On-Board Engine Exhaust Particulate Matter Sensor for HCCI and Conventional Diesel Engines

Matt Hall and Ron MatthewsThe University of Texas at Austin

DOE Project # DE‐FC26‐06NT42966Project Manager: Ralph Nine

U.S. DOE Merit Review Meeting, May 22, 2009

Project ID #ace_44_hall

This presentation does not contain any proprietary, confidential, or otherwise restricted information

2

Overview

• Project start date: 10/1/2006• Project end date: 12/31/2009• Percent complete: 80%

• Barriers addressed– “Develop new sensitive real-time

PM measurements diagnostics for developing engines with ultra low-PM emissions, especially for rapid transients, and for providing the engine-out emissions characterization needed for design optimization and life-cycle analysis of PM aftertreatment systems.”• Total project funding

– DOE share: $412,203– Contractor share: $392,305

• Funding received in • FY08 : $137,450• FY09: $136,745

Timeline

Budget

Barriers

• Interactions/ collaborations:• Emisense/Ceramatec Inc.• Cummins Engine Co.

• Project lead: UT-Austin

Partners

3

Study Objective

• Complete the development of an inexpensive, sensitive, accurate, and durable on-board particulate matter (PM) sensor, bringing it to a point where it can be commercialized and marketed.

Year 3 Objectives

• Continue development of PM sensor to further improve durability, sensitivity, and signal-to-noise ratio.

• Demonstrate PM sensor durability and response through on-board diesel vehicle studies.

• Explore sensitivity limits of PM sensor for ultra-low PM exhaust concentration levels upstream and downstream of DPF in 2008 model year Cummins 6.7 liter engine.

• Investigate velocity dependence of PM sensor signal and compensation techniques.

• Continue collaboration with Emisense/Ceramatec, Inc. to further the commercialization of the sensor.

4

5

MilestonesMonth/Year Milestone or Go/No-Go DecisionJune-08 Go/No-Go decision: Decision made to focus future development on foil-

electrode type sensors instead of wire electrode sensors. Foil-type sensors found to have greater sensitivity, less vibration noise, better durability.

February 2009 Milestone: Demonstration completed of higher durability (> 10s of hrs) PM sensor applied to diesel vehicle with validation of sensor sensitivity via filter measurements of PM emission concentrations from the 1.9 liter Fiat/Opel engine in the Chevrolet Equinox and with the opacity meter.

March 2009 Milestone: Compete set up of 2008 model year Cummins 6.7 liter diesel engine to begin ultra -low PM concentration testing of sensor with and without a DPF

6

Approach • New sensor designs that include active heating to

prevent soot fouling and different electrode sizes and shapes for enhanced sensitivity will be created in cooperation with Emisense/Ceramatec, Inc.

• May include proprietary configurations and installation geometries to minimize velocity effects on signal.

• Continue on-board vehicle testing to demonstrate durability and sort out any installation issues and velocity effects.

• Start ultra-low PM concentration level sensor testing and development with Cummins 6.7 liter engine, with and without DPF.

7

Background and Technical Accomplishments

Sensor Principle of Operation

• Soot particles in exhaust carrying a natural electric charge and are accelerated toward sensor electrode by a strong electric field and neutral particles are charged and detected.

• Rate of charge deposition on sensing electrode is proportional to PM content.

• Time resolution of 20 ms determined by charge amp electronics.

PM-Laden Exhaust

1000V

Charge Amplifying Circuitry

DAQ System

Metal ElectrodesInsulated Casing

Sensor electronics

Test platform for on-board vehicle studies:UT Austin Chevy Equinox Engine: 1.9 liter Fiat/Opel Turbo Diesel(with no exhaust after-treatment)

PM sensor and Opacity meter Installation

Instrumented for:PM sensor output voltage

Opacity meter output voltageVehicle speed (km/hr)Engine speed (rpm)

Pedal position (% of maximum)Engine Torque (N m)

PM sensor

Opacity meter

Test platform for ultra-low PM concentration level sensing with and without DPF:

2008 model year Cummins 6.7 liter engine

12 minute drive cycle in EquinoxPM sensor output, Opacity meter output, vehicle speed (km/hr),

Engine speed (rpm/100), Pedal Position (%)

0

1

2

3

4

5

-20

0

20

40

60

80

0 2 4 6 8 10

OpacitySensor Speed

Pedalrpm/100

PM s

enso

r and

Opa

city

met

er (v

olts

)

Pedal position, Vehicle speed, rpm/100

Time (min.)

Balances, sample bag, and filters used to calibrate PM sensor

Filter measurements of exhaust PM mass(Bag volume approximately 30 liters)

Bag contents forstrong tip-in

30 minutes continuoussampling at idle

Bag contents at idle

Sensor response and engine speed for an idle tip-in event

Gravimetric measured PM mass = 11.2 mg,yielding sensor sensitivity of 204 mg/m3 V

0

0.5

1

1.5

2

2.5

3

3.5

4

0

500

1000

1500

2000

2500

7 8 9 10 11 12 13

Sensorrpm

PM s

enso

r (v

olts

)Engine speed (rpm

)

Time (s)

Exploration of sensor resolution and sensitivity limitsSingle Cylinder Yanmar diesel engine

1500 rpm at a very low load of 1.5 Nm torqueSimultaneous filter measurement gave 17 mg/m3 dry PM mass,

yielding sensor sensitivity of 350 mg/m3 V

Engine shutdown

PM sensor, opacity and vehicle speed (km/hr) during drive cycle(No strong correlation of high emissions events with speed)

0

500

1000

1500

2000

0

10

20

30

40

50

60

70

80

3 4 5 6 7 8 9

Sensor Opacity * 10Speed

PM s

enso

r (m

g/m

3 )Vehicle Speed (km

/hr) and Opacity (volts *10)

Time (min.)

Calibrated PM sensor output (sensitivity = 200 mg/m3 V) compared with opacity meter output for 12 minute drive

(High emission events correlate well)

0

500

1000

1500

2000

-2

-1

0

1

2

3

4

5

0 2 4 6 8 10

Sensor

Opacity

PM s

enso

r (m

g/m

3 )O

pacity meter (volts)

Time (min.)

PM sensor output, opacity and pedal position during drive cycle(High emission events occur for pedal positions >40% of max.)

0

500

1000

1500

2000

-40

-20

0

20

40

60

80

4 4.5 5 5.5 6

Sensor

PedalOpacity * 10

PM s

enso

r (m

g/m

3 )Pedal position (%

max) O

pacity meter (volts *10)

Time (min.)

PM sensor and inverse of opacity meter(correlation between the two at lower PM levels)

0

200

400

600

800

1000

0.2

0.25

0.3

0.35

0.4

0.45

0.5

0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Sensor

1/opacity

PM s

enso

r (m

g/m

3 )1/O

pacity meter voltage

Time (min.)

0

200

400

600

800

1000

0.4

0.5

0.6

0.7

0.8

6 6.2 6.4 6.6 6.8 7

Sensor

1/opacity

PM s

enso

r (m

g/m

3 )1/O

pacity meter voltage

Time (min.)

PM sensor and Inverse of opacity signals(correlation differences for twin peak signals)

White smoke?

Black smoke?

PM sensor correlation with vehicle speed and engine speed

0

0.5

1

1.5

2

-20

0

20

40

60

80

6 6.5 7 7.5 8

SensorSpeed

rpm/100

PM s

enso

r (v

olts

) Vehicle speed (km

/hr) and rpm/100

Time (min.)

PM sensor correlation with vehicle speed and engine speed(sensor signal correlates with engine speed)

0

1

2

3

4

5

0

10

20

30

40

50

60

70

3.6 3.7 3.8 3.9 4 4.1

SensorOpacity

Speed

rpm/100

PM s

enso

r and

Opa

city

met

er (v

olts

) Vehicle speed (km/hr) and rpm

/100

Time (min.)

PM sensor output at different engine idle speeds(vehicle stationary)

0

0.5

1

1.5

2

2.5

3

3.5

4

0

1000

2000

3000

4000

0 0.2 0.4 0.6 0.8 1 1.2

Sensor

rpm

PM s

enso

r (v

olts

)Engine speed (rpm

)

Time (min.)

PM sensor output at different engine idle speeds (configuration to reduce exhaust velocity past sensor)

0

0.5

1

1.5

2

2.5

3

3.5

4

0

1000

2000

3000

4000

0 0.2 0.4 0.6 0.8 1 1.2

Sensor

rpm

PM s

enso

r (v

olts

)

Engine speed (rpm)

Time (min.)

27

Future Work

• Continue development of PM sensor to further improve durability, sensitivity, and signal-to-noise ratio.

• Demonstrate PM sensor durability and response through on-board diesel vehicle studies.

• Explore sensitivity limits of PM sensor for ultra-low PM exhaust concentration levels upstream and downstream of DPF in 2008 model year Cummins 6.7 liter engine.

• Investigate velocity dependence of PM sensor signal and compensation techniques.

• Continue collaboration with Emisense/Ceramatec, Inc. to further the commercialization of the sensor.

Summary

•UT PM sensor is capable of measuring time-resolved PM emissions from a diesel engine for steady-state and transient operation.•High PM emissions, in the range of 1000 mg/m3 were measured during hard acceleration and correlated with pedal positions greater than about 40% of maximum. •Good correlation between sensor and opacity was also found for much lower PM levels.•Previous Data suggest a PM concentration resolution of 3-4 mg/m3 dry mass is attainable and dynamic range of the sensor spans at least 2.5 orders of magnitude. •The effect of exhaust gas velocity on the sensor signal was evident by its response to changes in engine speed, both at idle and during driving.