United States EPA-600/7-81-127
Environmental Protection
Agency July 1981
v>EPA Research and
Development
ENVIRONMENTAL ASSESSMENT OF
COMBUSTION MODIFICATION CONTROLS
FOR STATIONARY
INTERNAL COMBUSTION ENGINES
Prepared for
Office of Air Quality Planning and Standards
Prepared by
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-81-127
July 1981
ENVIRONMENTAL ASSESSMENT OF COMBUSTION MODIFICATION
CONTROLS FOR STATIONARY INTERNAL COMBUSTION ENGINES
by
I. Lips, J. A. Gotterba, and K. J. Lim
Acurex Corporation
Energy & Environmental Division
485 Clyde Avenue
Mountain View, California 94042
Contract No. 68-02-2160
Program Element No. EHE 624A
Prepared for
EPA Project Officer -- J. S. Bowen
EPA Deputy Project Officer -- R. E. Hall
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park
North Carolina 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, DC 20460
-------
DISCLAIMER
This Final Report was furnished to the U.S. Environmental
Protection Agency by Acurex Corporation, Energy & Environmental Division,
Mountain View, California 94042, in partial fulfillment of
Contract No. 68-02-2160. The opinions, findings, and conclusions
expressed are those of the authors and not necessarily those of the
Environmental Protection Agency or of cooperating agencies. Mention of
company or product names is not to be considered as an endorsement by the
Environmental Protection Agency.
-------
PREFACE
This is the fourth in a series of five special reports to be
documented in the "Environmental Assessment of Stationary Source NO
A
Combustion Modification Technologies" (NO EA). Specifically, this
A
report documents the environmental assessment of combustion modification
controls applied to stationary reciprocating internal combustion engines.
The NO EA, a 36-month program which began in July 1976, is sponsored by
A
the Combustion Research Branch of the Industrial and Environmental
Research Laboratory of EPA (IERL-RTP). The program has two main objectives:
(1) to identify the multimedia environmental impact of stationary
combustion sources and combustion modification controls applied to these
sources, and (2) to identify the most cost-effective, environmentally
sound combustion modification controls for attaining and maintaining
current and projected NO- air quality standards to the year 2000.
The NO EA will assess the following combination of process
/\
parameters and environmental impacts:
Major fuel combustion stationary NO sources: utility
A
boilers, industrial boilers, gas turbines, internal combustion
(1C) engines, and commercial and residential warm air
furnaces. Other sources (including mobile and noncombustion)
will be considered only to the extent that they are needed to
determine the NO contribution from stationary combustion
A
sources.
Conventional and alternate gaseous, liquid and solid fuels
Combustion modification controls with potential for
implementation to the year 2000; other controls (flue gas
cleaning, mobile controls) will be considered only to estimate
the future need for combustion modifications.
t Source effluent streams potentially affected by NO controls.
A
ill
-------
Primary and secondary gaseous, liquid and solid pollutants
potentially affected by NO controls.
A
Pollutant impacts on human health and terrestrial or aquatic
ecology.
To achieve the objectives discussed above, the NO EA program
A
approach is structured as shown schematically in Figure P-l. The two
major tasks are: Environmental Assessment and Process Engineering (Task
B5), and Systems Analysis (Task C). Each of these tasks is designed to
achieve one of the overall objectives of the NO EA program cited
A
earlier. In Task B5, of which this report is a part, the environmental,
economic, and operational impacts of specific source/control combinations
will be evaluated. On the basis of this assessment, the incremental
multimedia impacts from the use of combustion modification controls will
be identified and ranked. Task C will in turn use the results of Task B5
to identify and rank the most effective source/control combinations to
comply, on a local basis, with the current ML air quality standards and
projected N0? related standards.
As shown in Figure P-l, the key tasks supporting Tasks B5 and C are
Baseline Emissions Characterization (Task Bl), Evaluation of Emission
Impacts and Standards (Task B2), and Experimental Testing (Task B3). The
arrows in Figure P-l show the sequence of subtasks and the major
interactions among the tasks. The oval symbols identify the major outputs
of each task. The subtasks under each main task are shown on the figure
from the top to the bottom of the page in roughly the same order in which
they will be carried out.
As indicated above, this report is a part of the Process
Engineering and Environmental Assessment Task. The goal of this task is
to generate process evaluations and environmental assessments for specific
source/control combinations. These studies will be done in order of
descending priority. In the first year of the NO EA, all the sources
A
and controls involved in current and planned NO control implementation
A
programs were investigated. The "Preliminary Environmental Assessment of
Combustion Modification Techniques" (Reference P-l) documented this effort
and established a priority rankings based on source emission impact and
potential for effective NO control, to be used in the current ongoing
A
detailed evaluation.
IV
-------
EMISSIONS
CHARACTERIZATION
-------
This report presents the assessment of combustion modification
controls for the fourth source category to be treated, internal combustion
engines. Other environmental assessment reports documented are:
Environmental Assessment of Utility Boiler Combustion
Modification NO Controls (Reference P-2)
/\
Environmental Assessment of Industrial Boiler Combustion
Modification NO Controls (Reference P-3)
/\
Environmental Assessment of Combustion Modification Controls
for Stationary Gas Turbines (Reference P-4)
Environmental Assessment of Combustion Modification Controls
for Residential and Commercial Heating Systems (Reference P-5)
-------
REFERENCE FOR PREFACE
P-l. Mason, H. B., et al., "Preliminary Environmental Assessment of
Combustion Modification Techniques: Volume II. Technical Results,"
EPA-600/7-77-1195, NTIS-PB 276 681/AS, October 1977.
P-2. Lim, K. J., et al., "Environmental Assessment of Utility Boiler
Combustion Modification NOX Controls: Volume I. Technical
Results; Volume II, Appendices," EPA-600/7-80-075a, b, April 1980.
P-3. Lim, K. J., et al., "Industrial Boiler Combustion Modification NOX
Controls: Volume I. Environmental Assessment," EPA-600/7-81-126a,
July 1981.
P-4. Larkin, R., et al., "Combustion Modification Controls for Stationary
Gas Turbines": Tolume I. Environmental Assessment,"
EPA-600/7-81-122a, July 1981.
P-5. Castaldini, et al., "Combustion Modification Controls for
Residential and Commercial Heating Systems: Volume I.
Environmental Assessment," EPA-600/7-81-123a, July 1981.
Vll
-------
CONTENTS
Section Page
Preface iii
Figures xii
Tables xiv
1 EXECUTIVE SUMMARY 1-1
1.1 Overview of Stationary Reciprocating Internal
Combustion Engines 1-2
1.2 Waste Streams and Pollutants of Major Concern . . . 1-2
1.3 Status of Environmental Protection Alternatives. . . 1-4
1.4 Data Needs and Recommendations 1-15
REFERENCES FOR SECTION 1 1-17
2 INTRODUCTION 2-1
2.1 Background 2-1
2.2 Role of Stationary Internal Combustion Engines . . . 2-3
2.3 Objective of This Report 2-4
2.4 Organization of This Report 2-6
REFERENCES FOR SECTION 2 2-7
3 SOURCE CHARACTERIZATION 3-1
3.1 Engine Design Types 3-1
3.1.1 Fuel Type 3-2
3.1.2 Method of Ignition 3-2
3.1.3 Combustion Cycle 3-3
3.1.4 Charging Method 3-5
3.2 Applications and Rank 3-7
3.2.1 Large Bore, High Power, Low-
and Medium-Speed Engines 3-7
3.2.2 Medium Power, High Speed Engines 3-8
3.2.3 Small Engines and Very Small Engines 3-9
3.2.4 Engine Users 3-9
3.3 Typical Engines in Each Size Class 3-10
3.3.1 Large Bore, High Power Engines 3-10
3.3.2 Medium Bore, Medium Power Engines 3-10
3.3.3 Small Bore Engines 3-11
REFERENCES FOR SECTION 3 3-12
-------
CONTENTS (continued)
Section Page
4 CHARACTERIZATION OF INPUT MATERIALS, PRODUCTS, AND
WASTE STREAMS 4-1
4.1 Baseline Emissions 4-1
4.2 Emissions Formation 4-2
4.3 Atmospheric Effects 4-4
4.4 Fuel Effects 4-5
4.4.1 Diesel Oil (No. 2) 4-5
4.4.2 Gasoline 4-7
4.4.3 Natural Gas 4-8
4.4.4 Other Fuels 4-8
4.5 Effects of Engine Design and Operation 4-8
4.5.1 Charging Method and Air-to-Fuel Ratio 4-8
4.5.2 Combustion Cycle 4-11
4.5.3 Combustion Chamber Design and
Ignition Timing 4-11
4.5.4 Load and Speed Effects 4-12
4.6 Products Characterization 4-12
REFERENCES FOR SECTION 4 4-14
5 PERFORMANCE AND COST OF CONTROL ALTERNATIVES 5-1
5.1 NOX Control Techniques 5-2
5.1.1 Air-to-Fuel Ratio Changes 5-3
5.1.2 Retarded Ignition Timing 5-4
5.1.3 Manifold Air Cooling and Turbocharging 5-6
5.1.4 Exhaust Gas Recirculation 5-11
5.1.5 Water Injection 5-12
5.1.6 Derating 5-15
5.1.7 Combining Control Techniques 5-15
5.1.8 Combustion Chamber Modifications 5-17
5.1.9 Catalytic Reduction 5-26
5.2 Cost of Pollution Controls 5-27
5.2.1 Cost Evaluation Procedures 5-28
5.2.2 Estimated Costs of Operational Adjustments .... 5-29
5.2.3 Exhaust Gas Treatment Techniques 5-35
5.2.4 Combustion Chamber Redesign 5-38
-------
CONTENTS (Concluded)
Section
5.3 Summary of NOX Emission Reductions and
Operation and Maintenance Impacts 5-38
5.3.1 NOX Reduction Techniques 5-38
5.3.2 Operational and Maintenance Impacts 5-39
REFERENCES FOR SECTION 5 5-43
6 ENVIRONMENTAL IMPACTS 6-1
6.1 Pollutant Formation Mechanism ... 6-1
6.1.1 NOX Formation 6-1
6.1.2 HC Formation 6-1
6.1.3 CO Formation 6-2
6.1.4 Smoke Formation 6-2
6.1.5 SOX and Trace Elements 6-3
6.1.6 Baseline Emissions 6-3
6.2 Incremental Environmental Impacts on Air 6-4
6.2.1 Derate 6-4
6.2.2 Retard 6-4
6.2.3 Air-to-Fuel Changes 6-11
6.2.4 Reduced Manifold Air Temperature 6-11
6.2.5 Exhaust Gas Recirculation 6-11
6.2.6 Flue Gas Treatment 6-12
6.2.7 Source Assessment Model Results 6-12
6.2.8 Concluding Remarks on Health Effects 6-13
6.3 Water and Solid Pollution Impacts 6-15
6.4 Summary of Environmental Impacts 6-15
REFERENCES FOR SECTION 6 6-18
7 SUMMARY OF NEEDS FOR ADDITIONAL DATA 7-1
7.1 Data Needs 7-1
REFERENCE FOR SECTION 7 7-3
APPENDIX A SAM/IA WORKSHEETS A-l
XI
-------
FIGURES
Figure Page
P-l NOX EA Approach v
2-1 Distribution of Stationary Anthropogenic NOX
Emission for the Year 1977
(Updated from Reference 2-19) 2-5
3-1 The Four-Stroke, Spark-Ignition (SI) Cycle. Four
Strokes of 180° of Crankshaft Rotation Each, or
720° of Crankshaft Rotation per Cycle
(Reference 3-4) 3-4
3-2 Cylinder Events for a Two-Stroke Blower-Scavenged
1C Engine (Reference 3-4) 3-6
4-1 Effect of Air-to-Fuel Ratio on Emissions from a
Gasoline Engine (Reference 4-1) 4-10
5-1 Effect of Air-to-Fuel (A/F) Changes on NOX
Emissions and Fuel Consumption (Reference 5-1) .... 5-5
5-2 Effect of Ignition Retard on NOX Emissions and
Fuel Consumption for Diesel Engines (Reference 5-1) . . 5-7
5-3 Effect of Ignition Retard on NOX Emissions and Fuel
Consumption for Gas and Dual Fuel Engines
(Reference 5-1) 5-8
5-4 Effect of Manifold Air Temperature Reduction on NOX
Emissions and Fuel Consumption (Reference 5-1) .... 5-10
5-5 Effect of Exhaust Gas Recirculation on NOX
Emissions and Fuel Consumption (Reference 5-1) .... 5-13
5-6 Effect of Varying Exhaust Gas Recirculation (EGR)
on NOX and Fuel Consumption for Gas 5-14
5-7 Effect of Derate on NOX Emissions and Fuel
Consumption (References 5-1 and 5-20) 5-16
5-8 Additive Effects of Controls for a Large Bore
Diesel Engine (Reference 5-1) 5-18
5-9 Additive Effects of Controls for a Large Bore
Dual Fuel Engine (Reference 5-1) 5-19
5-10 Additive Effects of Controls for Large Bore Gas
Engines (Reference 5-1) 5-20
-------
FIGURES (Concluded)
Figure
5-11 Fuel Injection Torch Ignition Concept
(Reference 5-21) 5-22
5-12 Alternative Pilot Injection Schemes (Reference 5-7) . . 5-25
6-1 HC Levels Versus Controlled NOX Levels
for Diesel Engines (Reference 6-3) 6-5
6-2 HC Levels Versus Controlled NOX Levels
for Dual-Fuel Engines (Reference 6-3) 6-6
6-3 HC Levels Versus Controlled NOX Levels
for Gas Engines (Reference 6-3) 6-7
6-4 CO Emissions Versus Controlled NOX Levels
for Diesel Engines 6-8
6-5 CO Emissions Versus Controlled NOX Levels
for Dual-Fuel Engines (Reference 6-3) 6-9
6-6 CO Emissions Versus Controlled NOX Levels
for Gas Engines (Reference 6-3) 6-10
xm
-------
TABLES
Table
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
1-9
1-10
4-1
4-2
5-1
5-2
5-3
5-4
5-5
Emissions Factors for 1C Engines, g/kWh
(References 1-2, 1-3)
NOX Reduction and Fuel Consumption Penalties for
Diesel, Dual-Fuel, and Gas Engines
Abbreviations for Engine Type and Emission Control
Technology used in Following Tables
Effects of Controls on Engines Larger than
5.7 x 10-3m3/cyl: NOX Emissions
Effects of Controls on Engines Larger than
5.7 x 10-3m3/cyl: CO Emissions
Effects of Controls on Engines Larger than
5.7 x 10-3m3/cyl: HC Emissions
Estimated Incremental Cost of Air-to-Fuel
Increase (1978)
Estimated Incremental Cost of External Exhaust
Gas Recirculation (1978)
Estimated Incremental Cost due to Retard (1978) ....
Estimated Incremental Cost of Combined Controls for
Large Bore Diesel Engines (1978)
Emissions Factors for 1C Engines, g/kWh
(References 4-1, 4-2)
Specifications for Diesel Fuels (ASTM D975)
(Reference 4-9)
Emissions from 1C Engines (1975)
Typical Costs for Uncontrolled Engines (1978)
Estimated Incremental Cost due to Retard (1978)
Estimated Incremental Cost of Air-to-Fuel Increase
(1978)
Estimated Incremental Cost of External Exhaust
Gas Recirculation (1978)
Page
1-3
1-5
1-7
1-8
1-9
1-10
1-11
1-12
1-13
1-14
4-3
4-6
5-2
5-30
5-32
5-33
5-36
XIV
-------
TABLES (Concluded)
Table Page
5-6 Estimated Incremental Cost of Combined Controls for
Large Bore Diesel Engines at 40 Percent NOX
Reduction (1978) .................... 5-37
5-7 NOX Reduction and Fuel Consumption Penalties for
Diesel, Dual Fuel, and Gas Engines ........... 5-40
6-1 Average Baseline Emissions from Stationary Internal
Combustion Engines, g/kWh ................ 6-3
6-2 SAM IA Results for Diesel Exhaust ........... 6-14
6-3 Effects of Controls on Engines Larger than
5.7 x lO-^/cyl: CO Emissions ............ 6-16
6-4 Effects of Controls on Engines Larger than
5.7 x 10-3m3/cyl: HC Emissions ............ 6-17
xv
-------
SECTION 1
EXECUTIVE SUMMARY
This is the fourth in a series of five special reports to be
documented in the "Environmental Assessment of Stationary Source NO
X
Combustion Modification Technologies" (NO EA). Specifically, this
/\
report documents the environmental assessment of stationary reciprocating
internal combustion (1C) engines with primary emphasis on NO combustion
/\
controls. The program has two main objectives: (1) to identify the
multimedia environmental impact of stationary combustion sources and
combustion modification controls applied to these sources, and (2) to
identify the most cost-effective, environmentally sound combustion
modification controls for attaining and maintaining current and projected
NOp air quality standards to the year 2000.
With more NO controls being implemented in the field and
/\
expanded control development anticipated for the future, there is
currently a need to: (1) ensure that the current and emerging control
techniques are technically and environmentally sound and compatible with
efficient and economical operation of systems to which they are applied,
and (2) ensure that the scope and timing of new control development
programs are adequate to allow stationary sources of NO to comply with
/\
potential air quality standards. The environmental assessment of
stationary reciprocating internal combustion (1C) engines helps to address
these needs by evaluating the operational, economic and environmental
impacts from applying combustion modification controls.
Internal combustion engines are the second largest contributors of
NO emissions from stationary anthropogenic sources in the U.S. --
X
constituting an 18.9 percent share (Reference 1-1). Because of this high
NO level and their potential for control, stationary 1C engines have
/\
been selected as one of the major source categories to be treated under
the NO EA program.
-------
1.1 OVERVIEW OF STATIONARY RECIPROCATING INTERNAL COMBUSTION ENGINES
This report concentrates on large bore engines since the smaller
engines tend to be adaptations of mobile engines and could thus
conceivably use the same control techniques. Also the largest amount of
NO emitted from stationary 1C engines comes from large bore engines.
A
These engines are used mainly as electrical generators, pumps, and
compressors. As electrical generators they are used by small public
utilities for baseload power and by some larger utilities as peaking
units, and at remote industrial sites and sites where a large amount of
power is needed. Their major use as pumps and compressors is in pipeline
service and oil and gas production.
The large bore engines ( 75 kW/cylinder) tend toward lower speeds,
usually less than 1000 RPM and burn three major types of fuel: diesel,
natural gas, and dual fuels (mixtures of diesel and gas). The natural gas
engines are spark ignited, while the diesel and dual fuel engines are
compression ignited. Some engines can burn residual oil and some use
sewage gas. Both two- and four-stroke models can be found in this size
range and the engine may be turbocharged, which usually increases
efficiency. Typical heat rates are 9 to 11 MJ/kWh (8500 to 10,500 Btu/
kWh). Smaller engines are not usually as efficient but have a lower
initial capital cost. Also the larger engines tend to be operated
unattended and at constant speed and are in use almost continuously.
1.2 WASTE STREAMS AND POLLUTANTS OF MAJOR CONCERN
The major waste streams are emitted through the exhaust pipe.
Hydrocarbons (HC) can be emitted from the fuel before combustion,
especially from natural gas-fired engines, but these emissions are minor.
There may also be some emissions from the crankcase caused by blowby but
this is a minor source. Also, the cooling system may release minor water
pollutant emissions. Liquid wastes in the form of used crankcase oil may
be another pollutant, but is not a major one.
Some average values for exhaust pipe emissions are listed in
Table 1-1 (References 1-2 and 1-3). The HC emissions listed are total
hydrocarbons; in the case of natural gas engines, these are mainly
methane. Although the table lists factors for all engine sizes, this
report focuses on the larger engines. Note that NO is the major
/\
pollutant for the large engines. Diesel engines may also emit polycyclic
1-2
-------
TABLE 1-1. EMISSIONS FACTORS FOR 1C ENGINES, g/kWh*
(References 1-2, 1-3)
Fuel
Gasoline >15 kW
<15 kW
Diesel >375 kWb
<375 kWc
Natural gas
Dual Fuel
NOX
11.9
7.5
17.3
16.6
15.4
11.0
CO
137
395
2.4
6.0
3.8
2.7
HC
11.2
27.5
0.6
2.8
6.5
4.1
Emission factors for gasoline and diesel engines are
modal averages; those for natural gas and dual fuel
are for rated conditions. Modal averages mean that
some of the NO^ numbers are taken from the constant
power out portion of mobile tests.
bBased on an average of rated condition levels from
engines considered
cWeighted average of two- and four-stroke engines.
Weighting factors = 2/3 for four-stroke and 1/3 for
two-stroke
1-3
-------
organic matter (POM). EPA is actively researching this area of concern
(Reference 1-4).
1.3 STATUS OF ENVIRONMENTAL PROTECTION ALTERNATIVES
There are three major types of pollution controls for 1C engines:
operational adjustment, catalytic exhaust gas treatment, and combustion
chamber redesign. Operational adjustment techniques are essentially
proven and can be applied now. Some combustion chamber redesign
techniques have been tested but most are unproven. Using catalysts to
reduce NO emissions has only been laboratory tested and not tested on
A
an actual engine under lean burning conditions. N0x control catalysts
have been used on engines operating under rich conditions. Catalysts have
also been used as CO and HC control devices.
The operational adjustment techniques are derate, ignition or
injection retard, air-to-fuel ratio change, reduced manifold air
temperature, and exhaust gas recirculation. Table 1-2 summarizes the
expected NO reduction along with the change in the brake specific fuel
/\
consumption (BSFC). Notice that most techniques have major effects on
BSFC.
Derating controls NO by lowering combustion temperatures, but
A
could require larger engines and increase carbon monoxide (CO) and HC
emissions. No major operation impacts are expected except for increased
fuel consumption.
For a lean burning engine, changes in the air-to-fuel ratio control
NO emissions by adding more air per unit of fuel and lowering the
s\
combustion temperature. Although altering this ratio may increase smoke
and HC emissions, no major problems could be expected unless the ratio is
adjusted to allow misfiring. The onset of engine misfiring is used as the
maximum limit of this control technique.
Increased manifold air cooling also lowers the combustion
temperature and may allow a higher air-to-fuel ratio. The manifold air
cooling system may require more maintenance since it would be important
that the cooling system operate at maximum efficiency for this control
technique to be applied. Also a larger cooling system could be required.
Ignition retard causes combustion to occur later in the power
stroke, thereby lowering both the peak combustion temperature and time
1-4
-------
TABLE 1-2. NOX REDUCTION AND FUEL CONSUMPTION PENALTIES FOR DIESEL,
DUAL-FUEL, AND GAS ENGINES
Control Approach
Derate (D) 3%
6%
10%
20%
25%
Retard (R) 2°
4°
8°
Air-to-Fuel (A) 2%
3%
5%
±10%
Manifold (M')311K(100°F)
Air 315K(107°F)
Temperature 318K(113oF)
Internal EGR
External EGR 10%
Retard and Manifold
Air Temperature
Retard & Air-to-Fuel
Retard and Manifold
Air Temperature and
Air-to-Fuel
Air-to-Fuel and
Manifold Air Temperature
Water Injection 50%
(H20/fuel ratio) 100X
Catalytic Reduction
(Projected)
Combustion Increased
Chamber Mixing
Modifications
(Projected) staged
Combustion
Engine Fuel Type
Diesel
% NOX
Reduction
--
--
<20
5-23
<20
<40
28-45
--
--
7-8
7-15
5
<20
33
<20
10-24
<20
<40
35-65
20
<20
20-30
25-35
50-80
10-30
10-30
ABSFC, %a
--
--
--
4
1-5
4
4
2-8
--
--
3
0-2
2
1
1
1
0-1
8
16
5-26
0
2
3
2-4
0
<5
0
Dual Fuel
% NOX
Reduction
--
<20
--
1-33
<20
<40
50-73
_-
<20
25-40
18-37
<20
;;
20
<20
25
<20
<40
56
<20
40
__
--
50-80
20-40
10-30
ABSFC, %a
--
4
--
1-7
3
1
3-5
--
0
1-3
0-1
1
;;
i
i
2
i
2
2
2
3
_.
--
0
<5
0-7
Natural Gas
% NOX
Reduction
<20
<40
--
--
5-90
<20
8-40
<20
--
<40
20-80
28
<20
<20
5-35
<20
33
<20
30-40
<20
<40
17-52
<20
<40
40-65
._
25-35
60-75
50-80
20-40
10-30
ABSFC, %a
2
3
--
2-12
-_
3
2-7
2
--
7
5-12
0
0
5
0-8
0
0
3
5-6
4
8
4-11
2
4
6-7
__
1-2
2-5
0
<5
0-2
aABSFC is increase in brake specific fuel consumption
1-5
-------
spent at high temperatures, but increases exhaust temperature. This
control method can reduce the life of exhaust valves and exhaust system.
Exhaust gas recirculation (EGR) reduces NO by lowering the
/\
combustion temperature. However, EGR increases maintenance problems,
especially for diesel engines. The exhaust gas recirculation system in
addition to the turbocharger can require periodic cleaning because of the
particulate loading in diesel exhaust units.
Although combustion chamber redesign techniques have not been
commercially demonstrated, they potentially have the least operational
impact. The goal of the combustion chamber redesign program is to design
an engine that will have a smaller fuel impact. Also it is hoped that,the
tuning of the new engine will not be as critical.
Because large engines are usually lean burning, catalytic reduction
of NO will most likely require a reducing agent such as ammonia. Thus
A
there are potential operational problems associated with ammonia
injection. There will be an additional system to maintain as well as the
potential for additional harmful emissions such as NH3 and HCN.
The effects of these controls on NO , HC, and CO are summarized
X
in Tables 1-3 through 1-10. Estimated costs for certain operational
adjustment techniques are also presented. The percentage cost increases
are based on engine cost only (e.g., the generator costs are not
included). The incremental initial capital cost of the available controls
range from zero to 5 percent of an uncontrolled engine. However, the
total annualized cost to control can increase the cost of power from an
engine by 1 to 14 percent, a significant impact because of additional fuel
and maintenance requirements.
Best Available Control Technology (BACT)
Currently the best available control technologies to control NO
/\
emissions from large bore engines are: (1) retarded ignition or fuel
injection, (2) air-to-fuel ratio changes, (3) increased manifold air
cooling, and (4) derating. These techniques can either be employed alone
or in combinations with each other. The best combination will be very
engine dependent. But in general retard is best for diesel-fueled
engines, air-to-fuel ratio changes for natural gas, and either control for
dual fuel. A 40 percent reduction in NO can usually be achieved
y\
without causing any major operational problems except for increased fuel
1-6
-------
TABLE 1-3. ABBREVIATIONS FOR ENGINE TYPE AND EMISSION CONTROL
TECHNOLOGY USED IN FOLLOWING TABLES
Abbreviation
Explanation
Air Charging
BS
NA
TC
Control Technology
D
R
TC
A/F
MAT
EGR(I)
EGR(E)
INJ
H20
CCR
Blower scavenged
Naturally aspirated
Turbocharged (and intercooled)
Derating
Retard
Turbocharged (and intercooled)
Increased air-to-fuel ratio
Decreased inlet manifold air temperature
Exhaust gas recirculation -- internal
Exhaust gas recirculation -- external
Modified injectors
Water induction
Combustion chamber redesign
1-7
-------
TABLE 1-4. EFFECTS OF CONTROLS ON ENGINES LARGER THAN 5.7 x lO'V/cyl: NOY EMISSIONS
A
Fuel
Strokes/Cycle
"""'""' ^^J\ir Charging
Control - ~^___^^
D
R
A/F
TC
MAT
INJ
EGR(I)
EGR(E)
H20
CCR-7
Diesel
Two
BS
4-
4-
4-
4-
4-
4-
TC
4-
*
i4-
4-
Four
NA
4-
TC
4-
4-
4-
4-
+
4-
4-
Dual Fuel
Two
BS
TC
4-
4-
.
4-
+
Four
NA
TC
4-
4-
4-
T-v
4-
t
4-
Natural Gas
Two
BS
4-
4-
4-
4-
4-
4-
TC
4-
4-
4-
4-
Four
NA
4-
4-
' 4-
f
4-
4-
TC
4-
4-
+
*
4-
I
CX3
-h Denotes emission increase with application of control
4- Denotes emission decrease with application of control
t-4- Denotes conflicting data with application of control
- Denotes no change in emissions with application of control
Blank indicates no data available on effect
-------
TABLE 1-5. EFFECTS OF CONTROLS ON ENGINES LARGER THAN 5.7 x 10~3m3/cyl: CO EMISSIONS
Fuel
Strokes/Cycle
- ^JUr Charging
Control ""~~~--^^^
D
R
A/F
TC
MAT
INJ
EGR(I)
EGR(E)
H20
CCR
Diesel
Two
BS
t
*
t
+
4-
4-
TC
t
i
H
t
Four
NA
TC
4
-f
i4-
t4-
4-
t4-
Dual Fuel
Two
BS
TC
t
+
t
f
Four
NA
TC
H
f
4-
1-4-
4-
4-
Natural Gas
Two
BS
t
t
4-
t
f
TC
i
t
i
t
Four
NA
t
4-
i
TC
t
4-
i
t
t Denotes emission increase with application of control
4- Denotes emission decrease with application of control
ti Denotes conflicting data with application of control
- Denotes no change in emissions with application of control
Blank indicates no data available on effect
-------
TABLE 1-6. EFFECTS OF CONTROLS ON ENGINES LARGER THAN 5.7 x 10-3m3/cyl: HC EMISSIONS
Fuel
Strokes/Cycle
"~ -^-^Air Charging
Control ^^^
D
R
A/F
TC
MAT
INJ
EGR(I)
EGR(E)
H20
CCR
Diesel
Two
BS
4-
+
4-
t
TC
+4-
+4-
4-
Four
NA
4-
TC
t
+4-
+ 4-
f4-
4-
C4-
Dual Fuel
Two
BS
TC
-h
h
t
4-
Four
NA
TC
+
4-
+
-n-
+
+
Natural Gas
Two
BS
+
t
4-
f
'^
TC
t
h
i
i
Four
NA
t
-h
-^
TC
-h
t
i
i
t
I
t«
o
t Denotes emission increase with application of control
4- Denotes emission decrease with application of control
t4- Denotes conflicting data with application of control
- Denotes no change in emissions with application of control
Blank indicates no data available on effect
-------
TABLE 1-7. ESTIMATED INCREMENTAL COST OF AIR-TO-FUEL INCREASE (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWh)
Total Annualized ($/kWh)a
Percent increase in initial cost
Percent increase in annual cost
Percent reduction in NOX
Engine/ Fuel Type
Diesel
(Electrical
Generation)
0
0.0002
0.0031
0.003
0
7.0
20
Dual Fuel
(Electrical
Generation)
0
0.0002
0.0004
0.001
0
3.0
40
Natural Gas
(Oil and Gas
Pipeline)
0
0.0002
0.0004
0.001
0
3.0
40
(Oil and Gas
Production)
0
0.0002
0.0004
0.001
0
3.0
40
aAssumes 8000 hours operating year.
-------
TABLE 1-8. ESTIMATED INCREMENTAL COST OF EXTERNAL EXHAUST GAS RECIRCULATION (1978)
ro
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWh)
Total Annual i zed ($/kWh)a
Percent increase in initial cost
Percent increase in annual cost
Percent reduction in NO
y\
Engine/ Fuel Type
Diesel
(Electrical
Generation)
12
0.002
0
0.0023
5
5.3
20
Dual Fuel
(Electrical
Generation)
12
0.004
0
0.0043
5
14
20
Natural Gas
(Oil and Gas
Pi pel ine)
12
0.004
0
0.0043
5
14
20
(Oil and Gas
Production)
4
0.004
0
0.0041
5
13
20
aAssumes 8000 hours operating year.
-------
TABLE 1-9. ESTIMATED INCREMENTAL COST DUE TO RETARD (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWh)
Total Annual! zed ($/kWh)a
Percent increase in initial cost
Percent increase in annual cost
Percent reduction in NOX
Engine/Fuel Type
Diesel
(Electrical
Generation)
0
0.0016
0.0012
0.003
0
7.0
20-30
Dual Fuel
(Electrical
Generation)
0
0.0016
0.0007
0.002
0
6.4
20-30
Natural Gas
(Oil and Gas
Pipeline)
0
0.0016
0.0008
0.002
0
6.1
20-30
(Oil and Gas
Production)
0
0.0016
0.0009
0.002
0
6.2
20-30
I
I"
co
^Assumes 8000 hours operating year.
-------
TABLE 1-10. ESTIMATED INCREMENTAL COST OF COMBINED CONTROLS FOR
LARGE BORE DIESEL ENGINES (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWH)
Total Annualized ($/kWh)a
Percent increase in capital cost
Percent increase in annual cost
Percent reduction in NOX
Control Technique
Retard
0
0.0016
0.0024
0.0040
0
9.3
40
Air-to-Fuel
Changes and
Manifold Air
Cooling
3.6
0.001
0.0015
0.0026
1.5
6.0
30-40
Air-to-Fuel
and Retard
0
0.0018
0.0015
0.0033
0
7.7
40
aAssumes 8000 hours operating year.
1-14
-------
consumption and possible increased maintenance. Since the baseline
emissions of these engines vary greatly, the actual control level achievable
also varies greatly. For dual fuel and gas engines a 40 percent NO
A
reduction causes an increase in the brake specific fuel consumption by about
2 to 7 percent while for diesel engines the increased fuel consumption would
be 4 to 8 percent based on the large bore engines tested. HC and CO
emissions may also increase by up to 50 percent. On a mass basis these
increases would be much less than the amount of NO reduced.
A
Best Promising Techniques
The best promising future control techniques are combustion chamber
redesign and catalytic exhaust gas treatment. Combustion chamber redesign
techniques have the potential of giving the same NO emission reduction
A
as the BACT technique but at a lower cost and smaller fuel penalty. If
very low NO emissions are required, catalytic exhaust gas treatment is
A
the only technique with that potential. For lean burning engines NH-
would probably have to be used as a NO reducing agent. The engine
X
would require greater operator attention because of the ammonia injection
equipment, and handling and storage facilities. Also the use of NH.,
reflects an energy consumption in itself since NhL is currently produced
from natural gas. Thus catalytic exhaust gas treatment of NO could
A
have a significant fuel impact.
1.4 DATA NEEDS AND RECOMMENDATIONS
There are two major weaknesses in the amount of available data for
combustion modification controls on 1C engines. The information on
operational effects and long term durability of these control techniques
is very limited, especially combustion chamber redesign data and catalytic
exhaust gas treatment. More information on combining these controls to
achieve the optimal low emissions/high efficiency is needed. Also the
available data on emissions other than NO , CO, and total hydrocarbons
A
are very limited. The amount and type of organics emitted from these
large bore engines are not very well understood. The potential
mutagenicity of organic emissions in diesel exhaust is of major concern.
Research is needed on designing a high efficiency low NO
A
emitting engine. Even with the best available controls applied, the large
bore stationary reciprocating internal combustion engine is the highest
NO emitter on a heat input basis of all the major combustion sources.
A
1-15
-------
EPA is currently sponsoring several programs in the health effects
area as well as new engine designs for low NO and high efficiency.
/\
These programs should help resolve many of the major data gaps in the
operational and environmental impacts of NO controls. One of these
^
programs is being conducted by A. D. Little under contract to the EPA.
They are presently testing new combustion chamber modification techniques
on large bore one cyclinder engines in the laboratory and plan to test
these techniques on full size engines in 1981 (Reference 1-5).
1-16
-------
REFERENCES FOR SECTION 1
1-1. Waterland, L.R., et al., "Environmental Assessment of Stationary
Source NOX Control Technologies -- Final Report," Acurex Draft
Report FR-80-57/EE, EPA Contract 68-02-2160, Acurex Corp., Mountain
View, CA, April 1980.
1-2. Youngblood, S. B., and G. R. Offen, "Emissions Inventory of
Currently Installed Stationary Reciprocating 1C Engines," Acurex
Internal Communication, Acurex Corporation, Mountain View, CA,
September 23, 1975.
1-3. "Stationary Internal Combustion Engines. Background Information:
Proposed Standards," EPA-450/3-78-125a, April 1979.
1-4. Barth, D. S., and S. M. Blacker, "The EPA Program to Assess the
Public Health Significance of Diesel Emissions," JAPCA, Vol. 28,
No. 8, 1978, pg. 769.
1-5. Personal communication with Jack Wasser, EPA, Research Triangle
Park, North Carolina, April 30, 1980.
1-17
-------
SECTION 2
INTRODUCTION
This report assesses the operational, economic, and environmental
impacts from applying combustion modification controls to stationary
internal combustion engines. With more NO controls being implemented
X
in the field and expanded control development anticipated for the future,
there is currently a need to: (1) ensure that the current and emerging
control techniques are technically and environmentally sound, and
compatible with efficient and economical operation of systems to which
they are applied, and (2) ensure that the scope and timing of new control
development programs are adequate to allow stationary sources of NO to
A
comply with potential air quality standards. The NO EA program
A
addresses these needs by (1) identifying the incremental multimedia
environmental impact of combustion modification controls, and (2)
identifying the most cost-effective source/control combinations to achieve
ambient N0? standards.
2.1 BACKGROUND
The 1970 Clean Air Act Amendments designated oxides of nitrogen
(NO ) as one of the criteria pollutants requiring regulatory controls to
A
prevent potential widespread adverse health and welfare effects.
Accordingly, in 1971, EPA set a primary and secondary National Ambient Air
Quality Standard (NAAQS) for N02 of 100 yg/m3 (annual average). To
attain and maintain the standard, the Clean Air Act mandated control of
new mobile and stationary NO sources, each of which currently emits
A
approximately half of the manmade NO nationwide. Emissions from light
A
duty vehicles (the most significant mobile source) were to be reduced by
90 percent to a level of 0.25 g N02/km (0.4 g/mile) by 1976. Stationary
sources were to be regulated by EPA Standards of Performance for New
Stationary Sources (NSPS), which are set as control technology becomes
available. Additional standards required to attain air quality in the Air
2-1
-------
Quality Control Regions (AQCRs) could be set for new or existing sources
through the State Implementation Plans (SIPs).
Since the Clean Air Act, techniques have been developed and
implemented that reduce NO emissions by a moderate amount (30 to 60
/\
percent) for a variety of source/fuel combinations. In 1971, EPA set NSPS
for large steam generators burning gas, oil, and coal (except lignite).
Recently, more stringent standards for utility boilers burning all gaseous
liquid and solid fuels have been promulgated, along with standards for
lignite fired utility boilers. In addition, NSPS have been promulgated
for stationary gas turbines and are currently being considered for
stationary internal combustion engines and intermediate size (industrial)
steam generators. Local standards also have been set, primarily for new
and existing large steam generators and gas turbines, as parts of State
Implementation Plans in several areas with NO problems. This
x\
regulatory activity has resulted in reducing NO emissions from
A
individual controlled sources by 30 to 60 percent. The number of
controlled sources is increasing as new units are installed with factory
equipped NO controls.
A
Emissions have been reduced comparably for light duty vehicles.
Although the goal of 90 percent reduction (0.25 g NO^/km) by 1976 has
not been achieved, emissions were reduced by about 25 percent (1.9 g/km)
for the 1974 to 1976 model years and in 1979 were reduced to 50 percent to
1.25 g/km. Achieving the 0.25 g/km goal has been deferred indefinitely
because of technical difficulties and fuel penalties. Initially, the 1974
Energy Supply and Environmental Coordination Act deferred compliance to
1978. The Clean Air Act Amendments of 1977 defined the 0.25 g/km emission
level as a research goal and set the standard of 0.62 g/km (1 g/mile) for
the 1981 model year and beyond.
Because the mobile source emission regulations have been relaxed,
stationary source NO control has become more important for maintaining
/\
air quality. Several air quality planning studies have evaluated the need
for stationary source NO control in the 1980's and 1990's in view of
/\
recent developments (References 2-1 through 2-10). These studies all
2-2
-------
conclude that relaxing mobile standards, coupled with the continuing
growth rate of stationary sources, will require more stringent stationary
source controls than current and impending NSPS provide. This conclusion
has been reinforced by projected increases in the use of coal in
stationary sources. The studies also conclude that the most
cost-effective way to achieve these reductions is by using combustion
modification NO controls in new sources.
/\
It is also possible that separate NO control requirements will
A
be needed to attain and/or maintain additional N02 related standards.
Data on the health effects of NO- suggest that the current NAAQS should
be supplemented by limiting short term exposure (References 2-4 and 2-11
through 2-14). In fact, the Clean Air Act Amendments of 1977 require EPA
to set a short term N02 standard for a period not to exceed 3 hours
unless it can be shown that such a standard is not needed. The need for a
short-term standard is currently under review by EPA.
EPA is continuing to evaluate the long range need for additional
NOV regulation as part of strategies to control oxidants or pollutants
X
for which NO is a precursor, e.g., nitrates and nitrosamines
J\
(References 2-4, 2-11, and 2-15 through 2-18). These regulations could be
source emission controls or additional ambient air quality standards. In
either case, additional stationary source control technology could be
required to assure compliance.
In summary, since the Clean Air Act, near term trends in NO
/\
control are toward reducing stationary source emissions by a moderate
amount; hardware modifications in existing units or new units of
conventional design will be stressed. For the far term, air quality
projections show that more stringent controls than originally anticipated
will be needed. To meet these standards, the preferred approach is to
control new sources by using low NOX redesigns.
2.2 ROLE OF STATIONARY INTERNAL COMBUSTION ENGINES
Internal combustion (1C) engines are the second largest contributor
of stationary source NO emissions in the U.S. Figure 2-1 shows that 1C
/\
engines were the origin of 18.9 percent of all stationary source NOX
2-3
-------
emissions for the year 1977 (Reference 2-19). In fact, NO emissions
A
are the principal pollutants from the larger 1C engines. Furthermore,
total stationary source NO emissions are projected to increase unless
A
adequate controls are developed (Reference 2-20).
Given this background and their potential NO control, 1C engines
A
were selected as the third source category to be treated under the NOX
EA program. The "Preliminary Environmental Assessment of Combustion
Modification Techniques" (Reference 2-8) concluded that modifying
combustion process conditions is the most effective and widely used
technique for achieving 20 to 60 percent reduction in oxides of nitrogen.
Nearly all current N0x control applications use combustion modifications.
2.3 OBJECTIVE OF THIS REPORT
This report provides comprehensive, objective, and realistic
evaluations and comparisons of the important aspects of the available
combustion control techniques, using a common and uniform basis for
comparison. The objective is to perform an environmental assessment of
combustion modification techniques for 1C engines to:
t Determine their impact on the achievement of selected
environmental goals, based on a comprehensive analysis from a
multimedia consideration
Ascertain the effect of their application on engine performance
and identify potential problem areas
Estimate the economics of their operation
t Estimate the limits of control achievable by combustion
modification
Identify further research and development and/or testing
required to optimize combustion modification techniques and to
upgrade their assessments.
Since larger engines are the primary emitters of NO , while the smaller
A
engines mainly emit CO, this report concentrates on the large bore
engines. Furthermore, medium to small engines are modified mobile engines
and could use the same emission control techniques as mobile engines.
Since controls for mobile engines have been discussed extensively
elsewhere (e.g., Reference 2-21) they will not be treated in detail here.
2-4
-------
Noncombustion 1.9%
Warm air furnaces 2.0
Gas turbines 2.0%
- Incineration 0.4%
Others 4.1%
Industrial process
heaters 4.U
Industrial
Boilers
14.4%
Reciprocating
1C Engines
18.9%
Total: 10.5 Tg/yr (11.6 x 10° tons/yr)
Figure 2-1.
Distribution of stationary anthropogenic NOX emissions
for the year 1977 (Reference 2-19).
2-5
-------
2.4 ORGANIZATION OF THIS REPORT
Evaluating the effectiveness and impacts of NO combustion
/\
controls applied to 1C engines requires assessing their effects on both
controlled source performance, especially as translated into changes in
operating costs and energy consumption, and on incremental emissions of
other pollutants as well as NO To perform such an evaluation, it is
/v
necessary to:
Characterize the source category with regard to equipment,
fuels, and emissions (Sections 3, 4)
Identify NO formation mechanisms and relate fuels to their
s\
emissions potential (Section 4)
Evaluate the performance of current and potential control
techniques available for implementation (Section 5)
Assess the operational and cost impacts, including energy
impacts, of implementing controls (Section 5)
Evaluate the environmental impact of controls through the
analysis of incremental emissions (Section 6)
Make an overall assessment and ranking of control techniques
for the major equipment/fuel categories (Section 1)
f Identify research and development needs (Section 7)
2-6
-------
REFERENCES FOR SECTION 2
2-1. Crenshaw, J. and A. Basala, "Analysis of Control Strategies to
Attain the National Ambient Air Quality Standard for Nitrogen
Dioxide," presented at the Washington Operation Research Council's
Third Cost Effectiveness Seminar, Gaithersburg, MD, March 1974.
2-2. "Air Quality, Noise and Health -- Report of a Panel of the
Interagency Task Force on Motor Vehicle Goals Beyond 1980,"
Department of Transportation, March 1976.
2-3. McCutchen, G. D., "NOX Emissions Trends and Federal Regulation,"
presented at the AIChE 69th Annual Meeting, Chicago, December 1976.
2-4. "Air Program Strategy for Attainment and Maintenance of Ambient Air
Quality Standards and Control of Other Pollutants," Draft Report,
U.S. EPA, Washington, D.C., October 1976.
2-5. "Annual Environmental Analysis Report, Volume 1 Technical Summary,"
The MITRE Corporation, MTR-7626, September 1977.
2-6. Personal communication with R. Bauman, Strategies and Air Standards
Division, Office of Air Quality Planning and Standards, U.S. EPA,
October 1977.
2-7. "An Analysis of Alternative Motor Vehicle Emission Standards," U.S.
Department of Transportation, U.S. EPA, U.S. FEA, May 1977.
2-8. Mason, H. B., et al., "Preliminary Environmental Assessment of
Combustion Modification Techniques: Volume II, Technical Results,"
EPA-600/7-77-119b, NTIS-PB 276 681/AS, October 1977.
2-9. Greenfield, S. M., et al., "A Preliminary Evaluation of Potential
NOX Control Strategies for the Electric Power Industry," EPRI
TR-13300, April 1977.
2-10. Waterland, L. R., et al., Environmental Assessment of Stationary
Source NOX Contro^Technologies -- Second Annual Report,"
EPA-600/7-79-147, NTIS-PB 300 469, June 1979.
2-11. French, J. G., "Health Effects from Exposure of Oxides of
Nitrogen," presented at the AIChE 69th Annual Meeting, Chicago,
November 1976.
2-12. "Scientific and Technical Data Base for Criteria and Hazardous
Pollutants -- 1975 EPA/RTP Review," EPA-600/1-76-023, NTIS-PB 253
942/AS, January 1976.
2-13. Shy, C. M., "The Health Implications of an Non-Attainment Policy,
Mandated Auto Emission Standards, and a Non-Significant
D terioration Policy," presented to Committee on Environmental and
Public Works, Serial 95-H7, February 1977.
2-7
-------
2-14. "Report on Air Quality Criteria for Nitrogen Oxides," AP-84,
Science Advisory Board, U.S. EPA, June 1976.
2-15. "Report on Air Quality Criteria: General Comments and
Recommendations," Report to the U.S. EPA by the National Air
Quality Advisory Committee of the Science Advisory Board, June 1976,
2-16. "Air Quality Criteria Document for Oxides of Nitrogen:
Availability of External Review Draft," Federal Register, Vol. 43,
pp. 55, 117, December 12, 1978.
2-17. Personal communication with M. Jones, Strategies and Air Standards
Division, Office of Air Quality Planning and Standards, U.S. EPA,
September 1976.
2-18. "Control of Photochemical Oxidants -- Technical Basis and
Implications of Recent Findings," EPA-450/2-75-005, July 1975.
2-19. Water land, L. R., et al., "Environmental Assessment of Stationary
Source NOX ControlTecHnblogies Final Report," Acurex Final
Report FR-80-57/EE, EPA Contract 68-02-2160, Acurex Corporation,
Mountain View, CA, April 1980.
2-20. Mason, H. B., ejt a/L_, "Utility Boiler NOX Emission
Characterization," in Proceedings: Second NOyControl Technology
Seminar, EPRI FP-1109-SR, July 1979.
2-21. Stork, E. 0., "Environmental Impact Statement and Economic Impact
Analysis Revised Heavy-Duty Engine Regulations for 1979 and later
Model Years," EPA Mobile Source Air Pollution Control, August 4,
1977.
2-22. Cadle, S. H., and 6. J. Nebel, "Control of Automotive Emissions,"
October 1978, G M Research Laboratories.
2-23. Klein, H. I., and R. W. Sbuschning, "Automotive Emissions and Fuel
Economy - A Primer," 1977 Ford Engineering Forum, June 1977.
2-8
-------
SECTION 3
SOURCE CHARACTERIZATION
Stationary reciprocating internal combustion (1C) engines are found
in a variety of applications where there is a requirement for mechanical
work in the form of shaft power. Installations vary greatly, ranging from
within large urban centers to remote areas. Because of their versatility,
1C engines range in size from 1 kW to over 10 MW power output.
Although reciprocating 1C engines are manufactured in almost every
size and serve industry throughout, a given industry in general will not
use all engine sizes. Small and very small engines are typically used as
consumer products around the home and small farm. At the other extreme
are the large engines, almost always found operating continuously in
utility applications such as electric power generation or pipeline pumping
(Reference 3-1). This report concentrates on the larger size engines
since they produce three quarters of all NO emissions from installed
/\
stationary engines (see Section 2), and their population, operation, and
emission characteristics can be quantified.
3.1 ENGINE DESIGN TYPES
All reciprocating internal combustion engines operate by the same
basic process. A combustible mixture is first compressed in a small
volume between the head of a piston and its surrounding cylinder. The
mixture is then ignited, and the resulting high pressure products of
combustion push the piston through the cylinder. This movement is
converted from linear to rotary motion by a crankshaft. The piston
returns, pushing out exhaust gases, and the cycle is repeated.
Although all reciprocating 1C engines follow the same basic
process, there are variations which classify engine types. Engines are
generally classified by their: (1) fuel burned, (2) method of ignition,
(3) combustion cycle, and (4) charging method.
3-1
-------
3.1.1 Fuel Type
The three primary fuels for stationary reciprocating internal
combustion engines are: gasoline, diesel (No. 2) oil, and natural gas.
Gasoline is used primarily for mobile and portable engines. Construction
sites, farms, and households typically use converted mobile engines for
stationary application because their cost is often less than an engine
designed specifically for stationary purposes (Reference 3-2). In
addition, mobile engine parts and service are readily available, and
gasoline is easily transported to the site. Thus gasoline is an essential
fuel for small and medium size stationary engines.
Diesel fuel oil is also easily transported, and therefore is used
in small and medium sized engines. Also, the generally higher
efficiencies exhibited by diesel engines makes diesel oil an ideal fuel
for large engines where operating costs must be minimized. Diesel is thus
the most versatile fuel for stationary reciprocating engines.
Natural gas is used more than any other fuel for large stationary
1C engines (Reference 3-3), typically operating pumps or compressors on
gas pipelines. This fuel may see decreasing usage if gas supplies
decrease.
Other fuels are also burned in stationary 1C engines, but their use
is limited. Some engines are burning heavy fuel oils, and a few burn
almost any other liquid fuel (Reference 3-3). Gaseous fuels such as sewer
gas are sometimes used at wastewater treatment plants where the gas is
available (Reference 3-4). Stationary 1C engines can be modified to burn
almost any liquid or gaseous fuel if the engine is properly designed and
adjusted.
3.1.2 Method of Ignition
Ignition is the means of initiating combustion in the engine
cycle. There are two methods used for stationary reciprocating 1C
engines: compression ignition (CI) and spark ignition (SI).
In compression ignition engines, combustion air is first
compression heated in the cylinder, and diesel fuel oil is then injected
into the hot air. Ignition is spontaneous as the air is above the
autoignition temperature of the fuel. Spark ignition engines initiate
combustion by the spark of an electrical discharge. Usually the fuel is
mixed with the air in a carburetor (for gasoline) or at the intake valve
3-2
-------
(for natural gas), but occasionally the fuel is injected into the
compressed air in the cylinder. Although all diesel fueled engines are
compression ignited and all gasoline and gas fueled engines are spark
ignited, gas can be used in a compression ignition engine if a small
amount of diesel fuel is injected into the compressed gas/air mixture to
initiate burning. Such dual fuel (DF) engines are usually designed to
burn any mixture ratio of gas and diesel oil, from 6- to 100-percent oil
(based on heating value).
CI engines usually operate at a higher compression ratio (ratio of
cylinder volume when the piston is at the bottom of its stroke to volume
when it is at the top) than SI engines because fuel is not present during
compression; hence there is no danger of premature autoignition. Since
engine thermal efficiency rises with increasing pressure ratio (and
pressure ratio varies directly with compression ratio), CI engines are
more efficient than SI engines. This increased efficiency is gained at
the expense of poorer response to load changes and a heavier structure to
withstand the higher pressures.
3.1.3 Combustion Cycle
As previously mentioned, the combustion process for stationary
reciprocating internal combustion engines consists of compressing a
combustible mixture by a piston, igniting it, and allowing the high
pressures generated to push the piston back. This process may be
accomplished in either four strokes or two strokes of the piston.
In the four-stroke cycle, the sequence of events may be summarized
as follows (see Figure 3-1):
1. Intake stroke suction of the air or air and fuel mixture
into the cylinder by the downward motion of the piston through
the cylinder.
2. Compression stroke -- compression of the air or air and fuel
mixture, thereby raising its temperature and reducing its
volume.
3. Ignition and power (expansion) stroke combustion and
consequent downward movement of the piston by pressure from the
expanding gases with energy transfer to the crankshaft.
4. Exhaust stroke -- expulsion of the exhaust gases from the
cylinder by the upward movement of the piston.
3-3
-------
Intake
Spark plug
Crank
(and crankshaft)
Intake Stroke
Intake valve opens
thus admitting
charge of fuel and
air. Exhaust valve
closed for most of
stroke.
Compression Stroke
Both valves closed
Fuel-air mixture
is compressed by
rising piston.
Spark ignites
mixture near end
of stroke.
Connecting
rod
(a)
(b)
Intake
manifold
Power or Work Stroke
Fuel-air mixture burns
increasing temperature
and pressure, expan
sion of combustion
gases drives pistoi
down. Both valves
closed -- exhaust
valve opens near
end of stroke.
Exhaust
manifold
Exhaust Stroke
Exhaust valve
open, exhaust
products are
displaced from
cylinder.
Intake valve
opens near end
of stroke.
Exhaust
Figure 3-1. The four-stroke, spark-ignition (SI) cycle. Four strokes of
180° of crankshaft rotation each, or 720° of crankshaft
rotation per cycle (Reference 3-4).
3-4
-------
A two-stroke cycle completes the power cycle in one revolution of
the crankshaft as compared to two revolutions for the four-stroke cycle
(see Figure 3-2). As the piston moves to the top of the cylinder, air or
an air and fuel mixture is compressed for ignition. Following ignition
and combustion, the piston delivers power as it moves down through the
cylinder. Eventually it uncovers the exhaust ports (or exhaust valves
open). As the piston begins the next cycle, exhaust gas continues to be
purged from the cylinder, partially by the upward motion of the piston and
partially by the scavenging action of the incoming fresh air. Finally,
all ports are covered (and/or valves closed), and the fresh charge of air
or air and fuel is again compressed for the next cycle.
Two-stroke engines have the advantage of higher horsepower-to-weight
ratio compared to four-stroke engines when both operate at the same
speed. In addition, if ports are used instead of valves, the mechanical
design of the engine is simplified. However, combustion can be better
controlled in a four-stroke engine and excess air is not needed to purge
the cylinder. Therefore, four-stroke engines tend to be slightly more
efficient, and may emit less pollutants (primarily unburned hydrocarbons)
than two-stroke engines (Reference 3-1).
3.1.4 Charging Method
Charging is the method of introducing air or the air and fuel
mixture into the cylinder. Three methods are commonly used: natural
aspiration, turbocharging and blower scavenged.
A naturally aspirated engine uses the vacuum created behind the
moving piston during the intake stroke to suck in the fresh air charge.
This process tends to be somewhat inefficient, however, since the actual
amount of air drawn into the cylinder is only about 50 to 75 percent of
the displaced volume (Reference 3-5). A more efficient method of charging
is to pressurize the air (or air and fuel) and force it into the cylinder.
This may be done with either a turbocharger or a supercharger. The
turbocharger is powered by a turbine that is driven by the energy in the
relatively hot exhaust gases, while a supercharger is driven off the
engine crankshaft. Air pressurization increases the power density, or
power output per unit weight (or volume) of the engine, since more air
mass can be introduced into the cylinder. As air pressure increases, its
temperature also rises because of the action of the compressor on the
3-5
-------
v Exhaust Valve
Fuel Injector
oo
en
Compression
A. Intake ports covered
B. Exhaust valve closes
Combustion Power Stroke
A. Fuel enters cylinder A. Piston moves down
by injection Q power delivered
B. Combustion by auto- to crankshaft
ignition
/ Scavenging
A. Air blown into cylinder
B. Exhaust gases purged
Figure 3-2. Cylinder events for a two-stroke blower-scavenged
1C engine (Reference 3^4).
-------
air. Therefore, the pressurized air is often cooled before entering the
cylinder to further increase power by allowing more air mass to be
introduced into the cylinder. This process is called interceding or
aftercooling.
Two-stroke engines are often air-charged by a blower, which also
aids in purging the exhaust gases. Such systems are called
blower-scavenged. This method is less efficient than turbocharging
because the blower produces less pressure than a turbine. However, high
volumetric flow rates are achieved, effectively purging the cylinder of
exhaust gases.
In a CI engine, fuel is injected into the cylinder near the end of
the compression stroke; whereas, in an SI engine, the fuel is usually
added to the air downstream of the turbocharger if any is used, and before
the mixture enters the cylinder. This is done with a carburetor.
However, some SI engines (particularly large natural gas fueled ones)
inject the fuel into the intake manifold just ahead of the valves, or into
the cylinder as done with CI engines.
Two methods of injection are commonly used. Direct injection
places the fuel directly into the cylinder and the principal combustion
chamber. These units are also called open chamber engines because
combustion takes place in the open volume between the top of the piston
and the cylinder. In contrast is indirect injection, where combustion
begins in a fuel rich (oxygen deficient) atmosphere in a smaller
antechamber and then expands into the cooler, excess air region of the
main chamber. These latter engines are also called divided or
precombustion chamber systems.
3.2 APPLICATIONS AND RANK
Stationary reciprocating 1C engines can be classified into four
characteristic size ranges, with common applications within each range.
3.2.1 Large Bore, High Power, Low- and Medium-Speed Engines
The large bore, high power (>75 kW/cylinder and less than 1000 rpm)
engines characteristically are four-cycle, compression ignition engines
designed to operate on either diesel oil or a mixture of oil and natural
gas (dual fuel) (Reference 3-1). Two cycle are also common. The
remainder of the engines are either spark ignited natural gas engines, or
engines which operate as either spark ignited or compression ignited (dual
3-7
-------
fuel or oil) and can easily be switched in the field in response to fuel
availability. These engines, therefore, typically find uses in industries
which seek best economy, and usually have sufficiently large installations
to provide several fuel types. Typical industries where these large bore
engines may be found are municipal electric power generation, oil and gas
pipeline transmission, and oil and gas production. In these industries,
the engine is run continuously. Based on 1976 data, only about 1000 to
2000 of these engines are sold per year, with a total production value of
$80 to $150 million (1976 dollars). Sales have generally been declining,
although sales of diesel engines for electric power generation are up
(References 3-2 and 3-4).
3.2.2 Medium Power, High Speed Engines
This size (7.5 to 75 kW/cyl and greater than 1000 rpm) class of
engines has the greatest variety, with some larger units equaling the
power of the large engine class. There are basic differences, however,
which separate the two groups. Large bore engines produce high power
output at low speeds due to their large displacement and consequent high
power per cylinder. Medium bore engines, in contrast, have lower power
per cylinder (and therefore more cylinders for the same engine
horsepower). They achieve high outputs by utilizing higher rotative
speeds. Thus, for the same power rating, high speed engines are smaller,
less expensive, and capable of running at a wider range of speeds. The
smaller engines within this size range are produced by truck and tractor
engine manufacturers and therefore can be considered mobile engines
modified for stationary installation and constant speed operation
(Reference 3-2). Fuels burned are typical mobile fuels, either diesel oil
or gasoline, although there are a very few (usually modified) natural gas
engines within this size range.
These engines are used in miscellaneous industrial, commercial,
nonpropulsive marine, and agricultural applications where shaft power is
needed and electric motors cannot be used. Unlike the large engines which
are typically sold directly to the user, these engines are most often sold
to Original Equipment Manufacturers who purchase engines for their
manufactured generators, compressors or pumps, or other equipment, and
sell the complete package to the user (References 3-1, 3-6, 3-7).
3-8
-------
Precise production data are unavailable because the number of
mobile engines modified for stationary use is unknown. However, in 1976,
sales of diesel medium power, high speed engines for stationary
applications have been estimated in the range of 60,000 to 80,000
units/year over recent years, with a total value of $150 to $200 million
per year (1976 dollars, FOB plant) (Reference 3-1). Annual sales for
gasoline medium power, high speed engines have been approximately 100,000
with a total value of $50 million (FOB plant) (1976). Sales of these
engines have been erratic from one year to the next, but the general trend
seems to be upward. The largest sales gains have been in the 1000 kW to
3500 kW range, for OEM generator packages (Reference 3-8).
3.2.3 Small Engines and Very Small Engines
Small engines are distinguished from the other engines in that they
are mostly one- and two-cylinder engines of less than 40 horsepower.
These engines are mostly diesel and gasoline, one- and two-cylinder
models, with some four-cylinder models. Almost all have four strokes per
power cycle and usually are air cooled.
Small engines are typically used in generator sets, small pumps and
blowers, off-the-road vehicles, and refrigeration compressors for trucks
and railroad cars. Very small engines find additional uses for lawn and
garden equipment, chain saws, and recreational vehicle generators. Sales
of small engines are in the range of $400 to $500 million per year
(1976). Diesel engines used for drip irrigation systems have been showing
the largest gain recently (Reference 3-9).
3.2.4 Engine Users
On the basis of installed capacity, the principal stationary
applications of 1C engines are: oil and gas pipelines, oil and gas
production, general industrial (including construction), electrical power
generation, and agriculture.
Pipeline installations are concentrated in the oil and gas
producing states on the Gulf Coast and in the Midwest. Electric
generation includes base load generation, principally for municipalities
in the plains and Midwest, peaking power, emergency standby for vital
public services and large buildings in urban areas, and remote generation
(for mines and homes) in rural areas. Agricultural applications are
primarily irrigation pumping, concentrated in those areas with irrigated
3-9
-------
farm lands (such as central and southern California, Arizona, and
northwestern Texas). Additional agricultural applications include frost
control, harvesting (auxiliary engines), and some remote electric
generation. Construction applications include portable compressors,
welders, pumps, electric generation, and material handling equipment.
3.3 TYPICAL ENGINES IN EACH SIZE CLASS
The previous section identified three engine size ranges. This
section summarizes those mechanical aspects considered typical of each
size category.
3.3.1 Large Bore, High Power Engines
These engines are almost always designed for permanent installation
and rarely sold as a package. They typically are multicylinder in "V" or
inline configurations, with large displacements per cylinder and
correspondingly high cylinder power (>75 kW/cyl). Rotational speeds are
less than 1000 rpm. Fuels for these engines are either diesel oil or
natural gas, with some engines capable of burning either or both fuels
(requiring at least 5 percent diesel oil). Engines can also be modified
to run on other fuels if the cost of the alternate fuel is less.
Large bore engines are always designed to give best economy (except
for emergency standby use) and therefore are usually turbocharged and
aftercooled to increase the air charge to the cylinders. Efficiencies
(energy out/energy in) are typically 35 to 40 percent based on the lower
heating value of the fuel (Reference 3-1).
3.3.2 Medium Bore, Medium Power Engines
Most medium power stationary 1C engines are modified heavy duty
mobile engines, although some manufacturers' units are only for stationary
use (Reference 3-2). Because sales are typically to small industries
which do not have extensive fuel supplies or process facilities or sales
for use in remote areas, most installations are packaged self-contained
units. This allows these engines to be semi-portable. Like the large
bore engines, these engines are multicylinder in "V" or inline
configurations, but have much smaller displacements and power per cylinder
(7.5 kW to 75 kW). Engine speed is usually greater than 1000 rpm.
Fuels are typical of mobile engines, either diesel oil or
gasoline. Engines are often turbocharged to increase thermodynamic
efficiency to as high as 38 percent. Naturally aspirated engines in this
3-10
-------
size range have typical efficiencies of 30 percent (References 3-1, 3-2,
3-6).
3.3.3 Small Bore Engines
Small engines are typically one- or two-cylinder models and are
usually air cooled. They are sold in packaged units or separately, and
rarely are permanently installed. Displacements per cylinder are small,
with total engine output less than 75 kW. Depending on use, engine speed
ranges from 500 rpm to over 3000 rpm.
Most of these engines are naturally aspirated. The prime fuel is
gasoline, although there are several manufacturers of diesel engines in
this size range. Efficiencies of these engines are generally low, ranging
from 25 percent to 35 percent (References 3-6 and 3-10).
3-11
-------
REFERENCES FOR SECTION 3
3-1. "Stationary Internal Combustion Engines. Background Information:
Proposed Standards," EPA 450/3-78-125a, April 1979.
3-2. Roessler, W. U., et al., "Assessment of the Applicability of
Automotive Emission "Control Technology to Stationary Engines,"
EPA-650/2-74-051, NTIS-PB 237 115, July 1974.
3-3. Salvesen, K. G., et £l_., "Emissions Characterization of Stationary
NOX Sources: Volume I. Results," EPA-600/7-78-120a, NTIS PB 284
520, June 1978.
3-4. Obert, E. F., "Internal Combustion Engines," Intext Educational
Publishers, New York, NY, 1973.
3-5. Baumeister, T., ed., Marks' Standard Handbook for Mechanical
Engineers, Seventh Edition, McGraw-Hill, New York, NY, 1976, pp.
9-144.
3-6. Letter from C. I. Taggart, White Motor Corporation, Eastlake, OH,
to R. L. Seiffert, U.S. Environmental Protection Agency,
Hopkins, MN, January 1975.
3-7. Hanley, G. P., "Impact and Potential for Improvement for Stationary
Engines," General Motors Statement at California Air Resources
Board Workshop on Suggested Rules for the Control of NOX
Emissions in the South Coast Air Basin, Los Angeles, CA, May 1978.
3-8. Diesel and Gas Turbine Progress Worldwide, Vol. X, No. 8, October
1978, p. 10.
3-9. Diesel and Gas Turbine Progress Worldwide, Vol. IX, No. 8, October
1977, p. 22.
3-10. Letter from W. Bazen, Briggs & Stratton Corporation, Wauwatosa, WI,
to S. Youngblood, Acurex Corporation, Mountain View, CA, 1974.
3-12
-------
SECTION 4
CHARACTERIZATION OF INPUT MATERIALS, PRODUCTS, AND WASTE STREAMS
Stationary reciprocating internal combustion engines are sources of
NO , HC, CO, and particulate emissions. Sulfur oxide emissions are also
A
possible if the fuel burned contains sulfur. Most of these pollutants are
formed during the combustion process and emitted through the engine
exhaust. However, some emissions also escape from the engine crankcase,
and others from fuel evaporation. This section presents the typical
emission levels from stationary engines from these sources, then describes
the effects of atmospheric conditions, fuel, and engine operation on these
emissions.
4.1 BASELINE EMISSIONS
Because stationary engine sizes vary widely, emissions have always
been presented on a mass-to-power basis. This allows comparisons of
emissions between engines of various sizes. In addition, since some
engines have already had emission controls applied, emissions are
presented as controlled or uncontrolled.
Typical uncontrolled exhaust emission rates, based on the heavy
duty engine 23-mode composite (see Federal Register, September 8, 1977,
page 45132), for four-stroke naturally aspirated gasoline engines in
g/brake kWh are: 3.2 to 17.6 for HC; 40 to 120 for CO; and 10.6 to 18.6
for NO (Reference 4-1). The 23-mode composite tests includes steady
/\
and transient operation. Similarly, brake specific emission rates from
all types of spark ignited natural gas fired engines (two- and
four-stroke, naturally aspirated and turbocharged engines) are: 0.5 to
9.4 for HC; 0.3 to 38.0 for CO; and 10 to 39 for NO (Reference 4-1).
A
For compression ignition engines, typical emission rates for all
types of uncontrolled diesel engines (including four-stroke, either
naturally aspirated or turbocharged and open- or divided-chamber, and
4-1
-------
two-stroke, either blower-scavenged or turbocharged) in g/brake kWh are:
0.1 to 3.9 for HC; 0.4 to 20 for CO; and 2.8 to 23 for NO
X
(Reference 4-1).
Average emission factors have been developed for engines of various
sizes and fuel types and are presented in Table 4-1 (References 4-1 and 4-2).
As previously mentioned, the engine exhaust is not the only source
of emissions. Hydrocarbons escape from the crankcase because of blowby
(gases which are vented from the oil pan after escaping from the cylinder
past the piston rings) and from the fuel tank and carburetor due to
evaporation. For diesel engines, crankcase blowby is minor (0.24 g/kWh
HC + N0v, 0.0084 g/kWh CO) because hydrocarbons are not present during
A
compression of the charge (Reference 4-1). Evaporative losses are also
insignificant due to the low volatility of diesel fuels. Evaporative
losses are also negligible in engines using gaseous fuels because these
engines usually receive their fuel continuously from a pipe rather than
via a fuel storage tank and fuel pump. However, in gasoline fueled
engines 20 to 25 percent of the total hydrocarbon emissions from
uncontrolled engines come from crankcase blowby and another 10 to 15
percent from evaporation of the fuel in the storage tank and the
carburetor (divided approximately equally between the two). Crankcase
blowby emissions can be virtually eliminated through the use of positive
crankcase ventilation as is demonstrated in the case of mobile engines.
Finally, oil and cooling water that are replaced during maintenance
represents additional waste streams that are periodically discharged.
These wastes need to be disposed of properly. However, combustion
modifications should not have any major effects on these waste streams,
and therefore disposal of these wastes is not discussed in this report.
4.2 EMISSIONS FORMATION
Oxides of nitrogen can be formed either from atmospheric nitrogen
in the combustion air (thermal NO ) or from nitrogen compounds in the
A
fuel itself (fuel NO ). The amount of thermal NO formed depends on
X X
the temperature, oxygen concentration, and residence time of the nitrogen
at high temperatures and/or oxygen concentration. Fuel NO formation
X
appears to be directly proportional to the amount of nitrogen compounds in
the fuel. For burner (continuous) type combustion, it has been found that
4-2
-------
TABLE 4-1. EMISSIONS FACTORS FOR 1C ENGINES, g/kWha
(References 4-1, 4-2)
Fuel
Gasoline
Diesel
Natural gas
Dual Fuel
>15 kW
<15 kW
>375 kWb
<375 kWc
NOX
11.9
7.5
17.3
16.6
15.4
11.0
CO
137
395
2.4
6.0
3.8
2.7
HC
11.2
27.5
0.6
2.8
6.5
4.1
aEmission factors for gasoline and diesel engines are
modal averages; those for natural gas and dual fuel
are for rated conditions. The modal averages includes
numbers from the constant output part of mobile
transient tests.
bBased on an average of rated condition levels from
engines considered,
cWeighted average of two- and four-stroke engines.
Weighting factors = 2/3 for four-stroke and 1/3 for
two-stroke.
4-3
-------
as fuel nitrogen compounds increase, the NO also increases, although
A
not as rapidly. The temperature, oxygen concentration, and residence time
also influence how much fuel nitrogen is converted to NO . Because of
A
the large amount of NO produced by an 1C engine as compared to a
/\
distillate oil or natural gas burner, most of the NO emitted from an 1C
A
engine is probably thermal NO . For example a residual oil boiler
A
burning oil with a nitrogen content of 0.5 percent would have a NO
A
level of about 350 ppm at 3 percent 0-, while a diesel engine burning a
diesel oil with a nitrogen content with less than 0.1 percent N, could
have a NO level of 3000 ppm at 3 percent 0_.
A £
Most of the other pollutants, HC, CO, and smoke are mainly the
result of incomplete combustion. Hydrocarbon emissions are believed to be
caused by three general mechanisms (Reference 4-3): wall quenching (fuel
impingement on the walls causing the fuel to be cooled below the
combustion temperature), variations in engine variables (mixing inside the
cylinder, wrong air-to-fuel ratio, defective ignition, etc.), and, in
two-cycle engines cooling down the exhaust gases by the scavenging air
before combustion is completed. CO emissions can also form by the same
general mechanisms.
Smoke formation is also related to incomplete combustion. The
color of the smoke indicates the cause of the smoke (Reference 4-1).
Bluish smoke occurs by incomplete combustion of crankcase oil forced past
worn piston rings into the cylinder. White smoke usually occurs at low
load or idle conditions and is mainly unburned liquid fuel or lubricating
oil. Black smoke consists of carbon particles formed by incomplete
combustion at high temperatures.
The other engine emissions such as SO , lead and other metals are
A
directly related to the amount of these compounds in the fuel.
Atmospheric conditions, fuels used, and engine design and operation
also affect emissions. These effects are described in the following
sections.
4.3 ATMOSPHERIC EFFECTS
The effects of the atmospheric conditions on NO emissions have
A
been evaluated by several sources, predominately by or for automotive
engine manufacturers. Their test results indicate changes in NO of up
A
to 25 percent caused by ambient temperature and humidity changes, and up
4-4
-------
to 40 percent changes in NO emissions caused by ambient pressure
A
changes (Reference 4-1). Most of these effects are caused by changes in
the air-to-fuel ratio as the density of the combustion air changes.
However, humidity has an additional effect on NO in that high moisture
A
conditions reduce the peak temperatures within the engine cylinder,
decreasing NO emissions (References 4-3 through 4-8).
A
Because one of the variables of engine design is the air-to-fuel
ratio, different engines respond differently to changes in atmospheric
conditions. Thus it is quite difficult to quantify atmospheric effects on
engine emissions. However, these general effects have been observed for
engines operating close to stoichiometric conditions:
t Increases in humidity decrease NO emissions
A
Increases in temperature increase HC and CO emissions
Decreases in pressure increase HC and CO emissions.
It should be noted that most large engines are not operated close to
stoichiometric conditions.
4.4 FUEL EFFECTS
Generally the difference in NO emissions from large bore engines
A
operating with any of the primary fuels (diesel oil, gasoline, natural
gas) is small. This is because the main source of NO is N in the
X £
combustion air, rather than from the fuel.
However, there are significant differences in other pollutants.
Hydrocarbons from natural gas engines can be five times greater than for
diesel oil, although most of the emissions are methane. On the other
hand, SO and particulate emissions are much greater with diesel oil.
A
CO emissions are generally always high with gasoline. Characteristics of
each primary fuel and its effects on emission are discussed in the
following subsection.
4.4.1 Diesel Oil (No. 2)
Diesel oil is an important fuel for stationary reciprocating 1C
engines, since its use is increasing as users become concerned about
natural gas supplies. Specifications for use in engines are well defined,
as indicated in Table 4-2.
Diesel fuel composition is complex and hence the exhaust may
contain between 9,000 and 12,000 different compounds (Reference 4-10). As
indicated, diesel fuels contain impurities such as sulfur (up to 10 times
4-5
-------
TABLE 4-2. SPECIFICATIONS FOR DIESEL FUELS (ASTM D975) (Reference 4-9)
Test
Limit
Flashpoint, K (°F), min
Water and sediment, vol percent,
max
Viscosity, kinematic, centistokes,
311K (100°F)
min
max
Carbon residue, wt percent, max
Ash, wt percent, max
Sulfur, wt percent, max
Ignition quality, cetane number, min
Distillation, temp, deg K (°F)
90% evaporated
min
max
325(125)
0.10
2.0
5.8
0.35
0.02
1.0
40
555(540)
575(576)
4-6
-------
more than gasoline Reference 4-10), ash, and carbon residue. Some fuel
bound nitrogen may be included, and detergents and metal based additives
may be added. All these fuel components affect emissions. The presence
of sulfur and nitrogen in the fuel affects SO and NO emissions
xx
respectively although the maximum effect of fuel bound nitrogen is
probably less than 15 percent (i.e., at least 85 percent of the NO is
A
thermal Reference 4-1). Probably the most critical parameter
indicating the fuel effects of diesel oils is the cetane number, or
ignition quality. Increase in cetane number decreases NO , HC, and CO,
/\
but increases smoke through higher carbon formation (Reference 4-11).
This is because the hydrogen of large fuel molecules burns away, leaving
carbon which cannot be oxidized during the rapidly decreasing temperatures
within the cylinder. Barium based additives can reduce smoke, but often
increase other pollutants when impurities or deposits form.
4.4.2 Gasoline
Gasoline is used mostly for mobile engines, and emissions from
mobile sources have been well documented. But gasoline has found use in
stationary applications because many medium and small engines are either
converted mobile engines or designed specifically for light stationary
use. Thus gasoline, although not a significant fuel on an installed
horsepower basis, is an important stationary fuel.
Like diesel oil, gasoline is a blend of many hydrocarbons. Because
this fuel is more volatile than diesel oil, a characteristic of this fuel
is an evaporative emission of 5 percent of the total HC emitted.
Composition effects related to the H/C ratio of the fuel have been
observed. The H/C ratio affects both NO and HC emissions, with NO
A A
being decreased as H/C increases because of changes in adiabatic flame
temperature. HC emissions also decrease as H/C increases, probably due to
changes in volatility and its effect on fuel combustion (References 4-3
and 4-12).
Fuel additive effects on emissions have also been observed.
Tetraethyllead (TEL), an additive for antiknock purposes, is a major
source of particulate, with smoke emission levels 10 to 20 times those for
unleaded fuels. Lead additives also increase HC emissions (References 4-11,
4-12, 4-13).
4-7
-------
CO emissions from gasoline are higher than from any other fuel.
However, these are not necessarily caused by fuel composition, but are a
function of stoichiometry, or combustion in fuel rich zones in the engine
cylinder. Because gasoline engines generally operate close to a
stoichiometric fuel ratio, CO emissions are therefore higher than emissions
from other more lean burning fueled engines (References 4-3, 4-12).
4.4.3 Natural Gas
Natural gas is used more than any other fuel for stationary
engines, although usage has been declining in recent years. There are no
specific specifications for natural gas because properties vary by
location. Natural gas is typically 85 percent methane with the rest of
its composition consisting of other low boiling hydrocarbons, hydrogen,
and nitrogen. NO emissions from natural gas fired engines are very
A
similar to diesel engine emissions, primarily because natural gas engines
and diesel engines often operate under similar conditions. However, total
hydrocarbons are five times greater and CO is half as much in gas fired
engines when compared to diesel engines. Most of the increase in HC
emissions is due to blowby and fugitive sources since the fuel is
gaseous. However, typically 80 percent of the HC emission is methane, a
noncriteria pollutant. CO emissions are less because usually the gaseous
fuel improves mixing within the engine cylinder, and oxidation is more
complete (Reference 4-1).
4.4.4 Other Fuels
Other gaseous and liquid fuels may be used for stationary engines.
Of course, emissions will vary depending on composition, additives, etc.
Little data are available for other fuel types, thus it is difficult to
predict general emission characteristics from other fuels.
4.5 EFFECTS OF ENGINE DESIGN AND OPERATION
Almost any variation in engine design or operation parameters will
affect emissions. These parameters may be divided into four classes:
(1) charging methods and air-to-fuel ratio; (2) engine combustion cycle;
(3) combustion chamber design and valve and ignition timing; and
(4) operating conditions of load and speed.
4.5.1 Charging Method and Air-to-Fuel Ratio
Charging method does not directly affect emissions, but it
influences the air-to-fuel ratio(A/F). Air-to-fuel ratio, in turn,
4-8
-------
affects emissions significantly. These effects are best surmiarized in
Figure 4-1. At air-to-fuel ratios below stoichiometric (rich), combustion
occurs under conditions of insufficient oxygen and thus unburned hydrocarbon
emission increase. CO increases because carbon is not sufficiently oxidized
to C09. NO decreases both because of insufficient oxygen and lower
£. A
temperatures.
At air-to-fuel ratios above stoichiometric (lean), combustion
occurs under conditions of excess oxygen, thus essentially all carbon is
oxidized to C0_. NO first increases rapidly with A/F near
£- X
stoichiometric, because of the excess oxygen and peak temperatures, then
decreases rapidly with A/F as the excess air cools peak combustion
temperatures. Hydrocarbons stay at a low level, then begin to increase as
the air-to-fuel ratio is increased because the lower temperatures inhibit
combustion.
Charging method is important because it often limits the range of
air-to-fuel ratio. Naturally aspirated carbureted engines generally must
operate with overall air-to-fuel equivalence ratios, defined as
^Whiometric/^actual' 9reater than °'7 because poor
distribution among cylinders will allow some cylinders to go excessively
lean. In contrast, turbocharged fuel injected engines, with precise
control of air-to-fuel ratio to each cylinder, can operate at equivalence
ratios of 0.5 to 0.3 without increasing hydrocarbon emissions
significantly. Some blower scavenged engines operate at equivalence
ratios below 0.25, although the actual ratio inside the cylinder is
usually higher (Reference 4-3).
The choice of lean or rich operation often depends on engine use.
Rich operating (meaning close to stoichiometry) engines give quicker
response to changing conditions, and also produce maximum power. In
addition, carbureted engines generally cost less than a comparable
turbocharged model. Thus, these engines typically find use in
construction and lightweight general industrial service.
Lean burning engines are much more economical to operate.
Therefore, large industry applications where the engine will be operated
for several thousand hours a year at constant load and speed typically
have lean burning engines.
4-9
-------
"aj
o>
|
w»
i/i
10
"oj
20
22
Air-to-fuel ratio
kg air/kg gasoline
Figure 4-1. Effect of air-to-fuel ratio
engine (Reference 4-1).
on
emissions from a gasoline
4-10
-------
4.5.2 Combustion Cycle
As discussed in Chapter 3, reciprocating 1C engines may be either
two- or four-stroke cycle. During combustion, emissions from either type
are essentially identical. However, during the charging of a two-cycle
engine, several events can take place. On noninjected engines, the
scavenging air, which purges the cylinder of exhaust gases and provides
the combustion air, can also sweep out part of the fuel charge. Thus
carbureted two-cycle engines often have higher HC emissions in the form of
unburned fuel.
The two-stroke engine can also have lower NO emissions. If the
y\
cylinder is not completely purged of exhaust gases, the result is internal
exhaust gas recirculation (EGR). The remaining inert exhaust gases absorb
energy from combustion, lowering peak temperatures, and thereby lowering
NOX.
4.5.3 Combustion Chamber Design and Ignition Timing
Almost any variation in cylinder design, valving, or ignition
timing will affect emissions. Unfortunately the effects cannot be
quantified since each engine is different and some design variables will
cancel any beneficial effects of others. However, some generalization can
be made. Design variables which improve mixing within the cylinder tend
to decrease emissions. Improvements in mixing may be accomplished through
swirling the air or fuel and air mixture within the cylinder, improving
the fuel atomization, and optimizing the fuel injection locations.
Decreasing the cylinder compression ratio reduces the maximum temperatures
achieved in the cylinder, lowering NO emissions.
/\
Stratifying the charge into a fuel rich zone and fuel lean zone
also reduces peak combustion temperature and NO . This usually requires
/\
a small antechamber apart from the cylinder itself. Fuel is injected or
combustion initiated in the smaller (rich) chamber, then expands into the
lean main chamber (cylinder). Combustion occurs slower with a lower
pressure rise, thereby reducing peak temperatures.
Another technique of reducing temperature is to retard spark or
injection timing. This is recognized as an effective NO control
A
technique. By initiating combustion later in the cycle, pressures are
reduced, thus lowering peak temperatures. However, there is a fuel
4-11
-------
penalty of 5 to 8 percent and the potential of excessive smoke from some
engines (Reference 4-1).
Finally, engine size affects emissions, and is another variable of
design. For instance, large bore, low- to medium-speed engines,
independent of the fuel type (diesel, dual fuel, natural gas), are
designed for low fuel consumption (i.e., high thermal efficiency). Their
low rotational speed maintains the products of combustion near peak
temperature for relatively long periods of time, so that fuel injection
timing may be adjusted for optimum results. The larger engines also tend
to have higher NO emissions. The large combustion volumes also allow
A
more aerodynamic design freedom than is available in small- or medium-bore
engines. Furthermore, these large engines are designed to operate under
steady conditions and almost always at more than 50 percent of rated
power; therefore, they do not have to contend with acceleration/deceleration
or low power requirements .
4.5.4 Load and Speed Effects
Load and speed effects on emissions also vary from engine to
engine. In general, diesel compression ignition engines exhibit
decreasing brake specific emissions of NO with increasing load at
A
constant speed. This is partly caused by changes in the air/fuel ratio.
Some turbocharged engines show the opposite effect of increasing brake
specific NO emissions as load increases. CO emissions first decrease
A
with increasing load (equivalent to increasing temperature) and then
increase as maximum load is approached. Brake specific HC emissions,
decrease with increasing load as a result of increasing temperature, but
smoke emissions reach their maximum at full load (References 4-1, 4-3,
4-11).
Natural gas engines follow the same trends as diesel engines for HC
and CO, but generally have maximum NO at maximum power.
A
Gasoline engine results vary greatly, but generally show the same
trends as above for HC and CO, with NO peaking at some intermediate
A
load. Small (less than 11 hp) SI engines (gasoline) exhibit relatively
high HC and CO emissions, particularly at low load operation. Speed
effects generally will decrease HC and CO, and increase NO emissions.
However, speed will also affect other design and operating variables which
may reverse the positive effects (References 4-1, 4-4, 4-11).
4-12
-------
4.6 PRODUCTS CHARACTERIZATION
The principal product of an 1C engine is shaft power. The exhaust
gas from an engine can also be considered a "product." The hot exhaust
gas can be used to supply waste heat via a heat recovery device or be used
in a supplementary fuel fired waste heat boiler. The product exhaust gas
from an 1C engine has been discussed earlier.
4-13
-------
REFERENCES FOR SECTION 4
4-1. "Stationary Internal Combustion Engines: Background Information
Proposed Standards," EPA 450/3-78-125a, April 1979.
4-2. Youngblood, S. B., and G. R. Offen, Acurex Internal Memorandum,
"Emissions Inventory of Currently Installed Stationary
Reciprocating Engines," Acurex Corp., Mountain View, CA, September
23, 1975.
4-3. Patterson, 0. J., and N. A. Henein, "Emissions from Combustion
Engines and Their Control," Ann Arbor Science Publishing, Inc., Ann
Arbor, MI, 1972.
4-4. Grinberg, L., and L. Morgan, "Effects of Temperature on Exhaust
Emissions," SAE 740527, June 1974.
4-5. Robson, J. A., "Humidity Effects on Engine Nitric Oxide Emissions
at Steady State Conditions," SAE 700467, May 1970.
4-6. "Effect of Humidity of Air Intake on Nitric Oxide Formation in
Diesel Exhaust," CRC Report No. 447, Coordinating Research Council,
Inc., New York, NY, December 1971.
4-7. Krause, S. R., "Effect of Engine Intake-Air Moisture, Temperature,
and Pressure on Exhaust Emissions," Ethyl Corp., Richmond, VA,
December 1970.
4-8. Brown, W. J., .et jil., "Effects of Engine Intake-Air Moisture on
Exhaust EmissionsT^SAE 700107, January 1970.
4-9. Baumeister, T., ed., Marks' Standard Handbook for Mechanical
Engineers, Seventh Edition, McGraw-Hill, New York, NY 1976.
4-10. Ketcham, B., and S. Pinkwas, "Diesels and Man, Are We Creating a
New Environmental Problem by Solving an Old One?" New Engineer,
Vol. 7, No. 4, pg. 23, April 1978.
4-11. Roessler, W. II., et al., "Assessment of the Applicability of
Automotive Emissions Control Technology to Stationary Engines,"
EPA-650/2-74-051, NTIS-PB 237 115, July 1974.
4-12. Harrington, J. A., and R. F. Shishu, "A Single-Cylinder Engine
Study of the Effects of Fuel Type, Fuel Stoichiometry, and
Hydrogen-to-Carbon Ratio on CO, NO, and HC Exhaust Emissions," SAE
730476, May 1973.
4-13. Salvesen, K. G., et al., "Emissions Characterization of Stationary
NOX Sources: VolumeTT Results," EPA-600/7-78-120a, NTIS-PB 284
520, June 1978.
4-14
-------
SECTION 5
PERFORMANCE AND COST OF CONTROL ALTERNATIVES
Emission controls for stationary reciprocating internal combustion
engines can be divided into two classes; those which reduce NO
A
emissions and those which reduce other pollutants. This is because the
formation mechanisms which generally produce minimum HC, CO, and smoke
emissions (high temperatures) often produce peak NO emissions, and vice
A
versa. Thus tradeoffs between emission levels of NO and the other
A
pollutants should be evaluated before any controls are applied.
The baseline emission levels reviewed in Section 4.1 indicate that
NO emissions for diesel oil and natural gas are at least 2 to 3 times
A
greater than other emissions. Gasoline engines also emit high NO
A
levels in addition to high levels of CO. Thus NO is an essential
A
emission control for all engines, and CO is important for stationary
gasoline engines. HC emission control for very small engines may also be
important.
From a nationwide emission standpoint, however, CO emissions become
less important. Stationary engines account for approximately 3 percent of
the total nationwide CO, versus nearly 7 percent for NO (Table 5-1)
A
(Reference 5-1). Furthermore, large engines (greater than 75 kW/cyl)
account for over 75 percent of all NO emitted from stationary engines.
X
Also CO or HC emissions can be lowered by efficient operation, while NO
A
reduction is harder to achieve. In addition, these large engines tend to
be in a class by themselves (large bore, low rpm, essentially no
transients), while the medium to small engines usually are modified mobile
engines (smaller bore, greater rpm, and operate under more transient
conditions). This division is not complete since some of the medium bore
engines are as large as the large bore engines. Since much has been
written about mobile engines and the larger size stationary engines are
the main source of pollutants, this section will concentrate on these
5-1
-------
TABLE 5-1. EMISSIONS FROM 1C ENGINES (1975)'
Size
>5700 cc/cylinder
375 kW to 5700 cc/cyl
75 kW to 375 kW
11 kW to 75 kW
11 kW
Total
All Stationary Sources
All Sources
Emissions (10^ metric tons/year)
NOX
1.15
0.02
0.16
0.16
0.04
1.53
14.5
23.8
CO
0.27
0.01
0.35
0.88
1.95
3.47
32.6
105.8
HC
0.45
0.00
0.05
0.08
0.14
0.72
10.4
24.2
Reference 5-1
larger engines. Also, since NO is the major pollutant from these
A
engines, the following sections focus on emission reduction controls and
their effectiveness on oxides of nitrogen. The performance and cost of
these controls are estimated.
5.1 NOV CONTROL TECHNIQUES
A
There are several demonstrated N0x control techniques for large,
stationary reciprocating internal combustion engines. Most of these
reduce NOX emissions by lowering the peak temperatures within the engine
cylinder or altering the burning rate. These techniques include changes
in air-to-fuel ratio or ignition timing, manifold air cooling,
modifications to the combustion chamber, exhaust gas recirculation, and
water injection. In addition, for many engines, derating the engine power
5-2
-------
is effective. Flue gas treatment controls such as catalytic reduction are
under development.
Each control technique, its effectiveness, effects on other
emissions, and operation and maintenance impacts are discussed in the
following sections. All data are from Reference 5-1 except as noted.
5.1.1 Air-to-Fuel Ratio Changes
The air-to-fuel (A/F) ratio is defined as the mass flowrate of air
divided by the mass flowrate of fuel. This ratio is termed stoichiometric
if precisely enough oxygen is present in the mixture to just completely
oxidize the fuel. Effects of air-to-fuel ratio changes were described in
Section 4, and were summarized in Figure 4-1.
When the engine is operated rich, the lack of excess oxygen
suppresses NO formation and so, despite the high cylinder temperatures,
/\
NO formation will drop sharply at increasingly rich mixtures. If the
/\
ratio is varied in the lean direction, the excess oxygen level will
increase but so will the mass of the combustion mixture, enabling it to
absorb more heat. This reduces the peak temperature, resulting in lower
NO formation.
/\
The most practical use of air-to-fuel ratio adjustment as a control
technique is to change the setting toward leaner operation, since
increasingly rich mixture operation increases both HC and CO emissions and
fuel consumption. This technique is better suited to injection type
engines. Carbureted engines will require better control of the
air-to-fuel ratio between cylinders before they can operate at leaner A/F
ratios. In fact, some current carbureted engines, when adjusted to leaner
than normal air-to-fuel ratios (but still rich), tended to increase their
NO emissions because they were moving towards the peak of the NO
X A
versus air-to-fuel ratio curve. They were not able to go beyond that
point to lower NO levels in the lean region without misfiring
/\
(Reference 5-2).
Injection type engines achieve leaner air-to-fuel ratios by either
reducing the fuel input (essentially derating) or by increasing the air
input. More air flow can be accomplished by either installing a
turbocharger or increasing the capacity of an existing turbocharger. For
blower scavenged engines, the same is true. These changes are limited by
increased parasitic horsepower which could increase fuel consumption.
5-3
-------
Also the chance of misfiring becomes greater, which may require more
sophisticated control of the air-to-fuel ratio such as Op sensors in the
exhaust and a feedback control.
For carbureted engines, air flows can be increased by changing the
venturi and fuel nozzles (Reference 5-3). Also, some modifications to the
intake manifold may be required to allow a uniform fuel-air mixture
distribution to each cylinder. The difficulty with distributing a uniform
mixture to each cylinder via an intake manifold is why injector type
engines allow leaner air-to-fuel ratios.
Figure 5-1 shows the effect of air-to-fuel ratio change on N0x
emissions and fuel consumption for diesel, gas, and dual fuel engines.
Small changes in the air-to-fuel ratio, approximately 10 percent,
generally reduce NO by about 30 percent with a fuel penalty of less
A
than 5 percent. As discussed earlier, the limit to changes in the
air-to-fuel ratio (i.e., the maximum ratio) is set by increased fuel
consumption, onset of misfiring, and increases in organic emissions.
Tests on installed gas pipeline engines have shown that increased
air-to-fuel ratio is the most significant source of NO reductions from
A
large bore gas engines (Reference 5-6).
5.1.2 Retarded Ignition Timing
Ignition in a normally adjusted engine is set to occur shortly
before the piston reaches its uppermost position (top dead center or
TDC). At TDC the air or air-fuel mixture is compressed to the maximum.
The timing of the start of injection or of the spark is given in the
number of degrees that the crankshaft must still rotate between this event
and the arrival of the piston at TDC. The extent of retard is then
expressed in degrees relative to normal ignition. Typical retard values
achievable are 2° to 6°, depending on the engine. Beyond these
levels, fuel consumption increases rapidly, power drops, misfiring
(erratic ignition) occurs, and smoke from diesel engines becomes excessive
(Reference 5-4). When ignition is retarded, the combustion process
duration does not change significantly but extends longer to the power
stroke. Thus combustion proceeds at a lower peak temperature, but raises
the exhaust temperature. This lower peak combustion temperature decreases
NO formation but does not affect CO and HC emissions, unless the
A
temperature is lowered significantly. Smoke from diesel engines may
5-4
-------
1.20
1.10
c
O 3
0 £ 1.00
-------
increase due to this lower combustion temperature. The higher exhaust
temperature could cause the exhaust valves and manifolds to require more
maintenance, or possibly replacement with ones that withstand the higher
temperatures. This higher exhaust temperature could also affect the
turbocharger.
Figure 5-2 shows the effect of retard on N0x emissions and fuel
consumption for diesel engines; Figure 5-3 presents the same information
for dual fuel and gas engines. On the average, diesel engines reduce
NO by 25 percent for 4° of retard and 40 percent for 8° of retard.
n o
Fuel usage increases approximately 2 percent at 4 of retard, while 8
of retard raises fuel usage by about 6 percent. Gas and dual fuel engines
show similar trends except the data are more scattered. Based on the
limited data available, retard appears to be a more effective technique to
reduce NO for dual fuel engines than gas engines.
/\
5.1.3 Manifold Air Cooling and Turbocharginq
Depending on how the engine is equipped, reducing manifold air
temperature requires some or all of the following devices:
t Aftercooler or intercooler
t Coolant circulation device (fan for air or pump for water)
Cooling tower or larger radiator if water cooled
t Temperature control mechanism.
Most turbocharged engines greater than 375 kW have an intercooler or an
aftercooler which are heat exchangers between the turbocharger and intake
manifold. Since compressing (turbocharging) air heats the air, cooling
the air and thus increasing its density before it enters the cylinder
allows a greater air mass flow rate. This in turn permits a higher fuel
flowrate and thus more engine power output. Aftercooling of the air to
the cylinders also reduces the peak combustion temperature. Aftercooling
is required on a turbocharged, carbureted natural gas engine to prevent
the hot gases from detonating the air fuel mixture before its entry into
the cylinder (Reference 5-4).
Adding an aftercooler to a turbocharged engine or increasing the
amount of aftercooling (lowering air temperature to the intake manifold),
decreases the peak combustion temperature and lowers NO emissions.
A
5-6
-------
1.20
4->
I ^
3 4-1
V) C
C O
S g
1.10
(O Ll_
t- 1/1
CD CD
1.00
#26 '
Diesel
2 Stroke-blower scavenged
2 Stroke-turbocharged
4 Stroke-naturally aspirated
4 Stroke-turbocharged
177
246
Retard,
10
12
14
160
16
1.0
0.8
0.6
X X
ii
0.4
0.2
6 8 10
Retard, °
12
Baseline NO
g/kklh x
14
i^26 )
. >Same engine
16
Figure 5-2. Effect of ignition retard on NOv emissions and fuel
consumption for diesel engines (Reference 5-1).
5-7
-------
1.10
1.00
G OF
D 01 2 Stroke-turbochargdd
A 4 Stroke-naturalIv
_ aspirated
V \r 4 Stroke-turbocharged
#20
2 i
\ 6
8
"Retard
10
12
14
16
#28
Engine
44
Baseline NO
g/kHh
28
21
16
44
46
20
13
16.8
18.2
10.4
38.9
22.1
23.5
11.8
16
Figure 5-3. Effect of ignition retard on NOX emissions and fuel consumption
for gas and dual fuel engines (Reference 5-1).
5-8
-------
However, this reduced peak combustion temperature can also increase CO,
HC, and possible smoke emissions. Lowering manifold temperature in some
of the diesels tested increased HC emissions but did not affect CO.
If the engine is equipped with an aftercooler, it may need to be
enlarged and/or require a larger coolant circulation system. An enlarged
coolant system could increase maintenance requirements. There is a limit
to how much the manifold air temperature can be reduced, especially on hot
days. Radiator cooling is limited by the ambient air temperature while
cooling tower effectiveness is restricted by the ambient air dew point.
Cooling tower systems use more water than radiators, which limits their
use in dry areas. In all cases, a refrigeration system could be installed
but might result in a large energy penalty.
Typical lower temperature limits for a hot, humid location with an
ambient air temperature of 310 K (100°F) are 320 K (115°F) for an air
cooled radiator, 310 K (100°F) for a cooling tower, and lower if a
refrigeration system is used. A typical gas exit design temperature from
an aftercooler is 330 K (130°F) (Reference 5-1).
The data presented in Figure 5-4 show that NO can be reduced 10
A
to 40 percent when manifold temperature is lowered from 330 K (130°F) to
310 K (100°F). This technique appears to be more effective on natural
gas engines. Also note that the effect on fuel consumption is slight when
compared to the other techniques.
Turbocharging an engine without air cooling sometimes reduces NO
A
emissions. As the inlet air temperature rises, the peak cylinder
temperature will be correspondingly higher; hence NO formation will
A
generally increase. However, depending on the location of the air-to-fuel
ratio of the nonturbocharged unit relative to the NO peak (as shown in
A
Figure 5-1), the increase in power from turbocharging has the net result
of a brake specific NO emissions reduction.
A
Adding a turbocharger requires installing the turbocompressor
discussed earlier and, depending on the strength of the originally
installed parts, may also require the replacement of a number of other
engine components (piston rings, connecting rods, wrist pins and cylinder
heads) with higher strength parts. This may be necessary because
turbocharged engines operate with higher cylinder temperatures and
pressures and thus experience higher thermal and structural loads.
5-9
-------
1.02
§ 1.01
0.99
0.98
G OF D
DO 2 Stroke-turbocharged
^ V V 4 Stroke-turbocharged
G -- Gas
OF Dual fuel
D Diesel
0.500.55 0.600.650.700.750.800.850.900.951TOO
VTu
1.00
0.90
0.80
1 o*
x 0.60
0.50
0.40
N0y Baseline
#13HT Engine
15
55
60
16
T -- uncontrolled temperature, °F ^0
T -- controlled temperature, °F
i i i i i i i
g/kWh
15.1
9.8
11.1
10.4
23.5
11.8
%
1 1 1 1
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00
VTu
Figure 5-4. Effect of manifold air temperature reduction on NOX
emissions and fuel consumption (Reference 5-1).
5-10
-------
An additional operational problem may occur if the turbocharged
engine must respond to rapidly varying loads and speeds. The problem is
increased smoke generation due to rich combustion during acceleration,
since the fuel flowrate increases much more rapidly than the air
flowrate. However, most large engines are not subject to rapidly varying
loads and speeds so this problem generally does not arise and the problem
can be designed away.
5.1.4 Exhaust Gas Recirculation
Exhaust gas recirculation (EGR) reduces peak combustion
temperatures by increasing the mass of gas available to absorb heat. As
compared to increasing the air-to-fuel ratio, this increased mass is
achieved without raising the amount of excess oxygen. EGR can be
accomplished by either restricting the exit of gases from the cylinder
(internal EGR) or by reintroducing the exhaust gases into the intake
manifold (external EGR). When external EGR is used the recirculated gases
can be cooled to further reduce peak temperature.
Since CO and HC oxidation depend upon the availability of excess
air and elevated temperatures, it might be expected that reducing both
oxygen and temperature by EGR would increase emissions of these two
pollutants. However, EGR traps or recirculates some of the unburned
hydrocarbons in the exhaust gas and thus HC emissions frequently decrease
when using EGR. Smoke levels increase with EGR, though, due to the
reduction in excess air.
The primary durability consideration for external EGR systems,
especially when applied to diesel engines, is the accumulation of solid
exhaust products in the recirculating system. When EGR is applied to
naturally aspirated engines, these deposits build up in the ducts, on any
valves used to control the recirculation rate, and possibly on the intake
valves. When external EGR is used in conjunction with a turbocharged and
interceded engine, further problems arise. Since the inlet charge, after
the compressor, is at a higher pressure than the exhaust, a separate
compressor and intercooler for the recirculating stream have to be
provided or the recirculated gases have to be mixed with the incoming air
before they pass through the turbocharger. Both approaches have similar
problems, namely fouling of the compressor blades and the heat exchanger
surface. If the compressor is designed to operate close to its optimum
5-11
-------
condition, its performance is very sensitive to the shape of the blades,
which would be slightly changed by deposit build-up. Similarly, the
effectiveness of heat exchangers is greatly reduced by any coating on
exchange surfaces. In fact, moderate deposits can make the heat exchanger
virtually useless. Such deposition problems do not necessarily preclude
the use of E6R on these types of engines, but they would require
significantly increased maintenance by the user.
Figure 5-5 shows the effects of internal EGR on a naturally
aspirated gas engine, a blower scavenged gas engine and a turbocharged
diesel engine. For these three engines, internal EGR reduced NO
/\
emissions from 4 to 37 percent.
External EGR results from three tests on gas, dual fuel, and diesel
turbocharged models are also shown on Figure 5-5 with reductions varying
from 25 to 34 percent. These reductions were obtained with exhaust gas
recirculation rates of 6.5 to 12 percent. The effect of varying EGR on
NO emissions and fuel consumption is shown in Figure 5-6. At 6 percent
A
EGR, NO reductions ranged from 10 to 22 percent. In general, fuel
/\
consumption remained unchanged for EGR rates less than 12 percent.
5.1.5 Water Injection
Water can be added to the fuel-air charge to lower the peak
combustion temperature via both the heat used in vaporizing the water and
the sensible heat absorbed by the water vapor. It is similar in principle
to EGR except that water absorbs heat by vaporization, also. H20
injection reduces NO emissions but increases HC emissions because of
/\
the lower peak temperature. In the engines tested, CO appeared to be
unaffected.
The large bore engine maufacturers who tested water injection
reported serious concerns about the adverse effects on engine durability.
Their concern is based on observations of water in the crankcase
(contaminated lubricating oil) and rapid buildup of mineral scale around
the valves, water injection nozzles, and other components such as
turbochargers (Reference 5-5). Because of these adverse effects, water
injection will probably never be considered a viable technique.
Demineralized water should decrease the mineral buildup problems but water
in the crankcase would still be a problem.
5-12
-------
in
i
EGR
£ 11000
^ 1000
9000
5
T 8000
Brake specific S ,m
fuel conjunction ^ loaa
Z 6000
22
-28
20
-24 18
16
-20
14
- 16 NOM level f 12
£
» 10
- 12
8
-8 6
4
- 4
2
K>x reduction (I)
External/internal
EGR rate (I)
Uncontrolled fuel consumption
(kJ/kHh)
Percent increase
A1r charging
Strokt/cycle
Engine number
Fuel
Gas
" t
A
-
-
4
-
-
: i
.
"
36.9 32.6 5.4
I E I
6.5
10,091 9691 11.511
7.7 0.03 1.1
BS TC NA
2 2 4
70 21
Dual
A
i
26. 0
E
10.7
9967
0.8
TC
2
13
Diesel
/\ *
i .
33. ( 4.1
E I
12.0
9456 9439
0.5 2.2
TC TC
2 4
12 27
Figure 5-5.
Effect of exhaust gas recirculation on NOX emissions
and fuel consumption (Reference 5-1).
-------
en
i
in
c
o
T3
0)
SO.9
co O
i
5
X
o
xO.8
^bO.7
£0.6
°0.5
NOTE:
All data from same
manufacturer
#13
l#12
D
G
DF
0
102 Stroke-turbocharged
Gas
Dual Fuel
Diesel
#13
6 8
Percent EGR
10
12
Figure 5-6. Effect of varying exhaust gas recirculation (EGR) on NOX
and fuel consumption for gas.
-------
5.1.6 Derating
An engine can be derated by restricting its operation to lower
levels of power production than normal for the given application.
Derating reduces cylinder pressures and temperatures and thus lowers NO
A
formation rates. Although NOV exhaust concentrations (i.e., moles of
A
NOX per mole of exhaust) are reduced, it is quite possible for this
reduction to be no greater than the power decrease. In such a case, brake
specific emissions (i.e., grams of N0v per kWh) are not reduced. This
A
is especially true for four-stroke turbocharged engines. In addition,
air-to-fuel ratios change less with derating for turbocharged engines than
for naturally aspirated or blower scavenged units. Thus NO emissions
A
are less responsive to derating for turbocharged engines. Derating also
reduces the engine's operating temperature, which can result in higher CO
and HC emissions.
Demonstrated NO emission reduction levels due to derating are
/\
shown in Figure 5-7 for a number of different engine types and fuels.
These data show emission reductions ranging from 1.6 to 30.8 g/kWh for
naturally aspirated or blower scavenged engines and from 0.3 to 14.0 g/kWh
for turbocharged units. Since these results were obtained with varying
amounts of derating, it is more informative to compare the effectiveness
of this emission control technique on a normalized basis i.e., percent
NO reduction per percent derate. On this basis, results for naturally
A
aspirated or blower scavenged engines varied from 0.25 to 6.2, whereas
those for turbocharged units varied from 0.01 to 2.6. No relationship was
found between normalized effectiveness and uncontrolled emission level,
number of strokes per cycle, or fuel. Fuel consumption increased by 0 to
15 percent. Some diesel engines showed increased NO emissions as load
was decreased.
One big disadvantage of derating is that spare engine capacity may
be needed which could require a large capital investment. For new
engines, derate can be applied by designing the engine to operate under
derated conditions. This could mean a larger, more expensive engine to do
the same job.
5.1.7 Combining Control Techniques
Experiments on combining more than one emission control technique
have been conducted by several engine manufacturers. Generally, the NO
A
5-15
-------
in
t
en
11000
10000
woo
24
20
16
12
8
4
9000
Brake
specific f g^
fuel £
consumption 3 75^
6000
22
20
18
16
14
. NOX level | ,2
I ,0
8
4 6
4
-
2
Percent derate
Normalized reduction 8
Uncontrolled fuel Btu
consumption hp-hr
Percent Increase
Air charging
Fuel
Gas
A
h
.. 4
t '
r
' 1
* "" * T 1 *
.
.
~
1
1
t
A
+ '
>
K
k
A
1
i
«
I
12 20 25 13 20 25 25 25 25 T4 25 25 25 25 25 25
6.2 J.O 2.6 1.4 2.1 2.J 0.3 1.3 0.4 0.3 1.1 1.5 2.3 0.6 2.2 0.5
7127 7400 6775 6760 7265 6942 8130 7760 784063747300 7440 7220 7590 7500 6500
2.6
K
2
1
«;1 6.1 0.3 4.3 14.4 6.4 9.6 15.3 0.8 3.0 1.7 2.8 - 3.5 2.0
«S TC TC TC TC IM KA NA TC TC TC TC TC TC TC
71
2 64 65 69 21 22 44 4 20 23* 25 29 45 46
aPercent NO* reduction/percent derate
bSpeed reduction, const, torque
Dual
- *
A
I
i
18 25
0.9 0.8
6400 6570
3.1 6.1
TC TC
4 4
7 61
Diesel
^
* t
V
1
M
* * *
*
* 4
25 5 25 25 38 27 25
0.9 (.2 0.3 0.5 0.2 0.3 0.2
7844 6951 7175 6930 6677 6366 6667
2.8 4.1 - -5.5 4.8 -0.7 --
K K TC M TC TC TC
22244 44
10 11 12 24* 5 15 27
Figure 5-7.
Effect of derate on NOX emissions and fuel
consumption (References 5-1 and 5-20).
-------
reductions are not additive, although diesel engines exhibit more additive
behavior than other engine types. Results of a set of tests are shown in
Figure 5-8.
For the large bore diesel engine shown in Figure 5-8, the maximum
NO reduction for a single control (retard) is 2.3 g/kWh. When all the
J\
controls tested (retard, reduced inlet manifold air temperature, increased
air-to-fuel ratio, and water injection) were applied simultaneously, NO
/\
was reduced 4.0 g/kWh. This is shown on the figure as an uninterrupted
downward arrow. For comparison, to the left of this, is a multiple arrow
line that represents the depiction in series of all the separate control
effects, as they were individually measured. The difference between the
length of these two lines is a measure of the relationship between the
additive effects of the controls when applied simultaneously and the sum
of their individual contributions. The figure shows the affects of the
controls tested on this engine were essentially additive.
Figure 5-9 shows that the combination of retard and manifold air
temperature reduction is nearly additive in dual fuel engines. However,
additional use of increased air-to-fuel ratio does not decrease NO at
X
much below the levels obtained by the first two controls as would be
expected from the results with air-to-fuel ratio alone. Furthermore,
adding water injection to these three controls has no effect whatsoever.
If moderate NO reduction were required (e.g., 25 to 30 percent) for
/\
this engine, the air-to-fuel ratio would be changed; however, if a greater
reduction were necessary, the combined controls of retard, manifold air
temperature decrease, and air-to-fuel ratio modification would be used.
Gas engines also do not respond additively to the simultaneous
application of several controls (Figure 5-10). Here, reducing manifold
air temperature decreased emissions from the blower scavenged engine by
36 percent, whereas applying reduced temperature plus retard lowered
emissions only 29 percent. Only the combination of the above two
techniques with increased air-to-fuel ratio could reduce emissions below
the level obtained by reduced manifold air temperature alone.
5.1.8 Combustion Chamber Modifications
Combustion chamber redesign is the control technique with the
greatest potential for reducing NO emissions from large bore engines
/V
with little or no loss in efficiency. It is probably also the technique
5-17
-------
en
i
oo
1
\
*~3
_^
|
\
en
10000 o_
f -= 7000
3 Q.
«/l .C
9000 °3
CD
| 6000
-16 12
11
-14
10
$ T
4> .c
« 1
O.
i"
-12 9
8
7
Control
Engine Number
Stroke-Conf igurat
A y V V V y V
-
LEGEND
Uninterrupted line - N0x reduction achieved ~
by simultaneous appli-
cation of several con-
trol techniques.
_ l A i L^ne w'tn intermediate arrows - Additive
1 combination of lines
^ indicating reductions
when each control ap-
V oc QC ce plied individually.
R - Retard
H - Decreased inlet manifold air temperature
\f \f \f \t A- Increased air-to-fuel ratio
a: z z H20 - Hater injection
- y
-------
on
i
.C
^
3
-C
3
^
C71
10000 o_
f-= 7000
1 0.
LO .C
£2 -^
9000 3 =
CD
j> 6000
-16 ^2
11
~
- 10
> i-
-------
en
ro
o
3
3
.c
^1
en
_ «-* ^
t^ 7000
5 o- '
VI -C
c - -
o s
V
2. 6000
-12 9
8
1 A
1Q
__ 7
"a! i-
> c-
Q.
SI
O O)
8 z " 6
5
- 6
4
Control
Engine Number
A A A A A A A
"1
>
^v
/
>
I ;
a:
r
>
f
\
f
LEGEND
\
y-
f
\
*
»
f
t
\
DC
\
x:
\
f «
t
Mr ^
O
CM
R M A H20 RM RMA
Stroke-Configuration
6
4-TC
f
t
\
Uninterrupted line - NOX reduction
achieved by simul-
taneous application
of several control
techniques. ~"
Line with intermediate arrows
Additive combination
of 1 ines indicating
reductions when each--
control applied In-
dividually.
R - Retard
M - Decreased inlet manifold air ~~
temperature
A - Increased air-to-fuel ratio
' H20 - Water Injection
f
f
RMA
H?0
Figure 5-10. Additive effects of controls for large bore
gas engines (Reference 5-1).
-------
requiring the greatest amount of research and development, and could
result in engine designs changing significantly. Thus, several years will
be required to implement the changes and obtain reliable performance and
durability data.
For some engines, the combustion process can be improved by
redesigning chamber geometries to increase turbulence, which is conducive
to good air-to-fuel mixing and, hence, efficient combustion. This
increase turbulence may ensure that most of the combustion takes place
under lean conditions rather than some combustion in regions where the
air-to-fuel mixture is stoichiometric. Combustion in lean regions usually
produces less NO
X
Staged combustion, in which the fuel is first burned rich in a
small chamber separate from the cylinder, then lean within the cylinder,
is another way of reducing emissions through combustion chamber
modifications. Rich combustion avoids excess oxygen at the time high
temperatures are needed for ignition, then completes combustion at a
temperature high enough for combustion but sufficiently low to limit NO
A
formation.
Arthur D. Little, Inc. (Reference 5-7) is currently under EPA
contract to evaluate combustion chamber modifications and other emission
control concepts for stationary engines. They have identified potential
chamber modifications which either improve mixing, enhance combustion, or
represent some form of staged combustion. For diesel engines, mixing can
be improved by circumferential injection, chamber shape, or a variable
area prechamber. Improved combustion in gas engines can be achieved
through torch ignition, multiple spark plugs, high energy spark, increased
turbulence through swirl or "squish," or diesel fuel injection. Diesel
fuel injection essentially converts the gas engine to a dual fuel engine.
Existing dual fuel engines tend to give less NO than existing gas
/\
engines. The average dual fuel engine produces 30 percent less NO than
/\
the average natural gas engine. Swirl or "squish", besides increasing
turbulence, is also a form of internal exhaust gas recirculation.
Staged combustion techniques include divided chambers, open
chambers, or degraded mixing for gas engines, and a prechamber or pilot
injection for diesel engines. Each technique is described below (from
Reference 5-7), first for natural gas engines, then diesel engines.
5-21
-------
5.1.8.1 Torch Ignition (Gas)
Torch ignition ignites a very lean mixture within the cylinder from
a small (2 to 5 percent of the total volume) antechamber. Ignition takes
placed throughout the torch boundary, rather than at a point source as
with a spark (see Figure 5-11). Leaner mixtures can be used, the burn
time is shortened, and turbulence is increased, all resulting in efficient
combustion at reduced temperatures. This may reduce NO emissions 30 to
A
40 percent. Potential disadvantages are that hydrocarbon emissions may
increase as a result of quenching, and fuel usage is projected to increase
no more than 3 percent.
5.1.8.2 Multiple Spark Plugs (Gas)
This concept achieves shorter combustion times and allows leaner
combustion by igniting the mixture at several points within the cylinder.
Although less costly than torch ignition, a high energy spark should be
used to allow reliable ignition. Emission reductions should be similar to
torch ignition, with a slightly higher (4 percent) fuel penalty.
Spark plug
Blind cavity
<^E
Fuel injector
(Specifically for gas fuel)
Main inlet valve
Main fuel injector
Main chamber
Figure 5-11. Fuel injection torch ignition concept (Reference 5-21)
5-22
-------
5.1.8.3 High Energy Spark (Gas)
This concept is similar to torch ignition, except a plasma jet acts
as the torch. The high energy spark is needed to ensure that a leaner
mixture can be ignited. The concept appears to be the lowest cost method
of achieving leaner combustion, but may have the greatest maintenance
requirements. Emission reduction performance should be similar to torch
ignition.
5.1.8.4 Increased Turbulence (Gas)
Increased turbulence also decreases the combustion duration, and
allows combustion of leaner mixtures. If sufficiently high, no fuel
penalty should be experienced. Turbulence can be increased by modifying
the intake passages and by placing shrouds or fins around valves to swirl
the mixture as it enters the cylinder. In addition, placing a cavity on
top of the piston produces "squish" (a circumferential swirl). The
disadvantages of increasing turbulence are the quench layer may be
increased since more surface/volume will be added with a piston cavity;
and increased heat transfer to the cylinder walls, requiring a larger heat
rejection system. A 20 percent reduction in NO emissions is estimated.
J\
5.1.8.5 Staged Combustion (Gas)
Staged combustion concepts all divide the mixture into a rich zone
and lean zone, both of which avoid air-to-fuel ratios that produce high
NO formation rates. The fuel rich zone is ignited first (it is much
X
easier to ignite). NO formation is reduced both because of lower
/\
temperatures and lack of oxygen. Combustion continues as the rich zone
ignites the lean zone, and NO formation is again reduced because of low
A
flame temperatures.
There are two basic approaches to staged combustion: divided
chamber, in which a smaller (20 percent to 30 percent of total volume),
separate chamber is used for the rich mixture; and open chamber, which
maintains a rich zone within the cylinders. Some type of fuel injection
is required to ensure that there are rich zones. Emissions reductions for
a large bore, divided chamber engine are estimated at 20 percent. For the
open chamber engine, a 10 percent reduction in NOX emissions is
estimated. However, reductions up to 30 percent are expected for engines
operating close to stoichiometric.
5-23
-------
5.1.8.6 Prechamber (Diesel)
A prechamber allows staged combustion for diesel engines. Fuel is
injected into a small chamber where it starts burning under rich
conditions (5 percent to 40 percent of the total volume) and then expands
into the main chamber where combustion is completed. (This differs from
gas engines which have combustible mixtures in both chambers at the time
of ignition.) NO emissions have been reduced up to 30 percent on small
/\
bore diesel engines. Large bore engines should achieve N0x emission
reductions of 20 percent with a 1 percent fuel savings based on computer
models. However, potential problems include cavitation in the cylinder
liner, shorter piston life, and some soot buildup.
5.1.8.7 Pilot Injection (Diesel)
Pilot injection places a small amount of the total fuel charge,
5 percent to 15 percent, into the cylinder before the main injection and
ignition take place (see Figure 5-12). A premixed combustion mixture is
formed from this pilot injection, and once ignition takes place ignition
delay is shorter. This effectively allows ignition retardation and
thereby lowers combustion temperatures, reducing the rate of NO
/\
formation. Timing of the pilot injection is critical; premature injection
may produce HC emissions, while delayed injection may increase smoke
emissions. Knocking is another potential problem since the pressure rise
within the cylinder is greater.
NO emissions have been reduced 15 percent on one test, and if
A
ignition is additionally retarded the reduction may reach 30 percent.
Fuel economy should also slightly improve.
5.1.8.8 Combustion Chamber Shape (Diesel)
Improved mixing of the fuel and air increases turbulence and allows
for greater ignition retard. Like gas engines, swirl, "squish", and
manifold and cylinder shapes are used to increase turbulence. NO
X
emission reductions of 20 percent are expected.
5.1.8.9 Circumferential Injection (Diesel)
This concept is similar to using multiple spark plugs for gas
engines, in that ignition takes place simultaneously at several points
within the cylinder. Fuel is injected tangentially which improves mixing,
and the multiple point ignition shortens combustion time. Thus retard may
again be used without a loss of fuel efficiency. The major disadvantage
5-24
-------
Main Pump
Pilot
(a) Single injector
Cylinder
(b) Dual injector
Vigom injection
Pilot injection
4
Mai
40 10
n injection
350 325 300
Top dead
(c) Fuel schedules for two types of preinjection center
Figure 5-12. Alternative pilot injection schemes (Reference 5-7)
5-25
-------
of this concept is the mechanical complexity of the ignition system.
NO reductions are estimated to be 20 percent.
x
5.1.8.10 Summary of Combustion Chamber Modifications
Modifications which increase turbulence may reduce NO emissions
A
20 percent to 40 percent, with less than a 5 percent consumption penalty
if all predictions are correct. Staged combustion techniques are
predicted to reduce NO emissions 10 percent to 30 percent, with little
A
or no fuel penalty.
5.1.9 Catalytic Reduction
Catalytic reduction is a flue gas treatment method of reducing
NO emissions. The catalyst is enclosed in a chamber within the engine
A
exhaust duct and the exhaust gases are passed over the catalyst. Under
controlled conditions of air-to-fuel ratio and temperature, reduction of
NO to N,, and 09 takes place. The concept has been tested by the
X L. C.
automobile industry for some time; one example is a rhodium catalyst
selectively reducing NO . However, this concept requires that
A
essentially no oxygen exist in the exhaust stream. Most large stationary
engines, except for carbureted engines which operate at air-to-fuel ratios
on the rich side of the NO peak, operate at lean air-to-fuel ratios
A
with large amounts of excess oxygen. Therefore, such catalysts cannot be
straightforwardly applied to these large engines.
One approach to overcome this limitation is to inject ammonia,
hydrogen, CO, or natural gas into the exhaust to create the required
reducing atmosphere. Ammonia appears to work best because it reacts
preferentially with the NO rather than the excess oxygen. Thus only
slightly more ammonia would be required for stoichiometric reaction with
the NO, whereas the other three substances need additional quantities to
deplete the excess oxygen in the exhaust.
The overall ammonia reaction is as follows:
2 NO + 2 NH3 + 1/2 02 ^2 N£ + 3 \\fi
This reaction proceeds with a sufficiently high rate at temperatures above
approximately 1090 K (1500°F), but since exhaust gas temperatures are
below this value, a catalyst is required. With the catalyst the reaction
5-26
-------
can be promoted in the temperature range 590 to 700 K (600° to
800°F). The actual temperature window is catalyst dependent.
Published emissions data from such catalyst systems are not yet
available. However Engelhard, the only current manufacturer of a catalyst
for this system, has reported NO emission reductions of 70 percent to
/>
80 percent at an industrial boiler installation in Japan (Reference 5-8).
This unit had flue gas oxygen levels similar to exhaust gas levels from
stationary engines, but operated at higher pressures (500 to 600 kPa).
The reaction is usually complete with the reduction in NO directly
A
dependent on the amount of ammonia injected.
Several potential problems must still be solved before such systems
are ready for commercialization. First, the catalyst is predicted to have
a limited lifetime (1 to 3 years), leading to high operation and
maintenance costs. A reliable ammonia storage and injection system must
still be developed. Materials and design of the exhaust system must also
be improved, because the reaction of NH3 and 02 is exothermic, and
heat may need to be removed from the system. Finally, toxic products such
as hydrogen cyanide and raw ammonia can result, and the allowable range of
parameters such as temperature, NHL/NO ratio, and catalyst surface
O A
area must be fully understood (Reference 5-6). Several engine users are
starting to test these catalytic systems on engines out in the field
(Reference 5-22). Both lean burning and rich burning engines are to be
tested, i.e., reduction catalyst with and without ammonia injection.
5.2 COST OF POLLUTION CONTROLS
Since this report concentrates on the large size stationary
engines, this cost section focuses on those size categories. Medium and
small size engines (which are essentially modified mobile engines) would
use mobile control techniques, and these costs may be found in reports on
engines of that size (e.g., References 5-18 and 5-19). Although it might
be assumed that mobile engine control costs would also apply to stationary
engines, this might not always be the case.
One special problem that might arise when applying mobile engine
control costs to stationary engines concerns certification requirements.
In many cases, stationary engine manufacturers satisfy user engine size
needs required by varying the number of cylinders. If each combination of
cylinders is considered a separate type of engine that must be certified,
5-27
-------
large certification costs may be incurred. This could greatly impact the
pollution control device cost (Reference 5-10).
5.2.1 Cost Evaluation Procedures
The two major cost components are capital costs and the cost of
"operating" the control. Capital cost is the cost to purchase and install
the pollution control device. In addition, capital cost could include
certification charges and factors for recovering research and development
cost for the control technique. This report does not include R&D,
recovery, and certification costs in the estimated capital costs.
However, the installation portion of capital costs include indirect
charges such as engineering and overhead in addition to charges for the
labor to actually install the control device.
The annualized incremental cost for pollution control devices
consists of annualized initial costs, charges resulting from maintenance
and operating cost changes, and alterations in fuel and lubrication use.
Annualized costs are calculated by the following formula:
AC = AIC + M + F
where
AC = annualized costs due to control device
AIC = annualized initial costs (capital cost)
M = maintenance and operating costs
F = incremental fuel and lubrication charges
In units of $/kWh, the terms are calculated by the following:
AIC = (Initial cost)(capita1 cost factor)
~ (Average power output)(hours/year operation)
M - will be given in terms of $/kWh (operating refers to operator
time)
F = (percent change in fuel consumption)(fuel consumption,
kJ/kWh) ' (fuel cost, $/kJ) + Changes in lubrication use
costs.
A capital cost factor of 20 percent will be used as is common in
industrial practice. Fuel costs used are $2.00/GJ (2.11 x 106 Btu) for
natural gas and $3.10/GJ (3.27 x 106 Btu) for diesel fuel (Reference 5-11).
These costs are based on the lower heating value of the fuel. To show how
5-28
-------
controls affect engine costs, the capital and annualized costs are given
as a percent of total engine costs.
Baseline costs are presented for four typical 1C engines in
Table 5-2. The annualized costs in the table do not include any charges
for operator attendance. Since these engines are designed to essentially
operate without any operator attention, this is not a large factor. Also,
costs are for the engines alone and do not include charges for devices
such as generators. The four engines described in Table 5-2 are: a large
bore diesel engine used to generate electricity; a large bore dual fuel
engine used to generate electricity; a large bore natural gas engine used
as a pipeline compressor; and a natural gas engine used at an oil and gas
production site. The average size of an engine used at the oil and gas
production site is about 750 kW, while the other engines are of average
size about 3000 kW. All the engines are assumed to operate for
8000 hours/year which is typical of baseload operation. The maintenance
costs, initial costs, and lubrication costs are based on estimates from
engine manufacturers as cited in Reference 5-1. All cost data are in 1978
dollars. Cost data that were not in 1978 dollars were updated using
Nelson cost indices (Reference 5-12).
As already discussed, the three main types of combustion
modification control techniques are: operational adjustments, exhaust
cleaning, and combustion chamber modifications. Since operational
adjustments have been studied in the greatest detail, the most cost
information exists for these control techniques. Cost data also exist for
the other control techniques, but since most of these techniques are in
the developmental stage, only rough cost estimates can be given.
5.2.2 Estimated Costs of Operational Adjustments
Operational adjustment techniques for NO control include retard,
A
derate, air-to-fuel ratio changes, manifold air temperature reduction and
external exhaust gas recirculation. The capital and maintenance cost of
these techniques were compiled from confidential communications with
engine manufacturers (Reference 5-1). The changes in efficiency or fuel
use were discussed in the preceding sections.
5.2.2.1 Retard
Retard involves only changing the ignition or injection timing,
thus, in principle, no capital equipment changes are necessary. However,
5-29
-------
TABLE 5-2. TYPICAL COSTS FOR UNCONTROLLED ENGINES (1978)
Parameters
Initial Cost, $/kWa
Capital Cost Factor
Operating Hours hr/yr
Maintenance, $/kWa
Fuel cost, $/GJb
Fuel Consumption, kJ/kWh
Lubrication, % Fuel Costa
Annualized Costs ($/kWh)
Capital
Maintenance
Fuel & Lubrication
Total
Typical Engine
3000 kW
Diesel
(Electrical
Generation)
240
0.2
8,000
0.005
3.10
9,900
5
0.006
0.005
0.032
0.043
3000 kW
Dual Fuel
(Electrical
Generation)
240
0.2
8,000
0.005
2.00
9,200
10
0.006
0.005
0.020
0.031
Natural Gas
3000 kW
(Gas
Transport)
240
0.2
8,000
0.005
2.00
9,900
10
0.006
0.005
0.022
0.033
750 kW
(Gas
Production)
80
0.2
8,000
0.005
2.00
11,300
10
0.002
0.005
0.025
0.032
Reference 5-1
bReference 5-11
because retard can raise exhaust temperatures, engine manufacturers may
want to replace existing exhaust valves or modify the exhaust system. If
this is done, there would be a capital cost. The cost estimates assume
that the valves are not replaced but maintenance increases by 33 percent.
Two-cycle diesel engines do not always have valves, but large four-cycle
engines do; thus maintenance costs for diesel engines are assumed to
increase 33 percent, also. For all engine types ignition retard reduces
5-30
-------
engine efficiency zero to 4 percent for a 20 to 30 percent reduction in
NO thus a 4 percent efficiency reduction will be assumed as a worst
/\
case. Table 5-3 shows that the estimated incremental annualized cost
would increase by about 6 to 7 percent. Recall that this is based on the
cost of the basic engine and thus the percent increase in operating a
generator or whatever would be less.
5.2.2.2 Air-to-Fuel Ratio Change
The incremental cost of air-to-fuel ratio changes depends highly on
whether a turbocharger is added. Since most large engines are equipped
with turbochargers, the initial cost of the engines would not change.
Adding a turbocharger to a medium size diesel engine would be ~5 percent
of the initial engine cost (assuming that there is space to add the
turbocharger), but this would be somewhat offset by the increased
efficiency (Reference 5-10). Since the turbocharger handles more air as
the air-to-fuel ratio increases, the maintenance charge for cleaning the
turbocharger could rise by ~0.0002 $/kWh. For a large bore diesel engine,
an air-to-fuel change to reduce NO by ~20 percent would increase fuel
A
consumption by about 10 percent. A dual fuel or natural gas engine
air-to-fuel change to reduce NO 40 percent should only increase fuel
X
consumption by ~2 percent.
Table 5-4 shows that the annualized cost of diesel engines would
increase by about 7 percent, with 3 percent increases for dual fuel or
natural gas engines. Remember that these costs are only for the basic
engine; the cost percentage increase for the total engine/generator
installation would be less.
5.2.2.3 Derate
Derate is a viable control technique for new units only if extra
power is available. However, derate can be used for new installations
by buying an engine larger than normally required. The initial cost of
this control technique is the cost of the additional capacity purchased.
Maintenance costs are almost directly proportional to the number of
cylinders to maintain; thus if the derated engine has more cylinders than
an engine that would normally be purchased, maintenance costs would
increase (Reference 5-13). Also, since derate usually involves operating
the engine at a lower rating than would provide maximum efficiency, brake
specific fuel consumption would increase and raise fuel operating cost.
5-31
-------
TABLE 5-3. ESTIMATED INCREMENTAL COST DUE TO RETARD (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWh)
Total Annual i zed ($/kWh)a
Percent increase in initial cost
Percent increase in total annual
past cost
Percent reduction in NOY
/\
Engine/ Fuel Type
Diesel
(Electrical
Generation)
0
0.0016
0.0012
0.003
0
7.0
20-30
Dual Fuel
(Electrical
Generation)
0
0.0016
0.0007
0.002
0
6.4
20-30
Natural Gas
(Oil and Gas
Pipeline)
0
0.0016
0.0008
0.002
0
6.1
20-30
(Oil and Gas
Production)
0
0.0016
0.0009
0.002
0
6.2
20-30
en
i
co
ro
aAssumes 8000 hours operating year.
-------
TABLE 5-4. ESTIMATED INCREMENTAL COST OF AIR-TO-FUEL INCREASE (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWh)
Total Annual i zed ($/kWh)a
Percent increase in initial cost
Percent increase in annual cost
Percent reduction in NOX
Engine/ Fuel Type
Diesel
(Electrical
Generation)
0
0.0002
0.0031
0.003
0
7.0
20
Dual Fuel
(Electrical
Generation)
0
0.0002
0.0004
0.001
0
3.0
40
Natural Gas
(Oil and Gas
Pipeline)
0
0.0002
0.0004
0.001
0
3,0
40
(Oil and Gas
Production)
0
0.0002
0.0004
0.001
0
3.0
40
en
CO
CO
^Assumes 8000 hours operating year.
-------
Specific figures are not presented here because of the difficulty in
specifying costs that are very site dependent.
5.2.2.4 Manifold Air Temperature Reduction
If the engine has an aftercooler or an intercooler, reducing
manifold air temperature could require a larger heat exchanger, more
coolant circulation, or perhaps a temperature control system. These
modifications would cost about 1.5 percent of the initial cost of the
engine (Reference 5-1). Maintenance costs would also rise because of the
additional cooling water which requires chemical treatment to prevent
sludge and scale buildup, or the increased service of radiators to
maintain the lower temperatures. However, these increased maintenance
costs are expected to be small: only on the order of $0.0008/kWh for a
diesel or dual fuel engine and $0.0002/kWh for a gas engine. There would
also be a small charge for pumping the additional cooling air or water.
As already discussed, manifold air cooling is expected to negligibly
affect engine efficiency. To reduce NO by 20 percent, the largest
/\
increase in brake specific fuel consumption for the tests made was only 1
percent. In some cases, brake specific fuel consumption even decreased
since manifold air cooling allows more air to be charged, thus increasing
efficiency. The largest increase in cost of power from these engines due
to increased manifold air cooling is expected to be about 1 percent.
5.2.2.5 External Exhaust Gas Recirculation
Installing an external exhaust gas recirculation system could
increase the initial engine cost 5 percent. That price includes adding
the required piping to recirculate the exhaust gases, a heat exchanger for
cooling the recirculated exhaust gases, and a control system. Maintenance
costs are expected to rise about 40 percent for a diesel engine and
80 percent for dual fuel or natural gas engines. This cost escalation
includes the additional system to maintain recirculating exhaust gases
which can foul various parts of the system (References 5-6, 5-14, 5-15).
The small number of tests has shown almost no change in fuel consumption
due to EGR (see preceding sections). Table 5-5 shows that E6R could
increase the cost of operating a diesel engine by over 5 percent and the
other types by over 13 percent.
5-34
-------
5.2.2.6 Combination of Operational Adjustment Controls
By combining control techniques it may be possible to achieve the
same NO reductions with a smaller fuel penalty, or reduce NOV levels
A X
more than could be achieved by each technique alone. Except for brake
specific fuel consumption (BSFC), the initial and operating costs can be
assumed to be additive. The cost advantage of combined controls over
single techniques will depend heavily upon the actual change in BSFC.
There are other reasons, such as operational impacts and control of other
emissions, that may favor combined control even if there are no cost
advantages. Table 5-6 compares three different methods for reducing NO
A
by 40 percent from a large bore diesel engine. Air-to-fuel changes
combined with manifold air cooling will only give 40 percent reduction in
special cases. For this case, there is a definite advantage to using
combined controls since the two combined techniques, air-to-fuel ratio
change and manifold air cooling or air-to-fuel ratio change and retard ,
had a lower BSFC than retard alone in the tests.
5.2.3 Exhaust Gas Treatment Techniques
Catalysts for oxidizing CO and HC are available, but when used with
diesel engines must be regenerated periodically because of particle matter
in the exhaust. The initial cost of the catalyst is about $10/kW
(Reference 5-1) and because of the regeneration problem, maintenance costs
can increase.
Catalysts for reducing NO are available for rich burning engines
/\
and have been used for purifying gases for injection into wells
(Reference 5-16). Engines are sometimes used as a source of inert gases
for injecting into gas wells because the engine can also be used to
operate the compressor. Because NO can cause corrosion problems in the
A
compressor (due to water formed when gas compresses), catalysts have been
used to reduce NO . The engines are usually specially tuned so that
X
very little oxygen is present in the exhaust. Thus these catalysts cannot
be applied to a lean burning engine such as most large stationary engines.
Some manufacturers are developing catalysts that use NH^ as a
reducing agent but these catalysts have not yet been tested in actual
engines (Reference 5-17). Since these catalysts are only in the
developmental stage, costs can only be roughly estimated. The catalyst
and container is estimated to be about 4 percent of the initial cost of
5-35
-------
TABLE 5-5. ESTIMATED INCREMENTAL COST OF EXTERNAL EXHAUST GAS RECIRCULATION (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWh)
Total Annual i zed ($/kWh)a
Percent increase in initial cost
Percent increase in annual cost
Percent reduction in NOV
A
Engine/ Fuel Type
Diesel
(Electrical
Generation)
12
0.002
0
0.0023
5
5.3
20
Dual Fuel
(Electrical
Generation)
12
0.004
0
0.0043
5
14
20
Natural Gas
(Oil and Gas
Pi pel ine)
12
0.004
0
0.0043
5
14
20
(Oil and Gas
Production)
4
0.004
0
0.0041
5
13
20
CO
aAssumes 8000 hours operating year.
-------
TABLE 5-6. ESTIMATED INCREMENTAL COST OF COMBINED CONTROLS FOR
LARGE BORE DIESEL ENGINES AT 40 PERCENT NOX
REDUCTION (1978)
Capital ($/kW)
Maintenance ($/kWh)
Fuel ($/kWH)
Total Annualized ($/kWh)a
Percent increase in capital cost
Percent increase in annual cost
Percent reduction in NOX
Control Technique
Retard
0
0.0016
0.0024
0.0040
0
9.3
40
Air-to-Fuel
Changes and
Manifold Air
Cooling
3.6
0.001
0.0015
0.0026
1.5
6.0
30-40
Air-to-Fuel
and Retard
0
0.0018
0.0015
0.0033
0
7.7
40
aAssumes 8000 in operating year.
5-37
-------
the engine (Reference 5-1), in addition to a charge for the ammonia
handling and injection system. Besides operating charges for maintaining
the NH3/catalyst system, there will be a charge for the ammonia and a
small cost for operating the injection system.
5.2.4 Combustion Chamber Redesign
Combustion chamber redesign techniques are only now being developed
and thus only very rough price estimates can be given. A. D. Little,
under contract to the EPA, is presently planning to test some combustion
chamber modifications and they have presented some cost estimates of these
techniques (Reference 5-7).
The cost estimates of the required hardware range from about $l/kW
to $47/kW with $8/kW being a typical estimate. Of course the price will
depend on exactly which technique is used. The operating cost will be
very dependent on the BSFC change. The goal is to keep the change in BSFC
to less than 4 percent. If this goal is met and the initial cost is about
what is estimated above, combustion chamber redesign could be the most
cost effective method of controlling NO for achieving 30 to 40 percent
/\
reduction.
5.3 SUMMARY OF NOX EMISSION REDUCTIONS AND OPERATION AND MAINTENANCE
IMPACTS
5.3.1 NO Reduction Techniques
/\
Table 5-7 shows the NO reductions and fuel penalties associated
A
with each control for each engine type. Combinations of controls which
may be readily applied are also shown. From this table it is apparent
that, with the exception of catalytic reduction, which is a flue gas
treatment concept, fuel injection retard is the most effective engine
control for diesel engines and E6R looks good if the additional
maintenance is not a problem. Air-to-fuel ratio changes are most
effective for gas engines and EGR and retard are also effective controls.
With combinations of controls, NO emissions reductions of 40 percent
/\
should be achievable with further reductions possible with a
catalyst-NH., system. Some engines may even achieve 60 percent NO
*J X
reduction.
On July 23, 1979 (Federal Register Vol. 44, No. 142, p. 43153), it
was proposed that diesel and dual fuel engines larger than 560 cubic
inches (0.0092 m ) per cylinder be controlled to 600 ppm at 15 percent
5-38
-------
3
CL, and gas engines larger than 350 cubic inches (0.0057 m ) per
cylinder be controlled to 700 ppm. Recall that the average emissions from
these engines is about 17 g/kWh (approximately 930 ppm at 15 percent 0~)
for diesel and slightly less for the others. Thus a reduction of
approximately 35 percent is required. As has just been discussed, there
are several techniques which should be able to achieve this level of
reduction.
5.3.2 Operational and Maintenance Impacts
Since engines are currently optimized for a minimum of maintenance
and fuel usage, any control technique which varies engine parameters from
standard conditions will impact operation and maintenance. Some of these
impacts have been well characterized, especially those from control
techniques which involve engine operational changes. Other control
techniques will require various degrees of evaluation before impacts are
clearly understood.
Derate, air-to-fuel ratio changes, manifold air cooling and
ignition retard present the least problems in operation and maintenance.
Derate has no impact mechanically and can improve durability because of
lower temperatures and pressures. However, additional engines may be
required to replace the lost power. Fuel penalties are usually low.
Air-to-fuel ratio changes in the lean direction cause a power loss
if larger blowers or turbochargers must be used. If the engine must be
operated richer to reduce NO emissions, other emissions such as smoke
X
and deposits can increase. This could cause an increase in engine
maintenance. Changes in air-to-fuel ratio in either direction will
increase fuel consumption. Finally, operation at air-to-fuel ratios where
misfiring or detonation occur can cause severe engine damage.
Manifold air cooling has little operational or maintenance impact
for a unit that is already interceded, but will increase the size of the
heat exchanger, water or air pump, control system, and other system
components. Of course, installing intercooling on an engine originally
without the system will add maintenance to achieve additional temperature
reductions. Changes in fuel consumption are small.
When properly applied, ignition retard has no serious mechanical
drawbacks. Some increase in operational and maintenance time would ensure
the degree of retard is always within safe limits. Increases in fuel
5-39
-------
TABLE 5-7. NOX REDUCTION AND FUEL CONSUMPTION PENALTIES FOR DIESEL,
DUAL FUEL, AND GAS ENGINES
Control Approach
Derate (D)
Retard (R)
3%
6%
10%
20%
25%
2°
4°
8°
Air-to-Fuel (A) 2%
3%
5%
±10%
Manifold (M')311K(100°F)
Air
Temperature
315K(107°F)
318K(113°F)
Internal EGR
External EGR
10%
Retard and Manifold
Air Temperature
Retard & Air-to-Fuel
Retard and Manifold
Air Temperature and
Air-to-Fuel
Air-to-Fuel and
Manifold Air Temperature
Water Injection 50%
H20/fuel ratio 10M
Catalytic Reduction
(Projected)
Combustion
Chamber
Modi fi cations
(Projected)
Increased
Mixing
Staged
Combustion
Engine Fuel Type
Diesel
% NOX
deduction
__
--
<20
5-23
<20
<40
28-45
--
..
7-8
7-15
--
5
--
<20
33
<20
10-24
<20
<40
35-65
20
.-
--
<20
20-30
25-35
50-80
10-30
10-30
ABSFC, %a
--
--
4
1-5
4
4
2-8
3
0-2
..
--
2
1
1
1
0-1
8
16
5-26
0
--
--
2
3
2-4
0
<5
0
Dual Fuel
% NOX
Reduction
--
<20
--
1-33
<20
<40
50-73
--
<20
25-40
18-37
__
<20
--
--
20
--
<20
25
<20
<40
56
<20
40
__
--
-_
50-80
20-40
10-30
ABSFC, %a
--
4
1-7
3
1
3-5
0
--
1-3
0-1
._
1
--
1
--
1
2
1
2
2
2
3
--
^
--
_
0
<5
0-7
Natural Gas
% NOX
Reduction
<20
<40
--
5-90
<20
8-40
<20
--
<40
20-80
28
<20
--
<20
5-35
<20
33
<20
30-40
<20
<40
17-52
<20
<40
40-65
..
25-35
60-75
50-80
20-40
10-30
ABSFC, %a
2
3
--
2-12
--
3
2-7
2
7
5-12
0
0
--
5
0-8
0
0
3
5-6
4
8
4-11
2
4
6-7
--
1-2
2-5
0
<5
0-2
ABSFC is Increase in brake specific fuel consumption.
5-40
-------
consumption are moderate. Excessive amounts of retard, however, can
create severe engine problems. Fuel consumption will increase rapidly,
power drops, misfiring can occur, and smoke levels increase. In addition,
mechanical maintenance will increase if the exhaust temperature exceeds
the safe limits for valves or the turbocharger, usually 920 K (1200°F).
More frequent engine teardown will be required, and higher initial costs
will result for higher temperature materials.
Exhaust gas recirculation and water injection are two effective
control techniques that will require some development before operation and
maintenance problems make the techniques useful. Exhaust gas
recirculation requires new hardware components which mandate new
maintenance techniques. Problems of fouling the flow passages of the
cooling heat exchanger, the engine turbocharger and aftercooler with
particulate must be solved, or frequent engine teardown will be required.
Under varying load conditions a sophisticated control system is required
or the engine may stall or emit unacceptable smoke levels. Fuel
consumption penalties with EGR are small.
Water injection can cause severe maintenance problems. Deposits
from untreated water will build up on internal engine surfaces, and also
foul the lubricating oil. These problems lead to major engine
maintenance. Water injection also adds another system to the engine which
must be maintained and controlled.
Although not demonstrated, combustion chamber modifications will
present the least impact to operation and maintenance. Maintenance can be
expected to increase slightly if additional injectors, spark plugs and
valves are added to the chamber. However, because this control technique
involves new design, many of the additional maintenance requirements can
be designed out. Fuel penalties are expected to be small.
Catalytic reduction will require no additional engine maintenance,
since it is a flue gas treatment technique, rather than an engine
modification. However, operating the catalyst may be expensive. Fouling
the catalytic surfaces with particulate may require frequent
regeneration. The catalyst may also have a relatively short life and need
to be replaced. Another system, ammonia injection, must be added and
maintained for most applications, and the cost of amnonia must be included
in engine operation. Finally, harmful products of the reaction may be
5-41
-------
produced if the catalyst temperature varies from the proper level, or if
excessive ammonia is injected. This catalyst must be installed and
operated on an engine before all these effects can be quantified.
5-42
-------
REFERENCES FOR SECTION 5
5-1. "Stationary Internal Combustion Engines. Background Information:
Proposed Standards," EPA-450/3-78-125a, April 1979.
5-2. Letter from W. E. Snyder, Waukesha, WI to D. Bell, EPA, RTP, N.C.,
May 18, 1978.
5-3. Patterson, D. J., and N. A. Henein, Emissions from Combustion
Engines and their Control, Ann Arbor Science Publishers, Inc.,
Ann Arbor, MI, 1974.
5-4. Roessler, W. U., ^t ^1_._, "Assessment of the Applicability of
Automotive Emission Control Technology to Stationary Engines,"
EPA-650/2-74-051, NTIS-PB 237 115, July 1974.
5-5. Shaw, J. C., "Emission Reduction Study on a Carbureted Natural Gas
Fueled Industrial Engine," Draft ASME Paper, White Superior
Division, White Motor Corporation, Eastlake, OH, November 1974.
5-6. Urban, C. M., and K. J. Springer, "Study of Exhaust Emissions from
Natural Gas Pipeline Compressor Engines," Southwest Research
Institute, PR-15-61, San Antonio, TX, February 1975.
5-7. "Potential Emission-Control Concepts for Large-Bore Stationary
Engines," Draft Report, EPA Contract 68-02-2664, Arthur D. Little,
Inc., Cambridge, MA, November 1978.
5-8. Acurex meeting notes, California Air Resources Board, "Public
Workshop for the Discussion of a Tentative Proposal for Control of
NOX Emissions from Stationary Internal Combustion Engines,"
Sacramento, CA, October 1978.
5-9. Personal communication with D. Brown, Briggs and Stratton,
Wauwatosa, WI, October 1978.
5-10. Personal communication with G. P. Hanley, General Motors,
Warren, MI, December 19, 1978.
5-11. "Monthly Energy Review," DOE/EIA-0035/7, NTIS-PB 127 007, July 1978,
5-12. "Nelson Cost Indices," The Oil and Gas Journal, Vol. 76, No. 40,
pg. 125, October 1978.
5-13. "Diesel Power MAN Supplement," Diesel and Gas Turbine Progress
Worldwide, Vol. 10, No. 9, pg. 1-44, November-December 1978.
5-14. "Characterization and Control of Emissions from Heavy Duty Diesel
and Gasoline Fueled Engines," Report under EPA-IAG-0219(D), Fuels
Combustion Research Group, Bartlesville Energy Research Center,
Bartlesville, OK, December 1972.
5-43
-------
5-15. Bosecker, R. E., and D. F. Webster, "Precombustion Chamber Diesel
Emissions A Progress Report," SAE 710672, August 1971.
5-16. Barstow, W. F., and G. W. Watt, "Fifteen Years of Progress in
Catalytic Treating of Exhaust Gas," Society of Petroleum Engineers
Paper 5347, 1975.
5-17. Personal communication with R. Leohelt, Engelhard Industries,
Newark, NJ, November 17, 1978.
5-18. "Environmental Impact Statement and Economic Impact Analysis,
Revised Heavy-Duty Engine Regulations for 1979 and Later Model
Years," Prepared by Office of Mobile Source Air Pollution Control,
EPA, Washington, DC, August 1977.
5-19. Lindgren, L., "Cost Estimations for Emission Control Related
Components/Systems and Cost Methodology Description,"
EPA-460/3-78-002, NTIS-PB 279 195, March 1978.
5-20. Hunter, S. C., "Application of Combustion Modifications to
Industrial Combustion Equipment," EPA-600/7-79-015a, January 1979.
5-21. Wyczalek, F. A., et aj_^, "EFI Prechamber Torch Ignition of Lean
Mixtures," SAE Paper 750351, February 1975.
5-22. Personal communication with Pat Bradley, Southern California Gas,
Los Angeles, California, February 22, 1980.
5-44
-------
SECTION 6
ENVIRONMENTAL IMPACTS
One very important criterion when evaluating the effectiveness of a
pollutant control technique is its effect on other pollutant emissions. A
NO control technique should limit NO emissions without adversely
A A
affecting the emissions of other pollutants such as smoke. This section
discusses pollutant formation mechanisms, and reviews the available data
on the effect of NO controls on incremental emissions. To help
A
quantify how NO controls affect the environmental impact of a
A
combustion source, a Source Analysis Model, SAM IA, was applied to typical
effluent stream emissions from a large bore diesel engine since a diesel
exhaust probably has the most pollutants.
6.1 POLLUTANT FORMATION MECHANISM
This section discusses how various pollutants may form during the
combustion process. The effect of postcombustion techniques such as
catalytic treatment will not be covered here since any effects are very
catalyst dependent.
6.1.1 NO.. Formation
x
As mentioned previously, the two main sources of NO are: NO
A A
formed from the oxidation of N£ introduced in the combustion air, or
thermal NO ; and NO formed from nitrogen containing compounds in the
X X
fuel or fuel NO . Because of the low nitrogen content of the fuels
A
normally used (see Section 4), most NO emissions are due to thermal
X
NO . The residence time of the N? at high temperatures in contact
X *-
with oxygen strongly influences the amount of NO formed. Internal
X
cylinder conditions such as local air-to-fuel mixture, fuel injection, and
heat loss could all affect NO formation.
A
6.1.2 HC Formation
The main sources of hydrocarbon emissions are unburned or partially
burned fuel components. Other sources such as blowby and evaporative
6-1
-------
losses should not be important in large bore engines. Diesel fuel has a
relatively low vapor pressure, and since natural gas is mainly methane
(which is not considered a reactive hydrocarbon) evaporative losses are
not expected to be significant. Blowby emissions should not be important
because diesel fuel is injected after the compression stroke. Therefore,
any blowby during the compression stroke would contain little fuel. For
natural gas engines blowby would be mainly methane. Also, for these large
bore engines the amount of surface area between the rings and cylinder
walls is relatively small.
Several mechanisms exist which could cause hydrocarbon emissions.
Fuel droplets could be transported or injected into the air layer near the
cylinder walls, which could be at too low a temperature to sustain
combustion. There could also be incomplete mixing or an improper air-to-
fuel ratio. For diesel engines, the fuel droplets could be too large and
thus not have enough time for complete volatilization and combustion. Any
mechanism that will keep the fuel from being in contact with oxygen at a
high enough temperature for a sufficient time can cause hydrocarbon
emissions.
6.1.3 CO Formation
CO emissions are caused by lack of oxygen or the CO cooling off
before it can be converted to C02. Mechanisms that cause hydrocarbon
emissions can also cause CO emissions. Since the CO comes from partial
oxidation of the fuel while HC emissions may be caused by completely
unoxidized fuel, there will not always be a 1 to 1 correspondence between
CO and HC emissions. But as a first approximation, mechanisms that
usually cause hydrocarbon emissions will also cause CO emissions.
6.1.4 Smoke Formation
As is the case for HC and CO emissions, smoke can be caused by the
lack of oxygen or low temperature. Smoke can be a mixture of soot
particles, which are mainly carbon particles, and fuel droplets. Smoke
can be caused by unburned or partially burned fuel from the center of
large fuel droplets and the core of the injector spray. The quenching of
the burning fuel by the low temperature region near the walls can also
cause smoke. Diesels may form polycyclic organic matter (POM) as
mentioned in Section 2, but the formation mechanisms are not understood.
6-2
-------
6.1.5 SO and Trace Elements
A
SO emissions are expected to be directly proportional to the
/\
amount of sulfur in the fuel. Trace elements emissions can come from the
fuel, from particles picked up by the combustion air, or from leaching of
engine surfaces. Trace element emissions are expected to be very small.
6.1.6 Baseline Emissions
Pollutant emission levels are given here in terms of g/kWh power
output instead of ng/J heat input since some NO control techniques can
A
cause large changes in fuel consumption (even up to 10 percent as
discussed in Section 5). Baseline emissions are listed for comparison in
Table 6-1 (Reference 6-1).
TABLE 6-1. AVERAGE BASELINE EMISSIONS FROM STATIONARY INTERNAL
COMBUSTION ENGINES9, g/kWh
Engine Fuel
Large bore engines
Diesel
Natural gas
Dual fuel
Medium and Small Bore Engines
Gasoline
Diesel
NOX
17.3
15.4
11.0
11.9
16.6
CO
2.4
3.8
2.7
13.7
6.0
HC
0.6
6.5
4.2
11.3
2.8
Reference 6-1
Recall that most of the hydrocarbon emissions from natural gas and dual
fuel engines are methane. Also, these emissions are only for an average
engine and emissions from a particular engine can be very different since
emissions depend on such factors as swirl, types of valves and ports, type
of injector or spark plug and their location, type of piston head,
combustion chamber and shape, operation conditions, etc. Also notice that
only NO , CO, and total hydrocarbon emissions are listed since the other
X
emissions and the breakdown of hydrocarbon emissions are not very well
quantified. As in the other sections, only the potential environmental
6-3
-------
impacts of combustion modifications on large bore engines are discussed in
the following.
6.2 INCREMENTAL ENVIRONMENTAL IMPACTS ON AIR
Figures 6-1 through 6-6 summarize the changes in CO and total HC
emissions as a function of controlled NO level for the large bore
A
engines measured. The graphs list the values for the emissions before and
after the controls were applied. The numbers by the dots refer to the
engine numbers as used in Section 5. Also recall that except for diesel
engines, most hydrocarbon emissions are methane as discussed in
Reference 6-2. These data are from publications and reports from engine
manufacturers (Reference 6-3).
6.2.1 Derate
Derate can lower the maximum engine temperature which could
increase the amounts of HC, CO, and smoke produced. For most of the large
bore diesel engines measured, CO only increased slightly, while two of the
engines actually showed large decreases in CO emissions. The natural gas
and dual fuel engines tested measured small to large increases in CO
emissions. The same effects were noticed for total hydrocarbon
emissions. For HC emissions, some engines showed large increases while
some showed large decreases as the engine was derated, with the unexpected
results that the engine showing large changes in HC emissions was not
necessarily the one showing large CO changes. For dual fuel and natural
gas engines tested, the total hydrocarbon emissions showed small to large
changes (some showed increases, some showed decreases) as the engine was
derated. The changes in CO and HC emissions from these engines were more
closely correlated than for diesel, though there was not a direct
correspondence for each engine. Again, for all types of engines the
preceding comments are based on only a limited number of tests. For all
tests conducted, diesel engines showed lower smoke levels as they were
derated, but results were based on three tests only. The fact that some
engines showed increased HC indicates that smoke might also increase.
6.2.2 Retard
Retard shortens the residence time, which would tend to increase HC
and CO but also increases the exhaust temperature, which would tend to
lower HC and CO emissions. As long as the engine is not retarded to the
point where misfiring occurs, tests have usually shown only small
6-4
-------
>
d)
cr>
en
IE #27
Diesel
O 2 Stroke-blower scavenged
D 2 Stroke-turbocharged
& 4 Stroke-naturally aspirated
« 4 Stroke-turbocharged
D Derate
R Retard
A Air-to-fuel ratio
M Manifold air temperature
IE Internal EGR
EE External EGR
CR Compression ratio
R&A
12
16 20
N0x level, q/kWh
24
28
32
Figure 6-1. HC levels versus controlled NO levels for diesel
engines (Reference 6-3). x
-------
01
I
o>
10
I
^
Oi
0)
R&M
Dual Fuel
2 Stroke-turbocharged
Stroke-turbocharged
D Derate
R Retard
A Air-to-fuel ratio
M Manifold air temperature
EE External EGR
I
I
12
16 18
NOX level, g/kWh
24
28
Figure 6-2. HC levels versus controlled NOX levels for dual-
fuel engines (Reference 6-3).
-------
EE&O
RirttA
I
12 16 20
NO, level. g/kWh
Gas
O 2 Stroke-blower scavenged
Q 2 Stroke-turbocntrged
D Derate
R Retard
A Air-to-fuel ratio
M Man Ho Id air temperature
EE External EGR
IE Internal EGR
Gas
£ 4 Stroke-naturally aspirated
4 Stroke-turbocharged
D Derate
Retard
Air-to-fue) ratio
Manifold air temperature
Internal EGR
145
.. 144
12
16 20
NOX level. g/kWh
24
28
Figure 6-3. HC levels versus controlled NO levels for gas engines
(Reference 6-3).
6-7
-------
en
oo
OJ
o
Diesel
O2 Stroke-blower scavenged
D 2 Stroke-turbocharged
&4 Stroke-naturally aspirated
^4 Stroke-turbocharged
D Derate
R Retard
A Air-to-fuel ratio
M Manifold air temperature
IE Internal EGR
EE External E6R
CR Compression ratio
12 16 20
NOX level, g/kWh
Figure 6-4. CO emissions versus controlled NO levels for diesel engines.
-------
10
en
-------
Gas
O 2 Stroke-blower scavenged
O 2 Stroke-turbocharged
D Derate
R Retard
A Air-to-fuel ratio
M Manifold air temperature
EE External EGR
IE Internal EGR
EE+D
77.
#29
16
12
8 12
NOX level. g/kWh
16
20
129
12
16 20
NOX level, g/kWh
24
Gas
4 Stroke-naturally
j aspirated
4 Stroke-turbocharged
D Derate
R Retard
A Air-to-fuel ratio
M Manifold air temperature
IE Internal EGR
28
32
Figure 6-6. CO emissions versus controlled NOX levels for
gas engines (Reference 6-3).
6-10
-------
increases in CO and HC emissions. In some tests, retard even decreased HC
emissions. Up to a certain point (2° to 6° retard) only a small
increase was noticed in smoke levels but if the diesel engine was retarded
further, smoke levels greatly increased (Reference 6-4).
6.2.3 Air-to-Fuel Changes
Most large bore engines operate on the lean side, i.e., greater
than stoichiometric air, and air-to-fuel chan-ges for NO control mean
A
operating under even leaner conditions. As the mixture becomes leaner,
the peak temperature decreases while the amount of excess oxygen
increases. Thus the HC emissions are expected to increase because of this
decreased temperature. However, CO emissions are not expected to increase
because the increased amounts of oxygen should counteract the effect of
lower temperature. Except for one diesel engine which showed a large CO
increase, the engines tended to follow the above, i.e., small HC increases
and very small CO increases. The test which showed large CO increases was
where the air-to-fuel ratio was decreased instead of increased as in the
other cases. NO was still reduced.
X
6.2.4 Reduced Manifold Air Temperature
Reduced manifold air temperature tends to lower the maximum
combustion temperature, which may raise CO and HC emissions. Only small
increases in HC and CO were measured from the engines tested, and in a
small number of cases, there was even a decrease. There are very few
smoke data but no significant change was noticed in the available data,
though smoke probably would increase along with increased HC emissions.
One engine manufacturer claims smoke increased as a result of reduced
manifold air temperature (Reference 6-4).
6.2.5 Exhaust Gas Recirculation
EGR will lower the combustion temperature and the amount of oxygen
present, which could lead to increased CO, HC, and smoke emissions. At
the same time, recirculating the exhaust gas could lower HC since some of
the HC in the exhaust would be put back into the cylinder and get a second
chance to burn. Tests on diesel engines used as locomotive engines showed
increased CO and smoke while HC was unchanged (Reference 6-5). Tests on
truck engines have shown the same trends (Reference 6-6).
6-11
-------
6.2.6 Flue Gas Treatment
The effect on CO and HC caused by a NO reduction catalyst will
A
be very catalyst dependent. Since most of these engines have large
amounts of excess oxygen, it is not expected that CO or HC would be
effected. If NH3 is used as the reducing agent, there is the danger of
HCN or NH3 emissions. In contrast, in one test on an oxidation catalyst
for reducing CO and HC emissions, NO emissions increased (Reference 6-7).
X
6.2.7 Source Assessment Model Results
To help quantify the change in potential environmental impact of an
1C engine which switches from baseline to low NO operation, a source
/\
analysis model, SAM IA (References 6-8 and 6-9), was applied to typical
air emissions data from a large bore diesel engine. EPA has been
developing a series of source analysis models to define methods of
comparing emission data to environmental objectives, termed multimedia
environmental goals (MEG's) (Reference 6-10). The model selected for the
level of data detail obtainable was SAM IA, designed for rapid screening
purposes. As such, it includes no treatment of pollutant transport or
transformation. Goal comparisons employ threshold effluent stream
concentration goals, termed discharge multimedia environmental goals
(DMEG's).
For the purposes of screening pollutant emissions data to identify
species requiring further study, a discharge severity (DS) is defined as
fo11ows:
n<- _ Concentration of Pollutant i in Effluent Stream
i DMEG of Pollutant i
The DMEG value, the threshold effluent concentration, is the
maximum pollutant concentration considered safe for occupational
exposure. When DS exceeds unity, more refined chemical analysis may be
required to quantify specific compounds present.
To compare waste stream potential hazards, a weighted discharge
severity (WDS) is defined as follows:
WDS = (? DS.) x Stream Mass Flow Rate,
where the OS's are summed over all species analyzed. The WDS is an
indicator of output of hazardous pollutants and can be used to rank the
needs for controls for waste streams. It can also be used as a
preliminary measure of how well a pollutant control, say a combustion
6-12
-------
modification NO control, affects the overall environmental hazard of
)\
the source. An extensive exposition of SAM IA and list of DMEG's are
presented in References 6-8 through 6-11 and will not be repeated here.
Using SAM IA, the effect of combustion modification controls upon
the total emissions from a large bore diesel engine was estimated. The
only pollutants for which there are quantified data showing these effects
are CO, NO , and total hydrocarbons. A partial hydrocarbon breakdown
A
has been measured for the baseline (uncontrolled) case but not for the
engine operating under NO control conditions (Reference 6-12). Because
/\
of this, SAM IA was applied assuming that the same kinds and distribution
of organics are emitted under NO control as those under baseline
/\
conditions. Only the total amount of HC emitted is assumed to change.
The results are summarized in Table 6-2. Appendix A presents the assumed
waste stream characterizations and SAM IA worksheet calculations. An
average NO control technique (such as retard, increased manifold air
/\
cooling, etc.) is assumed to be applied to the engine.
If the above assumption about organic emissions is correct, these
NO controls decrease the total toxic level of emissions from the
X
engine. The decrease in NO emissions is not offset by increases in CO
X
and HC. Some authors have measured mutagenic compounds in diesel exhaust
gases from mobile sources (Reference 6-9). Tests on the exhaust from a
150 kW stationary diesel have shown the exhaust to have a small mutagenic
effect but less than that found for mobile engines (Reference 6-14). The
tests on this engine operating under NOV control conditions has not been
X
completed yet. Thus more tests are needed before quantifying the results
to give the true health effects of lowering NOX emissions.
6.2.8 Concluding Remarks on Health Effects
An additional comment should be made about the toxicity of diesel
exhaust. As just mentioned, some authors have found mutagenic compounds
in exhaust from mobile diesel engines. A recent conference discussed this
potential problem (Reference 6-15). Tests where diesel exhaust was
assayed using bacterial mutagenecis tests elicited mutagenic responses.
But studies on workers in a London bus garage and on workers in mines
where diesels are used have showed no increases in the cancer rate as
compared to the general public. Still, benzo(a)pyrene, a potent
carcinogen, is one of the compounds found in the exhaust from diesel
6-13
-------
TABLE 6-2. SAM IA RESULTS FOR DIESEL EXHAUST
Pollutant
CO
NOX
HCa
Total
Discharge Severity *JJ*jJ "^^
Uncontrolled
3.5
86
22
112
Controlled
7
54
35
96
Uncontrolled
13650
333450
680000
4.3 x 106
Controlled
27300
212160
125000
3.7 x 106
Composition of organic emissions is approximated from limited
emission data and is assumed not to change under controlled
conditions, the total amount of organic emissions increases
under controlled operation.
6-14
-------
vehicles. Thus data indicate that diesel exhaust has the potential for
being a serious health hazard, but epidemiological data have not confirmed
this to date.
6.3 WATER AND SOLID POLLUTION IMPACTS
Combustion modification control techniques are not expected to
cause any additional water solid pollution impact. Since some, if not
all, of the techniques decrease fuel efficiency, the additional heat not
converted to mechanical energy must be disposed of. This could have a
small impact on the cooling water. Also, if reduced manifold air
temperature is achieved by the use of cooling water, there could be a
small impact on water discharges. These impacts are expected to be minor
except in very unusual cases.
6.4 SUMMARY OF ENVIRONMENTAL IMPACTS
Table 6-3 and 6-4 summarize the qualitative effects of NO
/\
control techniques on CO and HC emissions, respectively. In general, CO
and HC emissions are expected to slightly increase as NO is reduced.
X
This will, of course, not always be the case and the actual changes will
be very engine dependent. The increases in CO and HC emissions are
expected to be less than the decrease in NO emissions. The
A
distribution of hydrocarbon emissions is not very well quantified; thus it
cannot definitely be said that the health effect improvement caused by the
decrease in NO is not offset by the HC increase, although at present
/\
this does appear to be the case.
6-15
-------
TABLE 6-3. EFFECTS OF CONTROLS ON ENGINES LARGER THAN 5.7 x 10-3m3/cyl: CO EMISSIONS
Fuel
Strokes/Cycle
~~~~ ~ ^Air Charging
Control " - -.^^^
Derate
Retard
A/F
TC
MAT
INJ
EGR(I)
EGR(E)
H20
CR
Diesel
Two
BS
f
t
h
t
4-
4-
TC
+
4-
H
i
Four
NA
TC
4-
t
i4-
I-4-
4-
t4-
Dual Fuel
Two
BS
TC
t
i
t
t
Four
NA
TC
H
t
4-
t4-
4-
4-
Natural Gas
Two
BS
t
f
4-
t
t
TC
i
1
t
t
Four
NA
i
4-
t
TC
i
4-
+
t
h Denotes emission increase with application of control
i Denotes emission decrease with application of control
t4- Denotes conflicting data with application of control
Denotes no change in emissions with application of control
Blank indicates no data available on effect
-------
TABLE 6-4. EFFECTS OF CONTROLS ON ENGINES LARGER THAN 5.7 x 10-3m3/cy1 : HC EMISSIONS
Fuel
Strokes/Cycle
~--^Air Charging
Control ^~~~-^- -^^^^
Derate
Retard
A/F
TC
MAT
INJ
EGR(I)
EGR(E)
H20
CR
Diesel
Two
BS
4-
t
4-
i
TC
ti
t-4-
4-
Four
NA
4-
TC
+
t4-
t4-
t4-
4-
t-4-
Dual Fuel
Two
BS
TC
t
t
t
4-
Four
NA
TC
i
4-
t
f+
i
i
Natural Gas
Two
BS
t
t
4-
t
i
TC
t
t
i
+
Four
NA
i
t
t
TC
f
i
1
i
t
Ol
I
^ Denotes emission Increase with application of control
t Denotes emission decrease with application of control
f4- Denotes conflicting data with application of control
Denotes no change in emissions with application of control
Blank Indicates no data available on effect
-------
REFERENCES FOR SECTION 6
6-1. Youngblood, S. B., and 6. R. Offen, Acurex Interoffice Memorandum,
"Emissions Inventory of Currently Installed Stationary
Reciprocating 1C Engines," Acurex Corporation, Mountain View, CA,
September 23, 1975.
6-2. Urban, C. M., and K. J. Springer, "Study of Exhaust Emissions from
Natural Gas Pipeline Compressor Engines," Southwest Research
Institute Report PR-15-61, San Antonio, TX, February 1975.
6-3. "Stationary Internal Combustion Engines. Background Information:
Proposed Standards," EPA-450/3-78-125a, April 1979.
6-4. Letter from W. C. Passie, Caterpillar Tractor Co., Peoria, IL, to
H. Mason, Acurex Corporation, Mountain View, CA, March 15, 1979.
6-5. Storment, J. 0., and K. J. Springer, "Assessment of Control
Techniques for Reducing Emissions from Locomotive Engines,"
Southwest Research Institute Report AR-884, San Antonio, TX, April
1973.
6-6. Roessler, W. U., et al., "Assessment of the Applicability of
Automotive Emissions Control Technology to Stationary Engines,"
EPA-650/2-74-051, NTIS-PB 237 115, July 1974.
6-7. Wasser, J. H., and R. M. Statnick, "Emission Characteristics of
Small Stationary Diesel Engines," Proceedings of the Second
Stationary Source Combustion Symposium, Volume V,
EPA-600/7-77-073e, NTIS-PB 274 897, July 1977.
6-8. Schalit, L. M., and K. J. Wolfe, "SAM/IA: A Rapid Screening Method
for Environmental Assessment of Fossil Energy Process Effluents,"
EPA-600/7-78-051, NTIS-PB 277 088/AS, February 1978.
6-9. Hangebrauck, R. P., et al., "Nomenclature for Environmental
Assessment Projects: Part 1 Terminology for Environmental
Impact Analyses," U.S. Environmental Protection Agency, Industrial
Environmental Research Laboratory, Research Triangel Park, NC,
August 1979.
6-10. Waterland, L.R., and L.B. Anderson, "Source Analysis Modeling for
Environmental Assessment," Presented to Fourth Symposium on
Environmental Aspects of Fuel Conversion Technoloy, Hollywood, FL,
April 17, 1979.
6-11. Cleland, J. G. and G. L. Kingsburg, "Multimedia Environmental Goals
for Environmental Assessment," Volumes I and II, EPA 600/7-77-136a
and b, NTIS-PB 276 919 and NTIS-PB 276 920, November 1977.
6-12. Shin, C. C., et al., "Emissions Assessment of Conventional
Stationary ComFustion Systems; Volume II Internal Combustion
Sources," EPA-600/7-79-029c, NTIS-PB 296 390, February 1979.
6-18
-------
6-13. Barth, D. S., and S. M. Blacker, "The EPA Program to Assess the
Public Health Significance of Diesel Emissions," JAPCA, Vol. 28,
No. 8, pg. 769, August 1978.
6-14. Personal communication with J. Wasser, U.S. Environmental
Protection Agency, Research Triangle Park, NC, April 30, 1980.
6-15. Edited Transcript of Proceedings, "Workshop on Unregulated Diesel
Emissions and Their Potential Health Effects," DOT HS-803 27,
April 1978.
6-19
-------
SECTION 7
SUMMARY OF NEEDS FOR ADDITIONAL DATA
There are two major weaknesses in the available data. The
information on operational effects of these control techniques is not very
complete, especially information about combustion chamber redesign and
catalytic exhaust gas treatment. Also, the data available on emissions
other than NO , CO, and total hydrocarbons and specifications of HC
/\
emissions are very limited.
Research is needed on designing a low NO emitting engine (a
A
program sponsored by the EPA to address this need is discussed below).
Even with the best available controls applied, the large bore stationary
reciprocating internal combustion engine is the worst NO emitter on a
/\
heat input basis of all the major combustion sources. From an energy
viewpoint, these engines tend to be very efficient when compared to other
sources, especially if a heat recovery device is used on the exhaust from
the engine. Even with this high relative efficiency, the NO emitted
A
per power output is still very high. Other emissions from these engines
may also be important but more data are needed for quantification. In
spite of their high emissions, their engines are still preferred in many
operations because of their ease of operation (as compared to a boiler for
example).
7.1 DATA NEEDS
Operational effects and long term effects of combustion
modification control techniques need to be further evaluated. The effects
of retard, derate, air-to-fuel ratio changes, manifold air cooling and EGR
on engine operation are understood but some long term tests should be
conducted to verify that there are no unforeseen problems. Of perhaps of
more importance are tests on catalytic NO reduction and combustion
A
chamber redesign techniques. Since these last two techniques have not
been tested extensively even on a short term basis, much work is required
7-1
-------
to identify operation and maintenance problems, as well as to evaluate
emission control effectiveness. Combustion chamber redesign techniques
share the promise of giving the same emission reductions as the
operational adjustment methods but without the fuel penalties. Exhaust
gas catalytic NO reduction techniques have the potential of giving very
/\
large emission reductions.
A. D. Little under EPA contract has identified the most promising
combustion chamber modification techniques and a report on this work has
been published (Reference 7-1). These most promising techniques are at
present being tested on large bore one cylinder engines. The tests should
be finished by the end of 1981. Modification and testing of a full size
engine will follow (Reference 7-2).
Other than CO, NO , and total hydrocarbons, there is only a small
A
amount of information on emissions from these large bore engines. The
type of organics emitted both as vapor phase and absorbed on particulate
matter is not quantified under NO control conditions. Mobile diesels
J\
have been found to emit mutagenic compounds. Large bore stationary
diesels may also emit these compounds. The amount and size distribution
of the particulate matter need to be quantified.
7-2
-------
REFERENCE FOR SECTION 7
7-1. Wilson, Jr., R. P., "Single-Cylinder Tests of Emission Control
Methods for Large-Bore Stationary Engines," in Proceedings of the
Joint Symposium on Stationary Combustion NOX Control, EPA
IERL-RTP-1087, October 1980.
7.2 Personal communication with R. P. Wilson, Jr., Arthur D. Little Co.,
Cambridge, MA, February 2, 1981.
7-3
-------
APPENDIX A
SAM/IA WORKSHEETS
A-l
-------
TABLE A-2. SAM IA WORKSHEET FOR CONTROLLED DIESEL ENGINE EXHAUST
1. SOURCE/CONTROL OPTION: LEVEL
Diesel Engine/Controlled LEVEL
2. EFFLUENT STREAM
101 Exhaust
CODE NO. NAME
1 X
2
Page 1 of 1
3. TOTAL MASS RATE OF DISCHARGE (g/s)
Q= 3900 g/s
4. COMPLETE THE FOLLOWING TABLE FOR THE EFFLUENT STREAM OF LINE 2.
A
(3 tit
uj 2
B8
ar
co"*
UNITS
CO
NOY
A
Organics
Aliphatic
Hydrocarbons
Substituted
Benzene
Aldehydes
Ketones
POM
(IF MORE SPACE IS NE
B
MEG NUMBER
OH CATEGORY
42B100
47B150
0.1
1.5
0.7
21
EDED USE A
5. TOTAL DISCHARGE SE\
HEALTH BASED ( I col.
ECOLOGICAL BASED (
(ENTER HERE AND AT I
C
SPECIES OR
CATEGORY
CONCENTRATION
yg/m
2.8xl05
4.9xl05
6x1 04
3.7xl04
O.SxlO4
0.3xl04
CONTINUATION S
D
o
_J Ul t_
UJ Q J^J
o
o
yg/m3
4xl04
9x1 03
2xl05
IxlO5
2.5xl02
1.6xl03
HEET)
E
COMPOUND
MEG NO.
ASSUMED
FOR LEVEL 1
HEALTH
B120
A140
A060
A180
/ERITY
w,«a 96.0
I" r.n\ 1) Sh
.INE 8 OF THE SUMMARY SHEET)
F
ECOLOGICAL
DMEG
CONCENTRATION
G
COMPOUND
MEG NO.
ASSUMED FOR
LEVEL 1
- ECOLOGICAL
H
DISCHARGE
SEVERITY
HEALTH
(C/D)
7.0
54.4
0.3
0.4
32
1.9
6. TOTAL WEIGHTED I
HEALTH BASED ( I
ECOLOGICAL BASE
(ENTER HERE AND
i
DISCHARGE
SEVERITY
- ECOLOGICAL
(C/F)
J
V IF
HEALTH
DMEG
EXCEEDED
K
Q
u._i(3Q
QUIUJ
s.UIOU
L
M
WEIGHTED DISCHARGE SEVERITY
< to 3
g/s
27300
21 21 60
1170
1560
124800
7410
_i
-> < -1
g/s
DISCHARGE SEVERITY 5
col L)6a 37x10
D ( £ col. M) 6t
AT LINE 8 OF 1
»
rHE SUMMARY SHEET)
-------
TABLE A-l. SAM IA WORKSHEET FOR UNCONTROLLED DIESEL ENGINE EXHAUST
1. SOURCE/CONTROL OPTION: LEVEL 1
Diesel Enoine/Uncontrol led LEVEL:
2. EFFLUENT STREAM
101 Exhaust
CODE NO. NAME
x
>
3. TOTAL MASS RATE OF DISCHARGE (g/S)
o- 3900 a/s
Page 1 of 1
4. COMPLETE THE FOLLOWING TABLE FOR THE EFFLUENT STREAM OF LINE 2.
A
0UJ
UJ 5
II
12°
ty o
U UJ
UJ t-
o. <
«O
UNITS
CO
NOX
Organics
Aliphatic
Hydrocarbons
Substituted
Benzene
Aldehydes
Ketones
POM
(IF MORE SPACE IS NEI
B
MEG NUMBER
OR CATEGORY
12B100
17B150
01
15
07
21
EDED USE A
5. TOTAL DISCHARGE SEV
HEALTH BASED ( L col.
ECOLOGICAL BASED ( J
(ENTER HERE AND AT L
C
SPECIES OR
CATEGORY
CONCENTRATION
yg/m
1.4xl05
7.7xl05
4x1 04
2.5xl04
0.5xl04
0.2xl04
CONTINUATION S^
D
HEALTH
DMEG
CONCENTRATION
yg/m
4x1 04
9x1 03
2x1 05
IxlO5
2.5xl02
1.6xl03
1EET)
E
COMPOUND
MEG NO
ASSUMED
FOR LEVEL 1
HEALTH
.8120
A! 40
A060
AT 80
r no.6
r ml 1} fih
INE 8 OF THE SUMMARY SHEET)
F
ECOLOGICAL
DMEG
CONCENTRATION
G
COMPOUND
MEG NO.
ASSUMED FOR
LEVEL 1
- ECOLOGICAL
H
DISCHARGE
SEVERITY
- HEALTH
(C/D)
3.5
85.5
0.2
0.2
20
1.2
6. TOTAL WEIGHTED
HEALTH BASED ( I
ECOLOGICAL BASE
(ENTER HERE AND
i
DISCHARGE
SEVERITY
- ECOLOGICAL
(C/F)
J
u.£o§
-J UJ UJ
sps
, UJ
K
D
u. JC3Q
- QUJUJ
vais
UJ
L
M
WEIGHTED DISCHARGE SEVERITY
Z-
KO"J
_1 UJ Z
ass3
xS£
Q/S
13650
333450
780
780
78000
4680
_i
O "
o Q "J
X"JZ
8^3
gSx
UJ ~
g/s
DISCHARGE SEVERITY .. , n5
colL)6a 43x10
ID ( £ col. M) 6
AT LINE 8 OF '
-\
FHE SUMMARY SHEET)
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
Epa-600/7-81-127
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Environmental Assessment of Combustion Modification
Controls for Stationary Internal Combustion Engines
5. REPORT DATE
July 1981
B. PERFORMING ORGANIZATION CODE
. AUTHOR(S) ~
H.I. Lips, J. A. Gotterba, K.J. Lim, and
L.R.Waterland
I. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Acurex/Energy and Environmental Division
485 Clyde Avenue
Mountain View, California 94042
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
68-02-2160
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final: 9/78-7/79
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES IERL-RTP project officer is Joshua S. Bowen, Mail Drop 65,
919/541-2470.
16. ABSTRACT
repOrt gjves results of an environmental assessment of combustion
modification techniques for stationary internal combustion engines, with respect to
NOx control reduction effectiveness, operational impact, thermal efficiency impact,
capital and annualized operating costs, and effects on emissions of pollutants other
than NOx. Currently available operational adjustments for NOx control can reduce
emissions by about 40%, but significantly increase operating costs. The total annual-
ized cost to control can increase the cost of power by 3-14%, due to additional fuel
and maintenance requirements. Combustion modifications can reduce NOx emissions
without significantly increasing CO and hydrocarbon emissions; however, the kinds
and distribution of organic compounds emitted from stationary diesel engines are not
well characterized, and therefore are of concern.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a.
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Group
Pollution
Internal Combustion
Engines
Combustion Control
Assessments
Nitrogen Oxides
Pollution Control
Stationary Sources
Combustion Modification
Environmental Assess-
ment
13 B
21K
21B
14 B
07B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
141
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 <*-73)
-------
&EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-81 -127 July 1982
Project Summary
Environmental Assessment of
Combustion Modification
Controls for Stationary
Internal Combustion Engines
H. I. Lips, J. A. Gotterba, K. J. Lim, and L R. Waterland
This report gives results of an
evaluation of combustion modification
techniques for stationary internal
combustion (1C) engines, with respect
to NOx control reduction effective-
ness, operational impact, thermal
efficiency impact, capital and annual-
ized operating costs, and effects on
emissions of pollutants other than
NOX- Currently available operational
adjustments for NOX control can
reduce emissions by about 40 percent,
but significantly increase operating
costs. The total annualized cost to
control can increase the cost of power
by 3 to 14 percent, due to additional
fuel and maintenance requirements.
Combustion modifications can reduce
NOx emissions without significantly
increasing CO and hydrocarbon emis-
sions for most engines. However, the
kinds and distribution of organic
compounds emitted from stationary
diesel engines are ne
-------
combustion modification controls appli-
cable to these sources with potential for
implementation to the year 2000.
This summary outlines the environ-
mental, economic, and operational
impacts of applying combustion modifi-
cation controls to this source category.
Conclusions
Source Characterization
Stationary reciprocating 1C engines
are the second largest contributor of
NOx emissions from stationary sources in
the U.S. Figure 1 shows that this source
constituted about 19 percent in 1977.
Because of this high level of NOX
emissions and their potential for
control, stationary 1C engines represent
a priority source category for control
evaluation in the NOX EA.
Stationary 1C engines can be classified
into three characteristic size ranges:
large bore, high power, low to medium
speed; medium bore, high speed; and
small.
Large bore engines (>75 kW/cyl)
operate at lower speeds (usually less
than 1000rpm) and burn three major
types of fuel: diesel, natural gas, and
dual fuel (mixture of diesel and gas).
Natural gas engines are spark ignited,
and diesel and dual fuel engines are
compression ignited. Both two- and
four-stroke models are in this size
range, and the engine may be turbo-
charged, which usually increases
efficiency. Typical heat rates are 9 to 11
MJ/kWh (8500 to 10,500 Btu/kWh).
Typical industries using these large bore
engines are municipal electric power
generation, oil and gas pipeline trans-
mission, and oil and gas production. In
these industries, the engine is run
continuously. Based on 1976 data, only
about 1000 to 2000 of these engines
are sold per year, with a total production
value of $80 to $150 million (1976
dollars). Sales have generally been
declining, although sales of diesel
engines for electric power generation
are up.
Medium power engines (7.5 to 75
kW/cyl) exhibit the greatest variety;
some large units equal the power of
large bore engines. However, where
large bore engines produce high power
output at low speeds due to their large
displacement and consequently high
power per cylinder, medium bore
engines have lower power per cylinder
and, therefore, more cylinders for the
same engine horsepower. Fuels burned
in medium power engines are typical
Noncombustion 1.9%
Warm air furnaces 2.0%
Gas turbines 2.0%
Others 4.1%
Incineration 0.4%
Industrial process
Heaters 4.1%
Reciprocating
1C engines
18.9%
Total: 10.5 Tg/yr (11.6 x 10* tons/yr)
Figure 1. Distribution of stationary anthropogenic NO* emissions for the year 1977
(controlled NO* levels).
mobile fuels, either diesel oil or gasoline,
although there are a few (usually
modified) natural gas engines of this
size. These engines are used in miscel-
laneous industrial, commercial, nonpro-
pulsive marine, and agricultural appli-
cations where shaft power is needed
and electric motors cannot be used.
Small engines are mostly one- and
two-cylinder engines of less than 40 hp.
These engines are mostly diesel and
gasoline, one- and two-cylinder models,
with some four-cylinder models. Almost
all have four stroke cycles and are usual-
ly air cooled. Small engines are used
typically in generator sets, small pumps
and blowers, off-the-road vehicles, and
refrigeration compressors for trucks and
railroad cars.
This report focuses on large^ bore en-
gines since these represent the largest
NOX emitters in the category, and they
are most amenable to combustion
modification control.
Source Emissions
Air emissions in the form of exhaust
gases are essentially the only effluent
stream from stationary 1C engines.
Hydrocarbons (HC) can be emitted from
the fuel before combustion, especially
from natural-gas-fired engines, but
these emissions ar.e considered minor.
There may also be some emissions from
the crankcase caused by blowby, but
this is also a minor source. The cooling
system may release minor water pollu-
tant emissions, and liquid wastes in the
form of used crankcase oil may be
another pollutant. Neither of these is a
major release.
NOx, CO, and HC are the major
pollutants of concern in the exhaust
gases from stationary 1C engines. SOX
emissions are possible if thefuel burned
has appreciable sulfur content, but this
is rarely the case with the clean fuels
burned in these engines. Particulate
-------
emissions are low from stationary
engines. Diesel engines may also emit
polycyclic organic matter (POM) at low
levels, but even low level emissions of
these compounds would be of concern
because of their mutagenicity and
potential carcinogenicity.
NOx in 1C engines, as in all combustion
sources, is formed primarily by two
mechanisms thermal fixation and
fuel NOxformation. Thermal NOxresults
from the termal fixation of molecular
nitrogen and oxygen in the combustion
air, and the rate of formation increases
exponentially with local flame tempera-
ture. Fuel NOx resultsfromtheoxidation
of organically bound nitrogen found in
certain fuels and primarily depends on
the nitrogen content of thefuel. Since 1C
engines generally burn clean fuels, with
correspondingly low nitrogen contents,
thermal NOX predominates.
Of the other pollutants, HC and CO
are mainly the result of incomplete
combustion. HC emissions are believed
to be caused by three general mecha-
nisms: wall quenching (fuel impinge-
ment on the walls causing the fuel to be
cooled below the combustion tempera-
ture), variations in engine operation
(mixing inside the cylinder, wrong air-
to-fuel ratio, defective ignition, etc.),
and, in two-cycle engines, cooling the
exhaust gases by scavenging air before
combustion is completed. CO emissions
are also formed by the same general
mechanisms.
Typical uncontrolled emission factors
for 1C engines are listed in Table 1. The
HC emissions listed are total HCs; for
natural gas engines, these are mainly
Table 1.
Fuel
Gasoline
Diesel
Natural gas
Dual Fuel
Emissions Factors for 1C
Engines, g/kWha
> 75 kW
< 15kW
>375 k\A/°
<375 kW°
A/Ox
77.9
7.5
17.3
16.6
15.4
no
CO
137
395
2.4
6.0
3.8
2.7
HC
77.2
27.5
0.6
2.8
6.5
4.1
aEmission factors for gasoline and
diesel engines are modal averages;
those for natural gas and dual fuel are
for rated conditions. Modal averages
mean that some of the A/Ox numbers
are taken from the constant power out
portion of mobile tests.
b Based on an average of rated condition
levels from engines considered.
c Weighted average of two- and four-
stroke engines. Weighting factors =
2/3 for four-stroke and 1/3 for two-
stroke.
methane. Although Table 1 lists factors
for all engine sizes, this report focuses
on the larger engines. Note that NOX is the
major pollutant for large engines.
Control Alternatives
Since NOx is the major pollutant
emitted by stationary large bore 1C
engines, control development has
focused on limiting NOx emissions.
There are three major approaches to
controlling NOx from 1C engines: opera-
tional adjustment, combustion chamber
redesign, and catalytic exhaust gas
treatment. Operational adjustment
techniques can be considered demon-
strated and are finding current appli-
cation. Combustion system redesigns
are currently being developed and have
seen, at best, laboratory scale testing.
The use of catalysts to reduce NOX
emissions from lean-running engines
(selective catalytic reduction) has seen
only laboratory scale testing. Similarly,
early limited testing of NOX reduction
catalysts for rich-running engines (non-
selective catalytic reduction) has been
performed.
The operational adjustmenttechniques
are derate, ignition retard, air-to-fuel
ratio change, reduced manifold air
temperature, exhaust gas recirculation
(EGR) (both internal-restricting the exit
of exhaust gases from the cylinder, and
external-reintroducing exhaust gases
into the intake manifold), and water
injection. All these techniques essen-
tially act to lower the peak combustion
temperatures, thereby limiting thermal
NOx formation. These techniques can be
seen used in combustion, although NOX
reductions are not always additive.
Combustion system redesigns have
been aimed at improving cylinder
mixing, enhancing combustion, or
establishing some form of staged
combustion. The first two allowefficient
combustion to occur under leaner
lower-temperature conditions. The
third, in addition to lowering peak
temperature, lowers oxygen availability
at peak temperature.
For diesel engines, mixing can be
improved by circumferential injection,
chamber shape, or a variable area
prechamber. Combustion in gas engines
can be improved by torch ignition,
multiple spark plugs, high energy spark,
increased turbulence through swirl or
"squish," or diesel fuel injection.
Staged combustion techniques include
divided chambers, open chambers, or
degraded mixing for gas engines, and a
prechamber or pilot injection for diesel
engines.
Catalytic reduction is a flue gas
treatment technique in which exhaust
gas is passed over a reduction catalyst
which reduces N0xto NOg. Nonselective
reduction catalysts can be used with
rich-running engines since very little
oxygen exists in their exhaust. However,
lean-running engines require selective
reduction catalysts which further
require injecting a reducing agent,
ammonia, into the exhaust stream.
Table 2 lists the various combustion
modifications that have been investi-
gated for 1C engines and shows the NOX
reduction and fuel penalties associated
with these controls as a function of
engine type.
Currently, the best demonstrated
controls, the only ones sufficiently
demonstrated to allow meeting the
proposed 1C new source performance
standards (NSPS), are: (1) retarded
ignition or retarded fuel injection, (2)
air-to-fuel ratio changes, (3) increased
manifold air, or (4) in combinations with
the others. The best combination will be
very engine dependent. But in general,
retard is best for diesel-fueled engines,
air-to-fuel ratio changes for natural gas,
and either control for dual fuel. A 40
percent reduction in NOX can usually be
achieved without causing any major
operational problems, but there are fuel
consumption penalties.
For the future, combustion system
redesigns have the potential for obtain-
ing the same level of NOx reduction (40
percent) but with lower costs and fuel
penalty. For very low NOx emissions,
only catalytic reduction techniques
show promise.
Table 3 compares the estimated
annualized incremental costs of retard,
air-to-fuel increase, and exhaust gas
recirculation applied to various engines
to those of the corresponding uncontrolled
engines. Costs in Table 3 represent
annualized costs in mills/kWh (assum-
ing 8000 hours of operation per year)
and are in 1 978 dollars. Table 3 shows
that ignition retard increases the total
cost of power 6 to 7 percent, air-to-fuel
increase increases power costs 3 to 7
percent, and EGR increases power costs
5 to 14 percent. Though not shown in
Table 3, manifold air temperature
reduction should only have a small cost
impact, about a 1.5 percent increase in
initial engine cost and an increase in the
cost of power of about 1 percent. Derate
is a viable technique only if spare power
is available elsewhere. Though derating
-------
Table 2. /VOX Reduction and Fuel Comsumption Penalties for Diesel, Dual-Fuel, and Gas Engines
Engine Fuel Type
Control Approach
Derate 3%
6%
10%
20%
25%
Retard 2°
4°
8°
Air-to-Fuel 2%
3%
5%
±10%
Manifold 311k(WO°F)
Air 315k(107°F)
Temperature 318k(113°F)
Internal EGR
External EGR 10%
Retard and Manifold
Air Temperature
Retard & Air-to-Fuel
Retard and Manifold
Air Temperature and
Air-to-Fuel
Air-to-Fuel and
Manifold Air Temperature
Water Injection 50%
(HzO/fuel ratio) 100%
Catalytic Reduction
(Projected
Combustion Increased
Chamber Mixing
Modifications
(Projected) Staged
Combustion
%/VOx
Reduction
_
<20
5-23
<20
<40
28-45
.
7-8
7-15
5
.
<20
33
<20
10-24
<20
<40
35-65
20
<20
20-30
25-35
50-80
10-30
10-30
Diesel
&BSFC, %a
4
1-5
4
4
2-8
3
0-2
2
7
1
1
0-1
8
16
5-26
0
2
3
2-4
0
<5
0
Dual Fuel
%/VOx
Reduction ABSFC, %a
<20
7-33
<20
<40
50-73
<20
25-40
18-37
<20
20
<20
25
<20
<40
56
<20
40
50-SO
20-40
70-30
4
7-7
3
7
3-5
0
7-3
0-7
7
7
7
2
7
2
2
2
3
0
<5
0-7
Natural
%/VOx
Reduction
<20
<40
5-30
<20
5-40
<20
<40
20-SO
28
<20
<20
5-35
<20
33
<20
30-40
<20
<40
17-52
<20
<40
40-65
25-35
60-75
50-80
20-40
10-30
Gas
ABSFC, %a
2
3
2-72
3
2-7
2
7
5-72
0
0
5
0-8
0
0
3
5-6
4
8
4-11
2
4
6-7
7-2
2-5
0
<5
0-2
a Brake specific fuel consumption penalty.
would increase fuel consumption and
raise operating costs, specific figures
are not given because of the difficulty in
specifying highly site dependent costs.
In general, as shown in Table 3, the
incremental initial capital costs of the
available controls range from 0 to 5
percent of an uncontrolled engine's
cost. However, the total annualized cost
to control can increase the cost of power
from an engine by 3 to 14 percent, the
significant impact due to additional fuel
and maintenance requirements.
By combining control techniques, it
may be possible to achieve the same
NO, reductions with a smaller fuel
penalty, or reduce NOX levels more than
4
could be achieved by each technique
alone. Table 4 compares three different
methods for reducing NOX by40 percent
from a large bore diesel engine. For this
case, there is a definite advantage to
using combined controls since the two
combined techniques, air-to-fuel ratio
change and manifold air cooling (or air-
to-fuel ratio change and retard), had a
lower brake specific fuel consumption
penalty (BSFC) than retard alone.
Cost estimates for combustion system
redesign controls vary significantly due
to the developmental state of these
techniques. Estimates indicate that
these redesigns will fall between 0.5
and 20 percent of the capital cost of a
large engine, with 3 percent being
typical. Operational and maintenance
costs should increase very little because
the goal of development is to keep
BSFC changes negligible.
Operational Impacts
of Controls
Since engines are currently optimized
for minimum maintenance requirements
and fuel use, any control technique
which varies engine parameters from
standard conditions will impact opera-
tion and maintenance. Some of these
impacts have been well characterized,
especially from control techniques
which involve engine operational changes.
-------
Table 3. Annualized Control Costs for 1C Engines3
Control Techniques
Retard
Typical Engine
3000kW Diesel
(Electrical
Generation)
3000 kW Dual Fuel
(Electrical
Generation)
3000 kW Natural Gas
(Gas Transport)
750 kW Natural Gas
(Gas Production)
Uncontrolled Engine
Cost, Mills/ kWh
Capital
Maintenance
Fuel
Total
Capital
Maintenance
Fuel
Total
Capital
Maintenance
Fuel
Total
Capital
Maintenance
Fuel
Total
6
5
32
43
6
5
20
31
6
5
22
33
2
5
25
32
Percent
NO,
Reduction
20-30
20-30
20-30
20-30
Incremental
Cost,
mills/kWh
0
1.6
1.2
2.8
0
1.6
0.7
2.3
0
1.6
0.8
2.4
0
1.6
0.9
2.5
Air- to -Fuel
Ratio Change
Percent
NO,
Reduction
20
40
40
40
Incremental
Cost,
mills/kWh
0
0.2
3.1
3.3
0
0.2
0.4
0.6
0
0.2
0.4
0.6
0
0.2
0.4
0.6
External EGR
Percent
NO,
Reduction
20
20
20
20
Incremental
Cost,
mills/kWh
0.3
2
0
2.3
0.3
4
0
4.3
0.3
4
0
4.3
0.1
4
0
4.1
3Assumes 8000 hours of operation per year, 1978 dollars.
Table 4. Estimated Incremental Cost of Combined Controls for a Large Bore Diesel
Engine at 40 percent NO, Reduction
Incremental
Annua/ized
Control Cost,"
mills/kWh
Capital
Maintenance
Fuel
Total
Retard
0
1.6
2.4
4.0
(^urnrui i ecnniqut
Air-to-Fuel Changes and
Man/fold Air Cooling
0.1
1.0
1.5
2.6
Air-to-Fuel and Retard
0
1.8
1.5
3.3
3 Assumes 8000 hours in operating year, 1978 dollars.
Other control techniques will require
various degrees of evaluation before
impacts are clearly understood.
Derate, air-to-fuel ratio changes,
manifold air cooling, and ignition retard
present the fewest problems in operation
and maintenance. Derate has no impact
mechanically and can improve durability
because of lower operating temperatures
and pressures. However, additional
engines may be required to replace the
lost power. Fuel penalties are usually
low.
Air-to-fuel ratio changes in the lean
direction cause a power loss if larger
blowers or turbochargers must be used.
If the engine must be operated richer to
reduce NOx emissions, other emissions
(e.g., smoke, CO, and HC) can increase.
This could cause an attendant increase
in engine maintenance. Changes in air-
to-fuel ratio in either direction will also
generally increase fuel consumption.
Finally, operation at air-to-fuel ratios
where misfiring or detonation occur can
cause severe engine damage.
Increased manifold air cooling has
little operational or maintenance impact
for a unit that is already intercooled, but
will increase the size of the heat
exchanger, water or air pump, control
system, and other system components.
Of course, backfilling intercooling on an
engine will add maintenance attendant
to additional temperature reductions.
Changes in fuel consumption are small.
When properly applied, ignition retard
has no serious mechanical drawbacks.
Some increase in operational and
maintenance time would be needed to
ensure that the degree of retard is
always within safe limits. Increases in
fuel consumption are moderate. Exces-
sive amounts of retard, however, can
create severe engine problems. Fuel
consumption increases rapidly, power
drops, misfiring can occur, and smoke
levels increase. In addition, mechanical
maintenance will increase if the exhaust
temperature exceeds the safe limits for
valves or the turobcharger (usually 920
K-1200°F). More frequent engine
teardown will be required, and higher
initial costs will result for higher
temperature materials.
Exhaust gas recirculation requires
new hardware components which may
require added maintenance, Problems
of fouling the flow passages of the
cooling heat exchanger, the engine
turbocharger, and the aftercooler with
particulate must be solved, or frequent
engine teardown will be required.
Under varying load conditions a so-
phisticated control system is required
or the engine may stall or emit unac-
ceptable smoke levels. Fuel consump-
tion penalties with EGR are small.
Water injection can cause severe
maintenance problems. Deposits from
untreated water can build up on
internal engine surfaces, and also foul
the lubricating oil. The problem can
lead to major engine maintenance.
Water injection also adds another
system to the engine which must be
maintained and controlled.
Although not demonstrated, com-
bustion system modifications are
expected to present the least impact to
-------
operation and maintenance. Mainten-
ance requirements can be expected
to increase slightly if additional injectors,
spark plugs, and valves are added to
the chamber. However, because this
control technique involves new design,
many of the additional maintenance
requirements can be designed out.
Fuel penalties are expected to be small.
Catalytic reduction will require no
additional engine maintenance, since it
is a flue gas treatment technique,
rather than an engine modification.
However, operating the catalyst system
may be expensive. Fouling the catalytic
surfaces with particulate may require
frequent regeneration. The catalyst
may also have a relatively short life and
need to be replaced. Another system,
ammonia injection, must be included in
engine operation. Finally, harmful
products of the reaction may be
produced if the catalyst temperature
varies from the proper level, or if
excessive ammonia is injected. The
catalyst must be installed and operated
on an engine before all these effects
can be quantified.
Currently available operational ad-
justment NOx controls can only reduce
emissions by approximatley 40 percent,
while significantly increasing operating
cost and maintenance. Advanced com-
bustion chamber redesigns have the
potential of achieving similar NOX
reductions but at lower cost and smaller
fuel penalty. If very low NO* emissions
are required, catalytic exhaustgastreat-
ment is the only developing technique
with that potential.
Table 5 lists achievable control levels
and associated control techniques and
costs for typical diesel, natural gas, and
dual fuel engines, all assumed to be
turbocharged. In the case of natural gas
and dual fuel engines, the obvious
preferred approach from a cost-effec-
tiveness view would be to go directly to
the more stringent control level with air-
to-fuel adjustment. Note that all values
discussed a re typical, and may vary from
engine to engine.
Combustion modification controls
can reduce NOx emissions without
significantly increasing CO and HC
emissions from most engines. However,
the kinds and distribution of organic
compounds emitted from diesel engines
are not well characterized and, there-
fore, are of ootential concern.
Recommendations
There are two major weaknesses in
the data base for combustion modifica-
Table 5. Projected Control Requirements and Costs for Alternate NO* Emission
Levels
Type
Diesel
Natural Gas
Dual Fuel
/VOx Emission,
g NOx/kWh output
17
14
12
10
15
12
11
9
11
9
8
7
Control Techniques
Baseline
Exhaust gas
recirculation
Retard
A/F increase + retard
Baseline
Exhaust gas
recirculation
Retard
A/F increase
Baseline
Exhaust gas
recirculation
Retard
A/F increase
Control Cost,
mills/kWh
output
2.3
2.8
3.3
4.3
2.4
1.0
4.3
2.3
1.0
tion controls on 1C engines. The infor-
mation on operational effects and long-
term durability of these control tech-
niques is incomplete, especially con-
cerning combustion system redesign
and catalytic exhaust gas treatment.
I nformation on combining these controls
to achieve an optimum of lowemissions
other than NOX, CO, and total HC is very
limited. The amounts and types of
organics emitted from these large bore
engines are not very well characterized.
The potential mutagenicity of organic
emissions in diesel exhaust is of major
concern.
Research is needed on designing a
high efficiency low-NO* emitting engine.
Even with the best available control
applied, the large bore stationary
reciprocating 1C engine is the highest
NOX emitter on a heat input basis of all
major combustion sources.
EPA is currently sponsoring several
programs in the health effects area as
well as new engine designs for low-NOx
and high efficiency. These programs
should help resolve many of the major
data gaps in the operational and
environmental impacts of NOX controls.
H. I. Lips, J. A. Gotterba. K. J. Lim, and L. R. Water/and are with Acurex
Corporation, Energy and Environmental Division, Mountain View, CA 94042.
J. S. Bo wen is the EPA Project Officer (see below).
The complete report, entitled "Environmental Assessment of Combustion
Modification Controls for Stationary Internal Combustion Engines," (Order
No. PB 82-224 973; Cost: $ 13.50, 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:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
US.GOVERNMENT PRINTING OFFICE:1882-559-092-423
------- |