SRI/USEPA-GHG-QAP-33
August, 2004
Test and Quality Assurance
Plan
EnviroFuels
Diesel Fuel Catalyzer
Fuel Additive
Prepared by:
Greenhouse Gas Technology Center
Southern Research Institute
Under a Cooperative Agreement With
U.S. Environmental Protection Agency
-------
EPA REVIEW NOTICE
This report has been peer and administratively reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
-------
SRI/USEPA-GHG-QAP-33
July, 2004
Greenhouse Gas Technology Center
A U.S. EPA Sponsored Environmental Technology Verification ( ET₯ ) Organization
Test and Quality Assurance Plan
EnviroFuels
Diesel Fuel Catalyzer
Fuel Additive
Prepared By:
Greenhouse Gas Technology Center
Southern Research Institute
PO Box 13825
Research Triangle Park, NC 27709 USA
Telephone: 919/806-3456
Reviewed By:
EnviroFuels, L.P.
Environment Canada Environmental Technology Centre
Burlington Northern and Santa Fe Railway
NYS DEC Bureau of Mobile Sources and Technology Development
Southern Research Institute Quality Assurance
U.S. EPA Office of Research and Development
Kl indicates comments are integrated into Test Plan
-------
(this page intentionally left blank)
-------
Greenhouse Gas Technology Center
A U.S. EPA Sponsored Environmental Technology Verification ( ETw ) Organization
Test and Quality Assurance Plan
EnviroFuels
Diesel Fuel Catalyzer
Fuel Additive
This Test and Quality Assurance Plan has been reviewed and approved by the Greenhouse Gas
Technology Center Project Manager, Quality Assurance Manager, and Center Director, the U.S. EPA
APPCD Project Officer, and the U.S. EPA APPCD Quality Assurance Manager.
Stephen Piccot Date
Center Director
Greenhouse Gas Technology Center
Southern Research Institute
David Kirchgessner
APPCD Project Officer
U.S. EPA
Date
Tim Hansen Date
Project Manager
Greenhouse Gas Technology Center
Southern Research Institute
Robert Wright Date
APPCD Quality Assurance Manager
U.S. EPA
Ashley D. Williamson Date
Quality Assurance Manager
Greenhouse Gas Technology Center
Southern Research Institute
Test Plan Final: (date XX)
-------
(this page intentionally left blank)
-------
TABLE OF CONTENTS
Page
1.0 INTRODUCTION 1-1
1.1. BACKGROUND 1-1
1.2. DIESEL FUEL CATALYZER DESCRIPTION 1-2
1.3. TEST SITE AND LOCOMOTIVE DESCRIPTION 1-2
1.4. PERFORMANCE VERIFICATION PARAMETERS 1-4
1.5. PROJECT ORGANIZATION 1-4
1.6. SCHEDULE 1-6
2.0 VERIFICATION APPROACH 2-1
2.1. TEST DESIGN 2-1
2.2. REVENUE SERVICE BREAK-IN PERIOD AND FUEL TREATMENT 2-3
2.3. INSTRUMENTATION 2-3
2.3.1. Fuel Flow Metering 2-3
2.3.2. Power Metering 2-4
2.3.3. Radiator Cooling Fan Power Consumption 2-4
2.3.4. Emissions Measurements: General 2-5
2.3.5. Dynamic Off-road Emission Sampling (DOES2) Dilution Sampling
System 2-6
2.3.6. SO2 and Sulfate Sampling 2-7
2.3.7. Particulate Emission Sampling: Locomotive Particulate Sampling System 2-7
2.3.8. Opacity Measurements 2-8
2.4. ANALYTICAL METHODS AND NUMBER OF TEST RUNS 2-9
2.4.1. Engine Brake Horsepower 2-9
2.4.2. Emissions and Fuel Consumption 2-9
2.4.3. Baseline Versus Treated Fuel Statistical Significance 2-12
2.4.4. Sample Variance Similarity 2-12
2.4.5. Baseline Versus Treated Fuel Confidence Interval 2-13
2.5. COMPARISON WITH THE FTP 2-13
3.0 DATA QUALITY 3-1
3.1. DATA QUALITY OBJECTIVE 3-1
3.2. INSTRUMENT SPECIFICATIONS, CALIBRATIONS, AND QA/QC CHECKS 3-1
3.3. INSTRUMENT TESTING, INSPECTION, AND MAINTENANCE 3-5
3.4. INSPECTION AND ACCEPTANCE OF SUPPLIES AND CONSUMABLES 3-5
4.0 DATA ACQUISITION, VALIDATION, AND REPORTING 4-1
4.1. DATA ACQUISITION AND DOCUMENTATION 4-1
4.
4.
4.
4.
.1. Fuel Consumption and Power Data 4-1
.2. Emissions Data 4-1
.3. Locomotive Documentation 4-2
.4. QA/QC Documentation 4-2
4. .5. Field Test Documentation 4-2
4. .6. Corrective Action and Assessment Reports 4-2
4.2. DATA REVIEW, VALIDATION, AND VERIFICATION 4-3
4.3. DATA QUALITY OBJECTIVES RECONCILIATION 4-3
4.4. ASSESSMENTS AND RESPONSE ACTIONS 4-4
4.4.1. Project Reviews 4-4
4.4.2. Performance Evaluation Audit 4-4
-------
4.4.3. Technical Systems Audit 4-4
4.4.4. Audit of Data Quality 4-5
4.5. VERIFICATION REPORT AND STATEMENT 4-5
4.6. TRAINING AND QUALIFICATIONS 4-6
4.7. HEALTH AND SAFETY REQUIREMENTS 4-6
5.0 REFERENCES 5-1
APPENDICES
Page
Appendix Al: Sample Fuel Analysis Al
AppendixBl: ERMD Daily Testing Checklist Bl
Appendix B2: DOES2 Locomotive Sampling System Preparation and Emission Testing B2
Appendix B3: Locomotive Particulate Sampler B3
Appendix B4: Chain of Custody Form B4
Appendix B5: In-House Pre-testing Calibration Checklist B5
Appendix B6: Locomotive Test System Calibrations B7
Appendix Cl: GHG Center Field Data Cl
Appendix C2: ETC Field Data C2
Appendix C3: Results Summary C3
Appendix C4: Statistical Tests C4
Appendix C5: Method 2 Pitot Traverse Data C5
Appendix C6: Method 2 Traverse and FTP Exhaust Gas Volume Comparisons C6
Appendix C7: Cooling Fan Power Consumption C7
Appendix C8: Catalyzer and Fuel Ticket C8
LIST OF FIGURES
Page
Figure 1-1 EMD Model GP40-3 Locomotive 1-3
Figure 1-2 Project Organization 1-5
Figure 2-1 Exhaust Duct and Test Duct Schematic 2-5
Figure 2-2 DOES2 Schematic 2-6
Figure 2-3 LPSS Schematic 2-8
LIST OF TABLES
Page
Table 2-1 Locomotive Test Sequence 2-2
Table 2-2 FTP Duty-cycle Weighting Factors 2-10
Table 2-3 Selected T-distribution Values 2-12
Table 2-4 Selected FQ.OS Distribution Values 2-13
Table 3-1 Instrument and Accuracy Specifications 3-2
Table 3-2 Calibration Schedule 3-3
Table 3-3 System QA/QC Checks 3-4
-------
DISTRIBUTION LIST
U.S. EPA
David Kirchgessner
Robert Wright
Southern Research Institute
Stephen Piccot
Tim Hansen
Robert G. Richards
Ashley Williamson
EnviroFuels
Mark Lay
Genesee & Wyoming Railroad
Carl Belke
George King II
Environment Canada ETC
Fred Hendren
Greg Rideout
in
-------
List of Acronyms and Abbreviations
A
AC
ADQ
APPCD
CAR
CFR
CH4
CO
C02
cov
DC
DOE
DOES2
DQI
DQO
DVM
EPA-ORD
ETC
ETV
°F
FS
FTP
g/bhp-h
gph
GHG
HFID
hp
Hz
ID
kVA
kW
Ib/bhp-h
1pm
LPSS
mph
NDIR
NIST
NMHC
NO2
NOX
PEA
PLC
psia
QA
QA/QC
QMP
rpm
scfm
SOP
THC
TPM
V
ampere
alternating current
audit of data quality
Air Pollution Prevention and Control Division
corrective action request
Code of Federal Regulations
methane
carbon monoxide
carbon dioxide
coefficient of variation
direct current
Department of Energy
dynamic off-road emissions sampling system
data quality indicator
data quality objective
digital voltmeter
Environmental Protection Agency Office of Research and Development
Environmental Technology Centre
Environmental Technology Verification
degrees Fahrenheit
full scale
Federal Test Procedure
grams per brake horsepower-hour
gallons per hour
greenhouse gas
heated flame ionization detector
horsepower
Hertz
inner diameter
kilovolt-ampere (reactive power)
kilowatt (real power)
pounds per brake horsepower-hour
liters per minute
locomotive paniculate sampling system
miles per hour
non-dispersive infrared
National Institute of Standards and Technology
non-methane hydrocarbons
nitrogen dioxide
blend of NO, NO2, and other oxides of nitrogen
performance evaluation audit
programmable logic controller
pounds per square inch, absolute
quality assurance
quality assurance / quality control
Quality Management Plan
revolutions per minute
standard cubic feet per minute
standard operating procedure
total hydrocarbons (as carbon)
total paniculate matter
volt
IV
-------
1.0 INTRODUCTION
1.1. BACKGROUND
The U.S. Environmental Protection Agency's Office of Research and Development (EPA-ORD) operates
the Environmental Technology Verification (ETV) program to facilitate the deployment of innovative
technologies. The program's goal is to further environmental protection by accelerating the acceptance
and use of these technologies. Primary ETV activities are independent performance verification and
information dissemination. Congress funds ETV in response to the belief that many viable environmental
technologies exist that are not being used for the lack of credible third-party performance data. With
performance data developed under this program, technology buyers, financiers, and permitters will be
better equipped to make informed decisions regarding new technology purchases and use.
The Greenhouse Gas Technology Center (GHG Center) is one of several ETV organizations. EPA's ETV
partner, Southern Research Institute (Southern), manages the GHG Center. The GHG Center conducts
independent verification of promising GHG mitigation and monitoring technologies. It develops
verification Test and Quality Assurance Plans (test plans), conducts field tests, collects and interprets field
and other data, obtains independent peer-review input, reports findings, and publicizes verifications
through numerous outreach efforts. The GHG Center conducts verifications according to the externally
reviewed test plans and recognized quality assurance / quality control (QA/QC) protocols.
Volunteer stakeholder groups guide the GHG Center's ETV activities. These stakeholders advise on
appropriate technologies for testing, help disseminate results, and review test plans and reports. National
and international environmental policy, technology, and regulatory experts participate in the GHG
Center's Executive Stakeholder Group. The group includes industry trade organizations, environmental
technology finance groups, governmental organizations, and other interested parties. Industry-specific
stakeholders provide testing strategy guidance within their expertise and peer-review key documents
prepared by the GHG Center.
GHG Center stakeholders are particularly interested in transportation technologies with the potential to
increase fuel economy and reduce GHG and criteria pollutant emissions. The Department of Energy
(DOE) reports that transportation CO2 emissions were 32 percent of the total from all sectors during 2002
[1]. Railroad locomotives represent a significant fraction of the total. In 2002, railroads used
approximately 8.7 percent of all petroleum distillate fuels in the transportation sector, or about 1.8 percent
of all fuels consumed in the US. In 2000, railroad fuel consumption was about 3.071 x 109 gallons of
diesel fuel [2]. Even incremental fuel efficiency or emission rate improvements would have a significant
beneficial impact on nationwide air quality and railroad economics. Each 1 percent diesel fuel
consumption reduction would reduce CO2 emissions and fuel costs approximately 1 percent.
EnviroFuels, L.P. manufactures a diesel fuel additive and markets it to heavy-duty vehicle, off-road diesel
engine, and railroad locomotive operators as the Diesel Fuel Catalyzer (catalyzer). The catalyzer is a
suitable verification candidate considering its potential environmental benefits and ETV stakeholder
interest. Based on in-house testing on heavy-duty diesel vehicles, EnviroFuels claims that proper use of
the catalyzer can reduce:
fuel consumption (and corresponding CO2 emissions) by 5 percent
NOX emissions by 12 to 18 percent
unburned total hydrocarbon (THC) emissions up to 30 percent.
1-1
-------
The GHG Center plans to verify the potential fuel consumption and pollutant emission improvements
attributable to the catalyzer in a 3000 horsepower (hp) line-haul locomotive representative of much of the
nationwide roster. This test plan specifies catalyzer verification performance parameters and the rationale
for their selection. It contains the verification approach, data quality objectives (DQOs), and the relevant
QA/QC procedures. The test plan will guide test implementation, document creation, data analysis,
interpretation, and reporting.
The technology developer, testing subcontractor, expert peer reviewers, and the EPA-ORD QA team have
reviewed this test plan. Once approved, as evidenced by the signature sheet at the front of this document,
it will meet the GHG Center's Quality Management Plan (QMP) requirements. The GHG Center will
post the final test plan on their internet site at www.sri-rtp.com and the ETV program site at
www. epa. gov/etv.
The GHG Center will prepare an Environmental Technology Verification Report and Verification
Statement (report) upon field test completion. The same organizations listed above will review the report.
When the reviews and responses are complete, the GHG Center Director and the EPA-ORD Laboratory
Director will sign the Verification Statement, and the GHG Center will post the final documents as
described above.
1.2. DIESEL FUEL CATALYZER DESCRIPTION
EnviroFuels literature states that the key to the catalyzer's performance is a chemical reaction that creates
inorganic polymer complexes of phosphorus and nitrogen on the surface of ferrous and non-ferrous
metals. The formulators add the proprietary compound to refined mineral oil which, in turn, users
administer to diesel fuel at 1 part additive to 1280 parts fuel ratio during normal operation. An initial
dosing rate of 640:1 is utilized in most locomotive applications for a typically 6-8 week break in period.
The complexes, according to Envirofuels statements, smooth and passivate the metal surface, improve
reflectivity (or emissivity), and reduce oxygen reactivity. EnviroFuels states that the reduced oxygen
reactivity reduces NOX formation while the improved emissivity enhances combustion through reduced
radiative losses from the flame front. This, combined with improved lubricity, reduces fuel consumption.
EnviroFuels' research indicates that six to eight weeks of regular service are required from the initial fuel
treatment for the performance improvements to be fully realized in locomotive service. After that, the
fuel must be treated on an ongoing basis to maintain the effects.
1.3. TEST SITE AND LOCOMOTIVE DESCRIPTION
Locomotive testing methods differ from those involving other diesel-powered transportation modes, as
they are adapted to the locomotive's design and operating features. Locomotives are powered by an
engine through an electric alternator to electric motors that are connected to the drive wheels. This differs
significantly from road vehicles, where the relationship between engine revolutions per minute (rpm) and
vehicle miles per hour (mph) is mechanically dictated by the transmission and final drive gear ratios. A
locomotive engine is operated at a desired power output and corresponding engine rpm without being
constrained by locomotive speed because of the electrical coupling between the engine and drive wheels.
Power settings for railroad engines, or throttle position, generally include eight discrete positions or
notches on the throttle gate in addition to idle and dynamic brake. Each notch is numerically identified,
with notch one being the lowest power setting (other than idle) and notch eight being maximum power.
Each throttle notch corresponds to a discrete fuel delivery system setting. The engine can operate at only
1-2
-------
eight distinct combinations of fueling rate, power output and engine speed (in addition to idle and
dynamic brake). In the dynamic braking position, the traction motors act as generators, with the generated
power being dissipated as heat through an electric resistance grid. In dynamic braking mode, the engine
generates only enough power to operate the locomotive accessories and the resistance grid cooling fans.
EPA considered these design features while developing the Federal Test Procedure (FTP) for emission
measurements from diesel locomotives in Title 40 Code of Federal Regulations (CFR) 92, Subpart B [3].
The FTP is a steady-state test procedure as compared to the transient test procedures previously
established for on-highway heavy-duty diesel engines. This verification will employ test methods derived
directly from the FTP.
The GHG Center plans to verify the potential fuel consumption and pollutant emission improvements
attributable to the catalyzer in a 3000 horsepower model GP40-3 line-haul locomotive built by General
Motors' EMD division. Genesee and Wyoming's (G&W) St. Lawrence and Atlantic Railroad operates the
locomotive out of Auburn, ME. The GP40, shown in Figure 1-1, is a variant of the SD40 series, the most
common model pre-1990 line-haul locomotive in current use in the U.S. The unit to be tested has an
EMD 16-645E3 16-cylinder engine. It was built in 1980 and remanufactured to Tier 0 requirements
(§92.8(a)(l)(i))in2003.
Figure 1-1. EMD Model GP-40-3 Locomotive
1-3
-------
1.4. PERFORMANCE VERIFICATION PARAMETERS
The Envirofuels Diesel Fuel Catalyzer performance verification parameters are fuel economy, pollutant,
and GHG emissions changes due to catalyzer use in the test locomotive. Test personnel will measure the
parameters in successive tests using the 40 CFR 92 Subpart B test sequence, as described further below.
Changes in fuel economy and emissions will be calculated both on a "per notch" basis and as a weighted
average using the FTP's line-haul weightings. Reported parameters will consist of:
brake-specific fuel consumption rate change, ABSFCj, for each notch, pounds per brake
horsepower hour (Ib/bhp-h)
line-haul weighted brake-specific fuel consumption rate change, ABSFCDc, Ib/bhp-h
brake-specific mass emission rate change for each notch of each emitted pollutant or
GHG species, AEy, grams per brake horsepower hour (g/bhp-h)
line-haul weighted brake-specific mass emission rate change for each emitted pollutant
or GHG species, AEiDC, g/bhp-h
Pollutants and GHGs of concern are:
CO2
CO
NOX
total non-methane hydrocarbons (NMHC)
methane (CFy
total hydrocarbons (THC)
total particulate matter (TPM)
smoke opacity
The standard locomotive emissions testing procedure is the Federal Test Procedure (FTP) described in 40
CFR 92 Subpart B. Test results will be reported for each notch and as weighted for the locomotive's duty
cycle. While there is no comparable standard for fuel consumption measurement, the FTP does require
precise measurement of this quantity simultaneously with the emission measurements. This verification
will therefore report fuel consumption changes for each notch and duty cycle-weighted fuel consumption
changes. Duty cycle-weighted fuel consumption changes are useful because the FTP duty cycle
weightings are considered as reasonably representative locomotive use patterns. The change in duty
cycle-weighted fuel consumption should be a valid predictor of expected fuel cost savings.
The verification parameters presented here include only the line-haul duty cycle weighting since
locomotives of the size and configuration tested here would not generally be used in switching
applications. The switch cycle weighted emissions and fuel consumption will also be calculated and
reported for reference purposes, but they are not considered verification parameters.
1.5. PROJECT ORGANIZATION
Figure 1-2 presents the project organization chart.
1-4
-------
U.S. EPA
APPCD Project Officer
David Kirchgessner
Southern Research Institute
ETV GHG Center Director
Stephen Piccot
U.S. EPA
APPCD Quality Assurance Manager
Robert Wright
Southern Research Institute
Quality Assurance Manager
Ashley Williamson
GHG Center
Project Manager
Tim Hansen
GHG Center
Field Team Leader
Bob Richards
Testing Contractor:
Environment Canada
Emissions Technology Centre
Figure 1-2. Project Organization
The GHG Center has overall verification planning and implementation responsibility. The GHG Center
will coordinate all participants' activities; develop, monitor, and manage schedules; and ensure the
acquisition and reporting of data consistent with the strategies in this test plan. The GHG Center
Director, Mr. Stephen Piccot, will:
review the test plan and report for consistency with ETV operating principles
allocate appropriate resources for the verification
oversee GHG Center staff activities
Mr. Mark Lay of Envirofuels is the technology developer's primary point of contact. He will:
review the test plan and report especially with respect to accuracy in the technology
description and its application
secure the involvement of the locomotive owner, Genesee and Wyoming, Incorporated
(G&W), locomotive operating and maintenance personnel, and facilities where testing
will occur
provide a sufficient supply of the catalyzer to the locomotive owner for the six to eight
week break-in period and for the final test period
The GHG Center project manager is Mr. Tim Hansen. His responsibilities include:
drafting the test plan, report, and Verification Statement
overseeing the field team leader's activities
ensuring collection of high-quality data and that all DQOs are met
maintaining communications with all test participants
budgetary and scheduling review
The project manager will have authority to suspend testing for health and safety reasons and if the
QA/QC goals presented in Section 3.0 are not being met.
The field team leader is Mr. Robert Richards, who will supervise all field operations and the testing
contractor's activities. He will assess data quality and will have the authority to repeat tests as deemed
necessary to ensure achievement of data quality goals. He will:
1-5
-------
coordinate the installation of the electrical, and fuel metering equipment on the
locomotive with the owner
operate the electrical and fuel metering equipment during the tests
declare the beginning and end of each test run, with input from the testing contractor
collect interim test data for use in consultations with the project manager
supervise and coordinate subcontractor activities
perform other QA/QC procedures as described in Section 3.0
At the completion of each test run, the field team leader will communicate test results to the project
manager. The field team leader and project manager will collaborate on all major project decisions
including the need for further test runs or corrective actions.
The GHG Center QA manager, Dr. Ashley Williamson, will review this test plan. He will also review the
verification test results, report, and conduct the Audit of Data Quality (ADQ) described in Section 4.0.
The QA manager will report all internal audit and corrective action results directly to the GHG Center
Director who will provide copies to the project manager for inclusion in the report.
Environment Canada's Environmental Technology Centre (ETC) will act as the testing contractor. Their
responsibilities include:
coordination and installation of the temporary test duct, emissions testing, ambient
monitoring and other necessary equipment on the locomotive
performance of each test run in accordance with specifications of this test plan and
reference methods
reporting interim test data to the field team leader at the end of each test run
meeting the Quality requirements specified in this test plan and their subcontract SOW
analysis of all emissions test data and submittal of a test report within three weeks of the
end of the final test series
EPA-ORD will provide oversight and QA support for this verification. The Air Pollution Prevention and
Control Division (APPCD) project officer, Dr. David Kirchgessner, and QA manager, Mr. Robert Wright,
will review and approve the test plan and report to ensure that they meet EPA QA goals and represent
sound scientific principles. Dr. Kirchgessner will be responsible for obtaining final test plan and report
approvals.
1.6. SCHEDULE
The tentative schedule of activities for the catalyzer verification test is:
Verification Test Plan Milestones Dates
GHG Center internal draft development April 26 - July 9, 2004
EnviroFuels review July 9 - July 12, 2004
Industry peer review and plan revision July 14 - July 21, 2004
EPA review July 23 - August 6, 2004
Final test plan posted August 9, 2004
Verification Testing and Analysis Milestones Dates
Initial tests on untreated fuel August 16 - August 20, 2004
Revenue service break-in period on catalyzer-treated fuel August 20 - October 3,2004
Final tests on treated fuel October 4 - October 8, 2004
1-6
-------
Verification Report Milestones
GHG Center internal draft development
EnviroFuels review
Industry peer review and report revision
EPA review
Final report posted
Dates
November 1 - November 22,2004
November 22 - November 29, 2004
November 29 - December 13, 2004
December 13 - December 31, 2004
January 14, 2005
1-7
-------
(this page intentionally left blank)
-------
2.0 VERIFICATION APPROACH
This section describes the GHG Center's verification approach The following subsections describe the
test design, FTP details as applied here, and sampling and analytical methods. A final subsection
summarizes specific deviations from the FTP in the planned measurement series.
The FTP forms the basis for this test plan. The step-by-step procedures and tables of measurements
supplied in Appendices B and C, as referenced to §92.101 through §92.133 of the FTP, and ETC's Onsite
Collection Procedures for Chemical Analysis (for TPM, SO2, and sulfates; document number
11.23/1.2/S), will be the standard operating procedures (SOP).
2.1. TEST DESIGN
The GHG Center will first determine the locomotive's fuel consumption and emissions while operating
on untreated fuel. The mean of at least three and not more than six FTP test runs will serve as a baseline
for comparison. After the baseline test runs, railroad personnel will administer the fuel catalyzer to the
locomotive's fuel and return it to revenue service for a six to eight-week break-in period. GHG Center
personnel will then perform a final test series of at least three and not more than six test runs with the
catalyzer-treated fuel. During the break-in period, G&W personnel will record fuel and additive usage.
In order to assure that the test results are not biased by fuel properties other than Catalyzer use, the
supplier will provide fuel for both test series from a single fuel lot. Appendix A-l provides a sample
analysis of a recent fuel lot delivered to G&W. The analysis shows that this fuel is within §92.113 diesel
fuel specifications. The GHG Center will obtain a similar analysis for the test fuel actually used in this
verification and include it in the report. During testing, the locomotive will be fueled from its belly tank,
which will be drained, cleaned, and filled with fuel from the test lot prior to testing.
Each test run will consist of a set of emissions measurements while the engine operates at a series of
steady-state speed and load conditions. At each steady-state operating mode, measurements are taken of
CO2, CO, NOX, O2, THC, CH4, TPM, and smoke emissions, generated power, and fuel consumption rate.
Emissions are normalized to engine power in terms of brake horsepower-hour (bhp-h), which will be
measured electrically. Test personnel will also measure SO2 and sulfate emissions at notch 8 for
information only. These results are not considered part of this verification but will allow an assessment of
the Catalyzer's effects on the emissions partitioning of fuel-borne sulfur.
The test sequence is defined in terms of the throttle notches typically available on diesel-electric
locomotives. At the beginning of the sequence, the operator brings the engine to normal operating
temperature in accordance with the manufacturer's warm-up procedures for in-service locomotives.
Warmup will require approximately 1 hour (longer in cold weather). The engine is then operated at notch
8 (full power) for 5 minutes. The operator returns the engine to idle, or low idle if so equipped. Test
personnel then begin exhaust emission, fuel consumption, and other measurements. During each test
period, an external resistance load bank is connected to the locomotive's power distribution system to
dissipate the generated electricity. Test measurements continue while the operator cycles the locomotive
(and load bank) through each power setting to notch 8. Table 2-1 shows the test sequence and elapsed
time at each notch, as excerpted from Table B124-1 of §92.124.
2-1
-------
Table 2-1. Locomotive Test Sequence
Mode No.
Warmup
Warmup
la
1
2
3
4
5
6
7
8
9
10
Notch setting
Notch 8
Lowest Idle
Low Idle
Normal Idle
Dynamic Brake
Notch 1
Notch 2
Notch 3
Notch 4
Notch 5
Notch 6
Notch 7
Notch 8
Time in notch
5 ± 1 min
15 min maximum
6 min minimum
15 min minimum
Emissions,
power, and
fuel
consumption
measured
No
All
Particulate sampling will begin within ten seconds and end six minutes after the start of each test mode.
ETC will sample gaseous concentrations continuously and will calculate steady state concentrations per
the FTP as a one-minute average beginning after 300 seconds (840 seconds for notch 8) from the start of
each test mode. Sufficient time is allotted after each notch test to change particulate filters and perform
any other necessary activities.
If the FTP criteria for steady state concentrations are not fulfilled, the FTP requires an integrated
concentration calculation over the mode's minimum duration (6 or 15 minutes). ETC will correlate the
resulting concentrations with fuel consumption and power generation measurements during the last
minute of each FTP test mode (or last 3 minutes for idle modes). ETC will analyze and report the
gaseous and particulate emissions in conformance to these specifications. This will allow direct
comparisons with FTP results from other locomotive tests.
During each notch, test personnel will acquire the following measurements:
fuel consumption, gallons per hour (gph)
AR10 main generator power output, kilowatts (kW) and engine mechanical power, brake
horsepower (bhp)
exhaust emissions, grams per brake horsepower-hour (g/bhp-h)
sample system operating parameters
engine inlet and cooling air temperature, degrees Fahrenheit (°F)
ambient barometric pressure, pounds per square inch absolute (psia)
ambient temperature, °F
cooling fan(s) operating status, on or off
At the end of the revenue service break-in period, test personnel will repeat this sequence while operating
the locomotive on treated fuel.
2-2
-------
The field team leader will acquire the power consumption in kW for each cooling fan at each notch during
testing. These values, divided by the companion alternator's efficiency, contribute to the accurate
determination of engine bhp. During testing, additional parasitic loads (air conditioning, lights, radio,
etc.) will be shut off.
2.2. REVENUE SERVICE BREAK-IN PERIOD AND FUEL TREATMENT
At the conclusion of the baseline test runs, the field team leader will release the locomotive for a 6 to 8
week break-in period in regular revenue service during which time it is anticipated the locomotive will
consume approximately 30,000 gallons of fuel. If railroad operational issues prevent use of this much
fuel, the project manager may extend the break-in period. Envirofuels will supply a calibrated dosing
pump which will enable G&W personnel to administer the Fuel Catalyzer during routine locomotive
refueling operations.
The initial treated fuel dosing ratio during the break-in period will be 640:1. The dosing rate will be
reduced to the maintenance dosage rate of 1280:1 for a minimum of one week prior to the final test
period. The dosing pump has a totalizing readout which, when correlated with the fueling station's
records, will allow verification that the additive was properly mixed with the fuel. Appendix C8 provides
the procedure and a log form.
During the break-in period and final testing, the locomotive will not be scheduled for any maintenance
activities to ensure that no modifications are made to the engine that may affect its performance prior to
the final test period.
2.3. INSTRUMENTATION
2.3.1. Fuel Flow Metering
Fuel consumption during both test series will be measured using flowmeters supplied by the GHG Center
and installed by railroad maintenance staff under the field team leader's supervision. Fuel supply flow in
the typical EMD locomotive engine is approximately 2.5 times the anticipated net consumption rate at
notch 8, with the balance recirculated to the fuel tank. The excess flow provides injector and fuel system
lubrication and cooling. The fuel consumption measurement system therefore consists of two
temperature-compensated flowmeters, one each installed in the fuel supply and return pipelines.
This verification will employ a Flow Technologies, Inc. FuelCom model FC05 net fuel metering system.
The FuelCom flow meters use two rotating impellers driven by the flowing fuel. Magnets imbedded in the
impellers generate a pulsed output signal. Each pulse represents a known volume of fuel that is captured
between the impeller lobes. The system includes fuel temperature sensors. A microprocessor in the
FuelCom FC900 transmitter uses the pulse and temperature data to compensate for the temperature effects
on the fuel viscosity and density, thereby providing a linearized output signal.
The manufacturer specifies accuracy of ± 1.0 percent at higher fuel consumption rates (notches 2 through
8) and ± 2.0 percent at lower fuel consumption rates (idle and notch 1) of the net fuel consumption
reading. The flowmeter will be accompanied by a current (18-month) National Institutes of Standards
and Technology (NIST) - traceable calibration certificate. The manufacturer will calibrate the flowmeter
with diesel fuel at 84 data points from 50° F to 140° F over the full flow range.
2-3
-------
A datalogger will poll the sensors' 4-20 milliamp (mA) outputs once per second and compute and record
five-second temperature-compensated inlet, return, and net flow averages during each test run. A
separate datalogger and software package will monitor and record the sensors' RS-485 fuel temperature
outputs.
2.3.2. Power Metering
Brake-specific emissions calculations require the engine's total horsepower production at each notch.
This is the sum of the mechanical power input to locomotive accessories and the main "AR10" generator.
The locomotive service manual cites default power consumption values for the following accessories [8]:
main air blower, 13.5 hp
traction motor blowers, 75.0 hp
air compressor (unloaded), 17.0 hp
auxiliary direct-current (DC) generator, 4.0 hp
Test personnel will record actual power consumption at the radiator cooling fans as described in the
following subsection.
Engine mechanical power input to the main generator in bhp is the ARlO's real power output in DC kW
divided by the generation efficiency (default value of 0.715). AR10 power output is the product of the
current and voltage produced at each notch.
The current sensor will be a split-core Hall effect device connected to a signal conditioner. Rated
accuracy is ± 1.0 percent of reading; span is 4000 amperes (A). The voltage transducer rated accuracy is
± 0.25 percent full scale (FS); span is 1000 volts (V). The combined accuracy will be approximately ±
1.0 percent, which meets FTP power metering specifications. Both sensors will be accompanied by
current (18-month) NIST-traceable calibration certificates.
A datalogger will poll the sensors' 4-20 milliamp (mA) outputs once per second and compute and record
five-second averages during each test run.
An external resistance load bank will dissipate the power produced by the AR10 generator. Maintenance
personnel will connect the load bank to the proper test point in the locomotive's electrical distribution bus
according to the locomotive's standard load test procedure. Maintenance personnel will install the current
sensor around the two positive output cables at the AR10 terminal block and they will connect the voltage
sensor leads to the AR10 positive and negative output terminals or buses.
2.3.3. Radiator Cooling Fan Power Consumption
The field team leader will measure each cooling fan's power consumption as kW with a clamp-on true
power digital voltmeter (DVM) prior to the test campaign. This value at each notch divided by the D14
companion alternator efficiency yields the mechanical power in bhp required from the engine. Appendix
C7 provides a log form.
During testing, the field team leader will log the power consumption and on/off cycles for the cooling
fans. Analysts will add the total power consumption of all operating fans to the AR10 power output for
that notch.
2-4
-------
2.3.4. Emissions Measurements: General
Emissions measurements will be conducted by the Emissions Research and Measurements Division
(ERMD) of Environment Canada's Environmental Technology Centre (ETC). ETC will use three
sampling modules which generally meet 40 CFR 92.114 specifications: an extractive opacity monitor and
portable partial flow dilution sampling systems for gaseous and particulate emissions. FTP §92.114(d)(2)
permits sampling after dilution of part or all of the exhaust. ETC personnel will install a temporary test
duct onto the locomotive's single exhaust duct with the three sampling modules connected to the test
duct. Figure 2-1 illustrates the engine exhaust and provides a test duct schematic. The additional sample
ports shown in Figure 2-1 will allow ETC to perform the exhaust gas flow rate cross checks by pitot
traverse outlined in Table 3-4.
Ll
D -
Equivalent diameter is '' Ll+L2 ,
or 16.1". Approximately two diameters exist
below roof hatch.
Specify test duct Ll, L2, and egg crate clearance
to fit over upper lip of exhaust duct.
Sample ports not to scale. Specify number and
size to fit DOES2, LPSS, and smoke meter probes/
For pitot traverses, specify number, size, and spacing
per EPA method 1
L2
r- --i
\
c
r.
i -^
LJ U U LJ
U
Approx. 3/4 " horizontal clearance between roof
hatch opening and duct lip; exhaust duct lip extends
approximately 1 1/2" vertically above mounting flange
!=\ f\
""<".'.'" ..,--;......<.
Test Duct
32 1/2'
Egg-crate flow straightener
A
l.OxDeq
V
A
2.0 x Deq
Edge of Roof
Hatch
Exhaust Duct
Mounting Flange
Figure 2-1. Exhaust Duct and Test Duct Schematic
2-5
-------
2.3.5. Dynamic Off-road Emission Sampling (DOES2) Dilution Sampling System
The Dynamic Off-road Emission Sampling System (DOES2) is a partial flow portable sampling system
for gaseous emissions. ETC will configure the system according to 40 CFR 92.114(e) requirements.
The DOES2 collects a known quantity of the engine's raw exhaust and mixes it with a known quantity of
ambient dilution air. This dilution, while maintaining constant temperature and flow velocity, conditions
the sample and minimizes condensation. Sample pumps extract a secondary sample from the diluted
stream and route it to the gas analyzers. The analyzers and their detection principles will be:
THC by heated flame ionization detector (HFID)
CH4 by HFID
NOX by chemiluminescence detector
CO by non-dispersive infrared (NDIR)
CO2 by NDIR
O2 by paramagnetic detector or equivalent
The sample probe is fabricated from 3/8 inch stainless steel tubing, capped and drilled per Figure Bl 14-2
of §92.114(b)(3). A 25' heated line connects the sample probe to the DOES2 dilution tunnel. Figure 2-2
represents a schematic flow diagram of the dilution tunnel.
The dilution pump forces a controlled volume of air through a pre-filter into the dilution tunnel. That
volume combines with the exhaust gas drawn from the sample probe to produce the dilute sample. The
main pumps draw the dilute sample through the dilution tunnel and past the instrumental analyzer probes.
The dilute sample is thoroughly mixed as it travels more than 10 tunnel diameters before reaching the
instrumental analyzer sample probes. Sample pumps, two mass flow controllers, and a heated line convey
the dilute sample to the analyzers. Operators monitor the dilute sample temperature to ensure that no
moisture condensation occurs.
Heated Line
(exhaust)
Tunnel
Dilution
Pumps
Dilution
MFC
MFC Sulphur
Filter
I
Ambient
Heated Line
(to Inst. Bench)
Sample Tef,on
Pump R|ter
Sample
MFC W\ Filter
Vent
Tunnel
Temperature
Figure 2-2. DOES-2 Schematic
2-6
-------
A Programmable Logic Controller (PLC) will read and record all sensor signals at 0.2 second intervals,
calculate the dilution air requirement, control the variable flow solenoid valve and calculate the emission
rates. The PLC uses a Compact Flash Card as storage media. Test personnel will copy the data to floppy
disks as ASCII text files or Excel worksheets for the field team leader's review.
2.3.6. SO2 and Sulfate Sampling
The DOES2 system includes SO2 and sulfate sampling capabilities. Two separate treated filter media
collect gaseous and particulate samples from the diluted gas stream for later extraction by H2O2 for SO2 or
isopropyl alcohol and de-ionized water for sulfates. Analysis is by ion chromatography with a
conductivity detector for both species.
Sampling rate for SO2 and sulfate is 5.0 1pm for the duration of the mode (15 minutes at notch 8). ETC
will install a backup filter downstream of the teflon and coated filters to quantify any potential
breakthrough. ETC personnel will collect the sample filters and ship them to their laboratory under a
signed chain of custody form (Appendix B4) according to their Onsite Collection Procedures for
Chemical Analysis, document number 11.23/1.2/S. The ETC laboratory designates the analyses methods
as Determination of Anions and Cations on Multi (3) - Ion Chromatography System, document number
6.3/5.0/M, and Determination of Gaseous and Particulate Inorganic Air Pollutants by Ion
Chromatography. document number 6.5/1.0/M. The detection limit for sulfate on the teflon filter is 0.2
micrograms per sample. The detection limit for all sulfur species on the coated filter is 0.7 micrograms
per sample.
ETC will collect four SO2 and sulfate samples during notch 8 operations during valid test runs for each
fuel condition. Three of the samples will be analyzed with the fourth held in reserve for later analysis if
warranted. These measurements will be undertaken and reported for information only.
2.3.7. Particulate Emission Sampling: Locomotive Particulate Sampling System
ETC developed the Locomotive Particulate Sampling System (LPSS) to obtain particulate samples from
large engine exhausts. It is similar to the DOES2 but withdraws and dilutes larger amounts of exhaust
gas. The higher throughput provides larger amounts of particulate for sampling. The probe is designed
according to Figure B114-4 of §92.114(c)(2) to sample at points across the exhaust duct. This yields a
representative spatial average of the exhaust flow. As in the DOES2 system, a 25' heated line connects
the exhaust probe to the LPSS, where the exhaust is diluted with ambient air at a constant rate. Figure 2-3
presents a LPSS schematic flow diagram.
2-7
-------
Exnoust inlet
Temperature
Tunnel
Sample
Pump
-\
Dilution Control
['"'.I I
ion
wroture I
I m \
PoMiculoir
Filler
Dilution
Temper
Sample
Temperature
Heated Line
»nin Blower
Figure 2-3. LPSS Schematic
A hot wire anemometer is used to measure the tunnel flow. This flow is maintained by choking the outlet
of the main blower with a manual valve. The exhaust is drawn through the tunnel through a % inch
heated line connected to the probe in the exhaust duct. The dilution flow is introduced approximately 5
inches downstream from where the exhaust is entered. The dilution flow is generated by a smaller
blower, and is filtered to prevent ambient air particles from entering the tunnel. The dilution air is then
directed through a control splitter which varies the dilution air quantity. Test operators set the dilution
and tunnel flows manually because the LPSS was designed for use in steady-state testing.
A sample pump withdraws a metered quantity of the diluted sample through "back to back" 70mm
Fluorocarbon-coated glass fiber filters to collect TPM. The particulate filters conform to §92.114
specifications.
ETC personnel will collect the sample filters after each mode, seal the containers, and forward them to
their laboratory for gravimetric analysis under a signed chain of custody form (Appendix B4) according
to their Onsite Collection Procedures for Chemical Analysis, document number 11.23/1.2/S.
2.3.8. Opacity Measurements
A Bosch - RT100A opacity meter will measure smoke emissions. This opacity meter draws a sample of
exhaust from the test duct into a viewing chamber where a beam of light is passed through the sample. A
detector continuously measures the quantity of transmitted light and a datalogger records the results 2
times per second.
Locomotives generally produce their heaviest smoke emissions during transitions to higher operating
notches. ETC will therefore record opacity data continuously to measure smoke emissions during these
transients as well as during steady state operations at each notch, as specified in 40 CFR 92.131(b).
2-8
-------
2.4. ANALYTICAL METHODS AND NUMBER OF TEST RUNS
The FTP includes all the method's required calculations, so they are not reproduced here. This subsection
discusses the generalized emissions and fuel consumption calculations and introduces the statistical
methods the field team leader will use to finalize the number of test runs.
In general, more test runs will allow a better characterization of the catalyzer's effects. The GHG Center
will be unable to restore the locomotive to the baseline condition, so the field team leader must estimate
the proper number of baseline runs prior to releasing the unit for the break-in period.
He will evaluate the results after the 3rd baseline test run to estimate the statistical significance of potential
changes. If the baseline test results indicate that the expected 18 percent NOX emissions change and 5
percent fuel consumption change may be statistically significant based on sampling variability seen
during the first 3 runs, he may end the baseline tests. If not, he may call for additional test runs up to a
maximum of 6. A total of 6 baseline runs is the most likely scenario because his evaluation will consider
that particulate results will not be available on site. The field team leader will perform similar analyses
after the 3rd and any subsequent treated fuel test runs. The statistical tests are:
evaluate the statistical significance of any changes
establish that the test results have similar variability between the baseline and treated
fuel
calculate the confidence interval on the changes
The subsections below discuss these tests.
2.4.1. Engine Brake Horsepower
Engine bhp is:
kW
BHP = 1_ + bhp, +13.5 + 75 + 17+4 Eqn. 2-1
J 77e*0.7457 f
Where:
BHPj = mean mechanical power for mode j, bhp
kWARio = mean main generator power output, kW
T|e = AR10 electrical efficiency at mode j (default value is approximately 0.715)
0.7457 = horsepower per kilowatt
bhpfans = cooling fan kW divided by companion alternator efficiency (default 0.85), hp
13.5 = default main air blower mechanical power consumption, hp
75 = default traction motor blower mechanical power consumption, hp
17 = default unloaded air compressor mechanical power consumption, hp
4 = default auxiliary DC generator mechanical power consumption, hp
2.4.2. Emissions and Fuel Consumption
The following equations use the FTP nomenclature where possible. The normalized emission rates and
fuel consumption for each test mode are:
2-9
-------
BHP,
and:
Where:
BSFC; =-
BHP,
Eqn. 2-2
Eqn. 2-3
E;J
= brake -specific mass emission rate of pollutant i for mode j, g/bhp-h
My = mean mass emission rate for pollutant i during mode j test, grams per hour (g/h)
BHPj = mean brake horsepower for mode j, bhp
BSFQ = brake-specific fuel consumption for mode j, Ib/bhp-h
Wfj = mean fuel mass consumption rate for mode j, Ib/h
Measurements taken at each operating mode setting will be weighted according to two duty cycles
assumed typical of line-haul and switching operations. EPA derived these duty cycle weightings from
actual train time-in-notch measurements and considers them as representative. Table 2-2 summarizes the
weighting factors as excerpted from Table B132-1 of §92.132.
Table 2-2. FTP Duty-cycle Weighting Factors
Throttle Notch
Setting
Low idle
Normal idle
Dynamic brake
Notch 1
Notch 2
Notch 3
Notch 4
Notch 5
Notch 6
Notch 7
Notch 8
Fmode , line-haul
0.190"
0.190"
0.125
0.065
0.065
0.052
0.044
0.038
0.039
0.030
0.162
Fmode, SWitch
0.299"
0.299"
0
0.124
0.123
0.058
0.036
0.036
0.015
0.002
0.008
" For locomotives equipped with a single idle notch, the
combined idle Fmode is the sum of the two values shown
The weighted emissions for each duty cycle are:
Eioc =
Eqn. 2-4
Where the summation is over all modes j, and:
EiDC = (line-haul or switch) duty-cycle weighted brake-specific mass emission rate for
pollutant i, g/bhp-h
My = mean mass emission rate for pollutant i during mode j test, g/h
Fj = (line-haul or switch) duty cycle weighting factor for mode j (Table 2-2)
BHPj = mean brake horsepower for mode j, bhp
The weighted fuel consumption will be:
BSFCoc =
Eqn. 2-5
2-10
-------
Where:
BSFCDc = (line-haul or switch) duty-cycle weighted brake specific fuel
consumption, Ib/bhp-h
Wfj = mean fuel mass consumption rate for mode j, Ib/h
FJ = (line-haul or switch) duty cycle weighting factor for mode j (Table 2-2)
BHPj = mean brake horsepower for mode j, bhp
The verification parameters for this test are defined in terms of changes, delta (A), in these quantities after
introduction of the Catalyzer fuel additive. Thus, the brake-specific fuel consumption rate changes at each
operating mode will be:
ABSFG = BSFCibaselme - BSFC^catalyzer Eqn. 2-6
Where:
ABSFQ = brake-specific fuel consumption rate change for mode j, Ib/bhp-h
BSFQ baseime and BSFQ cataiyzer are computed using Equation 2-3.
The change in the duty-cycle weighted brake specific fuel consumption will be:
A BSFCflC = BSFCDCfasdim - BSFCDC^talyzer Eqn. 2-7
Where:
A BSFCDc = change in duty-cycle weighted brake specific fuel consumption, Ib/bhp-h
BSFCDc,baseime and BSFCDc,cataiyzer are computed using Equation 2-5.
The change in mode and duty-cycle weighted emissions will be calculated similarly:
AE = E-, r E t, Eon 2-8
ij ij, baseline ij,catalyzer T.
AE. =E. -E. Eqn. 2-9
IDC lDC,baseline lDC)Catalyzer
Where:
AEy = brake-specific emission rate change for pollutant i, and mode j, g/bhp-h
AEiDC = change in duty-cycle weighted brake-specific emissions rate for pollutant i,
g/bhp-h
Eij.baseiine and E1J)Cataiyzer are computed using Equation 2-2.
EiDc,baseime and EiDC,Cataiyzer are computed using Equation 2-4.
Note that at the low fuel burn rates expected during the lower modes (low idle through about notch 1), the
subtraction error caused by inaccuracies in the two fuel meters may be a large fraction of the burn rate. It
may be impossible to show a statistically significant fuel consumption rate change for those individual
modes. The overall measurement error for the duty-cycle weighted fuel consumption rates, however, will
be small. This is because the duty-cycle weighted fuel consumption rate calculation applies the Table 2-
2 weighting factors to each mode. These weighting factors, combined with the much lower fuel meter
subtraction error in the higher notches, will yield a satisfactory measurement error. For example, the
weighted fuel consumption rate from a pretest visit was 56.35 ± 0.14 gph, for a relative error of 0.25
percent. This is a small fraction of the estimated 5.0 percent fuel consumption change.
2-11
-------
2.4.3. Baseline Versus Treated Fuel Statistical Significance
The GHG Center will evaluate the statistical significance of the emissions and fuel consumption changes
between the baseline and treated fuel conditions. If fuel consumption changes are statistically significant,
the GHG Center will calculate the difference's confidence interval.
After the 3rd test run, and after each following run (up to the 6th), analysts will calculate a test statistic, ttest,
and compare it with the Student's T distribution value with (n] + n2 - 2) degrees of freedom as follows
[6]:
(^ 1 ~~ -A ? ) ( Li , U r.)
ttest = Eqn. 2-10
n,
Eqn. 2-11
Where:
X] = mean fuel economy with baseline fuel
X2 = mean fuel economy with treated fuel
(ii - (i2 = zero (H0 hypothesizes that there is no difference between the population means)
nj = number of repeated test runs with baseline fuel
n2 = number of repeated test runs with treated fuel
Si2 = sample standard deviation with baseline fuel, squared
s22 = sample standard deviation with treated fuel, squared
sp2 = pooled standard deviation, squared
Selected T-distribution values at a 95-percent confidence coefficient (t0 025, DF) appear in the following
table [6].
Table 2-3. Selected T-distribution Values
HI
3
4
5
6
n2
3
4
5
6
Degrees of
Freedom,
DF (n!+n2-
2)
4
6
8
10
to.025, DF
2.776
2.447
2.306
2.228
If ttest > to 025.DF, conclude that the data shows a statistically significant difference between the baseline and
treated fuel parameters. Otherwise, conclude that a significant fuel economy difference does not exist. If
significant, the the difference and its confidence interval will be reported.
2.4.4. Sample Variance Similarity
Use of equations 2-10 and 2-11 requires the assumption that the baseline and treated fuel test run results
have similar variance. The ratio of the sample variances (sample standard deviation squared) between the
two fuel test series is a measure of this similarity [7]. Analysts will calculate an Ftest statistic according to
2-12
-------
Eqn. 2-12 and compare the results to the values in Table 2-4 to determine the degree of similarity between
the sample variances.
S max -r-, ,-» i r*
Ftest= Eqn. 2-12
Where:
Ftest = F-test statistic
s2max = larger of the sample standard deviations, squared
s2mm = smaller of the sample standard deviations, squared
Table 2-4 [6] presents selected F0 Os distribution values for the expected number of test runs and the
acceptable uncertainty (a; 0.05).
Table 2-4. Selected F0.os Distribution Values
s min number of
runs
3
4
5
6
s2max number
of runs
Degrees of
Freedom
2
3
4
5
3
2
19.00
9.55
6.94
5.79
4
3
19.16
9.28
6.59
5.41
5
4
19.25
9.12
6.39
5.19
6
5
19.30
9.01
6.26
5.05
If the F-test statistic is less than the corresponding value in Table 2-4, then analysts will conclude that the
sample variances are substantially the same and the statistical significance evaluation and confidence
interval calculations are valid approaches. If the F-test statistic is equal to or greater than the Table 2-4
value, analysts will conclude that the sample variances are not the same and will consequently modify the
confidence interval calculation according to Satterthwaite's approximation [7]. The report will discuss
Satterthwaite's approximation if the actual test data indicate that it must be applied.
2.4.5. Baseline Versus Treated Fuel Confidence Interval
If a statistically significant difference in parameters is observed, the 95-percent confidence interval will
be calculated. The half width (e) of the 95 percent confidence interval is [6]:
Eqn. 2-13
Analysts will calculate and report per-notch AE1J5 and ABSFQ, as well as duty cycle-weighted ABSFCDc
and AEiDC results for both the FTP and extended steady-state test modes described above. All reported
results will include the 95 percent confidence interval, if the results are statistically significant.
2.5. COMPARISON WITH THE FTP
The 40 CFR 92 Subpart B FTP is the reference test method for this verification, and test personnel will
follow the FTP procedures in detail where practicable. There are, however, some differences which arise
because of the equipment used or this verification's goals.
2-13
-------
General equipment specifications
1. §92.105(e) specifies absolute pressure measurement gauges or transducers must have an accuracy
and precision of ± 0.1 percent of absolute pressure at point or better. ETC uses Viatran model
218 absolute pressure transmitters, which have an accuracy of ± 0.25 percent of full scale and a
repeatability of less than or equal to ± 0.2 percent of full scale. ETC and the GHG Center
consider this difference to be inconsequential.
2. §92.111 specifies an in-situ, duct-mounted smoke measurement system and indicates that the
light beam must pass through the longest axis of the entire exhaust smoke plume at right angles.
This differs from the way the Bosch RT100A measures opacity because a it draws a partial
exhaust gas sample from the exhaust stack. The length of !/2" sample line is kept to a minimum to
maintain accuracy. Both devices normalize opacity measurements to the path length. These
technique differences are minimal and inconsequential for this test, since any systematic results
difference would be common to both baseline and treated fuel tests.
Dilution system differences
ETC optimized the DOES2 and LPSS dilution sampling systems for in-use field measurements on
operating equipment rather than in a test cell environment. Some physical and functional differences
between these two systems and those specified in the CFR are the result. ETC and the GHG Center
reviewed the results of correlated DOES2 and LPSS tests as compared with 40 CFR 86 test cell
instrumentation. (Note that 40 CFR 92 incorporates 40 CFR 86 methods and instrumentation by
reference.) The results are comparable for emissions measurements on heavy duty diesel engines. The
maximum difference between the two methods was approximately 11 percent for TPM and 4.6 percent
for THC.
Some of the method differences are matters of choice and will not affect measurement results. Other
specific differences are:
1. The FTP specifies that TPM minimum dilution train inner diameter (ID) should be 4 inches and
the extraction probe should be "approximately 1.25 inch" ID. The LPSS dilution train ID is 3.75
inches and its extraction probe is approximately 0.75 inch ID. The LPSS meets the FTP
Reynolds number specifications. The GHG Center considers these differences to be
inconsequential, as confirmed by the previous correlation testing.
2. The formulas in §92.132(b)(3) provide for calculation of total exhaust flowrate based on the
diluted gas concentrations, sample, and dilution air flow rates reported from the DOES2. ETC
will modify the formula for TPM emission rate [§92.132(b)(4)J to reflect that TPM is sampled in
a different train with a different dilution factor. ETC will substitute the LPSS actual dilution air
and sample flowrates into the formula and correlate the measured emissions against the wet total
exhaust volume as calculated from the DOES2 sample. These calculations are functionally
equivalent to the FTP calculations and thus do not represent a true method difference.
3. The FTP [§92.114(c)(4)(ii)J specifies that the dilution air temperature be at 68 °F or greater.
While this can be controlled in a test cell, the dilution air temperature for the LPSS and DOES2 is
limited by the daily ambient conditions, and cannot be guaranteed to be above 68 °F during a
particular mode at a constant dilution ratio. This is of concern especially during tests in cool,
humid conditions. At the end of each test for each mode, ETC operators will inspect the LPSS
and DOES2 filter housings for moisture condensation. If any appears, that mode will be voided
and the dilution rate adjusted as necessary to eliminate the condensation during the repeated
mode.
2-14
-------
3.0 DATA QUALITY
3.1. DATA QUALITY OBJECTIVE
The GHG Center will perform this verification according to Title 40 CFR 92, subpart B (§92.101 ff)
procedures with certain modifications as discussed in Section 2.5. All measurement equipment
accuracies, analyses, and QA/QC procedures will equal or exceed those specified in the CFR or as
described in this test plan if different from the CFR. Achievement of this qualitative DQO will be
documented by the QA/QC checks discussed in the following subsections.
The actual fuel consumption and emissions reductions cannot be established until tests are concluded, so
this test plan does not adopt explicit quantitative DQOs. The estimated Catalyzer effects (5.0 percent for
fuel consumption and 18.0 percent for NOX reductions) and the analyses presented in Section 2.4 suggest
the following implicit data quality objectives for field personnel.
The implicit DQOs are that the data show statistical significance, variance similarity, and that the 95
confidence interval be refined as much as possible up to a maximum of six test runs.1 Although the field
data may show statistical significance at the 3rd test run, the GHG Center will consider additional runs up
to a maximum of 6. This is because, given a constant measurement variability, results from more test
runs allow a smaller 95 percent confidence interval calculation on the mean result. The ability to report a
"5.0 ± 2.0 percent fuel consumption change" will have more value than a "5.0 ± 4.5 percent fuel
consumption change" report. The field team leader will use equations 2-10 through 2-13 and the
Appendix C4 log forms to perform these evaluations."
3.2. INSTRUMENT SPECIFICATIONS, CALIBRATIONS, AND QA/QC CHECKS
Table 3-1 lists the instruments to be used in this verification test, their expected operating ranges, and
accuracy specifications. 40 CFR 92 Subpart B or 40 CFR 86 Subparts D or N are the sources for most of
the cited specifications. Table 3-2 provides instrument calibration specifications and schedules, as
provided by the FTP or as determined by ETC from other test programs. Table 3-3 summarizes field and
laboratory QC checks for the major systems. All calibrations, QA/QC checks, and acceptance criteria for
sampling equipment and analyzers that are described in 40 CFR 92, subpart B, are incorporated by
reference.
1 For reference, if the fuel consumption change is 5.0 percent, the sample standard deviation must be less than 2.2 percent (with
respect to the baseline results) if three runs are conducted or 3.9 percent if 6 runs are conducted. Similarly for the estimated
18.0 percent NOX reductions, the sample standard deviation must be less than 7.9 percent for 3 test runs and 14.0 percent for 6
test runs for the results to be statistically significant.
3-1
-------
Table 3-1. Instrument and Accuracy Specifications
Measure-
ment
Variable
Main traction
generator
voltage
Main traction
generator
current
Fuel flow rate
DOES2 main
flowrate
DOES2
dilution air
flowrate
DOES2
analyzer
sample
flowrate
LPSS main
flowrate
LPSS dilution
air flowrate
Temperature
LPSS main
Diff. pressure,
LPSS/DOES
Ambient
temperature
Ambient
pressure
Humidity,
ambient
CO
CO2
NOX
THC
PM Mass
Opacity
Expected
Operating
Range
0 - 900 V
0 - 4000 A
50 - 400
gph
35 - 45 1pm
5 - 40 1pm
91pm
400 1pm
275 - 325
1pm
<125°F
0-5 " H2O
40 - 90 °F
14.7 psia
40-100 %
0 - 300 ppm
0 - 3.0 %
0 - 300 ppm
0 - 10 ppm
n/a
0 - 40 %
Instrument Mfg.,
Model / Type
Flex-Core VT8-
014E
Flex-Core CTL-
502HS/4000; CTA-
215NYXX
FuelCom FC-050,
(paired meters for
supply, return)
MKS mass flow
controller model
1559
MKS mass flow
controller model
1559
Two parallel
MKS mass flow
controllers
model 179
Hot wire
anemometer
Mass flow meter
National
Semiconductor 1C
Temp Sensor
MKS 223B
VaisalaHM141
Viatran218
absolute pressure
transmitter
VaisalaHM141
Horiba
AIA-23ASWOPE-
15
Horiba
AIA-23/OPE115
Horiba
CLA 220
California
Analytical 300M-
HFIDCE
Sartorius MP5V1
microbalance
Bosch -RT100A
Instrument
Range
0 - 1000 V
0 - 4000 A
50-500 gph
0-100 1pm
0 -50 1pm
0-5 1pm, each
101pm, total
0 - 8500 1pm
0 - 500 1pm
32 - 392 °F
0- 10"H2O
39 - 212 °F
0-15 psia
0-100% RH
0 - 300 ppm
0 - 50 ppm
0 - 3 %
0 - 300 ppm
0 - 100 ppm
0-10 ppm
propane
0 - 2000 mg
0 - 100 %
Measurement
Frequency
Continuous, 1 Hz
Continuous, 1 Hz
Continuous, 1 Hz
Continuous, 2 Hz
Continuous, 2 Hz
Continuous, 2 Hz
Continuous, 2 Hz
Continuous, 2 Hz
Continuous, 2 Hz
Continuous, 2 Hz
1 per test run
1 per test run
1 per test run
Continuous, 1 Hz
Continuous, 1 Hz
Continuous, 1 Hz
Continuous, 1 Hz
2 per filter
Continuous, 2 Hz
Specification
+ 0.25 % FS
+ 1.0% of point
+ 1.0% of point
+ 1.0%FS
+ 1.0%FS
+ 1.0%FS
+ 1.0% of point
+ 1.0% of point
+ 0.9°F
+ 0.5%FS
+ 0.2%FS
+ 0.25% FS
+ 1.0%FS
+ 1.0% of point
+ 1.0% of point
+ 1.0% of point
+ 1.0% of point
+ 5.0ug
+ 1.0% opacity
How Verified /
Determined
Factory calibration
Factory calibration
Factory calibration
Factory /laboratory
calibration
Factory /laboratory
calibration
Factory/laboratory
calibration
Factory /laboratory
calibration
Factory/laboratory
calibration
Factory/laboratory
calibration
Factory/laboratory
calibration
Factory calibration
Factory/laboratory
calibration
Factory calibration
Factory, laboratory,
field calibration and
drift checks
Factory / laboratory
calibration
Factory / laboratory
calibration
3-2
-------
Table 3-2. Calibration Schedule
System or
Parameter
Main traction
generator
voltage
Main traction
generator
current
Fuel flow rate
DOES2 main
flowrate
DOES2
dilution air
flowrate
DOES2
analyzer
sample
flowrate
LPSS main
flowrate
LPSS dilution
air flowrate
Temperature
LPSS main
Diff. Pressure,
LPSS/DOES
Temperature,
ambient
Pressure,
ambient (BP)
Humidity,
ambient
CO
CO2
NOX
THC
CO
CO
CO2
NOX
PM mass
smoke
Description/ Procedure
NIST-traceable calibration
with as-found data
NIST-traceable calibration
with as-found data
Calibration with #2 diesel fuel
against NIST traceable std
meter at 7 temperatures (50 -
160 F) x 12 flow rates in
range (10% - 100% FS)
Calibration against Gilibrator
standard bubble flow meter
Calibration against Meriam
laminar flow element
Calibration against Gold seal
mass flow controller
Calibration against Omega
temperature calibrator
Calibration against Druck
pressure calibrator
Calibrated against laboratory
standard
Calibration against Druck
pressure calibrator
Calibrated against laboratory
standard
Gas divider calibration with
protocol calibration gases at
1 1 points evenly spaced
throughout span (including
zero)
CO2 interference check
Water interference check
Water interference check
Converter efficiency check
Balance calibrated by control
weights
calibration with NIST
traceable ND filters at 0, 10,
20, 40% opacity
Procedure
Reference
n/a
n/a
Fuelcom
internal; meets
§92.107
§92.117
§92.1
n/a
§92.105
n/a
§92.120
§92.120
§92.121
§92.119
§92.120,
§92.109
§92.120,
§92.109
§92.120,
§92.109
§92.120,
§92.109
§92.128,
§92.110
§92.122
Frequency
within 18
months
within 18
months
within 12
months
Before test
(check during)
Before test
(check during)
Every 4 weeks
or before
analyzer leaves
for field
Monthly
Daily
6 months
Allowable Result
All values within + 0.25 %
of point
All values within + 1.0 %
of point
All values within + 0.25 %
of point
+ 1.0 % of FS or + 2.0% of
point
Within +/-1. 7 °C
+ 0.1 % of absolute
pressure at point
Within +/-1. 7 °C
+ 1.0% of absolute
pressure at point
All values within + 2.0 %
of point or + 1.0% of FS;
zero point within + 0.2% of
FS
CO2 rejection ratio > 5000
to 1
water rejection ratio > 1000
to 1
water rejection ratio > 100
to 1
Converter efficiency > 90%
Per §92. HOe criteria
All values within + 1.0 %
of nominal opacity
3-3
-------
Table 3-3 System QA/QC Checks
System or
Parameter
Main traction
generator
voltage
Main traction
generator power
Test duct
cyclonic flow
Exhaust gas
flow rate
DOES2 leak
checks
DOES2 flowrate
check
LPSS leak
checks
LPSS flowrate
check
Temperature
LPSS main
LPSS / DOES2
moisture
condensation
Diff. Pressure,
LPSS/DOES
Tunnel blank
Ambient
Pollutant Levels
Analyzer zero
and span drift
check
QA/QC Check
Meter reasonableness check vs.
digital voltmeter (DVM)
Reasonableness: voltage and
current within manufacturer's
specifications
Method 1 cyclonic flow
determination
FTP exhaust gas flowrate
comparison with Method 2 pilot
traverse
Exhaust gas delta P monitoring
with stationary pilot at
representative sampling location
Tunnel is capped and drawn
from by main pump
pislon-lype calibrator
comparison
Tunnel is capped and drawn
from by sample pump
Each flow device is removed
from Ihe system and compared
lo a calibrated laminar flow
elemenl
Each lemperalure probe is
removed and calibrated againsl a
lemperalure calibrator. This is
only done in house.
Inspection of filler holders for
moislure
Each differential pressure
Iransmiller is calibrated againsl a
Druck pressure calibrator
Run simulation lesl sequence
Disconnecl from Exhausl probe
and run lesl Irace also serves as
warm up run.
Analyzer is zeroed and spanned
before each reading using on site
calibration gases
When Performed/
Frequency
Performed prior lo
and during lesl
series
Performed prior lo
and during lesl
series
Prior lo firsl lesl run
Al each nolch prior
lo and immediately
following each
series of baseline or
Irealed fuel tests
Throughoul all lesl
runs
Performed daily
prior lo lesl
Performed prior lo
testing
Performed daily
prior lo lesl
Performed prior lo
Iravel
Performed prior lo
Iravel
Immediately
following each lesl
run al each mode
Performed prior lo
Iravel
One blank taken per
day
One sample per lesl
series
Each lesl run
Expected or
Allowable Result
V values wilhin +
2.0 % of FS
Power wilhin 10%
of nominal for
nolch
< 20 ° cyclonic
flow
FTP Flowrate
agreemenl wilhin
+ 10%ofpilol
Iraverse resull
Wilhin + 15% of
Ihe mean Melhod
2 delta P al lhal
Iraverse poinl for
each nolch
< 1 1pm
+ 1.0%ofFSor +
2.0% of poinl
< 1 1pm
+ 1.0%ofFSor +
2.0% of poinl
+ 1.7°C
No visible
moislure on Ihe
internal surface of
any fitting,
housing, or filler
0.1 % of absolute
pressure al poinl
Blank musl nol
exceed 5.0 % of
sample weighl
Reasonable
ambienl levels
Posl-lesl zero or
span drift shall nol
exceed + 2.0 % FS
Response to Check
Failure or Out of
Control Condition
Find cause and correcl
or repair
Find cause and correcl
or repair
Modify lesl duel
Find cause and correcl
or repair
Find leak and repair
Adjusl MFC
accordingly and re-
calibrate
Find leak and repair
Adjusl accordingly and
re-calibrale
Replace lemperalure
probe
Void resulls for lhal
mode, adjusl dilution
ratio as necessary, and
relesl lhal mode
Adjusl Iransmitter
accordingly and re-
calibrale
If blank exceeds 5% of
sample weighl, ils
weighl is taken into
accounl in Ihe
calculations for Ihe
final sample resulls
Background corrections
in formulas
Assess impacls; Correcl
or void runs
Note that SO2 and sulfate QA/QC procedures will conform to the ETC methods 6.3/5.0/M and 6.5/1.0/M.
Sample handling procedures will conform to ETC method 11.23/1.2/S.
3-4
-------
The entries in Tables 3-2 and 3-3 will serve as the data quality indicators (DQI) for this test campaign.
Each DQI links with a corresponding measurement or determination which, in turn, contributes to
achievement of the overall qualitative and implicit DQOs. For example, the NOX converter must show
better than 90 percent efficiency for the NOX measurements to be valid. A valid NOX measurement then
contributes to the proper performance of the FTP, which is the test campaign's DQO.
3.3. INSTRUMENT TESTING, INSPECTION, AND MAINTENANCE
GHG Center personnel, the field team leader, or ETC will subject all test equipment to the QC checks
discussed earlier. Before tests commence, operators will assemble and test all equipment as anticipated to
be used in the field. They will, for example, operate and calibrate all controllers, flow meters, computers,
instruments, and other measurement system sub-components per the specified test methods and/or this
test plan. Test personnel will repair or replace any faulty sub-components before starting the verification
tests. Test personnel will maintain a small amount of consumables and frequently needed spare parts at
the test site. The field team leader, project manager, or ETC will handle major sub-component failures on
a case-by-case basis such as by renting replacement equipment or buying replacement parts.
3.4. INSPECTION AND ACCEPTANCE OF SUPPLIES AND CONSUMABLES
ETC calibrations will employ NIST-traceable or EPA Protocol 1 gases supplied either by a gas-divider
dilution system or directly from cylinders. Per EPA protocol gas specifications, the actual concentration
must be within ± 2 percent of the certified tag value. Gases certified to ± 1.0 percent will be used for
multipoint gas analyzer calibrations in accordance with 40 CFR 92 specifications. Copies of all EPA
protocol gas certifications will be available on-site.
The field team leader will provide technical oversight of the ETC field activities. The GHG Center QA
manager will review ETC calibration data and QA/QC check results to verify that emissions
measurements conform to 40 CFR 92, Subpart B requirements.
3-5
-------
3-6
-------
4.0 DATA ACQUISITION, VALIDATION, AND REPORTING
4.1. DATA ACQUISITION AND DOCUMENTATION
Test personnel (responsible parties are noted below in parentheses) will acquire the following types of
data and generate the following documentation during the verification:
fuel consumption and power data (GHG Center)
fuel emissions data (ETC)
manually acquired parameters and printed output data from the ETC sampling systems
such as sampling and dilution air flow rates, exhaust gas analyzer concentration, ambient
pressure, exhaust gas pressure, temperature, and ambient conditions (ETC)
QA/QC documentation as described in Section 3.0 (ETC, GHG Center)
field test documentation (GHG Center)
corrective action and assessment reports (GHG Center)
ETC will submit copies of all test-run printed outputs, calibration forms, analyses, certificates, etc. to the
Field Team Leader as each test run is completed. These submittals must be complete prior to the Field
Team Leader's departure after the final test run.
ETC will prepare and submit a report in printed and electronic format to the GHG Center Field Team
Leader within three weeks of the field activities' completion. The report will describe the test conditions,
document all QA/QC procedures, include copies of calibrations, calibration gas, and the verification test
results. The report will include a signed certification which attests to ETC's conformance with all
QA/QC procedures and the accuracy of the results. ETC will attach all relevant test data as appendices.
The following subsections discuss each of these items and their role in the test campaign. The GHG
Center will archive all electronic data, paper files, analyses, and reports at their Research Triangle Park,
NC office in accordance with the QMP.
4.1.1. Fuel Consumption and Power Data
The GHG Center Field Team Leader will obtain fuel consumption and power data during the multimode
tests. In addition to documenting the data for use in the report, he will supply these data to ETC staff for
their use in the following sections.
4.1.2. Emissions Data
ETC will be responsible for all emissions data, associated QA/QC log forms, paper, and electronic files
until they are accepted by the Field Team Leader.
ETC will report emission measurements for each test mode to the Field Team Leader as:
ppmv (percent for CO2) of emissions
g/bhp-h of pollutants
calculated exhaust flow rate based on carbon balance methods.
calculated exhaust flow rate based on Method 2 traverses.
4-1
-------
4.1.3. Locomotive Documentation
The Field Team Leader will document the applicable vehicle and engine specifications. Documentation
will generally conform to 40 CFR §92.133 and will include information such as:
O J o
locomotive engine family identification
alternator generator efficiency specifications
hourmeter readings prior to the baseline and treated fuel test series
general duty description
a description of the service during the break-in period
4.1.4. QA/QC Documentation
Upon completion of the field test activities, ETC will provide copies of calibrations, pre-test checks,
system response time, NO2 converter efficiency, and other QA/QC documents to the Field Team Leader.
Calibration records will include information about the instrument being calibrated, raw calibration data,
calibration equations, analyzer identifications, calibration dates, calibration standards used and their
traceabilities, calibration equipment, and names of participating staff. These records will provide source
material for the Verification Report's Data Quality section, and will be available to the QA Manager
during audits.
4.1.5. Field Test Documentation
The Field Team Leader will obtain copies of all manually and digitally logged data. He will take site
photographs and maintain a Daily Test Log which will include the dates and times for setup, testing,
teardown, and other activities.
The Field Team Leader will record test run information and observations in the Daily Test Log and on the
log forms in Appendix C. The Field Team Leader will submit digital and paper data files, ERMD test
results, and the Daily Test Log to the Project Manager.
4.1.6. Corrective Action and Assessment Reports
A corrective action will occur when audits or QA/QC checks produce unsatisfactory results or upon major
deviations from this Test Plan. Immediate corrective action will enable quick response to improper
procedures, malfunctioning equipment, or suspicious data. The corrective action process involves the
field team leader, project manager, and QA Manager. The GHG Center QMP requires that test personnel
submit a written corrective action request to document each corrective action.
The field team leader will most frequently identify the need for corrective actions. In such cases, he or
she will immediately notify the project manager. The field team leader, project manager, QA Manager
and other project personnel, will collaborate to take and document the appropriate actions.
Note that the project manager is responsible for project activities. He is authorized to halt work upon
determining that a serious problem exists. The field team leader is responsible for implementing
corrective actions identified by the project manager and is authorized to implement any procedures to
prevent a problem's recurrence.
4-2
-------
4.2. DATA REVIEW, VALIDATION, AND VERIFICATION
The project manager will initiate the data review, validation, and analysis process. At this stage, analysts
will classify all collected data as valid, suspect, or invalid. The GHG Center will employ the QA/QC
criteria specified in Section 3.0 and the associated tables. Source material for data classification include
factory and on-site calibrations, maximum calibration and other errors, audit gas analyses results, and lab
repeatability results.
In general, measurements which:
meet the specified QA/QC checks,
were collected when an instrument was verified as being properly calibrated,
are consistent with reasonable expectations (e.g., manufacturers' specifications,
professional judgment)
will form the basis for valid data.
The report will incorporate all valid data. Analysts may or may not consider suspect data, or it may
receive special treatment as will be specifically indicated. If the DQO cannot be met, the project manager
will decide to continue the test, collect additional data, or terminate the test and report the data obtained.
Data review and validation will primarily occur at the following stages:
on site ~ by the field team leader
before writing the draft report ~ by the project manager
during draft report QA review and audits ~ by the GHG Center QA Manager
The field team leader's primary on-site functions will be to monitor ETC activities and acquire fuel
consumption and power generation data. He will review, verify, and validate certain data (DOES2 file
data, QA/QC check results, etc.) during testing. He will plan to be on-site during all test activities. Log
forms in Appendix A provide the detailed information he will gather.
The QA Manager will use this test plan and documented test methods as references with which to review
and validate the data and the draft report. He will review and audit the data in accordance with the GHG
Center's QMP. For example, the QA Manager will randomly select raw data and independently calculate
the verification parameters. The comparison of these calculations with the results presented in the draft
report will yield an assessment of the GHG Center's QA/QC procedures.
4.3. DATA QUALITY OBJECTIVES RECONCILIATION
A fundamental component of all verifications is the reconciliation of the collected data with its DQO. In
this case, the qualitative DQO assessment consists of evaluation of whether the stated methods were
followed and satisfactory results obtained for the QC checks specified in Section 3.0. As discussed in
Section 4.2, the field team leader and project manager will initially review the collected data to ensure
that they are valid and are consistent with expectations. They will assess the data's accuracy and
completeness as they relate to the stated QA/QC goals. If this review of the test data show that QA/QC
goals were not met, then immediate corrective action is feasible, and will be considered by the project
manager. DQOs will be reconciled after completion of corrective actions. As part opf the internal Audit
of Data Quality (ADQ), the GHG Center QA Manager will include an assessment of DQO attainment.
4-3
-------
4.4. ASSESSMENTS AND RESPONSE ACTIONS
The field team leader, project manager, QA Manager, GHG Center Director, and technical peer-reviewers
will assess the project and the data's quality as the test campaign proceeds. The project manager and QA
Manager will independently oversee the project and assess its quality through project reviews, inspections
if needed, a scheduled PEA, and an ADQ.
4.4.1. Project Reviews
The project manager will be responsible for conducting the first complete project review and assessment.
Although all project personnel are involved with ongoing data review, the project manager must ensure
that project activities meet measurement and DQO requirements.
The GHG Center Director will perform the second project review. The director is responsible for
ensuring that the project's activities adhere to the ETV program requirements and stakeholder
expectations. The GHG Center Director will also ensure that the field team leader has the equipment,
personnel, and resources to complete the project and to deliver data of known and defensible quality.
The QA Manager will perform the third review. He is responsible for ensuring that the project's
management systems function as required by the QMP. The QA Manager is the GHG Center's final
reviewer, and he is responsible for assuring the achievement of all QA requirements.
Envirofuels, G&W, and selected GHG Center stakeholders and/or peer reviewers will then review the
report. Technically competent persons who are familiar with the project's technical aspects, but not
involved with project activities, will function as peer reviewers. The peer reviewers will provide written
comments to the project manager.
The GHG Center will submit the draft report to EPA QA personnel, and the Project Manager will address
their comments as needed. Following this review, the report will undergo EPA management reviews,
including the GHG Center Director, EPA ORD Laboratory Director, and EPA Technical Editor.
4.4.2. Performance Evaluation Audit
The GHG Center will conduct a performance evaluation audit (PEA) of the emission sampling system
and analyzers. The PEA will be performed by introducing a sample of audit gas of known concentration
to the system. The performance evaluation audit (PEA) gas will consist of a mixture of 0.5 to 4 percent
CO2 in air, but whose exact concentration is blind to the DOES2 system operator. The field team leader
will supply the audit gas to the DOES2 sample probe from the cylinder through one leg of a sample line
with a tee fitting. The remaining leg will be open to atmosphere through a rotameter. The cylinder
regulator will supply gas at the DOES2 system's normal sampling rate (approximately 40 1pm) with
enough surplus such that the rotameter shows flow to the atmosphere. The field team leader will submit
the data to the QA Manager, who will incorporate them into a PEA report to the GHG center.
4.4.3. Technical Systems Audit
The GHG Center QA Manager will perform a technical systems audit (TSA) to assess the implementation
of the Test/QA Plan. This TSA will include an evaluation of the following specific test equipment as well
as the implementation of the test plan requirements:
4-4
-------
Power metering equipment and calibrations
Fuel metering equipment and calibrations
ETC Instrumental analyzer system, calibrations
ETC DOES2 and LPSS sampling systems and calibrations
During the TSA, the QA Manager will verify that the equipment, procedures, and calibrations are as
specified in this Test Plan. Should the QA Manager note any deficiencies in the implementation of the
Test Plan, corrective actions will be immediately implemented by the project manager. The results of the
TSA will be documented in a separate TSA report.
4.4.4. Audit of Data Quality
The ADQ is an evaluation of the measurement, processing, and data analysis steps to determine if
systematic errors are present. During the ADQ, the QA Manager, or designee, will randomly select
approximately 10 percent of the data. He will follow the selected data through analysis and data
processing. The ADQ's scope is to verify that the data-handling system functions correctly and to assess
the quality of the analysis. The QA Manager will also include an assessment of DQO attainment.
The QA Manager will route the ADQ results to the project manager for review, comments, and possible
corrective actions. Project records will document the results. The project manager will take any
necessary corrective action needed and will respond by addressing the QA Manager's comments in the
report.
4.5. VERIFICATION REPORT AND STATEMENT
The project manager will coordinate report preparation. The report will summarize each verification
parameter's results as discussed in Section 2.0 and will contain sufficient raw data to support findings and
allow others to assess data trends, completeness, and quality. The report will clearly characterize the
verification parameters, their results, and supporting measurements as determined during the test
campaign. It will present raw data and/or analyses as tables, charts, or text as is best suited to the data
type. The report will also contain a Verification Statement, which is a 4 to 7 page document describing
the technology, the test strategy used, and the verification results obtained.
The report will also include the change in SO2 and sulfate emissions measured at notch 8, but these results
will be for information only and will not be considered as verification parameters.
The Project Manager will submit the draft Report and Statement to the QA Manager and Center Director
for review. A preliminary outline of the report is as follows:
Preliminary Outline
Envirofuels Diesel Fuel Catalyzer Verification Report
Verification Statement
Section 1.0: Verification Test Design and Description
Description of the ETV program
Catalyzer and Test Locomotive Description
Overview of the Verification Parameters and Evaluation Strategies
4-5
-------
Section 2.0: Results
Fuel Consumption Change
Emissions Performance
Section 3.0: Data Quality
Section 4.0: Additional Technical and Performance Data (optional) supplied by vendor
References:
Appendices: Raw Verification and Other Data
4.6. TRAINING AND QUALIFICATIONS
This test does not require specific training or certification beyond that required internally by the test
participants for their own activities. The GHG Center's field team leader is a licensed professional
engineer with approximately 15 years experience in field testing of air emissions from many types of
sources. He is also familiar with engine and vehicle testing, operations, maintenance, and repair. He is
familiar with the test methods and standard requirements that will be used in the verification test.
The project manager has performed numerous field verifications under the ETV program, and is familiar
with EPA and GHG Center QMP requirements. The QA Manager is an independently appointed
individual whose responsibility is to ensure the GHG Center's conformance with the EPA approved
QMP.
4.7. HEALTH AND SAFETY REQUIREMENTS
This section applies to GHG Center personnel only. Other organizations involved in the project have
their own health and safety plans - specific to their roles in the project.
GHG Center staff will comply with all known host, state/local and Federal regulations relating to safety at
the test facility. This includes use of personal protective gear (e.g., safety glasses, hard hats, hearing
protection, safety toe shoes) as required by the host and completion of site safety orientation (i.e., site
hazard awareness, alarms and signals).
4-6
-------
5.0 REFERENCES
1. Draft Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990 - 2002, U.S. EPA,
Washington, DC. 2004
(http://yosemite.epa.gov/oar/globalwarming.nsf/content/ResourceCenterPublicationsGHGEmissi
onsUSEmissionsInventorv2004.html)
2. Emissions of Greenhouse Gases in the United States 2002, Energy Information Administration,
Washington, DC. 2003
3. Title 40 CFR 92Control of Air Pollution from Locomotives and Locomotive Engines,
Environmental Protection Agency, Washington, DC. Adopted at Federal Register Vol. 63, No.
73, April 16, 1998
4. Test and Quality Assurance PlanConocoPhillips Fuel-Efficient High-Performance SAE 75W90
Rear Axle Gear Lubricant, SRI/USEPA-GHG-QAP-28, March 2003.
5. Generic Verification Protocol for Determination of Emissions Reductions Obtained by Use of
Alternative or Reformulated Liquid Fuels, Fuel Additives, Fuel Emulsions, and Lubricants for
Highway and Nonroad Use Diesel Engines and Light Duty Gasoline Engines and Vehicles,
Research Triangle Institute, EPA Cooperative Agreement No. CR826152-01-3, September 2003.
6. Statistics Concepts and Applications, D.R. Anderson, E.J. Sweeney, T.A. Williams. West
Publishing Company, St. Paul, MN. 1986
7. A Modern Approach to Statistics, R.L. Iman, W.J. Conover. John Wiley & Sons. New York,
NY. 1983
8. GP-40 Operators Manual and EMD 645E3 Turbocharged Engine Maintenance Manual. EMD.
LaGrange, Illinois. 1968.
5-1
-------
5-2
-------
APPENDICES
------- |