans. Am. Soc. Mech. Engrs. ~ (3), 191-211 (1941)
VERFEUERUNG GASREICHER KOHLE IN EINER NEUZEITLICHEN WANDER-
ROSTFEUERUNG (Combustion of Gas-Rich Coal on a Modern
Traveling Grate)
Albert Muller
Z. Reichshauptst. Techn. Uberwach. 1 (11/12), 61-72 (13 June);
(13/14), 76-81 (11 July 1942)
VII-18
Arthur D Little, Inc
-------
5.
Combustion on Fuel Beds (cont.)
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
VERIEUERUNG GASREICHER KOHLE IN EINER NEUZEITLICHEN WANDER-
ROSTFEUERUNG (Combustion of Gas-Rich Coal in Modern Traveling
Grate Firing)
Albert Muller
Feuerungstech. 31 (3), 68-9 (1943),
PRINCIPLES OF FUEL BEDS
P. Nicholls
Am. Inst. Mining Met. Engrs. Tech. Pub. No. 629, 1935, 17pp
THE OPERATION OF CHAIN-GRATE STOKERS
W.M. Park
Combustion 11, 24-8 (April 1940)
DIE GRENZ EN DER FEUERRAUMBELASTUNG UND IHRE RUCKWIRKUNG AUF
DIE AUSLEGUNG. DES KESSELS (Limits of the Combustion Chamber
Load and the Resulting Effect on the Design of the Boiler)
W. Pauer
Arch. Warmewirtsch. u. Dampfkesse1w. 20 (8), 197-202 (1939)
NOTES ON THE BURNING OF LOW-GRADE COAL ON CHAIN-GRATE STOKERS
V. Pickles and F.J. Redman
J.S. African Inst. Engrs. 12, 198-215 (1939)
DIE VERBRENNUNG VON BRAUNKOHLEN AUF DEM ARBATSKY-WANDERROST
(Combustion of Brown Coal on the Arbatsky Traveling Grate)
E. Praetorius
Braunkohle 30, 241-7, 266-9 (1931)
RESULTS OF EXPERIMENTS WITH STOKING WITH BAGASSE
H.C. Prinsen-Geerligs
Louisiana Planter 2i (5), 76-7 (1915)
UNDERFEED STOKERS BURN LOW GRADE OF COAL
C.E. Reese
Blast Furnace Steel Plant 10 (11), 588-91 (1922)
HEAT LIBERATION AND TRANSMISSION IN LARGE STEAM-GENERATING PlANTS
E.W. Robey, and W.F. Harlow
Proc. Inst. Mech. Engrs. 125, 201-89 (November 1933)
AERODYNAMICS AS A BASIS OF MODERN FUEL PRACTICE
P.O. Rosin
Fuel 15, 136-48 (1936)
ZUR PHYSIK DER VERBRENNUNG FESTER BRENNSTOFFE (The Physics of
Combustion of Solid Fuel)
P. Rosin and H.-G. Kayser
Z. Ver. deut. Ing. 12 (26), 849-57 (27 June 1931)
VII -19
Arthur D Little, Inc
-------
5.
Combustion on Fuel Beds (cont.)
178.
179.
180.
181.
182.
183.
184.
185.
186.
187.
188.
DIE RAUMLICHE UNO ZEITLICHE ENTWICKLUNG DER VERBRENNUNG IN
TECHNISCHEN FEUERUNGEN (The Change of Combustion with Respect
to Time and Space in Industrial Furnaces)
Kurt Rummel and Hellmuth Schwiedessen
Arch. Eisenhuttenw. ~ (12),543-9 (June 1933)
STAND UND ENTWICKLUNG DER FEUERUNGSTECHNIK. EIN QUERSCHNITT UND
UMRISS UBER DIE FORSCHUNG AUF DEM GEBIETE DER FEUERUNGSTECHNIK
IN DEN LETZTEN JAHREN (State and Development of Firing Techniques.
A Summary of Research in the Field of Firing Techniques in Recent Years)
Fr. Schulte and E. Tanner
Z. Ver. deut. Ing. 2l (21), 565-72 (27 May 1933)
VERBRENNUNGSPROBLEME IN BRENNKAMMERN VON HOCHLEISTUNGSDAMPFKESSELN
(Combustion Problems in Combustion Chambers of High Performance
Boiler Furnaces)
Karl Schwarz
Brennstoff-Warme-Kraft 1 (2),45-52 (May 1949)
A STUDY OF REFRACTORIES SERVICE CONDITIONS IN BOILER FURNACES
Ralph A. Sherman
U.S. Bureau of Mines Bull. No. 334, 1931, 141pp
DETERMINATION OF AIR FLOW THROUGH BERNITZ CLINKER PROOF FURNACE
LINING AND MEASUREMENT OF STATIC PRESSURE IN FUEL BED USING COAL,
COKE AND ASH AT DIFFERENT LEVELS
H.W. Shimmin and T. E. Waddell
B.S.M.E. Thesis, Massachusetts Institute of Technology, 1931, 59pp
AIR-SPACE AREA OF GRATES
R.T. Strohm
Elec. World 65 (2), 103-4 (9 January 1915)
COMBUSTION ON TRAVELING GRATES
B.M. Thornton
Eng. and Boiler House Rev. 56, 182-8, 204, 218-23, 237 (1942)
A METHOD FOR THE CONTROL OF COMBUSTION REACTIONS IN FUEL BEDS
M.W. Thring
Trans. Faraday Soc. ~, 366-77 (1946)
FURNACE DESIGN, VOLUME II, 3rd edition
W. Trinks
John Wiley & Sons, Inc., New York, 1963
NEW STOKER TECHNIC DEVELOPED FOR LIGNITE
G.W. Welch
Power Plant Eng. 38, 205-7 (May 1933)
VERBRENNUNGSVERLAUF BEl STEINKOHLEN MITTLERER KORNGROBEN
(Combustion of Coal of Medium Particle Size)
Helmut Werkmeister
Arch. Warmewirtsch. u. Dampfkesselw. 12 (8), 225-32 (1931)
VII-20
Arthur D Little, Inc
-------
5.
Combustion on Fuel Beds (cant.)
189.
190.
191.
6.
WINDVERTEILUNG UND FEUERGASBESCHAFFENHEIT BEl WANDERROSTFEUERUNGEN
(Blast Distribution and Combustion Gas Composition in Traveling
Grate Firing)
H. Werkmeister
Z. Ver. deut. Ing. ~ (26), 788 (30 June 1934)
FURNACE OPERATION ON BAGASSE FUEL
Wm. Whipple
Facts About Sugar 28, 321-2, 326 (1933)
BEKAMPFUNG DER VERSCHLACKUNG VON DAMPFKESSELN
Slagging in Boiler Furnaces)
Arthur Zinzen
Brennstoff-Warme-Kraft 1 (3), 63-8 (1950)
(Fighting of
See also:
References 106, 117, 214, 242
192.
Furnace Design
193.
194.
195.
196.
197.
198.
199.
THE "TURBINE" FORCED-DRAUGHT FURNACE
Anon
Engineering 110 (2863), 639-40 (12 November 1920)
BURNING A LOW GRADE OF FUEL ON UNDERFEED STOKERS
Anon
Power 1£, 247-51 (15 August 1922)
THE GILCHRIST BAGASSE FURNACE
Anon
Intern. Sugar J. ~, 257-60 (1922)
THE MODERN STEAM-JET FURNACE
Anon
Gas J. 163 (3149), 870-2 (19 September 1923)
THE MODERN TURBINE FURNACE
Anon
Eng. and Boiler House Rev. 35 (5), 149-50 (December 1923)
THE MODERN TURBINE FURNACE
Anon
Eng. and Boiler House Rev. lZ (5), 198-200 (January 1924)
FURNACES FOR BURNING WOOD REFUSE
Anon
Power Plant Eng. 28 (1), 68-70 (1 January 1924)
STOKERS AND FURNACES
Anon
Combustion 11 (5), 372-7 (Nov. 1924)
VII-21
Arthur D Little, Inc
-------
6.
Furnace Des ign (con t . )
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
211.
TYPES AND DESIGNS OF BAGASSE FURNACE
Anon
Planter Sugar Mfr. 12 (15), 287-9 (9 October 1926)
TYPES AND DESIGNS OF BAGASSE FURNACE -- II
Anon
Planter Sugar Mfr. 12 (16), 309-12 (16 October 1926)
PROPER BAFFLING INCREASES BOILER EFFICIENCY
Anon
Power Plant Eng. 38, 40-1 (January 1934)
OVERFEED STOKERS -- SPREADER AND INCLINED-GRATE
Anon
Power 80 (9), 476-7 (1936)
DISCUSSION ON "DOWNJET COMBUSTION"
Anon
J. Inst. Fuel ~ (129), 160-2 (January 1950)
DAMPFKESSELFEUERUNGEN (Stearn-Boiler Furnaces)
H. Berner
Z. Ver. deut. Ing. 65 (15), 371-5 (9 April 1921)
BOILER FURNACE CONSTRUCTION ACCORDING TO BERGMANS
Breidenbach
Deut. Zuckerind. 46, 346-8 (1921)
BAGASSE FURNACES
A.G. Budde
Intern. Sugar J. ~, 211-4 (1920)
THE AERODYNAMIC APPROACH TO FURNACE DESIGN
J.H. Chesters
Trans. Am. Soc. Mech. Engrs. 81, 361-70 (October 1959)
BESSERER FLAMMENAUSBRAND 1M FEUERRAUM DURCH FLAMMENW1RBELUNG
VERFAHREN, MOGLICHKE1TEN UND BETRIEBSERGEBNISSE (Superior Flame
Combustion in the Furnace by Means of Flame Turbulence)
Karl Cleve
Arch. Warmewirtsch. u. Dampfkesse1w. 20 (6), 149-53 (1939)
BAGASSE FURNACES
F. Coxon
Intern. Sugar J. ~, 496-502 (1917)
BAGASSE FURNACES
Frank Coxon
Intern. Sugar J. 20, 10-6 (1918)
VII - 2 2
Arthur D Little, Inc
-------
6.
Furnace Design(cont.)
212.
213.
214.
215.
216.
217.
218.
219.
220.
221.
222.
COMBUSTION SPACE AND SETTING HEIGHT FOR STOKER FIRED BOILERS
IS STUDIED
R.C. Cross, R.A. Sherman, and H.N. Ostborg
Heating, Piping, Air Conditioning 11 (11), 687-8 (1939)
NOTES ON THE COMBUSTION OF BAGASSE
J. Eigenhuis
Intern. Sugar J. ~, 474-7 (December 1937)
A NEW APPLICATION OF PERFECTLY STIRRED REACTOR (P.S.R.) THEORY
TO DESIGN OF COMBUSTION CHAMBERS
R.H. Essenhigh
U.S. Dept. of Navy, ONR - Power Branch, Contract Nonr 656(29),
Tech. Rept. FS67-1 (u) March 1967
DEVELOPMENT OF FUNDAMENTAL BASIS FOR INCINERATOR DESIGN EQUATIONS
AND STANDARDS
Robert H. Essenhigh and Ta-jin Kuo
Prepared for HEW by Combustion Laboratory, Dept. Materials
Science, Penn. State Univ., Tech. Rept. FS/PHS 8/69-4, August 1969
SPREADER STOKERS FOR FIRING STEAM PLANT BOILERS
Fairmount Coal Bureau
Nat1. Engr. 50 (6), 458-62 (1946)
EXPERIMENTS WITH FURNACES FOR A HAND-FIRED RETURN TUBULAR BOILER
Samuel B. Flagg, George C. Cook, and Forrest E. Woodman
U.S. Bureau of Mines Tech. Paper No. 34, 1914, 29pp
BAGASSE FEEDERS, FURNACE DESIGN, AND FURNACE CONTROL
A. Gartley
Louisiana Planter ~, 25-8 (1919)
BAGASSE-FURNACE DESIGN AND CONTROL
A. Gartley
Intern. Sugar J. ll, 232-6 (1919)
THE CONVERSION OF DISTILLERY BOILER-PLANTS TO LIGNITE-FIRING
Hermann Gesell
Z. Spiritusind 43, 251-2 (1920)
HOW TO BURN LIGNITE
Edward Green
Power ~, 322 (June 1935)
DIE WIRTSCHAFTLICHKEIT DES TORF-DAMPFKESSELBETRIEBES (The
Economy of Peat-Boiler Operation)
A.H.W. Hellemans
Feuerungstech. IV (11), 126-31 (1916)
VII - 2 3
Arthur D Little, Inc
-------
6.
Furnace Design (cont.)
223.
224.
225.
226.
227.
DOWNJET COKE FIRING FOR SMALL STEAM GENERATORS
F.B. Karthauser and G.C.H. Sharpe
J. Inst. Fuel ~ (129), 24-6 (January 1950)
UTILIZATION OF BAGASSE. XII. DRAUGHT FOR BAGASSE BOILERS
Haruji Kato
Cellulose Ind. (Tokyo) 13, 327-32 (1937)
BAGASSE FURNACES
E.W. Kerr
Trans. Am. Soc. Mech. Engrs. 61 (8), 685-91 (NovembeT 1939)
NEW GERMAN TECHNICAL DEVELOPMENT IN THE USE OF LOW-GRADE FUELS
B. Kramer
Engineering 142. 240 (28 August 1936)
TREND IN DESIGN AND OPERATION OF INDUSTRIAL PLANTS,
REFERENCE TO FURNACE VOLUME
H. Kreisinger
Proc. Eng. Soc. West. Penn. 45. 426-40 (1929)
WITH SPECIAL
228. COMBUSTION EXPERIMENTS WITH NORTH DAKOTA LIGNITE
Henry Kreisinger, C.E. Augustine, and W.C. Harpster
. U.S. Bureau of Mines Tech. Paper No. 207. 1919, 41pp
229.
230.
231.
232.
233.
234.
THE COMBUSTION OF COAL AND DESIGN OF FURNACES
H. Kreisinger, C.E. Augustine, and F.K. Ovitz
U.S. Bureau of Mines Bull. No. 135, 1917, 137pp
STAND U. ENTWICKLUNGSZIELE DER MOD ERN EN STEINKOHLENFEUERUNGSTECHNIK
(State and Development of Modern Coal Firing Practice)
W. Kretschmer
Intern. Bergwirtsch. u. Bergtech. ~, 169-72 (15 August);
192-6 (15 September); 211-15 (15 October 1931)
THE DEVELOPMENT OF A DESIGN OF SMOKELESS STOVE FOR BITIMINOUS COAL
B.A. Landry and R.A. Sherman
Trans. Am. Soc. Mech. Engrs. ~, 9-17 (1950)
DER NEUE MODERNE FEUERUNGSROST (New Modern Grate)
J. Lauf
Der Bergbau 38 (19), 329-33 (6 May 1925)
BURNING WOOD WASTE SMOKELESSLY
Richard B. Lemkuh1
Nat1. Engr. 53 (4), 15 (1949)
DESIGN AND OPERATION OF SPREADER STOKERS
William S. Major
Combustion l! (1), 41-3 (July 1949)
VIl-24
Arthur D Little, Inc
-------
6.
Furnace Design (cont.)
235.
236.
237.
238.
239.
240.
INFLUENCE OF CHEMISTRY UPON IMPROVEMENT IN STOKER DESIGN
C.H. McClure
E1ec. Rev. li, 620-1 (1919)
MODERN STOKERS DUE TO RESEARCH, WELL DESIGNED FOR CURRENT DEMAND
T. F.J. Moffett
Heating and Ventilating 40 (7), 62-4 (1943)
BURNING LOW-GRADE COALS OF THE SOUTHWEST
W.M. Park
E1ec. World ~ (17), 947-8 (24 April 1920)
DESIGNING BOILER BAFFLES
A.W. Patterson, Jr.
Combustion 11 (2), 122-4 (August 1924)
BOILER-FURNACE DESIGN
Edwin B. Ricketts
Mech. Eng. 45 (5), 299-302 (1923)
PROBLEMS IN BOILER FURNACE DESIGN
E.B. Ricketts
Power Plant Eng. ~ (9),473-6 (1 May 1924)
241. THE BURNING OF COKE BY THE DOWNJET METHOD
F.F. Ross and G.C.H. Sharpe
J. Inst. Fuel 23 (129), 20-4 (January 1950)
242. PEAT COMBUSTION PRACTICE IN THE U.S.S.R.
Scientific Experimental Institute of Peat (Russia)
Trans. Fuel Conf. World Power Conf. 1928, London 1, 1187-1210 (1929)
243. THE COKE-FIRED DOWNJET FURNACE IN INDUSTRY
G.C.H. Sharpe
J. Inst. Fuel ~ (129), 27-31 (January 1950)
244. HOW LARGE A FURNACE?
W.A. Shoudy
Power 90 (7), 460-2 (1946)
245. THE ART OF PARTIAL MODELING
D.B. Spaulding
Ninth Symposium (International)
1962, Academic Press, 1963, pp.
on Combustion, Cornell University
833-43
246. THE USE OF LOW GRADE FUELS ON TRAVELLING GRATE STOKERS
B.M. Thornton
Eng. and Boiler House Rev. 61 (2), 40-2 (August 1946)
VII-25
Arthur 0 little, Inc
-------
6.
Furnace Design (cont.)
247.
248.
249.
250.
251.
7.
GENERAL PRINCIPLES OF FURNACE CONSTRUCTION AND DESIGN
W. Trinks
Gas Age-Record 58, 419-22 (September); 453-6 (2 October 1926)
FURNACE DESIGN, VOLUME I, 5th edition
W. Trinks and M.H. Mawhinney
John Wiley & Sons, Inc., New York, 1961
BOILER-ROOM CHANGES DOUBLE OUTPUT OF EXISTING UNITS
George H. Urban
Power 86, 711-13 (October 1942)
A CHEAP IMPROVEMENT FOR EXISTING BAGASSE FURNACES
C.J. Van Ledden Hu1sebosch
Arch. Suikerind ll, 442-6 (1919)
EXPERIMENTAL STUDIES OF INCINERATION IN A CYLINDRICAL COMBUSTION
CHAMBER
Murray Weintraub, A.A. Orning, and C.H. Schwartz
U.S. Bureau of Mines Rept. Invest. No. 6908, 1967
See also:
References 21, 38, 40, 60, 131, 141, 156, 170, 175, 181,
252.
Furnace Gas Flow
:253.
254.
255.
256.
AERODYNAMICS AND COMBUSTION
Frederick Alton
Eng. and Boiler House Rev. ~, 37-41 (1942)
LES FOURS A FLAMMES (The Circulation of Hot Gases in Furnaces)
H. Drouot
Tech. Mod. 14, 151-7 (1922)
AERODYNAMICS OF THE OPEN-HEARTH FURNACE AS A MEANS OF CONTROLLING THE
PROCESSES OF COMBUSTION AND HEAT EXCHANGE
M.A. Glinkov
J. Iron Steel Inst. 196 (111), 1-14 (September 1960)
ISSLEDOVANIE PROTSESSA TURBULENTNOGO GORENIYA C UCHETOM
VTORICHNYKH REAKTSIYI (Investigation of Turbulent Combustion
with Calculation of Secondary Reactions)
S.A. Go1'denberg
Izvest. Akad. Nauk SSSR Otde1. Tekhn. Nauk (5), 657-66 (1951)
THE FLOW OF GASES IN FURNACES
W.E. Groume-Grjimai10
John Wiley & Sons, Inc., New York, 1923
VII - 26
Arthur D Little, Inc
-------
7.
Furnace Gas Flow (cont.)
257.
258.
259.
260.
261.
262.
263.
264.
265.
GRUNDLAGEN DER STROMUNGSTECHNIK DES INDUSTRIEOFENS
(Principles of Flow Techniques in Industrial Furnaces)
Michael Hansen
Arch. Eisenhuttenw. (11/12), 337-44 (November/December 1949)
FLOW OF GASES IN FURNACES
E.L. Harris
M.S.C.E. Thesis, Massachusetts Institute of Technology, 1925, 50pp
RECIRCULATION, AND ITS EFFECTS IN COMBUSTION SYSTEMS
A.B. Hedley and E.W. Jackson
Combustion 1I, 41-8 (December 1965)
THE EFFECT OF GAS FLOW PATTERNS ON RADIATIVE TRANSFER IN
CYLINDRICAL FURNACES
H.C. Hottel and A.F. Sarofim
MIT Preprint #489, no date given
STROMUNGSTECHNISCHE FRAGEN 1M DAMPFKESSEL- UND FEUERUNGSBAU
(Aerodynamic Problems in Boiler and Furnace Design)
W. Marcard
Warme 2£ (19), 291-4 (13 May 1933)
PROPORTIONING CHIMNEYS ON A GAS BASIS
A.L. Menzin
Paper No. 66578n, Presented at the Annual l1eeting,
Society of Mechanical Engineers, December 1915
American
STROMUNGSTECHNISCHE BETRACHTUNGEN 1M FEUERUNGS- UND DAMPFKESSELBAU
(The Hydromechanical Viewpoint in the Construction of Furnaces and
Boilers)
F. Michel
Feuerungstech.12 (23-4), 233-8 (15 December 1930)
REPRESENTATION AND EVALUATION OF RESIDENCE TIME DISTRIBUTIONS
Pinhas Naor and Reuel Shinnar
Ind. Eng. Chern. Fundamentals ~ (4), 278-93 (Nov. 1963)
DER STROMUNGSVORGANG IN DER BRENNKEMMER VON ROSTFEUERUNGEN.
EIN BEITRAG ZUR BERECHNUNG DER STROMUNGSVORGANGE AUF GRUND VON
MODELLVERSUCHEN (The Flow Process in the.Combustion Chamber on
Grate Firing. A Contribution to the Calculation of Flow Processes
from Model Experiments)
L. Schiegler
Z. Ver. deut. Ing. ~ (29), 849-55 (16 July 1938)
See also:
References 133, 208, 248
VII - 27
Arthur D Little, Inc
-------
8..
Miscellaneous
266.
267.
268.
FLUID FLOW THROUGH GRANULAR BEDS
P. C. Carman
Trans. Inst. Chern. Eng. 15, 150-66 (1937)
NOTES ON HEATING -- FURNACE ECONOMY AND OPERATIONS
M. H. Mawhinney
Metal Progr. 28 (4), 33-8 (1935)
DRAFT
Joseph G. Worker
Combustion 6 (5), 232-4, 240 (1922)
VII-28
Arthur D Little, Inc
-------
B.
ABSTRACTS OF FOREIGN ARTICLES
3.
MIT WELCHEM UNTERDRUCK WIRD DIE ZWEITLUFT DEM FEUERRAUM
AM ZWECKMAESSIGSTEN ZUGEFUEHRT? (At What Pressure
Should Secondary Air Be Supplied to a Furnace?)
Anon
Warme 61 (5), 97 (29 January 1938)
The pressure at which secondary air should be supplied
to combustion chamber is briefly discussed. Results
are given of investigations carried out on a sectional
boiler with 600 sq.m. heating surface equipped with a
zone controlled force draft stoker. It is shown that
nozzles are much more effective than simple wall
openings. [Eng. Index, 131 (1938)]
14.
UBER DIE WIRKUNGSWEISE VaN RUSSBLASERN (On the Way in Which
Soot Blowers Work)
K. Cleve and R. Muller
Arch. Warmewirtsch. u. Dampfkesselw. 21 (1), 17-9 (1940)
For many years soot blowers have been used for the operation
of boilers in order to avoid slag depositions and fly ash
agglomerations on the heating area. However, very little
is known about the flow and pressure history of the medium
which is blown - mostly overheated vapor - after it has
left the soot blowers. In this paper some tests are reported
which explain these conditions. In addition, the highest
possible jet velocity which is permissible without endangering
the sidewalls is briefly discussed.
20.
DIE TECHNIK DER ZUFUERUNG VaN VERBRENNUNGSLUFT (The Technique of
Supplying Combustion Air)
R. Fehling
Arch. Warmewirtsch. u. Dampfkesse1w. 11 (4), 119-23 (1930)
Thermal and aerodynamic conditions of combustion air admission
are discussed; nature of turbulence, its influence on combustion,
and relation between air admission and heat transfer in furnace.
[Eng. Index, 189 (1930)]
VII-29
Arthur D little, Inc
-------
24.
ZWEITLUFTZUFUHRUNG BEl ROSTFEUERUNGEN (Secondary Air for Grate
Firing)
Wilhelm Gumz
Feuerungstech. ~ (11), 123-4 (1935)
Among the topics discussed are the necessity for secondary
air for larger grate firings, preconditions as related to the
effect of secondary air, amount and spatial distribution of
the air, choice of pressure and temperature and its influence
on the penetration, advantages of secondary air.
25.
ZWEITLUFTZUFUHRUNG. DIE DUSENANORDNUNG UNTER BESONDERER
BERUCKSICHTIGUNG DES SYSTEMS BADER (Secondary Air Supply.
Arrangement with Special Regard to the Bader System)
Wilhelm Gumz
Feuerungstech. 30 (2), 32-6 (15 February 1942)
Jet
The advantages and implications of a secondary air supply
are discussed along with the amount of secondary to be used,
the influence of the pressure, influence of the jet diameter
and the local arrangement of the secondary air jets. The
Bader system is described. Comments are made on the intro-
duction of secondary air immediately over the coke layer and
introduction through a layer regulator which is built in the
form of a hollow tube.
27.
VERSUDHE AN FLAMMROHR-INNENFEUERUNGEN MIT ZWEITLUFTZUFUHRUNG
(Experiments with Secondary Air Supply in a Flame Tube Internal
Furnace)
H. Janissen
Warme 62 (12), 205-7 (25 March 1939)
The smoke and soot formation, which in recent years has become
noticeable again by the increased demand from the present boilers,
occurs particularly often if flue boilers [cornwall boilers]
with normal plane grate internal firings are used. To abolish
this shortcoming, several firms produce special provisions for
secondary air supply and offer them to be built into such
firings. Experiments with such installations yield a survey of
the way of their operation and the possible savings if the
required preconditions are given. It must be pointed out,
however, that before the installation, appropriate measure--
ments have to show whether an improvement can be achieved by
secondary air supply.
VII - 30
Arthur D Little, Inc
-------
29.
THEORETISCHES UBER ZWEITLUFTZUFUHR BEl ROSTFEUERUNGEN
(Theory of Secondary Air Supply for Grate Firing)
P. Koessler
Arch. Warmewirtsch. u. Dampfkesselw. 19 (6), 153-6 (1938)
For some time, the secondary air supply for grate firings
has received much attention; however, it seems that the
possible firing techniques which are noted are not always
clear. Therefore, the author treats here the problems of
secondary air supply according to processes in the combustion
chamber. In a second paper [Reference No. 30], the author
outlines the state of research and the problems which are
still to be solved.
30.
ZWECKMABIGE ZWEITLUFTZUFUHR BEl FEUERUNGEN- STAND DER FORSCHUNG
UND ZUKUNFTIGE AUFGABEN (Appropriate Secondary Air Supply in
Firing. State of Research and Future Problems)
P. Koessler
Arch. Warmewirtsch. u. Dampfkesselw. ~ (7), 169-73 (1938)
Secondary air is well apt to mix the gases in the combustion
chamber. By its application it is possible to achieve a far-
reaching improvement in the combustion process on the
traveling grate and to better adjust the firing to fuel and
load. With a suitable secondary air arrangement, it is
certainly possible to decrease the height of the combustion
chamber and increase the combustion chamber output. Suffi-
cient energy of the secondary air jets is the main require-
ment for a favorable effect. The amount of secondary air and
exit velocity can be guessed from the known experimental re-
sults and they can be calculated approximately from load, type
of fuel, and primary air supply. Secondary air amounts of
about 10% of the total air and velocities from 50 meters per
second upward have been shown to be effective. The influence
of the height of the nozzle installation above the grate has
not been investigated sufficiently. However, it is senseless
to install the nozzles very high. A downward direction of the
nozzles is advisable although here too, except from the theory
of Gumz, no exact investigations have been made. It is ad-
visable to install the nozzles in the front or rear wall of
the combustion chamber [not the sidewall]. Nozzles which blow
against one another, especially in a dislocated arrangement,
are undoubtedly good.
VII-3l
Arthur D Little,lnc
-------
36.
ZWEITLPFTZUFUHRUNG ZUM EINHALTEN DES WIRTSCHAFTLICHEN C02-
GEHALTS DER RAUDHGASE (Secondary Air Supply to Keep an Economic C02
Content of the Off-Gases)
Obering. Mortensen
Warme 60 (49), 799-801 (4 December 1937)
A new firing technique from Scandanavia is introduced and
experimental results from several plans are given. The new
technique consists of dividing the combustion process into
two combustion processes, in which the first stage is a
partial combustion with partial gasification and the complete
combustion takes place in the second stage by the use of
overfire air.
39.
SULLE POSSIBILITA DI MIGLIORARE IL RENDIMENTO DELLE CALDAIE
INSUFFLANDO ARIA SUPPLEMENTARE SOPRA IL COMBUSTIBILE (On the
Possibility of Improving Efficiency of the Boiler by Blowing
Supplementary Air Over the Fuel)
Antonio Rasi
Energia termica l, 202-6 (1939)
This paper reports some results obtained in combustion tests
carried out at the Instituto di Fisica Tecnica of the Uni-
versity of Padova. The boiler used was provided with a
regulator for controlling forced overfire air. The advantages
of its use are discussed.
42.
GASBLANDNINGENS INFLYTANDE PA FORBRANNINGSHASTIGHETIN I FLAMMOR
(The Influence of Gas Mixing on Combustion Velocity in Flames)
John Rydberg
Feuerungstech. 30 (11), 257-9 (15 November 1942)
From his investigations, Rydberg concludes that a high energy
consumption does not guarantee a good mixing result. It is
better to mix the secon~ary air into the combustion chamber through
many different nozzles in order to promote the mixing if the
minimum secondary air pressure and velocity can be kept. It
is particularly advantageous if the gas flow can be narrowed
down simultaneously by the addition of secondary air. Thereby,
the flame length can be shortened, a condition which otherwise
can only be obtained by a larger amount of excess air.
43.
WANDERROSTFEUERUNG MIT WIRBELLUFTZUFUHRUNG (Traveling Grate with
Turbulent Air Supply)
W. Schultes
Arch. Warmewirtsch. u. Dampfkesselw. 16 (5), 117-8 (1935)
A modern form of secondary air supply for traveling grate
firing is described. It uses sharp, narrow aerojets of high
VII-32
Arthur 0 Little, Inc
-------
.---- - I. .--
penetrating power for the whirling of combustion gases.
This method offers advantages for the elimination of smoke,
soot, and unburned gases.
57.
DIE WIRKUNGSWEISE VON WIRBELLUFTDUSEN (How Jets Promote Turbulence)
Karl Cleve
Feuerungstech. li (11), 317-22, 362 (1937)
The state of the knowledge is given of the characteristics
of air which is introduced by means of "vortex air jets" into
the combustion chamber. New experiments are described which
centralize velocity, and the expansion of the jet after the
exit from the air nozzle is discussed. The experimental
results have been applied and a proposal made for an appro-
priate supply of turbulent air.
58.
K VOPROSU 0 RASCHETE OSTROGO DUT'YA (On Overfire Air Jet Calculations)
1. K. Nairmark
Sovet. Kotloturbostroenie 2, 253-7 (1939)
The article gives formulae for calculating the trajectory of
an over fire air jet of a fixed size, the initial speed of the
jet and the direction of the nozzle. The trajectory of the
jet curls when there is a variation of specific gravity of
the air blast and hot gases, and also due to the action of
furnace gases on the jet stream.
67.
TURBULENZ (Report of
BERICHT UBER UNTERSUCHUNGEN ZUR AUSGEBILDETEN
Investigation of Developed Turbulence)
L. Prandtl
Z. angew. Math. u. Mech. l, 136-9 (1925)
In this article the author describes a model for the turbulent
flow near the wall. As an extension of the well-known model
of the turbulent mixing, the author outlines the procedure
to calculate free turbulent flow. The term free turbulent flow
is used for flows without bordering walls. An outline is given
only for stationary flows. The solution of the fundamental
differential equation will be published later from Dr. Tollmien.
VII - 3 3
Arthur D Little,lnc
-------
68.
69.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN. I. THEORETISCHE VORBEMERKUNGEN (The
Influence of the Mixing Process on the Combustion of Gases,
Fuel and Air in Furnaces. I. Introductory Theoretical Remarks)
Kurt Rummel
Arch. Eisenhuttenw. 10 (11), 505-10 (May 1937)
The different states of combustion and degrees of combustion
are explained and the importance of the combustion mechanism is
stressed. Visual flame and gas analysis are normally in-
sufficient to describe the degree of combustion. Therefore,
the state of combustion is described by the figures L-bar
for the amount of air, G-bar for the amount of unburned gas,
R-bar for the amount of reacted gas. Using these figures,
the areas of the flame are described.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN TElL. II. VERSUCHE AN DER BRENNERSTRECKE
(The Influence of the Mixing Process on the Combustion of Gases,
Fuel and Air in Furnaces. II. Experiments with the Firing)
Von Kurt Rummel
Arch. Eisenhuttenw. 11 (1), 19-30 (July 1937)
From previously described experiments, it can be derived
that:
a.
The velocity of the mechanical mixing is decisive for
the combustion in normal industrial furnaces. When-
ever gas and air mix, they burn with a velocity which
is infinitely greater than the one obtained normally.
b.
As a result of this, the influence of the temperature
on the velocity of the combustion in the furnace
should not be shown between 900 and 1300°. Probably,
there is no influence of the temperature even in
larger temperature limits, for example, between 600°
and the dissociation temperature. At high tempera-
tures we have a stronger expansion of the burnt gases
which requires an accordingly increased space for the
mixing.
c.
Design of the body of the furnace has an influence on
the shape of the space in which the mixing processes
take place.
d.
Space and time required for combustion decrease with an
increase in the difference of the velocity of the para-
llel jets of a burner with gas and air. This is due
to the increased turbulence at the boundary of gas and
air.
VII-34
Arthur D Little, Inc
-------
e.
Space and time required for combustion
a decrease of the exit velocity of gas
the burner as less energy is available
process.
increase with
and air from
for the mixing
f.
Excessive air lessens the combustion of the gases with
respect to space and time during the whole combustion
process, not only at its end.
g.
The space in which the combustion of gas and air takes
place, that is, the space in which air, gas and burnt
gas can be simultaneously found in the analysis, forms
a cone which starts at the mouth of the burner. Ex-
cessive gas or air diverge from this burner.
h.
The buoyancy forces are very small in a vivid jet of
given specific weight in the body of the furnace in
which the temperature differences are on the order of
a few hundred degrees.
i.
If an air jet lies on top of a gas jet of lower speci-
fic weight and the difference of the specific weight
counteracts the buoyancy forces, this appearance can be
used to guide the flame. The difference of the specific
densities causes the stronger mixing and shortens the
space and time required for combustion. The composi-
tion of the "atmosphere" is extremely different as
long as the mixing of gas and air is incomplete. It
should be determined whether flat burners should be
used instead of round burners in cases where the re-
ducing or oxidizing atmosphere is warranted at parti-
cular points of the furnace.
70.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN TElL. III. MODELLVERSUCHE UBER DIE MISCHUNG
VON GAS- UND LUFTSTRAHLEN (The Influence of the Mixing Process
on the Combustion of Gases, Fuel and Air in Furnaces. III. Model
Attempt for the Mixing of Gas and Air Jets)
Von Kurt Rummel
Arch. Eisenhuttenw. 11 (2), 67-80 (August); (3), 113-23 (September);
(4), 163-81 (October~937)
This paper deals with the following topics:
Similarity of the Models - the relationship between the
mixing factors and the composition of the gas and air
atmosphere; The Experimental Plan - the influence of the
dimensions, the load, the velocities, the jet direction and
repulsion, the influence for burners of the design of the
SM-furnaces, before the entrance to the combustion chamber,
of the dead corners of the combustion chamber, etc.
VII-35
Arthur D Little, Inc.
-------
71.
DER EINFLUSS DES MISCHVORGANGS AUF DIE VERBRENNUNG VON GAS UND
LUFT IN FEUERUNGEN TElL. IV. NACHPRUFUNG DER ERGEBNISSE DER
MODELLVERSUCHE (The Influence of the Mixing Process on the
Combustion of Gasest Fuel and Air in Furnaces. IV. Summary of
the Results of the Furnace Investigations)
Von Kurt Rummel
Arch. Eisenhuttenw. 11 (5)t 215-24 (November 1937)
The theory which was developed in References 68-70 was
examined in a furnace and found to be correct in practice.
AdmittedlYt the judgment of the practical cases was often
difficult. Only fundamental cases could be examined in this
work. Most of tent howevert numerous influences work to-
gether and oppose one another in industrial furnaces. The
aerodynamics of firing is the result of complicated relations.
However, in addition to the investigations of fundamental
cases, the work has shown the applicability of the model
attempts and representst thereforet an important mean. The
investigators design a simple burner for gas which allows,
in a wide range of furnace conditions, the quick adjustment
of the flame length and of the atmosphere over the furnace
bottom.
72.
LAMINARE STRAHLAUSBREITUNG (Laminar Jet Expansion)
H. Schlichting
Z. angew. Math. u. Mech. ~t 260-3 (1933)
Theoretical mathematical analysis of laminar spread of flui.d
jett issuing from small rectangular or circular orifice, within
surrounding fluid. [Eng. Index, 479 (1933)]
77.
BERECHNUNG TURBULENTER AUSBREITUNGSVORGANGE (Calculation of
Turbulent Expansion Processes)
Wal ter Tollmien
Z. angew. Math. u. Mech. ~ (6)t 468-78 (1926)
The article treats the turbulent mixing of the homogeneous
air jet with its surrounding air.
81.
UBER DIE STROMUNGSVORGANGE 1M FREIEN LUFTSTRAHL (On the Flow
Processes of Free Air Jets)
W. Zimm
VDI-Forsch. Gebiete Ingenieurw. 11 (234), 5-35 (1921)
A free air jet is investigated whose velocity is low enough
so that the compressibility of the air does not have to be
regarded. The expected flow processes are derived in a
theoretical way from an expanding nozzle whose outer walls
are increased and taken away in the process of transition
VII-36
Arthur D Little, Inc
-------
102.
11.9.
to a free jet. The "secondary air movement" occurs as
a characteristic appearance of the free flow. Description
is made of the instruments used in the experiment to produce
the air jet and to measure the velocity. (The pitot stagnation
instrument and electrical heat wire instrument are from
Professor Weber). For precise measuring of the flow in the
downstream region, only the sensitive heat wire instrument is
used which is calibrated with a pitot-tube and which is con-
tinuously re-examined. Experimental investigation of the
turbulent jet region goes up to 3.5m actual width and up to
.1m/sec velocities. The analysis and plotting of the achieved
experimental results with respect to the flow process, the
participating amounts of air, and the kinetic energy were
disturbed by the secondary air which influences the measure-
ments in the outer boundaries by vortexes. The experimental
results prove the theoretical assumption of a secondary air
movement in the case of a free air jet. The amounts of
secondary air increase with an increasing distance from the
center and an increased velocity of the primary air and dis-
charge the kinetic energy of ' the original jet. The initial
energy is transferred with considerable losses onto the
secondary air jet. This one represents then with its single
elements a new primary gas which continues its energy dis-
sipation into the space. The general validity and the broad-
ening of the appearances, which have been found for a jet with
a circular cross-section, is now considered. The picture,
which has been found in the framework of these investigations
for the flow processes in a free unhindered expanding air jet
of low velocity can be valuable for the purposes of the tech-
nique and can be regarded as a preparatory work for the complete
analysis of the free air jet.
GRAPHISCHES BERECHNUNGSVERFAHREN FUR MEHRSTUFIGE DAMPFSTRAHLAPPARATE
(Graphical Method for Computing Multiple Steam-Jet Installations)
Ladis1aus Zimmermann
Chem.-Ing.-Tech. ~ (11), 665-71 (1953)
A comprehensive treatise containing nomographs relating
important operating variables. Examples are given for a
and five-stage unit. [Chern. Abstracts 48, 1736 (1954)]
four
BETRIEBSERFAHRUNGEN MIT DEM NEUEN TURBINENROST
Experiences with New Turbine Grate)
Anon
Arch. Warmewirtsch. ~ (11), 299-301 (1925)
(Operating
Describes new furnace and grate design for uniform air
distribution over the entire grate; results show that this
furnace is very suitable for low grade fuels. [Eng. Index,
84 (1925)]
VII - 3 7
Arthur D Little, Inc
-------
122.
1.27.
131.
FEUERUNGSTECHNIK DES WANDERROSTES EINFLUSSE VON BRENNSTOFFt
LUFT UND ROSTBAUART (The Firing Technique of the Traveling Grate.
The Impact of Fuelt Air and Grate Design)
Karl Beck
Arch. Warmewirtsch. u. Dampfkesselw. 20 (4)t 93-8 (1939)
This paper is written mainly for men from power stations
who want to achieve a clearer view of the present state-of-
the-art of combustion processes on traveling grates and the
influences of varying plant conditions. Factors which are
mainly regarded while designing the grate will not be con-
sidered here. The article discusses some findings of A.R.
Mayer and is, in additiont a very good introduction to the
operation of the traveling grate.
ROSTWIDERSTAND VERSCHIEDENER KOHLENSORTEN (Combustion Resistance
of Various Types of Coal)
Roman Dawidowski
Z. oberschles. bert- u. huttenmann. Ver. Katowice 65 (lO)t 660-7
(October); 728-34 (November 1926)
Discussed flow of air of combustion in fuel bed for various
kinds of coal; adaptability of grates to coals; gas friction in
fuel beds; velocity of flow, results of experiments. [Eng.
Indext 82 (1926)]
PHYSIKALISCHE THEORIE DER VERBRENNUNG (Physical Theory of
Combustion)
W.H. Fritsch
Warme 60t 749-57 (13 November); 768-73 (20 November 1937)
Factors influencing the efficiency and rate of combustion in
a restricted combustion space, e.g., a boiler furnacet are
discussed. The velocity of the chemical reactions involved is
such that the time necessary for their completion is only about
0.001% of that actually taken. The observed velocity must
therefore be determined by physical factorst e.g't by the rate
of mixing of combustible with air. The mechanism of air pene-
tration through a fuel bed and of the mixing of gases or
liquids by turbulence is discussed in relation to data afforded
by experiments with models. The introduction of secondary air
is not an effective method of improving the mixing of gases in
a combustion space. It is concluded that more efficient combus-
tion in a boiler furnace is to be effected only by radical
changes in furnace-grate designt thus permitting better mixing
of combustible gases. By overcoming the physical resistance to
combustion, much higher rates of heat release per hr. per cu.
m. of combustion space could be obtained. [Chern. Abstracts 33t
2310 (1939)] -
VII-38
Arthur D Little, Inc
-------
137.
138.
143.
144.
DER VERBRENNUNGSVORGANG IN DER WANDERROSTFEUERUNG
of Combustion with Chain Grates)
W. Gumz
Feuerungstech. ~t 10-2 (1936)
(The Process
Meier investigated the combustion process on a traveling
grate. Experiments were made in a normal industrial furnace.
Thust the mistakes or side influences of small lab instru-
mentation could be avoided. The grate was operated in the
zoneless way and secondary air was added in various small
amounts. Brown coals were used. In the dissertationt
the percentages of coalt volatile compounds and ash have
been determined and measurements of the C02 concentration
in the combustion chamber have been made. It was concluded
that the speed of combustion and pyrolysis mainly depends
on the heat transfer within the layers.
ZONENEINTEILUNG UND ZWEITLUFTZUFUHRUNG BEl WANDERROSTEN (Division
into Zones and Secondary Air Supply for Chair Grate Firings)
Wilhelm Gumz
Feuerungstech. 30 (ll)t 256-7 (1942)
The advantages of the division into zones has been illustrated
by showing the composition of the combustion gas in the body
of the furnace. Complete burnout for lignite could not be
obtained and the additional help of secondary air was employed
to reach a complete burnout.
DE LA COMBUSTION
Materials)
Victor Kammerer
Bull. Soc. indus.
DES MATIERES VOLATILES (Combustion of Volatile
Mulhouse ~ (2)t 111-33 (1926)
A discussion of the mechanism of combustion on mechanical
stokerst in the light of the results of a certain number of
boiler testst with practical indications on conditions re-
quired to obtain optimum results with coals having medium
and high volatile contents. [Chern. Abstracts 20t 2241 (1926)]
ZUR AERODYNAMIK DER BRENNSTOFFSCHUTTUNG IN ROSTFEUERUNGEN (On the
Aerodynamics of Fuel Charging in Grate Firing)
H.G. Kayser
Forsch. Gebiete Ingenieruw. B. ~ (2), 89-100 (March-April 1935)
Aerodynamics of fuel layer in grate furnaces; theoretical
considerations; mulitgrain mixtures; fuel layer as discon-
tinuous system; it is claimed results of investigation can
be applied to other fields of engineering where flow through
granular materials playa role. [Eng. Index, 105 (1935)]
VII-39
Arthur D Little, Inc
-------
146.
151.
153.
154.
VERSUCHE AN WANDERROSTFEUERUNGEN (Experiments
Furnaces)
W. Koeniger
Arch. Warmewirtsch. 10, 243-8 (1929)
on Chain-Grate
In earlier experiments with traveling grate plants, variation
in load had almost no influence on the degree of efficiency
of the grate firings if the pressures were not varied sub-
stantially. The efficiency varied only slightly under con-
ditions ranging from half-load to 25% overload. The in-
fluence of storage heat on the load variations, grate and
combustion chamber loads, law of draft resistance and change
of draft by buoyancy forces are discussed. [Chern. Abstracts
23, 4106 (1929)]
DIE SCHUTTHOHE
Grate Firing)
A.R. Leye
Brennstoff- u.
EINER ROSTFEUERUNG (The Height of the Bed in
Warmewirtsch. 1I (2), 15-21 (1935)
Regulation of load on boiler grates by depth of fuel bed and
passage of air; laws governing combustion process in coke layer;
furnace should be designed to dispense as far as possible with
secondary air admission. [Eng. Index, 105 (1935)]
VERBRENNUNGSVERLAUF VON STEINKOHLE AN EINER WANDERROSTFEUERUNG
(The Combustion of Bituminous Coal on a Traveling Grate)
Rud Loewenstein
Warme~, 97-101 (17 February); 121-5 (24 February 1934)
An experimental apparatus and an experimental procedure have
been developed for traveling grate firings. Experiments were
carried out with different fuel layer heights at equal grate
loads on the traveling grate of a 30-atmosphere-vertical-tube
boiler with natural draft. Measurements were made of the air
temperature under the grate, the grate rod temperatures, the
air velocity and the gas composition over the layer. The
findings of the combustion process are, compared with the
results of former investigations of experimental firings.
Conclusions for zone division and underwind blast pressure
for gas to coal ratio and advantages and disadvantages of
large fuel layer heights are given.
BEITRAGE ZUR FEUERUNGSTEDHNIK VON STEINKOHLEN AUF DEM WANDERROST
(Contribution to Firing Practice of Bituminous Coal on a
Traveling Grate)
Marcard
Warme ~, 397-401 (11 June 1932)
See Reference No. 154
VII - 4 0
Arthur D Little, Inc
-------
155.
156.
160.
161.
BEITRAGE ZUR FEUERUNGSTEDHNIK VON STEINKOHLEN AUF DEM WANDERROST
(Contribution to Firing Practice of Bituminous Coal on a
Traveling Grate)
Marcard
War me 55, 417-22 (18 June 1932)
Contributions to technique of firing bituminous coals on
traveling-grate stoker development of traveling grate for
high-capacity combustion equipment; influence of fuels,
air distribution and furnace design on ignition and combustion.
[Engineering Index~, 1282 (1932)] .
DIE VERBRENNUNG ALS STROMUNGSVORGANG (Combustion as a Flow Process)
W. Marcard
Warme 60, 257-66 (24 April 1937)
Topics discussed are: the importance of fluid dynamics for
the combustion technique; general form of the combustion
process; individual combustion processes; change of state for
liquid and gaseous fuels; air flow under and in the grates;
elimination of combustion gases and mixing; off-gas explosion,
explosion weight; percentage of parts participating in the
heat transfer; calculation of temperatures at furnace end;
new ways of boiler design.
DIE WIRKUNG DER ZWEITLUFT IN DER WANDERROSTFEUERUNG (D84)
(The Effect of Secondary Air on Traveling Grate Firing)
Albert R. Mayer
Z. bayer. Revisions-Ver. ~ (4), 31-3 (February 1938)
This reference covers the first part of a thesis discussion
from Mayer. It deals with combustion processes on the
traveling grate firing. It points out that the combustion
process can only take place after gas and air have come to'-
gether by processes like diffusion, turbulence and "free
whirl formation." These processes are discussed.
DIE VORGANGE 1M FEUERRAUM EINES KESSELS MIT WANDERROSTFEUERUNG UND
IHRE ANDERUNG DURCH ZWEITLUFTZUFUHR (The Reactions in a Boiler
Furnace and Their Alteration with Twin Air Supply)
Albert R. Mayer
Feuerungstech. 1£ (5), 148-50 (1938)
It is shown by analyses of gas samples taken from numerous
positions in the combustion chamber that conditions therein
are unsatisfactory with a great load of 80-90 kg/sq m/hr.,
corresponding to 75% of the normal boiler capacity. The gas
composition can be made uniform throughout the chamber by
VII-41
Arthur D Little, Inc
-------
162.
166.
introducing secondary air, which also accelerates the
combustion can be obtained by suitable adjustment of the
quantity and velocity of the secondary air, thus affording
a means of controlling combustion and increasing the effi-
ciency of the combustion chamber. [Chern. Abstracts 33,
4405 (1939)]
UNTERSUCHUNGEN UBER ZWEITLUFTZUFUHR IN WANDERROSTFEUERUNGEN
(Investigation of Secondary Air Supply in Traveling Grate Firings)
Albert R. Mayer
Feuerungstech. 26 (7), 201-10 (1938)
Gas composition is determined in the body of a furnace
burning coal with high and low grate load. Also noted are
the caloric values of the combustion gases, the influence
of secondary air on the burnout of the combustion gases,
relationship of secondary air to the required height of
the combustion chamber and side effects of secondary air.
VERFEUERUNG GASREICHER KOHLE IN EINER NEUZEITLICHEN WANDER-
ROSTFEUERUNG (Combustion of Gas-Rich Coal on a Modern Traveling
Grate)
Albert Muller
Z. Reichshauptst. Techn. Uberwach. 1 (11/12), 61-72 (13 June);
(13/14), 76-81 (11 July 1942)
On account of the numerous experiments and investigations,
it has been realized that the combustion process on the
grate as well as in the combustion chamber of modern traveling
grate firings has still to be improved, particularly if gas-
rich brands with long flames are used. Experiments were
conducted with this in mind. These experiments were carried
out to determine the advantage of the zone regulation of the
traveling grate or the combustion process on the grate and
the burnout of the gas in the combustion chamber versus the
zone length operated grate. In addition, the influence of
the secondary air on the combustion of the gas in the com-
bustion was investigated. Experiments show that it is parti-
cularly difficult to achieve complete combustion while using
gas-rich coals in high combustion chambers of modern radiator
boilers with short ignition bodies at an economical air excess.
The role of the combustion chamber height was to be investigated
with respect to the burnout of the combustion gases for primary
air as well as secondary air operation of the firing. In order
to answer these different problems, a special Bavarian coal
with 36% volatile compounds was burned in a modern traveling
grate firing and the combustion process on the grate as well
as in the combustion chamber was observed. To achieve this,
the composition of the coal was analyzed at different loads
from the layer regulator to the slag stopper. In addition,
VII - 4 2
Arthur D Little, Inc
-------
167.
the pyrolyzing gases from the coal layer were analyzed in
several heights above the grate. Thereby) it was shown to
be efficient to compare certain tests with zone regulation
with one of the same loads but without the zone regulation
[from W. Meier]. Knowledge of coal and gas analysis allows
one to theoretically obtain) most importantly) the amount of
air which enters through the grate and to obtain, especially)
the adaptation of the air throughput to the locally required
amount of air for the whole grate length theoretically. In
this way) the great advantage the regulated air supply to the
fuel has over the zoneless grate operation is shown. In addi-
tion) the total heat history of the coal burned on the grate
and in the combustion chamber can be shown. Secondary air on
the gas burnout was investigated for larger load variations
as a function of amount and entrance velocity by gas samples
from different heights above the grate. The findings of the
new investigations allow one to determine which amounts of
secondary air and which air entrance velocities achieve a
favorable operation for gas-rich coals in a traveling grate
firing. Finally) proposals are made for the choice of the
average combustion chamber height.
VERIEUERUNG GASREICHER KOHLE IN EINER NEUZEITLICHEN WANDERROSTFEUERUNG
(Combustion of Gas-Rich Coal in Modern Traveling Grate Firing)
Albert Muller
Feuerungstech. 31 (3») 68-9 (1943)
The plant of the Technical University of Munich) which has been
investigated by W. Meier - dissertation Munich) 1935) as a
zoneless traveling grate firing, has now been investigated with
zone division and with secondary air supply. This was done in
order to distinguish the impacts of the zone division and the
secondary air supply. With zone division the burnout of the
fuel layer is substantially different and travels more to the
entrance than could have been expected. The processes in the
combustion chamber are improved but the mixing effect is not
yet sufficient to prohibit that in a combustion chamber) which
is 5.75 meters high. Unburned gas can be found in the last
measuring plant which is 4.65 meters over the grate. By addi-
tion of secondary air which amounted to 20% of the total air
supplied, the CO could not be abolished completely but it
decreased substantially. The result was improved with 30%
secondary air. The effect of the secondary air was decreased
with an increased combustion chamber load. However) the en--
trance velocity of the secondary air was from 10.9 to 32.6
meters per second which is obviously too low. In spite of
this) we could increase, under these conditions and with a
moderate air excess) the combustion chamber load to 295)000 kcals
per cubic meter and hour) and we could decrease the combustion
chamber height to 4 meters. The zone division loses a little
of its importance if secondary air is supplied into the com-
bustion chamber.
VII-43
Arthur D Little, Inc
-------
170.
1;12.
177.
DIE GRENZEN DER FEUERRAUMBELASTUNG UND IHRE RUCKWIRKUNG AUF
DIE AUSLEGUNG DES KESSELS (Limits of the Combustion Chamber
Load and the Resulting Effect on the Design of the Boiler)
W. Pauer
Arch. Warmewirtsch. u. Dampfkesselw. 20 (8), 197-202 (1939)
The problem of the combustion chamber load which is determined
by the burning time of the dust particles and the allowed com-
bustion chamber temperature is solved substantially for coal
dust firings. The applicability of the relations developed
for coal dust firings to gas and oil firings has not been
pointed out clearly. The generalization that the permissible
load is only a time function, which is valid for gas- and
oil-firing, is not confirmed. It is the purpose of this
paper to examine the validity of the famous permissible load
formulas and clarify their applicability for the different
kinds of firings.
DIE VERBRENNUNG VON BRAUNKOHLEN AUF DEM ARBATSKY-WANDERROST
(Combustion of Brown Coal on the Arbatsky Traveling Grate)
E. Praetorius
Braunkohle 30, 241-7, 266-9 (1931)
Tests at the Power Station Gruenberg on a provisionary
Arbatsky-grate have shown that the combustion of brown coal on
traveling grates with a high grate space and width load is not
only possible, but that width loads can be obtained which up
to now could only be obtained from coal firing and coal high
performance grates. It should not be difficult to improve the
degree of efficiency over one obtained under unfavorable test
conditions, from 63% to about 80% and higher. It will be
possible to obtain, with the use of the Arbatsky grate for
brown coals and other low-caloric fuels, 200 square meter
heat transfer surfaces for one meter width of boiler and it
will be possible to achieve heat transfer surface outputs
of 50-60 kilograms per square meter.
ZUR PHYSIK DER VERBRENNUNG FESTER BRENNSTOFFE (The
Combustion of Solid Fuel)
P. Rosin and H.-G. Kayser
Z. Ver. deut. Ing. ~ (26), 849-57 (27 June 1931)
Physics of
The following topics are discussed: physical relationship
between the process of coal combustion and the solution of
solid bodies; model experiments for dissolving salt in
water currents; weight and time of the reactions for single
bodies; the aerodynamics of the grate; the behavior of multiple
body heats at high air velocities; instability and resistance
to fly coke; aerodynamics of the dust firing; limiting load
VII-44
Arthur D Little, Inc
-------
178.
179.
and particle size; load regulation as a function of the
amount of air; fuel layer and grate steering, regulation of
fire zoning; air regulation according to conditions of the
combustion process and behavior of the ash.
DIE RAUMLICHE UND ZEITLICHE ENTWICKLUNG DER VERBRENNUNG IN
TECHNISCHEN FEUERUNGEN (The Change of Combustion with Respect
to Time and Space in Industrial Furnaces)
Kurt Rummel and Hellmuth Schwiedessen
Arch. Eisenhuttenw. ~ (12), 543-9 (June 1933)
The combustion process is not sufficiently described by
measuring temperature, velocity and composition; parti-
cularly the state of the combustion in one point or in
one plane of the furnace cannot be predicted. Qualitative
description is only possible by introducing a new definition
for the degree of combustion in one point and by plotting
the curves of equal degree of combustion. Mathematical
derivations show, in addition, that qualitative descriptions
can be made too if the degree of combustion is related to
the volume which is enclosed by areas of equal degree of
combustion [volume of reaction]. Additional derivations
allow predictions of the velocity with which the combustion
proceeds to the practical degree of combustion [average re-
lated load] and the combustion time and weight. The deriva-
tions are explained and proved by practical examples.
STAND UND ENTWICKLUNG DER FEUERUNGSTECHNIK. EIN QUERSCHNITT UND
UMRISS UBER DIE FORSCHUNG AUF DEM GEBIETE DER FEUERUNGSTECHNIK
IN DEN LETZTEN JAHREN (State and Development of Firing Techniques.
A Summary of Research in the Field of Firing Techniques in Recent
Years)
Fr. Schulte and E. Tanner
Z. Ver. deut. Ing. 12 (21), 565-72 (27 May 1933)
The heater tube imperical development of the firing technique
requires an extension towards the physical side of the up-to-
now almost purely chemical combustion research. New knowledge
of the combustion process is discussed by limiting ourselves
mainly to the discussion of coal. The chemical reaction of
the layer and the air required in the individual steps of
combustion are discussed in a way suited for the practical
preparation of firings. The processes of ignition and igni-
tion throughout the fuel layers are discussed, and important
results for cold dust firing are summarized. Results supported
by weight and temperature measurements are mentioned for the
whirling of the ash from the grate, the formation of slag,
and the wear of the grate. It was found that a certain mini-
mum content of ash is necessary for the burning of the coal
on the grate to achieve low wear-and-tear of the grate. The
VII - 4 5
Arthur D Little, Inc
-------
180.
188.
189.
flow pattern in the body of the furnace and in the fuel
layer, the gas velocity, the gas composition and the
temperature distribution are briefly discussed.
VERBRENNUNGSPROBLEME IN BRENNKAMMERN VON HOCHLEISTUNGSDAMPFKESSELN
(Combustion Problems in Combustion Chambers of High Performance
Boiler Furnaces)
Karl Schwarz
Brennstoff-Warme-Kraft 1 (2), 45-52 (May 1949)
Problems created in a high performance boiler furnaces,
like increased danger of clogging and corrosion, are viewed
according to firing technique. These problems result from
the increased use of fuels with a high content of ash and water.
In particular, those cases are discussed which are due to
incomplete combustion. The flow and mixing problems in the
off-gas combustion air flow are explained for two practical
cases, that of the modern grate boiler and that of the dust
boiler.
VERBRENNUNGSVERLAUF
(Combustion of Coal
Helmut Werkmeister
Arch. Warmewirtsch.
BEl STEINKOHLEN MITTLERER KORNGROBEN
of Medium Particle Size)
u. Dampfkesselw. 11 (8), 225-32 (1931)
The methods used hitherto for calculating the firing technique
and examining the fuel chemistry produced imprecise knowledge
of the process of combustion of solid fuels. For the design
and operation of firing, it is, however, very important to
control the combustion process. In an extensive test series,
the combustion of nut coal with different gas contents was
investigated with a special process. The results are given
in the form of characteristic burning lines as a function of
the combustion time. Approximate formulas for calculating
the technical combustion process have been used.
WINDVERTEILUNG UND FEUERGASBESCHAFFENHEIT BEl WANDERROSTFEUERUNGEN
(Blast Distribution and Combustion Gas Composition in Traveling
Grate Firing)
H. Werkmeister
Z. Ver. deut. lng. ~ (26), 788 (30 June 1934)
Blast pressure is discussed as a function of height of the
coke layer, particle size and combustion of ~he layer.
VII-46
Arthur D Little, Inc
-------
191.
205.
209.
222.
BEKAMPFUNG DER VERSCHLACKUNG VON DAMPFKESSELN (Fighting of
Slagging in Boiler Furnaces)
Arthur Zinzen
Brennstoff-Warme-Kraft 1 (3), 63-8 (1950)
The general improved melting diagram for fuel ashes allows
one to judge about the behavior of ashes. The chemical re-
actions of combustion have been investigated for the luminous
flame and the zones in which the unburned gases, the sulfur
compounds, which originate in the flame, and the fly coke
have to be burned. The conclusions for the planning and the
carrying out of boiler furnace firings are meant to show how
to prevent boiler slagging.
DAMPFKESSELFEUERUNGEN (Steam-Boiler Furnaces)
H. Berner
Z. Ver. deut. Ing. ~ (15), 371-5 (9 April 1921)
Among the topics discussed are: the changed requirements of
furnaces; fuels of low quality as seen from the furnace tech-
nical viewpoint; the combustion process and loss of output
with fuels of low quality; the possibility of the general
furnace; mechanical draft; variation in furnace losses;
the different types of grates.
BESSERER FLAMMENAUSBRAND 1M FEUERRAUM DURCH FLAMMENWIRBELUNG
VERFAHREN, MOGLICHKEITEN UND BETRIEBSERGEBNISSE (Superior Flame
Combustion in the Furnace by Means of Flame Turbulence)
Karl Cleve
Arch. Warmewirtsch. u. Dampfkesselw. ~ (6), 149-53 (1939)
Greater furnace capacity can be obtained by the use of a
turbine-type burner, by the constriction of the furnace
cross section above the grate, and by the use of high-
pressure secondary air, properly directed. These all depend
on good mixing of the flame parts. A diagram is given showing
air velocities in front of a three-part turbine-type burner,
when the parts are used separately and together. [Chern.
Abstracts 34, 241 (1940)]
DIE WIRTSCHAFTLICHKEIT DES TORF-DAMPFKESSELBETRIEBES (The Economy
of Peat-Boiler Operation)
A.H.W. Hellemans
Feuerungstech. IV (11), 126-31 (1916)
Proper design and operation of boilers for burning fuel is
discussed. [Eng. Index, 259 (1916)]
VII-47
Arthur D Little, Inc
-------
230.
232.
253.
255.
STAND U. ENTWICKLUNGSZIELE DER MODERNEN STEINKOHLENFEUERUNGSTECHNIK
(State and Development of Modern Coal Firing Practice)
W. Kretschmer
Intern. Bergwirtsch. u. Bergtech. 24, l69~72 (15 August);
192-6 (15 September); 211-15 (15 October 1931)
The ignition and combustion operation of the coal is first
explained in order to ascertain the proceedings in the
firing. Finally the different constructions of coal firings
are dealt with and it will be seen to what extent these
already comply with the previously laid down requirements.
DER NEUE MODERNE FEUERUNGSROST (New Modern Grate)
J. Lauf
Der Bergbau 38 (19), 329-33 (6 May 1925)
Details of turbine furnace, a forced draft furnace regulated
by means of steam-jet nozzle; manual stoking; claimed to
possess all advantages of economic operation. [Eng. Index,
86 (1925)]
LES FOURS A FLAMMES (The Circulation of Hot Gases in Furnaces)
H. Drouot
Tech. Mod. 14, 151-7 (1922)
W.E. Groume-Grjimailo works out the theory of the circulation
of the hot gases in furnaces by analogy from the laws of
hydraulics. Small-sized sections of furnaces have been built
and enclosed between parallel glass plates, and the laws veri-
fied experimentally by means of water and colored petroleum
(representing the light hot gases), taking into account the
facts that the hot gases and air are miscible, and that their
density varies with temperature. The findings have been
applied to furnaces already in operation or under construction,
and on the whole they have given satisfactory results. Modern
practice in furnace. construction is discussed in the light of
this theory, showing what features are defective and how to
improve them. [Chern. Abstracts 16, 1997 (1922)]
ISSLEDOVANIE PROTSESSA TURBULENTNOGO GORENIYA C UCHETOM
VTORICHNYKH REAKTSIYI (Investigation of Turbulent Combustion
with Calculation of Secondary Reactions)
S.A. Gol'denberg
Izvest. Akad. Nauk SSSR Otdel. Tekhn. Nauk (5), 657-66 (1951)
Turbulent heterogeneous combustion, as carried out experi-
mentally by flow of 0 at 500-1000° through cylindrical C tubes,
VII-48
Arthur 0 Little, Inc
-------
257.
261.
263.
265.
secondary reactions of
Under these conditions
4.3 x 107 exp-(-28,500/RT).
is affected significantly by the
C02 reduction and CO combustion.
the coefficient of gas formation =
[Chem. Abstracts 46, 2775 (1952))
GRUNDLAGEN DER STROMUNGSTECHNlK DES lNDUSTRIEOFENS
(Principles of Flow Techniques in Industrial Furnaces)
Michael Hansen
Arch. Eisenhuttenw. (11/12), 337-44 (November/December 1949)
The factors determining the flow of heat and gases, namely,
continuity, turbulence, viscosity, Reynolds number, and
design factors, are treated theoretically and methods for
their determination are discussed. [Chem. Abstracts 44,
4294 (1950))
STROMUNGSTECHN1SCHE FRAGEN 1M DAMPFKESSEL- UND FEUERUNGSBAU
(Aerodynamic Problems in Boiler and Furnace Design)
W. Marcard
Warme 56 (19), 291-4 (13 May 1933)
The importance of aerodynamics as an auxiliary science is
stressed. The fundamental laws of aerodynamics are used in a
model experiment in which the investigated media are water
vapor, air and combustion gases. The experiments have been
carried out in important parts of the boiler.
STROMUNGSTECHNlSCHE BETRACHTUNGEN 1M FEUERUNGS- UND DAMPFKESSELBAU
(The Hydromechanica1 Viewpoint in the Construction of Furnaces and
Boilers)
F. Michel
Feuerungstech. 12 (23-4), 233-8 (15 December 1930)
Designers of boilers and furnaces have not learned to apply
modern discoveries .in the field of fluid flow. Proper atten-
tion to these principles should make possible better utili-
zation of the available combustion volume reduction in friction
losses and an improvement in heat transfer. [Chem. Abstracts
25, 1708-9 (1931))
DER STROMUNGSVORGANG IN DER BRENNKEMMER VON ROSTFEUERUNGEN. EIN
BEITRAG ZUR BERECHNUNG DER STROMUNGSVORGANGE AUF GRUND VON
MODELLVERSUCHEN (The Flow Process in the Combustion Chamber on
Grate Firing. A Contribution to the Calculation of Flow Processes
from Model Experiments)
L. Schiegler
Z. Ver. deut. lng. 82 (29), 849-55 (16 July 1938)
Examination of the combustion gas flow in all the furnace
VII-49
Arthur D Little, Inc
-------
firings is a difficult task for which the general solution
has not yet been' found. In this papert which will be con-
tinued in a second articlet an approximate solution has been
given for certain flow conditions resulting from model in-
vestigations. This solution has been confirmed by practical
experience and allowst under certain preconditions, the
calculation of given flow problems in the firing technique.
VII-50
Arthur D Little, Inc
-------
APPENDIX A
EQUATIONS FOR INVISCID NON-CONDUCTING FLOW IN TWO ZONES
Arthur D Little, Inc
-------
r-------
APPENDIX A
EQUATIONS FOR INVISCID NON-CONDUCTING
FLOW IN TWO ZONES
Referring to Figure 111-1 in Chapter III, we will assume that:
1.
Pressure is constant across Section a.
Z.
ula and uZa are different but constant across Section a in
each zone.
3.
h is constant across Section a.
PI is less than Pz as the temperature is higher in Zone 1.
Because of Assumptions 1, Z and 3, the total head, H, though differ-
ent j.n the two zones is constant everywhere within each zone. At Section b,
the streamlines are assumed parallel, so that the quantity (p + pgh) is
constant across the section; and as a result, the velocity is also constant.
Consldering the streamlines just on each side of the interface between
the two zones, conditions at Section b can be related to those at Section a
by the following equations:
4.
2 2
ulb ula (Pa - Pb)
2g=2g+
PI g
gc - (Yb + AZb)
(A-I)
Z Z
ulb ula (Pa - Pb)
-=-+
2g Zg P Z g
gc - (Yb + AZb)
(A-Z)
Conservation of mass flow rate yields the following equations:
ula Ala PI = ulb (A - AZb) PI
(A-3)
uZa AZa Pz = uZb AZb P2
(A-4)
Equations (A-I) through (A-4) relate the variables ulb' u2b' AZb
and J) and can be used to determine these quantities, given the
condi.~ions at Section a. Combination of the equations to eliminate
all variables but AZb results in:
A-I
Arthur D Little, Inc
-------
A~b [-(P2 - P1)g] + Aib [(~2a - q1a) - (P2 - PI)g(Yb - 2A)]
3
+ A2b [-2A(q2a - q1a)
- (P2 - P1)gA(A - 2Yb)]
2 2
+ w 1 - w 2 - (p - P ) gy A 2 ]
2p 1 2p 2 2 1 b
(A-5)
+ A;b [A2(q2a - qla)
2 2A2
w2A w2
+ A2b [-] - - = 0
P2 2P2
where:
(q2a - qla)
2
1 w2
= "2 [ 2
P2A2a
2
w1
2 ]
PlAIa
WI = flow rate per unit width in Zone 1.
w2 = flow rate per unit width in Zone 2.
Solution of the above equation can be accomplished with a standard computer
program for determining the roots of polynomials. Such a program has been
used to obtain the results shown in Table 111-2.
A-2
Arthur D Little, Inc
-------
APPENDIX B
INSTABILITY IN STRATIFIED FLOW
Arthur D Little, Inc
-------
APPENDIX B
INSTABILITY IN STRATIFIED FLOW
As noted in Section C-l of Chapter III, in a linear feed incinerator,
thE~ hottest gases are produced very near the input end of the burning
grclte, while the gases at the discharge end are cool because of the large
peI:cent excess air. If the flow of overfire gases is in the same direc-
ticm as the fuel bed, the gases will naturally stratify with the hot,
fUE~l-rich components forming a blanket along the roof of the combustion
chamber. This hot blanket would also have a rather high velocity because
of the accelerations produced by the rise to the roof (see Section D of
Chapter III). An important question is whether the stable density
distribution will significantly inhibit mixing of the hot gas with the
cocller, oxygen-rich gases below.
The question can be answered by considering the stability of a
stratified fluid of infinite extent with an initially parallel velocity
field, as in Figure B-1. We are primarily interested in examining the
competing effects of density in promoting stability and shear in generating
instability, so it will be sufficient, at first, to take a simple two-
strata model with the shear confined to a single interface. It is a con-
venient approximation to take the dense fluid to be infinitely deep.
ThE! mathematical solution to this problem is a modification of that given
in Reference 2 pp. 159-163. The modification consists of inserting an
upper boundary and dropping the effect of interface surface tension.
In each of the two regions, the disturbance velocity field is irro-
tational and may be subsumed in a potential function. These functions
arE~ denoted by
-------
y
--
AOl-' 16-370
upper boundary
u2
Yo
..
interface
x
.
density. P'2
u2
FIGURE B-1
SCHEMATIC OF STRATIFIED FLOW CONDITION
B-2
Arthur D uttle.lnc
-------
~---
At: the upper boundary, the vertical velocity component must vanish
a
-------
t:, PI e2Ky 0 + 1 PI
G = - = - coth Ry 0
1 P2 e2Kyo - 1 P2
(B-9)
and another describing the effect of gravity acting on the density differ-
ential:
G ~.& (p 2 - PI)
2 K P2
(B-lO)
The solution of the quadratic is given by:
.£=
K
~.~ Glul ~ - Gl (ul -
1 + Gl
2
u2) + (1 + Gl) G2
(B-ll)
Stability of the stratified flow is characterized by disturbances
which propagate but do not grow in time, i.e., by real a. If a has an
imaginary component, then the flow is unstable. From Equation (B-ll) ,
it is clear that the shear, ul - u2' contributes to instability while a
positive G2 contributes to stability. The threshold of stability is found
when the sum of terms inside the square root is zero, or when
2 1
(ul - u2) = (1 + G; )G2
1
P2
= (1 + - Tanh
PI
.& (P2 - PI)
KYo)K P
2
(B-12)
If we regard PI' P2 and (ul - u2) as specified parameters, then (B-12) repre-
sents a conditIon on wavenumer, K. The wavenumber which satisfies (B-12)
is a critical wavenumber which we denote by K. For K > K , the waves are
unstable. A sketch of the growth rate (imagiaary part of g) versus Kyo
PI 2
is shown in Figure B-2 for - = .25 and (ul - u2) /gyo = 6.
P2
This result is unrealistic for large K because it predicts growth rate
increasing without limit. The unreality is due to the assumption of an
infinitely thin shear layer. What is needed is a problem specification which
includes a continuous velocity profile and density profile as well. The solu-
tion of such a problem is not generally possible in closed form and requires
a computer calculation. Since any velocity or density profiles would be highly
conjectural, it does not seem worthwhile to pursue this in detail. There are,
B-4
Arthur D Little, Inc.
-------
o
KIU1 - U2]
0.4-- - - - - - - - -
1/2
(p l/P 2)
= .4
1 + P1/P2
---
o
.25
1
FIGURE B-2
GROWTH RATE OF DISTURBANCES IN NON-ISOTHERMAL
PARALLEL FLOW
B-5
ADL -116-370
Kyo
Arthur D Little.lnc
-------
however, certain well-known results which can be invoked to obtain an
wlderstanding of the relevant physics. For a fluid of uniform density,
it is known1 that if a shear layer is spread out over a thickness 1jJ,
then all very small disturbances (very high K) are stable. There is a
high wavenumber cut-off, K ,above which stability is found, and K
~ co co
= 1/1jJ. Even relatively large density differences have very small ettect
on this result.
The question at hand, then, is whether K is approximately equal to
co
(or less than) K found previously, in which case the combined effects
of density stratIfication and finite thickness of shear layer stabilize
disturbances over the entire range of sizes. A rough answer can be given
rather simply. Assuming that the dime~sion of the shear layer is roughly
equal to the dimension h, we have Kco = l/yo. If we introduce a charac-
tE!ristic length
v =
2
(u1 - u2)
2g
(B-13)
then (B-12) can be written
P2 2 P2
1 + -- tanh(K Yo) = ~ ( ) (Kcyo)
P1 c Yo P2 - Pl
(B-l4)
and since tanh (K Yo) is never greater than unity, an upper bound for K
c c
is given by
2 2
P2 - Pl
K Yo <
c - 2 P1P2
Yo
v
(B-15)
In. order for the stratified flow to be completely stable, it is necessary
that
2 2
Yo P2 - Pl
v 2 P lP 2
> 1
(B-l6)
or
2 2
P2 - P1
P1P2
>
2
(u1 - u2)
gyo
(B-l7)
This requirement is a stringent one which will not ordinarily be
satisfied. This can readily be seen if we imagine the dense fluid to
be at rest and the heated fluid to have achieved its velocity, u2' by
rising through a height L. Then
B-6
Arthur D Little, Inc
-------
U 2
1
PI 2 = (PZ - Pl)gL
U = 0
2
(B-18)
and (B-17) becomes
PI + P2
Yo 2 > L
P2
(B-19)
In other words, the total rise must be less than the thickness of the
heated layer iteslf. For most continuous-feed incinerator configurations,
tru~ rise of heated gases is larger than this, so the effect of density
gradient will not eliminate turbulent mixing.
REFERENCES
1.
"Dynamics of Non-Homogeneous Fluids", C.S. Yih, McMillan Company (1965).
B-7
Arthur D Little, Inc.
-------
APPENDIX C
DECAY OF SHEAR VELOCITY IN A CHANNEL
Arthur D Little, Inc
-------
APPENDIX C
DECAY OF SHEAR VELOCITY IN A CHANNEL
To obtain an estimate of the rate of shear velocity decay as a result
of turbulent mixing, we assume that the momentum exchange across the mid-
planes of unconfined and confined jets are the same. For simplicity, we
consider the approximations to Gaussian profiles shown in Figure C-I: M-M
is the midplane. We will treat a "linearized" problem, where density and
velocity gradients are relatively small compared to average values.
The momentum decrease in the high velocity region of the unconfined
shear layer in the distance dx and, hence, the momentum transfer across
M-M, is proportional to the shaded area in Figure C-l. Therefore,
the rate of momentum transfer is proportional to
1 ul - u2 db
- 2" ( 2 ) dx
(C-I)
Silularly, for the confined shear layer, the rate of momentum transfer
across M-M with distance downstream is proportional to
[(!:. - b) + E.] du - 1 (A - b) . du
2 2 dx-2 dx
( C- 2 )
For the same fluids and velocities in the two cases, the proportionality
constants will be the same, so that our assumption that momentum exchange
is the same yields
1 du I ul - u2 db
2" (A - b) dx = - 2" ( 2 ) dx
( C- 3)
For an unconfined shear layer, the rate of spreading is constant, 1. e. ,
db - ~o= constant
dx -
(C-4 )
where ~ois the spreading angle for an unconfined shear layer.
earized case, we can use the approximation
For a lin-
d(UI - uZ)
dx
~ Z du
dx
(C-5)
C-I
Arthur D Little, Inc
-------
db-1
M
~-b
u,
u,
(U,;U,)
FIGURE C-1
I
I
M
(a) Unconfined
M
u,
M
(b) Confined
u,
DIAGRAM IllUSTRATING MOMENTUM
EXCHANGE IN A SHEAR lAYER
C-2
Arthur D Little.lnc
-------
Combining (C-3, 4, and 5), we get
d(Ul - uZ)
(A - b) dx = ~ 0 ( ul - uZ)
or
1
(ul - uZ)
d(Ul - uZ)
dx
= -
~o
b
A(l - -)
A
( C- 6 )
If we assume that a typical Gaussian velocity profile can be represented
by the simple profile of Figure C-l(b), with
~-.!
A - 3
then
1
(ul - uZ)
d(Ul - uZ)
dx
3~o
= - ZA
( C- 7 )
C-3
Arthur D Little, Inc
-------
1---
APPENDIX D
EXPERIMENTAL APPARATUS FOR JET-IN-CROSSFLOW STUDIES
Arthur D Little, Inc
-------
APPENDIX D
EXPERIMENTAL APPARATUS FOR JET-IN-CROSSFLOW STUDIES
Experiments to provide experimental data on the behavior of a gas
jet in a crossflow of lower density gas (as for a cold overfire air jet
discharging into a heated furnace atmosphere) were conducted with the
apparatus shown schematically in Figures D-I and D-2. Figure D-I shows
the low velocity wind tunnel, in which uniform upflow of air is produced
in. a vertical test section. Velocities in the test section ranging from
about 0.5 to 2 FPS could be attained. A jet of nitrogen gas at -320°F
with a density about 3.7 times that of the air was injected horizontally
in.to the test section, as shown in Figure D-l. The cold nitrogen mixing
wi.th the moist room air causes condensation of water vapor and produces
visible flow patterns.
The nitrogen jet was produced with the arrangement shown in Figure
D-.2. By dissipating a measured electrical heat input in the liquid
ni.trogen, a known volume flow rate of saturated nitrogen vapor can be
ge.nerated. Hence, the jet velocity can be readily determined from the
re:lation
Uo =
4q
H
2
P hf 'lTdo
g g
where:
Uo = jet velocity (feet/see)
qH = heater input power (Btu/sec)
p =
g
h =
fg
3
density of saturated nitrogen vapor (lbm/ft )
latent heat of vaporization (Btu/lb )
m
do = diameter of nozzle (feet)
The heat input to the liquid nitrogen, and hence the velocity of the jet,
is controlled by the voltage input to the heater and is easily varied by
mE:ans of the Variac. The arrangement shown in Figure D-2 provides excellent
thermal isolation of the liquid so that boil-off due to ambient heat leak
is very small compared to that produced by the heater, for jet velocities
of interest. The insulated, short flow-length passage from the vapor space in
the dewar to the jet nozzle exit insures minimal heat transfer to the
flowing cold vapor. Temperature measurements show that it issues from
the nozzle at very close to the boiling point temperature.
Air upflow velocities in the test section are measured with a Hastings-
Raydirt hot-wire anemometer whose calibration has been checked against flow
rates determined by pilot tube traverses across the diffuser duct at a plane
nl~ar its inlet.
D-I
Arthur D Little, Inc.
-------
I", 10' .1. 3' --1
1
- 2'
1
TOP VIEW
t1
I
N
»
"'"
~
:r
c
"'"
---
U
C'
~
~
?F"
~
o
Jet
Apparatus
""
Test
Section
Inclined
Manometer
Flow Straight-
ening Grid
Plenum
Pitot
Tube
Di ffuser
1 0% Open
Perforated
Plate
Floor
SIDE VI EW
fiGURE D-1
LOW VELOCITY WIND TUNNEL
-------
Volt Meter
110V-AC
FIGURE D-2
Variac
i~~J~ii:~~;i~;~i~
Seal
Nitrogen
Vapor
--
Liquid Nitrogen
Heater
Nozzle Tube
. v.
J
TEST SECTION
Silvered
Glass Dewar
APPARATUS TO PRODUCE COLD NITROGEN-VAPOR JET
D-3
Arthur 0 Little.lnc.
-------
The front and side panels of the test section are Plexiglass to per-
mit photographing flow patterns of the jet in crossflow (such as shown in
Se,:tion D of Chapter IV) indicated by the condensed water vapor fog.
A photograph of the assembled test apparatus is shown in Figure D-3.
D-4
Arthur D Little, Inc
-------
-,
I
L
\
(
~ .
FIGURE D-3
PHOTOGRAPH OF TEST APPARATUS
D-5
Arthur D Little, Inc
-------
APPENDIX E
DERIVATION OF COMBINED EFFECT MODEL
Arthur D Little, Inc
-------
1-
APPENDIX E
DERIVATION OF COMBINED EFFECT MODEL
The model is based on conservation of momentum in the coordinate
system shown on Figure E-l.
In the axial (y) direction, conservation of momentum requires that:
2 -2
Pj Uj S sinS Uo S sinS.
n no
(E-l)
Rearrangement of Equation (E-l) gives:
2
Ex-
dt - Uj sinS =
Po Uo S sinSo
no
P. Uj S
J n
(E-2)
In the cross flow (x) direction, conservation ot momentum requires
that change in momentum flux be equal to the drag and buoyant forecast-
ing on the volume element dy in thickness.
d Mx = d FxD + d FxB
(E-3)
where:
. 2
Mx = Pj Uj Sn cosS
(E-4)
1 2
d F xD = "2 C x Paul h dy
( E- 5 )
(C is the normal drag coefficient)
x
d FxB ~ g(O a - 0 j) (-.:~~ dy
Substitution of Equations (E-4), (E-5) and (E-6)
(E-3) and integration from y = 0 to Y = Y yields:
(E-6)
into Equation
2 2
P j Uj Sn cos S - Po u 0
y
Sno cosSo = ~ Cx P a uii hdy
o
y
+gf
o
(E-7)
S
( P a - P j) s i: S d y
E-l
Arthur [) Little, Inc
-------
1- -
(a) SIDE VIEW
y
\
\
\~(3
,
u'
J,
p.
J
t t t t
Crossflow
Conditions: Ta U, Pa
(b) TOP VI EW
y'~
Sn
h
FIGURE E-1
TRAJECTORY OF DEFLECTED JET
E-2
~
z
~
Uo
Po
x
Sno
Arthur 0 Little Inc
-------
Rearrangement yields:
dx ( 1
~lt = Uj cosS = \fj Uj
[ y y
SJ t Cx Pa u~ ~ hdy + gS
o 0
S
) n d
Pa - Pj sinS y
+ po
Uo
S
no
coss .J
(E-8)
Division of Equation (E-8) by Equation (E-2) yields the slope of the jet
centerline.
1 C
"2 x Pa
dx -
dy -
2 Jy fy ( ) Sn d
ul hdy + g Pa - Pj sinS y
o -2. 0 +
po
Uo
S
no
sinSo
c + 13
o 0
(E-9)
In order to evaluate Equation (E-9) and allow integration to des-
cribe the jet trajectory, we must establish the functional relationships
hey), Pj(Y) and Sn(y).
For an undeflected jet, Abramovich assumed that:
h = 2.25 do + a2 A
(E-10)
where A is the
value of about
now indication
arc length along the jet path and the constant a2 has a
0.22. Deflected jets spread more rapidly and there is
that for these jets a2 may be as high as 0.32.
The relationship between A and y is unknown at this point.
simplicity, we assume that:
For
'V
A = ay
1
(E-ll)
where slis taken to be a constant of the order of l/sinS. Substituting
E~uations (E-ll) into Equation (E-10) yields (taking a2= 0.22):
h = 2.25 do + 0.22 slY
(E-12)
so that the first integral in Equation (E-9) becomes:
y y
J hdy = J
o 0
(2.25 do + 0.22 alY) dy
(E-13)
2
= 2.25 do y + 0.11 alY
E-3
Arthur D Little, Inc
-------
The second integral in Equation (E-9) can be treated as follows,
us:lng Equation (E-l):
y
J (Pa
o
S
) n d - -2
P j sinS y - P a u 0
S
no
y
sinS f
o
(p a - P j) dy
2 2
P j Uj sin B
(E-14)
We make use of the following relationships which hold for undeflected
jets:
Tj - Tl ~
"'-
To -Tl"'uO
(E-15)
and:
~ = do do
3.2T"" = 3.2-
alY
(E-16)
u.
together with the physical relationship:
Pa = ~
Pj Ta
(E-17)
Substitution of these relationships into Equation (E-14) yields
Y
J (p a - P j )
o
S
n
-dy-
sinS -
Po
S sinSo (T
no a
3.2 do Tl
- T) Yay
1 f 1
2 dy
sin a
o
(E-18 )
Ma.king use of the fact that al~ l/sinS, the integral in Equation (E-18)
ca.n be approximated by:
fy alY '" fy 3
sin2S dy = al ydy
o 0
=
2 2
al y
2
(E-19 )
Substitution of Equations (E-18), (E-19) and (E-13) into Equation
(E-9) yields:
E-4
Arthur D Little, Inc
-------
C
dx - x
--
dy
P U2l (2.25 d y + 0.11 ay2)
a 0
-2
2 P 0 u 0 S sin 8 0
no
(E-20)
+
3 2
g (T a - T 1) a 1 y
.....2
6.4 u 0 doT 1
Integration of Equation (E-20), recognizing that y = 0 at x = 0 yields:
(d:) -
~2' 25 ~x t a u~ d~ ~ f{0)2
4 P au 0 S sint3 0 ~
no
~' 2 2
0.11 C Paulald
+ x 0 +
2
6 P u 0 S sin 80
a no
g (T. -2 T 1) a3 d2j f{0)3
19.2 u 0 doT ~
1
(E-2l)
+ (1:) ceq.
In order to evaluate Equation (E-2l), numerical values for Cx and a must
be selected. The effective drag coefficient Cx is believed to be in the
range of 4 to 5, based on analysis of experimental crossflow data.
Since al~ l/sin~, an average value of sin8 should be used. To des-
cribe the trajectory near the jet mouth, the value of sin80 should be
used.
E-5
Arthur D Little, Inc
-------
APPENDIX F
RESULTS OF FLUE SAMPLING TESTS AT MUNICIPAL INCINERATOR
NEWTON, MASSACHUSETTS
,-
Arthur [) Little, Inc
-------
APPENDIX F
RESULTS OF FLUE SAMPLING TESTS AT MUNICIPAL INCINERATOR
NEWTON, MASSACHUSETTS
It has been suggested (Chapter II) that although a large fraction of
the combustibles are emitted in the pyrolysis zone of the grate, some slow-
burning refuse components continue to burn (or begin to burn) on the dis-
charge grate. The high excess air levels often found over the discharge
grate could lead to quenching of combustion reactions and consequent pollu-
tant carry-over. To obtain an indication of the importance of this source
of combustibles, a test was carried out at the municipal incinerator in
NE~ton, Massachusetts.
Samples of flue gas and gas temperature measurements were taken using
uncooled probes in the flue immediately downsteaam of the furnace (see
Figure VI-I, Chapter VI). The results are reported below:
1.
SUMMARY
a.
CO level is greater near the wall, and at the bottom of the
flue in the low-temperature region.
b.
C02 level is greatest in the center of the duct, and at higher
temperatures.
c.
Temperature fluctuations of greater than 100°F occur in less
than a 5-minute period.
2.
DUCT AND PORT DIMENSIONS
The duct is 11.1 feet across by 12 feet high (outside) - 1 foot thick
wall. Five metal sleeves 12 inches long x 5.5 inches in diameter had been
previously inserted into the sidewall in a vertical row on 2.2 foot centers
with the uppermost port 23 feet from floor level and 11 feet from duct
bottom level.
3.
SAMPLING
Samples were taken from the upper three ports only. The bottom two
ports are below the fly ash level in the duct. Two sets of samples were
taken: a) at 2 feet in from the duct sidewall and b) at 5.5 feet in
from the duct sidewall. The samples were taken simultaneously from the
upper three ports due to rapid temperature fluctuations of about 100°F over
a 5-minute period. Temperatures were monitored by shielded chromel-alumel
thermocouple located at the tip of the gas sampling probe.
F-l
Arthur 0 Little, Inc
-------
4.
ANALYSIS
Analysis was by gas chromatography. The CO content of several samples
was also determined by color indicator tubes made by the Bacharach Indus-
tri.al Instrument Company. Oxygen, nitrogen and carbon monoxide levels
were determined using a Linde 5A molecular sieve column and thermal con-
ductivity detection.
Carbon dioxide was determined using a Porapak Q (cross-linked poly-
styrene) column and thermal conductivity detector. Instrument response to
each gas was determined by external calibration.
5. DATA
b %
ppm CO
GC Bacharach 02 B-2 C02 Sum c Temp
*Ala 540 23.5 83.0 0.14 106.6 ll60°F
(d) A2 250 400 l385°F
(e) A3 55 80 16.8 80.0 3.9 100.7 l470°F
A4 <50 <50 l355°F
(d) Bl 320 12.9 83.8 7.6 104.3 l360°F
B2 55 100 l675°F
(e) B3 <50 17.5 83.8 7.5 108.8 1600°F
B4 l6l0°F
(d) Cl <50 70 l470°F
C2 <50 12.2 77.0 7.0 96.2 1680°F
(e} C3 <50 10.3 79.9 9.1 99.3 1635°F
C4 1655°F
(a)
Two sets were taken at each point. All samples were not analyzed.
Samples identified as 3&4 were taken in center of duct~
(b)
Detection limit = 50 ppm.
Bacharach and GC values.
Note consistent difference between
(c) Total of 02 + N2 + C02' Theoretical = 100%.
(d) Duplicates of samples 'V 1 foot inside wall.
(e) Duplicates of samples about in middle of duct.
*A = Lowest port; C = Highest port.
6.
DISCUSSION
The data show the presence of carbon monoxide (CO) in the low-temperature
gases. Based on reasoning presented in Chapter III, this suggests their
generation in the discharge grate region, perhaps by quenching of the com-
bustion reactions.
F-2
Arthur D Little, Inc
-------
The absence of CO in the hot gases is not conclusive proof that they
did not exist there (in fuel-rich eddies). It would be expected that
at the temperatures noted in the upper regions of the duct (about l600°F)
the~ mixing obtained in the uncooled sampling probe would result in rapid
oxfdation of the CO to C02. For the test program, it should be noted,
water-cooled sampling pro6es will be used.
/
F-3
Arthur D Little, Inc
------- |