Application of Particulate Filter Technology to a CNG Engine

Jonathan HarrisWestport Innovations Inc.

Introduction

• Project had three main areas of work;– Investigation into the effect of the addition of a

Particulate Filter (PF) on the emissions from a CNG-fueled Heavy-duty engine.

– Investigation into the impact of lubricants on emissions from CNG engines.

– Assess emission-reduction potential, performance, and reliability of the catalyzed particulate filters during a six-month in-use demonstration program.

• Funded by SCAQMD, Sempra Utilities and NREL.

Introduction

• Project was lead by Westport Innovations with support from the following;– West Virginia University (WVU)

• Chassis Dynamometer facilities and regulated emissions

– University of Wisconsin (UWI) • Chemical speciation

– University of Minnesota (UMN) • Aerosol concentrations and size distributions

– SunLine Transit Agency • Field Trial Partner

Contents

• Suitability of Particulate Filters for use with CNG engines

• Aftertreatment Configuration Selection• Testcell Setup and Investigation • Vehicle Installation• Chassis Dynamometer Testing and Results• Field Trial Results• Conclusions

PF Suitability for CNG Engines.

• Composition of PM from CNG engines is significantly different to that from Diesel Engines

• CNG PM in general has: – Lower engine out PM levels – Larger fraction of Ash and VOF – Lower elemental carbon levels

• For Lean Burn SI (LBSI) engines– Higher NOx to carbon ratio (vs. diesel)– Higher minimum exhaust temperatures (vs. diesel)– O2 in exhaust stream

• LBSI engine exhaust conditions should be favorable to PF continuous regeneration

Selected Engine

• Engine selected is the CWI C Gas Plus. • Launched in 2001• Lean-burn Spark Ignited (LBSI) engine.• Turbocharged • Fuel metered into intake (fumigation)• Air to Fuel (AFR) control via oxygen sensor in

exhaust stream. • Drive-by-wire throttle.

Selected Engine

• The CWI C Gas Plus is certified to the following emissions levels:– US EPA 2004– EPA CFFV ULEV– CARB Optional Low NOx (1.8 gm/bhp-hr)– Euro 3 (with THC specific catalyst)

• Certification PM emissions < 0.01 g/bhp-hr • Peak Torque 850 lb-ft (1153 Nm) at 1400 rpm • Rated Power 280 HP (209 kW) at 2400 rpm.

Aftertreatment Configuration Selection

• Selection to find best combination for CNG operation.

• 2 PF’s selected:– Engelhard DPXTM [Catalyzed Diesel Particulate Filter]– Engelhard PF [Uncatalyzed Particulate Filter]

• 2 Oxidation Catalysts selected:– Standard C Gas Plus formulation– HEX-1107 [predominantly Palladium]

Testcell Setup•Testcell located at Westport Innovations in Vancouver, Canada.

•PM emissions measured using Sierra BG-2 Mini-dilution tunnel.

•Horiba emissions bench used for gaseous emissions measurement.

•Modular exhaust system allows for combinations of PF and oxidation catalyst to be tested.

Testcell Investigation

• Engine was base-lined for engine out emissions and with standard OEM Oxidation Catalyst fitted.

• Various aftertreatment configurations fitted and compared by:– Measured PM mass emissions.– Effect on gaseous emissions.– PF mass increase over 100 hours.

• Results allowed selection of configuration for field trial.

Testcell Investigation

• Selected Configuration combined in series:– Standard C Gas Plus Oxidation Catalyst + Uncatalyzed

Engelhard PF PM Mass flow AVL 8 Mode

0

0.5

1

1.5

2

2.5

3

3.5

4

Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6 Mode 7 Mode 8

AVL Mode

Mas

s flo

w g

/h

Standard C+ CatStd C+ & DPF2

Vehicle Installation

• SunLine Transit supplied a 40-ft transit bus fitted with a CWI C Gas Plus.

• Aftertreatment system fitted for a 6 Month Field Trial in revenue service.

Vehicle Installation

• Available space for packaging was limited in this chassis.

Oxidation Catalyst

Particulate Filter

Engine

Chassis Dynamometer Testing

• “Real-world” testing of selected Aftertreatment system completed on WVU portable Chassis Dynamometer.

• “Clean” CVS tunnel (reserved for low-emitting applications) was used for this study.

Chassis Dynamometer Testing

• Test cycles used:– 3 steady state modes (idle, 20 mph, 40 mph)– Quad Central Business District Cycle (QCBD)

CBD Cycle Profile

Instrumentation

• UMN (physical characterization)– Condensation Particle Counter (CPC)

• to measure total particle number concentrations]

– Engine Exhaust Particle Spectrometer (EEPS)• Aerosol size distributions

– Scanning Mobility Particle Sizer (SMPS)• 10 to 300 nm for steady state testing only

• UWI (chemical characterization)– Media was collected on-site and processed once

testing was completed.

Chassis Dynamometer Results• Gaseous emission results

HC Mass Flow

0

50

100

150

200

250

300

350

400

Idle Run 1 Idle Run 2 20 MPH 40 MPH

Mas

s Fl

ow [g

/hou

r]

OC only.

OC + PF

CO2 Mass Flow

0

10000

20000

30000

40000

50000

60000

Idle Run 1 Idle Run 2 20 MPH 40 MPH

Mas

s Fl

ow [g

/hou

r]

OC OnlyOC + PF

NOx Mass Flow

0

50

100

150

200

250

300

Idle Run 1 Idle Run 2 20 MPH 40 MPH

Mas

s Fl

ow [g

/hou

r]OC OnlyOC + PF

• PF does not adversely affect gaseous emissions from engine.

Chassis Dynamometer Results• PM mass emissions

PM Mass Emissions on a g/mile basis from WVU measurements.

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

OC Only OC + PF

PM g

/mile

Run 1

Run 2

Run 3

Chassis Dynamometer Results• Particulate Filter reduces particle number.

1.0E+00

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

0 200 400 600 800 1000 1200 1400 1600

Elasped Time, s

CPC

Num

ber,

par

ticle

s/cm

3

End-Of-Life, OCEnd-Of-Life, OC+PF

Plot of CPC number concentration for the average quad-CBD cycle for the end-of-life oil condition, with the OC and with the OC and PF together.

Chassis Dynamometer Results• Start up spikes removed with addition of PF

1.0E+00

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

1.0E+06

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Elasped Time, s

CPC

Num

ber,

par

ticle

s/cm

3

New, OCMid-Life, OCEnd-Of-Life, OCEnd-Of-Life, OC+PF

Particle number measured during 40 mph tests, not corrected for dilution ratio.

Chassis Dynamometer Results• Distribution of particles unaffected by addition of PF

1.0E+10

1.0E+11

1.0E+12

1.0E+13

1.0E+14

1.0E+15

1 10 100 1000

Dp, nm

dN/d

logD

p pe

r mile

, par

ticle

s/m

ile

Oil: End-Of-Life, Aftertreatment: OC

Oil: End-Of-Life, Aftertreatment: OC + PF

40 mph cruise condition SMPS size distribution with and without the particle filter (PF).

Chassis Dyno Results Summary

• The addition of the PF does not produce a statistically significant change in PM mass emissions

• In many cases and under many conditions, particle number concentrations were not detectable above background.

• The addition of a particulate filter is effective at removing start-up spikes in particle number concentration

• By adding the particulate filter the size distribution is unchanged but the overall number is reduced.

Field Trial Results

• SunLine Bus 531 covered over 30,000 miles in 6 months. • No mechanical issues incurred. • Fuel consumption effect minimal.

0

20

40

60

80

100

120

Janu

ary

Febru

ary

March

April

May

June

July

August

Septem

ber

October

Novembe

rDec

ember

Janu

ary

Month

Tem

pera

ture

(deg

rees

F)

0.5

1

1.5

2

2.5

3

3.5

4

Fuel

Eco

nom

y (g

.g.e

.)

Daily Average High Temperature

Bus 531 Fuel Economy with PF

Bus 531 Fuel Economy without

Conclusions

• The experience of the 6-month field trial was positive with no serious issues encountered during operation.

• The addition of the Particulate Filter to the Aftertreatment system reduced the particle number emissions from the engine.

• During engine and vehicle dynamometer emissions testing, the addition of the Particulate Filter did not adversely affect the standard engine operation.

• The filter used in the 6-month field trial has been stored and could be examined (for ash content / composition etc.) further if a suitable opportunity arose.