-------
TABLE 2-2. NOX EMISSIONS AND TOTAL ANNUAL HEAT INPUT FOR INDUSTRIAL
PROCESS EQUIPMENT BEING CONSIDERED FOR TESTING
NOX Emissions
Gg/y (tons/y)
Total Annual Heat Input
J (Btu)
Natural draft process heaters
Forced draft process heaters
Cement kilns
Steel soaking pits and
reheat furnaces
Glass container furnaces
Wood-bark boilers
93.4 (103,000)
28.1 (31,000)
704 (776,000)
28.6 (31,500)
38.9 (42,900)
141 (157,000)
1.37xl018 (1.3xl015)
211xl015 (200xl012)
513xl015 (486xl012)
538xl015 (SlOxlO12)
105xl015 (99.6xl012)
915xl015 (873xl012)
2-4
KVB 6015-798
-------
On the basis of the data presented here KVB decided to concentrate
its efforts on selecting petroleum process heaters, cement kilns, wood-bark
boilers, and steel furnaces as potential test sites.
Subscale test work has now been completed on a vertically-fired,
rectangular process heater, a rotary dry process cement kiln, and a subscale
steel furnace. The process heater subscale tests were conducted in conjunction
with a major process heater burner manufacturer. The subscale steel furnace
was tested in conjunction with a major steel furnace burner manufacturer.
The research kiln tests were conducted at the laboratory of a major cement
industry association.
Full-scale test sites at which KVB has completed testing include two
wood-bark boilers and a lime kiln. Other units scheduled for testing are
a refinery process heater, a steel furnace, and a cement kiln.
2.2 DEFINING COMBUSTION MODIFICATIONS
A list of candidate combustion modifications was prepared for each of
the three subscale test sites. These modifications involved minor hardware
changes as well as operating changes in the case of the subscale process
heater. Only a limited number of modifications including the injection of
various materials into the flame zone were tried with the subscale cement
kiln because of the small size of the kiln.
2.2.1 Process Heater Combustion Modification Concepts
There are two general approaches that merited consideration in per-
forming combustion modifications on process heaters. In natural draft
devices, the air flow, mixing, and flame shape are intimately interrelated
in that air flow adjustments to alter burner stoichiometry have a negative
impact on the other parameters. Therefore, some consideration was given to
varying burner stoichiometry by controlling the fuel flow and injection
pressure or velocity. Another approach to limiting NO formation in the flame
X
zone is to control the local mixing and flame zone intensity by minor burner
design modifications to achieve more desirable temperature and stoichiometry
conditions.
2-5 KVB 6015-798
-------
In chronological order, the modifications tested in the subscale
process heater were the following:
1. Lowered excess air
2. Low-NOx burners (two designs)
3. Steam injection
4. Staged combustion (two methods)
5. Flue gas recirculation
6. Modified fuel injection
Each of these concepts is defined in Section 4.1 of this report. A
summary of test results for each modification is also given in that section.
2.2.2 Cement Kiln Combustion Modification Concepts
Rotary kilns present a difficult task for combustion modification
because they have only a single burner, and the product quality is very depen-
dent upon temperature. Because of the unique combustion system design and
process operating constraints, combustion modification requires careful con-
sideration of process temperature requirements. Modified combustion system
operating techniques which were evaluated included lowered excess air, and
injection into the flame zone of steam, sulfur, and fly ash. These modifica-
tions are explained along with test results in Section 4.2 of this report.
2.2.3 Steel Furnace Combustion Modification Concepts
The small size and easy accessibility of the research steel furnace
tested on this program made it a logical choice for the generally more
cumbersome modifications of HO injection and flue gas recirculation. The
burner manufacturer also had existing steam and water-injection capability
as well as a suitable fan for the flue gas recirculation tests. There, the
modifications tested in the subscale steel furnace were the following:
1. Lowered excess air
2. Varied heat input rates
3. Steam injection (firing No. 2 oil fuel)
4. Water injection (firing natural gas)
5. Flue gas recirculation (firing No. 2 oil and NG)
2-6 KVB 6015-798
-------
J-t is believed that these modifications were preferable to those
involving burner stoichiometry changes because of product quality considerations.
Although perhaps not quite as much as in a cement kiln, the required operating
conditions in a steel furnace place restrictions on the oxidizing/reducing
characteristics of the flame; hence, there is little flexibility for making
stoichiometry changes, particularly in a single-burner application.
2.3 EXAMINING PROCESS CONSTRAINTS
2.3.1 Natural Draft Process Heaters
The following constraints pose limitations on any modifications
tried.
1. Because of the very small pressure drop across a natural draft
burner, special care must be taken to avoid an unstable flame
when making modifications which alter the structure of the
burning zone.
2. Process heaters generally run at or near full capacity so
that any modification which would result in a lowering of
process rate is undesirable (particularly in retrofit
applications).
3. Flame impingement on walls or tubes should be, avoided.
2.3.2 Cement Kilns
The process constraints are considerably more restrictive for cement
kilns than they are for process heaters. Some of these constraints are
identified as follows:
1. The crystal structures of the cement clinker components tricalcium
silicate, 3CaO SiO (= C S) and dicalcium silicate, 2CaO SiO
( = C S) are a function of temperature profile. The size and shape
of each of the two types of crystals depend upon the residence
times at certain temperatures of the kiln feed mix. C S crystal
size increases with the amount of time spent over 1450 °C. C S
2-7 KVB 6015-798
-------
2.
crystal size increases with the amount of time spent between
1200 °C and 1450 °C. The kiln temperature profile depends on the
flame length. A long flame implies a long residence time between
1200 °C and 1450 °C and a short time over 1450 °C, whereas a short
flame implies a short residence time between 1200 °C and 1450 °C
but a long time over 1450 °C (as shown in Figure 2-1). A long
flame, often preferable to a short flame from the standpoint of
NO emissions, is usually detrimental to product quality in a
X
cement kiln. With a long flame, the C S crystals grow to be too
large and the C S crystals are too small. In addition, the C_S
crystals lose their circular shape and become jagged as a result
of slow cooling in a long flame. These jagged crystals are not
hydraulically active and, therefore, in .a long flame the 28-day
compression strength of the cement can be reduced to as little
as half of the desired value. If the cement strength is too low,
the user must mix more cement with the aggregate in order for
the concrete strength to meet specifications.
When firing oil or natural gas 1-1.5% excess O is generally
maintained. As a rule, when firing coal 1% of primary air is
used for each 1% of volatile matter in the coal.
H
w
w
H
H
J
u
1450 °C
1200 °C
C S grows
KILN LENGTH
Feed
Lame
Figure 2-1. Effect of kiln temperature profile on crystal structure of the
clinker.
2-8
KVB 6015-798
-------
3. The temperatures required in the flame zone are determined by
-the burnability of the feed mix as characterized by four modulae:
the lime saturation factor, silica modulus, iron modulus, and
percent liquid phase modulus. With the first three modulae, the
higher the number, the harder the mix is to burn; hence higher
temperatures are associated with higher modulae values. The
reverse is true of the percent liquid phase modulus, i.e., a
lower value of this modulus implies a harder-burning mix. The
burnability modulae are defined as follows:
1. Lime Saturation Factor (LSF) =
% CaO
2.8(% SiO ) + !.!(% Al 0 ) + 0.72(% Fe 0 )
= .0.87 - 0.95 typically
Silica Modulus (SM) = % M 0% f^Fe 0
^ J £ J
Iron Modulus (IM) = % A12°3
4. Percent Liquid Phase Modulus = 1.13(C A) +1.35 (C AF) + MgO
+ alkalies
= 23 - 26% typically
Note: -C A - = tricalcium aluminate , 3CaO.. Al O
C.AF = tetracalcium alumina ferrite, 4CaO Al_0 Fe_
" £ J £,
(Both are liquids above 200 °C)
An overabundance of the liquid phase will erode the kiln coating.
The burning modulae are measured on the feed mix, but if coal is
burned the fly ash should also be included in determining the
modulae .
2-9 KVB 6015-798
-------
2.3.3 Steel Furnaces
Steel furnace combustion modifications are subject to the following
important process limitations:
1. The temperature in a reheat furnace or soaking pit must be maintained
at a constant level in each zone of the furnace. This level is often
about 1533K (2300°F). Proper temperature distribution assures uniform
heating of the steel with no localized overheating of an ingot, slab,
billet, or bloom.
2. The heating rate in the furnace may not exceed the ability of the
steel to safely absorb heat. Variations in firing rates may cause
problems in furnace heat distribution.
3. The range of fuel-air ratios is limited by scaling problems at the
steel surface. Too much or too little oxygen can result in undesir-
able scale losses.
4. Furnace draft is sometimes used to control heat distribution in
steel furnaces. There may be times when this draft setting will
limit the excess air variations which could be made on a furnace.
2-10 KVB 6015-798
-------
SECTION 3.0
GASEOUS AND PARTICULATE EMISSIONS TEST METHODS AND INSTRUMENTATION
The process heater emission measurements were made with instrumentation
carried in a 32 ft x 8 ft mobile laboratory which was described in detail in
the EPA Report Application of Combustion Modification to Industrial Combustion
Equipment, Contract No. 68-02-2144.
All emission measurement instrumentation was transferred and reinstalled
in a new 8 x 42 ft laboratory trailer. This trailer was used for the tests at
the subscale cement kiln. A plan view of the trailer is shown in Figure 3-1.
The gaseous species measurement's are made with analyzers located in the trailer,
while the particulate, particulate size, smoke spot, and sulfur oxides measure-
ments are made at the sample port, and the weighing and titration are done in
or near the trailer.
The emission measurement instrumentation used is the following:
TABLE 3-1. EMISSION MEASUREMENT INSTRUMENTATION
Species
Manufacturer
Measurement Method
Model
No.
Hydrocarbon
Carbon Monoxide
Oxygen
Carbon Dioxide
Nitrogen Oxides
Particulates
Sulfur Dioxide
Sulfur Oxides
Smoke Spot
Particulate Sizing
Particulate Sizing
Beckman Instruments
Beckman Instruments
Teledyne
Beckman Instruments
Thermo Electron Co.
Joy Manufacturing Co.
DuPont Instruments
KVB Equipment Co.
Bacharach
Andersen 2000, Inc.
Monsanto Chemical
Flame lonization
IR Spectrometer
Po1arographi c
IR Spectrometer
Chemiluminescent
EPA Method 5 Train
UV Spectrometer
Controlled Condensation
ASTM D2156-65
Cascade Impactor
Cascade Impactor
402
865
326A
864
IDA
EPA
400
21-7006
Mark III
BMS-11
3-1
KVB 6015-798
-------
Calibration Gas
. Bottles
Door and Stairs
Spare Calibration
Gas Bottles
Air
Conditioning/
Heater
oooo
r\ Sample Handling/
^ Conditioning
O Room
TT
Counter Top/
lsink | Cabinets
(Instrument
Console
Air Conditioning/Heating Duct and Ven
jj ...nr r
Counter Top/Cabinets
ts
r
'Over
! »
_.
Fume
Hood
OOOO
. £rfibe. stotaas.
/Storage
Room
Fruehauf 42' x 8' Double Axle
Semi Trailer
Figure 3-1. Instrumentation trailer floor plan.
UD
00
-------
3.1 GAS SAMPLING AND CONDITIONING SYSTEM
& flow schematic of the flue gas sampling and analyzing system is
shown in Figure 3-2. The sampling system uses three positive-displacement dia-
phragm pumps to continuously draw flue gas from the stack into the laboratory.
The sample pumps pull from six unheated sample lines. Selector valves allow
composites of up to six points to be sampled at one time. The probes are con-
nected to the sample pumps with 0.95 cm (3/8") or 0.64 cm (1/4") nylon line.
The positive displacement diaphragm sample pumps provide unheated sample gas
to the refrigerated condenser (to reduce the dew point to 35 °F), a rotameter
with flow control valve, and to the O , NO, CO, and CO2 instrumentation. Flow
to the individual analyzers is measured and controlled with rotameters and
flow control valves. Excess sample is vented to the atmosphere.
To obtain a representative sample for the analysis of NO , SO and
hydrocarbons, the sample must be kept above its dew point, since heavy hydro-
carbons may be condensible, and SO. and NO_ are quite soluble in water. For
this reason, a separate electrically-heated sample line is used to bring the
sample into the laboratory for analysis. The sample line is 0.64 cm (1/4-inch)
Teflon line, electrically traced and thermally insulated to maintain a sample
temperature of up to 400 °F. Metal bellows pumps provide sample to the
hydrocarbon, SO and NO analyzers.
, £ X
3.2 INSTRUMENTAL CONTINUOUS MEASUREMENTS
The laboratory trailer is equipped with analytical instruments to
continuously measure concentrations of NO, N0_, CO, CO , 0 , SO , and hydro-
ft 2 ^ ^
carbons. All of the continuous monitoring instruments and sample handling
system are mounted in the self-contained mobile laboratory. The entire system
requires only connection to on-site water, power, and sampling lines to
become fully operational. The instruments themselves are shock mounted on a
metal console panel. The sample flow control measurement, and selection,
together with instrument calibration are all performed from the console face.
3-3 KVB 6015-798
-------
Hot Pump
Pressure
Hot Punp
Vacuum
U)
o
M
01
Heated Line
Flowmeters (6)
| Manifold |
Vacuum
Pump
Hot
Sample Dry Sample Lines
Line (Typical Set-up Six Lines):
Filters (6)
(7 microns)
k-4
Sample
Pumps
(3)
Condenser
16
* Hot/Cold
^ Switch
Refrigeration Condenser
ample Pressure
^ zerol
Zero | pspan |fi]Span
°2
CO
co2
Figure 3-2. Flue gas sampling and analyzing system.
-------
Three-pen recorders provide a continuous permanent record of the data taken.
The sample gas is delivered to the analyzers at the proper condition and flow
rate through the sampling and conditioning system described in the previous
section. The sections below describe the analytical instrumentation.
3.2.1 Nitric Oxide (NO) and Total Nitrogen Oxides (NO )
~ " " " ~ - - - _--. - .T5.J--T.--._»- _m . _j XL"
Both the total nitrogen oxides (NOx) and nitric oxide (NO) concentra-
tions are measured from a sample gas obtained using a heated sample line at
394 K (250 °F). In addition, the nitric oxide concentrations are measured
sequentially from samples obtained using the unheated sample line that is
connected to the same analyzer in the laboratory trailer. In the latter case,
water is first removed from the sample gas by a refrigeration unit. The
analytical instrument that is used for all of these measurements is the
Thermo Electron Model IDA chemiluminescent gas analyzer.
For NO analyses, the sample gas is passed directly into the reaction
chamber where a surplus of ozone is maintained. The reaction between the NO
and the ozone produces light energy proportional to the NO concentration
which is detected with a photomultiplier and converted to an electrical signal.
Air for the ozonator is drawn from ambient through an air dryer and a 10-
micron filter element. Flow control for the instrument is accomplished by
means of a small metal bellows pump mounted in a heated box.
The chemiluminescent reaction with ozone is specific for NO. To
detect NO , a thermal converter has been designed to dissociate the NO_ to NO
A
by the bi-molecular 'reaction: 2 NO -> 2 NO + O . A model 700 thermal con-
verter is used in conjunction with the chemiluminescent gas analyzer as shown
in Figure 3-3. The converter is a coil of resistance-heated stainless steel
tubing whose purpose is to drive the NO /NO ratio to its chemical equilibrium
value at the converter temperature and pressure. The unit is designed to
operate at a temperature of 923 K (1200 °F) and pressure of 1.3 kPa (10 torr).
For these conditions and typical stack gas O concentrations, the equilibrium
NO concentration is 0.2% of the total NO concentration. Therefore, when a
£ X.
gas sample containing any NO is passed through the converter, essentially
all the NO - would be converted to NO. The resulting total NO is then measured
using the cherailuminescent analyzer and the difference between the actual NO
and the "total NO" would be the sample NO? concentration. The "total NO" is
interpreted as NO .
x 3-5 KVB 6015-798
-------
|
e __ __ _
_ ;| _ __ -ne
il^oA i
Heated
Sample
Line
J Model 700 !
Figure 3-3. Schematic of NO /NO chemiluminescent analysis system.
3-6
KVB 6015-798
-------
ISO may react upon contact with HO (liquid phase) to form HNO
(nitric acid). Under field test conditions, the exhaust gas may contain
significant HO (depending upon the process and the ambient meteorological
conditions), and it is necessary to convert the NO to NO before the HO
is allowed to condense in the sampling system. By using the heated sample
line and the Thermo Electron Model 700 heated NO module, NO concentra-
X A
tions will effectively be measured. In reference to Figure 3-3, the sample
is maintained above the HO dew point up to and through the 127 ym (0.005
in.) capillary in the heated module. Downstream of this capillary, the
flow network is maintained at 1.3 kPa (10 torr), where the partial pressure
of the H_O in the sample is sufficiently low to prevent any condensation at
ambient temperature.
When using the heated system, NO, NO , and NO are measured on a
wet basis. When not using the heated system, a condenser is placed up-
stream of the analyzer and NO is measured on a dry basis.
Specifications
Accuracy: 1% of full scale
Span stability: +_ 1% of full scale in 24 hours
Zero stability: +_ 1 ppm in 24 hours
Power requirements: 115 +_ 10V, 60 Hz, 1000 watts
Response:--96%-of full scale in 1 sec (NO mode);
0.7 sec (NO mode) X
Output: 4-20 ma
Sensitivity: 0.5 ppm
Linearity: +_ 1% of full scale
Vacuum detector operation
Range: 2.5, 10, 25, 100, 250, 1000, 2500, 10,000 ppm
full scale
3.2.2 Carbon Monoxide and Carbon Dioxide (CO and CO?)
Carbon monoxide and carbon dioxide concentrations are measured using
Beckman Model 864 and 865 short-path-length nondispersive infrared analyzers
(see Figure 3-4) . These instruments measure the differential in infrared
3-7
KVB 6015-798
-------
MFKAKIDSOUKCI
7*7*4-*JtMpi* "* '*OM iou«c«
Figure 3-4. Schematic of NDIR analyzer.
3-8
KVB 6015-798
-------
energy absorbed from energy beams passed through a reference cell (con-
taining a gas selected to have minimal absorption of infrared energy in
the wavelength absorbed by the gas component of interest) and a sample
cell through which the sample gas flows continuously. The differential
absorption appears as a reading on a scale of 0% to 100% and is then related
to the concentration of the species of interest by calibration curves supplied
with the instrument. A linearizer is supplied with each analyzer to provide
a linear output over the range of interest. The operating ranges for the
CO analyzer are 0-100 and 0-2000 ppm, while the ranges for the CO analyzer
are 0-5% and 0-20%.
Specifications
Span stability: +_ 1% of full scale in 24 hours
Zero stability: +_ 1 ppm in 24 hours
Ambient temperature range: 273 to 322 K (32 °F to 120 °F)
Line voltage: 115 j^ 15V rms
Response: 90% of full scale in 0.5 or 2.5 sec
Linearity: Linearizer board installed for one range
Precision: +_ 1% of full scale
Output: 4-20 ma
3.2.3 Oxygen (0 )
A Teledyne Model 326A oxygen analyzer is used to automatically and
continuously measure the oxygen content of the flue gas sample. The analyzer
utilizes a micro-fuel cell which is specific for oxygen, has an absolute
zero, and produces a linear output from zero through 25% oxygen. The micro-
fuel cell is a sealed electrochemical transducer with no electrolyte to
change or electrodes to clean. Oxygen in the flue gas diffuses through
a Teflon membrane and is reduced on the surface of the cathode. A corres-
ponding oxidation occurs at the anode internally and an electric current
is produced that is proportional to the concentration of oxygen. This
current is measured and conditioned by the instrument's electronic circuitry
to give an output in percent O by volume for operating ranges of 0% to 5%,
0% to 10%, and 0% to 25%.
3~9 KVB 6015-798
-------
Specifications
Precision: +_ 1% of full scale
Response: 90% in less than 40 sec
Sensitivity: 1% of low range
Linearity: +_ 1% of full scale
Ambient temperature range: 273 K to 325 K (32 to 125 °F)
Fuel cell life expectancy: 40,000%+-hrs
Power requirement: 115 VAC, 50-60 Hz, 100 watts
Output: 4-20 ma
3.2.4 Total Hydrocarbons (HC)
Hydrocarbon emissions are measured using a Backman Model 402
high-temperature hydrocarbon analyzer. The analyzer utilizes the flame
ionization method of detection which is a proven technique for a wide
range of concentrations (0.1 to 120,000 ppm). A flow schematic of the
analyzer is presented in Figure 3-5. The sensor is a burner where a
regulated flow of sample gas passes through a flame sustained by regulated
flows of air and a premixed hydrogen/nitrogen fuel gas. Within the flame
the hydrocarbon components of the sample stream undergo a complex ionization
that produces electrons and positive ions. Polarized electrodes collect
these ions, causing current to flow through electronic measuring circuitry.
Current flow is proportional to the rate at which carbon atoms enter the
burner.
The analysis occurs in a temperature-controlled oven. The sample
is extracted from the stack with a stainless steel probe which has been
thermally treated and purged to eliminate any hydrocarbons existing in
the probe itself. An insulated heat-traced teflon line is used to
transfer the sample to the analyzer. The entire heated network is main-
tained at a temperature to prevent condensation of heavier hydrocarbons.
The flame ionization detector is calibrated with methane, and the
total hydrocarbon concentration is reported as the methane equivalent.
FID's do not respond equally to all hydrocarbons but generally provide a
measure of the carbon-hydrogen bonds present in the molecule. The FID
does not detect pure carbon or hydrogen.
3_1Q KVB 6015-798
-------
AIR
INLET
FILTER
INLET
VALVE
Figure 3-5. Flow schematic of hydrocarbon analyzer (FID).
3-11
KVB 6015-798
-------
Specifications
Full-scale sensitivity: adjustable from 5 ppm CH to 10% CH
Ranges: Range multiplier switch has 8 positions: XI, X5, XlO,
X50, X100, X500, X1000, and X5000. In addition, span
control provides continuously variable adjustment
within a dynamic range of 10:1
Response time: 90% full scale in 0.5 sec
Precision: +_ 1% of full scale
Electronic stability: +_ 1% of full scale per 24 hours with
ambient temperature change of less than
10 °F
Reproducibility: +_ 1% of full scale for successive identical
samples
Analysis temperature: ambient
Ambient temperature: 273 K to 317 K (32 °F to 110 °F)
Output: 4-20 ma
Air requirements: 250 to 400 cc./min of clean, hydrocarbon-free
air, supplied at 2.07 x 105 to 1.38 x 10
n/irr (30 to 200 psig)
Fuel gas requirements: 75 to 80 cc/min of fuel consisting of
100% hydrogen supplied at 2.07 x 105
to 1.38 x 106 n/nr (30 to 200 psig)
Electric power requirements: 120 V, 60 Hz
Automatic flame indication and fuel shut-off valve
3.2.5 Sulfur Dioxide (SO )
A Dupont Model 400 photometric analyzer is used for measuring SO,,.
This analyzer measures the difference in absorption of two distinct wave-
lengths (ultraviolet) by the sample. The radiation from a selected light
source passes through the sample and then into the photometer unit where
the radiation is split by a semi-transparent mirror into two beams. One
beam is directed to a phototube through a filter which removes all wave-
lengths except the "measuring" wavelength, which is strongly absorbed by
the constituent in the sample. A second beam falls on a reference photo-
tube, after passing through an optical filter which transmits only the
3-12 KVB 6015-798
-------
"reference" wavelength. The latter is absorbed only weakly, or not
at all, by the constituent in the sample cell. The phototubes translate
these intensities to proportional electric currents in the amplifier.
In the amplifier, full correction is made for the logarithmic relation-
ships between the ratio of the intensities and concentration or thickness
(in accordance with Beer's Law). The output is therefore linearly pro-
portional, at all times, to the concentration and thickness of the sample.
The instrument has a lower detection limit of 2 ppm and full scale ranges
of 0-200 and 0-2000 ppm.
Specifications
Noise: Less than 1/4%
Drift: Less than 1% full scale in 24 hours
Accuracy: (+ 1% of analyzer reading)+(+_ 1/4% of full scale range)
Sample cell: 304 stainless steel, quartz windows
Flow rate: 6 CFH
Light source: Either mercury vapor, tungsten, or "Osram"
discharge type lamps
Power rating: 500 watts maximum, 115 V, 60 Hz
Reproducibility: 1/4% of scale
Electronic response: 90% in 1 sec
Sample temperature: 378 K (220 °F)
Output: 4-20 ma d.c.
3.3 SULFUR OXIDES (SO )
X
Goksoyr-Ross MethodWet Chemical Method
The Goksoyr-Ross Controlled Condensate (G/R) method is used for the
wet chemical SO /SO determination. It is a desirable method because of its
simplicity and clean separation of particulate matter, SO and K SO (SO ) .
This procedure is based on the separation of H SO (SO ) from SO by cooling
the gas stream below the dew point of H SO but above the HO dew point.
Figure 3-6 illustrates schematically the G/R test system.
3~13 KVB 6015-798
-------
Adapter for Connecting Hose
Rubber
Vacuum
Hose
Vacuum
Gauge
Asbestos Cloth
Insulation
Glass-Cloth Heating
Mantle >^"~-
Stack
Gas Flow
Recirculator
Thermometer
Styrofoam Ice Chest
3-way
Valve
Drierite
Figure 3-6. Schematic of Goksoyr-Ross Controlled Condensation
System (CCS).
3-14
KVB 6015-798
-------
Participate matter is first removed from exhaust gas stream by means
of a quartz glass filter placed in the heated glass filter holder. Tissue-
quartz filters are recommended because of their proven inertness to I^SO^.
The filter system is heated by a heating tape so that the gas out temperature
of 260 °C (500 °F) is maintained. This temperature is imperative to ensure
that none of the H SO will condense in the filter holder or on the filter.
The condensation coil where the H SO is collected is cooled by water
which is maintained at 60 °C (140 °F) by a heater/recirculator. This tempera-
ture is adequate to reduce the exhaust gas to below the dew point of H SO^.
Three impingers are shown in Figure 3-6. The first impinger is
filled with 3% HO to absorb SO ; the second impinger is to remove carry-
over moisture; and the third contains a thermometer to measure the exhaust
gas temperature to the dry gas meter and pump. The sampling rate is 2.3 1pm
(0.08 CFM).
Analysis Procedure
For both SO and H SO determination, the analytical procedure is
£ £ 4i
identical. The H SO sample is washed from the back part of the filter holder
and the coil using distilled water. The sample from the first impinger which
is assumed to be absorbed and reacted SO in the form of H SO is recovered
with distilled water washing. The amount of H SO. in the condensate from the
coil and from the HO impinger is measured by H+ titration. Bromphenol Blue
is used with NaOH as the titrant.
3.4 PARTICULATE MATTER TOTAL MASS CONCENTRATION
Particulate matter is collected by filtration and wet impingement in
accordance with US-EPA Method No. 5. Nomograph techniques are utilized to
select the proper nozzle size and to set the isokinetic flow rates.
Gas samples for particulate sampling can be taken from the same
sample port as those for gas analysis and passed through the Joy Manufacturing
Company Portable Effluent Sampler. This system, which meets the EPA design
specifications for Test Method 5, Determination of Particulate Emissions from
Stationary Sources (Federal Register, Volume 36, No. 27, page 24888, December 24,
1971, and revisions thereof) is used to perform both the initial velocity
traverse .and the particulate sample collection.
3-15 KVB 6015-798
-------
Dry particulates are collected in the heated case that may contain
a cyclone to separate particles larger than 5 ym and a 125-rnm glass-fiber
filter to retain particles as small as 0.3 ym. Condensible particulates
are collected in four Greenburg-Smith impingers immersed in a chilled
water bath.
The sampling probe is positioned through an exhaust port and
attached to the sampling box. The probe consists of a sampling nozzle,
heated probe, gaseous probe, thermocouple, and pitot tube. The ball
joint from the heated probe connects to the cyclone and glass filter
holder assembly. These assemblies are positioned in the heated sampling
box which is maintained at 433 K (320 °F) above the predicted SO dew point,
in order to eliminate condensation. The sample then passes from the heated
section to four Greenburg-Smith impingers immersed in an ice bath. Only
the second impinger has the original tip, the other three have had the
tip removed to decrease the pressure drop through them. The first and
second impingers are filled with 250 and 150 milliliters of distilled/
deionized water, respectively. The third impinger is left dry. The
fourth impinger is filled with approximately 200 grams.of indicating
silica gel to remove entrained water. The use of silica gel assures that
a dry sample is delivered to the meter box. After sampling, the spent
silica gel is discarded and not used for any further analysis.
An umbilical cord connects the last impinger, the pitdt tube, and
the heating elements to the meter box which is located in a convenient
place within 15 m of the sampling ports. The meter box contains a
vacuum pump, regulating valves, instantaneous and integrating flow meters,
pitot tube manometers, vacuum gauge, and electrical controls.
Particulate matter (solids and condensibles) is collected in three
discrete portions by the sampling train: the probe and glassware upstream
of the filter; the filter; and the wet impingers. The probe and glass-
ware are brushed and rinsed with acetone; the matter is captured for
gravimetric analysis. The probe and glassware are then rinsed with
distilled water and the rinsings transferred to a second container for
3_15 KVB 6015-798
-------
analysis. The filter is desiccated and analyzed gravimetrically. The com-
bined impinger liquid is heated to drive off uncombined water and the residue
retained for analysis. The particulate matter analysis is illustrated
schematically in Figure 3-7.
US EPA Method 5 considers the particulate matter captured' in containers
(1) and (3); the filter, probe brushing, and probe acetone rinse. As EPA
source standard are based on solid particulates only, care is taken to differ-
entiate between solid and the total (including condensible) particulates.
The water wash is performed because KVB's test experience has shown that a
significant amount of water-soluble material may sometimes be captured by the
probe.
The dry sample volume is determined with a dry test meter at a measured
temperature and pressure and then converted to standard conditions. The volume
of condensed water in the impingers is measured in milliliters and the corre-
sponding volume of water vapor is then computed at standard conditions. The
dry sample volume and water vapor volume are then summed to give the total
sample volume. The dry sample volume is used in the data reduction procedures.
A point of interest is the. method chosen to calculate particulate
emissions in ng/J or lb/10 Btu from the experimental data. The particulate
sampling train, properly operated, yields particulate mass per unit flue gas
volume. Having measured g/m , it is necessary to establish the flue gas
volume per unit heat input if emissions in ng/J are desired. The original
Method 5 involved determining a velocity traverse of the stack, the cross-
sectional area, the flue flow rate, and fuel heating value. A revised and
more accurate method has been promulgated by the Environmental Protection
Agency that utilizes a fuel analysis (carbon content, hydrogen content, high
heating value, etc.) and the measured excess O in the exhaust to calculate
the gas volume generated in liberating 1.055 GJ (a million Btu's). The
velocity traverse approach generally results in a 20 to 30% higher value and
is believed to be less accurate.
3-17 KVB 6015-798
-------
PARTICULATE MATTER MASS DETERMINATION
sampling
train
component
particulate
matter
transfer
procedure
container
processing
analysis
result
Probe Cyclone
I 4.
Brushing
Acetone
Rinse
Distilled
Wa t e r
Rinse
^V j/ '
r
I Bake at 215°F to drive off uncombined H20 and Acetone
I
!
O 0
Y- y'
Gravimetric to 0.1 milligrams
i
nig
I Samples stored for Compositional Analysis
Figure 3-7. Processing and analyzing particulate matter.
3-18
KVB 6015-798
-------
3.4.1 ^articulate Size
J>articulate matter size distribution is determined using a cascade
impactor to collect the sample and a Cahn Model G-2 Electrobalance to weigh
the sample. When light fuels, i.e., No. 2 oil, are used and the flue gas
is relatively clean, a high volume type impactor, the Anderson 2000 Mark III,
is used. When the grain loading of the flue gas is heavy, as when coal is
burned, a low volume impactor, the Brink, as shown in Figure 3-8, is used.
To improve the accuracy of the weighing, lightweight substrate made
of aluminum foil or glass fiber is placed in or on each steel collection
stage. The particles are collected on these lightweight discs, and the
original steel collection stages are used only as a backing for these
substrata.
A common problem with impactors is that the particles do not adhere
to the stage surface, but strike it, rebound, and are re-entrained in the
flow through the slots down to the next stage. Re-entrainment has not proved
to be a problem with the cascade impactor measurements KVB currently is
making. The flue gas flow rate has been reduced from the nominal 46.7x10
m /s (2.8 liters per minute) to 33x10 m /s (2.0 LPM) or less. Visual
examination of the collection stages has found no evidence of scouring or
re-entrainment. One set of stages was further examined under an electron
microscope and there was no sign of a significant number of particulates
that were larger than the aerodynamic diameter cut point (D ) of the
preceding stage. There was, however, a considerable amount of sponge-like
material that appeared to be an agglomeration of small particles.
If rebound proves to be a problem that cannot be solved by reducing
the throughput, the substrate is coated with an adhesive. Workers in the
field currently are using a solution of 5% polyethelene glycol 3000 in
benzene as the substrate coating substance. If a coating is used the
substrates are baked at 473 K (200 °C) for two hours or until the volatiles
have vaporized, and the weight ceases to change. At least one additional
substratum is .processed as a blank.
3-19 KVB 6015-798
-------
DIMENSIONS OF CASCADE IMPACTOR JETS
Dimensions, Cm
Jet No. Jet Diam.
1
2
3
4
5
*From
0.249
0.1775
0.1396
0.0946
0.0731
collection c.up
Spacing of
Jet Opening*
0.747
0.533
0.419
0.282
0.220
surface .
COLLECTION
CUP
SPRING
JET SPINDLE
GASKET
-3 SLOTS
The in-line inspector has fire Jfoges. Particles in the range of :0.3 to 3:0
-microns are collected by successive impingement
Collection cups are "positioned so that
:H» .distance from the jel .decreases
«$ the jet diameter become* ;smtjll*r.
^Annular -ilols .around cup -minimize
Jurbulence
Figure 3-8. Design of a single stage from a Brink type cascade impactor.
3-20
KVB 6015-798
-------
Back-up filters are used on all impactors to collect the
material that passes the last impaction stage. Binderless, glass-
fiber filter material, such as high-purity Gelman Type A Glass Fiber-
Filter Webb, is employed for this purpose. For the Brink brand of impactor,
25-mm-diameter circular filters are placed under the last spring in the outlet
stage of the impactor. The filter is protected by a Teflon O-ring, and a
second filter disc is placed behind the actual filter, acting as a support.
The Andersen brand impactor uses 625-mm-diameter filter discs placed above
the final "F" stage.
For accurate weighing of collected material, a Cahn G-2
Electro-balance with a sensitivity of 0.05 micrograms is used. This
sensitivity is needed for the lower stages of the high loading impactors
where collection of 0.3 mg or less is not uncommon. KVB currently is
using this balance in the field and has found it to be insensitive to
vibration.
The flow through the impactor is measured to determine the
cut points of the individual stages. The flow through the impactor
assembly is monitored by the pressure gauges on the EPA train control
box. The pump on the control box is used to maintain the flow. This
technique is being used successfully in the field by KVB, Inc. at
present.
To ensure proper measurement by the dry gas meter and to pro-
tect the vacuum pumps from damage by water condensing from the flue
gas, the sample stream is chilled and the water dropped out by a
commercially available condenser of the type available for use with
the Western Precipitation, Inc. EPA Train.
If the stack pressure is less -than the ambient pressure it is
possible for backflow to occur through the impactor when the pump is
turned off. This can cause the collected material to be blown off the
collection substrates and onto the underside of the jet plate above.
KVB avoids this problem by ensuring that no gas flow through the impactor
takes place, except when sampling, by using a check valve to close off
the impactor from the pump while removing the impactor from the duct.
3-21 KVB 6015-798
-------
The impactor is carefully loaded with the stage cups and the pre-
weighed stage substrates. The Andersen type impactor requires that extra
attention be paid to the substrate stage and stage-to-stage alignments to
ensure that the jets are not blocked by the substrate and that the jets of
one stage are above the collection surface of the next stage. After all
stages are loaded and the cap and nozzle are added, the assembled Brink
is tightened with wrenches to make certain the high temperature No. 116
asbestos gaskets are seated. Hand tightening suffices for the Andersen
impactor.
KVB has found that supplemental heating of the impactor is not
necessary to prevent the condensation of flue gas water inside the case.
If it is found with industrial combustion equipment that heating is necessary
to prevent water vapor from condensing in the impactor, heating tape and the
necessary insulation are employed. A thermocouple mounted in the sample gas
flow immediately downstream of the impactor outlet is used to monitor and
control the impactor temperature. This measurement also yields the tempera-
ture needed for calculating impactor cut points.
The impactor is preheated for at least 30 minutes before sampling.
The inlet nozzle is pointed downstream of the flow field during this heating
phase to prevent the premature accumulation of particulates in the impactor.
A predetermined flow rate is established immediately and maintained
constant throughout the test. Attempts to modulate flow to compensate for
changes in the duct flow rate and to maintain isokinetic sampling would
destroy the utility of the data by changing the cut points of the individual
stages. Establishing the correct flow rate quickly is especially important
for the short sampling times typical of coal fuels. If a non-standard flow
is necessary, the true cut points will be calculated for the actual flue
gas temperature and impactor pressure drop.
3-22 KVB 6015-798
-------
3CVB has found that the post-test procedure is very important in
obtaining accurate measurements. The crucial part is to make sure the
collected material stays where it originally impacted. After the test, the
impactor is carefully removed from the duct without jarring, unscrewed from
the probe, and allowed to cool. Proper disassembly is critical as discussed
below.
1. Brink Impactor: Careful disassembly of a Brink is a necessity
for obtaining good stage weights. If a precollector cyclone has been used,
all material from the nozzle to the outlet of the cyclone is included with
the cyclone catch. All of this material is brushed onto a small 3 cm x 3 cm
aluminum foil square and saved for weighing. Cleaning the nozzle is also
important, especially if it is a small bore nozzle.
All material between the cyclone outlet and the second stage nozzle
is included with material collected on the first collection substrate. All
adjacent walls are brushed off, as well as around the underside of the nozzle
where a halo frequently occurs on the upper Brink stages. All material
between the second stage nozzle and third stage nozzle is included with that
on the second collection substrate. This process is continued down to the
last collection substrate. Finally, care is exercised in taking out the
filter.
2. Andersen Impactor: The foil to hold the stage 1 substrate is
laid out. Next the nozzle and entrance cone are brushed out and onto the
foil. Then the material on stage 0 is brushed off. Next, any material on
the top 0-ring and bottom of stage 0 is brushed onto the foil. The stage 1
filter substrate material is then placed on the foil and, finally, the top
of the stage 1 plate 0-ring and cross piece are brushed off. Depending on
how tightly the impactor was assembled, some filter material may stick to
the 0-ring edge contacting the substrate. This is carefully brushed onto
the appropriate foil. This process is continued through the lower stages
and the filter.
3-23 KVB 6015-798
-------
All substrates, the backup filter, and the control blanks are cooled
to room temperature in a desiccator and weighed to +_ 0.01 mg. The weighing
chamber of the balance also will be desiccated. Samples and blanks are
returned to the desiccator overnight and reweighed until constant weight is
established. The substrates are weighed soon after the end of the test so that
the data will be available for setting up the following test.
Upon their arrival, the field test crew undertakes the combustion
modification testing, including total particulate measurement. While this
initial testing is being done an estimate of the grain loading and particle
size is made. The data used to select an isokinetic nozzle for the EPA train
are also used to select a nozzle for the impactor. In no case is an impactor
nozzle of less than two millimeters diameter to be used.
Measurements are made at a sufficient number of points across the
flue or smoke stack, as specified by EPA Method 5, to make certain that a
representative sample of particulates is obtained. Whenever possible, the
impactor is oriented vertically so that the flow through it is directed down-
ward. This minimizes the tendency of the particulates to fall off the stages.
When horizontal orientation is unavoidable, extra care is taken to prevent
the impactor from being jarred during removal from the flue.
When coal fuel is fired and sampling is done upstream of the dust
collector, the percentage (by weight) of material with sizes larger than ten
micrometers is appreciable. In such cases a precutter cyclone, such as that
shown in Figure 3-9 and currently used with the Brink impactor, is used to
prevent the upper impactor stages from overloading. A precutter cyclone is
used during the preliminary orientation run, and if the weight of material
obtained by the precutter is comparable to that 'on the first stage, the
precollector is used on subsequent runs.
The required sampling time is dictated by grain loading and the
particulate size distribution. An estimate can be made from the following
typical data gathered during previous KVB test programs.
3-24 KVB 6015-798
-------
.350-
.50 R
#4-40 Tap
our Places
On 1-1/4 Circle
Inlet
Nozzle
2.695
#7 Drill
.201 Tangent
To Bore
PKECUTTER CYCLONE
«£- Complete
Stage
3 Slots
Single
Collection^
Cup
STAGE
Figure 3-9.
Detail of one stage and of precutter cyclone for cascade
impactor.
3-25
KVB 6015-798
-------
Sampling Duration
Fuel and/or Burner (min.)
No. 6 oil 120-240
Spreader stoker 59
No. 2 oil 300
The flow rate and nozzle size are closely coupled, and requirements
for isokinetic or near-isokinetic nozzle flow sometimes force a compromise
on nozzle selection. The general order of priorities used by investigators
to determine nozzle size in the field is (1) nozzle diameter (minimum only),
(2) last stage jet velocity, (3) isokinetic flow rate required, and (4)
nozzle diameter if greater than 2.0 mm.
It is preferable to use as large a nozzle diameter as possible to
minimize sampling errors resulting from nozzle inlet geometry. Investiga-
tors have reported that when very small nozzles have been used with the
Brink impactor, there have been some cases in which large amounts of material
were retained in the nozzle or the nozzle has been completely blocked. The
smallest diameter nozzle KVB uses is 2.0 mm. In some instances, a
90-degree elbow may be necessary due to port location and gas flow direction,
but these situations will be avoided when possible. Problems in cleaning
elbows may occur as well as difficulties in determining the size interval (s)
from which the deposited material originated. When these problems
cannot be avoided, nozzle bends are made as smooth as possible
and of sufficiently large radius to minimize the disturbance of
the flow.
For light oil fuel, a duration of 300 minutes was required to
collect a measurable sample. On the other hand, with coal, only 59
minutes was required. The long test time for No. 2 oil was necessary
because a low-flow-rate Brink brand impactor was used. To avoid long
test time KVB used a high-flow-rate impactor when the flue gas grain
loading was low. However, in no case will the test duration be less
than 60 min. in order to allow for short-term variations in the
operation of the combustion device.
3-26 KVB 6015-798
-------
3.5 SMOKE SPOT
On combustion, equipment where smoke numbers normally are taken,
such as oil-fired boilers, KVB, Inc. determines the smoke number using
test procedures according to ASTM Designation: D 2156-65. The smoke
number is determined at each combustion modification setting of the
unit. Examples are baseline, minimum excess air, low load, etc., and
whenever a particulate concentration is measured.
Smoke spots are obtained by pulling a fixed volume of flue gas
through a fixed area of a standard filter paper. The color (or shade) of
the spots that are produced is visually matched with a standard scale.
The result is a "Smoke Number" which is used to characterize the density
of smoke in the flue gas.
The sampling device is a hand pump similar to the one shown
in Figure 3-10. It is a commercially available item that can pass 36,900
+_ 1650 cu cm of gas at 16°C and 1 atmosphere pressure through an enclosed
filter paper for each 6.5 sq cm effective surface area of the filter
paper.
Sampling Tube
A
Filter Paper
Plunger
Handle'
Figure 3-10. Field-service-type smoke tester.
The smoke spot sampler is provided with a motor-driven
actuator to ensure a constant sampling rate independent of variations
in stroke rate that can occur when the sampler is operated manually.
3~27 KVB 6015-798
-------
The smoke scale required consists of a series of ten spots numbered
consecutively from 0 to 9, and ranging in equal photometric steps from white
through neutral shades of gray to black. The spots are imprinted or other-
wise processed on white paper or plastic stock having an absolute surface
reflectance of between 82.5 and 87.5%, determined photometrically. The smoke
scale spot number is defined as the reduction (due to smoke) in the amount of
light reflected by a soiled spot on the filter divided by 10.
Thus the first spot, which is the color of the unimprinted scale, is
No. 0. In this case there is no reduction in reflected incident light directed
on the spot. The last spot, however, is very dark, reflecting only 10% of the
incident light directed thereon. The reduction in reflected incident light
is 90%, and this spot is identified as No. 9. Intermediate spot numbers are
similarly established. Limits of permissible reflectance variation of any
smoke scale spot will not exceed +_ 3% relative reflectance.
The test filter paper is made from white filter paper stock having
absolute surface reflectance of 82.5 to 87.5%, as determined by photometric
measurement. When making this reflectance measurement, the filter paper is
backed by a white surface having absolute surface reflectance of not less
than 75%.
When clean air at standard conditions is drawn through clean filter
paper at a flow rate of 47.6 cu cm per sec per sq cm effective surface area
of the filter paper, the pressure drop across the filter paper falls between
the limits of 1.7 and 8.5 kPa (1.3 and 6.4 cm of mercury).
The sampling procedure is exactly that specified in D 2156. A clean,
dry, sampling pump is used. It is warmed to room temperature to prevent
condensation on the filter paper. When taking smoke measurements in the
flue pipe, the intake end of the sampling probe is placed at the center line
of the flue. When drawing the sample, the pressure in the flue gas stream
and the sampler is allowed to equalize after each stroke.
3-28 KVB 6015-798
-------
"She smoke density is reported on the Mobile Lab Data Sheet as the Smoke
Spot Number on the standard scale most closely corresponding to test spot.
Differences between two standard Smoke Spot Numbers are interpolated to
the nearest half number. Smoke Spot Numbers higher than 9 are reported
as "Greater than No. 9."
This procedure is deemed to be reproducible to within +_ 1/2 of a
Smoke Spot Number under normal conditions where no oily stain is deposited
on the disk.
KVB's field experience with industrial boilers has been that the
human factor involved in the interpretation of the smoke spot by an experi-
enced observer does not cause a significant lack of precision.
3/6 OPACITY
Opacity readings are taken by a field crew member who is a certificated
graduate of a U.S. Environmental Protection Agency approved "Smoke School."
Observations are made at the same time that particulate measurements are
made and as often in addition as deemed necessary to gather the maximum
amount of information. The procedures set forth in EPA Method 9, "Visual
Determinations of the Opacity of Emissions for Stationary Sources," are
followed.
Observations are made and recorded at 15-second intervals while
particulate concentration is being measured and after the unit has stabilized
at other times. Before beginning observations, the observer determines that
the feedstock or fuel is the same as that from which the sample was taken
for the fuel analysis.
Before beginning opacity observations, the observer makes arrangements
with the combustion unit operator to obtain the necessary process data for the
standard KVB Control Room Data Sheet. The control room data are recorded for
3-29 KVB 6015-798
-------
the entire period of observations, as is customarily done by KVB during an
emissions test. The process unit data that are obtained include:
a. Production rates
1. maximum rated capacity
2. actual operating rate during test
b. Control device data
1. recent maintenance history
2. cleaning mechanism and cycle information
The observer requests the appropriate plant personnel to
briefly review and comment on the opacity measurements and process
data, and the observer comments on:
a. the basis for choosing the observation periods used.
b. why it is believed the periods chosen constitute periods
of greatest opacity.
c. why the observations span a time period sufficient to
characterize the opacity.
Consideration is given to postponing the EPA Method 5 particulate
tests during periods of cloudy or rainy weather because of the inability
of the observer to monitor the smoke.
3-30 KVB 6015-798
-------
SECTION 4.0
PETROLEUM PROCESS HEATER SUBSCALE BURNER RESEARCH
This section summarizes the emission and efficiency data collected
for the subscale heater tests. Also discussed is the cost effectiveness study
conducted for the natural draft heaters. The results presented herein
summarize the equipment characteristics, special instrumentation requirements,
gaseous emissions data, combustion modifications, efficiency, cost effective-
ness and conclusions and recommendations.
4.1 SUBSCALE TEST - PETROLEUM PROCESS HEATER
The testing covered in this section was conducted either in a research
furnace of a major manufacturer of natural draft burners or in a cold flow
model in the KVB laboratory.
4.1.1 Process Heater (Location 1) - Equipment Characteristics
The process heater subscale testing was conducted in the research
furnace of a major manufacturer of natural draft burners. The furnace was
a refractory lined, uncooled rectangular box type furnace 2.4 m (81) wide
by 1.8 m (61) deep by 9.8 m (32') high.
The natural draft burner was installed in the furnace floor firing
vertically upward. Furnace draft was controlled manually with a damper in
the stack. View ports for observing flame shape were provided.
The furnace had the capability of firing either oil or natural gas,
and both flows were measured with flow meters. Thermocouples were installed
in the side of the furnace to indicate the vertical thermal gradient and to
show when the furnace was up to operating temperature. Figure 4-1 shows the
placement of thermocouples in the furnace.
4-1 KVB 6015-798
-------
12" *1 Stack "T"
1
i
/~\ /""\
/r\ xr\ /*^\
. . .
V
©0©
Burner
Opening
i ' ,i ..,^_
o. /yir.
(2
^
0.£
I")
r
.
4m
(33"
4
0.76m
(30")
0.76m
(30")
_f
0.69m
(27")
O.tf4m
(33")
f
0.76m
(3?"}9.8 »
-t (32'0")
1.4m
(5-i
tmt
3.
(11
'
" )
Cm
7")
r i
t 1.0m t
r~(4o"rl
1 2.4m
1 (96")
Figure 4-1. Research furnace thermocouple location.
4-2 KVB 6015-798
-------
4.1.2 Burner Cold Flow Tests
A cold flow burner model simulating the natural draft burner was
fabricated at the KVB laboratory. The cold flow model was built to the same
dimensions as the burner tested in the research furnace. Three sets of gas
tips were supplied by the burner manufacturer for evaluation.
The purpose of the cold flow tests was to develop an analytical mixing
model to provide insight into fuel .injection modifications which could lead to
lowered NO emissions.
The cold flow model was a natural draft model with the same air flow
and velocity as the actual burner. The natural gas was simulated by gaseous
CO . Fuel injection momentum was the same as in the actual burner. Measure-
ments of fuel concentration were made in two axes across the firebox at three
axial planes. Contour maps of constant concentration were prepared to compare
the mixing characteristics of the different gun tips. After the mixing model
was prepared, modifications to the gas tips were made and tested.
Tests were conducted with three standard gas tips supplied by the
burner manufacturer. The gas fuel was simulated by CO and the concentration
measured with an NDIR CO analyzer. Actual burner fuel air ratio and the cold
flow simulation are related by the following expression:
(F/A)Burner = CF/?'TSim. air Burner
A schematic of the flow system is shown in Figure 4-2. Sample ports were
located at three axial positions approximately 5, 36, and 66 cm (2, 14, and
26 inches) above the gas injection plane. At each axial position sample ports
were on the burner center line and 5 cm (2 inches) apart to the edge of the
burner, then 10 cm (4 inches) apart from the burner to the walls. Figure 4-3
presents the pattern for the sample ports. Sample ports not being used were
covered to prevent influx of air.
The results of the measurements for nozzle configuration No. 2 are
shown in Figures 4-4 and 4-5.
4~3 KVB 6015-798
-------
Gas Spuds
(Typ.)
1/2" = 13 mm
3/8" = 9.5 mm
o
.1/2" Pipe
3/8" Tubing
\
O
1/2" x 3/8" Tubing
O
.3/8" Tubing
Hand Valve
Flowmeter
Pressure
Hand Valve
co2
Bottle
Figure 4-2. Schematic of cold flow model.
4-4
KVB 6015-798
-------
South
Exhaust Fan
Face
Sample Ports
Symmetrical
About Center'
Line
ioOOO G G O G
q>OOOO O G G O
I
i n
'GOOG O G G G
f
1
23 cm
121? 7 cm
i
j Sleeve
Burner
Elevation (Typ. for
2 sides)
1
30 cm
(12")
30 cm
(12")
28 cm
(11")
10 cm
(4")
Air Flow
Figure 4-3. Schematic of burner cold flow model showing sampling locations.
4-5
KVB 6015-798
-------
1,2
cn
o
M
ui
-j
i£>
oa
Hpzzle #2 South to North
I I I
DISTANCE, cm (inch)
Figure 4-4. CQ2 concentration as a function of radial position for three axial positions.
-------
1.2
en
o
ui
03
Nozzle #2 East to West
i i
DISTANCE, cm (inch)
Figure 4-5. CO2 concentration as a function of radial position for three axial positions.
-------
Figure 4-4 shows the concentration gradient across the simulated
burner for the three axial positions. The lower level, which is two inches
above the gas injection plane, shows a large gradient in concentration. The
high points are adjacent to the injection points. Inside the burner (10-.
20 cm, 4-10 inches) the fuel concentration is low. Examination of the center
and upper level concentration curves shows that the gas is almost completely
mixed within 36 cm (14 inches) of the burner. Figure 4-5 presents similar
measurements made across the burner but at 90 deg. from the previous data.
This figure shows similar trends, indicating that the burner mixing is uniform
circumferentially.
The three nozzles are compared in Figures 4-6 and 4-7. Figure 4-6
shows the concentration gradient at the burner exit plane for all three gas
tips tested. Very little difference is noted .among the three patterns.
Figure 4-7 shows the concentration gradient 36 cm (14 inches) downstream from
the injection plane for all three nozzles with only minor differences among
the gas tip patterns.
Several modifications to the fuel injection geometry were evaluated
with cold flow simulation. Modifications which looked promising from the
cold flow model were then evaluated in the hot firing -tests.
The modifications tested were the following.:
1. Turning the-gas nozzles so that "the center firing port was aimed
radially outward such that the gas stream impinged upon the 41 cm
(16 in.) diameter cylindrical sleeve.
2. Placing a 20 cm (8 in.) diameter 'staging' cylinder whose vertical
centerline coincided with that of the burner into the flow such
that roughly 25%-30% of the 'combustion' air flow was introduced
through the cylinder. Two cylinders of different length were
used in separate tests. In one case, the top of the cylinder was
5.4 cm (2-1/8 in.) above the gas tips. In the second case, the
cylinder top was 30.5 cm (12 in.) above the gas tips.
4-8 KVB 6015-798
-------
VD
8
Measured at lower level
I I
0.2 _
DISTANCE, cm (inch)
Ul
-J
Figure 4-6.
CO concentration as a function of radial position for three gas tip patterns.
00
-------
I
h-1
o
Measured at center level
DISTANCE, cm (inch)
(Tv
O
CO
Figure 4-7. CO0 concentration as a function of radial position for three qas tip patterns,
-------
3. Placing a 7.6 cm (3 in.) wide, 15.2 cm (6 in.) long deflector
upstream of each of the gas nozzles inclined at a 45-degree
angle from vertical and extending from the 'burner' sleeve
to the plane of the gas tip orifices.
All of these modifications were expected to delay mixing of fuel and
air, thereby lengthening the flame in a hot-firing application, lowering peak
temperatures and, thus, lowering NO emissions. The concentration of the
cold flow test gas (CO_) was measured at various positions along the north-
south centerline across the simulated burner. The results for the four
different test cases at each of three heights above the gas nozzles are shown
in Figures 4-8, 4-9, and 4-10. These curves indicate that the concentric
'staging' cylinder and the radially-outward-facing injection orifices produce
a significant delay in the mixing of the test gas and air. The mixing pattern
with the deflectors in place did not vary appreciably from the patterns
obtained for the nozzles without modification.
The mixing patterns of each of the three configurations which appear
promising for hot-firing application are shown in Figures 4-11, 4-12, and
4-13. Based on these results, KVB made similar modifications to the conven-
tional burner at the manufacturer's research facility.
4.1.3 Hot Firing Test Results
Tests were conducted to evaluate the effect of combustion modifica-
tions on emissions from a natural draft process heater. The reduction in
NO emissions and the change in efficiency were evaluated for (1) lowered
X
excess air, (2) staged combustion air, (3) low-NO burners (tertiary air
JC
injection and recirculating tile designs), (4) flue gas recirculatiori, (5)
steam injection and (6) altered fuel injection geometry. The tests were con-
ducted with natural gas and No. 6 oil. Only burner baseline measurements were
made with No. 2 oil. Fuel samples were taken for all tests, and the analyses
are summarized in Table 4-1.
4-11 KVB 6015-798
-------
CTi
O
cn
vo
CD
2.0
1.6
1.2
O
U
.8
i r
Upper Level
T
T
T
Tips Toward Walls
Cylinder 5.4 cm (2-1/8") Above
Tips
ilDeflectors
/Scylinder 30.5 cm (12") Above
Tips
r
10.2
(4)
1
20.3
(8)
Q>^
I
30.5
(12)
^Kf*
I
40.6
(16)
^5T
1 1
50.8 61.0
(20) (24)
u
1
71.1
(28)
L
_l
81.3
(32)
DISTANCE, cm (in.)
- Burner Opening
Figure 4-8.
CO concentration versus centerline distance at one axial position with four
different modifications.
-------
cn
o
oo
2.0
1.6
1.2
o
u
0.8
I
I
T I
Middle Level
Gas tips toward walls
Cylinder 5.4 cm (2-1/8 in.) above tips
Deflectors
Cylinder 30.5 cm (12 in.) above tips
DISTANCE, cm (in.)
Burner Opening
Figure 4-9. CO concentration vs. centerline distance at one axial position with four different
modifications.
-------
2.0
I
M
it*
en
o
1.6
1.2
o
u
0.8
0.
I I
toward walls
^Cylinder 5.4 cm (2-1/8 in
above tips
Deflectors
£"\Cylinder 30.5 cm (12 in.)
above tips
-Ij Figure 4-10.
DISTANCE, cm (in.)
Burner Opening
CD
CO concentration vs. centerline distance at one axial position with four different
modifications.
-------
Ln
r
1 1
10.2 20.3
(4) (8)
ff*
1
30.5
(12)
"-^J7l -
<^p^
- ^7- ~
i
40.6
(16)
-8
"\J
1
50.8
(20)
\
Upper Level
1 1
61.0
(24)
71.1 81.3
(28) (32)
en
o
tn
00
DISTANCE, cm (in.)
- Burner Opening
Figure 4-11.
CO concentration versus centerline distance at three axial positions with gas tip orifices
turned toward walls.
-------
en
en
o
ID
co
2.0
1.6
1.2
O
U
0.8
0.4
Lower Level
Middle Level
Upper
Level
I
10.2
(4)
20.3
(8)
30'. 5"
(12)
40.6
(16)
DISTANCE, cm (in.)
- Burner Opening
50.8
(20)
61.0
(24)
71.1
(28)
81,3
(32)
Figure 4-12.
CO concentration vs. centerline distance at three axial positions with cylinder top
located 5.4 cm (2-1/8 in.) above gas tips.
-------
<*>
o
u
I
M
-J
Middle
Level
I
I
10.2
(4)
20.3
(8)
30.5
(12)
40.6 50.8
(16) (20)
DISTANCE, cm (in.)
61.0
(24)
71.1
(28)
81.3
(32)
en
o
Burner Opening
VD
oo
Figure 4-13. CO concentration versus centerline distance at three axial positions with cylinder
top located 30.5 cm (12") above gas tips.
-------
TABLE 4-1. FUEL OIL AND NATURAL GAS ANALYSES
Ultimate Analysis:
Carbon , %
Hydrogen , %
Nitrogen, %
Sulfur, %
Ash, %
Oxygen, % (by difference)
API Gravity at 60 °F
Heat of Combustion:
Gross, kJ/kg (Btu/lb)
Net, kJ/kg (Btu/lb)
Component
Helium
Nitrogen
Carbon Dioxide
Methane
Ethane
Propane
Iso-Butane
N-Butane
Iso-Pentane
N-Pentane
Hexane
Oxygen
Heating Value, dry, Gross, J/m
Fuel Oil
1
No. 6 Oil, 1/18/78
Test 1/1-7
86.29
10.07
0.31
2.14
0.042
1.15
11.4
42 082 (18,090)
39 942 (17,170)
Natural Gas
Mol
0.
2.
0.
92.
3.
0.
0.
0.
0.
0.
0.
0.
2
No. 2 Oil, 1/20/78
Test 1/1-10
86.36
13.48
0.012
0.11
0.001
0.04
38.7
45 688 (19,640)
42 827 (18,410)
%
04
23
53
14
96
60
06
10
03
03
03
25
(Btu/CF) 38.18X106 (1025)
Specific Gravity of Gas (relative to air) 0.
Carbon, %
Hydrogen , %
Nitrogen, %
Sulfur, %
Ash, %
Oxygen, % (by difference)
API Gravity at 60 °F
Heat of Combustion:
Gross, kJ/kg (Btu/lb)
Net, kJ/kg (Btu/lb)
Fuel Oil
No. 6 Oil, 2/23/78
85.98
10.33
0.31
2.18
0.042
1.16
12.4
40 763 (18,190)
38 571 (17,250)
4-18
5987
Shale Oil, 2/22/78
83.96
11.17
2.10
0.52
0.051
2.20
20.3
40 851 (18,270)
38 571 (17,250)
(continued)
KVB 6015-798
-------
TABLE 4-1 (Continued).
Fuel Oil
No. 6 Oil, 3/23/78
Carbon, % 85.37
Hydrogen, % 10.54
Nitrogen, % 0.28
Sulfur, % 1.87
Ash, % 0.037
Oxygen, % (by difference) 1.90
API Gravity at 60 °F 13.7
Heat of Combustion:
Gross, kJ/kg (Btu/lb) 40 606 (18,120)
Net, kJ/kg (Btu/lb) 38 455 (17,160)
Fuel Oil
No. 6 Oil, 5/4/78
Carbon, % 86.57
Hydrogen, % 10.62
Nitrogen, % 0.29
Sulfur, % 1.92
Ash, % 0.037
Oxygen, % (by difference) 0.56
API Gravity at 60 °F.... 13.5
Heat of Combustion:
Gross, kJ/kg (Btu/lb) 42 594 (18,310)
Net, kJ/kg (Btu/lb) 40 338 (17,340)
4-19 KVB 6015-798
-------
A. Baseline Tests
Tests were conducted with each burner prior to implementing any combus-
tion modification., These baseline measurements were made with the burner firing
natural gas, No. 6 oil and No. 2 oil. A summary of baseline gaseous emissions
data is presented in Table 4-2. A complete tabulation of all of the process
heater test data is given in Appendix A.
Initial tests were conducted with a standard MA-16 natural draft
burner. A schematic (Fig. 4-14) of a section through the burner at the injec-
tion plane shows the relative position of the four natural gas tips. An oil
gun was installed at the center of the burner. Three configurations of gas
tips were evaluated with this burner. Details of the gas tip hole drill pat-
terns are shown in Figure 4-15 for the. three configurations tested. The
modified configuration II tip and the configuration IV tip result in a more
tangential injection of the gas jet.
B. Lowered Excess Air
The effect of burner operation at lowered excess.air was evaluated
for three unmodified burnerstwo standard natural draft burners and a low
NO burner design which incorporates tertiary air injection. The effect of
X
excess oxygen on NO emissions is shown in Figure 4-16 for the three burners
tested. The low-NO burner (tertiary air injection) exhibited the lowest
level of NO at the nominal 3% 0 condition and showed the most dependence on
O level. NO dropped sharply as excess O decreased. The NO emissions at
£ X £ X
2.7% O were 100 ppm, which dropped off to 76 ppm at 2.1% O .
The effect of excess O on NO firing No. 6 oil with three different
spray patterns is shown in Figure 4-17 for the MA-16 natural draft burner.
The spray angle is defined as the total included angle of the conical jet
produced by the oil gun. The effect of 0_ on NO emissions is not very
pronounced over the range of O tested. For the 30-deg. spray angle, the
maximum NO reduction was 21% from the baseline condition. The minimum NO
level was 222 ppm (at 3% 0 , dry) with the burner operating at 0.85% 0 , the
CO limit. With the 40-deg. spray angle nozzle, the maximum reduction was 9%
from the baseline condition. The minimum value for NO for this nozzle was
4-20 KVB 6015-798
-------
TABLE 4-2. SUMMARY OF AVERAGE BASELINE GASEOUS EMISSIONS FOR UNMODIFIED BURNERS
*.
Natural Gas
MA-16
DBA- 16
Low-NOx Burner
(Tertiary Air
Injection)
Low-NOx Burner
(Recirculating
Tile)
No. 6 Oil
MA-16
Low-NOx Burner
(Tertiary Air
Injection)
No. 2 Oil
MA-16
Low-NOx Burner
(Recirculating
Tile)
Heat
MW
1.53
1.52
1.49
1.47
1.47
1.43
1.41
1.49
Input Rate
(106 Btu/h)
(5.2)
(5.2)
(5.1)
(5.0)
(5.0)
(4.9)
(4.8)
(5.1)
O2 CO2 NO
% % ppm* ng/J
3.0 10.7 107 54.6
3.0 10.3 131 67.0
3.2 10.4 92 47.1
3.1 9.9 104 53.0
3.0 13.3 285 159
3.1 13.7 265 149
3.1 12.7 112 63
3.9 12.6 110 61.7
NO
ppm* ng/J
103 53.8
127 64.9
87 44.4
104 53.0
278 156
261 147
108 61
105 58.9
CO
ppm*
0
0
0
0
0
0
0
0
S02
ppm*
0
0
0
0
1015
1334
46
38
*Corrected to 3% O , dry
o
M
in
oo
-------
Tile in 18 Sections
Gas Tips
Pilot
Oil Tip
Figure 4-14. Plan view of the MA-16 burner for natural draft process heater.
4-22
KVB 6015-798
-------
60° Off Vert CL 45° Off Vert
15° 15°
i/.
3d*
60° Off
Vert^ 10
60° Off Vert
30° Off
Vert
1-0.16 cm
(1/16")diam Ign
Port @ 45° Off
Vert (
45° Off Vert
1-0.16 cm
(1/16") diam Ign
Port @ 45° Off
Vert C
30° Off Vert
CONFIGURATION II
.CONFIGURATION IV
1-0.16 cm (1/16")
diam Ign Port @
45° Off Vert C
CONFIGURATION II (MODIFIED)
Ul
I
Figure 4-15. Gas tip hole drilling patterns. Note that vertical centerline is perpendicular
to the plane of the page.
-------
a
6
cu
a
180
160
140
120
100
80
60
40
20
I I I
Fuel: Natural Gas
Firing Rate: 1.52 MW (5.2x10
(CO concentration)
- denotes CO limit
Btu/h) nom.
TL39
I
'O MA-16 unmodified
1 & DBA-16 unmodified
i ^ Tertiary Air
Burner, all reg.
100% open
I 1
012 345
STACK EXCESS OXYGEN, %, dry
Figure 4-16- Summary of NO emissions as a function of excess oxygen for
subscale natural draft furnace firing natural gas.
4-24
KVB 6015-798
-------
300
13
200
4J
(0
C.
O
Z
100
40 deg Spray Angle
(1/1-15)
O
.... ... (1/1-11)
(1/1-12) (1/1-16)
(1/1-17)
(1/1-24)
(1/1- d/1-21)
CO Limit ^ 3Q deg Spray Angle
(1/1-22)
CO Limit
MA-16 Burner
Fuel: No. 6 Oil (Open Symbols)
No. 2 Oil (Shaded Symbol)
Input: ^ 1.41 MW Cv 4.8xl06 Btu/h)
-~ 40-deg Spray Angle Burner Tip
Q 30-deg Spray Angle Burner Tip
Q 50-deg Spray Angle Burner Tip
( ) Test Number
- (?/;-
20)
50 deg Spray Angle
(1/1-25)
EXCESS OXYGEN, %, dry
Figure 4-17. The effect of excess oxygen on NO emissions for a natural draft
burner firing oil.
4-25
KVB 6015-798
-------
282 ppm {3% O_, dry) at 0.75% O . The 50-deg. nozzle produced an NO measure-
ment comparable to the 30-deg. nozzle but the flame appearance was very bad;
it was impinging on refractory tiles, producing sparks, and smoking, and was
very non-uniform. No further testing was conducted with the 50-deg. nozzle.
The baseline NO emission for the MA-16 burner firing No. 2 oil using a
40-deg. nozzle is also shown in Figure 4-17 foir the sake of comparison.
C. Staged Combustion'
Staged combustion is a technique for emissions reduction wherein a
portion of the flame zone is operated fuel-rich, and secondary air is injected
subsequently to bring the overall air-fuel ratio to the desired level to
assure complete combustion. Staged combustion hag been shown to be an
effective method of NO_ reduction in other applications. In order to develop
X
staged combustion in a natural draft heater, two techniques were evaluated,.
In the first method, four staged air lances were inserted through the furnace
floor positioned 90 degrees apart outside the burner tile on a diametet of
61 cm (24 inches) -. This modification is shown schematically in figure 4-18.
The staged air lances were fabricated from 3.2 cm (1-1/4 in.) diameter
stainless steel pipe with an orifice plate of 3.0 cm (1-3/16 in.) diameter on
the end. The end of the lance was angled 45 degrees inward to provide better
penetration of the flame by the secondary air. Adjustment of the insertion
depth was provided by a locking collar outside the furnace floor. Adjustment
in depth up to 1.52 m (5 ft) was possible. An MA-16 burner was used for
these tests.
Nineteen tests were conducted with the staged air configuration to
evaluate the effect on NO emissions and burner performance. Nine tests were
with natural gas, and the remaining 10 were with No. 6 fuel oil* The first
tests with natural gas consisted of varying the injection depth for the staged
air. The effect of staged air injection depth on NO emissions is shown in
Figure 4-19 for natural gas firing. Combustion air was held constant and the
injection depth varied for tests 1/6-1 through 1/6-5. These data show that
no significant reduction in NO is experienced with injection beyond 1.22 m
4-26 KVB 6015-798
-------
I
KJ
Air
Supply
Manifold
Pilot
Air Supply
Tube
Tile in 18
Sections
as Tips
7.6 cm (3")
diam. Air
Manifold
Pilot Gas Conn.
Air Supply Tube
'Length adjustable
'0.3-1.5 m (1-51)
1.3 cm (1/2") Gas Conn.
1.3 cm (1/2") Steam Conn.
o
k>
U1
Figure 4-18. Schematic of staged combustion burner.
03
-------
K)
00
o
M
Ln
00
100 I 1 d/6-1) 4>B = 1-17
.3.2% -
a
n
4J
rt
B
a,
a,
50
(1/6-8) 4> = 0.87
2.7% O
6
(1/6-2) =
3.0% 0
Normal
O
O
(1/6-4) A =
2.6% O_
B
(1/6-5 )<|> =0.84
0.8 3.0% 0
Fuel: Natural Gas
Firing Rate: 1.5 MW
(S.lxlO6 Btu/hr)
Gas Tip: Pattern II
(Test No.)
*B- (A/F)actual/(A/F)stoich.
(1/6-7) <)> = 0.78
1.1% 0,
B
(1/6-6) 4> = 0.75
0.9% O
Low O
Tests
I
I
I
J_
I
0.3(1)
0.6(2) 0.9(3) 1.2(4)
SECONDARY AIR TUBE INSERTION DEPTH, m (ft)
1.5(5)
1.8(6)
Figure 4-19. NO emissions as a function of staged air injection depth and burner air for natural gas firing.
-------
(4 ft) approximately. With the staged air tubes at the maximum depth of
1.52 m £5 ft), the excess oxygen was reduced to 0.9% Q^, which was the CO
limit. The combination of low 0 and staged combustion air gave the maximum
reduction of 67%.
The effect of the burner air-to-fuel ratio is shown in Figure 4-20 for
natural gas firing. The parameter <{> is the ratio of actual air-fuel ratio to
stoichiometric air-fuel ratio. Values of on NO emis-
X
sions is shown in Figure 4-21. .At the normal 0 condition, a reduction of
35% was achieved. With unit operating in the low O mode a reduction of 51%
from the baseline condition was achieved.
An alternative method of producing staged combustion was developed
from the cold flow tests described in an earlier section. This technique
employed a central cylinder which introduced the secondary air into the flame
zone after the primary combustion zone.
Figure 4-22 illustrates the staging cylinder concept based on cold
flow model work. For this modification the orifice plate was removed from
a DBA-16 burner (a conventional burner differing only in tile design from the
MA-16 burner) and a 19.1 cm (7-1/2 in.) I.D., 21.6 cm (8-1/2 in.) O.D. cylinder
placed in the burner such that its longitudinal axis coincided with the verti-
cal centerline of the burner. The bottom of the cylinder rested on the base
of the secondary air section of the burner. Thus, all of the primary air flow
(approximately 1/3 of the total air flow) was routed through the cylinder and
the rest through the secondary air registers.
The staging cylinder concept was tested in the KVB cold flow model
and demonstrated to be effective in delaying mixing of fuel gas and air in the
model and was expected to lower NO emissions under actual operating conditions.
X
The effect of the cylinder is to produce fuel-rich zones in the vicinity of
the gas tips and relatively lean regionsfurther downstream.
4-29 KVB 6015-798
-------
100
50
Fuel: Natural Gas
Firing Rate: 1.5 MVJ (5.1 Btu/hr x 10 )
Gas Tip: Pattern II
Low O,.
I
I
Normal O,
Test No.
(l) 1/6-1, 3.2% O,
i^ ^
[l) 1/6-2, 3.0% E (A/F) ./(A/F) . . . ).
actual stoich
4-3.0
KVB 6015-798
-------
400
U)
0
df>
n
B
a
300
200
100
0.6
Fuel: No. 6 Oil
Firing Rate: 1.51 MW (5.1 Btu/hr x 10 )
Tip: 764
Staged Air Pipes at 1.2 m (4 ft)
0.7
I
I
I
0.8 0.9 1.0 1.1
BURNER EQUIVALENCE RATIO, <}>
1.2
1.3
en
o
en
Figure 4-21. NO emissions as a function of burner <|> for No. 6 oil firing.
CX3
-------
(1/2 in.,) 1.27 cm
'Staging' Cylinders
(Tops of cylinders to be
7.6 cm (3 in.),, 15.2 cm
(6 in.); and 22.& cm
(9 in.) above gas tips)
Primary
Tile
19.1 cm
r7-l/2"TnT5*
Secondary
Air Register
Primary Air Register
Figure 4-22.
Conventional burner with central cylinder modification.
4-32 KVB 6015-798
-------
Several different cylinder lengths were tried in this series of tests
The height of the top of the cylinder above the gas tips was, varied from 7.6 cm
(3 in.) to 109 cm (43 in.). Figure 4-23 summarizes the NO emissions from a
DBA-16 burner equipped with different staging cylinder lengths for approximately
3% O . The graph suggests that an optimum height above the gas tips lies
between 23 cm (9 in.) and 94 cm (37 in.). The lowest NO emissions of the
' X i
cylinder heights tested occurred at 94 cm (37 in.) above the gas tips and was
88 ppm, down 33% from the standard DBA-16 burner average baseline of 131 ppm.
At a cylinder height of 109 cm (43 in.), excess oxygen w&s varied from
4.9% to the CO limit of 0.5% (with a CO concentration of 439 ppm). At 1.2%
t
excess 0 , the NO concentration was 66 ppm, a reduction of 42% from the cor-
<- X "''-.,_, -r «.- '
responding 0 point for the standard DBA-16 burner. At the CO limit, the NO
.; 2 '' -' ' '' X
level dropped to 54 ppm for a reduction of 48% from the CO limit concentration
emitted by the standard burner. Figure 4-24 shows the variation of NO con-
centration with excess 0 for this cylinder height.
D. ; Low-NO Burner (Tertiary Air Injection)
A low-NO burner similar to the conventional MA-16 burner was tested.
x .
This low-NO . design incorporated a tertiary air register above the primary
and secondary air registers. Figure 4-25 is an overall schematic of the
tertiary air burner, and Figure 4-26 is a plan view of the burner.
The tertiary air register allows a certain amount of staging of the
combustion process. Under ordinary operating conditions all registers are
100% open, and 60% of the burner air flow comes through the primary and secondary
air registers while 40% of the air comes .through the tertiary air register.
The:tertiary air' is introduced1 in an annular ring outside the primary and
secondary air flows. Normally, the primary arid secondary air registers
communicate, but for the purposes of these tests, they were banded off so
that each air stream could be independently varied.
Figure 4-27 shows the gas tip hole drilling patterns used in the
tertiary air burner for tests with natural gas. For the first pattern tried
(Configuration IIB), NO levels were high. Baseline emission was 153 ppm
X
(corrected to 3% O , dry).
4-33 KVB 6015-798
-------
150
125
100
o
f\
o
<*>
+J
id
S,
ft
O
2
75
50
25
(1/10-5)
3.1%
(1/10-4)
3,0%
I
I
DBA-16 Burner
Fuel: Natural Gas
Firing Rate: 1.52MW
Gas Tip: Pattern II
(Test No.)
Excess O %
I
(5.18x10
Btu/hr)
25
(10)
51
(20)
76
(30)
102
(40)
127
(50)
STAGING CYLINDER HEIGHT ABOVE GAS TIPS, Cin ,(iu.)
Figure 4-23.
NO emission as a function.of staging cylinder height above
gas tips.
4-34
KVB 6015-798
-------
125
100
s-i
o
+J
(0
ft
ft
75
50
25
(1/10-6)
(1/10-5)
(1/10-9)
/(1/10-8)
Fuel: Natural Gas
Fuel Tips: Pattern II _
Firing Rate: 1.56MW
(5.31x10 Btu/hr)
(Test No.)
Secondary Air Cylinder 109cm
(43") above gas tips
I
I
2 3
EXCESS OXYGEN, %, dry
Figure 4-24.
NO emissions as a function of excess oxygen while firing
natural gas.
4-35
KVB 6015-798
-------
Tertiary Air Register
Furnace
Floor
\
Register Controls
Secondary Air
Register
Primary Air
Register
Not to Scale
Figure 4-25. Schematic of tertiary air burner for natural draft process heater.
4-36
KVB 6015-798
-------
TILE I N 19
SECTIONS
Figure 4-26. Plan view of tertiary air burner.
4-37
KVB 6015-798
-------
i
LO
00
2 - 0.44 cm (11/64 in.)dianj
Firing Ports as shown
45° Off Vert.
30° Off Vert. C
1 - 0.24 cm (3/32 in.)diam
Ign. Port at 45° Off
Vert. Q
- 0.44 cm (11/64 in.) diam
Firing Ports as shown
30° Off Vert. Q
20°
- 0.24 cm (3/32 in.) diam
Ign. Port at 45° Off
Vert. Q
CONFIGURATION IIB
CONFIGURATION I1C
o
M
Ol
<£>
CO
Figure 4-27. Gas tip hole drilling patterns for the tertiary air burner.
-------
Figure 4-28 shows the effect of excess O. changes on NO emissions
2 X
for the tertiary air burner. With the Configuration IIB gas tips, excess 0
was varied from 4.0% down to 0.4%. CO concentration at the low O point was
83 ppm (corrected to 3% O , dry). NO concentration at that point was 122 ppm,
a reduction of 20% from the baseline value, but still higher than the baseline
NO emission from the conventional MA-16 burner.
x
A second gas tip hole pattern (Configuration IIC) was installed in
an attempt to lower NO emission. This configuration had more radially-
oriented injection orifices and produced a much longer, narrower flame.
Baseline NO measurements were about 100 ppm (corrected to 3% O^,
dry). Excess O was varied from 4.1% down to 2.1%. CO concentration at the
minimum 0 was 47 ppm (corrected to 3% 0 , dry). NO at that 0 setting was
£, &, X ^-
76 ppm (corrected to 3% O , dry), a reduction of 24% from the baseline con-
centration. The results of these tests are also shown in Figure 4-28.
Firing rate changes were also made with the Configuration IIB gas
tips. Figure 4-29 shows NO emissions as a function of firing rate for the
x g
tertiary air burner. At 100% of capacity (6.5x10 Btu/hr) NO emission was
X
155 ppm (corrected to 3% O , dry) and dropped to 109 ppm (corrected to 3% 0 ,
£ £
dry) at 37% of capacity.
A series of air register adjustments were made at approximately 3% 0
with the tertiary air burner but produced no appreciable reduction in NO
X
levels. This is illustrated in Figure 4-30. Excess oxygen was varied at
register condition 3. As Figure 4-28 shows, the NO levels were about the same
as those obtained when all registers were 100% open.
The effect of furnace temperature on NO emissions with natural gas
fuel and with the pattern IIC tips is shown in Figure 4-31. The NO level
X
tends to rise until a stack temperature of about 1200 K (1700 °F) is attained.
Since many tests were conducted with stack temperatures less than 1200 K due
to the length of time required for furnace heat-up (about 4 hours) some tempera-
ture-related effects were unavoidable in the data. However, the effects were
fairly small and were also minimized where possible by conducting a related
series of tests (e.g., different excess O points) over the shortest time
possible and making baseline checks periodically during the day.
4-39 KVB 6015-798
-------
Fuel: Natural Gas
Firing Rate: 1.5 MW (5x10 Btu/hr)
PAR = Primary Air Register
SAR = Secondary Air Register
TAR = Tertiary Air Register
180
160
140
120
CM
0 100
o
-------
-M
(0
t
a
180
160
140
120
100
80
60
40
20
(1/3-9)
Baseline
O
(1/3-8)
Fuel: Natural Gas
100% Capacity:
Excess 02: 3%
( ) Test No.
1.9 MW (6.5x10 Btu/hr)
O Tertiary Air Burner, Pattern IIB Tips
1
1
1
20 40 60 80
FIRING RATE , % of Capacity
100
120
Figure 4-29. The effect of firing rate changes on NOX emissions for the low-NO
tertiary air natural draft burner firing natural gas.
4-41
KVB 6015-798
-------
o\°
ro
120
100
80
60
a
o
2
40
20
Fuel: Natural Gas
Firing Rate: 1.47 MW (5x10 Btu/h)
Gas Tips: Pattern IIC
I
I
I
234
STACK EXCESS OXYGEN, %, dry
AIR REGISTER SETTINGS:
@All Registers 100% Open
(l/3-24b)
x-vPAR = Closed (1/3-25)
^SAR=TAR= 100% Open
(|)PAR = 50% Open (1/3-26)
SAR = TAR =100% Open
= Closed (1/3-27)
^PAR=TAR=100% Open
©BAR = 50% Open (1/3-28)
PAR = TAR =100% Open
PAR = Primary Air Register
SAR = Secondary Air Register
TAR = Tertiary Air Register
PAR= SAR= 50% Open
TAR = 100% Open
(1/3-29)
PAR=Closed, SAR=50%
Open, TAR =100% Open
(1/3-30)
PAR=Closed, Tar= 50%
Open, SAR =100% Open
(1/3-31)
PAR = SAR = TAR = 50%
Open (1/3-32)
Figure 4-30. The effects of air register adjustments and stack excess oxygen
on NO emissions for the tertiary air burner firing natural gas.
X
4-42
KVB 6015-798
-------
120,
100
^80
60
O
m
B
§
£ 40
20
0
(1/3-33)
(l/3-24a)
Fuel: Natural Gas
Excess O
2:
3%
Firing Rate: 1.47 MW (5x10
Registers: All 100% Open
Gas Tips: Pattern IIC
( ) Test No.
Btu/hr)
1089
(1300)
1144
(1GOC)
1200
(1700)
1256
(1C 00)
STACK TEMPERATURE, K (°F)
Figure 4-31. The effect of furnace temperature on NO emissions for the
tertiary air burner firing natural gas.
4-43
KVB 6015-798
-------
The data indicate that there was no large temperature effect on
NO emissions with No. 6 oil.
X
Tests on No. 6 oil with the tertiary air burner consisted of excess
O variation at the normal air register settings, air register adjustments,
and excess 0 changes at the register setting which gave the lowest NO
^ X
emissions. An oil tip with a 40-deg. spray angle was used for all of the
tests with No. 6 oil.
The effect of excess 0 on NO emissions for the tertiary air burner
£ X
using No. 6 oil is shown in Figure 4-32. The curve is fairly flat, showing
baseline NO emissions to be 272 ppm (corrected to 3% O_, dry) and dropping to
X ^
235 ppm (corrected to 3% O , dry) at 0.5% O , for a reduction of 14%. The
CO level at 0.5% O was 57 ppm (corrected to 3% O , dry). These baseline
NO values were '^ 15% less than the baseline NO emissions from the MA-16
x x
burner with a 40-deg. spray angle tip.
The effect of air register adjustments is shown in Figure 4-33. The
setting which produced the lowest NO had the primary air register 10% open
X
and the secondary and tertiary air registers 100% open. As with natural gas
fuel, decreasing the primary air seemed to produce the largest effect from
the standpoint of NO emissions. At 2.9% 0.,, NO emission at this register
X ^ X
setting was 200 ppm (corrected to 3% 0 , dry) , a reduction of 26% from
tertiary air burner baseline or 37.5% from the MA-1'6 burner baseline.
The variation of excess O at this register setting could not be
completed because of a severe coking problem which was encountered in the
burner. It was discovered that the oil tip had been placed about 1/4 inch
too low in the burner throat. In addition, because of the small primary air
register opening, less primary air was available for oxidation of the fuel.
These two conditions resulted in the "dropout" of a large amount of unburned
fuel and forced an early stoppage of the tests. Thus, the reduction in
NO emissions achieved with the primary air register 10% open might not be
X
practically attained in actual application.
A few tests on the tertiary air burner were also conducted using a
shale oil of high nitrogen -content (2.1% by weight). Excess 0 changes
coupled with relatively minor register adjustments were made. NO emissions
4-44 KVB 6015-798
-------
350
300
250
T3 200
CN
o
*
115°
a
o
3
100
50
(1/3-3)
Baseline
(1/3-6)
(1/3-5)
(1/3-4)
CO limit
(1/3-45)
(1/3-44)
Fuel: No. 6 Oil
Firing Rate: 1.47 MW (5xl05 Btu/hr)
Tip: 864
( ) Test No.
PAR = Primary Air Register
SAR = Secondary Air Register
TAR = Tertiary Air Register
O All Registers 100% Open
D PAR = 10% Open
SAR = TAR = 100% Open
(Lowest NOX Condition)
J_
234
EXCESS OXYGEN, %, dry
Figure 4-32. The effect of excess oxygen at two register settings on
emissions for the tertiary air burner firing No. 6 oil.
4-45
KVB 6015-798
-------
350
300
250
a
..
t
6
e#>
200
ft 150
ft
g
100
50
(1/3-43) («
(1/3-42)
(1/3-39)
(1/3-41)
'(1/3-40)
Fuel: Ho. 6 Oil
Firing Rate: 1.47 MW (5x10 Btu/hr)
Oil Tip: 864
( ) Test No.
PAR = Primary Air Register
SAR = Secondary Air Register
TAR = Tertiary Air Register
All Registers 100% Open
PAR =10% Open
SAR = TAR = 100% Open
PAR = 50% Open
SAR = TAR = 100% Open
PAR = SAR = 50% Open
TAR = 100% Open
SAR = Closed
PAR = TAR = 100% Open
234
EXCESS OXYGEN, %, dry
Figure 4-33. The effects 'of air register adjustments on NO emissions for
the tertiary air burner firing No. 6 oil.
4-46
KVB 6015-798
-------
varied from 526 ppm (corrected to 3% O , dry) at 6.5% C>2 to 200 ppm (corrected
to 3% 0 , dry) at 0.35% O . The CO concentration at the latter C>2 was
> 2000 ppm. At an optimum-low O of 1.2%, NO emission was 295 ppm (corrected
to 3% O , dry), or 33% less than the emission at 3.2% O2 (439 ppm, corrected
to 3% O , dry). Also, at this point all registers were closed somewhat,
increasing the draft and lengthening the flame.
For one test, major register adjustments were made: the primary air
register was very nearly closed, the secondary air register was 25% open, and
the tertiary air register was 100% open. In that case, at 3.0% O , the NO
concentration was 329 ppm (corrected to 3% O , dry), or approximately 25% less
than the 439 ppm measured at 3.2% O with all registers 50% open. Figure
4-34 summarizes the shale oil test data.
Samples of the shale oil and No. 6 oil were taken, and the analyses
are shown in Table 4-1.
E. Low-NO Burner (Recirculating Tile)
A low-NO burner incorporating a self-recirculating tile was evaluated
in the research furnace. A special tile was used to achieve some recircula-
tion of fuel vapors and the products of combustion in the immediate vicinity
of the burner. The recirculation of these gases is intended to lower the
flame zone temperature and, thus, lower thermal NO . The tile was located
X
on the furnace floor just above the fuel injection plane.
Instead of a throat (used in the conventional burner) the burner had
a 25-cm-diameter (10-in.) diffuser cone attached to the oil gun to create a low
pressure zone near the fuel injection plane.
Figures 4-35a and 4-35b are a plan view of the burner showing the
positioning of the diffuser cone and oil and gas tips, and a schematic of
the recirculating tile, respectively.
Figure 4-36 shows the gas tip configurations used in the tests with
natural gas. Configuration III was used initially but resulted in hiqh NO
x
levels at baseline conditions (129 ppm corrected to 3% 0 dry). In an
attempt to lower the NO by increasing the amount of recirculation, the gas
tips were lowered 1.3 cm (1/2 in.). The NO level increased, however, probably
because of the increased recirculation of high temperature combustion products
without sufficient entrainment of cooler gases.
4-47 KVB 6015-798
-------
T3
4-1
IT)
300
700
600
500
400
0,
o 300
200
100
Fuel: Shale Oil
Firing Rate: 1.29 MW (4.4x10
Oil Tip: 764
( ) Test No.
PAR = Primary Air Register
SAR = Secondary Air Register
TAR = Tertiary Air Register
Btu/hr)
(1/3-15)
(1/3-20)
(1/3-18)
CO Limit
All Registers 100% Open
All Registers 50% Open
PAR = 25% Open, SAR = 40% Open
TAR = 50% Open
PAR = Closed, SAR
TAR = 100% Open
25% Open
I
345
EXCESS OXYGEN, %, dry
Figure 4-34. The effects of excess oxygen and register adjustments on NO
emissions for the tertiary air burner firing shale oil.
4-48
KVB 6015-798
-------
T//>
7?MOT"
Figure 4-35a. Plan view of recirculating tile burner for natural draft process
heater. . .
r~\
Figure 4-35b. Schematic of recirculating tile showing cross-section (left) and
0,,.. . . elevation (right).
4_49 KVB 6015-798
-------
4 Firing Ports
As Shown
60° off Vert. C
lb>
o
45° off Vert.
15° 15
20
1 - 0.16 cm
(1/16 in.) diam Ign
Port at 45° off
Vert. C
2 - 0.2705 cm (0.1065 in.) diam.
Firing Ports at 45° off
Vert. C
1 - 0.16 cm (1/16 in.) diam Ign
Port at 45° off Vert. C
CONFIGURATION III
CONFIGURATION I
o
M
U1
Figure 4-36. Gas tip hole drilling patterns for recirculating tile burner.
05
-------
A tip pattern having injection orifices producing less swirl (Configura-
tion I) was tried next. The intent was to decrease mixing and lengthen the
flame. The NO level decreased to a baseline value of 104 ppm (corrected to
3% O , dry) with these gas tips, still only slightly lower than the baseline
NO value of 114 ppm (corrected to 3% O , dry) found on the conventional MA-16
X ^
burner.
Using the Configuration I tips, excess 0 was varied from 4.4%
down to the limit at which CO appeared, 0.6%. At 0.6% O , NO emission
^ X
decreased by 20% to 83 ppm (corrected to 3% 0 , dry). The CO concentration
at this condition was 44 ppm (corrected to 3% 0 , dry). The effects of
excess oxygen as well as the change in gas tips on NO emissions is shown
X
in Figure 4-37.
The effect of firing rate changes on NO emissions for natural gas
fuel is shown in Figure 4-38. NO emission levels remained constant (115 ppm,
X
corrected to 3% 0 , dry) at firing rates from 50% to 100% of capacity (100%
capacity = 6.5xl06 Btu/hr) and dropped somewhat (to 110 ppm, corrected to
3% O , dry) at 28% of capacity. Excess O was necessarily high, however, at
this low capacity to provide flame stability. One explanation for the
flatness of the NO vs. firing rate curve is that a greater degree of mixing
offsets the lower firebox temperatures at low firing rates. The most likely
explanation, however, is that the burner tile temperature is probably
relatively constant over most of the load range. Thus, as the firing rate
was decreased, the air flow decreased while the residence time in the burner
tile increased. These offsetting trends probably resulted in the constant
NO levels.
x
Tests on the recirculating tile burner were also conducted firing
No. 2 oil using an oil tip with a 60-deg. spray angle (30 deg. either side
of the vertical centerline of the furnace). No. 2 oil was used rather than
No. 6 oil, for which tests had originally been planned, because of coking
problems which had been experienced with No. 6 oil on this particular type of
burner. The wide angle oil tip was expected to result in better recirculation
by having the flame impinge on the burner tile just below the top of piece C.
4-51
KVB 6015-798
-------
180
160
140
120
TJ
(N
0 100
I
& 80
60
40
20
(1/2-2)
(1/2-1)*
(1/2-9)
Baseline
(1/2-8)
(1/2-11)
(1/2-10)
CO limit
*Gas tip moved 13 mm (1/2 in.) deeper axially into
air stream
( ) Test Number
Recirc. Tile Brn., Patt. Ill Gas Tip"
Recirc. Tile Brn., Patt. I
Fuel: Natural Gas
Firing Rate: 1.5 MW (5xl06 Btu/hr) "~
I
Figure 4-37.
0123 456
EXCESS OXYGEN, %, dry
The effect of excess oxygen on NOX emissions for the recirculating
tile low-NOx natural draft burner firing natural gas.
4-52
KVB 6015-798
-------
180
160
140
120
n
V>
(0
I 80
60
40
20
(1/2-2)
(1/2-1)*
(1/2-15)
8.9% O. |
(1/2-14) (1/2-12) (1/2-13)
Baseline
Fuel: Natural Gas
100% Capacity: 1.9 MW (6.5x10 Btu/hr)
Excess O2: 3%
( ) Test No.
*Gas Tips 13 mm (1/2") deeper axially into
air stream
Recirculating Tile Burner, Pattern III Tips
Recirculating Tile Burner, Pattern I Tips
_L
20 40 60 80
FIRING RATE, % of Capacity
1
100
120
Figure 4-38. The effect of firing rate changes on NOX emissions for recircula-
ting tile low-NOx natural draft burner firing natural gas.
4-53
KVB 6015-798
-------
Baseline NO emissions (at 4% excess O.J for the recirculating tile
X &,
burner firing No. 2 oil were 110 ppm (corrected to 3% 0 , dry). Excess O
was varied from 5.1% down to 0.5% at which point the CO concentration was
147 ppm (corrected to 3% 0 , dry). The lowest NO emission occurred at an
excess 0 of 1.4% and was 98 ppm (corrected to 3% 0 , dry), down 11% from the
baseline value. Further reduction of the excess 0 appeared to have no
significant result on NO emissions. Figure 4-39 summarizes the effect of
excess 0 on NO emissions.
Z. X
Figure 4-39 also shows a significant drop in NO emissions at 50%
X
capacity for excess 00 of 5.7%. The NO concentration at this condition was
£ X
85 ppm (corrected to 3% O , dry), or 23% less than baseline. Time constraints
prevented testing at other firing rates.
F. Flue Gas Recirculation
Flue gas recirculation has been demonstrated to be an effective
method of NO reduction for industrial boilers. Thus far, flue gas recircula-
tion has not been applied to process heaters for NO reduction. The objective
of this test series was to evaluate the effect of flue gas recirculation on
gaseous emissions and thermal efficiency in a process heater. It was not
possible to duct actual flue gases from the stack to the burner because a high
temperature fan was not available. In order to simulate FGR, a system was
installed as shown schematically in Figure 4-40. An auxiliary burner was
installed which exhausted into a combustion air duct leading to the burner
plenum. A gas-gas heat exchanger was installed to control the temperature of
the combustion air-flue gas mixture. The percentage of recirculated flue gas
was varied by adjusting the firing rate for the auxiliary burner. Flue gas
recirculation rates were varied up to a maximum of 40% approximately when
firing natural gas and No. 6 oil.
The effect of FGR rate is shown in Figure 4-41 where NO is plotted as
a function of percent FGR for natural gas firing. The flue gas recirculation
is defined by the following expression:
Recirc. mass flow rate x 100
s FGR
Combustion air flow + Recirc. mass flow + Fuel flow
4-54 KVB 6015-798
-------
T)
-P
a
a
140
120
100
80
60
40
20
1 T
1 T
I T
(1/2 - 6)
(1/2 - 5)
CO limit
(1/2 - 3)
Baseline
(1/2 - 4)
50%
Capacity Q
(1/2 - HI
Fuel: Ho. 2 Oil
Firing Rate: -1.5 MW (5x10 Btu/hr)
Tip: 766
( ) Test No.
J \ \ I I I
234
EXCESS OXYGEN, %, dry
Figure 4-39. The effect of excess oxygen on NO emissions for the recirculating
tile burner firing No. 2 oil.
4-55
KVB 6015-798
-------
w.
10"
DUCT
STACK PROBE
RESEARCH
FURNACE
BLOWER
AUX. BURNER
COMB.
AIR
BLOWER
GAS-GAS
HEAT EXCHANGER
Figure 4-40. Schematic of FGR setup at Location 1.
4-56
KVB 6015-798
-------
100
^^ Baseline
a
dp
ro
-P
iti
04
8
50
en
o
Fuel: Natural Gas
Firing Rate: 1.48 MW
(5.1 Btu/hr x 106)
Gas Tip: Pattern II
Low O
Condition
I
I
VD
00
10
20 30
RECIRCULATED FLUE GAS,
40
50
60
Figure 4-41. The effect of flue gas recirculation on NO emissions (natural gas)
-------
A reduction in NO of 57% from the baseline condition was achieved with FGR
x
at the normal O level. The overall O level was reduced until the CO limit
£ £
was reached. This limiting value of excess 0 was 0.7% O. A reduction in
NO of 62% was measured with the combination of FGR and low 0 operation.
Figure 4-42 presents NO as a function of FGR rate for No. 6 oil firing.
X
FGR alone resulted in a reduction of 34% at the maximum recirculation rate.
The combination of FGR and low 0 operation yielded a reduction of 40% in NO
^ X
emissions.
G. Steam Injection
The effect of steam injection on NO and NO emissions was evaluated for
natural gas firing with an MA-16 burner. Two methods of steam injection were
tried. In the first method, steam was injected into the gas manifold and the
steam/gas mixture then injected radially inward through the normal gas tips.
The steam flow rate was varied up to a maximum of 0.0098 kg/sec (78 Ib/hr).
The effect of steam injection flow rate on NO emissions is shown in Figure 4-43.
The maximum reduction in NO occurred with the maximum steam flow rate. NO
emissions were reduced 32% from the baseline condition at 0.0098 kg/sec (78
Ib/hr) flow rate.
An alternate method of steam injection was evaluated to determine the
influence on NO emissions. In this method the steam was injected through the
oil gun of a DBA-16 burner at the burner centerline. Since steam for fuel oil -
atomization is already supplied to the oil gun, injection through the oil gun
is a simpler modification than steam injection through the gas tips. Further,
it was hoped that by experimenting with the positioning of the oil tip relative
to the gas tips, NO emissions could be reduced below the levels of the previous
X
tests.
Figure 4-44 shows the effect of steam injection through the oil gun
on NO emissions for the DBA-16 burner with Pattern II tips, normal orienta-
tion. Maximum reduction in NO was achieved at the highest rate of steam
X
injection114 ppm at 0.0095 kg/s (75 Ib/h) steam flow. Very little difference
in NO production was observed at the other steam flow rate used (0.0067 kg/s
X
or 53 Ib/h). Thus, the lowest NO emissions for steam injection through the
oil gun were 16% less than the normal baseline (3% O ) NO levels for the
DBA-16 burner.
4_58 KVB 6015-798
-------
200
150
dP
n
41
(0
I
a
8
100
50
Fuel: No. 6 oil
Firing Rate: 1.49 MW
(S.OxlO6 Btu/hr)
Burner Tip: 764
J_
I
10 20
RECIRCULATED FLUE GAS, %
30
40
Figure 4-42. Tiie effect of flue gas recirculation on NO emissions (Wo. 6 oil).
4-59
KVB 6015-798
-------
100
<*>
f>
50
(1/5-1)
3.4%
(1/5-2)
3.3%
Fuel: Natural Gas
Firing Rate: 1.58 MW (5.4xl06 Btu/hr)
Gas Tip: Pattern II
(Test No.)
Excess O
Steam injection through gas tips
1
I
I
0.0025
(20)
0.0050
(40)
0.0076
(60)
0.0101
(80)
0.0126
(100)
STEAM INJECTION, kg/s (Ib/hr)
Figure 4-43.
The effect of steam injection on NO emissions for the MA-16
burner firing natural gas.
4-60
KVB 6015-798
-------
140
125
100
<*>
m
50
25
3.1%
2.9%
Fuel: Natural Gas -
Firing Rate: 1.53MW (5.22x10
Btu/hr)
Gas Tip: Pattern II (impinging on
burner tile)
(Test No.)
Excess O
Steam injection through oil gun
I
I
0.0031 0.0063
(25) (50)
STEAM FLOW, kg/s (Ib/hr)
0.0094
(75)
Figure 4-44.
The effect of steam injection on NO emissions for the
DBA-16 burner firing natural gas.
4-61
KVB 6015-798
-------
The influence of steam injection on NO emissions was not nearly as strong
X
for steam injection through the oil gun as it was for steam injection through the
gas tips as is seen by comparing the slopes of the curves in Figures 4-43 and 4-44.
H. Altered Fuel Injection Geometry
Previous work with boilers has shown that NO emissions can be reduced
X
by altering fuel injection geometry to produce locally fuel-rich zones in the
flame. The fuel-rich zones are at a lower temperature and result in lower
overall NO production. Based on the results of cold flow tests in KVB's
laboratory the fuel injection geometry was modified for a DBA-16 natural
draft burner.
A DBA burner was selected because the straight-sided tile more closely
resembled the cold flow conditions than did the MA-16 tile. Figure 4-45
presents a schematic comparison of the two types of burner tiles.
Tests revealed that NO emissions at baseline conditions (with no
X
modifications) for the DBA-16 burner were higher than the emissions from the
MA-16 burner tested (see Figure 4-46) , contrary to expectations. Apparently,
the straight-sided tile of the DBA-16 confines high-temperature combustion
gases to a smaller volume, resulting in more intense burning and, consequently,
higher HO emissions.
x
In the first test series standard Pattern II gas tips were installed
in the DBA-16 with the center firing port facing radially outward, perpendicular
to the burner circumference. On the basis of the cold flow test results
reported by KVB, this tip orientation was expected to delay mixing of fuel
and air, thus producing a longer, less intense flame and lower NO emissions.
The tests showed that NO emissions were indeed lower for this tip
configuration than for the standard configuration (compare Figures 4-46 and
4-47). At 3% excess 0 , the NO concentration with outward-facing firing
^ X
ports was approximately 94 ppm (dry, corrected to 3% O ), about 33% lower
than the NO emissions from the standard tip orientation. At an excess 0 of
x 2
1.1%, the NO level was 78 ppm, a 32% reduction from the standard orientation
X
NO level. The CO limit occurred at 0.5% O , compared'with 0.3% O for the
x z 2
standard configuration with a CO concentration of 615 ppm. NO at this point
was down to 73 ppm, 29% below the standard configuration value.
4-62 KVB 6015-798
-------
Not to Scale
_ 46 CITL_
(18 in.)
41 cm
~16 in.)
-H
Standard Tile for MA-16 Burner
Standard Tile for DBA-16 Burner
o
h-1
tn
Figure 4-45. Comparison of the tiles used in the conventional natural draft process heater
burners tested at Location 1.
00
-------
150
125
100
4J
(d
I
75
50
25
(1/1-6)
Fuel: Natural Gas
Firing Rate: <\J.49MW
10 Btu/hr)
(Test No.)
I
MA-16
DBA-16
I
2 3
EXCESS" OXYGEN, %, dry
rigure 4-46. NOX emissions as a function of excess 02 for natural draft
burners firing natural gas, normal tip configuration.
4-64
KVB 6015-798
-------
150,
125
100
o
df
n
ft
O
2
75
50
25
(1/8-10)
(1/8-9)
(1/8-8)
(1/8-7)
Fuel: Natural Gas ,
Firing Rate: 1.49MW (5.08x10
Btu/hr)
Gas Tip: Pattern II
(Test No.)
1
2 3
EXCESS OXYGEN, %, dry
Figure 4-47. NO emission as a function of excess O_ for DBA-16 burner
firing natural gas with gas tips radially outward.
4-65
KVB 6015-798
-------
The flame shape with the reverse tip orientation was shorter than the
normal flame and segmented into four fuel-rich regions, one above each of the
gas tips. The flame appeared to be quite lazy at low firing rates.
I. Summary of Hot-Firing Test Results
In Figures 4-48 and 4-49 the NO emissions from the conventional MA-16
X
and DBA-16 burners and the tertiary air burner for various test conditions are
graphed as a function of stack excess oxygen, firing natural gas and No. 6 oil.
The unmodified burners (all registers 100% open in the case of the tertiary
air burner) are represented by heavy symbols and curves, and modifications to
each burner are represented by light symbols and curves.
It is important to note that the percent reductions in NO shown in
Table 4-3 for the lowest O conditions are probably not attainable in an actual
process heater since they were obtained at or very near the CO limit. Operation
with such low excess air would not be possible in a natural draft process
heater because it is likely to result in positive stack static pressures which
would violate plant safety codes. In addition, the lack of fine control of
the fire at low drafts increases the chances for operation with high levels
of combustibles. Thus, these figures probably represent an ideal (but
practically unattainable) upper bound for NO reduction potential for the
various combustion modification techniques considered.
Table 4-3 shows that the largest percent reductions in NO occurred
with staged air or flue gas recirculation techniques. With SCA, these reduc-
tions seem to be a relatively strong function of excess air whereas with FGR
they are a rather weak function of excess air. With natural gas fuel, all
modifications (except the tertiary air burner) appeared to increase furnace
efficiency. With No. 6 oil, efficiency decreased slightly with SCA and
decreased with FGR, but increased when FGR was coupled with low excess air.
4-66 KVB 6015-798
-------
360
o
dP
CO
a
ft
o
2
320
280
240
200
160
120
80
40
I I I I I
Fuel: No. 6 Oil g
Firing Rate: 1.52 MW (5.2x10 Btu/h) nom.
(CO concentration)
- denotes data_at or below the CO limit
MA-16
Unmodified
- A Staged Air (4 tubes)
I""! Flue Gas Recirculation (40%
i J nom . )
LOW-NO (TERTIARY AIR) BURNER
^
registers 100% open (unmod.)
Extended Secondary tile,
All registers 100% open
_L
234
STACK EXCESS OXYGEN, %, dry
Figure 4-49. Summary of NO emissions as a function of excess oxygen for
subscale natural draft furnace firing No. 6 oil.
4-68
KVB 6015-798
-------
o
OP
180
160
140
120
100
80
60
40
20
I I I I
Fuel: Natural Gas
Firing Rate: 1.52 MW (5.2x10 Btu/h) nom.
(CO concentration)
- denotes CO limit
""D-39 ppm
(615 ppm) QS
(47
ppm)
~
-------
TABLE 4-3. SUMMARY OF NOX REDUCTION AND EFFICIENCY CHANGE AS A FUNCTION OF COMBUSTION MODIFICATION
TECHNIQUE FOR NATURAL GAS AND NO. 6 OIL FOR NATURAL DRAFT BURNERS.
fuel
Average Baseline NOX
MA-16
DBA- 16
Combustion Modification Technique
Lowered Excess Air
Staged Combustion Air
Floor Lances, Normal O?
Floor Lances, Low O
Central Cylinder, Normal ©2*
Central Cylinder, Low O2*
Tertiary Air Burner, Lowest NOX
Configuration (relative to average
baseline NOX for the MA-16)
Flue Gas Recirculation
Normal O_
Low O
Steam Injection, Normal O_
Altered Fuel Injection Geometry
Normal O *
Low O *
Natural Gas
ppm,dry @ 3% O
107
131
NOX Reduction
27
46
67
31
59
30
59
63
33
31
44
ng/J
54.6
66.8
Efficiency Change
+ 4.7
+ 0.7
+ 2.6
0.0
+ 3.4
- 2.0
+ 4.7
+ 4.9
0.0
+ 3.4
No. 6 Oil
ppm,dry @ 3% 0
285
NOx Reduction
10
35
51
42
31
39
ng/J
160
Efficiency Change
+ 0.1
- 0.7
- 0.4
0.0
- 2.6
+ 2.0
en
o
Ul
i
NO reduction is relative to average baseline NO for the DBA-16.
X X
ID
oo
-------
In cases where baseline data could not be taken on the same day during
which modifications were tested, the average NO concentration for the modi-
fication is compared to the average NO from the unmodified burner at baseline
X
conditions. The average unmodified burner baseline NO concentrations are
X
given in Table 4-3. NO emission reduction for the tertiary air burner was
x
reported relative to the average unmodified baseline emissions from the MA-16
burner. The efficiency changes reported in Table 4-3 result from a comparison
of the efficiencies occurring at modified conditions with efficiencies measured
at baseline conditions at similar stack temperatures using a conventional
MA-16 burner.
The percent reductions in NO observed for modifications to the DBA-16
X
burner are expected to occur for the same modifications to the MA-16 burner
with the possible exception of AIG (where the difference in burner tiles may
play an important role in the mixing patterns resulting from the modified
injection scheme).
4.1.4 Cost Effectiveness of Combustion Modifications to Natural Draft Process
Heater Burners
A. Summary
The cost effectiveness of the combustion modifications applicable to
natural draft process heaters has been evaluated, and the results are
summarized in Table 4-4. All costs are based on 1978 dollars.
The initial installed cost for each of the modifications is shown in
Table 4-5 for three different heater sizes. The largest and smallest sizes
147 MW (500x10 Btu/h) and 2.9 MW (10x10 Btu/h), respectivelyrepresent
the two extremes in firing rate for refinery process heaters. The intermediate
size of 73.3 MW (250x10 Btu/h) was chosen for this cost analysis because it
is the current size limit above which steam boilers are regulated by federal
emission standards.
The total annualized cost per 10 kg of NO reduction shown in Table
X
4-4 was determined by amortizing the initial fixed capital costs given in
Table 4-5 at 20% (corresponding to straight-line depreciation of the capital
equipment over 12 years, and assuming a .10% cost of capital, state and federal
taxes totalling approximately 11%, and insurance charges of 0.5%). The annual
4~70 KVB 6015-798
-------
TABLE 4-4. COST EFFECTIVENESS ($/10 kg of NOX reduction) OF COMBUSTION
MODIFICATIONS TO A NATURAL DRAFT PROCESS HEATER
(NOT INCLUDING ANNUAL FUEL COSTS/SAVINGS)
Modification
Low Excess Air
Altered Injection
Geom.
(Normal O )
Altered Injection
Geom.
(with LEA)
Staged Air - Central
Cyl. (Normal 0 )
Staged Air - Central
Cyl. (with LEA)
Staged Air - Floor
Lances (Normal 0 )
Staged Air - Floor
Lances (with LEA)
Flue Gas Recircula-
tion (Normal O_)
Flue Gas Recircula-
tion (with LEA)
Steam Injection
(no initial cost)
Steam Injection
(incl. initial cost)
Tertiary Air Burner
Heater Size
2.9 MW
(lOxlO6 Btu/h)
Natural No. 6
Gas Oil
0 0
$6.60
$4.70
$200
$100
$1000 $460
$710 $320
$1800 $1200
$1700 $940
$990
$1400
$250 $60
73.3 MW
(250xl06 Btu/h)
Natural No. 6
Gas Oil
0 0
$0.78
$0.55
$43
$23
$160 $70
$110 $48
$320 $200
$300 $160
$970
$1100
$250 $60
147 MW
(SOOxlO5 Btu/h)
Natural No. 6
Gas Oil
0 0
$0.78
$0.55
$33
$18
$130 $57
$87 $39
$320 $200
$300 $160
$960
$1000
$250 $60
4-71
KVB 6015-798
-------
TABLE 4-5. INITIAL INSTALLED COSTS OF COMBUSTION MODIFICATIONS TO
A NATURAL DRAFT PROCESS HEATER (in $)
Modification
Low Excess Air
Alternate Injection
Geom.
Staged Air (central
cyl.)
Staged Air (floor
lances)
Flue Gas Recirculation
Steam Injection*
Tertiary Air Burner
Heater Size
2.9 MW
(lOxlO6 Btu/h)
0
50
1500
9500
18500
3000
1500
73.3 MW
(250xl06 Btu/h)
0
150
8300
36500
54000
16300
37500
147 MW
(SOOxlO6 Btu/h)
0
300
12800
59000
94000
27800
75000
*If no existing capability
4-72
KVB 6015-798
-------
capital charge was added to the annual operating and maintenance cost to
obtain the total annualized cost. These costs are shown in Table 4-6. (Annual
operating costs did not include projected fuel savings or costs resulting from
modifications for reasons explained below.) The total annualized cost was then
divided by the annual reduction in NO emissions to obtain the cost effective-
x
ness values in Table 4-4.
The annual reduction in NO emissions was calculated for each modifica-
x
tion from the percent NO reduction listed in Table 4-3 using the formula
X
* ron rt Average baseline emissions
Annual NO reduction (10 kg) = x-,^ UC 1On * from conventional burner
x 1UU . . _.
(ng/J)
x heat input rate (W) x 31.536xl06 sec/y x 0.8 (use factor) x metric tons
ng
NO emission reductions were determined relative to the conventional MA-16
burner for the following modifications:
1. Lowered Excess Air (LEA)
2. Flue Gas Recirculation (FGR)
3. Staged Combustion Air - Floor Lances (SCA-L)
4. Steam Injection {STM)
5. Tertiary Air Burner (TAB)
NO emission reductions were determined relative to the conventional DBA-16
X
burner for these modifications:
1. Altered Fuel Injection Geometry (AIG)
2. Staged Combustion Air - Central Cylinder (SCA-C)
Note that although Table 4-3 gives values for efficiency changes
associated with each modification, these values are not appropriate for
estimating annual fuel costs or savings. They are useful only inasmuch as
they indicate expected trends in fuel consumption. This is so because the
research heater tested by KVB at Location 1 had no process tubes and, therefore,
the data do not reflect any inefficiencies or variabilities due to changes in
heat transfer to a process stream.
4-73 KVB 6015-798
-------
TABLE 4-6. TOTAL ANNUALIZED COSTS (IN $) NOT INCLUDING FUEL COSTS (SAVINGS)
OF COMBUSTION MODIFICATIONS TO A NATURAL DRAFT PROCESS HEATER
(AMORTIZING INITIAL CAPITAL COSTS AT 20%)
Modification
Low Excess Air
Altered Injection
Geom.
Staged Air - Central
Cyl.
Staged Air - Floor
Lances
Flue Gas Recircula-
tion
Steam Injection
(if initial instal-
lation necessary)
Steam Injection
(no initial instal-
lation required)
Tertiary Air Burner
Heater Size
2.9 MW
(lOxlO5 Btu/h)
0
10
300
1900
4300
1900
1300
300
73.3 MW
(250xl06 Btu/h)
0
30
1660
7330
18800
35400
32200
7500
147 MW
(SOOxlO6 Btu/h)
0
60
2560
11800
38100
69900
64300
15000
4-74
KVB 6015-798
-------
Figures 4-50, 4-51, 4-52, and 4-53 illustrate the relationships
between estimated costs of NO removal and heater size.
x
These figures show that the simplest modifications are the most cost
effective. The least expensive modifications, AIG and SCA-C, were tested only
in gas-firing application. It is possible that both techniques may be adapted
to handle oil-firing applications as well. The more involved modifications,
FGR and SCA-L, are less cost effective although they produced the largest
percent NO reductions.
Most modifications result in lower costs per metric ton of NO removed
X
as heater size increases. Only STM and TAB cost effectiveness ratios appear
to be relatively independent of size. For the other modifications, both on
natural gas and No. 6 oil-firing, the cost effectiveness decreased as heater
size increased from 2.9 MW (10x10 Btu/h) to 73.3 MW (250x10 Btu/h) according
to the relation
CE at 73.3 MW /73.3'a
CE at 2.9 MW \ 2.9
where - 0.67 _< a _< - 0.47, a = - 0.56, and S (standard deviation) = 0.07.
(Note that a is the slope of the line segments in Figures 4-50 to 4-53.)
Since
-0.56
CE « (size)
and since
1.0
therefore,
NO reduction a (size)'
X
total annualized cost a (size) .
For example, using the total annualized cost for a FGR system for
a 2.9 MW (10x10 Btu/h) heater given in Table 4-6 at $4300, one can calculate
approximately the total annualized cost of FGR for a 73.3 MW heater as
follows:
73'" ' x (4300) = $17,810
4-75 KVB 6015-798
-------
100.00
.0 10.00
o
d
o
2
Cn
W
O
u
1.00
0.10
T r i 111111
T IT
l-
Altered Fuel Injection Geom.
Normal O,,
Low 0,
I I I I I I III
I I I i I I 111
10
HEATER SIZE,
100
MW
Figure 4-50. Estimated cost as a function of heater size for altered fuel
injection geometry modification to natural draft process heaters
firing natural gas only.
4-76
KVB 6015-798
-------
10000
3 looo
u
p
D
V
K
O
Z
CP
o
rH
v>
o
100
10
I I I II III!
I I II I Ml I
I -
Flue Gas Recirculation
Staged Combustion
Air - Floor Lances
Normal O.
Low O,
Normal O,
Low CL
Normal 0
Staged Combustion
Central-Cylinder
I I I I I I III
I I I I I I II I
10
HEATER SIZE, MW
100
Figure 4-51. Estimated cost as a function of heater size for three combustion
modifications to natural draft process heaters firing natural
gas only.
4-77
KVB 6015-798
-------
10000
c
o
o
D
"3
1000
O
Z
Cn
O
u
100-
10
^ I I! I I M III
II I I I I I II
Steam Injection
No Existing
Steam Lines
With Existing
Steam Lines
Tertiary Air Burner
II I I I I I III
I I I I I I III
10
HEATER SIZE, MW
100
Figure 4-52. Estimated cost as a function of heater size for steam injection
modification and for changeover from conventional to tertiary air
burner in natural draft process heaters firing natural gas only.
4-78
KVB 6015-798
-------
10000
.3 1000
in-
EH
10
o
u
100
10
1 I I I I I III
1 I I I I IIII
Flue Gas Recirculation
Normal O,
Staged Air -
Floor Lances
Tertiary Air Burner
Low 0,
Normal O_
Low 0,
I I I I I I III
I I I I I I III
10
HEATER SIZE, MW
100
Figure 4-53. Estimated cost as a function of heater size for two combustion
modifications and for changeover to tertiary air burner in
natural draft process heaters firing No. 6 fuel oil only.
4-79
KVB 6015-798
-------
This compares well with the actual value of $18,800 given in Table 4-6 for
a 73.3 MW heater.
For heaters larger than 73.3 MW (250x10 Btu/h) there is greater
variability in the cost/size relationship among the various modifications.
In all cases, however, total annualized costs increase more rapidly with
size than they did for heaters between 2.9 MW and 73.3 MW. In the cases of
FGR and AIG, the total annualized cost becomes directly proportional to the
first power of heater size. For SCA-C and SCA-L the total annualized cost
is roughly proportional to heater size to the two-thirds power.
One is cautioned, however, against applying these power laws
indiscriminately. They may be used to give a first approximation of
modification costs in cases where the modified system incorporates the
changes and additions to the original unit outlined in part B of this section.
In Figures 4-54, 4-55, and 4-56 the cost effectiveness versus the
potential NO reduction possible for each modification is plotted for the
X
three heater sizes. Where two values are plotted for the same modification
and the same fuel, the one associated with the larger NO reduction corresponds
X
to operation at low excess air and the other corresponds to normal excess air.
Where there is only a single value for a particular modification (other than
LEA), operation at normal excess air is understood.
B. Determination of Initial Capital Costs and Annual Operating Costs
1. Lowered Excess AirThere are no initial costs associated with LEA
provided damper controls and accurate oxygen-measuring instruments are
available. Measurement of CO is also required.
Zero annual operating costs (or savings) are reported here although
some fuel savings results with the use Of LEA. There are some probable
costs associated with monitoring heaters firing closer to stoichiometric,
but these are likely to be small and variable from plant to plant.
2. Flue Gas RecirculationFor the purposes of cost estimating, all burners
are assumed to be the same size, i.e., 2.9 MW (10x10 Btu/h). Thus, the small
heater has one burner, the intermediate size has 25 burners, and the large
4_80 KVB 6015-798
-------
2000
1800
1600
1400
c
3 1200
u
3
o
&
X
o
53 1000
en
ro
w
8
800
600
400
200
1
1
1
1
2
3
4
5
6
7
I I
- Low Excess Air
Altered Injection Geometry
- Staged Combustion Air - Central Cylinder
- Staged Combustion Air - Floor Lances
- Flue Gas Recirculation
- Tertiary Air Burner
- Steam Injection - Existing Steam Lines
~ Natural Gas Firing
I 1- No. 6 Oil Firing
©
m
m
20 Vi/3(>ix 40
NO REDUCTION, %
50
60
70
Figure 4-*54. Estimated cost of combustion modifications as a function of
percent NOx reduction for a 2.9 MVJ (lOxlO6 Btu/h) process heater.
4-81
KVB 6015-798
-------
900
800
700
Reduction
&
o
0
X
i 500
CP
fl
O
^ 400
o
u
300
200
100
0
C
1 '0 ' '
1 - Low Excess Air
2 - Altered Injection Geometry
3 - Staged Combustion Air - Central Cylinder
4 - Staged Combustion Air - Floor Lances
5 - Flue Gas Recirculation
6 - Tertiary Air Burner
7 - Steam Injection - Existing Steam Lines
fj- Natural Gas Firing
1 1- No. 6 Oil Firing
©
CD
CD ©
,JD E n
© bJ
m i {Ti v^ i (?} i
' 1 20 ^^^30-^ 40 ^"^ 50
1
_
GL)
© ~~
0L
\M
60 70
NO REDUCTION, %
Figure 4-55. Estimated cost of combustion modifications as a function of percent
NO reduction for a 73.3 MW (250xl06 Btu/h) process heater.
4-82
KVB 6015-798
-------
J.UUU
900
800
700
c
0
H
-M 600
0
3
0)
X
i 500
n
o
^ 400
w
O
u
300
200
100
0
' ' '© ' ' '
1 - Low Excess Air
2 - Altered Injection Geometry
3 - Staged Combustion Air - Central Cylinder
4 - Staged Combustion Air - Floor Lances
5 - Flue Gas Recirculation
"" 6 - Tertiary Air Burner ~~
7 - Steam Injection - Existing Steam Lines
(/ - Natural Gas Firing
~" 1 1 - No. 6 Oil Firing """
"^ ^"
._
09
©
-ED
QD
©
^JD Q rn ^
i \ r\r< i r\ i Q
0 10 20 V-'30V^/ 40 ^ 50 60 7
NO REDUCTION, %
Figure 4-56. Estimated cost of combustion modifications as a function of percent
NO reduction for a 147 MW (SOOxlO6 Btu/h) process heater.
4-83
KVB 6015-798
-------
heater has 50 burners. The number of burners involved is important in
determining lengths of pipe or ductwork required, the number of tees, ells,
crosses, etc., for each modification technique.
The initial fixed costs for all modifications were estimated using
References 10 to 14 and best judgment. Cost values presented in this section
which were determined without reference to published cost data will be
identified as such.
For FGR the initial fixed costs were determined as shown in Table 4-7.
TABLE 4-7. INITIAL INSTALLED COSTS (IN $) OF FLUE GAS RECIRCULATION
2.9 MW
Item/Heater Size (lOxlO6 Btu/h)
Fan , Motor & Drive
Damper
Ductwork & Burner Plenum
Duct Insulation
Instrumentation & Control
Systems
Engineering/Design
Totals
2,000
500
2,000
2,000
10,000
2,000
?18,500
73.3 MW
(250xl05 Btu/h)
10,000
500
10,000
8,500
15,000
10,000
$54,000
147 MW
(SOOxlO6 Btu/h)
23,000
1,000
20,000
15,000
15,000
20,000
$94,000
The fan, motor, and drive costs above were obtained from References
11 and 12 along with fan power requirements and other operating data. Table
4-8 summarizes the cost data on fans applicable to either FGR or SCA-L. These
fan sizes and costs are based on the air flow requirements for approximately
30% FGR.
Ductwork costs are based on 0.32 cm (1/8") steel duct material,
30 cm x 60 cm (I1 x 2'), and 15 m (50') long for the 2.9 MW heater, 30 m
(100') long for the 73.3 MW heater, and 60 m (2001) long for the 147 MW
heater. Based on past experience, the cost per foot of length of ductwork
was estimated to be $20. Installation cost of ductwork hangers was assumed
to be equal to the cost of the duct material itself.
4-84
KVB 6015-798
-------
TABLE 4-8. DELIVERED/INSTALLED* COSTS (IN $) OF FANS, ASSOCIATED MOTORS, AND
DRIVES AND POWER REQUIREMENTS AS A FUNCTION OF GAS TEMPERATURE
AND VOLUME FLOW RATE
Gas
Temperature
K (°F)
294 (70)
533 (500)
810 (1000)
470 (1000)
300/500
(1/2 hp)
1000/1600
(1.4 hp)
3500/5600
(8 hp)
Volume Flow
m3/s (SCFM)
9400 (20000)
1200/1900
(10 hp)
5600/8900
(25 hp)
17000/27000
(50 hp)
18800 (40000)
2200/3500
(30 hp)
14000/22300
(60 hp)
43000/68400
(150 hp)
*Installed values = 1.59 x Delivered Values, Rounded to nearest $100
Installed costs include equipment foundations, electrical, paint and field
labor (according to Ref. 10) .
4-85
KVB 6015-798
-------
In the case of the small heater a burner plenum requires a rather
small amount of duct material and installation material and labor. In the
larger heaters, however, the plenum may represent a considerable fraction
of the overall installation. It must be designed so as to supply the
appropriate mixture of the recirculated flue gas and the combustion air to
each burner. Thus, in the case of the larger two heaters, the cost of the
burner plenum was presumed to be equal to the cost of the ductwork plus the
cost of ductwork hangers and installation as a first approximation.
The cost of dampers to control recirculated flue gas flow was
assumed to be small and rather independent of heater size up to 73.3 MW
(250x10 Btu/h). A one-day installation time by two men at $150/man/day
was assumed along with $200 in material costs for the smaller two heaters.
These values were doubled to arrive at the $1,000 value shown in Table 4-7
for the 147 MW heater.
Cost data on duct insulation was obtained from Reference ID (pp. 155-6)
2 2
At $3.25/ft , the insulation cost for the 600 ft of ductwork required for
the 73.3 MW (250x10 Btu/h) heater was $1,950 in 1968 which, when appreciated
at 7%/year for 10 years, is equivalent to approximately $4,000 in 1978
dollars. Labor costs, including a factor for fair working conditions, were
approximately $500 in 1978 dollars for the installation of this same amount
of ductwork.
The basic cost of the insulation material was doubled in order to
include the costs of lagging, studs, and the cost of insulating the plenum
not included in the $4,000 figure. Thus, when added to the labor cost a
total cost for all insulation material and installation of $8,500 was
obtained for the 73.3 MW heater. Since the ductwork area to be insulated
increases linearly with heater size for the FGR systems envisioned here,
the cost of the insulation for the large 147 MW heater was assumed to be
roughly double that of the 73.3 MW heater with some allowance for economy
of scale. A figure of $2,000 was used for the 2.9 MW heater because of the
relative ease of installation and the disproportionately small amount of
insulation required for the burner plenum.
4_86 KVB 6015-798
-------
Basic controls required for the simplest FGR system are a recirculated
gas temperature, indicator, flow indicator, and damper and fan controls
(including alarms and an automatic safety shut-off switch). The installed
cost of the temperature and flow indicators in 1961 dollars according to
Reference 10 (p. 152) is $1,500 not including fired, variable, and semi-
variable costs as defined in that reference. To include those costs the
figure of $1,500 was doubled as a first approximation. In 1978 dollars,
assuming an average rate of inflation of 6% over the 17-year period, this
sum amounts to approximately $8,000. The cost of fan and damper controls
and alarms is estimated by KVB to be approximately $2,000, thus bringing
the total instrument and control costs to $10,000 for the smallest heater.
For the larger heaters, an additional $5,000 was assumed in order to account
for the added complexity of those systems.
The engineering/design cost figures were based roughly on the
following man-hour requirements estimated by KVB:
2.9 MW (IQxlO6 Btu/h) 73.3 MW (250xlQ6 Btu/h) 147 MW (SOOxlQ6 Btu/h)
50 m-h at $40/h 250 m-h at $40/h 500 m-h at $40/h
The total fixed initial costs for FGR do not include costs for
back-up fans or extra controls which might be required in certain applications.
The annual operating costs for FGR include only the fan motor
electrical requirement and maintenance since annual fuel costs or savings
are impossible to project on the basis of the data collected thus far. The
amount of electricity used to drive the fan, based on 75% electrical-to-
mechanical conversion efficiency and an 80% operating factor, as well as
the cost of electricity, based on a price of 4C/kW-h determined from
Reference 13, are tabulated below:
4-87 KVB 6015-798
-------
TABLE 4-9. ANNUAL FAN ELECTRICAL REQUIREMENTS IN 1000 kW-h AND COSTS
Flue Gas
Temperature
K (°F)
294
533
810
(70)
(500)
(1000)
Recirculated Flue Gas Flow
m /s (SCFM)
470 (1000) 9400 (20000) 18800 (40000)
3.50 ($140)
9.81 ($392)
56 ($2293)
70.1 ($2803)
175 ($7008)
470 ($18781)
210
420
701
($8410)
($16819)
($28032)
FGR costs in Table 4-6 are based on a 533 K (500 °F) flue gas stream.
Thus, $400, $7000, and $17000 are used as annual fan electrical requirements
for the small, medium, and large heaters, respectively. The values in Table 4-6
also include annual maintenance costs of 10% of the initial fixed capital costs
(in Table 4-5).
3. Staged Combustion Air - Floor LancesInitial costs for SCA-L were
determined using data from Reference 10 (pp. 163-173) and some of the cost
data developed for FGR.
The costs of piping, valves, fittings, and bends were computed
separately, and then the costs of staged air fans, dampers, controls, and
engineering/design were added. A 7% annual inflation rate was assumed to
convert 1968 dollars to 1978 dollars.
4_es KVB 6015-798
-------
The cost breakdown is shown below for the 2.9 MW (10x10 Btu/hr)
heater.
Component Material
12 m (40') of 5 cm (2")
ceramic pipe 240
(4 pipes at 10' each)
12 m (401) of 5 cm (2")
carbon steel pipe 24
6 m (20') of 10 cm
(4") carbon steel pipe 40
(4) 5 cm (2") control
valves 600
(1) 5 cm (2") cross 20
(4) 5 cm (2") 90° ells 12
(4) ceramic-steel
fittings (like raised
face flange) 14
Subtotal
Fan
Damper
Fan & Damper Controls
& Air Flow & Temperature
Measurements
Engineering/Design
Total
+ Installation = Total (1968.$)
1978 $
60
60
40
48
15
36
16
300
84
80
648
35
48
30_
1225
2500
500
500
5000
1000
$9500
4-89
KVB 6015-798
-------
The costs for the 73.3 MW (250x10 Btu/h) process heater size were
developed in a similar manner as follows:
Material + Installation = Total (1968 $) 1978 $
Component
61 m (200*) of 5 cm (2")
ceramic pipe 1200
(20 pipes at 10' each)
300
1500
61 m (2001) of 5 cm (2")
carbon steel pipe 120
15 m (50') of 23 cm (9")
carbon steel pipe 250
(20) 5 cm (2") control
valves 3000
(20) 5 cm (2") tees,
unequal 200
(20) 5 cm (2") 90° ells 60
(20) ceramic-steel
fittings (like raised
face flange) 70
Subtotal
Fan
Damper
Fan s Damper Controls
& Air Flow & Temperature
Measurements
Engineering/Design
Total
300
225
240
320
160
80
420
475
3240
520
220
150
7025
14000
2000
500
15000
5000
$36500
4-90
KVB 6015-798
-------
The costs for the 147 MW (500x10 Btu/h) heater were determined as
follows;
Component
Material + Installation = Total (1968 $)
91 m (300') of 7.6 cm (3")
ceramic pipe 3000
(30 pipes at 10" each)
600
3600
1978 $
91 m (3001) of 7.6 cm (3")
carbon steel pipe
30 m (1001) of 28 cm (11")
carbon steel pipe
(30) 7.6 cm (3")
control valves
(30) 7.6 cm (3")
tees, unequal
(30) 7.6 cm (3")
90° ells
(30) ceramic-steel
fittings (like raised
face flange)
Subtotal
Fan
Damper
Fan & Damper Controls
& Air Flow & Temperature
Measurement
Engineering/De s ign
Total
330
600
6300
500
150
600
500
600
100
330
930
1100
6900
600
480
135
180
315
15925
32000
3500
1000
15000
7500
$59000
4-91
KVB 6015-798
-------
The annual operating costs for SCA-L excluding fuel costs (savings)
are assumed to be negligible for the purposes of this cost estimate. Annual
fan electrical requirements and fan maintenance costs are not included in
Tables 4-4 and 4-6. The annual FGR fan electrical costs for 294 °K shown in
Table 4-9 may be used to estimate the additional costs. Thus, fan power
requirements and maintenance for the 2.9 MW, 73.3 MW, and 147 MW heaters
add 1.5%, 7.7%, and 14.3%, respectively, to the costs shown in Tables 4-4
and 4-6 for SCA-L.
It is possible to design an SCA-L modification which does not require
a fan but uses natural draft to introduce air into the firebox through the
lances. Naturally, such a system would be cheaper than the system considered
here, although some further testing would be required to optimize pipe sizes
and configurations.
4. Steam InjectionFor facilities already having steam piped to the
burners the initial fixed costs are assumed to be negligible. For plants in
which no steam capability currently exists the initial costs were determined
with the aid of Reference 10 (pp. 163-173) as follows:
2.9 MW (10x10 Btu/h) heater
Component
15 m (50') of 5 cm (2")
carbon steel pipe
(1) 5 cm (2") control
valve
(1) 5 cm (2") 90° ell
(1) 5 cm (2") tee,
unequal
Subtotal
(1) flowmeter (1961 $)
Total
Material + Installation = Total (1968 $) 1978 $
30
150
3
10
460
75
12
8
16
150
105
162
11
26
304 -» 700
610 (1961 $) -> 2300
$3000
4-92
KVB 6015-798
-------
73.3 MW (250x10 Btu/h) heater
Component
Material + Installation = Total (1968$) 1978 $
152 m (500') of 5 cm (2")
carbon steel pipe 300
(26) 5 cm (2") control
valves 3900
(25) 5 cm (2") 90° ells 75
(25) 5 cm (2") tees,
unequal 250
Subtotal
(1) flowmeter
Total
750
312
200
400
1050
4212
275
650
6187
14000
147 MW (500x10 Btu/h) heater
Component Material + Installation
229 m (7501) of 5 cm (2")
carbon steel pipe 450
(51) 5 cm (2") control
valves 7650
(50) 5 cm (2") 90° ells 150
(50) 5 cm (2") tees,
unequal 500
Subtotal
(1) flowmeter
Total
1125
612
400
800
= Total (1968 $) 1978 $
1575
8262
550
1300
11687
25500
2300
$27800
4-r93
KVB 6015-798
-------
The annual operating cost for the STM modification was assumed to be
equal to the cost of the steam used. Maintenance was assumed to be a minor
cost and, in the case of existing steam capability, zero cost.
The annual steam use was calculated for each heater size assuming an
80% use factor and a 0.01 kg/s (75 Ib/h) steam injection rate per burner.
The cost of steam was determined by adding together the cost of water and
the cost of converting it to steam (assuming a heat requirement of 2.79x10
J/kg steam (1200 Btu input/lb steam) . The cost of water used was a figure
obtained from the Los Angeles Department of Water and Power of 34.8 C/1000 ft .
As an example, the annual cost of steam for the 2.9 MW heater is calculated
below.
Annual Steam Consumption = 75 Ib/h x 1 burner x 0.80 x 8760 h/y
= 525,600 Ib steam/y
Annual Cost of Water Used = 525,600 Ib x - *"
- TT n ,^
oz.4 J-b HO 10U
= $29
Annual Cost of Energy = 525,600 Ib x 1200 Btu/lb x $2.00/106 Btu
to Generate Steam
$ 1262
Total Annual Steam Cost = $1262 + $29 = $1291 - $1300
The annual steam costs for the 73.3 MW (250x10 Btu/h) and 147 MW (500x10
Btu/h) were calculated in the same manner as $32,200 and $64,300, respectively.
5. Tertiary Air BurnerThe initial costs of a TAB were estimated on the
basis of information obtained from the burner manufacturer which designed
the TAB (Ref. 14). Using the highest figure given for the cost of each
tertiary air burner ($1500) and multiplying by the number of burners in
each heater, the initial installed cost of a TAB was determined. (Capital costs
make up about 2/3 of the investment; installation costs make up the balance.)
4-94 KVB 6015-798
-------
She annual maintenance costs of the tertiary air burner are not
expected to be significantly different from the maintenance costs associated
with conventional burners. No other annual costs were considered.
6. Altered Injection GeometryThis modification requires very little
initial investment or annual expenditure. Only the initial cost of making
the necessary minor burner adjustments is included in the present estimate.
7. Staged Combustion Air - Central CylinderThe initial installed costs
of SCA-C were calculated as shown in Table 4-10.
TABLE 4-10. INITIAL COSTS OF SCA-C MODIFICATION
Item
Cylinder material
Labor to fabricate
cylinders
Labor to install
(Includes shipping
and handling costs)
2.9 MW
200
500
(16 m-h at
$31.25/h)
300
(10 m-h at
$31.25/h)
1000
x 1.5
$1500
73.3 MW
3000
1500
(48 m-h at
$31.25/h)
1000
(32 m-h at
$31.25/h)
5500
x 1.5
$8250
147 MW
5000
2000
(64 m-h at
$31.25/h)
1500
(48 m-h at
$31.25/h)
8500
x 1.5
$12750
Annual operating costs are expected to be negligible for SCA-C.
C. Conclusions
The most cost-effective combustion modification for NO reduction in
X
natural draft process heaters appears to be staged combustion air. The central
cylinder technique is the least expensive type of staged air modification,
although the largest percent NO reduction was obtained using the floor lance
X
technique. Optimization of the central cylinder concept may further improve
its NO reduction potential, however.
4~95 KVB 6015-798
-------
FGR is an effective but more costly modification. TAB, AIG, and STM
are all moderately effective in reducing NO . STM costs were the highest of
X
any modification for large heater sizes. AIG in the present form applies only
to gas fired units, although the concept is adaptable to oil firing. TAB is
currently available and represents moderate NO reduction capability at
moderate cost.
4~96 KVB 6015-798
-------
SECTION 5.0
SUBSCALE TEST - ROTARY CEMENT KILN
5.1 INTRODUCTION
KVB completed a series of tests on a small pilot cement kiln. The
cement kiln, located at a major cement industry association facility, has a
13 cm (5 inch) ID, 30 cm (12 inch) OD, and is 4.6 m (15 feet) in length. The
maximum kiln feed rate is 0.0015 kg/s (12 Ib/h), and the unit has no air pre-
heat capability.
All tests were conducted with natural gas fuel. The objectives of the
tests were the following: to determine the effects of (1) sulfur addition either
with the fuel or with the feed, (2) water injection at the burner, and (3) kiln
dust injection at the burner, and (4) fly ash injection at the burner on gaseous
emissions, kiln operating conditions (temperature), and clinker quality.
Table 5-1 summarizes the effects of sulfur addition, water injection,
and fly ash injection on gaseous emissions and kiln operating temperatures.
The analysis of the clinker material from the kiln for each set of conditions
is being carried out by the cement association, and that information will be
supplied to KVB in a forthcoming report.
5.2 EMISSIONS SAMPLING
All emission measurements were taken from the center of the dustbox
(at the back end of the kiln upstream of the cyclone as illustrated in Figure
5-1). Flame zone temperature readings were taken with an optical pyrometer,
and the cyclone inlet temperature was measured with a thermocouple. Dustbox
excess oxygen measurements were verified using a portable oxygen analyzer.
The kiln feed used in the tests was pelletized from a difficult-to-
burn mix. This mix was high in limestone content and contained a relatively
large amount of binder material to lower the dust loading. The hard-burning
5-1 KVB 6015-798
-------
TABLE 5-1. SUMMARY OF GASEOUS EMISSION DATA - LOCATION 2, RESEARCH ROTARY CEMENT KILN]
01
to
Kiln
Test Date, Peed Rate
No. 1978 q/s (Ib/h)
2/3-1 8-17 1.06 (8.4)
2/3-2
2/3-3
2/3-4
2/3-5 i
1
0.78 (6.2)
+
2/3-6 8-18 0.93 (7.4)
2/3-7 \ if
2/4-1 8-18 0.45 (3.6)
2/4-2 1 J
2/4-3 0.76 (6.0)
2/4-4 1 +
2/5-1 8-18 0.76 (6.0)
2/5-2
2/5-3
2/5-4
2/5-5
2/6-1 8-21 0.44 (3.5)
2/6-2
2/6-3
2/6-4
2/6-5
2/6^6
2/6-7
2/6-8
0.61 (4.8)
1
1
1
0.76 (6.0)
1
1
t
2/7-1 8-21 0.76 (6.0)
2/7-2
2/7-3
2/7-4
2/7-5
2/7-6
2/7-7
Heat
Input Rate
kW(lo6Btu/h)
78.5 (0.268)
1
1
79.7 (0.272)
78.5 (0.268)
75.7 (0.258)
I
75.7 (0.258)
1
1
79.7 (0.272)
I
79.3 (0.271)
1
\
71.7 (0.245)
70.9 (0.242)
1
1
1
73.3 (0.250)
1
1
»
73.7 (0.252)
74.9 (0.256)
73.3 (0.250)
t
73.7 (0.252)
1
73.3 (0.250)
S'
0.20
0.15
0.10
0.20
0.40
1.8
2.0
2.1
3.75
3.1
2.4
2.55
2.05
2.2
2.05
1.7
1.3
1.6
1.5
1.55
0.25
0.10
0.15
0.30
0.4
0.3
0.3
1.5
1.5
1.8
1.6
C%*
13.4
12.4
12.4
11.9
11.9
12.0
11.5
9.9
9.4
10.2
10.4
10.2
10.6
10.6
10.6
12.0
11.4
11.2
11.6
11.2
12.0
12.2
12.4
12.8
13.2
13.2
12.8
11.8
11.7
11.7
11.6
NO
ppm*
64
1.0
VI. 0
0
3.1
66
58
63
55
52
58
51
58
54
63
77
66
67
78
36
17
44
76
103
91
119
82
73
71
99
*ng/J
33.
0.5
M).5
0
1.6
34
30
32
28
27
30
26
30
28
32
40
34
34
40
19
8.8
23
39
53
47
61
42
38
37
51
NO
ppm« ng/J
64
1.0
VI. 0
0
2.6
65
57
44
35
47
46
45
44
53
45
55
73
65
66
73
35
16
40
72
100
89
116
82
73
71
96
33
0.5
V0.5
0
1.3
33
29
23
18
24
24
23
23
27
23
28
38
33
34
38
18
8.2
21
37
51
46
60
42
3B
37
49
CO
ppm*
407
>1727
>1722
>1731
830
28
19
19
21
21
48
24
24
24
52
23
32
28
23
28
226
1068
1470
296
227
1077
198
2B
14fl
38
202
SO2
ppm*
36
35
^B60
685
350
23
11
0
VL25
17
485
66
22
25
27
12
0
0
0
0
19
20
24
0
22
0
11
0
0
0
0
HC
ppm*
31
22
52
77
85
40
T-104
134
153
104
88
88
99
23
13
11
9
8
26
18
37
21
14
12
21
13
10
13
%.
0
7.3
25
14
0
0
8.1
0
19
0
18
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
H20
-J-
0
0
0
0
0
0
0
0
0
0
0
0
13
24
59
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Kiln
Dust
InJ.
%5
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3.1
8.6
9.8
0
3.4
9.3
0
0
0
0
0
0
0
0
Fly
Ash
-I'
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2.2
6.6
0
0
2.4
7.3
0
Flame Zone
Temp.
K (°F)
1839(2850)
1805(2790)
1789(2760)
1797(2775)
1800(2780)
1761(2710)
1761(2710)
1755(2700)
1755(2700)
1722(2640)
1739(2670)
1733(2660)
1755(2700)
1758(2705)
1744(2680)
1766(2720)
1755(2700)
1733(2660)
1694(2590)
1678(2560)
1761(2710)
1766(2720)
1772(2730)
1800(2780)
1766(2720)
1783(2750)
1789(2760)
1789(2760)
1791(2765)
1755(2700)
1778(2740)
Cyclone
Inlet
Temp.
K (°F)
849(1068)
849(1068)
849(1068)
839(1050)
844(1060)
850(1070)
843(1058)
833(1040)
836(1045)
833(1040)
836(1045)
830(1035)
832(1038)
838(1048)
839(1050)
836(1045)
805 (990)
803 (985)
805 (990)
803 (985)
816(1010)
811(1000)
808 (995)
808 (995)
794 (970)
791 (965)
794 (970)
794 (970)
800 (980)
794 (970)
794 (970)
Comments 3
Baseline - LSF
Sulfur Injection-
LSF,
I
Sulfur burn-out -
LSF
Baseline - LSF
Sulfur Injection-
LSF
Baseline - HSF
Sulfur Injection-
HSF
Baseline - HSF
Sulfur Injection-
HSF
Baseline - HSF
Water Injection -
HSF
1
Baseline - HSF
Baseline - LSF
Kiln Dust Injec-
tion - LSF
1
1
Baseline - LSF
Kiln Dust Injec-
tion - LSF
*
Baseline - LSF
Fly Ash Injection
-LSF
Baseline - LSF
Baseline - LSF
Fly Ash Injection
- LSF
Baseline - LSF
Natural gas fuel used for all tests.
2Percent by mass of kiln feed rate
3LSF = Low-Sulfur Feed; HSF = High-Sulfur Feed
*dry, corrected to 3% O
-------
Exhaust
Ln
I
Gas
Burner
Ambient
Air
3 Rotating Kiln
Feed
Chute
Probe
\/
Dustbox
I.D. Fan
Feed
Hopper
Cyclone
3O
O
M
Ln
Figure 5-1. Schematic of subscale dry process rotary cement kiln (not equipped with air preheat).
id
00
-------
mix was selected so that flame zone temperatures would be abnormally high, thus
providing a worst case situation from the standpoint of NO emissions.
X
The fuel analysis for all tests is given in Table 5-2 below.
TABLE 5-2. NATURAL GAS FUEL ANALYSIS (TYPICAL)
Component
Nitrogen
Hydrogen
Carbon Dioxide
Methane
Ethane
Propane
Butane
High Heating value, dry,
J/m3 (Btu/CF)
Specific gravity
Volume
1.7
0.1
0.5
95.0
2.0
0.5
0.2
37.89xl06
%
(1017)
0.5816
The following sections discuss each of the combustion modifications
and the results obtained.
5.3 COMBUSTION MODIFICATION
5.3.1 Sulfur Addition
Sulfur was injected with the fuel at different rates for two different
feed sulfur contents. The sulfur was injected through a screw feeder and
blown in with air. The sulfur injection rate was determined after each test
by measuring the total mass of sulfur injected and the time taken to inject
it.
Under ordinary operating conditions the dustbox excess oxygen is
maintained at 1.0%-2.0%. At approximately this oxygen level the maximum NO
reductions were ^20% below a baseline value of 63 ppm (dry, corrected to
3% 02) with the higher sulfur feed and 12% below a baseline of 66 ppm (dry,
corrected to 3% O ) with the lower sulfur feed. The NO levels at this 0
z x 2
level did not appear to be affected by the change in feed sulfur content
although a greater proportion of the total NO occurred as NO with the
x 2
5-4 KVB 6015-798
-------
high-sulfur feed. (The actual feed sulfur content has not yet been made avail-
able to 3CVB.) The injection of sulfur produced significant increases in SO
emissions when the sulfur injection rate was greater than 10% of the kiln feed
rate.
At lower excess oxygen (< 0.4%) the NO dropped sharply on the low-
sulfur feed. A full 100% reduction in NO concentration was observed at
x
0.2% O on the low-sulfur feed. At the same time, SO emissions increased
^ *^
from a baseline level of 36 ppm (dry, at 3% 0 ) to 685 ppm (dry, at 3% O ).
The large decrease in NO emissions may be partially a result of oxygen
X
consumption by sulfur (to form S00). However, the decrease in NO may also
£ X
have been caused by a drop in excess oxygen which occurred during the test.
The excess O values reported in Table 5-1 for Tests 2/3-1 to 2/3-5 are
nominal values; the lag time in measuring the oxygen concentration may have
resulted in readings which did not match kiln conditions precisely. (In
tests subsequent to the sulfur injection tests it was determined that small
changes in oxygen concentration at low 0 [<0.5%] produced significant changes
in NO emissions. Special effort was made in those later tests to hold oxygen
levels constant.
At the low excess oxygen conditions with the low sulfur feed, CO
concentrations went off scale during sulfur injection, up from an initial
baseline at 0.20% O of 407 ppm (dry, corrected to 3% O ). At higher excess
oxygen conditions CO concentrations were generally < 30 ppm.
Figures 5-2 and 5-3 graph the relationship of NO emissions to
(1) dustbox excess oxygen and (2) sulfur injection rate for the two different
kiln feed sulfur contents. (Low-O_ conditions were not tested with the
high-sulfur feed because of a shortage of feed.) Figure 5-4 shows NO
x
emissions versus SO emissions. However, no direct relationship between the
two is implied by this graph.
5.3.2 Water Injection
Water was sprayed into the flame zone at three different flow rates for
one feed sulfur content and at approximately 2% excess oxygen. Water was
metered through a pipette and entered the kiln through a pipe next to the
burner pipe.
5-5 KVB 6015-798
-------
ou
(N
0 60
(0
% 40
ft
^
o* 20
E5
0
c
1
0(2/3-l)
"^ .
/
/
(2/3-3)
/ .12/3-2)
1 ft (2/3-4)
K*L^ '
)^ 0.8
Figure 5-2a. NO emis
research
80
1 1 1 1
(2/3-6)
/ O (2/3-7)
t
Fuel: Natural Gas
Firing Rate: ^ 80 kW
(0.27xl06 Btu/h)
Feed: Low-Sulfur
(Test Number)
till
1.6 2.4 3.2 4.0
DUSTBOX EXCESS OXYGEN, %, dry
sions as a function of dustbox excess oxygen for a
cement kiln with low-sulfur kiln feed.
1
1(2/3-6) 02
(N V
0 60
*
*
C 40
ft
0* 20
2:
C
1 1 1 I
= 1.8%
|"(2/3-l) """^O (2/3-7)0 = 2.0% ~
1 °2 = 0.20%
-\
\
- \
v (2/3-5) ^
) 0.05
(0.40)
Fuel: Natural Gas
Firing Rate: ^ 80 JcW
(0.273xl06 Btu/h)
Feed: Low-Sulfur
(Test Number)
0 = 0.15% O = 0.10%
(273-2) 02 = 0.20% .
>^Lnl2/3-^_ _L_ _ur>
0.10^ 0.15 0.20 0.25 0.30
(0.79) (1.19) (1.59) (1.98) (2.38)
Figure 5-2b.
SULFUR INJECTION RATE, g/s (Ib/h)
NO emissions as a function of sulfur injection rate for a
research cement kiln with low-sulfur kiln feed.
5-6
KVB 6015-798
-------
CN
O
60
ro
4J
(0
>i
0 40
B
a
a
o* 20
0
1 1 1
__ O
(2/4-4)
Fuel: Natural
Firing Rate:
(o
1 1
O
(2/4-3)
Gas
^ 75 kW
.26xl06 Btu/h)
Feed: High-Sulfur
(Test Number)
1 1 1
1 1
0.8 1.6 2.4 3.2 4.0
DUSTBOX EXCESS OXYGEN, %, dry
Figure 5-3a. NO emissions as a function of dustbox excess oxygen for a
research cement kiln with high-sulfur kiln feed.
80
60^
(N /
o v,
n
±>
a 40
d
i
a 20
"x
o
2
0
c
1
(2/4-1) 02 =
^(2/4-3)
0 = 3.1%
1
0.05
(0.40)
1 1
2.1%
O (2/4-4)
02 = 2
Fuel: Natural
1 1
.4%
Gas
Firing Rate: ^ 75 kW
(0.256xl06 Btu/h)
Feed: High-Sulfur
(Test Number)
I 1
0.10 0.15
(0.79) (1.19)
1 I
0.20 0.25 0.30
(1.59) (1.98) (2.38)
SULFUR INJECTION RATE, g/s (Ib/h)
Figure 5-3b. NO emissions as a function of sulfur injection rate for a
research cement kiln with high-sulfur kiln feed.
5-7
KVB 6015-798
-------
Ul
00
<*>
n
M
O
a
i
80
(
60
40
20
0
C
1 1
^
(2/4-1) 02 = 2.1%
/n (2/3-6) 0 = 1.8%
SO (2/3-1) 02 = 0.20%
5(2/3-7) 0 = 2.0%
U (2/4-3) 02 = 3.1%
0 = 0.15%
(2/3-2)
n 1 I
) 100 200
1 1 1 1 1 1
Fuel: Natural Gas
Firing Rate : 'v- 80 kW
(0.27x10 Btu/h)
(Test Number)
o
(2/4-4)
0_ = 2.4%
2 _
O = 0.40%
2 O = 0.20% O = 0.10%
, o3"5! , , (2/3-4) <2/3-3>
I ° I I 1 oi in
300 400 500 600 ^00 800 900
SO , ppm, dry at 3% O
CTI
o
Figure 5-4.
NO emissions as a function of SO emissions for several sulfur addition rates
and dustbox excess oxygen conditions.
-J
ID
oo
-------
Small reductions in NO of 12-14% below baseline levels of 58-63 ppm
X
(dry, corrected to 3% O ) occurred during the water injection tests. The NO^
concentration did not appear to vary significantly with the water injection
rate. At the highest injection rate, however, the CO concentration was twice
the baseline value (52 ppm, dry, corrected to 3% 0 up from 24 ppm, dry,
corrected to 3% O ).
Figure 5-5 shows the relationship between NO emissions and water
X
injection rate at a nominal 0 level of 2% for the high-sulfur kiln feed.
5.3.3 Kiln Dust Injection
Kiln dust containing 6.76% sulfur (by weight) was injected at various
rates and at two different excess oxygen conditions while burning the low-
sulfur kiln feed. The injection technique was the same as that used for
sulfur addition.
At a baseline excess oxygen level of approximately 1.5% the maximum
NO reduction of 14% below the baseline of 77 ppm (dry, corrected to 3% 0_)
X £,
occurred with the lowest rate of the kiln dust injection (approximately 3%
of kiln feed rate). Increases in dust injection rate caused the NO to
X
increase rather than decrease. CO, SO , and hydrocarbon emissions were all
very low at this O_ level.
At the low excess oxygen conditions, the maximum reduction of NO
X
again occurred at the smallest kiln dust injection rate (again approximately
3% of kiln feed rate). This reduction, however, was accompanied by a slight
drop in excess oxygen similar to the drop which occurred during the sulfur
injection tests. Thus, changes in excess 0 may have been responsible,
at least in part, for the reduction in NO concentration observed at less
than 0.3% excess oxygen.
During the dust addition at the low 0 levels the CO concentration
rose to greater than 1000 ppm. SO and hydrocarbon concentrations were
low, however, although they were slightly higher than they had been at the
higher 0_ condition.
Figure 5-6 graphs NO emissions versus dustbox excess oxygen and
X
kiln dust injection rate.
5-9 KVB 6015-798
-------
70
4-1
re
50
40
30
10
(2/5-1)
= 2.55%
(2/5-2)
0 = 2.05%
0 = 2.05
Fuel: Natural Gas
Firing Rate: ^ 80 kW
(0.27xl06 Btu/h)
Feed: High-Sulfur
(Test Number)
1
1
1
1
0.10
(0.79)
0.20
(1.59)
0.30
(2.38)
0.40
(3.17)
0.50
(3.96)
WATER INJECTION RATE, g/S (lb/h)
Figure 5-5. NO emissions as a function of water injection rate for a
research cement kiln.
5-10
KVB 6015-798
-------
BO
fN
0 60
n
m
$ 40
1
ft
"x
§ 20
0
0"
(2/6-8)
«
^X"
(2/6-7)/
- /
/O (2/6-5)
O (2/6-6)
1
1 (2/6-1)1 O O'(2/6-4) 1
^^^ " "
^ (2/6-lPO (2/6-2)
_
Fuel: Natural Gas
Firing Rate: ^75 kW __
(0.26xl06 Btu/h)
Feed: Low-Sulfur
(Test Number)
1 1 1 1
0.4 0.8 1.2 1.6 2.0
DUSTBOX EXCESS OXYGEN, %, dry
2.4
Figure 5-6a.
NO emissions as a function of dustbox excess oxygen for a
research cement kiln
O- = 1.3%
on ^
(N
0 60
Of
ro
n)
0 40.
1
a
S 20
*z
0
C5 1-^5(2/6-3)0- = 1.
(2/6-2) *
" 02 = 1.6%
(2/6-5)
o2 = o.
-Vw-
v>-
25%
/
^^
?
/
(2/6-7) 0
Fuel:
Firing
(2/6-6)
o2 = o
0 0
(0
.10%
|
.04
.32)
0.
(0.
Feed:
i (Test
08 0
63) (0
1.55* 1
5%
2 = 0.15%
1
Natural Gas __
Rate: ^
(0.26xl06
75 kW
Btu/h)
Low- Sulfur
i Number) i
.12 0.
.95) (1.
1
16 0.20
27) (1.59)
Figure 5-6b.
KILN DUST INJECTION, g/s (Ib/h)
emissions as a function of kiln dust injection for a
5-11
NO
research cement kiln.
KVB 6015-798
-------
5.3.4 Fly Ash Injection
ash containing 0.16% sulfur by weight was injected at various
rates and at two different excess oxygen levels while firing the low-sulfur
feed. The injection method was that used for sulfur and kiln dust addition.
Figure 5-7 shows the effects on NO emissions of fly ash injection rate and
X
dust box eKcess oxygen.
At the baseline oxygen level of approximately 1.5% the maximum NOx
reduction of 28% below a baseline of 99 ppm (dry, corrected to 3% 0 ) occurred
at the maximum fly ash injection rate (approximately 7% of kiln feed rate) .
CO concentrations rose somewhat during fly ash injection to 100-200 ppm
from a baseline level of 28 ppm (dry, corrected to 3% 02) . Other emissions
were low.
At low excess oxygen conditions (approximately 0.3%) NO values
X
dropped a maximum of only 24% from a baseline level of 119 ppm (dry,
corrected to 3% 0_) . This reduction occurred at the greatest water injection
rate (again approximately 7% of kiln feed rate) . The CO concentration rose
to 1077 ppm (dry, corrected to 3% 0 ) from a baseline value of 198 ppm (dry,
corrected to 3% O ) . SO and hydrocarbon emissions were low.
Special effort was made during the fly ash injection tests to
maintain constant excess oxygen levels throughout and, especially, to prevent
the oxygen concentration from dropping below 0.3% at the low 0 condition.
The results showed that NO reduction potential may not be any greater at
X
very low O than it is at the baseline O level.
5.4 CONCLUSIONS
Operation of the cement kiln at very low excess oxygen levels
(below 0.5%) does not seem to be practical. Very low NO levels may be
attained", but the accompanying CO concentrations are high. In addition, when
special care was taken to hold the oxygen level constant, the results
indicated that a. modification applied at baseline O (approximately 1.5%)
has nearly the same effect on NO emissions when applied at low 0 conditions;
^ 2
5-12 KVB 6015-798
-------
140
(2/7-2)
02 = 0.4%
f*» *
4J 8°
*
O
1 6°
i*
40
20
0
(
72/7-4) ^* ^
0-1.5% O" ' -. 0
(2/7-5) V
02 = 1.5% (2/7-6)
Fuel: Natural Gas
_ Firing Rate: ^ 75 kW __
(0.26xl06 Btu/h)
Feed: Low- Sulfur
(Test Number)
1 1 1 1 1
D 0.01 0.02 0.03 0.04 0.05 0.
(0.08) (0.16) (0.24) (0.32) (0.40) (0.
06
48)
FLY ASH INJECTION RATE, g/s (Ib/h)
Figure 5-7. NO emissions as a function of fly ash injection rate at
baseline and low excess oxygen conditions.
5-13
KVB 6015-798
-------
She maximum practical NO reductions attained in the research kiln
are shown- in Table 5-3. These reductions all occurred at baseline oxygen
conditions. Sulfur, water,, and kiln dust injection seem to produce similar
results. Fly ash injection; produced the largest practical NO reduction.
X
TABLE 5-3. MAXIMUM PRACTICAL NOX REDUCTIONS FOR FOUR COMBUSTION
MODIFICATIONS TO A RESEARCH CEMENT KILN
Combustion Modification
Maximum NO Reduction (%)
Sulfur Injection
Water Injection
Kiln Dust Injection
Fly Ash Injection
12 - 20
14
14
28
5-14
KVB 6015-798
-------
SECTION 6.0
SUBSCALE TEST, STEEL FURNACE
This section deals with the emissions and efficiency testing of a
small research steel furnace with a maximum firing rate of 0.6 MW (2x10
Btu/hr) located at the test facility of a major manufacturer of steel furnace
burners. Both natural gas fuel and No. 2 oil were fired in a standard radial/
axial burner provided by the manufacturer. A schematic of the experimental
apparatus is presented in Figure 6-1.
6.1 TEST APPARATUS AND EMISSIONS SAMPLING
The test apparatus consisted of a burner firing into a test furnace
with provisions for flue gas, steam, and water to be introduced into the
burner flame (see Figure 6-1) .
The test furnace served as a combustion chamber with a residence
time of about two seconds when firing at 0.6 MW. The furnace can
operate at 1978K (3100°F) and is outfitted with numerous access ports for
visual observation and temperature measurement. The furnace temperature
was maintained at 1533K (2300°F) throughout the test by exposing more or
less of the water cooled probes to the furnace interior. This was done
to simulate desired conditions in an actual furnace.
The 0.6 MW (2x10 Btu/hr) burner of the axial-radial type was used to
simulate the commonly used 2.4 MW (8x10 Btu/hr) version. The burner can be
fired on natural gas. No. 2 fuel oil, or both simultaneously. When firing on
gas, the unused oil port was used for water injection. When firing on oil,
one of the unused gas ports was used for steam injection.
The introduction of flue gases into the burner flame was accomplished
by passing a portion of the furnace exhaust's flue gases by means of a blower
6-1
-------
a\
i
I
GENERAL ARRANGEMENT
Combustion Air
1IH&
Atomizing Air /Cs
Qy~
No. 2
Fuel Oil
Injector
Water
I
To Atmosphere
Flue Gas Recirculation
; 533K
(500°F)
Water Cooled Probes
I.
20 cm (8") Dia. Orifice
() Measurement
Burner
Gas
I I
|T
I i
i i
i i
i i
-I -L
I I
I I
i i
Test Furnace
Furnace
Temperature
U 1533K (2300°F)
G
Qr
Steam
(Furnace Interior: 0.6 m (21) I.D. x
6.1 m (201) length)
Ui
Figure 6-1. Subscale steel furnace test schematic.
CD
-------
through a stainless steel air-to-air cooler and combining this flow with the
combustion air flow. The temperature of the recycled flue gases was maintained
at ~533K (500°F) by adjusting the flow of cooling air through the cooler. Good
mixing of the combustion air and flue gases was assured by employing a diffuser
between the combustion air/flue gas plenum and the burner.
A 20-cm (8-inch) diameter orifice was added to the furnace exhaust
stack to stop ambient air entfainment in the exiting flue gases resulting
in inaccurate flue gas O readings and place the furnace and the flue gas
recirculation (FGR) ductwork under positive pressure to reduce the infiltration
of ambient air through the cooler and blower.
All gas flows (combustion air, FGR, steam, atomizing air, and natural
gas) were measured with the aid of orifices and manometers and are considered
to be accurate to within 5 percent. All liquid flows (No. 2 fuel oil and
water) were measured with rotameters which had been calibrated with the fluid
to be measured.
The gas samples were taken from the exhaust stack with an aspirated
stainless steel tube. The flue gas samples were pumped to continuous gas
analyzers housed in the KVB mobile laboratory.
The installation of an 0 analyzer in the FGR ductwork just upstream
of the combustion air plenum became necessary to determine the degree of
flue gas dilution with infiltrating ambient air.
Temperature measurements of all flows, including the flue gas temperature,
were made using type "K" (61K to 1589K) thermocouples. Flame temperature profiles
were obtained using a type "R" (256K to 2033K) aspirated thermocouple.
Gaseous emissions were measured at baseline conditions and at various
excess air settings at full capacity and half capacity (nominally). In addition
to excess air variation, steam injection and water injection were tried at full
capacity firing No. 2 oil and natural gas, respectively, to reduce NO emissions.
Flue gas recirculation was also tested firing each of the two fuels. The FGR
rates reported in this section were calculated according to the expression
used in Section 4, page 4-54 for process" heaters.
6-3 KVB 6015-798
-------
Flame temperature profile measurements were made for each of the
modified conditions as well as for baseline conditions firing both natural
gas and No. 2 oil. All the data obtained from the subscale steel furnace
tests are presented in Appendix B.
6.2 COMBUSTION MODIFICATIONS
The overall results of the combustion modification tests are most
encouraging from the standpoint of NO emission reduction potential. The
maximum NO reductions obtained for each modification are summarized in
Table 6-1. The average baseline NO emission for a steel furnace burner
X
firing natural gas and No. 2 oil is given in Table 6-2.
Figures 6-2 and 6-3 show the effect of excess oxygen on NO emissions
when firing natural gas and No. 2 oil. For both fuels NO emission peaks at
about 2 percent O . As the 0 was increased beyond 2 percent, the NO con-
£ £
centration tended to decrease. The NO concentration also decreased at excess
0? levels below 2 percent, but the trend is less pronounced. (The high furnace
temperatures which occurred at low excess 0 conditions on several occasions
caused NO emission at these conditions to be higher than it would have been
if the temperatures had been held constant.)
There is an apparent considerable spread in the data for NO emission
versus stack excess oxygen. Figures 6-2 and 6-3 suggest a family of curves
for NO vs. O_. This indicates that another important factor is involved
in determining NO levels. It is believed that this factor in the combustion
air humidity and that each curve in the "family" of curves represents a
constant combustion air moisture content. Unfortunately, precise moisture
data was unavailable at the test site. Dry bulb and relative humidity
data was obtained from a weather station approximately ten miles from the
test site. This data was used in the construction of Figures 6-4 and 6-5.
The moisture in the combustion air was added to the HO injected through
the burner in these figures.
6-4 KVB 6015-798
-------
TABLE 6-1
SUMMARY OF SIGNIFICANT TEST RESULTS,
SUBSCALE STEEL FURNACE BURNER
Test Combustion
Number Fuel Modification
4/3-11 NG Water Injection
4/4-13 NG FGR
4/3-12 NG FGR + Water Inj .
4/7-2 No. 2 Steam Injection
4/8-10 No. 2 FGR
Firing % Reduction in NO
Rate O2 NO From
(% Cap.) % (ppm) * Nearest Baseline
100 2.2 98
100 2.0 38
100 1.8 24
100 2.1 24
100 2.0 57
47
88
87
89
77
*NO corrected to 3% 0 , dry
TABLE 6-2. AVERAGE BASELINE NO EMISSION,
X
SUBSCALE STEEL FURNACE BURNER
(Including All Baseline Tests at Location 4)
Fuel
NG
No. 2
ppm*
222
277
NO
ng/J
114.6
153.4
Number
of Tests
11
8
a.
Coefficient
of Variation
0.19
0.23
*ppm corrected to 3% O , dry
tCoefficient of variation =
Std. deviation
Mean
6-5
KVB 6015-798
-------
300
o
dP
4J
re
TD
i
250
200
150
100
50
T
(CO=140ppm)
4/4-21
(C0=312ppm)
_ 4/2-4
4/10-3
(CO=131ppm)
4/4-23
4/2-3
4/2-1
4/4-22
4/2-2
Fuel: Natural Gas
T = 1533K ± 89K (2300°F ± 160°F)
O Firing Rate = 0.59 MW (2.0 x 10 Btu/h)
Firing Rate = 0.29 MW (1.0 x 10 Btu/h)
Stack Excess Oxygen, %, dry
Figure 6-2. NO emission as a function of stack excess oxygen for a
subscale steel furnace firing natural gas.
6-6
KVB 6015-798
-------
400
350
300
250
<*»
n
T3
200
150
100
50
4/14-3
(C0=362ppm)
4/13-2
4/5-1
4/14-1
13-1
4/6-4
NOTE: For tests 4/6-1 to 4/6-6
T rroMTv =1460±35°K
FURNACE(2168±63oF)
O Firing Rate = 0.55 W (1.9 x 106 Btu/h)
iJ Firing Rate = 0.30 MW (1.0 x 10 Btu/h)
Stack Excess Oxygen, %, dry
Figure 6-3. NO emission as a function of stack excess oxygen
for a subscale steel furnace firing No. 2 oil.
6-7
KVB 6015-798
-------
300
250
CM
0 200
d*>
m
i-i
10 150
50
0
.100
nn V
O
I
I
Fuel: KG
Firing Rate:
2% 0,,
2x10 Btu/hr
(0.59 MW)
TFURNACE" 2300°F * 5°°
FURNACE (1533K ± 28K)
o
I
I
I
.200 .300 .400 .500 .600
Water Mass Flow Rate/Fuel Mass Flow Rate
.700
Figure 6-4. NO emission as a function of water injection rate for
subscale steel furnace firing natural gas.
6-8
KVB 6015-798
-------
tM
dP
m
I
D,
400
350
300
250
200
150
100
50
i
No. 2 Oil
1.8x10 Btu/hr (0.55 MW)
;
1
I
I
I
I
100 .200 .300 .400 .500 .600
Steam Mass Flow Rate/Fuel Mass Flow Rate
.700
Figure 6-5. NO emission as a function of steam injection rate
for a subscale steel furnace firing No. 2 oil.
6-9
KVB 6015-798
-------
Figures 6-4 and 6-5 reveal the sensitivity of NO to change in E^O
injection rate, particularly when firing No. 2 oil. The maximum percent
NO reduction obtained by injecting water with natural gas was 47 percent
as compared with 89 percent reduction of NO obtained by injecting steam
with No. 2 oil. It was not practical to try steam injection with natural
gas or water injection with No. 2 oil with this particular burner design.
Flue gas recirculation resulted in large NO reductions for both
natural gas and No. 2 oil fuels (see Figures 6-6 and 6-7). The greatest
decrease in NO using the FGR technique was observed when firing natural
gas (88 percent reduction).
6.3 COST ANALYSIS OF COMBUSTION MODIFICATIONS
6.3.1 Initial Capital Costs
A. Capital Costs of Steam and Water Injection for Steel Furnaces
For a plant which has steam generating capability but no steam piping
to the furnace to which the steam injection modification is to be applied,
the capital costs were determined in Section 4.1. For three heater sizes
those costs are shown below in 1980 dollars:
2.9 MW 73.3 MW 147 MW
(lOxlO6 Btu/hr) (250xl06 Btu/hr) (SOOxlO6 Btu/hr)
$3,500 $19,000 $32,000
Although these costs were developed for a process heater modification
it is not expected that they will differ substantially for a steel furnace
modification. They involve only straightforward piping changes to get the
steam from existing headers to the furnace itself.
B. Flue Gas Recirculation Capital Costs for Steel Furnaces
"The capital costs determined for flue gas recirculation systems for
process heaters are used here to estimate the costs of an FGR system for
steel furnaces. There are only two substantial differences between an FGR
system for a refinery process heater and a steel furnace system:
1. A heat exchanger may be needed in a steel furnace application in
order to cool the flue gases to a temperature which can be sent
6-!° KVB 6015-798
-------
250
200 _
<*>
n
150
O
2
100
50
2.0%, Water
Injection
4.8 gph (5.0 g/1)
Fuel: NG
Firing Rate: 2x10
Btu/hr. (0.59 MW)
5 10 15
% Flue Gas Recirculated
Figure 6-6. NO emission as a function of percent flue gas
recirculated for a subscale steel furnace
firing natural gas.
6-11
KVB 6015-798
-------
300
250
\ ' '
1
No. 2 Oil;
O 4%02
D2%°2
1.9xl06Btu/hr
(0.55 MW)
200 _
df
n
150
100 _
10 15 20
% Flue Gas Recirculated
25
Figure 6-7. NO emission as a function of percent flue gas
recirculated for a subscale steel furnace firing
No. 2 oil.
6-12
KVB 6015-798
-------
through the recirculating fan. This heat exchanger could act as
joegenerator, thereby increasing the efficiency of the unit and
offsetting its cost at least in part.
2. A burner plenum would not be required in a forced-draft steel
furnace whereas it was required in a natural draft process heater
for which there was not existing forced air injection capability.
Although the costs previously developed for FGR on a process heater
were based on 30% FGR, those 1978 costs will probably not differ significantly
from the costs of a 20% FGR system built in 1980. Thus, for the purposes of
the present study, those cost figures will be used without escalation to
represent the costs of a 20% FGR system installed on a steel furnace. The
initial installed costs were shown in Table 4-7.
6.3.2 Annual Operating Costs
A. Annual Operating Costs for Steam and Water Injection in Steel Furnaces
The total annual steam cost for process heater modfiication at 0.01 kg/s
(75#/hr) = $l,300/burner (mid-1978 $). In that calculation the steam cost
varies with the steam injection rate (m=lb/hr) as follows:
Total annual steam cost = Nm (C + C + C ) x P.I.
j- *L j
where N is the number of burners, P.I. is the price index and C.. , C , C_ are
constants relating to 1978 water supply costs, steam generation costs, and
water treatment costs, respectively. For an 80% use factor (i.e., assuming
an average steam mass flow rate of 80% over a single year) and an 80%
efficiency for the conversion of heat input to steam, C.. = 0.391; for NG,
C2 = 18.501; for No. 2 oil, C2 = 38.263; for No. 6 oil, C2 = 32.797; and
C3 = 2.383. These numbers will give costs in 1978 dollars for a furnace
containing N burners. To get 1980 dollars one must multiply by the price
index relative to mid-1978. The Chemical Engineering Plant Cost Index
(January 14, 1980) is 1.15. This value is used here as a suitable price
index.
Thus, for m in Ib/hr,
Total Annual Steam Cost = 21.275 x m N (NG)
= 41.037 x mSTMN (No. 2)
= 35.571 x m N (No. 6)
6-13 KVB 6015-798
-------
This relationship is shown in Figure 6-8. These costs do not include
maintenance costs since those are expected to be a small fraction of the
total annual operating costs. If mSTM = Kg/s, multiply the above equations
by 7936.6.
Similarly, the total annual water cost to calculated below. (Since
ordinary domestic water could be used in a water injection system, water
treatment costs are not included here.)
Total Annual Water Cost =NmH Q x C^ x P.I. = 0.450 NmH Q (mH 0=lb/hr)
These figures do not include maintenance costs. The total actual water cost,
including the cost of maintenance, may be somewhat greater than the cost
calculated by this equation. To allow for this the values plotted in
Figure 6-9 are increased by 20% over the values determined from the
above relationship.
Additional annual costs in the form of increased fuel requirements
brought about by steam and water injection must also be considered. The
additional fuel requirement is calculated in Appendix A for a subscale
steel furnace with a maximum firing rate of 0.59 MW (2.0 x 10 Btu/hr)
and 0.005 Kg/s (40 Ib/hr) steam injection. The additional heat required is
directly proportional to the steam or water flow rate. The relationship
is given below:
Ahs " NCSTM
w ~ NCWATER
Where CWATER = 2'387 Btu/lb and CSTM = 1275 Btu/lb, N is the number of burners
in the furnace, and Ah is the incremental heat input requirement in units of
Btu/hr. The cost increase on an annual basis may be determined, assuming
an 80% use factor, as follows:
Increase in Total Annual Fuel Cost = Ah x - ^^ - x 8760 -x 0.80
Unit Heat Input y
The cost per unit heat input for typical natural gas fuel is $2.20/106 Btu
(Ref. 15), the cost for No. 2 oil is $4.55/106 Btu, and the cost of No. 6
oil is $3.90/106 Btu (Ref. 16). The increase in total annual fuel cost is
6~14 KVB 6015-798
-------
30,00
28,OOC
26,000
24,000
22,000
£ 20,000
J
§ 18,000
o
ch 16,000
ii
^
w 14,000
O
u
£ 12,000
3
2
Z
10,000
8,000
6,000
4,000
2,000
1 I I
Natural Gas Fuel
No. 2 Oil Fuel
No. 6 Oil Fuel
N = Number of Burners
280,000
240,000
200,000
160,000
120,000
80,000
40,000
Figure 6-8.
STEAM MASS FLOW, g/s (Ib/hr)
Annual steam cost as a function of steam injected
per burner for different numbers of burners and
different fuels.
6-15
KVB 6015-798
-------
g
o
00
CTi
400
350
300
250
w 200
K
1
Hi 15°
100
50
N = Number of Burners
4000
3500
3000
2500
2000
1500
1000
500
WATER MASS FLOW, g/s (Ib/hr)
Figure 6-9. Annual water cost as a function of water injected
per burner for different numbers of burners, N.
6-16
KVB 6015-798
-------
graphed in Figures 6-10, 11, and 12 as a function of steam or water flow
rates for different numbers of burners. The calculation of Ah is explained
in Appendix C.
The total annual costs of steam and water injection, including water
cost, steam generation cost, additional fuel requirement cost, and maintenance,
are shown in Table 6-3 for three heater sizes using 0.005 Kg/s/burner (40 lb/
hr/burner) injection rate. (In order to use Figures 6-10, 11, and 12 each
burner is considered to have an average heat input capability of 2.9 MW
(10x10 Btu/hr) ) . One observes that the annual operating costs of steam
and water injection are, for all practical purposes, equal. Thus, the
average of the total annual costs of steam and water injection is used here
for costing both modifications.
B. FGR Annual Operating Costs
1. Electrical Costs A cost analysis of FGR annual operating costs for a
process heater has been presented in Section 4. The electrical cost of fan
operation is one of the chief components of the annual operating costs of a
flue gas recirculation system. Incremental fuel costs and maintenance costs
make up the bulk of the remaining annual costs. In the special case of steel
furnaces, heat exchanger maintenance costs would be added to those used
in applications involving process heaters since the flue gases used would
be much hotter (1366K or 2000°F) . The annual maintenance costs are estimated
to be -10% of initial fixed capital costs.
The additional fuel costs resulting from the use of FGR are determined
for 20% FGR and 2% excess O in the stack. We emphasize that efficiencies
calculated here assume that the flue gas temperature change from combustion
zone to reinjection point is all due to heat loss to the external environment;
i.e. , there is no regenerative capability of the FGR system.
The additional fuel costs are directly proportional to the mass flow
rate of the recirculated flue gas. The relationship used to calculate those
costs is the following:
A Cost = £!! x 8760 x 0.8 x
. .
R Unit Heat Input
6-17 KVB 6015-798
-------
60,000
50,000
w
K
3
tJ
§ 40,000
o
oo
o^
w
8
,j
w
fe
)J
<
30,000
20,000
10,000
Steam Injection
Water Injection
FUEL: No. 2 Oil
N = Number of burners
600,000
500,000
400,000
300,000
200,000
100,000
2.5 5.0 7.6 10.1 12.6
(20) (40) (60) (80) (100)
STEAM OR WATER MASS FLOW, g/s (Ib/hr)
Figure 6-10. Annual additional fuel requirement cost with
steam or water in a steel furnace firing No.
2 oil.
6-18
KVB 6015-798
-------
30,000
Steam Injection
Water Injection
FUEL: Natural Gas
N = Number of burners
12.6
(100)
STEAM OR WATER MASS FLOW, g/s (Ib/hr)
Figure 6-11. Annual additional fuel requirement cost with
steam or water injection in a steel furnace
firing natural gas.
300,000
250,000
200,000
150,000
100,000
50,000
6-19
KVB 6015-798
-------
K
§
§
O
CD
O1
EH
w
8
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
Steam Injection
~"^~~~"^~ Water Injection
FUEL: No. 6 Oil
N = Number of burners
400,000
350,000
300,000
250,000
200,000
150,000
100,000
50,000
12.6
(100)
STEAM OR WATER MASS FLOW, g/s (Ib/hr)
Figure 6-12.
Annual additional fuel cost with steam
or water injection in a steel furnace
firing No. 6 oil.
6-20
KVB 6015-798
-------
TABLE 6-3. TOTAL ANNUAL COSTS OF STEAM AND WATER INJECTION
I
CO
Costs
Water
Additional Fuel
Total
Steam
Additional Fuel
Total
Average
2.9 MW (lOxlO6
No. 2 Oil
$ 22
3,000
$3,022
$1,700
1,500
$3,200
$3,100
Btu/hr)
NG
$ 22
1,400
$1,422
$1,000
750
$1,750
$1,600
73.3 MW (250x10
No. 2 Oil
$ 600
75,000
$81,000
$41,000
40,000
$81,000
$81,000
6 Btu/hr)
NG
$ 600
37,000
$37,600
$21,000
20 , 000
$41,000
$39,300
147 MW (500 x
No. 2 Oil
$ 1,100
155,000
$156,100
$ 83,000
80,000
$163,000
$159,500
io6 Btu/hr j
NG
$ 1,100
73,000
$84,000
$43,000
39,000
$82,000
$83,000
CTl
o
Ul
VD
00
-------
Ah is calculated in Appendix C for these typical test conditions:
FGR
flue gas temperature of 558K (546°F) , flue gas density of 0.673 Kg/m (0.042
Ibm/ft ), and specific heat of 1.089 kJ/kg-°C (0.26 Btu/lbm-°F) . At these
conditions, the above expression may be written in terms of nu^r as follows:
A Cost = 560.04 x m___ x 8760 x 0.8 x .^ _ - -
FGR Unit Heat Input
where nu, = Ib/hr of flue gas. If the percent FGR and the fuel and combustion
and mass flows are known, one can use the following expression to convert
percent FGR to n
_ $ FGR .
"'FGR 100-% FGR FUEL mcoMBusTiON AIR
At 20% FGR this expression reduces to
mFGR ~ " (mFUEL + mCOMBUSTION AIR
where mFUEL and mCOMBUSTION MR are the mass flows of fuel and combustion air
for a given size heater. Using the cost per 10 Btu of each of the three
fuels natural gas, No. 2 oil, and No. 6 oil the incremental annual fuel
costs of 20% FGR, valid at the flue gas conditions described, are given as
follows:
>. 2 Oil)
3.827 m_-__ _ . (No. 6 Oil)
FUEL + C.A.
Note that 1250 x * E + c A = Thermal Heat Input (Btu/hr). This is a close
approximation for all three fuels and is used in conjunction with the three
preceding expressions in dveloping Figure 6-13.
The cost of electricity to operate the FGR fan may be estimated from
the data in Table 4-9. The costs of electricity are graphed as a function of
furnace size (m^^ + mCOMBUSTION AJR) for 20% FGR in Figure 6^14.
6-22 KVB 6015-798
-------
10,000,000
8
Q
o
CO
01
8
ij
o
H
EH
H
D
D
I
Z
1,000,000
100,000
10,000
0.29
(1)
20% Flue Gas Recirculation
I I I I I I I
2.9
(10)
29
(100)
290
(1,000)
FURNACE HEAT INPUT, MW (10 Btu/hr)
Figure 6-13. Annual additional fuel cost of flue gas recirculation as a
function of furnace size in a steel furnace for three
different fuels.
6-23
KVB 6015-798
-------
100,000
w
g
Q
o
CO
2
O
H
H
a
D
U
K
K
O
O
w
EH
W
8
10,000
1,000
100
' I I I I I III
I I I I I I 111
I I
20% Flue Gas Recirculation
Minimum Cost
Minimum Cost
Minimum Cost
I I I I I I II I
I I I » I I II
I III I I
1,260
(1,000)
12,600
(10,000)
126,000
(100,000)
1,260,000
(1,000,000)
MASS FLOW RATE OF FUEL AND COMBUSTION AIR, g/s (Ib/hr)
Figure 6-14. Cost of electricity to operate the fan of a FGR system as a
function of furnace size and different flue gas temperatures.
6-24
KVB 6015-798
-------
TABLE 6-4. ANNUAL OPERATING COSTS OF 20 PERCENT FLUE GAS RECIROTATION
FOR A STEEL FURNACE FIRING NATURAL GAS OR NO. 2 OIL (1980 DOLLARS)
FLUE GAS TEMPERATURE = 533K (500°F)
Costs
2.9 MW (10x10 Btu/hr)
733 MW (250x10 Btu/hr)
147 MW (500x10 Btu/hr)
Additional Fuel (NG/No.2)
Fan Electricity
Maintenance
TOTAL (NG/No. 2)
17,100/35,380
400
1,850
19,350/37,630
428,000/885,000
3,600
5,400
437,000/894,000
856,000/1,770,000
7,700
9,400
873,100/1,787,100
NJ
01
en
o
M
Ul
-j
ID
00
-------
The total annual costs (not annualized) taking into account the cost
of additional fuel requirements, fan electrical costs, and maintenance cost
are given in Table 6-4 for three furnace sizes (m + mCOMBUSTION MR) for
natural gas or No. 2 oil firing.
Again, it is emphasized that the incremental fuel costs shown here were
determined for the worst case in which nearly all of the sensible heat of the
recirculated flue gas is lost to the furnace surroundings. This situation
would probably not prevail in a practical, full-size steel furnace. However,
it is impossible to predict, at this time, how much heat may be retained in
the furnace.
6.3.3 Total Annualized Costs
The initial fixed capital costs of combustion modifications are
annualized making the following assumptions:
1. Straight-line depreciation of capital assets over a 12-year life
span.
2. After-tax rate of return of 15 percent.
3. State and federal property taxes totalling 11 percent of the
initial capital cost.
4. Insurance charges of 0.5% of the initial capital cost.
5. Debt /equity ratio of 0 (100% equity) for financing of initial
fixed capital costs.
6. Annual income tax rate (state and federal) of 50 percent.
7. Annual investment tax credit of 10 percent (applies only to the
first year of operation) .
The annualized capital costs must then be added to the annual operating
costs to give the total annualized cost of combustion modifications.
A. Total Annualized Costs of Water or Steam Injection
The calculation of total annual expenses and total annualized cost
of the water or the steam injection modifications to a steel furnace are
shown on the next page for No. 2 oil firing and natural gas firing.
6-26 KVB 6015-798
-------
TOTAL ANiNOALIZED COSTS OF WATER OR STEAM INJECTION
Annual Operating $3,100/1,600 $81,000/39,300 $159,500/83,000
Cost (No. 2/Natural Gas)
State and Federal 385 2,090 3,520
Taxes (11% of IFC)
Insurance (0.5% of IFC) 18 95 160
Depreciation (Straight 290 1,585 2,667
Line over 12 years)
Total Annual Expenses $3,793 $84,770 $165,847
(No. 2)
Total Annual Expenses $2,293 $43,070 $ 89,347
(Natural Gas)
INITIAL FIXED COSTS (IFC) 3,500 19,000 32,000
(WATER OR STEAM)
ROR=i=15%,n=12
Capital Recovery
Factor=.1845=CR
Annual Income
Tax Rate=50%
Investment Tax
Credit=10%=i
(1st year only)
Total Annual
Capital Factor*
(ACF)=.2773
Annual Capital
Charge (=IFCxACF) 971 5,269 8,875
TOTAL ANNUALIZED COSTS (1980 DOLLARS)
No. 2 Oil 4,764 90,039 174,722
Natural Gas 3,264 48,339 98,222
1 1c
*ACF = CR + T (CR )-
n n
where CR = capital recovery factor =
and T = 1.0 (for debt/equity ratio of O)
6-27 KVB 6015-798
-------
B. Total Annualized Costs of Flue Gas Recirculation
The total annual expenses and total annualized costs of flue gas
recirculation on a steel furnace are determined below for No. 2 oil-firing
and natural gas firing.
TOTAL AHNUALIZEB COSTS OF FGR
Annual Operating Costs 37,630/19,350 894,000/437,000 1,787,100/873,100
(No. 2/MG) =
State and Federal Taxes 2,035 5,940 10,340
(11% of IFC)
Insurance 100 270 470
(0.5% of IFC)
Depreciation (Straight 1,540 4,500 7,830
Line over 12 Years)
TOTAL ANNUAL EXPENSES 41,305 904,710 2 ,,678,840
(No. 2)
TOTAL ANNUAL EXPENSES 23,025 447,710 891,740
(NG)
INITIAL FIXED COSTS 18,500 54,000 94,000
(ROR =i=15%,n=12
Capital Recovery
Factor=.1845
Annual Income Tax
Rate=t=50%
Investment Tax
credit=i =10%
Q
(1st year only)
Total Annual
Capital Factor
=.2773
Annual Capital
Charge 5,131 14,976 26,069
TOTAL ANNUALIZED COSTS (1980 DOLLARS)
No. 2 46,436 919,686 2,704,909
NG 28,156 462,686 917,809
6-28 KVB 6015-798
-------
The cost effectiveness of a combustion modification is defined as the
total airrraalized cost of the modification divided by the annual NO emission
X
reduction potential of the modification (in thousands of Kg). The annual
NO emission reduction potential for steam and water injection and for flue
X
gas recirculation firing No. 2 oil and natural gas is given in Table 6-5.
The equation at the bottom of the table was derived from the expression used
in Section 4, page 4-73 for NO reduction potential from process heaters.
X
The cost effectiveness of steel furnace combustion modifications
for two different fuels and three furnace sizes including the annual
incremental fuel costs is given in Table 6-6.
It is important to note that while the steel furnace cost effectiveness
values include annual fuel costs due to combustion modification, those values
calculated for the cost effectiveness of combustion modifications made on a
process heater do not. The annual fuel cost turns out to be the most
significant item in the cost effectiveness calculation for steel furnaces.
These costs were calculated for steel furnaces based on the annual incremental
fuel requirements of combustion modifications. Certain assumptions were
made in the calculation of those fuel requirements. They are explained
along with those calculations in Appendix C.
It was felt that similar incremental fuel calculations would not
be meaningful for process heaters because data concerning the effects of
combustion modifications on heater efficiency were inconclusive.
It should be observed that the calculation of the annual incremental
fuel cost of FGR in steel furnaces is a worst-case calculation. This is so
because one of the assumptions made in the calculation is that all of the
heat of the recirculated flue gas is lost to the surroundings between the
point of extraction near the flame zone and the point of reinjection into the
furnace. Most likely, in a practical application, some of that heat would
be used to preheat combustion air or in some other waste heat recovery
scheme. Thus, fuel costs of FGR could very well be considerably less than
reported here.
6-29 KVB 6015-798
-------
TABLE 6-5. BASELINE NO EMISSIONS FROM A STEEL FURNACE
x
Annual
Modification Fuel
T
Ul
o
en
o
Steam
Steam
Steam
Water
Water
Water
FGR
FGR
FGR
FGR
FGR
FGR
Annual
No. 2
No. 2
No. 2
NG
NG
NG
No. 2
No. 2
No. 2
NG
NG
NG
_ io3
Heat Input NO Concentration
MW ng/J
2.93
73.2
147
2.93
73.2
147
2.93
73.2
147
2.93
73.2
147
V ft MO
y IMW __ ^ /-i ^ r » .. \*r.i ..
y
153.
153.
153.
114.
114.
114.
153.
153.
153.
114.
114.
114.
ng
J
4
4
4
6
6
6
4
4
4
6
6
6
Annual Emission
103Kg NO
11.3
283
568
8.5
211
425
11.3
283
568
8.5
211
425
Reduction
Percent
89
89
89
47
47
47
77
77
77
88
88
88
Reduction
10 3 Kg NO
10.
252
506
4.
99
200
8.
218
437
7.
186
374
1
0
7
5
VD
00
-------
CTl
O
Ul
Modi fication
TABLE 6-6. COST EFFECTIVENESS OF COMBUSTION MODIFICATIONS
ON A STEEL FURNACE ($/103 Kg OF NO REDUCTION)
INCLUDING ANNUAL INCREMENTAL FUEL COSTS
2.9 MW (10x10 Btu/hr)
Furnace Heat Input
73.3 MW (250xl06 Btu/hr)
147 MW (500x10 Btu/hr)
STEAM INJECTION
No. 2 Oil
NG
472
323
357
192
345
194
en
i
WATER INJECTION
No. 2 Oil
NG
1,191
816
909
488
874
491
FLUE GAS RECIRCULATION
No. 2 Oil 5,337
NG 3,754
4,219
2,488
6,190
2,454
vo
03
-------
6.4 CONCLUSIONS
The results of the tests at the subscale steel furnace are summarized
below:
1. Large NO emission reductions were obtained when firing natural
gas and No. 2 oil by the method of HO injection and by the flue
gas recirculation technique.
2. Excess air variations did not affect NO emissions significantly
except at ;
operation.
except at a high O level, which is a less efficient mode of
3. From the standpoint of NO reduction capability, without regard to
efficiency considerations, the steam injection technique appeared
to give the best results when firing No. 2 oil, and FGR gave the
best results when firing natural gas.
4. Measured flame temperature profiles indicate that NO increases
directly with the temperature in the flame zone; average flame
zone temperatures ranged from 1367 to 1922K (2000 to 3000°F).
5. Final calculations of the relative cost of combustion modifications
indicate that steam or water injection offers the best NO removal
capability for the least cost.
6-32 KVB 6015-798
-------
SECTION 7.0
REFERENCES
1. Unpublished results from API NOx study at KVB
2. Schorr, J. R. et al., "Science Assessment: Glass Container
Manufacturing Plants, " EPA-600/2-76-269, October, 1976.
3. Ketels, P.A. et al., "Survey of Emissions Control and
Combustion Equipment in Industrial Process Heating," EPA 600/
7-76-022, October, 1976.
4. Hunter, S. C. et al., "Application of Combustion Modifications
to Industrial Combustion Equipment," KVB, Inc., presented to the
2nd Symposium on Stationary Source Combustion, August 29-Sept.
1, 1977-
5. Allen, K. C., Directory of Iron and Steel Works of the United
States and Canada, 33rd edition, American Iron and Steel
Institute, July, 1974
6. Private Communication with Max Hoetzl, Surface Combustion, Inc.,
November 17, 1977.
7. Private communication with Chuck Mellus, Surface Combustion, Inc.,
November 18, 1977.
8. Sittig, Marshall, Practical Techniques for Saving Energy in the
Chemical, Petroleum, and Metals Industries, Noyes Data Corp.,
Park Ridge, New Jersy, 1977.
9. National Emissions Data System, Emissions by SCC, Oct. 27, 1977,
provided by EPA, Nov. 1977.
10. Popper, Herbert, Modern Cost-Engineering Techniques, McGraw-Hill
Book Co., New York, 1970.
11. Private communication from Vern Sharpe, Sharpe Heating and Ventilating,
Alhambra, CA to R. J. Tidona (KVB), June 22, 1978..
12. Private communication from Industrial Gas Engineering, Westmont, IL,
to R. J. Tidona (KVB), June 22, 1978.
13. Typical Electric Bills_, 1977, Federal Power Commission, FPC R90.
14. Private communication from refinery heater burner manufacturer to
S. S. Cherry (KVB), March 21, 1978.
7-1 KVB 6015-798
-------
15. American Gas Association Quarterly Report of Gas Industry Operations,
American Gas Association, Second Quarter, 1979.
16. Energy User News, October 22, 1979, p. 15.
7-2 KVB 6015-798
-------
APPENDIX A
SUMMARY OF GASEOUS EMISSION DATA,
LOCATION 1,
PROCESS HEATER RESEARCH FURNACE
KVB 6015-798
-------
APPENDIX A
TABLE A-l. SUMMARY OF GASEOUS EMISSION DATA, LOCATION 1, PROCESS HEATER RESEARCH FURNACE
Test No.
1/1-1
1/1-2
1/1-3
1/1-4
1/1-5
1/1-6
1/1-7
1/1-8
1/1-9
1/1-10
1/1-11
1/1-12
1/1-13
1/1-14
1/1-15
1/1-16
1/1-17
1/1-18
1/1-19
1/1-20
Fuel
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
'No. 6
Date,
1978
1/12
1/13
1/13
1/13
1/13
1/13
1/13
1/13
1/16
1/16
1/17
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/19
1/19 '
Heat Input
Rate
HH (106Btu/h)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.38 (4.7)
1.40 (4.8)
1.37 (4.7)
1.35 (4.6)
1.38 (4.7)
1.37 (4.7)
1.39 (4.8)
1.36 (4.6)
1.43 (4.9)
1.35 <4.(.)
°2.
2.8
3.9
2.9
2.1
1.1
0.3
3.7
3.0
3.5
3.1
3.2
3.0
2.0
1.1
0.8
3.9
0.7
3.0
3.0
3.2
C02,
11.0
10.2
10.9
10.9
12.8
12.0
11.0
11.0
10.3
10.8
13.1
13.8
14.2
14.4
15.2
13.5
15.2
13.8
11.7
13.3
NOX
ppm« ng/J
114 58
111 57
108 55
105 54
93 47
83 42
118 60
113 58
118 60
120 61
321 180
311 174
312 175
297 167
284 159
306 172
280 157
285 160
284 159
2H4 15")
NO
ppm* ng/J
107 55
104 53
104 63
104 53
89 45
82 42
115 59
111 57
115 59
117 60
313 176
305 171
301 169
282 158
284 159
305 171
270 151
280 157
274 lr>4
277 I'j5
CO
ppm'
0
0
0
0
0
139
0
0
0
0
0
0
0
0
706
0
17-
2(>(>
0
0
c
S02
ppm*
0
0
0
0
0
0
0
0
0
0
1354
1310
1467
1352
1796
1284
If.M
1431
125(1
1490
Stack Temp.
K <°F)
1167 (1641)
1115 (1547)
1125 (1567)
1142 (1596)
1166 (1640)
1179 (1662)
1183 (1669)
1185 (1674)
1099 (1518)
1163 (1634)
1093 (1507)
1042 (1416)
1069 (1464)
1089 (1500)
1096 (1513)
1108 (1535)
1115 (1548)
1135 (1584)
1000 (1340)
1059 (1447)
Heater
Effi-
ciency
t
49.3
49.6
50.7
51.5
51.6
52.2
46.8
47.5
51.0
48.6
55.7
58.3
58.6
59.0
59.1
54.0
58.4
54.1
59.2
57.2
Smoke
Spot
3
3
4
5
7
2.5
6
2
2
2
*B
Staging
Height
m (ft)
__
__
--
__
__
FGR
%
~
Comments
Baseline, HA- 16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
O2 swing, MA-16 Brn
Pattern II
Baseline, MA-16 Brn
Pattern IV
Baseline, MA-16 Brn
Pattern II Modified
Baseline, MA-16 Brn
Tip 764
O2 swing, MA-16 Brn
Tip 764
02 swing, MA-16 Brn
Tip 764
O2 swing, MA-16 Brn
Tip 764
O2 swing, MA-16 Brn
Tip 764
O2 swing, MA-16 Brn
Tip 764
O2 swing, MA-16 Brn
Tip 764
Baseline, MA-16 Brn
Tip 763
Baseline, MA-16 Brn
Tip 763
O2 swing, MA-16 Brn
Tip 763
ID
CD
Corrrc-tiM Lo 3». O,, Dry
(continued)
-------
TABLE A-l (Continued).
Test No.
1/1-21
1/1-22
1/1-23
1/1-24
1/1-25
1/2-1
1/2-2
1/2-3
1/2-4
1/2-5
1/2-6
1/2-7
1/2-8
1/2-9
1/2-10
Fuel
No. 6
No. 6
No. 6
No. 6
No. 2
NG
NG
No. 2
l;
No. 2
No. 2
No. 2
No. 2
NG
NG
NG
Date,
1978
1/19
1/19
1/19
1/19
1/20
2/13
2/13
2/14
2/14
2/14
2/14
2/14
2/15
2/15
2/15
Heat Input
Rate
MW (106Btu/h)
1.46 (5.0)
1.44 (4.9)
1.38 (4.7)
1.45 (4.9)
1.41 (4.8)
1.43 (4.9)
1.43 (4.9)
1.49 (5.1)
1.43 (4.9)
1.35 (4.6)
1.32 (4.5)
0.94 (3.2)
1.47 (5.0)
1.47 (5.0)
1.47 (5.0)
\'
3.5
0.9
2.2
3.1
3.1
2.8
2.9
3.9
1.4
0.5
5.1
5.7
3.1
1.3
O.o
C02,
13.3
14.8
13.8
13.6
12.7
10.6
11.2
12.6
14.4
14.4
12.0
11.9
9.9
10.6
12.0
NOX
ppm* ns/J
282 158
229 128
273 153
289 162
112 63
129 65.8
146 74.5
110 61.7
98 55.0
100 56.1
119 66.8
85 47.7
104 53.0
100 51.0
83 42.3
NO
ppm* ng/J
276 155
222 125
268 150
283 15*
108 61
126 64.3
141 71.9
105 58.9
96 53.9
96 53.9
109 61.2
83 46.6
104 53.0
88 44.9
78 39.8
CO
ppm*
0
18
0
0
0
0
0
0
0
147
0
0
0
0
44
S02
ppm*
1417
1386
1354
1274
46
0
0
38
--
112
0
0
0
Stack Temp.
K CP)
1081 (1486)
1095 (1512)
1107 (1533)
1111 (1541)
958 (1264)
998 (1336)
1071 (1468)
1122 (1560)
1130 (1574)
1124 (1564)
1119 (1555)
1071 (1467)
1104 (1528)
1153 (1616)
1160 (1629)
Heater
Effi-
ciency
55.0
58.3
55.8
53.0
61.4
55.5
52.3
52,9
56.3
57.4
49.8
49.4
50.6
51.4
51.8
Smoke
Spot
2
4
2
2
0
0
0
1.5
1
2
0
4
0
0
0
^
--
Staging
Height
m (ft)
~
--
-_
_-
--
FGR
t
-
~
Comments
O swing, MA-16, tip 763
02 swing, MA-16, Tip 763
O swing, MA-16, Tip 763
O swing, MA-16, Tip 763
Baseline, MA-16 Brn
Tip 764
Baseline, Low-NOx
(Recirc. Tile) Brn -
Pattern III, (Gas
tips flush with top
of Piece A of burner
tile)
Baseline, Low-NOx
(Recirc. Tile) Brn -
Pattern III (Gas
tips 1/2" below top
of piece A of burner
tile)
Baseline, Low-NOx
(Recirc. Tile) Brn -
Tip 766
O2 swing, Low-NOx
(Recirc. Tile) Brn -
Tip 766
O2 swing, Low-NOx
(Recirc. Tile) Brn -
Tip 766
O2 swing, Low-NOx
(Recirc. Tile) Brn -
Tip 766
50* Capacity, Low
NOX (Recirc. Tile)
Burner , Tip 766
Baseline, Recirc.
Tile Burner, Pattern
I
O2 swing, Recirc.
Tile Burner, Pattern
I
O2 swing, Recirc.
Tile Burner, Pattern
I
tn
I
-j
<£
CD
(continued)
-------
TABLE A-l (Continued).
Test No.
1/2-11
1/2-12
1/2-13
1/2-14
1/2-15
1/3-1
1/3-2
1/3-3
1/3-4
1/3-5
1/3-6
1/3-7
1/3-8
Fuel
NG
NG
NG
NG
NG
NG
No. 6
No. 6
No. 6
No. 6
No. 6
NG
NG
Date,
1978
2/15
2/15
2/15
2/15
2/15
2/16
2/17
2/17
2/17
2/17
2/17
2/18
2/18
Heat Input
Kate
MW (106Ktu/h)
1.49 (5.1)
1.52 (5.2)
1.90 (6.5)
0.94 (3.2)
0.52 (1.8)
0.85 (2.9)
1.43 (4.9)
1.47 (5.0)
1.47 (5.0)
1.38 (4.7)
1.38 (4.7)
0.70 (2.4)
0.94 (3.2)
°2.
%
4.4
2.8
2.9
3.5
8.9
11.2
3.2
3.1
0.5
1.4
4.2
3.3
3.0
CO2,
%
10.0
10.6
10.4
10.4
6.8
5.7
13.8
13.6
15.3
14.4
13.2
10.2
10.7
NOX
ppm* ng/J
109 55.6
115 58.7
114 58.1
115 58.7
110 56.1
89 45.4
260 146
272 153
235 132
248 139
274 154
109 55.6
126 64.3
NO
ppm* nq/J
109 55.6
106 54.1
113 57.6
112 57.1
110 56.1
87 44.4
256 144
266 149
223 125
248 139
273 153
102 52.0
(110 60.2
CO
ppm*
0
0
0
0
0
0
0
0
37
0
0
0
0
S02
ppm*
0
0
0
0
0
0
1265
1468
1447
1519
1391
0
0
Stack Temp.
K (°F>
1159 (1626)
1163 (1633)
1191 (1684)
1122 (1559)
1053 (1436)
748 (886)
1111 (1539)
1171 (1647)
,
1178 (1661)
1168 (1643)
1168 (1542)
872 (1109)
963 U?73)
Heater
Effi-
ciency
%
46.0
48.6
48.2
47.1
33.3
51.2
54.1
51.6
55.5
54.3
49.5
56.3
54.9
Smoke
Spot
0
0
0
0
0
0
4
1
2.5
2
0.5
0
0
*B
~
Staging
Height
m (ft)
._
_.
..
FGR
%
Comments
O2 swing, Recirc.
Tile Burner, Pattern
I
Firing Rate Swing,
Recirc. Tile Burner,
Pattern I
Firing Rate Swing,
Recirc. Tile Burner,
Pattern I
Firing Rate Swing,
Recirc. Tile Burner,
Pattern I
Firing Rate Swing,
Recirc. Tile Burner,
Pattern I
Heat-up, Low-NOx
(Tertiary Air) Brn,
Pattern IIB (ports
at 15° and 45° off
radial)
Baseline, Low-NOx
(Tertiary Air) Brn,
Tip 864
O2 swing, Low-NOx
(Tertiary Air) Brn,
Tip 864
O2 swing, Low-NO,,
(Tertiary Air) Brn,
Tip 864
O2 swing, Low-NOx
(Tertiary Air) Brn,
Tip 864
O2 .swing, Low-NOx
(Tertiary Air) Brn,
Tip 864
35% Capacity,
Tertiary Air Burner
Pattern IIB (ports
at 15° and 45° off
radial)
50* Capacity,
Tertiary Air Burner
Pattern IIB
s
M
I/I
-O
U>
oo
Corrected to 3% o.)( Dry
(continued)
-------
TABLE A-l (Continued).
Tesl No.
1/3-9
1/3-10
1/3-11
1/3-12
1/3-13
1/3-14
1/315
I/ 3 16
1/3-17
1/3-18
1/3-19
1/3-20
Fuel
NG
NG
NG
NG
NG
NG
Shale
c
Shale
oil
Shale
Oil
Shale
Oil
Shale
Oil
Shale
Oil
Date,
197B
2/20
2/20
2/20
2/20
2/20
2/20
7/77
^/ t.4.
7/77
f./ ££,
7/77
£,/ ££,
7 /77
£./£.£
Heat input
R^ttT.
MW (lO-^tu/h),
1.55 (5.3)
1.55 (5.3)
1.55 (5.3)
1.55 (5..3)
1.55 {5.3)
1.90 (6.5)
1\C (A C\
* J3 (4 * ml
11 A t 1 Ql
. 1* I .? 3 J
1 74 f4 4)
i. , &j i^ . n i
1 7Q 14 4 1
1 . £^y \ *. * J
1 20 (4 1)
1 741 / A d't
J * 45 l^ . H 1
<-!
3.5
.4
o.»
2.1
4.0
3-2
Se
. ;>
3.2
2 0
0 3C
i 2
3£
.
C°2<
^; '
10.2
11.3
11. «
11.2
10.2
16.3
NOX
ppm? 'ng/J
1S3 7B.O
122 62.2
133 *7.«
14» 7*.0
1*1 12.1
155 79. 1
349 196
- , w .
ppin*.'. ng/il -
143 72.9
117 '/)./
124 63.2
141 71.9
149 76.0
14* 74.5
526 295
439 246
364 204
200 112
J
l nl
GO
'nan*
0 -
MI
0
0
0
0
0
0
Q
-.Spg
!'£&*'
^%r/?. ^
0
0
. 0
p
0
0
SJtack TejJ(p. '
' -'.;*|( -^ "f^F ) '
116? (1631)
12QXI (1700)
1200 (1701)
1200 (1700)
119,3 dill)
1232 (1751)
1099 (1519)
1 106 (1531)
1111 ( 1540)
1113 ( 1544 )
1117 /l^mi
lilt { IJDll
Heater
Effi-
ciency.
" IT '"i
47.6
50.9
50.2
41.6
45.5
45.1
Sritoke
Sp^Jt
0
0
0
0
0
0
*
*B
-
Staging
Height
m UO
-
--
-_
FGR
%
--
--
Co5iTven.ts.
Baseline (80* Cap. )
Pattern IIB
02 swing, Tertiary
Pattern III
02 swing, Tertiary
Air Burner,
Pattern IIB
02 swing. Tertiary
Air lurner.
Pattern 11»
02 swing, Tertiary
Air Burner,
Pattern IIB
100» Capacity,
Tertiary Air Irn,
Pattern III
High Q-y i Tertiary
Air Burner, Tip 764,
All Registers 100%
Open
Baseline 02 >
Tertiary Air Brn,
Tip 764, All
Registers 100% Open
Tertiary Air Brn,
Tip 764, All
Registers 50% Open
Low 02 , Tertiary Air
Burner, Tip 764
Op imum-Low 2'
Tip 764, PAR1 = 25%
Open, SAR2 = 38%
Open, TAR3 = 100%
Open
i 1-1
0sei i ne 02 i
Tip 764, PAR = Closed
SAR=25% Open, TAR =
100% Open
I
(Jl
o
M
Ul
oo
(continued)
-------
TABLE A-l (Continued).
Test No.
1/3-21
1/3-22
1/3-23
l/3-24a
l/3-24b
1/3-25
1/3-26
1/3-27
1/3-28
1/3-29
1/3-30
1/3-31
1/3-32
Fuel
NG
No. 6
No. 6
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
Date,
1978
2/22
2/22
2/22
2/23
2/23
2/23
2/23
2/23
2/23
2/23
2/23
2/23
2/23
Heat Input
Race
MM (106BT;u/h)
1.41 (4.8)
1.41 (4.8)
1.38 (4.7)
1.47 (5.0)
1.47 (5.0)
1.41 (4.8)
1.41 (4.8)
1.47 (5.0)
1.47 (5.0)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
1.49 (5.1)
°2.
%
3.4
3.0
3.2
3.1
3.7
3.2
2.9
3.0
3.2
3.2
2.9
3.4
3.5
CO2,
?
..
13.6
12.6
10.3
10.2
10.6
11.0
11.0
10.2
10.2
10.8
10.6
10.3
NOX
ppm* ng/i.1
..
260 146
194 109
79 40.3
98 50.0
85 43.4
84 42.8
88 44.9
94 47.9
95 48.5
86 43.9
89 45.4
96 49.0
NO
Dpm* ng/J
148 75.5
260 146
192 108
75 38.3
r
93 47.4
114 42. 11
80 40.8
83 42.3
92 4f>.'l
90 45.9
84 42.8
86 43.9
96 49.0
CO
pj'm*
0
0
0
0
0
II
0
0
)
0
0
0
0
S02
ppm*
0
1357
1122
0
0
0
0
0
0
0
0
0
0
Stack Temp.
K <°F)
1090 (1500)
1177 (1658)
1179 (1662)
1129 (1572)
1188 (1678)
1190 (U.H2)
1200 (1700)
1210 (1718)
1209 (1717)
1216 (1729)
1223 (1741)
1228 (1750)
1232 (1758)
Heater
Effi-
ciency
%
50.7
51.4
,
50.9
49.5
45.8
46.5
46.6
46.1
45.8
45.6
45.9
44.7
44.4
Smoke
Spot
_
0.5
1
0
0
0
0
0
0
0
0
0
0
*B
..
Staging
Height
m (ft)
._
__
_-
_-
-
FOR
%
Comments
Baseline, Tertiary
Air Burner, Pattern
IIB (ports at 15°
and 45° off radial)
Baseline, Tertiary
Air Burner, Tip 864
PAR = 1/8 open.
SAR = TAR - 100%
open. Tertiary Air
Burner, Tip 864
Baseline, Tertiary
Air Burner, Pattern
IIC (2 firing ports.
each 15° either side
of radial)
Baseline, Tertiary
Air Burner, Pattern
IIC (2 firing ports,
each 15° either side
of radial) , Higher
Firebox Temperature
Air Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
Air Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
Air -Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
Air Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
Air Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
Air Register Adjust-
Burner, Pattern IIC
Air Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
Air Register Adjust-
ments, Tertiary Air
Burner, Pattern IIC
>
cr>
ID
00
Corri'cti-'il t-u 3* O.,, Dry
(continued)
-------
TABLE A-l (Continued).
Test No.
1/3-33
1/3-34
1/3-35
1/3-36
1/3-37
1/3-38
1/3-39
1/3-40
1/3-41
1/3-42
1/3-43
Fu,el
NG
NG
NG
NG
NG
NG
No. 6
No. 6
No. 6
No. 6
No. 6
Date,
1978
2/23
2/23
2/23
2/23
2/23
2/23
2/24
2/24
2/24
2/24
2/24
Heat Input
H«te
W U06»tu/h)
1.52 (5.2)
1.49 (5.1)
1.49 (5.1)
1.49 (5.2)
1.49 (5.2)
1.49 (5.2)
1.41 (4.8)
1.38 (4.7)
1.38 (4.7)
1.32 (4.5)
1.35 (4.6)
»2-
%
2.7
2.1
4.1
3.2
3.»
2.2
3.1
3.3
2.9
2.7
2.6
C02,
t
10.6
10. t
9.8
10.6
10.3
11.0
13.6
14.0
14.0
14.0
14.0
NOX
ppm* ng/J
100 51.0
76 38.8
103 52.5
101 51.5
105 53.6
75 38.3
268 150
198 111
251 141
261 146
275 154
NO
ppm* ng/J
93 47.4
76 38.8
102 52.0
92 46.9
105 53.6
75 38.3
262 147
198 111
248 139
251 141
27r. 1S4
CO
Ppm*
0
47
0
0
0
57
0
0
0
0
11
SO2
ppm*
0
0
0
0
0
0
1246
1237
llf>6
1258
f)f,7
Stack Temp.
K <*F)
1236 (1765)
1240 (1772)
1241 (1773)
1245 (17«1)
1244 (1780)
1244 (1780)
1104 (1527)
1115 (1547)
1134 (1581)
1136 (15B5)
mn (15B9)
Heater
Effi-
ciency
%
45.8
46.8
42.9
44.5
43.2
46.4
54.5
53.6
53.4
53.5
53.7
Smoke
Spot
0
0
0
0
0
0
2
3
1
0.5
*.
Staging
Height
m (ft)
._
FGR
%
Comments
02 swing, Tertiary
Air Burner, Pattern
IIC, All Registers
100% Open
O2 swing. Tertiary
Air Burner, Pattern
IIC, All Registers
100% Open
Op swing. Tertiary
Air Burner, Pattern
IIC, All Registers
100% Open
O2 swing at Optimum
Reg. Setting,
Tertiary Air Burner,
Pattern IIC, PAR =
50% Open, SAR =
TAR » 100% Open
O2 swing at Optimum
Reg. Setting,
Tertiary Air Burner,
Pattern IIC, PAR =
50% Open, SAR =
TAR 100% Open
O2 swing at Optimum
Reg. Setting,
Tertiary Air Burner,
Pattern IIC, PAR »
50% Open, SAR =
TAR - 100% Open
Baseline, Tertiary
Air Burner, Tip 864
Air Register Adjust-
ments, Tertiary Air
Burner, Tip 864
Air Register Adjust-
ments, Tertiary Air
Burner, Tip 864
Air Register Adjust-
ments, Tertiary Air
lurner. Tip 864
Air Register Adjust-
ments, Tertiary Air
Burner, Tip 864
en
o
oo
(continued)
-------
TABLE A-l (Continued).
Test No.
1/3-44
1/3-45
1/4-1
1/4-2
1/4-3
1/4-4
1/4-5
1/5-1
1/5-2
1/5-3
l /; A
L/ D *l
1/6-1
1/6-2
1/6-3
1/6-4
1/6-5
1/6-6
1/6-7
1/6-B
1/6-9
1/6-10
1/6-11
1/6-12
Fuel
No. 6
No. 6
NG
NG
NG
NG
NG
wr1
Nti
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NO. 6
No. 6
No. 6
Date,
19VS
2/24
2/24
3/13
3/13
3/13
3/13
3/13
3/13
1/1 i
J/-IJ
1/11
J/l J
1/11
J/l J
3/14
3/14
3/14
3/14
3/14
3/14
3/14
3/14
3/14
3/15
3/15
3/16
Heat Input
Kate
MW (106Btu/h)
1.38 (4.7)
1.38 (4.7)
1.55 (5.3)
1.52 (5.2)
1.55 (5.3)
1.58 (5.4)
1.58 (5.4)
1 58 (5.4)
ICO 1C A\
DO ID . *t )
1 . 58 (5.4)
1 58 (5 4)
1.58 (5.4)
1.52 (5.2)
1.47 (5.0)
1.38 (4.7)
1.47 (5.0)
1.58 (5.4)
1.55 (5.3)
1.52 (5.2)
1.44 (4.9)
1.61 (5.5)
1.58 (5.4)
1.41 (4.8)
°2.
%
2.9
3.4
3.4
2.6
2.8
2.9
3.0
3-4
3.3
2.8
39
. £
3.2
3.0
3.7
2.6
3.0
0.9
1.1
2.7
2.5
3.0
2.7
3.2
C02,
%
14.0
13.2
10.4
10.2
10.6
11.0
9.9
10. 0
10 . 0
10* 6
11 3
10.1
10.2
9.9
10.4
10.6
11.3
13.6
10.6
10.6
12.8
14.8
12.2
NOX
ppm* ng/.l
200 112
205 115
99 50.5
100 51.0
105 53.6
110 56.1
116 59.2
64 32 . 6
60 30*
69 35.2
89 45 4
103 52.5
74 37.7
75 38.3
62 31.6
65 33.2
40 20.4
41 20.9
79 40.3
120 61.2
294 164.9
183 102.7
307 172.2
NO
ppm* ng/.I
194 109
205 115
94 47.9
98 50.0
104 53.0
108 55.0
108 55.0
64 32 . 6
60 30. 6
69 35 . <
102 52.0
73 37.2
72 36.7
62 31.6
65 33.2
39 19.9
41 20.9
75 3H.3
116 59.2
292 163.8
177 99.3
299 167.7
CO
ppm*
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
S02
ppm*
963
844
-
~
1230
Stack Temp.
K (°F)
1141 (1594)
1141 (1594)
941 (1234)
1005 (1350)
1033 (1399)
1074 (1473)
1113 (1544)
1041 (1415)
1120 (1556)
1130 (1574)
1157 (1623)
1164 (1636)
1198 (1696)
1215 (1727)
1220 (1736)
1202 (1704)
1119 (1555)
1151 (1613)
980 (1305)
Heater
Effi-
ciency
%
53.1
52.1
57.9
56.3
54.7
52.8
50.9
53.7
50.6
49.1
49.7
48.6
50.5
49.5
46.9
47.9
54.6
53.7
60.5
Smoke
Spot
1
0.5
-
--
-
0
3
4
*B
1.17
0.85
0.88
0.8
0.84
0.75
0.7B
0.87
1.12
1.26
0.92
1.28
Staging
Height
m (ft)
--
_-
0.61 (2)
0.91 (3)
1.22 (4)
1.52 (5)
1.52 (5)
1.52 (5)
1.52 (5)
_-
__
0.91 (3)
FGR
%
~
Comments
02 swing at LoU-NOx
Register Settirig,
Tertiary Air Burner,
Tip 864, PAR = 10%
Open, SAR = TAR =
100% Open
©2 swing at Low-NOx
Register Setting,
Tertiary Air Burner,
Tip 864, PAR = 10%
Open, SAR - TAR =
100% Open
Baseline MA-16 Brn
Pattern II
PAR = 100% open,
SAR = 50% open
PAR = 100% open,
SAR = 20% open
PAH - 50% open.
SAR - 50% open
Baseline MA-16 Brn
Pattern II
50 Ib/h steam in j
78 Ib/h steam inj
Baseline MA 16 Brn
Pattern II
Baseline, MA-16 Brn
Pattern II
Staged air
Staged air
Staged air
Staged air
Staged air, low O-^
Staged air, low O^
Staged air
Baseline, MA-16 Brn
Pattern II
Baseline, MA-16 Brn
Tip 764
Staged air
Baseline, MA-16 Brn
Tip 764
CO
x)
VD
00
Corrected to 3% O^, Dry
(continued)
-------
TABLE A-l (Continued).
Test No.
1/6-13
1/6-14
1/6-15
1/6-16
1/6-17
1/6-18
1/6-19
1/7-1
1/7-2
1/7-3
1/7-4
1/7-5
1/7-6
1/7-7
1/7-8
1/7-9
1/7-10
1/7-11
1/7-12
1/7-13
1/7-14
1/7-15
1/7-16
1/7-17
1/7-18
1/7-19
1/7-20
FUL'l
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
NG
NG
NG
NG
NG
NG
NG
NG
NG
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
No. 6
Date,
l"7tl
3/16
3/16
3/16
3/16
3/16
3/16
3/16
3/21
3/21
3/21
3/21
3/21
3/21
3/21
3/21
3/21
3/22
3/22
3/22
3/22
3/22
3/22
3/22
3/22
3/22
3/22
3/22
Heat Inpuf
'. 'Bate'
HW' .(10uDVu/h)
1.44 (4.9)
1.44 (4.9)
1.49 j(5.1)
1.61 (5.5)
1.58 (5.4)
1.52 (5.2)
1.55 (5.3)
1.52 (5-2)
1.47 (5.0)
1.49 (5.1)
1.52 (5.2)
1.41 (4.8)
1.41 (4.8)
1.41 (4.8)
1.52 (5.2)
1.5B (5.4)
1.47 (5.0)
1.47 (5.0)
1.47 (5.0)
1.49 (5.1)
1.52 (5.2)
1.52 (5.2)
1.47 (5.0)
1.47 (5.0)
1.47 (5.0)
1.44 (4.9)
1.47 (5.0)
°2-
%
3.1
1.0
1.0
3.3
3.9
4.1
3.2
3.0
2.8
2.8
2.9
2.1
3.0
2.1
0.7
3.4
3.0
3.2
3.0
3.0
2.0
2.0
2.0
0.8
1.0
1.0
2.5
CO,
%
14.6
15.6
15.6
13.6
12. i
12.8
13.6
10.7
10.6
10.6
10.6
11.2
10. t
11.2
12.8
10.8
13.6
13.2
13.2
13.2
13.6
13.0
13.4
15.0
14.1
13.8
13.8
«ox
jjpm* ncj/.l
200 112.2
149 83. 6
166 93.1
203 113.9
226 126.8
224 125.7
266 149.2
91 47,4
79 40.3
77 39.3
84 42.8
43 21.9
48 24.5
47 24.0
43 21.9
116 59.2
252 141.4
194 108.8
183 102.7
174 97.6
156 87.5
172 96.5
194 108.8
171 95.9
148 83.0
155 87.0
241 135.2
, j)i>
ppni* ncj/.l
196 110.0
144 80.8
162 90.9
203 113.9
226 126.8
224 125.7
263 147.5
92 46.9
70 39.8
74 37.7
93 42.3
42 21.4
47 24.0
47 24.0
42 21.4
110 56.1
241 135.2
192 107.7
18.0 101.0
172 96.5
155 87.0
170 95.4
189 106.0
169 94.8
142 7'J.7
153 85. H
236 132.4
CQ>
npm*
0
0
0
0
0
0
0
0
0
0
:)
0
0
0
31
0
0
0
0
0
0
0
n
9
')
)
0
'/jSPj'
ppni*
1214
1232
1213.
102-7
1067
1030
958
--.
~
672
-
~
, St^cK Temp.
K (°F)
998 (1336)
1060 (1448)
1077 (1479)
943 (1237)
1001 (1355)
1035 (1403)
1073 (1471)
' 880 U125)
1037 (1407)
1132 (1578)
1142 (1596)
1096 (1514)
1131 (1577)
1155 (1620)
1210 (1719)
1223 (1741)
1056 (1441)
1103 (1526)
1110 (1538)
1113 (1544)
1135 (1583)
1151 (1613)
1174 (1653)
1166 (1639)
1180 (1665)
1171 (1041!)
1173 (1G52)
Heater
Effi-
ciency
%
59.9
60.1
5S.3
62.1
58.4
56.7
56.3
61.1
54.6
50.5
49.8
53.2
50.1
50.7
50.3
45.4
57.4
55.0
55.1
54.9
55.6
54.9
53.9
56.1
55.1
55.5
53.2
Sinoke
Spot
a
7
7
3
8
3
2
~
-.-
-
4.25
2.5
4
4
5
4
4
6
4
2
4
.
B
0.90
0.78
0,85
0.9S
1.03
O.98
1.25
~
Staging
Height
m (ft)
1.22 (4)
1.22 (4)
1.22 (4)
1.22 (4)
1.22 (4)
1.22 (4)
~
__
~
~
-_
FGR
%
~
-
15.7
15.4
11.6
40.2
36.9
38.2
36.9
19.3
30.1
37.5
36.8
27.8
18.4
20.6
28.8
37.6
Coiments
Staged air
Staged air, low O2
Staged air, low O.
Staged air
Staged air
Staged air
Baseline, HA- 1 6 Brr.
Tip 764
laseline
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc,
Low O_
Flue Gas Recirc
"lue Gas Recirc,
jow ;()
2
Flue Gas Recirc,
Low P2
Baseline
Baseline
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc
Flue Gas Recirc,
Low O.,
2
Flue Gas Recirc,
Low O
Flue Gas Recirc,
Low O_
Baseline
O
M
Ul
(continued)
-------
TABLE A-l (Continued).
Test No.
1/8-1
1/8-2
1/8-3
1/8-4
1/8-5
1/8-6
1/8-7
1/8-8
1/8-9
1/8-10
1/9-1
1/9-2
1/9-3
Fuel
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
Date,
1978
4/26
4/26
4/26
4/26
4/26
5/2
5/2
5/2
5/2
5/2
4/27
4/27
4/27
Heat Input
Rate
MW (10fcBtu/h)
1.47 (5.01)
1.46 (4.99)
1.54 (5.24)
1.5J (5.22)
1.52 (5.19)
1.49 (5.10)
1.49 (5.07)
1.47 (5.03)
1.49 (5.07)
1.46 (5.00)
1.52 (5.18)
1.49 (5.08)
1.54 (5.21)
°2-
%
3.0
1.1
0.3
4.5
3.1
2.9
0.5
1.1
4.05
3.0
3.1
3.1
2.<1
C02,
10.2
11.7
12.0
9.5
9.9
10.4
12.0
11.4
9.6
10.6
10.1
10.2
10. C>
NOX
ppm* ng/J
66 33.7
55 28.1
86 43.9
73 37.2
78 39.8
95 48.5
94 47.9
123 62.7
119 60.7
115 5B.7
NO
ppm* ng/J
64 32.6
54 27.5
48 24.5
58 29.6
55 28.1
84 42.8
70 35.7
77 39.3
93 47.4
90 45.9
122 62.2
116 59.2
114 5R.1
CO
ppm*
0
0
152
0
0
0
f>lr>
0
0
0
0
0
0
S02
ppm*
0
0
0
0
0
0
0
0
0
0
()
0
0
Stack Temp.
K (°F>
1104 (1527)
1158 (1625)
1179 (1663)
1169 (1644)
1171 (1648)
1098 (1517)
1113 (1543)
1119 (1554)
1115 (1547)
1118 (1552)
1133 (1581)
1156 (1622)
1172 (1651)
Heater
Effi-
ciency
%
50.8
51.3
51.5
45.2
47.6
51.2
53.6
52.9
48.5
50.2
49.3
48.3
47.9
Smoke
Spot
0
1
2
0
0
0
0.5
0
0
0
0
0
0
*B
_
Staging
Height
m (ft)
--
-_
__
__
__
FGR
%
CommentB
Baseline, DBA- 16 Brn
Pattern II, firing
ports toward tile.
data questionable.
sample line leak
O2 swing, same con-
figuration as 1/8-1,
data questionable,
sample line leak
O2 swing, same con-
figuration as 1/8-1,
data questionable.
sample line leak
O2 swing, same con-
figuration as 1/8-1,
data questionable.
sample line leak
O2 swing, same con-
figuration as 1/8-1,
data questionable.
sample line leak
Baseline, DBA- 16 Brn
Pattern II, firing
ports toward tile
O2 swing, same con-
figuration as 1/8-6
O2 swing, same con-
figuration as 1/8-6
O2 swing, same con-
figuration as 1/8-6
O2 swing, same con-
figuration as 1/8-6
Baseline, DBA- 16 Brn
Pattern II, gas tips
normal
0.0067 Kg/s (531b/h)
steam injection thru
oil. gun, oil gun
20 cm (8") below gas
tips, DBA-16,
Pattern II
0.0067 Kg/s (53 Ib/h)
steam injection, oil
gun 10 cm (4") below
gas tips
3
en
o
M
(Jl
-J
«>
00
Corrected to 3» O,, Dry
(continued)
-------
TABLE A-l (Continued).
Test No.
1/9-4
1/9-5
1/9-6
1/10-1
1/10-2
1/10-3
1/10-4
1/10-5
1/10-6
1/10-7
1/10-8
1/10-9
Fuel
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
Date,
1978
4/27
4/27
4/27
4/28
4/28
4/28
5/1
5/1
5/1
5/1
5/1
5/1
Heat Input
Rate
MW (106Btu/h)
1.54 (5.24)
1.55 (5.28)
1.55 (5.28)
1.49 (5.10)
1.51 (5.15)
1.48 (5.04)
1.50 (5.12)
1.61 (5.51)
1.55 (5.29)
1.57 (5.36)
1.54 (5.27)
1.56 (5.31)
°2-
%
3.0
3.0
3.0
3.0
3.1
3.2
3.0
3.1
4.9
1.2
0.5
3.1
C02,
%
10.6
10.6
10.3
10.2
9.9
9.9
10.3
10.6
9.6
11.8
12.0
10.5
NOX
pom* ng/J
118 60.2
114 58.1
136 69.4
145 74.0
115 58.7
98 50.0
88 44.9
94 47.9
114 58.1
66 33.7
54 27.5
87 44.4
NO
ppm* ng/J
115 58.7
110 56.1
130 66.3
142 72.4
113 57.6
94 47.9
84 42.8
88 44.9
110 56.1
66 33.7
53 27.0
06 43.9
CO
ppm*
0
0
0
0
0
0
0
0
c
0
439
0
SO2
ppm*
0
0
n
0
o '
c
0
0
o
0
0
0
Stack Temp.
K (°F)
1183 (1670)
1186 (1676)
1193 (1687)
1136 (1585)
1140 (1593)
1092 (1506)
1037 (1408)
1056 (1442)
1068 (1462)
1078 (1480)
1086 (1496)
1089 (1502)
Heater
Effi-
ciency
%
47.3
47.1
46.9
49.4
49.0
51.0
53.7
52.7
49.3
54.5
54.8
51.2
Smoke
Spot
0
0
0
0
0
0
0
0
0
0
0
0
*B
Staging
Height
m (ft)
_-
-_
__
FOR
%
~
--
Comments
0.0067 kg/s (53 Ib/h)
steam injection, oil
gun 5 cm (2") below
gas tips
0.0095 kg/s (75 Ib/h)
steam injection, oil
gun S era (2M) below
gas tips
Repeat baseline, no
steam injection
Baseline, staging
cyl. 7.6 cm (3")
above gas tips ,
Pattern 11
Baseline, staging
cyl. 15.2 cm (6")
above gas tips.
Pattern II
Baseline, staging
cyl. 22.9 cm (9")
above gas tips ,
Pattern II
Baseline, staging
cyl. 94.0 cm (37")
above gas tips,
Pattern II
Baseline, staging
cyl. 109 cm (43")
above gas tips.
Pattern II
02 swing, staging
cyl. 109 cm (43")
above tips. Pattern
II
02 swing, staging
cyl. 109 cm (43")
above tips. Pattern
II
OT swing, staging
cyl. 109 cm (43")
above tips, Pattern
II
02 swing, staging
cyl. 109 cm (43")
above tips, Pattern
IT
Ul
I
-J
10
00
Corrected to 3* O , Dry
(continued)
-------
TABLE A-l (Continued).
Test No.
1/11-1
1/11-2
1/11-3
1/11-4
1/11-5
1/12-1
1/12-2
1/12-3
1/12-4
1/12-5
1/12-6
1/12/7
1/12-8
Fuel
HG
NG
NG
NG
NG
NG
NG
NG
NG
NG
No. 6
No. 6
No. 6
Date,
1978
5/2
5/2
5/2
5/2
5/2
5/3
5/3
5/3
5/3
5/3
5/4
5/4
5/4
Heat Input
Rate
MW (106Btu/h)
1.50 (5.12)
1.50 (5.11)
1.51 (5.14)
1.43 (4.88)
1.51 (5.14)
1.54 (5.26)
1.53 (5.23)
1.55 (5.29)
1.52 (5.20)
1.52 (5.18)
1.41 (4.8)
est.
1.41 (4.8)
est.
1.41 (4.8)
est.
°2.
S
3.0
1.1
4.5
0.3
3.0
3.0
1.3
0.5
3.0
4.2
3.0
0.8
0.2
CO-,,
%~
10.3
11.1
9.5
12.0
10.6
10.6
11.5
11.7
10.2
9.9
13.7
14.7
15.2
NOX
ppm* ng/J
125 63.8
114 5B.1
146 74.5
103 52.5
141 71. »
101 51.5
90 45.9
88 44.9
111 56.6
114 58.1
239 134.1
201 112.8
164 92.0
NO
ppm* ngAl
122 62.2
110 56.1
142 72.4
97 4».5
135 68.9
96 49.0
87 44.4
82 41.8
108 55.1
111 56.6
229 128.'j
200 112.2
156 87.5
"
CO
ppm*
0
0
0
678
0
0
0
111
0
0
0
0
2')4
SO?
ppm-
0
0
0
0
0
0
0
0
0
0
1135
1127
1422
Stack Temp.
K (°F)
1059 (1447)
1107 (1533)
1108 (1535)
1114 (1546)
1115 (1547)
1067 (1460)
1117 (1551)
1126 (1567)
1131 (1577)
1131 (1577)
1123 (1561)
1150 (1610)
1153 (1616)
Heater
Effi-
ciency
%
52.7
53.3
4«.l
53.8
50.2
52.4
52.7
53.3
49.5
47.6
54.0
56.2
56.9
Smoke
Spot
0
0
0
1
0
0
0
0
0
0
1
1
2
*B
Staging
Height
m (ft)
-_
__
__
FGR
%
~
Comments
Baseline, DBA-16 Brn
Pattern II
O2 swing, DBA-16 Brn
Pattern II
02 swing, DBA-16 Brn
Pattern II
62 swing, DBA-16 Brn
Pattern II
02 swing, DBA-16 Brn
Pattern II
Baseline, tertiary
air brn. w/extended
secondary tile.
Pattern IIC
02 swing, tertiary
air brn. w/extended
secondary tile.
Pattern IIC
O2 swing, tertiary
air brn. w/extended
secondary tile,
Pattern IIC
02 swing, tertiary
air brn. w/extended
secondary tile.
Pattern IIC
02 swing, tertiary
air brn. w/extended
secondary tile,
Pattern IIC
Baseline, tertiary
air brn. w/extended
secondary tile. Tip
784
02 swing, tertiary
air brn. w/extended
secondary tile, Tip
784
OT swing, tertiary
air brn. w/extended
secondary tile. Tip
784
O
M
Ui
10
00
Corrected to 3% O , Dry
(continued)
-------
TABLE A-l (Continued).
Test No.
1/12-9
1/12-10
fuel
No. 6
Mo. 6
Date,
1978
5/4
5/4
Heat Input
Rate
MW (106Btu/h)
1.41 (4.8)
est.
1.41 (4.8)
est.
O2,
%
3.1
4.2
C02,
13.6
13.2
NOX
ppm« ng/J
237 133.0
24fi 118.0
NO
ppm' ng/J
231 130.0
2T> 11/1.1
CO
ppm«
0
0
SO2
ppm'
972
l?on
Stack Temp.
K (°F)
1158 (1624)
11 '.ft (lf,J2)
Heater
Effi-
ciency
%
52.2
10.1
Smoke
Spot
1
o.r,
B
Staging
Height
m (ft)
_-
FOR
%
--
Comments
O2 swing, tertiary
air brn. w/extended
secondary tile, Tip
784
O2 swing, tertiary
air brn. w/extended
secondary tile, Tip
784
Corrected to 3% O2> Dry
'PAR = Primary Air Register
2SAR = Secondary Air Register
3TAR = Tertiary Air Register
Ln
ID
00
-------
APPENDIX B
SUMMARY OF GASEOUS EMISSION DATA
LOCATION. 4,
SUBSCALE STEEL FURNACE
KVB 6015-798
-------
TABLE B-l. SUMMARY OF GASEOUS EMISSION DATA, LOCATION 4, SUBSCALE STEEL FURNACE
test No.
4/1-i
4/1-2
4/1-3
4/1-4
4/2-1
4/2-2
4/2-3
4/2-4
4/3-1
4/3-2
4/3-3
4/3-4
4/3-S
4/3-6
4/3-7
4/3-8
4/3-9
4/3-10
4/3-11
4/3-12
4/4-1
4/4-2
4/4-3
4/4-4
4/4-5
4/4-6
4/4-7
4/4-8
4/4-9
4/4-10
4/4-11
4/4-12
4/4-13
4/4-14
4/4-15
4/4-16
4/4-17
4/4-1B
4/4-19
4/4-20
4/4-21
4/4-22
4/4-23
Fuel
NG
NG
HG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
Date,
1979
7/12
7/12
7/12
7/12
7/12
7/12
7/12
7/12
7/13
7/13
7/13
7/13
7/13
7/13
7/13
7/27
7/27
7/27
7/27
7/27
7/18
7/18
7/18
7/18
7/18
7/18
7/18
7/18
7/18
7/19
7/19
7/19
7/19
7/19
7/19
7/19
7/20
7/20
7/20
7/20
7/20
7/20
7/20
Heat Input
Rate
HW (10 Btu/h)
0.29 (1.0)
0.29 (1.0)
0.29 (1.0)
0.29 (1.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
O.59 (2.0)
0.59 (2.0)
O.S9 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.56 (1.9)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
*
4.17
1.90
0.45
4.25
3.1
4.1
2.0
0.2
1.90
1.95
1.92
1.98
2.00
2.05
1.90
2.02
2.00
2.17
2.20
1.81
4.10
2.08
0.25
2.05
2.02
2.00
4.15
0.30
0.20
2.05
2.05
2.04
1.95
2.10
4.05
4. OS
0.40
0.26
0.33
0.21
0.41
4.14
2.00
CO,
9.2
10.2
11.6
9.2
9.9
8.9
10.8
11.5
10.3
10.2
10.1
10.1
10.2
9.9
10.2
11.1
11.3
10.1
10.1
10.6
9.4
-
11.7
11.3
11.0
11.4
10.0
11.5
11.6
11.8
10.6
10.8
9.9
10.1
9.5
8.9
11.3
11.4
11.2
11.3
11.4
9.2
10.6
NO
ppm* ng/J
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
186 96.1
130 67.2
105 54.2
98 50.6
24 12.4
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
NO
ppm* ng/J
128 66.1
216 111.6
180 93.0
160 82.7
181 93.5
168 86.6
189 97.6
192 99.2
203 104.9
118 61.0
141 72.8
156 80.6
166 85.8
185 95.6
207 106.9
186 96.1
130 67.2
105 54.2
98 50.6
24 12.4
117 60.4
157 81.1
162 83.7
213 11O.O
178 92.0
102 52.7
75 38.7
75 38.7
251 129.7
52 26.9
280 144.7
46 23.8
38 19.6
310 160.2
239 123.5
38 19.6
22T 117.3
68 35.1
51 26.3
35 18.1
258 133.3
216 111.6
258 133.3
CO
PP»«
64
56
473
11
23
11
57
312
52
57
61
39
43
43
56
10
6
6
7
7
48
43
302
76
10
10
9
152
86
33
10
14
38
28
21
16
166
239
109
173
140
0
0
HC
ppm'
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
10
13
13
11
11
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
SO
PP">*
I/A
N/A
N/A
N/A
N/A
N/A
N/A
«/A
N/A
ri/A
N/A
N/A
N/A
N/A
M/A
0
0
0
0
0
N/A
K/A
N/A
N/A
N/A
M/A
II/A
N/A
N/A
t'/A
1VA
(./A
N/A
N/A
»VA
t/A
K/A
N/A
N/A
n/A
K/A
N/A
M/A
S°3.
pp..
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Stack Temp
K (°F)
1458 (2165)
1559 (2346)
1594 (2410)
1406 (2071)
1547 (2325)
1523 (2282)
1578 (2380)
1604 (2428)
1514 (2265)
1519 (2275)
1516 (2270)
1522 (2280)
1533 (2300)
1536 (2305)
1533 (2299)
1551 (2332)
1551 (2333)
1544 (2320)
1533 (2300)
1555 (2340)
1550 (2330)
1547 (2325)
1553 (2335)
1547 (2325)
1514 (2265)
1500 (2240)
1485 (2213)
1571 (2368)
1579 (2382)
1559 (2347)
1529 (2293)
1485 (2214)
1510 (2258)
1610 (2439)
1531 (2297)
1507 (2253)
1533 (2300)
1562 (2352)
1569 (236S)
1561 (2351)
1623 (2461)
1538 (2309)
1539 (2310)
Furnace
Effic.
%
24.5
24.1
27.9
26.5
26.9
23.2
29.0
30.7
30.0
27. 0
27.3
27.6
28.2
27.4
28.7
31.3
29.8
24.7
24.0
-
22.0
-
30.9
32.1
29.8
29.3
25.0
27.3
31.7
26.3
30.3
25.6
19.4
25.2
25.7
14.8
32.7
27.3
24.4
22.2
29.5
24.1
30.0
FGR
t
16.9
4.9
5.1
5.4
0
5.1
9.9
9.3
10.3
O
14.0
0
15.5
19.3
0
0
16.6
0
10.1
12.9
17.1
0
0
0
Hater/Steam
Injection Rate
g/s (Ib/h) cement*
Excess O variation at half capacity.
2% O , half capacity
Excess O2 variation at half capacity.
Excess O variation at half capacity.
Excess O variation at full capacity.
Excess O variation at full capacity.
Baseline
Excess O variation at full capacity.
0 Baseline
2.52 (20.0) Mater Injection
1.89 (15.0) Mater Injection
1.26 (10.0) Hater Injection
0.63 (5.0) Hater Injection
0 Atomizing air only - no Hater Injection
0 Baseline
0 Baseline
2.52 (20.0) Hater Injection
3.78 (30.0) Hater Injection
5.04 (40.0) Hater Injection
5.04 (40.0) Maximum Mater Injection plus maximum FGR
5% FGR, High 0
5% FGR, Normal O
5% FGR, low Oj
Baseline
5% FGR, Normal O
10% FGR, Normal O
10% FGR, High O-
10% FGR, Low O
No FGR, Low O
15% FGR, Normal O
Baseline
15% FGR, Normal O
Maximum FGR, Normal O
Baseline
No FGR, High O
Maximum FGR, High O
No FGR, Low O_
10% FGR, LOW 6
15% FGR, Low 0
Maximum FGR, Low O
No FGR, Low O
No FGR, High 6
Baseline
V
I dry corrected to 3%
2
N/A - Not available -(Hot line out of service)
Continued
-------
TABLE B-l. Continued
TMt NO.
4/5-1
4/5-2
4/5-3
4/5-4
4/6-1
4/6-2
4/6'3
4/6-4
4/6-S
4/6-6
4/7-1
4/7-2
4/7-3
4/7-4
4/7-S
4/7-6
4/8-1
4/6-2
4/8-3
4/8-4
4/8-5
4/8-6
4/8-7
4/8-8
4/8-9
4/8-10
4/8-11
4/8-12
4/8-13
4/8-14
4/8-15
4/8-16
4/8-17
4/8-18
4/8-19
4/9-1
4/9-2
4/9-3
4710-1
4/10-2
4/10-3
Fuel
NO, 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No; 2
No. 2
No. 2
NG
NG
NG
NG
NG
NG
bate,
1979
7/23
7/23
7/23
7/23
7/23
7/23
7/23
7/27
7/27
7/27
7/24
7/24
7/24
7/24
7/24
7/24
7/25
7/25
7/25
7/25
7/25
7/25
7/25
7/25
7/26
7/26
7/26
7/26
7/26
7/26
7/26
7/26
7/26
7/26
7/27
7/31
7/31
7/31
7/31
7/31
7/31
Heat Input
Rate 0
MW (10 Btu/h) %
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.28 (1.0)
0.33 (1.1)
0.33 (1.1)
0.30 (1.0)
0.30 (1.0)
0.30 (1.0)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.52 (1.8)
0.55 (1.9)
'0.55 (1.9)
0,55 (1.9)
O.SS (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0.55 (1.9)
0,55 (1.9)
0.55 (1.9)
,0.59 (2.0)
0.59 (2.0)
0.59 (2.0)
0.29 (1.0)
0.29 (1.0)
0.29 (1.0)
4.05
2.05
1.3O
2.95
4.10
2.15
0.35
4.10
1.95
0.30
2.00
2.05
2.07
2.03
2.10
2.07
3.97
3.98
4.00
4,00
4.10
4.00
2.10
2.00
2.10
1.95
2.10
2.10
1.75
0.90
0.60
0.60
0.65
0.80
0.72
Except
3.90
2.15
0.55
4.10
2.00
0.4O
CO,
12.1
13.8
14.0
12.9
12.9
12.8
14.8
12.1
13.6
15.1
13.0
14.4
14.0
14.0
14.0
13.2
12.5
13.6
13.6
12.5
11.9
14.1
13.0
13.4
14.8
14.2
14.3
14.3
15.0
15.4
15.2
14.9
14.9
14.8
-
for Test
9.5
10.6
11.4
9.9
11.0
12.0
ppm* ng/J
N/A
N/A
N/A
N/A
N/A
N/A
N/A
226 124.9
232 128.3
191 105.6
N/A
N/A
N/A
N/A
N/A
N/A
243 134.3
116 64.1
149 82.4
76 42.0
72 39.8
233 128.8
219 121.1
66 36.5
246 136.0
57 31.5
90 49.8
124 68.5
140 77.4
181 100.1
55 30.4
77 42.6
109 60.2
339 187.4
359 198.5
4/12-1, Flam
126 65.1
172 88.9
185 95.6
149 77.0
180 93.0
179 92.5
NO
ppm* ng/J
329 181.9
375 207.3
384 212.3
364 201.2
181 100.1
272 150.4
205 113.3
226 124.9
232 128.3
191 105.6
223 123.3
24 13.3
47 26.0
78 43.1
141 77.9
271 149.8
243 134.3
116 64.1
149 82.4
76 42.0
72 39.8
233 128.8
219 121.1
66 36.5
246 136.0
57 31.5
90 49.8
124 68.5
140 77.4
181 100.1
55 30.4
77 42.6
109 60.2
339 187.4 .
359 198.5
CO
ppm*
10
57
238
10
21
14
227
11
9
43
11
19
28
28
24
24
11
11
11
11
11
11
10
9
8
9
9
9
9
103
132
256
345
339
169
HC
ppm*
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0
0
2
N/A
N/A
N/A
N/A
N/A
N/A
22
20
19
25
22
18
23
7
19
6
12
13
a
14
-
16
17
16
2
SO SO
ppm* ppm*
N/A
N/A
N/A
N/A
N/A
N/A
N/A
-
-
-
N/A
N/A
N/A
N/A
N/A
N/A
340
212
259
229
205
285
236
272
235
257
211
313
235
271
258
319
316
280
-
e Temperature Profile Measuiaments
126 65.1
172 88.9
185 95.6
149 77.0
180 93.0
179 92.5
3
5
80
2
2
131
0
5
4
3
2
2
0
0
0
0
0
0
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
were
-
-
-
-
-
Furnace
Stack Temp Effic.
K (°F) %
1530
1544
1525
1547
1426
1494
1425
1466
1475
1491
1533
1533
1514
1551
1553
1559
1533
1567
1523
1529
1520
1615
1545
1556
1553
1535
1578
1547
1541
1541
1553
1569
1543
1566
1522
made
1557
1521
1535
1499
1535
1530
(2294)
(2320)
(2285)
(2325)
(2107)
(2229)
(2106)
(2180)
(2195)
(2225)
(2300)
(2300)
(2265)
(2332)
(2335)
(2346)
(2300)
(2361)
(2292)
(2292)
(2276)
(2448)
(2322)
(2341)
(2335)
(2304)
(2380)
(2325)
(2315)
(2315)
(2336)
(2365)
(2318)
(2360)
(2280)
during all
(2343)
(2278)
(2303)
(2239)
(2303)
(2294)
28.8
34.1
35.5
31.1
32.1
29.3
37.5
27.8
32.5
36.0
32.0
33.8
34.1
33.4
33.8
31.7
30.1
27.9
31.3
21.6
20.1
32.4
31.5
25.2
36.6
25.3
27.3
30.7
35.0
35.8
27.9
28.9
33.1
36.3
37.8
of the
24.6
30.6
32.9
25.3
28.0
31.3
Hater/Steam
FGR Injection Rate
% g/B (Ib/h)
0
5.38 (42.7)
3.78 (30.0)
2.52 (20.0)
1.32 (10.5)
0
0
9.0
5.4
15.7
15.9
0
0
18.0
0
21.2
14.7
10.2
5.3
6.1
19,8
16.8
10.0
0
0
following tests.
Comments
Excess O_ variation at
Baseline
Excess O variation at
Excess O variation at
Excess O2 variation at
2% O , low firing rate
Excess O. variation at
Excess O variation at
2% O , low firing rate
Excess O variation at
Baseline
Steam Injection
Steam Injection
Steam Injection
Steam Injection
Baseline
No FGR, High O,
10% FGR, High 6
5% FGR, High O
15% FGR, High 6.
Maximum FGR, High 02
No FGR, High O
Baseline
Maximum FGR, Normal O
Baseline
Maximum FGR, Normal O
15% FGR, Normal O
10% FGR, Normal O
5% FGR, Normal O
5% FGR, Low 0
M»viMi|m FGR, LOW O_
15% FGR, Low 0
10% FGR, Low O
No FGR, Low O
No FGR, Low O2
Excess O variation at
Baseline
Excess O variation at
Excess O. variation at
2% O , half capacity
Excess O variation at
full capacity.
full capacity.
full capacity.
low firing rat*.
low firing rate
low firing rat*
low firing rate
full capacity
full capacity
half capacity
half capacity
dry corrected to 3% O-
N/A - Not available - (Hot line out of Mirvic*)
Continued
-------
TABLE B-l. Continued
tm*t Ho.
4/11-1
4/11-2
4/11-3
4/12-1
4/12-2
4/13-1
4/13-2
4/13-3
4/14-1
4/14-2
4/14-3
4/15-1
4/15-2
4/16-1
4/16-2
Fuel
NG
NG
NG
NG
NG
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
Date,
1979
7/31
7/31
7/31
8/1
8/1
8/1
8/1
8/1
8/1
8/1
8/1
8/2
8/2
8/2
8/2
Heat Input
Rate 0
MW (10 Btu/h) %
0.59
0.59
0.59
0.59
0.59
0.30
0.30
0.30
0.55
0.55
0.55
0.55
0.55
0.55
0.55
(2.0)
(2.0)
(2.0)
(2.0)
(2.0)
(1.0)
(1.0)
(1.0)
(1.9)
(1.9)
(1.9)
(1.9)
(1.9)
(1.9)
(1.9)
1.97
2.00
1.90
1.90
2.00
4.10
1.97
0.40
4.00
2.00
1.15
1.90
2.05
2.00
2.00
"2
11.0
11.0
11.0
11.0
10.6
12.3
14.1
15.6
12.0
13.2
13.6
13.6
13.7
14.0
14.0
NO
ppm* ng/J
180 93.0
116 59.9
221 114.2
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
NO
ppra* ng/J
18O 93.0
116 59.9
221 114.2
200 103.3
52 26.9
250 138.5
303 167.9
258 142.9
307 170.1
374 207.2
354 196.1
262 145.2
52 28.8
248 137.4
95 52.6
CO
ppm*
5
5
5
9
4
5
5
39
5
38
362
38
95
14
9
HC
ppm*
4
2
2
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
SO
ppm*
0
0
0
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
351
SO Stick Temp
ppm" K <°F)
1511
1505
1489
1521
1510
1525
1516
1531
1518
1526
1513
1556
1525
1533
0 1515
(2260)
(2250)
(2220)
(2278)
(2259)
(2286)
(2270)
(2297)
(2272)
(2288)
(2263)
(2342)
(2285)
(2300)
(226*)
furnace
Efflc.
%
30.4
28.4
33.4
31.9
24.9
26.2
32.5
35.9
28.9
32.9
34. B
33.1
32.3
35.1
29.8
Mater/Steam
FGR Injection Rate
% g/s (Ib/h)
2.52 (20.0)
5.04 (40.0)
0
0
12.6
-
-
-
-
-
-
-
5.04 (40.0)
0
12. »
Comment!
Water Injection
Hater Injection
Baseline
Baseline
15% FGR, Normal O2
Excess o Variation at low firing rate
2% O* low firing rate
Excess O variation at low firing rate
Excess O- variation at full capacity
Baseline
Excess O. variation at full capacity
Baseline
Steam Injection
Baseline
15% FGR, Normal O , Gokeoyr-Ross (SO ) Test
* dry corrected to 3% 0
N/A - Mot available - (lot line out of service)
-------
APPENDIX C
CALCULATION OF INCREMENTAL FUEL REQUIREMENTS OF
COMBUSTION MODIFICATIONS TO A STEEL FURNACE
The calculation of the incremental heat requirement of steam or water
injection or flue gas recirculation when applied to a steel furnace necessitates
the assumption that there is no effect of these modifications on furnace
thermal efficiency other than the additional thermal losses caused by having
to heat the injected materials to combustion temperatures. Thus, the effects
of the combustion modifications on convective or radiative heat transfer
rates, which also affect furnace efficiency, are not considered in this
report.
In an actual application, the convective heat transfer rates from
the combustion gases to the steel itself will probably increase because of
the higher mass flow of gases through the furnace brought about by the
injection of additional material. This may partially offset the efficiency
degradation associated with the added thermal load of the injected material.
In addition, the radiative heat transfer rate may also be increased
by flue gas recirculation or by steam or water injection. The increase in
the partial pressure of CO and H»0 resulting from the injection of these
£. £
materials would tend to increase the emissivity of the combustion gases
and, therefore, the radiative heat transfer rate to the steel.
The effects of altered convection and radiative heat transfer need
to be studied further in order to develop more meaningful efficiency
assessments of the combustion modifications discussed in this section.
The incremental heat requirements for steel furnace combustion
modifications are calculated in the order of increasing complexity, beginning
C-l KVB 6015-798
-------
with steam injection, followed by water injection and, finally, flue gas
recirculation. The percent increase in heat load is equal to the percent
increase in fuel required.
The incremental heat requirement per burner of steam injection at
5.0 g/s (40 Ib/hr) injected steam flow rate is defined as follows:
Ah = heat required to take steam from the injection conditions
to the furnace bulk gas temperature, T
For P = 1 atm, T =273K, Pp=l atm, and T =1755K (typical measured
operating temperature)
h^ = 3384 kJ/kg (1,455 Btu/lb) and
F
Thus,
hIN = 419 kj/k9 (18° Btu/lb>
Ah = 0.015 MW (51,000 Btu/hr)
= 2.56% of experimental burner capacity
of 0.586 MW (2.0xl06 Btu/hr)
The incremental heat requirement per burner for 5.0 g/s (40 Ib/hr)
injected water flow rate is determined below. This requirement includes
the heat necessary to raise the water temperature to the boiling point,
complete the phase change to steam, and heat the steam to the furnace bulk
gas temperature.
WATER ~ STM + "H 0 fg + PH O(£) B.P. ~ TIN
Where
Ahj. = heat of vaporization of water at p = 1 atm,
fg
T = 373K (212°F) = 2256 kJ/kg (970 Btu/lb)
Cp = specific heat of water
T = 373K (212°F) = boiling point of water at p=l atm
B. P.
T = injected water temperature
C-2 KVB 6015-798
-------
For T = 294K (70°F),
IN
All = 0.028 MW (95,480 Btu/hr)
= 4.75% of experimental burner capacity
The heat losses associated with flue gas recirculation in a steel
furnace arise from the cooling of the flue gas
temperature to the reinjection temperature. In the experimental arrangement
most of this cooling occurred in an air-gas heat exchanger, and no heat
was recovered, i.e., all of the heat was lost to the ambient air. It is
emphasized here that in a practical application of flue gas recirculation
much of this heat could be retained within the furnace proper by combustion
air preheat or some other means of waste heat recovery. The following diagram
illustrates the calculation of furnace efficiency for the steel furnace.
CONTROL
VOLUME
FUEL"
HEAT EXCHANGER
TFGR=559K(546°F)
STACK
(T =T )
^ STACK F'
FURNACE
T =1755K(2700°F)
r
_[ __j I
(Q )
^
RADIATION
C-3
KVB 6015-798
-------
One observes from this drawing the increased heat loss from the steel
furnace with FGR as compared to a furnace without FGR. In the calculations
which follow we assume that there are no factors influencing the furnace
efficiency other than this heat loss. This assumption may not be strictly
valid, however, for two reasons:
1. The convective heat transfer coefficient of the combustion gases
in the furnace should increase with the increased mass flow through
the furnace due to FGR, thereby increasing the convective heat
transfer to the steel.
2. The recirculation of flue gases containing large amounts of the
radiative species CO and HO may increase the emissivity of the
combustion gases within the furnace, thus increasing radiative
heat transfer rates to the steel. (This is also especially true
for the case of HO injection where the volume fraction of HO
in the combustion gases is significantly increased. )
The incremental heat (fuel) requirement for 20 percent FGR is
determined below:
AhFGR = ^FGR^WVW
where (Cp) = specific heat of the flue gas
FGR
T = flue gas temperature at point of injection
into the furnace
= recirculated flue gas mass flow rate
For Test #4/4-13,
T = 1755K (2700°F), T = 559K(546°F),
F r GR
(C ) =1.09 kJ/Kg-K (0.26 Btu/lbm- °R) ,
p FGR
and nW^0-051 k9/s (400.8 Ibm/hr)
and so
Ah = 0.066 MW (224,464 Btu/hr)
FGR
= 11.2% of burner heat input capacity.
Thus, with no waste heat recovery, and neglecting the effects of
the combustion modifications on convective and radiative heat transfer, the
additional fuel requirements are summarized in Table C-l.
C-4 KVB 6015-798
-------
TABLE C-l. INCREMENTAL FUEL REQUIREMENTS OF
COMBUSTION MODIFICATIONS TO A STEEL FURNACE
Modification
Percent Increase in Fuel Consumption
Steam Injection
Water Injection
Flue Gas Recirculation
2.56
4.75
11.2
C-5
KVB 6015-798
-------