Multicylinder Diesel Engine for Low Temperature Combustion ... file1 2008 DEER Dearborn, MI August...

20
1 2008 DEER Dearborn, MI August 4-7 Multicylinder Diesel Engine for LTC operation Acknowledgements: DOE LTC consortium project, Low Temperature Combustion Demonstrator for High Efficiency Clean Combustion (DE-FC26- 05NT42413) . Industrial Partners: UCB, LLNL, Ricardo, Siemens, ConocoPhillips, BorgWarner, Mahle. William de Ojeda Phil Zoldak, Raúl Espinoza, Chunyi Xia, Dan Cornelius Navistar, Inc Raj Kumar University of Windsor Diesel Engine Development DOE DEER CONFERENCE Dearborn, Michigan August 4-8, 2008

Transcript of Multicylinder Diesel Engine for Low Temperature Combustion ... file1 2008 DEER Dearborn, MI August...

1

2008 DEERDearborn, MI August 4-7

Multicylinder Diesel Engine for LTC operation

Acknowledgements: DOE LTC consortium project, Low Temperature Combustion Demonstrator for High Efficiency Clean Combustion (DE-FC26- 05NT42413) .

Industrial Partners: UCB, LLNL, Ricardo, Siemens, ConocoPhillips, BorgWarner, Mahle.

William de OjedaPhil Zoldak, Raúl Espinoza, Chunyi Xia, Dan Cornelius

Navistar, Inc Raj Kumar

University of Windsor

Diesel Engine DevelopmentDOE DEER CONFERENCE

Dearborn, MichiganAugust 4-8, 2008

2

2008 DEERDearborn, MI August 4-7Goals and Objectives

• Demonstrate the application of low temperature combustion to:– Yield 2010 NOx and Soot in-cylinder emissions– Study is carried out on the Navistar 6.4L engine using today’s Diesel fuel– Target load 12.6 bar– Improve engine thermal efficiency

• Develop technology capable for production implementation.

2008 DEER

DevelopPredictive Capability

KIVA-CFD

CFDLLNL

Cylinder Pressure Feedback

PHASE IIIMulti-Cylinder Steady State Optimization

Dec 06June 06 Dec 07June 07 Dec 08June 08

Transfer Engine to Navistar

PHASE IV Control:Transient Response

Dec 05

PHASE IEngine Design

PHASE IIEngine Build

HCCI on 1 cyl

RICARDOBURR RIDGE

Combustion System OptimizationCombustion PhasingInjection StrategyCR sweepAir System

VVA testing

June 09

LTC2010 Emissions

AIRSystem MatchingBorg Warner, Mahle

Injection System SelectionSiemens

Misfire DetectionKnock SensorIon sensorCyl Pressure

ECU developmentIn-Cyl PressureCombustion Diagnostics

HP/LP EGR loop

Fuel Efficiencyimprovements

2008DEER

Fuel testingFuelselectionConocoPhillips

CN rangeand

FACE FUELS

2007DEER

Milestones(0.2gNOx engine out) BMEP~6bar 12bar 16bar

3

2008 DEERDearborn, MI August 4-7Goals and Objectives

Improve Thermal Efficiency

Optimize TC system / operationReduced back pressure

Optimize CR and ERVVA

Supporting work:

Engine Testing

1-D Cycle SimulationKIVA-CFD Modeling

Detail Energy BalanceDisc

usse

d He

re

2007baseline

Optimizedcombustion

40%

50%

55%

AirSystem

CoolingSystems

Parasitic Losses

DOE 2010

DOE 2013

Electrification Waste HeatRecoveryAdvanced

Platforms(VVA, LP EGR)

Fuels

Short combustion durationCombustion Phasing

Injection strategy (PPCI)VCR

Injector Bowl MatchInjection pressure

Emission Driven

4

2008 DEERDearborn, MI August 4-7Engine Build

Baseengine

V8Test Engine SCTE

Displac.BoreStroke

6.4L98.5mm105mm

6.4L98.5mm105mm

0.75L95mm105mm

FIE DICommon Rail

DICommon Rail

DICommon Rail

CR 16 12-16.5 15-16.5

Turbo Charger

Single StageVNT

Dual StageVNT Surge tank

EGR system HP loopSingle Cooler

HP loopDual Cooler cooled

IVCEVO

-133 BTDC132 ATDC

-133 BTDC132 ATDC

-133 BTDC132 ATDC

CAC

Engine

EGR

TexhPexh

Tw

TCAC

Pim Tim

Mair = f(DP)Tamb Pamb

Pc2 Tc2

T1

Tstack

Pstack

Hea

terT3T5T7

T2T4T6T8

TBP

CO2

CO2

CO2

SmokeNOxHCCOCO2

N

V(reg)

Tcharge Pcharge

P2P4P6P8

P1P3P5P7

N

emissions

IVC

IVC

FIE

FIE

EGR

PegrTegr

PegrTegr

5

2008 DEERDearborn, MI August 4-7Definition of LTC

0

0.2

0.4

0.6

0.8

1

0 2 4 6 8 10 12 14Exhaust O2 Concentration [%]

Indi

cate

d N

Ox

[g/k

W-h

r]

0

0.05

0.1

0.15

0.2

Indi

cate

d S

oot [

g/kW

-hr]

IMEP 7bar NOxIMEP 9bar NOxIMEP 7bar SootIMEP 9bar Soot

80 70 60 50 40 30EGR% ratio

Low Temperature CombustionPCI

Representative data from SCTEIMEP 7, 9 barN 1200 rpmPintake 2 barMf 28, 33 mg/stke

Ref. AVL

HCCI

HCCI: Lean, Homogenous (Φ ~ 0.2), Temp controlled ignition

LTC: Similar to Conven’l Diesel Diffusion (Φ ~ 1 -1.2) Very-high EGR (~60%)

PCCI: Improved mixture, increased ignition delay Higher injection pressuresFuel injection timing (closer to TDC)Less dependent on very-high EGRFurther bowl – injector matching

[Herzog et. al 1992][Akihama et. al 2001]

6

2008 DEERDearborn, MI August 4-7

Engine Mapping Optimization

02

46

81012

1416

1820

0 1000 2000 3000 4000rpm

BM

EP (b

ar)

Work in ProgressExtend LTC rangeFocus of Presentation

PCCI - LTCInjection pressureFuel injection strategyTemperature management

LTC - High EGRorHCCIMultiple shotsMedium EGR rate

Pres

ent R

ange

of te

stin

g DOE TARGET LINE

2007 Lug Line

Target of Program:Implement LTC trough 12.5 bar BMEP on a production platformMeet NOx and SOOT 2010 targets without after-treatmentPromote BSFC to meet 2007 levels and wok towards DOE targets of 50 to 55% efficiency.

7

2008 DEERDearborn, MI August 4-7

Combustion System Optimization Approach

PCCI – LTC BMEP load gains:

Fundamental understanding1-D Thermodynamic Models

CFDKIVA

Multiple ShotROI and Spray models

LTC

Temperature EQR

Emerging Technologies

Fuel Properties

Turbo SystemsCoolingFuel Injection EquipmentVVA

Diesel SurrogatesFACE Fuels

Engine Testing

8

2008 DEERDearborn, MI August 4-7

Combustion System Optimization Highlights

Injection strategy – Single ShotEffects of early timing vs. load increaseEffective NOx controlModeling and experimental correlations

Transition to Multiple ShotsMultiple PilotsEffective soot control (safeguard bsfc)Modeling and experimental correlations

Air System optimizationEffects on Injection Strategy on BSFCCoordinate EGR / TurboAdvantages of VVA

Summary

Current StatusOperating at ~ 16 bar BMEP

Can run with 40 - 50% EGR Yield 0.2gNOx/bhp-hrReasonable soot (0.05-0.1g/bhp-hr)High combustion Efficiency

Next steps Combustion optimization towards BSFC goals

Take a piece-wise approach (vs one-step to 50%)

FUP ~ comb durationFuel specific formulations (e.g. non-sooting)

Highlights

9

2008 DEERDearborn, MI August 4-7

Positive Impact of Early Injection (early PCCI)Single Shot Injection

Injection Strategy Combustion Phasing Optimization

Path to reduce in-cylinder PM• Optimize piston-injector match• Increased boost• Increase ignition delay (premixed burn)

- Increased injection pressure- Early injection timings

Path to reducing flame temps• Dilute charge O2 concentration (AFR)• Raise heat capacity of mixture (EGR)• Reduce fuel rich zones (strength)

CA earlyCA late

Emission Driven

CA late

CA early

Multi-Cylinder Engine Testing

Focus on Combustion mechanism at 1750 10.7 bar reducing NOx and SOOT with injection timing

At higher loads:- Multiple injections- Lower combustion temperatures

10

2008 DEERDearborn, MI August 4-7

Positive Impact of Early Injection (early PCCI)1. The local chemistry shows leaner and cooler combustion.2. The early injection aids to better entrainment and vaporization of the fuel.3. The rate of combustion is more rapid.

Injection Strategy Combustion Phasing Optimization

1750 rpm – 10.7 bar BMEPKIVA vs. Engine Data

Early SOI

Late SOI Line

Late SOI Line

1750 rpm 10.7 bar BMEP

-100

0

100

200

300

400

-30 -20 -10 0 10 20 30

Crank Angle (Ѳ)

Hea

t Rel

ease

(Jou

les/

deg)exp

sim

exp

sim

Early InjectionImage corresponds to Peak in HR trace

Temp Φ

Late Injection

Temp Φ

11

2008 DEERDearborn, MI August 4-7

Positive Impact of Early Injection (early PCCI)

Injection Strategy Combustion Phasing Optimization

1750 rpm – 10.7 bar BMEP

1750 rpm 10.7 bar BMEP

-500

50100150200250300

-30 -20 -10 0 10 20 30CAD

HR

(J/d

eg)

-5 +6

Simulations highlight the impact of injection timing in the throughout the combustion process including, ignition delay, premixed burn, mixing control burn.

Point of max hear release

rate

Late injection timings show cells longer residence time within the NOx formation island.

Temp φ

12

2008 DEERDearborn, MI August 4-7

162

164

166

168

170

172

174

176

0.16 0.18 0.20 0.22 0.24 0.26 0.28 0.30

NOx(g/hp-hr)

BSF

C (g

/hp-

hr)

1

2

3

4

5

6

Injection Strategy Transition to multiple-shot

0.05

0.10

0.15

0.20

0.25

0.30

0.16 0.18 0.20 0.22 0.24 0.26 0.28 0.30

NOx(g/hp-hr)

Smok

e (g

/bhp

-hr)

MainPilot+Main (35ppm NOx)Pilot+Main (32 ppm NOx)Pilot+Main+Post (32 ppm NOx)Retarded+High Injection PressureSoot limitNOx limit

1

2

3

4

5

6

EGR

1-23456

Combustion optimization is driven by emission requirements. Engine Data at1750 rpm – 10.7 bar BMEP

Next focus on pilot strategy

Options become reduced when fuel efficiency is to be preserved or improved.

The injection strategy can be tailoredto reduce soot and NOx.

13

2008 DEERDearborn, MI August 4-7

Injection Strategy Pilot-shot optimization

Optimization of pilot quantity

Pilot quantity is evaluated via soot and NOx tradeoff with respect to BSFC:

• Pilot quantity can reduce SOOT at constant fueling and NOx.• The effect of pilot is sensitive to a minimum quantity• Injection pressure and combustion phasing may weigh in to

further optimize the present tradeoff without excess penalty to fueling

• It is effective to reduce the max rate of pressure rise (A vs. B)

HC and CO with respect to BSFC• Combined contribution appears to be minimum vs. BSFC

A

B

Engine Data at1750 rpm – 10.7 bar BMEP

A B

14

2008 DEERDearborn, MI August 4-7

Injection Strategy Pilot-shot optimization

Optimization of pilot quantity

Simulations captured the ignition characteristics of the pilot events.Pilot liquid phase (A) vaporizes prior to main injection eventResults corraborate the BSFC – (HC,CO) balance from previous slide

1750 rpm – 10.7 bar BMEPSIMULATION vs. Experiments

pilot main

(A)

Next Slide

Cylinder Pressure

Heat release trace

Detail of Injection trace and fuel phase composition

Heat release trace

15

2008 DEERDearborn, MI August 4-7

Injection Strategy Pilot-shot optimization

1750 rpm – 10.7 bar BMEP14 mg pilot

SIMULATIONPilot mixture preparation

318

320

322

325

Temperature EQR

Pilot injection vaporizes

High temperaturePilot Combustion begins at 355º

Low temperature combustionBased on temperature distribution and heat release curve shows initial combustion at 340º

335

340

348

355

Start of main injection358

359

362

Suppression of the heat release from the main injection fuel occurs for approximately 4º

Temperature EQR

364

365

366

367

Main reaction take placePilot appears to ignite the main spray

0

50

100

150

200

250

300

-30 -20 -10 0 10 20 30

Crank Angle Degree

HR

R (J

/deg

)

16

2008 DEERDearborn, MI August 4-7

Injection Strategy high load testing

1670 rpm – 16 bar BMEP

140145150155160165170175180185190

0.1 1 10BSNOx (g/bhp-hr)

BS

FC (g

/bhp

-hr)

00.10.20.30.40.50.60.70.80.91

BS

OO

T (g/bhp-hr)

Baseline Engine

LTC@ <0.2gNOx

3%

5%

10-15%

17

2008 DEERDearborn, MI August 4-7

Air System Optimization Model Based Control

• Air system management coordinated EGR, ITH, VVA, VNT

• Fuel Injection management • In cylinder diagnostics • Modeling and simulation • Controls Rapid Prototype Systems

Systems Approach

2250 rpm - BMEP 9 barMultiple shotStrategies

Single shot

BSFC to DP:1% ~ 10kPa

18

2008 DEERDearborn, MI August 4-7

Air System Optimization VVA design and Impact on Engine Performance

VVA advantages:• Improve volumetric efficiency at low

speeds and loads• Improve emissions by increasing ignition

delay and reducing in-cylinder temperatures

• Maximize efficiencies and PCCI regions by using Miller-type cycles at part loads

BSFC improvements ~ 4%

FSN improvement islands ~ 2 FSN

Volumetric efficiency gains

Electro-hydraulic Intake Valve System control installed on the

Navistar 6.4 L V8 engine.

19

2008 DEERDearborn, MI August 4-7Summary

• Applied low temperature combustionTo target 2010 NOx engine out emissions using today’s Diesel fuel without active aftertreatment.

• Approach focused on the combustion system primarily looking to optimizea. The fuel injection strategy to favor premixing fuel into the charge cylinder mass. b. Optimize the operating boundary conditions, such as in-cylinder temperatures, EGR / in-cylinder O2 content.c. Improve brake thermal efficiency.d. Rely on CFD to understand behavior of with pilot injection (prediction of ignition delay, heat release)

• LTC was achieveda. With EGR, temperature management for BMEP load levels of 6 bar.b. With EGR, high injection pressure and early injection timing for loads from 6-12 bar.c. Transitioning ot a multi-shot strategy at 12-16 bar load levels.

• The technology gathered is capable for production implementation

• Future work will examine a. The impact of VVA in the engine emissions and BSFC.b. The effects of a variety of fuel formulations will be tested in this platform.

20

2008 DEERDearborn, MI August 4-7Acknowledgements

Project Partners

TM

Lawrence Livermore National Laboratory

CFD Fuels Enabling Technologies Combustion Diagnostics