BAE, Choongsik - OSTI.GOV

33
BAE, Choongsik Engine Laboratory Department of Mechanical Engineering, KAIST KAIST Korea Advanced Institute of SoleAte and T>: t hnoingy Cerc

Transcript of BAE, Choongsik - OSTI.GOV

BAE, ChoongsikEngine Laboratory

Department of Mechanical Engineering, KAIST

KAIST Korea Advanced Institute of Sole Ate and T>: t hnoingy Cerc

Backgroundy

Objectivey

Experimental Resulty

Conclusiony

ackground Homogeneous Charge Compression Ignition

nn

h#

Diesel Engine(compression ignition}

Gasoline Engine(homogeneous premixed charge)

HCCI Engine(Homogeneous Charge Compression Ignition)

3

Background - Advantages & Disadvantages

□ Advantages and disadvantages of HCCI engine

HCCIEngine

Advantages

Usage of the different type of fuels

Ultra low NOx & PM emissions

Improved fuel economy

Excessive combustion rate

Engine noise

HC and CO emissions

HCCIEngine

Disadvantages

r V

Increase of Negative

work

Y

Ja

I—BMBIH rj Co IHW

Background - Research Fields of HCCI Engine

HCCI Engine research

i

Combustion control

Direct injection

EGR and internal EGR

Compression ratio

Intake charge temperature

1

Fuel characteristics

r Gasoline

DME + —LPG

Hydrogen

Methanol and Ethanol

6

Anti-knock property

Extend high load limit

LPC.Caolinc.Hydrogen

y

LPG

-42.07 Boiling temperature (°C) -0.5KC\A 1 nnitinn lomnarafnro

•-v v v

114 Octane number 91

Dual-Fuel HCCI CombustionBased on SI Engine

- LPG Port and DME Direct HCCI- Comparing Propane and Butane

Experimental apparatus

Nitre gen

*1

Nitroga

i L

DME

LPG Proyamm able ECUu

-tSs*'

0

Programmable ECU

Proyamm able ECU

Comb u*tion Pm: smj i c Trait

rLdmbtiu meter (LA4)

LxKdutlPressure ^jaye

Exhaust Gas Emission

(MEXAl 50CD)

Number of Cylinder 1

Bore (mm) 82

Stroke (mm) 93.5

Displacement volume (cc)

493.7

Compression ratio 13.0

RPIH 1000

Intake Valve Open (CAD)*

1

AC Dynamometer

9

Mot

ored

in-c

ylin

der p

ress

ure

Experimental ConditionsiNESE LAB.

- LPG Injection Timing

DIME injection Timing

120 180 240 300

Crank angle degree (CAD)

Engine Speed(RPIM)

IQQQ

DME injection Quantity

(J/stroke)

413677

LPC injection 276Quantity 386

(J/stroke) 448

LPC composition PropaneButane

Experimental Result and Discussion

Output Performance - Indicated Mean Effective Pressure (IMEP)

DIKE combusted with Propane, Butane

DME injection: 413 J/stroke, at Optimal timing (240 CAD)

IMEP was decreased In spite ofincrcascd Injection propane quantity

280 320 360 400 440

LPG injection quantity (J/stroke)

• The maximum IMEP was increased around 360 kPa• Average IMEP was higher when propane was used

PropaneButane

400 -

350 -

300 -

1

J

Experimental Result and Discussion

Output rcrfonnancc—lidlcatcd RHeai Effective Pressure (IMEP)

DIRE combusted wltlPmpanc, Butane

DIME injection: 677 J/strokc, at Optimal timing (240CAD)

Propane

400 -

350 -

300 -

LPG injection quantity (J/stroke)

W m

IMEP was kept increasing as addition ofLPG

Si H

• The maximum IMEP was increased around 400 kPa

• Average IMEP was higher when propane was used 12

Experimental Result and Discussion

Output rcrfonnancc—lidlcatcd RHeai Effective Pressure (IMEP)

8.0

7.0ro

CL 6.0

2 5.03V)V) 4.02

Q_t— 3.0d)"Oc 2.0

o 1.0c

0.0

Increased ignition May ■> Portion of positive work is increased

6z

Propane : 276 J/stroke 'T TTTT Butane: 276 J/stroke o, J/stroke

13 J/stroke

330 Soppressed entire combustion reaction -> Total amount of work is decreased

380 390

• AdditLotV-hi i ru ■ inirucKjuu ±yi u.u.ui i uuiay a uuimjckjuu iii am iiiu iii pr hmmlv u

• Better anti-knocking property of propane -> More suppression of combustion

P

13

Experimental Result and Discussion

Output Performaucc - Induced Mean Effective Pressure (IMEPI

DIME: 677 J/stroke, injected at Optimal timing (240 CAD)

Propane

6.0 -

4.0 -

2.0 -

— Propane : 276 J/stroke- - Propane : 386 J/stroke

Propane : 448 J/stroke

Crank Angle Degree (CAD)

Butane

— Butane: 276 J/stroke— - Butane: 386 J/stroke— Butane: 448 J/stroke

Crank Angle Degree (CAD)

• Enough amount of DUE -> no suppression of entire reaction

-> increased ignition delay & maximum pressure

14

Experimental Result and Discussion ENGINE LAB.

Output Performance - liductd Mcai Effective Pressure (IMEP)

8.0

03CL

7.0 -

6.0 -

§ 5.01-(/)P 4.01-

o) 3.01-"OI 2 0 I-

i.or0.0

330 34C

Enough ^

Increased ignition delay, maximum pressure . maxkamn pressure rise rate

"TOrJjijjj ^ Jitiiil iJjjjJiuiJLUJ

— propane : z/o j/stroKe- - Propane : 386 J/stroke £ 1.0

TOran^TTrrl7strokeButane: 386 J/stroke

hk'strokeBO 390

Severe noise and knocking occurs

9-> increased ignition delay & maximum pressure

Summary - DME HCCI combustion in Single Fuel Engine

The effect of LPG composition in a DME-LPG dual-fueled HCCI engine at various injection quantity and injection timing was observed

1. In a DME HCCI engine, higher load limit was extended by using LPG as an ignition inhibitor

2. If injection quantity of LPG exceeds certain level compared to injected quantity of DME, IMEP was decreased

3. Propane was more effective way to increase IMEP in this study

16

Summary - DME HCCI combustion in Single Fuel EngineENGINE LAB.

Propane

Dual-Fuel HCCI CombustionBased on SI Engine

- Gasoline + DME HCCI- LPG + DME HCCI

Experimental apparatus

Pressure regulator

n

DME supplying system

Combustion

Pressure Transducer

DME direct injection system

Programmable ECU

fuel injection system

■ Lambda meter (LA4)

1----------JExhaust Pressure gauge

Exhaust Gas Emission

Analysers

(MEXA1500D)

LPG port fuefInjection system

Experimental conditions

□ Experimental conditions

Engine speed, rpm 1000

IVO [ATDC] -29, -19, -9, 1, 11

DME injection timing [ATDC]

110

^TOTAL2.12, 2.41,2.57,

2.77, 2.91

^DME 3.7

Intake air temperature, °C 30

Coolant/Oil temperature, °C 80/80

IVO timings

o.io -

0.05 -

EVC timing

-35-30-25-20-15-10 -5 0 5 10 15

DME injection timing

50 100 150 200 250 300 350 400 450 500

Crank angle degree [CAD]

20

IMEP of HCCI engineENGINE LAB.

IMEP in Gasoline HCCI [bar]

IreT3-Q

Ere

2.9

IVO timing

IMEP in LPG HCCI [bar]2.9

-20 -10 0 10IVO timing

Maximum IMEP region IMEP was decreased due to negative work

Heavy knock region imep W',

Lam

bda

tota

lC02 emissions of HCCI

The difference of C02 emission [%]

-20 -10 0 10IVO timing

emission of than that of Blasolinelow carbon contained

high octane number

HCCI was HCCI

Handmi

Difference = LPG

Theemission

difference

2.

wasincreased

as theIVO timing

retardedand

Bwasdecreased

2.

- gasoline

2.La

mbd

a to

tal

Lam

bda

tota

l

Summary - HCCI combustion in Dual Fuel EngineENGINE LAB.

LPG-DME HCCI engine

Combustion characteristics

Gasoline-DME HCCI engine

Combustion characteristics

1. The operating range of LPG-DME H gasoline-DME HCCI engine due to hi octane number.

CCI engine was wider than that of gher latent heat of vaporization and

2.The C02 emission of LPG-DME HCCI engine was lower than that of gasoline-DME HCCI engine.

23

Dual-Fuel HCCI CombustionBased on Cl Engine

- Hydrogen Port and DME Direct HCCI

Intake air diesel engine

| Common-rail injection system (igniter)ruvi injection type | | CNG injector (port fuel injection)

25

Experimental setup - Hydrogen injector

Hydrogen injection system

Hydrogen injector: Bosch CNG injector

- Peak and Hold (2A f 0.5A)- Supply voltage : 6 - 16V- Operating temperature : -20 - 70°C

Hydrogen injector: 2 Ch CNG injector driver

- Operate up to 2 injectors at once- Supply voltage : 220V

Operating conditions - Effect of DME Injection Timing

Fuel DME / H2

Engine speed 1200 rpm

Compression ratio 14.8

Load 0.2 MPa

Injection timing of DME 40-2 CAD BTDC

Injection timing of H2 360 CAD BTDC

Injection quantity of DME 5mg(fixed)

Injection quantity of H2 varied(meet 0.2MPa IMEP)

Coolant temperature 80°C

Intake air temperature 30°C

27

HR

R[J

/CA

]Experimental results - In-cylinder pressure and HRR

Engine speed: t1 intake 'IMEP:

1200rpm 30°C 0.2 MPa

• DME injection Before BTDC 10 CAD

CtenkAncJe[GAE|

- In-cylinder pressure and HRR were increased with retarding DME injection

- LTR and HTR were observed

DME injection After BTDC 10 CAD

- In-cylinder pressure and HRR were increased with advancing DME injection

- only HTR was observed

28

DME injection timing [CAD BTDC]-40 -35 -30 -25 -20 -15 -10 -5 0

1200 -

- 4000

- 3000 T

♦ - 2000 >=

- 1000

-40 -35 -30 -25 -20 -15 -10 -5 0

DME injection timing [CAD BTDC]

• CO and HC were not burned when DME was injected before BTDC 30 CAD due to low combustion temperature

• After BTDC 10 CAD DME injection CO and HC emissions started increase

29

Exper

DME injection timing [CAD BTDC]

■40 -35 -30 -25 -20 -15 -10 -5 0

- 90 E

• NOx Emission was maximum when DME is injected at BTDC 10 CAD

• At BTDC 10 CAD DME injection, Heat release rate was Maximum

30

1. Combustion phase was retarded with higher portion of Hydrogen due to consumption of OH radial.

2. DME injection timing can control the ignition timing of Hydrogen-DME HCCI combustion.

3. With DME injection before BTDC 10 CAD, Low temperature reaction and high temperature reaction were observed.

4. With DME injection after BTDC 10 CAD, only high temperature reaction was observed

5. CO and HC were not burned when DME was injected before BTDC 30 CAD due to low combustion temperature

6. NOx Emission was maximum when the HRR was maximum.

7. The strategies to increase the combustion temperature such as fuel stratification are needed to improve the combustion efficiency and combustion stability.

31

Summary

Single fuel HCCI High cetane „ IMEPincrease gy^aust gas recirculation

Early

Fuel (DME) Late burn(Dilution, Heat capacity)

£Ignition liming

vCombustion phase control

» Increased IMEP

High Octane Fuel (LPG, Gasoline,

Hydrogen)

Dual fuelBDC

tLimited by knocking

TDC

Direct injHtion timingHCCI PCCI

Late

Conventional ClLocal rich region and temperature drop

UDoarote ^®oo8