United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-90/008 Aug. 1990
&EPA Project Summary
Verification of Simplified
Procedures for Site-Specific
SO2 and NOX Control Cost
Estimates
Thomas E. Emmel and Mehdi Maibodi
Detailed retrofit studies were con-
ducted for 12 coal-fired plants in Ohio,
Kentucky, and the Tennessee Valley
Authority system. Because detailed
studies are expensive and time-consum-
ing, the results from the 12-plant study
were used to develop simplified proce-
dures for estimating sulfur dioxide (SO?)
and nitrogen oxides (NOx) retrofit control
costs and performance for 200 SOa-emit-
ting coal-fired power plants in the 31-
state eastern region: The simplified
procedures require less time, data and!
preparation effort. This report docu-
ments the results of an evaluation to
verify the accuracy of the simplified pro-
cedures. The evaluation compared the
costs for a number of plants estimated
using the simplified procedures to costs
estimated using detailed procedures, ac-
tual retrofit costs, and more detailed cost
estimates provided by utility companies.
Based on the evaluation, recommenda-
tions for changes to the simplified proce-
dures were developed. Control
technologies addressed in this report are
conventional lime/limestone flue gas
desulfurization, lime spray drying, fur-
nace sorbent injection, duct spray drying,
coal switching, physical coal cleaning,
and selective catalytic reduction. In
general, it was found that the simplified
procedures can be used to generate im-
proved cost performance estimates
based on generally available information:
U. S. Geological Survey photographs and
Energy Information Agency Form 767
data.
This Project Summary was developed
by EPA's Air and Energy Engineering Re-
search 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
The objective of the National Acid
Precipitation Assessment Program
(NAPAP); retrofit cost project is to develop
improved (site-specific) cost estimates for
retrofitting sulfur dioxide (SOa) and nitrogen
oxide (NOx) controls on 200 of the largest
emitting coal-fired utility power plants. This
project was conducted in the four phases
depicted in Figure 1. In Phase I, procedures
were developed based on plant-specific in-
formation obtained from site visits (detailed
procedures). Twelve plants in Kentucky,
Ohio, and the Tennessee Valley Authority
(TVA) system were evaluated underthis pro-
gram phase, and the procedures and results
were reviewed by a Technical Advisory
Committee and the participating utility com-
panies.
Because detailed studies are expensive
and time consuming, the results of the 12-
plant study were used to develop simplified
procedures for estimating SOg/NOx control
technology retrofit cost/performance for 200
SOg-emitting coal-fired power plants in the
31-state eastern region. In simplifying the
detailed procedures, the major retrofit factor
"drivers" were identified and only the retrofit
difficulties associated with these drivers
-------
PHASE 1
Develop Detailed Procedures
PHASE II
Select 12 Plants and Develop Cost/Performance
Estimates
Revise Procedures and Cost Estimates and
Develop Simplified Procedures
PHASE III
Evaluate 50 Plants
Verify Simplified Procedures with Plant Visits
Modify Procedures and Evaluate Remaining 138
Plants
PHASE IV
Finalize Top 200 Study Report
TECHNICAL ADVISORY COMMITTEE
Electric Power Research Institute
U.S. Environmental Protection Agency
U.S. Department of Energy
Utility Air Regulatory Group
Tennessee Valley Authority
National Resources Defense Council
Vendors
UTILITY COMPANIES
Ohio Edison
American Electric Power
Ohio Electric Utility
Tennessee Valley Authority
Kentucky Utilities
Union Electric
Cincinnati Gas and Electric
All Represented
Utility Companies
Figure 1. Top 200 plant study technical approach.
(e.g., site access and congestion) were es-
timated. To estimate retrofit factors using
the simplified procedures, a plot plan and/or
aerial photograph were used instead of in-
formation gathered during a site visit. How-
ever, assumptions were made to streamline
these procedures. For example, average
underground obstructions as well as some
scope adjustments were always used (e.g.,
chimney liner, demolition and relocation,
draft controls, and new rails) unless one or
more of these items were unnecessary or
Incorrect. After developing cost/perfor-
mance estimates for 50 plants using the
simplified procedures, the procedures were
evaluated to verify their accuracy. The
results of the simplified procedures were
compared to: results obtained by using the
detailed procedures, actual retrofit cost, and
more detailed estimates provided by par-
ticipating utility companies.To accomplish
this objective, 6 of the 50 plants were
selected to evaluate the simplified proce-
dures. Additionally, for 17 boilers, detailed
cost estimates or reported actual costs for
retrofit FGD systems were available to com-
pare to the detailed and simplified procedure
results generated under this program.
The following procedures were used to
verify the simplified procedures for the six
plants selected for site visits: 'conduct plant
visits and collect site-specific^data, develop
boiler/control-specific retrofit difficulty fac-
tors, develop boiler/plant-specific cost and
performance estimatesforthe SOa and NOx
controls selected using the detailed proce-
dures, and compare these results with
results developed using simplified proce-
dures. Utility interest in participation in the
-------
study was a major consideration in plant
selection. Table 1 presents the boiler char-
acteristics for the six plants visited to verify
and quantify the accuracy of the simplified
procedures.
Simplified Procedures
Verification Methodology
Table 2 presents the SOa/NOx control
technologies that are being evaluated under
the NAPAP retrofit control cost study. As
this table shows, detailed and simplified pro-
cedures were developed for the lime/lime-
stone (L/LS) and lime spray drying (LSD)
flue gas desulfurization (FGD) technologies.
However, for the other technologies only
simplified procedures were developed. The
focus of the simplified procedure verification
Table 1. Boiler Characteristics of the Plants Evaluated
Plant
Name
Rush Island
Sioux
Meramec
Labadie
Beckjord
Miami Fort
Boiler
Number
1
2
1
2
1
2
3
4
1
2
3
4
1
2
3
4
5
6
5
6
7
Gross Operating
Capacity (MW)S
621
621
550
550
137
137
289
359
620
620
620
620
100
100
135
162
255
449
85
175
524
Installation
Date
1976
1977
1967
1967
1953
1954
1959
1961
1970
1970
1973
1973
1952
1952
1954
1958
1962
1969
1949
1960
1975
Capacity
Factor
(%)
42
58
43
32
10
12
11
11
59
52
57
58
6
8
10
22
29
57
40
71
67
Firing
Type
Tangential
Tangential
Cyclone
Cyclone
Tangential
Tangential
Wall
Wall
Tangential
Tangential
Tangential
Tangential
Tangential
Tangential
Wall
Tangential
Tangential
Tangential
Wall
Tangential
Wall
Coal
Sulfur
(%)
1.1
1.1
1.7
1.7
1.1
1.1
1.1
1.1
2.6
2.6
2.6
2.6
1.0
1.0
1.0
1.0
2.5
2.5
1.0
1.0
3.0
a Maximum unit design capability.
Table 2. Emission Control Technologies Evaluated
Development Status
Limited Near
Species Controlled Commercial Commercial
SOz A/Ox Commercial Experience Demonstration
Lime/Limestone (L/LS) flue
gas desulfurization (FGD)
Additive enhanced LfLS-FGD
Lime spray drying (LSD) FGD a
Physical coal cleaning (PCC)
Coal switching and blending (CS/B)
Low-NOx combustion (LNC)
Furnace sorbent injection
(FSI) with humidification
Duct spray drying (DSD)
Natural gas reburning (NGR) b
Selective catalytic reduction (SCR)
X X
X X
X X X
X X
X X
X X
X X
X X
XX X
X X
Type of Procedures
Detailed Simplified
X X
X X
X X
X
X
X
X
X
X
X
a Commercial on low sulfur coals, demonstrated at pilot scale on high sulfur coals.
b For wet bottom boilers and other boilers where LNC is not applicable.
-------
evaluation was twofold. First, the accuracy
of the publicly available data and the as-
sumptions used to develop the inputs to the
simplified procedures were reviewed.
These inputs were then compared to the
inputs developed using the more detailed
data obtained from the site visits.
Second, for the FGD technologies, retrofit
factors and indirect costs were developed
using the detailed procedures. These
retrofit factors and indirect costs were com-
pared to the simplified procedure results
developed before the site visits as well as the
simplified procedure results developed with
the site visit information. This was done so
that differences in the detailed procedures
and the simplified procedures could be clas-
sified as being due to errors in the quality of
the publicly available information or due to
the different level of detail between the two
procedures.
Additionally, when actual cost data or
more detailed cost estimates were available,
these costs were compared to the costs
estimated using the simplified and detailed
procedures. Actual retrofit costs were ob-
tained from the Energy Information Agency
(EIA)-767 forms for seven units at six plants.
Detailed engineering cost estimates were
obtained for 10 units at 6 plants.
Cost estimates developed under this
study were generated using the Integrated
Air Pollution Control System (IAPCS) cost
model. These cost estimates were made to
reflect site-specific retrofit impacts by input-
ting capital cost multipliers, scope adder
costs, and capital cost indirect factors that
were developed using the simplified and
detailed procedures.
Lime/Limestone FGD Procedures
Comparison of Simplified and
Detailed Procedure Results
For L/LS and LSD-FGD, retrofit factors and
cost estimates were developed using both
the simplified and detailed procedures
based on the detailed information collected
during the site visits. The retrofit factors and
cost estimates were then compared to those
previously developed using the simplified
procedures and based on EIA-767 data and
plant plot plans or aerial photographs.
The percent differences between the
simplified and detailed procedures based on
the data obtained from the site visits are
small. The difference for L/LS-FG5D varied
between -6 and +5% with an average of
2.7%. The difference for LSD-FGD varied
between <1 and 5% with an average of
1.9%. These differences are due to simplify-
ing assumptions made when developing the
simplified procedures.
The differences between the detailed pro-
cedures based on site visit data and the
simplified procedures based on publicly
available data were also determined. The
difference for L/LS-FGD varied from -9 to
+16% with an average of 8.8%. The dif-
ference for LSD-FGD varied from -24 to
+17% with an average of 9.8%.
The difference between the simplified and
detailed procedure results is due to proce-
dural differences and errors in assumptions
due to having less detailed and incorrect
data. The procedural differences result from
simplifying the detailed procedures by using
average factors for flue gas duct distance,
underground obstruction difficulties, and
scope adder costs. The incorrect data
resulted in changing the need for wet to dry
ash handling at the four Missouri plants.
Additionally, the absorber locations for one
of the Missouri plants, Meramec units 1-4,
were changed. Duct work distance and ab-
sorber/flue gas handling area access/con-
gestion changes were also made at Labadie
and Rush Island when more detailed infor-
mation was available. A comparison of the
simplified procedure results, developed
before and refined after the site visits for the
two Ohio plants, revealed no significant dif-
ferences.
Based on these results, it was concluded
that the use of EIA-767 form data and U.S.
Geological Survey (USGS) photographs
can be used to accurately determine the
retrofit difficulty for FGD. However, ex-
perience is needed to accurately evaluate
these sources of information. Additionally,
communications with the plantgenerally can
be very helpful in confirming the best loca-
tion of the FGD process areas and retrofit
difficulty issues.
Comparison to Other Cpst
Estimates
Capital cost estimates using the simplified
procedures were developed and compared
to estimates developed by architectural and
engineering (A&E) firms for six plants.
These cost estimates were obtained through
the data collection effort for the program.
Table 3 summarizes the results of this com-
parison. As can be seen, the simplified pro-
cedure results compare well with the A&E
estimates. The difference varied from -12 to
+20%, well within the ±30% accuracy of the
Electric Power Research Institute (EPRI)
FGD cost estimating guidelines.
Table 3. Comparison of Capital Cost Estimatese
Plant and Unit
Asbury 1
Sibley3
Thomas Hill 1
Thomas Hill 2
New Madrid 1
New Madrid 2
Beckjord 5 + 6
Miami Fort 6
Miami Fort 7
A&E Estimate Simplified Procedure
($/kW) Estimate ($/kW)
274
199
363
282
259
215
203
276
210
264
222
355
297
221
229
202
283
169
Absolute b
Difference (%)
4
12
2
5
15
7
<1
3
20
In 1987 dollars.
(A&E estimate - simplified procedure estimate)/A&E estimate x 100.
-------
Comparison of IAPCS Cost
Estimates to Actual Installed
Costs
A review of the Flue Gas Desulfurization
Information System identified 11 retrofit FGD
systems that came on-line after 1980. For
seven of these units, actual costs reported
in EIA-767 forms and aerial photographs
were available and were used to compare
the simplified procedure cost estimates to
actual reported costs. For these seven
units, FGD system cost estimates have been
developed using the IAPCS cost model with
retrofit factors developed from the simplified
or detailed procedures for the following
units: Paradise 1 and 2, Widows Creek 7,
Mill Creek 1 and 2, and Four Corners 4 and
5. Table 4 summarizes the results of this
effort with the actual costs being escalated
to the same year dollars as the cost es-
timates (January 1988).
The cost estimates developed for Four
Corners, Paradise, and Mill Creek, unit 1,
compare well with the reported actual costs.
However, the cost estimates for the Mill
Creek, unit 2 and Widows Creek, unit 7, differ
significantly. At least two reasons are pos-
sible to account for the differences:
1. Mill Creek units 1 and 2 share a common
feed and waste handling facility, as do
Widows Creek units 7 and 8.
2. The Mill Creek and Widows Creek units
were not current state-of-the-technology
designs and have undergone significant
design changes since start-up.
All of these factors would reduce the ac-
tual cost compared to the estimates
developed based on current new source
performance standard (NSPS) design.
However, despite these differences, the
reported costs for Mill Creek 2 and Widows
Creek 7 appear to be very low. Note that the
most recent FGD system actual costs com-
pare favorably to the IAPCS estimates and
that the cost of the more recent units ranges
from $180 to $300/kW. Additional com-
parisons should be made for the other four
units at the Cromby, Eddystone, and
Mitchell plants if sufficient data are obtained
to conduct the evaluation.
Lime Spray Drying (LSD) FGD
With Reuse of Electrostatic
Precipitators (ESPs)
As part of the simplified procedure
methodology for LSD-FGD, it is assumed
that the existing ESP will be used for particu-
late control if the specific collection area
(SCA) is greater than 225 sq ft of plate area
per 1000 acfm (28 actual m3/min) of flue gas.
This was the case for all of the Missouri
plants and for several of the Ohio units.
However, for many of these units, reusing
the ESP would result in very difficult and
lengthy flue gas duct runs from the outlet of
the air heater to the spray dryer absorbers
and back again to the ESP inlet. For units
that have space immediately adjacent to the
air heater and ESP, this option is more prac-
tical. Table 5 presents the ESP SCA and
comments on applicability of this technol-
ogy for the six plants visited.
Coal Switching (CS) and
Physical Coal Cleaning (PCC)
To estimate the cost of CS, the simplified
procedures used one of the two low sulfur
base coals contained in the IAPCS cost
model. One of the coals is a low sulfur West
Virginia bituminous coal and the other is a
Montana subbituminous coal. The currently
used coal characteristics and cost are input
to the model, and the model estimates the
cost of CS based on the user specifying the
fuel price differential between the current
coal and the future price of one of the two
low sulfur base coals. Future coal price
differentials of $5 and $15/ton were input to
the model to span the range of possible low
sulfur coal prices over high sulfur coals.
IAPCS cost estimates also include the
cost of ESP upgrades and changes in waste
disposal costs due to differences in coal ash
content. The model does not evaluate the
impact of CS on the following boiler operat-
ing parameters: pulverizer capacity, slag-
ging, fouling, erosion, and flue gas flow
rates. Also, costs for additional coal receiv-
ing, storage, and handling facilities are not
included. These factors are discussed
qualitatively for each plant and boiler situa-
tion.
Under this effort, the primary objective
was to obtain additional information from the
six plants visited to better identify and quan-
tify boiler and plant parameters that would be
negatively impacted by CS. Table 6 sum-
marizes the results of this effort. As can be
seen, the only major issue identified was
switching to a subbituminous coal at the
Sioux plant. The low sulfur subbituminous
IAPCS base coal option was chosen be-
cause the plant currently uses a low sulfur
subbituminous coal. The procedure
methodology assumes that the base coal
should be similar to the low sulfur coals
currently being used by the plant. However,
completely switching a boiler designed to
fire a bituminous coal to a subbituminous
coal would result in a major unit derating.
Currently, the CS methodology does not
quantitatively address capital costs as-
sociated with coal receiving, storage, and
handling facilities that may be required to
blend coals on-site or to bring coal in by a
different transportation method. For the
Labadie plant, the cost to upgrade its
facilities to blend coals was on the order of
$22.5 million (1987 dollars). However, the
cost of upgrading facilities would typically be
much less than a $5/ton impact on the coal
fuel price differential.
Based on the six plant evaluation of the
simplified procedures, a more careful ap-
proach should be made in evaluating CS to
subbituminous coals on boilers designed
for firing bituminous coals. In general,
Tab/e 4. Comparison of Retrofit Factors and Estimated and Reported Costs
Plant and Unit
Paradise 1 or 2
Four Corners 4 a
Four Corners 5 a
Mill Creek 1
Mill Creek 2
Widows Creek 7
Retrofit IAPCS Estimated b
Factor Cost($/kW)
1.35
1.66
1.66
1.26
1.26
1.24
182
316
316
180
180
170
Reported Actual
Cost ($/kW)
184
303
303
173
75
140
Absolute °
Difference (%)
1
4
4
4
140
21
a Cost of fabric filters are also included in the FGD cost.
b Includes allowance for funds during construction.
0 (Actual cost- estimatedcost)/actual costx 100.
-------
Table 5. Summary of LSD-FGD ESP Reuse Applicability
Plant Name Unit ESPSCAa Comments
Applicability
Stoux 1 and 2 268
Labadle land 4 319
Labadle 2 and 3 319
Rush Island 1 and 2
Meramec
Moramoc
1-3
279
492-522
545
Marginal SCA; space available on one side;
reasonable accessibility to existing flue gas ducting.
Good ESP performance; space available on one side;
access to existing flue gas ducting is difficult.
Good ESP performance; space not available at sides or
behind chimney; access to existing flue gas ducting
is very difficult.
Good ESP performance but marginal SCA; reasonable
access to existing flue gas ducting; space available on
one side and behind chimney.
Good ESP performance and SCA; long duct runs and difficult
access to existing flue gas ducting because space is only
available on far side of unit 4 for absorbers.
Good ESP performance and SCA; moderate duct runs and
accessibility to existing flue gas ducting; space
available for absorbers at one side of unit.
Marginal due to SCA
Moderate
Marginal due to SCA
Mariginal due to SCA
Marginal (good, if [
absorbers are located
where original ESPs are)
Good :
Beckjord
Miami Fort
1-4
5
6
5
6
7
205-244
247
646
354
205
275
Marginal ESPs; very poor access for ducting.
Same as above.
Good candidate; easy access.
Poor access for duct runs.
Very small ESP SCA.
Marginal ESP SCA; space available on both sides.
Poor
Poor
Good
Marginal
Poor
Marginal due to SCA
' Electrostatic precipitator (ESP) specific collection area (SCA) in terms of square feet of area per 1000 acfm.
Table 6. Summary Of Coal Switching Issues
Plant Name
Units
Coal Switching and Blending Issues
Sioux 1-2 Minor boiler derate up to 40% SB3, major derate >50%
SB.
Labadie 1-4 Switch coal may reduce severe slagging problems at
high load but will increase ash disposal costs. Switching
could also increase slagging problems and cause unit
derates.
Rush Island 1-2 Low SOz emission reduction potential. Noon-site
facilities for blending.
Meramec 1-4 Low SOa emission reduction potential. No on-site
facilites for blending.
Beckjord 1-6 Low sulfur coal on units 1-4. No on-site blending
facilities for units 5-6.
Miami Fort 5-7 Compliance coal on unit 8. No on-site blending facilities
for units 5-7.
* SB = subbituminous.
boilers designed for bituminous coals
should continue to burn the same percent-
age of bituminous coal as currently is being
fired.
Due to the lack of coal washability data, it
has not been possible to evaluate quantita-
tively the cost of deep PCC. As such, the
costs generated under this study are very
qualitative. However, no feedback has been
given to date regarding the costs and per-
cent removals projected with the current
methodology.
Selective Catalytic Reduction
(SCR)
Based on the information obtained during
the plant visit, the assumptions used to
develop the retrofit factors and scope ad-
ders for SCR were reviewed for accuracy.
The most important parameters that will im-
pact the cost of retrofitting SCR are: reactor
access and congestion, flue gas duct length
to and from the reactor, and demolition costs
of existing equipment and general facilities
percentage.These parameters are impacted
by the location chosen for the SCR reactor.
Table 7 summarizes the changes made to
access/congestion difficulty, scope adder
costs, and general facilities percentages due
to the site visits for the four Missouri plants.
The major problem with the simplified proce-
dures identified by this analysis is selecting
-------
the correct location based on the informa-
tion on hand. For the Meramec and Labadie
plants, inaccuracies in the available plot plan
or the poor resolution of the aerial
photograph resulted in incorrectly locating
the reactors. For the two Ohio plants, no
changes to the reactor locations, duct run
lengths, and access/congestion difficulty
were needed for Miami Fort. However, for
Beckjord units 1-4, the reactor location was
changed, resulting in significant increases in
duct length.
One option used in the study for several
plants has been to call the plant and discuss
potential equipment locations. This has
been done several times when it was ap-
parent that the quality of the aerial
photograph or plot plan was poor. General-
ly, plant personnel have been cooperative.
In the future, it may be desirable to contact
every plant after reviewing the available data.
Sorbent Injection Technologies
Based on the information obtained during
the plant visit, the assumptions used to
develop the retrofit factors and scope ad-
ders for the sorbent injection technologies
were reviewed for accuracy. The most im-
portant parameters that will impact the cost
and performance of the sorbent injection
technologies are:
• paniculate control device size, perfor-
mance, and difficulty of upgrade (ac-
cess/congestion associated with adding
plate area);
• the need to convert a wet ash handling
system to a dry handling system; and
• sufficient flue gas residence time between
the air heater and particulate control
device to allow for duct spray drying
(DSD), droplet drying (2 seconds of
straight duct run), or humidification with
furnace sorbent injection (FSI).
The impact of superheat and economizer
gas lane pluggage for FSI was not evaluated
under this program because of the limited
commercial demonstration data and boiler
information.
Table 8 summarizes the results of the
changes made to the simplified procedure
inputs after the site visits. In general,, the
overall qualitative determination of the ap-
plicability of the sorbent injection tech-
nologies was the same before and after the
site visit, the major differences being Be-
ckjord, units 1-5, and Miami Fort, unit 6.
These units have small ESP SCAs, and the
ducting configuration for units 1 -5 was incor-
rectly interpreted. As a result, because
these units have a very short flue gas
residence time between the air heater and
the ESP inlet and small ESPs, the ap-
plicability of any of the sorbent injection
technologies is questionable.
With regard to factors that directly impact
the cost estimates, the access/congestion
difficulty factor changed for 7 of the 21 units
and the wet to dry ash conversion assump-
tion changed for 12 of the 21 units. As
discussed previously, the error in the ash
conversion assumption was due to a
misunderstanding. The EIA-767 forms ac-
curately indicate that the Missouri units cur-
rently use wet ash handling/disposal
systems.
Based on these findings, it appears that
the errors in the procedure methodology are
related to incorrectly interpreting the data
TaWe 7. Summary of Changes Made to SCR Retrofit
Plant Name
Units
Reason for Changes
Sioux 1-2 Scope adder cost changes due to longer duct run.
Labadie 1-4 Change in location of reactor resulted in changes to
access/congestion, general facilities, and duct run
lengths for all Labadie units.
Rush Island 1-2 No major changes were made to the SCR retrofit
compared to the results obtained prior to the site visit.
Meramec 1-4 Changes in location' of reactors resulted in. significant
changes in access/congestion and duct rurr lengths for
all Meramec units.
Table 8. Summary of Changes or Sorbent Injection Technologies
Applicability Before Visit Applicability After Visit
Plant Name
Unit
DSD
FSI
DSD
Sioux
Labadie
Rush Island
Meramec
Beckjord
Miami Fort
1-2
1-4
1-2
1-2
3-4
1-5
6
5
6
7
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Poor
Good
Good
Poor
Poor
Good
Good
Good
Good
Poor
Poor
Very good
Poor
Poor
Good
Good
Poor to good b
Moderate
Good
Poor
Poor
Good
Poor
Poor
Poor
Does not consider superheat/economizer gas line pluggage due to insufficient data.
Part of flue gas bypassed old ESPs resulting in adequate residence time for 50% of gas flow.
-------
available. It appears that the publicly avail-
able data are sufficient to accurately
evaluate the retrofit difficulty of the sorbent
Injection data for most plants. However, as
mentioned earlier, contacting the plant to
verify the accuracy of these data and to
confirm the accuracy of basic assumptions
should reduce the potential for incorrectly
Interpreting available data. This is particular-
ly true for small, older units because of the
resolution of the photograph and the
likelihood that unusual ESP flue gas con-
figurations exist due to the retrofit of addi-
tional ESP plate area. The Beckjord units 1 -5
and Miami Fort unit 6 are good examples of
the problem associated with unusual flue
gas configurations due to the retrofit of ad-
ditional ESP plate area. Additionally, a
Department of Energy database containing
duct and flue gas information upstream of
the ESP has been made available to this
study.
Economic and Financial
Assumptions
At the outset of the NAPAP Task Group I
effort, a decision was made to use economic
and financial data consistent with accepted
industry practices. The accepted standard
for the electric utility industry is published in
the EPRI's Technical Assessment
Guidelines (TAG). The EPRI TAG provides
the economic factors and financial data on
which the cost estimating procedures used
in the electric utility industry are based.
Table 9 presents the 1986 TAG values.
Retrofitting a plant with SO2 and/or NOx
control technologies with' possibly more
stringent control limits could cause addition-
al costs for compliance that are not reflected
in this study.
Table 9. 1986 EPRI TAG Values - Capital and Financial Structure
Type of Security
Percent of total
Current Dollar0
Cost
Return
Debt (Bonds)
Preferred
Common stock
Discount rate, $/yr
50
15
35
11.0
11.5
15.3
5.5
1.7
5.3
12.5
Inflation rate:
Federal and state income tax rate:
Investment tax credit:
Property taxes and insurance:
Book life
6.0%/yr
38.0%
0.0%
2.0%/yr
30 yr
Item
Unit Cost Data
January 1985 value
Units
Operating labor
Water (river)
Lime
Limestone
Land 6,500 $/acre
Disposal Charges
Sludge
Dry, granular solids
Gypsum disposal -
By-Product Credits
Sulfuric acid
Sulfur
Ammonia (anhydrous)
Special Items
Electric power (in plant)
Steam (in plant)
0-70 psia b
70-250 psia
250-2400 psia
19.70
0.60
65
15
9.25
8.0
4.75
50
75
150
5.0
2.85
3.50
5.30
$/person hour
$/1,OOOgalb
$/ton "
$/ton
$/ton (dry basis)
$/ton
$/ton
$/ton
$/long ton
$/ton
0/kWh
$/1000lbl
$/1000lb
$/1000lb
Levelization Factors
O&M
Carrying charges, %
1.75
17.5
8 For constant dollars, inflation rate is set to zero.
b 1 acre = 4047 mz; 1 gal. = 3.79 L; 11b = 0.45 kg; 1 long ton = 1.02 metric tons;
psia = psig + 14.7 (1 psi = 6.89 kPa; 1 aim = 101.3 kPa); 1 ton= 0.907 metric ton.
-------
The economic and financial assumptions
of the EPRI TAG, as applied in this stage, are:
1. Cost-estimating premises adhere to the
cost methodology described in the 1986
EPRI TAG.
2. The indirect capital cost factors were as-
signed to each technology in accordance
with the EPRI TAG. These values were
varied in accordance with site-specific
conditions and are presented in Table 10.
3. Allowance for funds during construction
(AFDC) is estimated by adjusting the total
plant cost by an allowance factor (AF)
that is a function of the idealized con-
struction period and economic
parameters.
4. For annual operating costs, the unit costs
for consumables per the EPRI TAG are
summarized in Table 9.
5. Total annual maintenance costs are es-
timated per the 1986 EPRI TAG as a per-
centage of installed capital cost depend-
ing on the nature of the processing con-
ditions and the type of design. The factors
assigned to the various technologies are
summarized in Table 11.
6. Economic premises for existing electric
power generating plants were updated by
EPRI and released as a supplement is-
sued May 1, 1983. Revised schedules
were published for book life, tax life, and
levelization factors. For power plants in
operation before January 1,1979, Internal
Revenue Service (IRS) Code Section 169
allows 5-year tax depreciation of new
identifiable pollution control facilities
completed or acquired after December
31, 1982. This depreciation schedule is
applied on a straight-line basis. Book life
of retrofitted equipment is equal to the
years of remaining life of the power plant
rounded off to the nearest 5-year incre-
ment. All of these items are identical in
the 1986 TAG.
7. The financial and economic premises
significantly influence the levelization fac-
tors calculated for operating and main-
tenance (O&M) and carrying charges.
Using the 1986 guidelines recommended
by EPRI-a 12.5% discount rate (or
weighted cost of capital), 6.0% inflation
rate (long-term average), 30-year book
life (existing facility), and 20-year tax life
(straight-line depreciation)-the leveliza-
tion factors computed for O&M and capi-
tal carrying charges are 1.75 and 0.175
for current dollars and 1.0 and 0.105 for
constant dollars, respectively.
8. All costs are presented in current and
constant dollars (in the current dollars,
the effect of escalation due to inflation is
accounted for) and reflect June 1988 dol-
lars. Capital costs are escalated using
Chemical Engineering indices. Current
dollar costs account for inflation; con-
stant dollar costs do not.
Table 10. Nominal Indirect Cost Schedule
Indirect Component a PCC LNC
General facilities, %
Engineering and home
office fees, %
Project contingency, %
Process contingency, %
Sales tax, %
Royalty allowance, %
Preproduction cost0
Inventory capital d
Initial catalyst e
Idealized construction period, yrf
0
0
0
0
0
0
c
d
0
0
10
10
30
10
0
0
c
d
0
1
FSI
10
10
30
20
0
0
c
d
0
1
NGR
10
10
30
10
0
0
c
d
0
1
SCR
10
10
30
20
0
0.5
c
d
0
1
LSD
10
10
30
4.3
0
0
c
d
0
3
DSD
10
10
30
30
0
0
c
d
0
1
ESP
10
10
30
0
0
0
c
d
0
1
FFb
10
10
30
0
0
0
c
d
0
1
FGD
10
10
30
T.4
0
0
c
d
0
3
a Applied to process capital except as noted.
bFF = fabric filter.
° 1 month of fixed operating cost.
1 month of variable operating cost.
2 percent of total plant investment
60-day supply of consumables.
e SCR catalyst costs are estimated based on unit size and desired NOX removal efficiency.
1 Used for estimating allowance for funds during construction (AFDC).
Table 11. Maintenance Cost Factors
PCC
LNC
FSI
NGR
SCR
LSD
DSD
ESP
FF
FGD
EPRI schedule
(percent of total process capital).
-------
Conclusions
Based on the six-plant evaluation effort to
verify the accuracy of the simplified proce-
dures, the following changes to the proce-
dures were made:
1. After receiving the publicly available infor-
matlon, identifying the equipment
layouts, and selecting inputs to the
simplified procedures to develop retrofit
factors and scope adder costs, the plant
was contacted to confirm the accuracy of
the data and assumptions.
2. To continue to verify the accuracy of the
procedures and cost estimates
generated, the plant evaluation and cost
estimates were made available to the
plant for review and comments. Ap-
propriate comments and comparative
cost estimates were incorporated into the
report. Additionally, where actual costs
of retrofit are available, these costs were
compared to estimates generated using
the simplified procedures.
3. Evaluation of retrofitting LSD-FGD was
eliminated for units that have marginally
sized ESPs (SCA = 225-300 sq ft/1000
acfm) and do not have space for plate
area addition.
4. Evaluation of PCC was eliminated except
for mine mouth plants with moderate to
high sulfur coal, because the emission
reductions that are achievable are low,
and insufficient data are available to ac-
curately estimate deep coal cleaning
costs.
5. Because the current CS methodology as-
sumes total switching to a low sulfur coal,
units designed to fire bituminous coals
were not switched to a subbituminous
coal unless the plant indicates that this is
a reasonable option and the cost of unit
derating is included in the cost of CS.
6. Hot side SCR was evaluated for units that
have space near the boiler.
7. Evaluation of retrofitting FSI and DSD
technologies were eliminated for units
that have marginally sized ESPs (SCA =
225-300 sq ft/1000 acfm) ^and do not have
space available for plate area addition
and sufficient space between the air
heater and particulate control device to
allow for DSD droplet drying or
humidification with FSI.
Making these changes to the simplified
procedures improves the accuracy of the
cost estimates, ensure continued feedback
from industry regarding the accuracy of the
procedures/estimates, and eliminates the
evaluation of technologies that are not likely
to be used, that have limited SOa reduction
potential, or those for which insufficient data
are available to accurately develop cost and
performance estimates.
T. Emme/ and M. Maibodi are with Radian Corporation, Research Triangle Park, NC
27709.
Norman Kaplan Is the EPA Project Officer (see below).
The complete report, entitled "Verification of Simplified Procedures for Site-Specific
S02 and NOx Control Cost Estimates," (Order No. PB 90-187 261/AS; Cost:
$23.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
10
-------
-------
m
1
I
o
o
8
CD
3
§
a
o"
Q
,
3!
q|
x I
I
8
§!
------- |