Electro motive EMD 645 User manual

NAVAL FACILITIES ENGINEERING SERVICE CENTER
Port Hueneme, California 93043-4370
USER GUIDE
UG-2022-ENV
USER GUIDE FOR CONVERSION TO DUAL FUEL
OPERATION OF EMD 645 ENGINES ON NAVY MUSE
GENERATOR SETS
by
Norman L. Helgeson, Ph.D.
Scott E. Jensen, Eugene Conversions, Inc.
Allen E. Willey, POl
Delbert Terrell, P02
Dave Vawter, POl
Michael Childers, POl
March 1997
Sponsored by
Naval Facilities Engineering Command
19970825 130
DTIG QUALITY INSPECTED S
Approved for public release; distribution is unlimited.
^\ Printed on recycled paper

REPORT DOCUMENTATION PAGE
Form Approved
OMB No. 0704-018
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching
existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this
burden estimate or any other aspect of this collection information, including suggestions for reducing this burden, to Washington Headquarters Services,
Directorate for Information and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget,
Paperwork Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE
March 1997
3. REPORT TYPE AND DATES COVERED
Final
4. TITLE AND SUBTITLE
USER GUIDE FOR CONVERSION TO DUAL FUEL
OPERATION OF EMD 645 ENGINES ON NAVY MUSE
GENERATOR SETS
5. FUNDING NUMBERS
6. AUTHOR(S)
N.L. Helgeson, Ph.D, S.E. Jenson (Eugene Conversions, Inc.),
A. Wiley, D. Terrell, D. Vawter, and M. Childers
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESSE(S)
Naval Facilities Engineering Service Center
1100 23rd Ave
Port Hueneme,CA 93043-4370
8. PERFORMING ORGANIZATION REPORT
NUMBER
UG-2022-ENV
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESSES
Naval Facilities Engineering Command Chief of Naval Operations
200 Stovall Street 2000 Navy Pentagon (Code N45)
Alexandria, VA 22332-2300 Washington, DC 20350-2000
10. SPONSORING/MONITORING AGENCY REPORT
NUMBER
11. SUPPLEMENTARY NOTES
12a. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution is unlimited.
12b. DISTRIBUTION CODE
13. ABSTRACT (Maximum 200 words)
Conrversion of the Navy's Mobile Utilities System Equipment (MUSE) diesel generators to dual fuel (natural gas plus
diesel fuel) operation to reduce their NOx (nitrogen oxide) emissions is described. The Navy maintains a fleet of 60 such
units, powered by EMD 645 engines, that range in size from 1,500 kW to 2,500 kW. A means for providing low-NOx
MUSE generators that will not compromise the operational capability of the fleet is required.
This guide is based upon the experience gained from converting a 1,500 kW MUSE unit to dual fuel operation (the
unit is currently installed and operating at King's Bay Naval Station in Georgia). Information on the conversion process,
operation, maintenance, and the cost of the dual fuel conversion are provided.
14. SUBJECT TERMS
Diesel generators, MUSE, dual fuel conversions, low-NOx emissions, EMD 645 engines, and
natural gas operations.
15. NUMBER OF PAGES
166
16. PRICE CODE
17. SECURITY CLASSIFICATION
OF REPORT
I Unclassified
18. SECURITY CLASSIFICATION
OF THIS PAGE
Unclassified
19. SECURITY CLASSIFICATION
OF ABSTRACT
Unclassified
20. LIMITATION OF
ABSTRACT
UL
NSN 7540-01-280-5500
Standard Form 298 (Rev. 2-89)
Prescribed by ANSI Std. 239-18

EXECUTIVE SUMMARY
Steps are described for converting the Navy's Mobile Utilities System Equipment
(MUSE) diesel generators to dual fuel (natural gas plus diesel fuel) operation to reduce their NOx
(nitrogen oxide) emissions. The Navy maintains a fleet of 60 such units, powered by EMD 645
engines, that range in size from 1500 kW to 2,500 kW. As the current MUSE diesel units meet
few existing local NOx emission regulations, newer more-restrictive regulations (a trend now
ensured with the proposed further tightening of the National Ambient Air Quality Standards)
will ensure that the areas accessible to MUSE diesel generators will continue to decrease.
Therefore a means for providing low-NOx MUSE generators that will not compromise the
operational capability of the fleet is required.
This guide is based upon the experience gained from converting to dual fuel operation a 1500
kW MUSE unit that is now installed and operating at the King's Bay Naval Station in Georgia.
Information on the conversion process, operation, maintenance and the cost of the dual fuel
conversion are provided.
NOx emission reductions of 70 percent were demonstrated with this unit; NOx reductions of
greater than 90. per cent are achievable using an additional secondary ignition cell. Capital
requirements for the retro-fit dual fuel conversion of a 2500 kW MUSE engine generator are
shown to be $158. per kilowatt (including the secondary ignition cells). This is compared to the
cost of replacing this same MUSE unit with a new spark-ignited, low-NOx, natural gas engine.
The cost for the latter is $560. per kilowat. For a 2500 kW MUSE unit, these numbers translate
into a capital savings of over $1.0 million if the option of retro-fitting for dual fuel operation is
chosen. Operating costs for dual fuel (natural gas) operation of MUSE generating sets are also
substantially reduced over that of unmodified diesel operation. For the case evaluated, the cost
for natural gas firing is 4.1 cents per kW-hr while that for diesel firing is 8.9 cents kW-hr.
A further major operational (strategic) advantage of the retro-fit dual fuel unit over that of a
new, spark-ignited, natural gas engine is that it can be made to operate as either a full diesel or a
full natural gas unit.
DTIC QUALFiT ESBPi30'.i.'iSD 3
in

ACKNOWLEDGEMENTS
XT ,?
S
. ?
r0jeCt WaS fonded by the Pollution
Abatement Ashore Progam, managed by the
Naval Facilities Engineering Command, and sponsored by the Environmental Protection Safety
and Occupational Health Division (N45) of the Chief of Naval Operations.
This project would not have been possible without able assistance of many people A
technically challenging aspect of the project was automation of the controls which was required
to ensure the safe handling of natural gas within the engine house. Master Chief Petty Officer
(retired) Ronald Kluender's intimate knowledge of the original MUSE control system and Scott
Jensen s thorough understanding of the ECI dual fuel conversion engine control unit (ECU)
made possible the successful integration of those two control systems. John Pesar Program
Manager for MUSE, made available a 1,500 kW MUSE engine generator set to serve as the
prototype and approved use of MUSE personnel to assist with the conversion to dual fuel
operation. MCPO Russell Dominy, MCPO Jim Riley, POl Anthony Fourage, and POl Rodney
Hood also assisted.
J
WAVS??
Keene
n
0 f
L
S UBASE Kin
§'
s Ba
y P
rovi
ded a field test site along with funds from
NAVSEA to install the dual fuel unit at King's Bay. He and Bill Strickland arranged for
installation of the unit there, oversaw its operational schedule, and provided a strong presence in
resolving field operational problems. Their assistance was invaluable. Ralph Kerwin of Gage
£ A°^C?
Safety COnsuItants
)
heI
P
ed t0 defi
"e hardware and procedures for safely operating
he MUSE engine generator while using natural gas. Kevin Beaty of Southwest Research
Institute (SwRI) recognized the potential value of the results of this project to the efforts to
GasRail, Inc for developing lo-NO
x
engines for locomotive use, and provided flowrate
measuring and data logging instrumentation for the project. Jack Smith and Butch Quip of SwRI
installed that instrumentation.
_ George Warren and Ed O'Neil of NFESC provided structural recommendations for the
installation and mounting of a new compressor and water pump. Wayne Tanaka of the Natural
Gas Vehicle Division of the Southern California Gas Company (SCGC) made special
arrangements for weekend use of the SCGC compressed natural gas tanker along with a high-
capacity pressure reduction manifold and the personnel to operate them for shakedown tests
Manny Perez, Bob Saunders, Ken Hanzlick, and Galen Marks of NFESC gave up their weekends
to assist with measurements during the shakedown tests.
pS

CONTENTS
Page
1.0 INTRODUCTION 1-1
1.1 Environmental Compliance and Navy Need 1-1
1.2 Selection of NO
x
Control Technology 1-2
1.3 Installation and Testing of Prototype Dual-Fuel MUSE Diesel Generator 1-2
1.4 Organization of the Guide 1-3
2.0 NO
x
REDUCTION FOR LARGE, TWO-STROKE DIESEL ENGINES 2-1
2.1 Exhaust Gas Treatment 2-1
2.2 NO
x
Reduction From EMD 645 Dual-Fuel Engines 2-2
2.3 Further Rduction of NO
x
Emissions 2-3
3.0 ELEMENTS OF THE ECI/EMD DUAL-FUEL CONVERSION 3-1
3.1 Natural Gas Fuel 3-1
3.2 Dual Fuel Characteristics 3-1
3.3 Engine Operating Sequence 3-2
3.4 Conversion Kit Components 3-3
3.4.1 Duel Fuel Heads and Pistons 3-3
3.4.2 Gas Inlet Valves 3-3
3.4.3 Pilot Fuel Control and Diesel Injection 3-3
3.4.4 Low Emission Idle (LEI) 3-4
3.4.5 Electronic Control Unit 3-4
4.0 SUB-SYSTEM DESCRIPTIONS AND INSTALLATION PROCEDURES 4-1
4.1 Engine Modifications 4-1
4.1.1 Power Pack Assembly 4-1
4.1.2 Cylinder Relief Valves 4-1
4.1.3 Diesel Fuel Supply 4-1
4.2 Natural Gas Supply System 4-2
4.3 Control Air System 4-2
Vll

CONTENTS
Page
4.4 Air Throttle
4
"
3
4.5 Engine Cooling System
4
"
3
4.6 Sensors
4
"
4
4.6.1 Thermocouples
4
"
4
4.6.2 Water Temperature
4
~
4
4.6.3 Control Air Pressure
4
"
4
4.6.4 Flywheel Sensors
4
"
4
4.6.5 Air Box Temperature
4
~
4
4.6.6 Air Box Pressure
4
"
4
4.6.7 Natural Gas Sensors
4-4
4.6.8 Gas Header Pressure
4
"
4
4.7 Natural Gas Safety System
4
"
5
4.8 Electrical System and Controls
4_5
4.8.1 Engine Control Unit
4
"
6
4.8.2 Air Service Cabinet and Engine Control Panel 4-6
4.8.3 GIV Wire Harness
4
"
6
4.9 Switchgear, External Power Hookups and External Communications 4-6
4.10 Modification of MUSE Schematic Drawings 4-7
5.0 OPERATIONAL PROCEDURES 5"
1
5.1 Diesel Only Operation 5-1
5.2 Natural Gas (Dual Fuel) Operation
5_1
5.3 Automatic Control
5
"
2
5.3.1 Autostart Software
5-2
5.3.2 ECU Alarm Logic
5
"
4
5.3.3. ECU Status Messages
5_4
5.4 The ECU •
5_6
5.4.1 ECU Status Screen
5_6
5.4.2 Computer Interface
5
"
6
5.4.3 Using LCD Screen Editor
5
"
6
5.4.4 ECU Dial-Up Software
5
"
7
5.4.5 Static Protection Precautions ••• ••
5_7
Vlll

CONTENTS
Page
6.0 FIELD SITE INSTALLATION, MAINTENANCE AND TRAINING 6-1
6.1 Site Requirements "
-1
6.2 Maintenance Schedule and Procedures
6_1
6.3 Training
6_1
7.0 CAPITAL AND OPERATING COSTS 7-1
8.0 REFERENCES
8_1
9.0 GLOSSARY
9_1
APPENDICES
A - Letter Report Regarding Fire Safety Measures for Conversion of
USN MUSE Diesel-Generator Units From Diesel to Natural Gas Fueling A-l
B - Schematic Drawings of Dual Fuel 1,500 kW Conversion of
Diesel Power Plants
B_1
C - Maintenance Bulletins and Instructions
c_1
IX

1.0 INTRODUCTION
This user data package describes steps in the conversion of the Navy's Mobile Utilities
System Equipment (MUSE) transportable diesel generator sets from diesel to dual fuel (natural
gas plus diesel fuel) operation. These diesel generator sets are part of a fleet of 60 such units
maintained by the Navy's MUSE detachment for deployment to Navy, other military, and
sometimes civilian uses throughout the world (Ref 1-1). The units range in size from 750 to
2,500 kW and are powered, largely, by EMD 645 engines manufactured by the ElectroMotive
Division of the General Motors Corporation. The engines are medium-speed (750 or 900 RPM),
have 8 to 20 cylinders with cylinder displacements of 645 cubic inches, and use a 2-stroke
operational cycle. This guide is based on the experience gained from converting to dual fuel
operation a 1,500-kW MUSE unit that is now installed and operating at the King's Bay Naval
Station in Georgia (Fig. 1-1). The technology used for this conversion was provided by ECI Inc.,
Tacoma, Washington. MUSE personnel, assisted by ECI technicians, installed the required
engine modification hardware and the electrical and engine skid modifications.
1.1 Environmental Compliance and Navy Need
Of the Navy's many diesel engines, those installed in its MUSE generating equipment for
temporary (<4 years) stationary electrical power were the first off-road Navy engines to be
affected by environmentally-mandated nitrogen oxide (NO
x
) emission regulations. As MUSE
units are transportable but are typically employed at a given site for periods exceeding that
normally used to define 'temporary power' (temporary power is often defined as service not
exceeding one year), they must comply with air pollution regulations that apply to stationary
power generating equipment. But regulatory emission standards for stationary power units are
highly variable as they both depend upon the degree of nonattainment in the surrounding air
control region and upon the air control district in which the region is located. Therefore MUSE
units must have a capability for meeting these variable standards as the units are moved from one
location to another. Such regulations can vary from none to those applicable in the South Coast
Air Quality Management District (SCAQMD) of California, where emission limits are < 1.0
gram/horsepower-hour (gm/HpH). As NO
x
emissions from current MUSE units range anywhere
from 12 to 16 gm/HpH, attempts to operate them in areas in which they do not comply with
emission regulations could result in the imposition of heavy fines and as well as ordered shut-
down by civilian authorities. Therefore a means for reducing NO
x
emissions from them is
required.
The following factors and criteria were determined to be important in evaluating
approaches for reducing NO
x
emissions from MUSE diesel generator sets: (a) most MUSE
engine generator units are driven by an EMD 645 engine, (b) when burning diesel fuel, these
EMD engines do not meet the emission requirements in many areas of the country where the
Navy operates, (c) although transportable, MUSE units must comply with emission regulations
that apply to stationary devices because of their extended time on site, (d)the required
transportability of the MUSE units limits the types of NO
x
emission control technology that can
1-1

be used with them, (e) the large number of MUSE units in inventory and the cost of then-
replacement requires that a retrofit rather than a replacement technology be acquired for reducing
NO
x
emissions, (f) a capability for firing diesel fuel must be maintained to meet the Navy's
operational commitments, (g) full generator power capability must be maintained, (h) reliability
equal to or better than that for the diesel-only configuration is needed, (i) the technology selected
must be within the technical expertise of MUSE and onsite personnel who will be required to
operate and maintain the modified units, and (j) operational costs must not escalate appreciably
beyond those incurred when the MUSE unit is operating as a diesel-only engine.
1.2 Selection of NO
x
Control Technology
Two general approaches can be used for reducing NO
x
emissions from diesel engines:
(a) modification of the engine combustion process so that fewer NO
x
species (nitric oxide, NO,
and nitrogen dioxide, N0
2
) are generated, and/or (b) treatment of the exhaust gases to destroy the
NO
x
species generated to prevent their emission into the atmosphere. Of the many variations of
these approaches that have been investigated, two were considered for application to MUSE
diesel generator sets: (a) conversion of the engine to dual fuel operation, and (b) the use of
selective catalytic reduction (SCR) for treatment of the engine exhaust stream. Although SCR
offers the advantage of bringing the MUSE engine generators into immediate full compliance
with the most restrictive of NO
x
emission regulations, its application would add significant cost,
complexity, and size (requiring that an additional module be transported) to the transportable
MUSE engine generator set, and would also introduce ammonia (NH3, a hazardous chemical) at
the operating site. NH3, which is used as the chemical reductant for NO
x
, is not only a hazardous
chemical, but, itself, presents a threat to the environment. Therefore the SCR process using NH3
as the reductant was not considered to be a practical solution for reducing NO
x
emissions from
MUSE diesel generator sets.
The recommended application of dual-fuel technology (or alternative fuel) is based upon
a retrofit conversion kit designed, specifically, for the EMD 645 engine by Energy Conversions,
Incorporated, Tacoma, WA. The performance, reliability, and emissions reduction of that
conversion technology have been demonstrated in a single application on two tandem
locomotives used for coal line-haul operations on the Burlington and Northern Railway between
Montana and Wisconsin (Ref 1-2). Liquefied natural gas (LNG) from a separate LNG fuel car
was used to fuel the tandem locomotives, and NO
x
reductions of nearly 70 percent (to 4.0 from
the 12.0 gms./Hph produced when burning diesel fuel) were achieved. Because of these
demonstrated results, because no other competing technology could make similar claims with the
EMD 645 engine, and because this technology satisfied the criteria for selection of a MUSE NO
x
reduction technology (discussed above), it was selected for installation and testing on a 1,500 kW
MUSE diesel generator set.
1.3 Installation and Testing of Prototype Dual Fuel MUSE Diesel Generator
The retrofit, dual fuel conversion was completed in July 1995 at the Construction
Battalion Center in Port Hueneme, California. Although previously tested in a locomotive
application, it had not been used for stationary power. And although the application to stationary
power is, in ways, not as complex as that for locomotives, it is different and presents other
1-2

significant challenges. To meet these new challenges, to accommodate the use of natural gas
within the MUSE unit, and to provide a capability for the remote operation of the engine
generator unit, a new automated control system was designed and installed along with the dual-
fuel conversion. Operational checkout tests of the modified unit provided the emissions data that
are summarized in Table 1-1. The tests also showed: (a) that the unit could be started,
electrically synchronized with a grid, stopped, and otherwise controlled remotely, (b) that full
engine generator set power could be produced, and (c) that NO
x
emissions could be reduced by
65 percent, the approximate level anticipated by the dual fuel conversion. The modified unit was
then transported to SUBASE King's Bay Georgia where it is now undergoing field testing.
Further engine efficiency, fuel usage, power, and emissions data will be obtained during the field
test period along with maintenance and other operational and control data.
1.4 Organization of the Guide
This guide contains the information necessary for installing, operating, and maintaining
the dual fuel conversion hardware supplied by ECI Inc., Tacoma WA, when used on MUSE
diesel generator units that utilize the EMD 645 engine. Section 2 reviews some of the factors
involved in NO
x
production in 2-stroke diesel engines and some of the efforts that have been
undertaken previously to minimize those emissions. Suggestions are included for making further
important reductions in NO
x
emissions from the EMD 645 engines. Section 3 describes the
fundamental elements of the ECI dual fuel conversion kit, and Section 4 provides detailed
descriptions and installation instructions for the conversion sub-systems, as defined for this new
application. Section 5 describes operational procedures of the converted unit and Section 6
discusses field site installation, maintenance, and training requirements. Section 7 provides
capital and operating costs for the retrofit, dual-fuel MUSE diesel generator.
1-3

(a) Side view of switchgear, interconnecting cable housing, and engine house.
IM*fc*fci&i
(b) End view showing natural gas supply line.
Figure 1-1. MUSE 1,500 kW dual-fuel engine generating set at SUBASE, King's Bay, Georgia.
1-4

Table 1-1. Comparison of Measured Emissions Before and After Dual-Fuel Conversion
Engine
Emissions
Fuel
Power
(hp)
NO
x
(grams/hph)
CO
(grams/hph)
Particulates
(grams/hph
Before
Conversion
Diesel
2,119
10.5
0.25
0.35
After
Conversion
Dual Fuel
2,119
2.40
9.2 0.211
Dual Fuel
2,402
3.42
11.4
—
1-5

2.0 NO
x
REDUCTION FOR LARGE, TWO-STROKE DIESEL ENGINES
Of the many combustion devices having large commercial usage, diesel engines have
provided the greatest technical challenges to achieving significant NO
x
reductions. Two general
approaches can be used to reduce NO
x
emissions from them: (a) modification of the engine
combustion process so that fewer NO
x
species (nitric oxide, NO, and nitrogen dioxide, N0
2
) are
generated, and/or (b) treatment of the exhaust gases to destroy the NO
x
species generated and to
prevent their emission into the atmosphere. As diesel engines are lean-burn (stoichiometrically,
fuel lean), the catalytic reactors used by spark-ignited gasoline engines to reduce NO
x
and
hydrocarbon emissions and which require near-stoichiometric combustion, are not effective for
reducing the concentrations of the same pollutant species from diesel exhausts. Further, diesel
engines operate at higher compression ratios than gasoline engines and combustion in them
proceeds according to a complex, heterogeneous diffusion process as opposed to the simpler,
easier-to-control, homogeneous flame propogation that characterizes gasoline engine combustion
(see Fig. 2-1). The higher compression ratio of the diesel engine leads to its greater efficiency
and the diffusion-controlled combustion is what permits its operation at higher pressures without
the occurrence of destructive knocking. However, both of these factors lead to higher
combustion temperatures and increased NO
x
production in the diesel engine.
Most of the work for reducing NO
x
emissions from diesel engines has been with smaller
sized engines. Little research has been carried out on engines as large as the EMD 645. And as
the complexity of combustion increases and one's ability to control those processes decreases
with engine size, progress in NO
x
emissions reduction from larger engines has been comparably
slow. However, as characterization of the fundamental processes in diesel engines improves, the
results for controlling NO
x
emissions from both large and small engines may improve. The
Navy's Office of Naval Research (ONR) is sponsoring fundamental research (see Refs 2-1, 2-2,
and 2-3) on the control of NO
x
emissions from the larger engines used for boat and ship
operations. Those results could also find an application with MUSE engines. However, the
practical application of that research, if successful, is several years in the future. A further
complication with the EMD 645 engines is that they are 2-stroke engines. This reduces the
number of engine parameters available for controlling or altering the internal engine processes
for the purpose of NO
x
reduction. Further, most work on NO
x
reduction from diesel engines has
been on those of the 4-stroke design. Therefore, the quantity of engine research to draw on for
reducing NO
x
emissions from the EMD 645 engine is limited.
2.1 Exhaust Gas Treatment
Exhaust gas treatment provides one option for significantly reducing NO
x
emissions from
diesel engines, although the additional equipment and chemicals required for its implementation
works against its application on MUSE units. Two such approaches are available commercially;
others are being developed. Those available commercially use a chemical reductant either with
catalysts (selective catalytic reduction (SCR)), or without catalysts (selective non-catalytic
reduction (SNCR)). The reductants used are usually ammonia or a related compound such as
2-1

urea. Other than the identification of a useful chemical additive, the development of an
appropriate catalyst is the major technical problem in the development of a useful SCR approach.
Although SNCR requires no catalyst, other chemical routes and temperatures for the destruction
of NO
x
must be used. The general features and flow paths of the two approaches are contrasted
in Figure 2-1.
Ammonia has been the chemical most used for NO
x
reduction, but recent years have
seen development of the use of related chemicals (e.g., urea and cyanuric acid). Ammonia and
urea have both been used in SNCR and SCR processes, but the temperature window of their
application for SNCR is too high to be useful with diesel exhausts. The use of urea in SCR
processes is a rather recent development and offers significant advantages over that of ammonia.
SCR (urea) provides at least the NO
x
reduction that SCR (NH
3
) provides (both >90 percent), but
without the burden of being either a hazardous chemical or an environmental threat. The use of
SCR (urea) was developed in Europe and is currently available only from European suppliers.
Cyanuric acid has been used for SNCR (non-catalytic) processes. Its major claim is that it is
more effective than either ammonia or urea at the lower temperatures characteristic of diesel
exhausts.
2.2 NO
x
Reduction From EMD 645 Dual-Fuel Engines
Several efforts have been undertaken to use natural gas with 2-stroke diesel engines.
Those of interest to the MUSE engine generator program are summarized in Table 2-1. The first
line provides operating data from current MUSE EMD 645 diesel engines as a baseline for
comparison. The remaining cases are for engines from the same locomotive-type engine series
(the EMD's 567, 645, and 710, cylinder displacement in cubic inches) and the Detroit Diesel 92
(a truck-sized engine). Fuel type is either dual fuel (DF) or natural gas (NG), and charging of the
engine cylinders with natural gas is described as either early or late in the compression stroke.
For early injection (El), a low injection pressure (100 to 300 psi) can be used. Air flow and
mixing patterns within the cylinder then produce a nearly homogeneous fuel/air mixture prior to
ignition. For late cycle injection (LCI), a much higher pressure (>3,000 psi) is required, and as
pre-mixing of the fuel and air is not achieved prior to ignition, a diesel-type combustion takes
place. Ignition of the natural gas charge in both of these cases is by injection of a pilot quantity
of diesel fuel (4 to 7 percent of the amount required for full diesel operation) into the cylinder at
the time of ignition. Significant NO
x
reduction is achieved with the El process. NO
x
reduction is
much more difficult to achieve for the LCI process.
Anticipated advantages of LCI are increased engine thermal efficiency (a higher
compression ratio diesel cycle can be used) and avoidance of the combustion "knock" that limits
power production in Otto cycle (homogeneous combustion) engines. Disadvantages of LCI are
the technical problems and high cost of providing natural gas at pressures greater than 3,000 psi
and the difficulty in achieving reduced NO
x
emissions in higher-pressure, higher-temperature
combustion environments. The advantage of reduced fuel costs due to the use of natural gas
rather than diesel fuel is provided by both approaches.
The data of Case 1 were obtained in 1982 from a locomotive engine fueled from
compressed natural gas cylinders carried on a railroad flatcar (Ref 2-4). The major limitation
observed in that test operation was the reduction (20 to 30 percent) of available engine power
2-2

caused by the onset of combustion knocking. In Case 2, the Department of Energy (DOE)
sponsored single-cylinder engine tests (Refs 2-5 and 2-6) to determine if improved engine
thermal efficiencies and power production could be obtained with a diesel-type (late cycle
injection) combustion process for dual fuel operation. Although substantial progress was made in
these tests, engine efficiency and exhaust emission objectives were not achieved. The DD 92
spark-ignited, truck-sized engine (Case 3) had significantly different requirements than those for
the larger locomotive engines. Its manufacture was discontinued after several years' production.
Case 4 was undertaken by many of the same parties involved in Case 1, but with several
changes. These included: using turbocharging rather than positive displacement compressors for
charge-air compression; changing the piston crown design to improve fuel-air mixing and to
reduce the compression from 14.5 to 12.8; altering the head configuration to allow admission of
natural gas into the chamber using electronic rather than mechanical controls; incorporating
additional charge-air cooling for the engine; modifying the controls of the injector rack; and
incorporating single-bank engine idling to reduce emissions and improve engine efficiency (Ref
2-7). The results achieved were sufficient to convince the Burlington and Northern Railway to
install retro-fit packages designed, specifically, for the EMD 645 engine on two tandem
locomotives for coal line-haul operations between Montana and Wisconsin. Liquefied natural gas
(LNG) was used as the fuel, and a separate LNG fuel car was constructed to provide fuel. An
LNG refueling station for the round trip was provided in Minnesota. The performance,
reliability, and emissions reduction of that technology were demonstrated in a program that
continued for over two years. The technology represented by this demonstration test appears to
be the only engine-related retro-fit technology available for achieving significant NO
x
reductions
with the EMD 645 engine.
The last entry in Table 2-1 is for development tests that have been underway at the
Southwest Research Institute (SwRI) for development of the late, high-pressure injection of
natural gas. Data regarding the performance, emissions, cost, and reliability of this technology
have not yet been published.
2.3 Further Reduction of NO
x
Emissions
Although the 70 percent reduction in NO
x
demonstrated by the results of Case 4 is an
important step forward, further NO
x
reductions appear to be available with some additional
modification of the dual-fuel EMD 645 engine. These have centered on the use of pre-ignition
chambers and/or separate diesel injection igniters (see Table 2-2). The results presented in this
table show that NO
x
emissions for dual fuel injection can be reduced to the level of 1.0 gram/hph
while the required pilot fuel is reduced to about 1.0 percent (Reference 2-8 and 2-9 indicate that
CO, hydrocarbon, and particular emission are also significantly reduced.). Although the use of
pre-chambers in engines is a familiar concept, their use with dual fuel engines has been limited
and their significant advantages have not been widely explored. In the cases shown, the
Wartsilla is a somewhat complex multi-fueled application requiring high pressure injection (Ref
2-10). It would not be useful for Navy MUSE units. Cooper Bessemer (Cases 2 through 5)
conducted tests that started with the objective of improving the operation of large spark-ignited
natural gas engines and resulted in major improvements in their dual fuel engines as well (see
Fig. 2-2, Ref 2-8). The Cooper Bessemer engines were all four-stroke units. Fairbanks-Morse
(Ref 2-9) achieved results similar to those of Cooper Bessemer with a 900-rpm, two-stroke,
2-3

opposed-piston engine. A plan view of the injector arrangement for the natural gas, diesel, and
pre-chamber injectors of the Fairbanks Morse application is shown on Fig. 2-3. When the engine
is operating in the diesel mode, both diesel injectors are used. In the natural gas mode, a single,
low-pressure gas injector is used along with a diesel-fired pre-ignition chamber. Figure 2-4
shows Fairbanks-Morse data for four modes of engine operation: standard diesel, standard dual
fuel, spark ignition, and Enviro Design pre-chamber. The numbers within each region of the
figure refer to the number of data points used to define that region, NO
x
emissions and thermal
efficiency (brake-specific fuel consumption) are significantly superior for the Enviro-Design.
Pre-chamber parameters affecting the reliability and effectiveness for achieving ignition,
minimum emissions, and minimum pilot fuel are shown on Figure 2-5 (Ref 2-8).
Combustion pre-chambers can be used with dual fuel engines to:
• Further lower NO
x
, hydrocarbon, CO, and particulate emissions.
• Improve engine thermal efficiency.
• Reduce the required diesel pilot fuel for both two- and four-stroke dual fuel engines.
This technology is now available and could be applied to the EMD-645 to achieve further
important improvements in exhaust emissions and thermal efficiency for dual fuel operation.
Such efforts would require a modest extension of the existing dual fuel technology (low-pressure
injection) already demonstrated on B&N RR locomotives and the Navy MUSE unit. Successful
demonstration of a dual fuel EMD-645 engine with pilot ignition would bring the Navy's
stationary dual fuel units to the level of NO
x
required by the most restrictive emissions
regulations in the country for stationary engines.
2-4

Combustion-
Gases '.
Heat
Extraction
Heat
Extraction
Exhaust
Gases
1500 - 2000 F
V
NH3 —
Injection
(a) Selective non-catalytic reduction (SNCR).
Combustion-
Gases -
Heat
Extraction
<
<
< 600 - 800 F
<
<
x
x
X
NH3 —
Injection
Exhaust
Gases
Catalyst Bed
(b) Selective catalytic reduction (SCR).
Figure 2-1. General features of SCR and SNCR NO
x
reduction processes.
2-5

6.0
I
4
'
0
'
2.0
LO-
CO
Standard dual fuel
. Pre-chamber spark ignition
Dual fuel pilot \
+
+ ■
6100 6300 6500 6700
BSFC
Figure 2-2. Results of tests conducted by Cooper Bessemer showing NO
x
■ P re-Chamber
with Pilot
Injector
•Diesel
Injection
Nozzle
■ Diesel Injection
Nozzle
■ Gas Valve
Figure 2-3. Plan view of arrangement of natural gas, diesel, and pre-chamber
injection for Fairbanks-Morse application.
2-6

10. -
8.
_ fl H Standard diesel
„
g/hph
at
_ ■■ Standard dual fuel
° A
z 4.
—
Enviro-Design
\
2.
6.
^B ^ ■ H Spark
1 ^^^ l^^^l
1 1 1 1 1
0 6.2 6.4 6.6 6.8 7.0
BFSC
Figure 2-4. Results of tests conducted by Fairbanks-Morse showing NO
x
emissions and brake specific fuel consumption (Btu/hph).
Diesel injector
Pre-chamber
parameters
Volume
Volume/Orifice area
Orifice shape
Location/Orientation
Quanity pilot fuel
F/A
Pilot fuel injector (1-5%)
N. G. injector
in
M£±
■ Head
. Pre-chamber
Main combustion
chamber
Piston
Figure 2-5. Schematic outline of dual fuel cylinder and parameters
that affect pre-chamber performance.
2-7

Table 2-1. N0
X
Reduction Technologies Related to EMD-645
Case
Mfgr
Number
Strokes
Fuel
Type*
Natural Gas
Injection
Pilot Diesel
NO
x
g/hph
Other
Time
Pressure
%
Location
1
EMD-567
2
DF
Early
200
7
Main
—
Low power,
1982
2
EMD-567
2
DF
Late
3,500
2.0 knock
limit
Main,
side wall
5
Heavy smoke,
reduced
efficiency
SwRI**
3
Detroit
Diesel 92
2
NG
Early
250
N/A
N/A
N/A
Discont'd
production
4
EMD-645
2
DF
Early
100
6
Main
4
Full power
locomotive,
1992-1995
5
EMD-710
2
DF
Late
High
9
Main
?
Results not
available
*DF = Dual Fuel, NG = Natural Gas.
**Southwest Research Institute, Inc.
Table 2-2. Further NO
x
Reductions Using Pre-Ignition Chambers
Mfgr
Number
Strokes
Fuel
Type*
Natural Gas
Injection
Pilot Diesel
NO
x
g/hph
Other
Time
Pressure
(psig)
%
Location
Wartsilla
4
MF
Late
4,000
5
Main
5
New and
Retrofit
Cooper
Bessemer
4
NG
Early
Low
Spark
Pre-
1.5
StdforDF
Cooper
Bessemer
4
DF
Early
Low
5
Main
5
Optimized
Cooper
Bessemer
4
DF
Early
Low
1
Main
4
Unstable
combustion
Cooper
Bessemer
4
DF
Early
Low
0.9
Pre-
1
Full power,
improved
opacity
Fairbanks-Morse
Enviro Op
2
DF
Early
Low
—
Main
4.5 —
Fairbanks-Morse
Enviro Op
2
DF
Early
Low
1.4
Pre-
1
Full power
*MF = Multi-fuel, NG
= Natural Gas, D
F = Dual fi id.
2-8