Measurement of Real-World Emissions from Heavy-Duty Diesel Vehicles:

The State-of-the-Art

Mridul Gautam, Greg Thompson, Nigel Clark, Dan Carder, Wes Riddle, Don Lyons

Department of Mechanical and Aerospace EngineeringWest Virginia University

NATIONAL RESEARCH CENTER FOR ALTERNATIVE TRANSPORTATION FUELS, ENGINES AND EMISSIONS

West Virginia UniversityFuels, Engines, and Emissions Research

Purpose of In-use Emissions Measurements

• Technology Development and/or Assessment

• Enforcement

• Compliance

• I/M

• Screening

• Inventory

Available Tools• Engine Test Cells

– Simulated Routes– FTP

• Chassis Dynamometers

• On-road, On-board Emissions Measurement Systems– AEI, Columbus, IN– Horiba, Ann Arbor, MI– Sensors, Saline, MI

Challenges to Measurement of On-board, On-road Diesel Emissions

• False positives ; Error minimization• Torque (or percent load) broadcast• Exhaust flowrate measurement• Fuel quality variability• Emissions characteristics from current and future engines/exhaust

aftertreatment systems (NO, NO2, OC dominated PM emissions)• Current definition of particulate matter.

• Obsession with brake-specific emissions– It is recognized that the FTP (brake-specific emissions) is essential – However, in-use fuel-specific emissions would eliminate majority of

challenges associated with collecting brake-specific emissions in the field (application dependent)

• Instrumentation– Advances in systems development have not been fast enough– Portability; Bulk

West Virginia University, Morgantown,WV 26506

In-Use Emissions Work at WVURelated to Consent Decrees

• PHASE I: DEVELOPED MOBILE EMISSIONS MEASUREMENT SYSTEM FOR ON-BOARD, IN-USE HEAVY-DUTY VEHICLE APPLICATIONS

• PHASE II: DEVELOPED IN-USE EMISSIONS TESTING PROCEDURES, AND TEST ROUTES

• PHASE III: CONDUCTED EMISSIONS TESTING ON A VARIETY OF IN-SERVICE DIESEL ENGINES (≤MY1998)USING THE WVU MOBILE EMISSIONS MEASUREMENT SYSTEM (MEMS) TO CHARACTERIZE REAL-WORLD EMISSIONS FROM SUCH ENGINES

West Virginia University, Morgantown,WV 26506

In-Use Emissions Work at WVURelated to Consent Decrees (…Cont’d)

• PHASE IV: CONDUCTING ON-ROAD COMPLIANCE MONITORING OF HEAVY-DUTY DIESEL VEHICLES (≥MY2002) USING THE MONITORING TECHNOLOGY, AND PREVIOUSLY DEFINED TESTING PROCEDURES (AND DRIVING ROUTES) DEVELOPED BY WVU, AND APPROVED BY THE US EPA.

Mobile Emissions Measurement System

(MEMS)

Exhaust Samplingand Conditioning

FlowMeter

EngineInterface

AmbientSensors

ExhaustConstituentsNOx & CO2

Differential Pressure,Absolute Pressure,and Temperature

Engine Speed,Vehicle Speed,

& Torque

Ambient Temperature,

Pressure,& Humidity

GPS

Vehicle Speed & Location

Mobile Emissions Measurement System

West Virginia University,Morgantown, WV 26506

NOX MASS EMISSION RATES ON SAB2SW ROUTE – MEMS AND

LABORATORY: CUMMINS ISM 370

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Time (s)

NO

x (g

/s)

Dilute LaboratoryRaw MEMS

Comparison of Brake Specific Emissions Results

from the FTP Test Cell and MEMS

-4.39%

524.0

548.0

CO2(g/bhp-hr)

-0.18%Percent Difference

4.389MEMS

4.397Laboratory

NOx(g/bhp-hr)FTP Cycle

Exhaust Stack Measurements

Current Status and Future Direction

• CO2– Solid State NDIR

• Current “garage-grade” NDIR microbenches have served the purpose

– New Direction – “Sensor-on-a-chip” (a hot-bolometer). Silicon microbridge elements with photonic bandgap modified surfaces.

• CO– Improvements are needed to current solid-state “garage-grade”

NDIR’s ability to measure low levels of CO from diesel engines

t (m sec)

I (m

A)

t (m sec)

T (K

)

λ (µm )

Flux

(a.u

.)

λ (µm )

Flux

(a.u

.)

λ (µm )

Flux

(a.u

.)

t (m sec)

T (K

)

Source driver

Source temperature

Source emission

Optical efficiency

Gas attenuation

Dual wavelength source/detector

Detector response

Sensor-on-a-Chip (NDIR)

Current Status and Future Direction

• NOx

– Zirconium oxide sensors• Results are better with a NOx converter. • Avoid using the sensor in the raw exhaust stream

– Current “garage-grade” NDUV may not fully account for noise attenuation, interferences, lamp decay

– New Direction – UV Resonance Absorption Spectroscopy

UV-Resonance Absorption Spectroscopy• Simple photometer technology• Excellent agreement with other

standard methods• No critical components:

– Hot measurement, no chiller– Direct measurement of NO2 ,

no converter– No ozone generator, no

vacuum pumps etc.• Calibration with long-term stable,

gas-filled calibration cells• Simultaneous measurement of

up to 3 gas components NO, NO2, NH3 0.0

0.1

0.1

0.2

0.2

0.3

0.3

0.4

200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580

WL in nm

Abs

orpt

ion

0

10

20

30

40

50

60

70

80

90

100

Lam

p In

tens

ity

NO NO2 EDL

NO

NO2

Current Status and Future Direction• Total Hydrocarbons

– NDIR detectors are not recommended for diesel exhaust– Heated Flame Ionization Detectors serve well– New Direction – Reduce the size and complexity of the FIDs.

Systems are available, but prone to breakdowns• Exhaust Flowrate

– Considering the system accuracy, turn-down ratio, meter drift, measurement frequency, response time, size and weight, robustness (including operation in harsh environments), low backpressure on the engine, etc.

• Averaging pitot tubes (e.g. Annubar), the AEI system• Others:

– Ultrasonic flowmeters (size/temperature limitations), – Hot-wires (response time limitations), – ECU based value (calculated value), – Intake measurements (system leaks, time delays, secondary air

pump during cold start, positioning in the intake system)

Current Status and Future Direction

• Torque and Engine Speed

– Engine Speed is an Accurate Measurement– Inference of Engine Power is Possible via Publicly

Broadcast ECU Information– Engine Torque/Power Can Be Inferred to Within ±10% of

Laboratory Measurements• Engine Maintenance History• Lug Curve• Accessory Loading• Ambient Conditions Limitation• Operate Within the NTE Zone• Use Good Engineering Judgment

Current Status and Future Direction

• Uncertainties associated with Exhaust Flowrateand Torque measurements can be avoided by Fuel Specific Measurements

• Only concentration measurements will be required

Brake Specific Emissions

NOx

CO2

=(g/bhp-hr) NOx

(g/bhp-hr) CO2

=

(Density)*(Exhaust Mass Flow Rate)*(Concentration)NOx

bhp-hr

(Density)*(Exhaust Mass Flow Rate)*(Concentration)CO2

bhp-hr

NOX Index(Fuel Specific Emissions)

(Concentration of NOx) x (Exhaust flow rate) x MWNOx

(Concentration of CO2) x (Exhaust flow rate) x (12.011+1.008*(H:C))

grams of NOx / kg of Fuel

•NOx concentration•CO2 concentration•Fuel H:C ratio

Engine Operation Over an On-road Route

0

200

400

600

800

1000

1200

1400

1600

1800

500 1000 1500 2000

Engine Speed(RPM)

Engi

ne T

orqu

e(ft-

lb)

lug curveTorqueN loN hi15% ESC Speed30% Torque30% Power Torque

NTE Zone: Engine Operation and Brake Specific NOxas a function of Engine Speed and Engine Load

0

200

400

600

800

1000

1200

1400

1600

1800

500 1000 1500 2000

Engine Speed(RPM)

Engi

ne T

orqu

e(ft-

lb)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

NTE

bsN

Ox

(g/b

hp-h

r)

lug curveNTE TorqueN loN hi15% ESC Speed30% Torque30% Power TorqueNTE bsNOx

3-D plot of Brake Specific NOx as a function of Engine Speed and Engine Load

1 2 0 01 4 0 0

1 6 0 01 8 0 0

2 0 0 0

2

3

4

5

8 0 01 0 0 0

1 2 0 01 4 0 0

1 6 0 01 8 0 0

NTE

bsN

Ox

(g/b

hp-h

r)

Engine Load (ft-lb

)E ng ine S peed (R P M )

Errors in Torque Broadcast(Simulated SAB-to-BM Route)

-500

0

500

1000

1500

2000

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Time (s)

Engi

ne L

oad

(ft-lb

)

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

100.00Inferred Torque from Lab Gen. Lug Curve (ft-lb)Lab Torque (ft-lb)Inferred Torque from Manf. Lug Curve (ft-lb)% Difference

Simulated SAB-to-BM Route – NTE Zone

0

200

400

600

800

1000

1200

1400

1600

1800

500 1000 1500 2000

Engine Speed(RPM)

Engi

ne T

orqu

e(ft-

lb)

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

lug curveTorqueN loN hi15% ESC Speed30% Power Torque%Difference

Ratio of bsNOx/bsCO2 vs. Time

0.0000

0.0020

0.0040

0.0060

0.0080

0.0100

0.0120

0.0140

0 500 1000 1500 2000 2500 3000

Time (s)

ratio

of b

sNO

x/bs

CO

2

ratio from on-road data

ratio from ftp cycle

Ratio of bsNOx/bsCO2 vs. Power

0.000

0.005

0.010

0.015

0.020

0.025

0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 450.00

Power(hp)

Rat

io(N

Ox/

CO

2 )

0.000

2.000

4.000

6.000

8.000

10.000

12.000

bsN

Ox

(g/b

hp-h

r)

Ratio on road

Ratio from FTP cycle

bsNOxStandard bsNOx

.00928

5.0g/bhp-hr

In-use Particulate Matter EmissionsMajor Challenge:

• Definition of particulate matter• Current definition of particulate matter emissions is

valid in a multi-million dollar brick-and-mortar engine/chassis dynamometer test cell

• EPA is hoping to demonstrate “equivalency” between portable PM instruments and test cell methods.– On integrated PM measurements

• Several years down the road unless definition of PM is modified.

Real-Time Particulate Mass Monitor (Quartz Crystal Microbalance)

• In-use, On-board applications (capable of handling severe vibrations)

• Test cell applications• Ultra-clean (US EPA 2007

Standards) engines• Older “dirty” engines• Sample Conditioning

System provides accurate dilution up to 1:2000

• NOx, CO2, CO, HC• CAN network

communication & RS232, GPS

Continuous TPM Measured withTPM Trace vs. Power: FTP Cycle

-1000

-500

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0 200 400 600 800 1000 1200 1400

Time(s)

Con

cent

ratio

n (u

g/m

3)

-100

-50

0

50

100

150

200

Pow

er(h

p)

The RPM-100 (QCM) in a Backpack and on a Caterpillar D11 Dozer with a 3508 V-8

MEMS & RPM 100

Gravimetric PM Comparisons Between the RPM-100 and

the Full-flow Dilution Tunnel

Test 01 Test 02 Test 03 MARI RPM 100 Integrated PM Mass 0.74 1.97 1.73

Full-Flow Dilution Tunnel Gravimetric Integrated PM Mass 0.71 1.75 1.79

Percent Difference 4.2% 12.6% -3.4%

Note: The RPM-100 sampled from the raw exhaust stream. The resultant error includes all sources of errors in the emissions measurement systems (exhaust flow rate; concentration; data acquisition; etc.)

Conclusions

• Re-visit the definition of particulate matter• Give serious consideration to fuel-specific

emissions measurements• Move all on-board emissions measurement

systems out of the truck cab and onto the exhaust stack – make accurate and precise measurements more vehicle driver/owner friendly.

• New, accurate sensors are available• Need to focus on measurement of emissions

(species and concentrations), which will be encountered in 2007.

West Virginia University,Morgantown,WV 26506

West Virginia University,Morgantown,WV 26506

The End

Engine Speed Histogram for the Candidate Routes Mack CH Tractor and Trailer

With a Nominal 60,000 lbs GVW

0

10

20

30

40

50

Eng. Spd. (rpm)

Per

cent

age

of T

ime

(%)

SAB2SAB 16.22 1.69 1.49 3.42 12.86 24.30 31.43 8.59

SAB2SW 9.99 1.23 1.80 1.14 5.81 21.95 43.17 14.90

SW2SAB 10.50 1.97 3.34 2.20 7.19 14.42 45.40 14.98

SAB2BM 9.81 0.53 1.96 1.69 8.68 16.77 45.31 15.27

BM2SAB 10.86 1.83 1.51 2.20 10.31 18.48 32.86 21.93

WAHSPA1 17.04 1.87 2.12 3.79 12.35 28.13 25.14 9.57

WASHPA2 20.46 2.87 3.40 4.14 10.79 30.78 19.10 8.45

WASHPA3 4.36 0.56 0.89 0.87 7.84 35.04 45.44 4.99

0 - 700 700 - 900 900 - 1100 1100 - 1300 1300 - 1500 1500 - 1700 1700 - 1900 1900 - 3000

ECU-Derived Engine Power Histogram for the Candidate Routes Mack CH Tractor and Trailer With a Nominal

60,000 lbs GVW

0

10

20

30

40

50

ECU Power (hp)

Per

cent

age

of T

ime

(%)

SAB2SAB 21.95 24.97 6.62 4.81 4.78 4.97 5.38 4.69 4.53 17.01 0.28

SAB2SW 19.56 15.18 3.62 2.40 2.14 2.48 3.16 3.34 5.26 42.29 0.56

SW2SAB 25.75 17.65 2.96 2.05 1.46 2.24 2.91 4.35 5.13 34.89 0.62

SAB2BM 21.96 14.13 2.40 2.69 2.61 2.60 2.59 2.89 2.86 44.36 0.90

BM2SAB 31.52 18.16 2.91 2.01 1.47 1.80 2.23 2.58 2.62 34.13 0.58

WAHSPA1 27.31 27.88 3.27 3.26 2.82 4.42 4.58 5.01 4.91 16.19 0.35

WASHPA2 23.72 29.83 4.55 4.08 3.47 4.88 5.43 5.47 5.08 13.18 0.32

WASHPA3 21.01 9.85 5.71 4.92 5.13 4.92 4.89 5.70 4.55 32.96 0.36

-200 - -25 -25 - 25 25 - 75 75 - 125 125 - 175 175 - 225 225 - 275 275 - 325 325 - 375 375 - 425 425 - 475

Inference of Engine Power

0

100

200

300

400

500

600

500 1000 1500 2000 2500 3000Engine Speed (rpm)

Shaft

Tor

que (

ft-lb)

0

20

40

60

80

100

120

ECU

Perce

nt Lo

ad (%

)

LugCurve

Lug Curve Percent Load

Curb Idle Load

Curb Idle Percent Load

Ratio of bsNOx/bsCO2 vs. Power

0.000

0.005

0.010

0.015

0.020

0.025

0.00 100.00 200.00 300.00 400.00 500.00

Power(hp)

Rat

io (N

ox/C

O2 )

0.000

0.500

1.000

1.500

2.000

2.500

3.000

3.500

4.000

4.500

bsN

Ox

(g/b

hp-h

r)

Ratio on road

Ratio from FTP cycle

bsNOxStandard bsNOx

0.00628

0.00864

4.0 g/bhp-hr

0

400

800

1200

1600

30 60 90 120 150 180Time (s)

Shaft

Tor

que (

ft-lb)

LaboratoryMeasured

ECU Inferred,LaboratorySupplied

Engine Torque

0

500

1000

1500

2000

0 100 200 300 400Time (s)

Engin

e Spe

ed (r

pm)

LaboratoryECU

Engine Speed

Results – Steady State Instantaneous Engine Speed and Torque

• Constant 1500 rpm• 10 Second Period • Torque Varied From 325 ft-lb to Maximum•

0%

0%0.1%

0.1%

0.2% 9.7% 13.1%

4.7%