United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park, NC 27711
Research and Development
EPA/600/S7-88/019 Jan. 1989
&EPA Project Summary
Evaluation of Low-Emission
Coal Burner Technology on
Industrial Boilers
B. A; Folson, A. R. Abele, J. L. Reese, and T. M. Sommer
The Distributed Mixing Burner
(DMB) is a low-NOx pulverized-coal
burner for wall-fired boiler
applications. The burner operates
under reducing conditions in the
primary flame zone to minimize NOX
emissions while an overall oxidizing
environment is maintained in the
furnace to minimize slagging and
corrosion. This operation is
accomplished by using air ports
around the burner throat to provide
staged combustion conditions at
each individual burner.
This report gives results of a field
evaluation of the DMB on a 98 kg/hr
(215,000 Ib/hr) steaming capacity,
four-burner, front-wall-fired
boiler. Prior to the DMB retrofit, field
tests established baseline operating
conditions with the pre-NSPS (New
Source Performance Standards)
burners originally installed in the
boiler. Following DMB installation,
the boiler was operated and tested
with the new burners for 17 months.
Under routine operation, the DMBs
reduced NOX emissions by about
50%-from a baseline condition of
about 0.96 to about 0.46 lb/1Q6 Btu
(418 to 200 ng/J). Under carefully
controlled, optimized conditions, NOX
emissions were further reduced
another 20%-to about 0.3 lb/106 Btu
(131 ng/J).
This Project Summary was
developed by EPA's Air and Energy
Engineering Research Laboratory,
Research Triangle Park, NC, to
announce key findings of the
research project that is fully
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
Staged combustion is one of the most
effective techniques for reducing NOX
emissions from fuels containing high
levels of nitrogen. Its objectives are to:
(1) reduce the peak flame temperature
which significantly reduces the oxidation
of nitrogen introduced by the combustion
air.and (2) produce an oxygen-deficient
(fuel-rich) flame zone that inhibits
oxidation of all nitrogen species present,
including those introduced by the fuel
itself. The concept involves firing the fuel
under oxygen conditions (primary flame
zone) initially followed by secondary air
addition to complete the combustion
process in an air-rich environment
(secondary flame zone). In the fuel-rich
primary zone, nitrogen compounds are
reduced to N2 preferentially prior to
adding the secondary air. The
effectiveness of this staged air addition in
reducing NOX emissions depends on
combustion conditions, particularly in the
fuel-rich primary zone. The optimum
primary zone stoichiometry is typically in
the range of 70% theoretical air (TA).
Staged combustion has been
demonstrated to be effective in reducing
NOX emission on full-scale wall-fired
boilers through the use of overfire air
ports. However the primary zone
stoichiometry must be maintained
approximately at stoichiometric
conditions (100% TA) to avoid slagging
and corrosion in the lower furnace and to
-------
achieve adequate char burnout. This
limits the effectiveness of NOX control.
For the last several years, the EPA
has been developing a Iow-N0x
pulverized-coal burner that achieves
staging by means of air ports around a
circular burner. This Distributed Mixing
Burner (DMB) allows a fuel-rich primary
zone to be established in the furnace
adjacent to the burner while maintaining
an overall oxidizing environment farther
into the furnace to minimize slagging,
corrosion, and char burnout.
Figure 1 shows how the fuel/air mixing
is staged sequentially. The combustion
process occurs in three zones. In the first
zone, pulverized coal transported by the
primary air combines with the inner
secondary air to form a very fuel-rich
(30 to 50% TA) recirculation zone which
provides flame stability. The coal
devolatilizes, and fuel nitrogen
compounds are released to the gas
phase. Outer secondary air is added in
the "burner zone" where the
stoichiometry increases up to about 70%
TA. Air to complete the combustion
process is supplied through tertiary air
ports outside the burner throat. This
allows substantial residence time in the
burner zone for decay of bound nitrogen
compounds to N2 and radiative heat
transfer to reduce peak temperatures.
The tertiary air ports surrounding the
burner throat provide an overall oxidizing
atmosphere and minimize interactions
between adjacent burners.
The objective of this program was to
integrate the DMB concept with
commercial burner components and to
field test the DMB performance on a full
scale boiler. The goal was to attain NOX
emissions less than 87 ng/J (0.2 lb/106
Btu) or as low as possible without
adverse effects on boiler operability and
durability, thermal efficiency, and the
emission of other pollutants.
The characteristics of industrial and
small utility boilers in the U.S. were
analyzed, and several Foster Wheeler
boilers were identified as candidate host
units. Western Illinois Power
Cooperative's (WIPCO's) Pearl Station 1
was selected as the host boiler. This unit
is a 98 kg/hr (215 x 103 Ib/hr) steam
front-wall-fired utility boiler with four
pre-NSPS Foster Wheeler Intervene
Burners. A field test was conducted to
establish the baseline performance.
A prototype DMB was designed by
integrating the DMB design criteria with
Foster Wheeler burner components.
Some compromises were necessary to
meet the space and geometrical
requirements imposed by the host boiler.
Comprehensive burner tests were
conducted in two research furnaces with
an Intervene Burner (to establish baseline
performance) and with several
configurations of the prototype DMB. The
tests of the initial prototype DMB
achieved low NOX emissions but the
flame stability was unacceptable at the
burner operating point, probably due to
the differences in design between the
research and commercial burner
components. To improve flame stability,
the prototype burner exit was modified to
incorporate Foster Wheeler's proprietary
Controlled Flow Burner exit geometry.
The Controlled Flow Burner is Foster
Wheeler's first generation Iow-N0x
design which was offered commercially
until 1979. It has since been replaced by
a second generation design. Subsequent
tests demonstrated that this DMB
configuration met all operational
requirements necessary for the field
installation. Consequently, the field-
operable DMBs were based on this
design. Subsequently, other prototype
DMB configurations which did not
include proprietary burner design
parameters were tested and optimized.
These DMB configurations achieved low
NOX emission and adequate flame
stability.
The host boiler was retrofitted with
DMBs in the spring and summer of 1981
and the burner/boiler performance was
evaluated for 17 months.
Host Boiler
Figure 2 shows the general
arrangement of the host boiler, WIPCO's
Pearl Station Unit 1. Table 1 lists the
design and operating characteristics.
This unit is representative of a large
number of medium-pressure pre-
NSPS industrial and small utility boilers.
There is no reheater or economizer, and
a portion of the steam is generated in the
convective pass. This unit was erected in
1965 by Foster Wheeler and was
equipped with four Intervane Burners in a
2x2 array. The Intervane Burner is
Foster Wheeler's standard commercial
pre-NSPS burner. Pearl Station is the
principal power plant of WIPCO and
normally operates base-loaded.
Pearl Station fires a variety of fuels.
Small quantities of coal are purchased
from several mines based on cost ($/106
Btu) primarily. Since the beginning of this
program, coals from four mines in
Indiana, Illinois, and Missouri have been
fired. All are high volatile bituminous
coals but their compositions vary over a
considerable range.
To establish the DMB operatir
requirements and to determine tr
baseline performance with the origin
burners, a comprehensive field test w?
conducted. In general, the measure
baseline performance was typical i
many pre-NSPS units. Carbo
utilization, as measured by carbc
content of the particulate emissions usir
coal ash as a tracer was typicall
99.65% CO emission was less than A
ppm, and hydrocarbon and ammon
emissions were negligible. NC
emissions ranged from 807 to 847 ppn
which corresponds to 395 to 412 ng,
(0.91 to 095 lb/106 Btu) Since this boik
was constructed before 1971, it is n<
subject to the NSPS emissio
regulations.
DMB Design
Figure 3 shows the design of th
central portion of the prototype DM
which integrates the DMB design criteri
with Foster Wheeler commercial burne
components and meets the host boik
requirements. This burner design meet
all of the DMB design criteria except fc
the burner exit. The wmdbox depth an
burner-to-burner spacing of the hos
boiler, which are typical of domesti
boiler designs, required a short burne
exit The exit geometry configuration
illustrated is the initial configuratio
tested in the research furnace. Th
burners installed at Pearl Station ar
similar except for the exit geometr
which was modified to match Foste
Wheeler's proprietary Controlled Flo\
Burner parameters.
DMB Installation
The primary difference in thi
installation requirements for the DMB an<
conventional burners is the tertiary ai
port array required for the DMB. Basei
on the research furnace tests, tertiary ai
ports should be arranged to distribute thi
air uniformly around each burner n
discrete jets. This requires tertiary ai
ports between the burners and outboan
of the burner array The preferred DME
application is in new boilers when
adequate provision for the tertiary ai
ports can be provided as part of th<
boiler design. Since the DMB installatioi
in the host boiler was a retrofit, som<
compromises in both the DMB desigi
and the host boiler equipment wer<
required.
After considerable analysis an<
discussion, the compromise por
arrangement illustrated in Figure 4 w
selected. Thirteen tertiary air ports
installed, four above the burners, fou
-------
Tertiary Air
Inner
Secondary Air
Coal and
Primary Air
/ Progressive Air Addition Zone \
Very Fuel-Rich (Overall Stoichiometry 70%) \
Zone (A verage f ,na/ Air Addition Zone for Burnout
Stoichiometry 40%) (Overall Stoichiometry 120%l
Figure 1. Distributed mixing burner concept.
Spray
Control
Header
Secondary __
Superheater
Primary
Superheater
Feeders —
Pulver-
izers
Regenerative
Air Heater
— FDFan
Primary
Air Fan
Figure 2. WIPCO's Pearl Station Unit 1.
below the burners, four outboard of the
burners, and one in the center of the
burner array. This arrangement
distributes the tertiary air around and
between the burners while minimizing
structural problems. The ports above the
burner are the maximum size that will
clear the buck stay. The port in the
center was reduced in diameter to allow
clearance for its control mechanism
between the burner registers. The ports
along the sides have been moved to
allow free flow into the pulverizer ducts.
The bottom of the windbox has been
lowered to provide a plenum for
installation of the lower ports.
The modifications to the host boiler
equipment included:
• Front Wall Replacement • Due to
the large number of complex tube
bends, the entire front wall had to
be replaced.
• Buck Stay Removal - The buck
stay between the burners had to be
removed, and additional windbox
structures were constructed to
handle the pressure loads.
• External Hopper Support • The
front wall strength was inadequate
to support the hopper due to the
large number of tube bends. An
external hopper support was
constructed to handle the hopper
weight.
Since one pulverizer would be taken out
of service at low loads, the burners were
designed to operate with the tertiary
ports both closed and open. Electrical
drives were provided to control all
tertiary air ports and registers, a total of
18 drives. With the ports closed, leakage
airflow (for cooling) resulted in an
average burner zone Stoichiometry of
about 100% TA. Opening the ports fully
and restricting the airflow through the
burner throats using the sleeves and
registers reduced the burner zone
Stoichiometry to near the DMB design
point, 70% TA.
Field Test Results
Since the completion of the DMB
retrofit, the host boiler has operated in
commercial service for 45 months
(except for brief outages). This includes
17 months of intensive testing with the
tertiary air ports open as part of this
program and 28 months of normal
operation by the WIPCO operators with
closed or open tertiary air ports.
The test program included short-
term tests over a range of burner settings
and operating conditions to establish
optimum settings and extended periods
of continuous monitoring while the unit
-------
Table 1. Initial Demonstration Site Characteristics
Parameter
Value
Boiler
Configuration
Capacity MCR
Peak
Burners
Type
Array
Capacity MCR
Throat Diameter
Spacing Horizontal
Vertical
Furnace
Construction
Depth
Width
Burner Zone Liberation Rate
(BZLR)
Fuel
Coal Type
Pulverizers Type
Number
Air Supply
Air Heater Type
Secondary Air Temperature
Draft
Windbox Depth
Burner Pressure Drop (nominal)
Single-Wall-Fired
98 x 103 kg/hr (215 x 103 Ib/hr)
111 x 103 kg/hr (245 x 103 Ib/hr)
Foster Wheeler Intervane
2x2
20 Thermal MW (69 x 1Q6 Btu/hr)
61 cm (24 in.)
182cm (71.5 in.)
198 cm (78.0 in.)
Membrane Wall
6.28 m (20.6 ft)
5.01 m (16.4ft)
492 Thermal kW/m2
156x103Btu/hr-ft2)
Indiana Bituminous
Foster Wheeler MB
2
Regenerative
265°C (510°F)
Pressurized
132 cm (52 in.)
8.9 cm H20 (3.5 in H20)
Core Air
Valve
Perforated Plate
Air Hoods
Inner
Register
Outer
Register
Firing
Face
Telescoping
Inner Nozzle
Cast Refractory
Exit
Coal Inlet
Figure 3.
Initial prototype DMB based on Foster Wheeler components (tertiary ports not
shown).
was operating by WIPCO following load
Measurements included boiler contro
room data, burner settings, emissions
boiler efficiency, and corrosion.
The large number of burner controls
on the DMBs allowed the burner to be
adjusted to provide wide ranges of flow
conditions, flame patterns, and hence
emissions and boiler efficiency. Minimum
NOX emissions were produced by
adjusting the burners to achieve: (1)
short flame standoff distance, (2) low
burner zone stoichiometry near the DMB
design point (70% TA), and (3) balanced
air and coal flows to the four burners.
Figure 5 shows data from several
tests over a range of burner zone
stoichiometries at two excess air levels.
NOX emissions were slightly higher for
the higher excess air level, but this effect
was small compared to the impact of
burner zone stoichiometry. NOX
emissions were minimized at low burner
zone stoichiometries as expected from
the research furnace tests.
Under controlled conditions, NOX
emissions as low as 259 ppm @ 0% 02
or 129 ng/J (0.29 lb/106 Btu) were
achieved at full load. This is a reduction
of 69% from the NOX emissions
measured during the 30-day baseline
test with the pre-NSPS Intervane
Burners. A wide range of burner settings,
was used. Burner zone stoichiometry
was in the range of 72 to 77% TA, and
excess Og was in the range of 2.87 to
3.94%. These low NOX operating
conditions were obtained over a 9-
month period and represent the best
operating conditions achieved on each
specific day.
Maintaining low-NOx emissions for
extended periods proved difficult for the
operators due to burner balance
problems. Based on all available data,
the burner balancing problems were
probably due to non-uniform
entrainment of tertiary air into some of
the flames. It is expected that the
operation of the burners with the
registers almost fully closed and the
specific tertiary air port array interact to
result in non-uniform entrainment of
tertiary air into the four flames. Figure 6
illustrates the three positions of tertiary
air mixing patterns.
Due to these burner balancing
problems, the extended tests were
conducted with more conservative burner
settings which resulted in higher NO*
emissions. Four extended tests were
conducted including two 30-day tests.
Table 2 compares the emissions
measured during 30-day tests with
baseline pre-NSPS burners and the
-------
Firing Face
Windbox
Cylindrical
Tertiary Control
Valves (9)
Valve
Actuators
(10)
Plenum
Sliding Tertiary
Control Valve (1)
Figure 4. Tertiary air ports for WIPCO DMBs.
Primary Air
Ducts (2)
DMBs. The DMBs reduced NOX
emissions by 49%. CO emissions and
the carbon content of the fly ash were
higher for the DMB. The hydrocarbon
emissions were negligible for both tests.
S02, SOa, COa, and total particulate
emissions are sensitive to fuel
composition, and the differences in the
table should not be attributed to burner
performance. The particulate carryover,
expressed as the particulate-to-coal
ash mass ratio, was similar for the two
tests.
Boiler performance data obtained
during the 30-day test are presented in
Table 3. Based on these and other data,
the DMBs had no discernible impact on
air heater outlet temperature and final
earn temperature. The only efficiency
.osses which can be impacted by the
burners are the dry gas and combustible
losses. The dry gas loss in Table 3 is
slightly higher for the DMB since the
boiler was operated at higher excess 62
during the test. Note that the DMBs
could be set up to operate at lower
excess C>2, and this would result in dry
gas losses comparable to the pre-NSPS
burners. Combustible loss was higher for
the DMBs, and this accounts for most of
the reduction in boiler efficiency.
To assess the impacts of the DMBs
on tube wall metal wastage,
comprehensive measurements were
made, including, installation/removal/-
analysis of four corrosion panels in the
furnace and field measurements of tube
wall thickness using ultrasonic techniques
over a 47 month period (12 months with
the pre-NSPS burners and 35 months
with the DMBs). The results of these
tests showed that the DMBs had no
discernible impact on tube wall metal
wastage.
Summary and Conclusions
A field evaluation of the DMB has
been completed on a 98 kg/hr (215 x 103
Ib/hr) front-wall-fired boiler. This DMB
installation was necessarily a retrofit
application. To meet the requirements of
the host boiler, some compromises in the
DMB design were required, especially
the tertiary air ports. In addition to burner
replacement, it was necessary to replace
the front wall, modify the windbox,
replace the burner management system,
and provide an alternate support
structure for the hopper. However, the
preferred DMB application is in new
boilers where it should be possible to
adjust the windbow and furnace wall
designs to avoid many of these
problems.
The field tests demonstrated that, in
comparison to the original equipment
pre-NSPS burners, the DMBs reduced
NOX emissions by about 50% under
routine operation and up to 70% under
carefully controlled conditions. There
was no impact on steam temperature, air
heater discharge temperature, boiler
capacity, or furnace tube wall metal
wastage. Boiler efficiency was lower with
the DMBs due principally to increased
carbon loss. This loss was offset partially
by reductions in fan power due to the low
windbox-to-furnace pressure dif-
ferential required by the DMBs.
Overall, the host boiler operators
have been pleased with the DMBs and
have elected to leave the DMBs in
operation rather than restoring the
original burners. At present, the host
boiler remains in commercial service with
the DMBs operating with the tertiary air
ports open but not at low burner zone
stoichiometry. Under this condition,
carbon loss is in the same range as with
the original burners.
-------
yuu
800
700
f 600
Q 500
O
SS 400
@
O 300
200
roo
0
I ' \
• 2.9-3.3% O2
-D 4.0-4.4% O2
.
-
• *~
— (0
0
a -L ^ ^
F? *§"
S
~
t t >
l.U
0.9
0.8
0.7
'0.6
0.4
0.3
0.2
0.1
0.0
65 70 75 80 85
Burner Zone Stoichiometry (Percent TA)
350
300
25°
Q 200
150
O
O
roo
so
• %
•k-P *
S3^
°D EbfcP
65 70 75 SO 55
Burner Zone Stoichiometry (Percent TA)
Figure 5. Effect of burner zone Stoichiometry on emissions for the DMBs in the host boiler.
Small Flames
(Leanj
Large Flames
(Rich)
Small
Flame
(Lean)
Large Flame
(Rich)
A. Unstaged with balanced flames
C. Staged with imbalanced flames
B. Staged with balanced flames
Figure 6. Tertiary air mixing patterns.
-------
Table 2. Emissions Measured During 30-Day Tests of the Host Boiler
Averages Over 30-Dav Period
Load
02 (%)
NOX @ 0% O2 (ppm)
CO @ 0% O2 (ppm)
SO2 @ 0% 02 (ppm)
C02 @ 0% O2 (%)
Short Term Measurements
SO3 @ 0% 02 (ppm)
HC as C3 @ 0% O2 (ppm)
Total Particulate @ 0% O2 (gr/scf)
Particulate/Coal Ash (mass ratio)
Carbon in Fly Ash (%)
Baseline
Pre-NSPS
Burners
18.5
3.6
847
37
3299
18.5
25
<1
5.93
0.828
3.85
DMB
16.6
4.0
432
85
3682
18.3
47
<1
5.95
0.734
6.62
Table 3. Boiler Performance Measured During 30-Day Tests of the Host
Boiler
Baseline
Pre-NSPS
Burners
DMB
Operating Conditions
Steam Flow
Steam Temp
Excess Air (%)
Air Heater Gas Outlet
Temp, Undiluted
Coal Calorific Value
Efficiency Losses (%)
Dry Gas
H2O from Fuel
H2O in Air
Combustibles
Radiation
Manufacturer's Margin
Boiler Efficiency (%)
81,633 kg/hr
180,000 Ib/hr
473°C
884 °F
25.4
186°C
367 °F
5347 kcal/kg
9622 Btu/lb
6.308
5.605
0.158
0.370
0.436
1.500
85.623
81,905 kg/hr
180,600 Ib/hr
466 "C
870°F
26.5
207 °C
404 °F
5,709 kcal/kg
10,274 Btu/lb
6.478
5.419
0.162
1.070
0.440
1.500
84.931
------- |