GASOLINE COMPRESSION IGNITION – A PROMISING …
Transcript of GASOLINE COMPRESSION IGNITION – A PROMISING …
![Page 1: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/1.jpg)
GASOLINE COMPRESSION IGNITION – A PROMISING TECHNOLOGY TO MEET FUTURE ENGINE EFFICIENCY AND EMISSIONS TARGETS
drhgfdjhngngfmhgmghmghjmghfmf
STEPHEN CIATTIPrincipal Mechanical EngineerArgonne National Laboratory
2nd CRC Advanced Fuels and Engine Efficiency WorkshopWednesday, November 2, 2016Work funded by DOE Office of Vehicle Technologies – Gurpreet Singh and Leo Breton
This presentation does not contain any proprietary, confidential or otherwise restricted information
KHANH CUNGPostdoctoral FellowArgonne National Laboratory
![Page 2: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/2.jpg)
INTRODUCTION Advanced combustion concept: HCCI, PPC
or GCI utilizing LTC process to reduce soot formation, and avoid high NOx (more premixing and dilution) GCI has shown to have advantages over
traditional diesel compression ignition due to: low soot and emission, high thermal efficiency Initial hardware tends to influence desired
strategies and tradeoffs Many dependent variables (Pin, Tintk, mixture
stratification) on the ignition, combustion and the emissionRecently explored: Effect of multiple parameters (injection pressure, mixture composition, boost,
injection timing, and fuel reactivity) on engine outputs (ignition, combustion phasing, emission, etc.) using design of experiment approach
Distinguish the transition from “quasi” HCCI to GCI as injection timing is swept with constant air/fuel mixture for two different fuels
Kitamura, T., et al. International Journal ofEngine Research 3.4 (2002): 223-248.
![Page 3: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/3.jpg)
ENGINE SPECIFICATION4-cylinder light duty diesel engine (Euro IV
General Motors 1.9 L)Diesel fuel injection system (Bosch CRIP2)Eaton supercharger equipped for study with
stable boost pressure
1234
Supercharger
Uncooled EGR Loop
Cooled EGR Loop
Heat Exchanger
Exhaust Outlet
Ambient Intake Air (from LFE)
Bypass Valve
Heat Exchanger
Intake Air (+ EGR)
Exhaust
Turbocharger
Engine GM 1.9 L (4 cylinders)
Bore 82 mmStroke 90.4 mmConnecting rod 145.4 mmCompression ratio 17.8:1IVO 359 deg. aTDCIVC -149 deg. aTDCEVO 130 deg. aTDCEVC -357 deg. aTDCEngine speed 1000 RPM
2000 RPMIntake pressure (abs.)
1.15 to 1.55 bar
Intake temperature 48 deg. CInjectors Bosch
Numbers of hole 7Nozzle diameter 0.141 mmSpray angle 120 deg.
Injection pressure 400 to 600 bar
![Page 4: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/4.jpg)
Graphics courtesy ORNL (Curran & Dempsey)
OBJECTIVES OF WORK - MULTI-CYLINDER, HIGH EFFICIENCY GASOLINE COMPRESSION IGNITION
Current Specific Objectives:1. Evaluate effect of Low Pressure EGR upon auto-ignition
and engine performance characteristics 2. Quantitatively study effect of injection strategy upon auto-
ignition to develop approach for transient operation and reduced fuel sensitivity
3. Perform factorial experiments to quantify the effect of important input parameters upon engine performance, noise and emissions
4
Long-Term ObjectiveUnderstand the physical and chemistry characteristics ofGasoline Compression Ignition (GCI) in a multi-cylinder engine toaid industry in developing a practical high efficiency, lowemission combustion system
Mixing Limited GCIHCCI PFS
Majority Premixed GCI
Majority Stratified GCI
![Page 5: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/5.jpg)
INJECTION STRATEGY: E10 MINIMUM FUELING – FULL SOI SWEEP (-141 ATDC TO TDC)
5
SOI [deg. aTDC]
-140 -120 -100 -80 -60 -40 -20 0
Lam
bda
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Lambda
Minimum Fueling approach: least fuel requirement for stable combustion (COVIMEP < 3%)
Combustion mode (HCCI vs. GCI) characterized by SOIs
Discontinuity due to fuel rate changed
A-period Location of ignition (CA10), and combustion phasing (CA50) seems stay constant
B-period More fuel (smaller lambda) is needed to have stable combustion, but CA10/CA50 also seems constant
C-period IMEP shows a drop near -60 deg. aTDC due to possible fuel entering squish region, as indicated by simulation
A B C
Fuel in squish region
Both SOI and Lambda show to have effect on CA10/CA50, but lambda is more effective
There seems to be a condition with constant lambda to fix CA10/CA50
GM 1.9 L 17.8:1 (CR)Engine speed 1000 rpm
Injection pressure 400 bar
Injector-Bosch
7 hole 120 deg.
cone angleFuel E10
Constant lambda at some SOI range
![Page 6: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/6.jpg)
SOI [deg. aTDC]
-140 -120 -100 -80 -60 -40 -20 0C
A [d
eg. a
TDC
]-10
-5
0
5
10
15
20 CA10CA50CA90
E10 CONSTANT LAMBDA – SOI SWEEP
Early to near TDC SOI effect
Minimum fuel provides fuel requirement (least fuel) for combustion stability, but does not give same mixture to study SOI effect explicitly on ignition and combustion constant lambda approach
Fuel rate was adjusted to keep same lambda through all SOIs
λ calculated from emission bench
“Sweet spot” (early ignition)
High noise level near
TDC
quasi HCCI Transition GCI
Less fuel in squish region
Almost similar ignition location in “quasi HCCI” (also similar CA50/CA90)
Existing region where earliest ignition occurs (near -30 deg. aTDC) – reduction in fuel in squish region
Near TDC, short residence time for ignition
IMEP increases slightly near TDC (less fuel in squish)
It was harder to control noise level with late injection
very lean combustion (λ = 4.4)
6
COVIMEP<3%, max = 5%)
![Page 7: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/7.jpg)
LOW SPEED/LOAD E10 CONSTANT LAMBDA – SOI SWEEP: EMISSION
Early injection: Low level of NOx in highly
homogeneous mixture (quasi HCCI region)
Incomplete combustion (high HC)
Low combustion temperature high CO
Smoke number (FSN) was very low (<0.1) due to lean condition at all SOIs
Late injection (GCI mode): Insufficient time for air/fuel mixing High NOx due ~ high combustion
temperature; richer zones for ignition
Less HC and CO (aggressive reaction leads to more complete combustion)
GM 1.9 L 17.8:1 (CR)Engine speed 1000 rpm
Injection pressure 400 bar
Injector-Bosch
7 hole 120 deg cone
angleFuel E10
7
SOI [deg. aTDC]
-140 -120 -100 -80 -60 -40 -20 0H
C[g
/kw
-hr]
0
2
4
6
8
10
HC
SOI [deg. aTDC]
-140 -120 -100 -80 -60 -40 -20 0
T ex
haus
t [de
g. C
]
110
120
130
140
150
160
170
NO
x [g
/kw
-hr]
0246810121416
TexhaustNOx
SOI [deg. aTDC]
-140 -120 -100 -80 -60 -40 -20 0
CO
[g/k
w-h
r]
0
20
40
60
80
100
120
CO
Low combustion temperature
More fuel enters squish zoneLong residence
time, well-mixed combustion
![Page 8: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/8.jpg)
PARAMETRIC STUDY: HIGHER ENGINE SPEED CONDITION (2000 RPM)
Engine Speed 2000 rpmLevel Low High
Constant Boost 0.45Injection pressure [bar] 400 600
SOI [deg. BTDC] 70/20 70/40Lambda 2.7 3.7
Constant Lambda 3.1Injection pressure [bar] 400 600
SOI [deg. BTDC] 70/20 70/40Boost [bar] 0.35 0.55
Double injection: Same duration for
pilot & main Fixed pilot Helpful for meeting
COV, noise levels
Sample AHRR of long vs short dwell: Pinj=600 bar, PIntk=0.55 bar, λ = 3.1
8
Considered parameters
P Injection PressureS Start of injectionL LambdaB Boost
![Page 9: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/9.jpg)
FACTORIAL STUDY - CONSTANT BOOST TESTS SHOW INFLUENCE OF COMBUSTION MODE ON EMISSIONS, LAMBDA ON COMBUSTION PHASING
EffectP S L PS PL SL PSL
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5
IMEPCOVNoiseBSFC
EffectP S L PS PL SL PSL
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5 NOx
HCCO
B [bar] 0.2P [bar] 400 600S [deg. bTDC] 15 141L 3.6 4.5
P Injection PressureS Start of injectionL LambdaB Boost
1000 RPM
2000 RPM
λ has strong impact on ignition
Lower performance with leaner mixture
Lower emissions with shorter dwell
B [bar] 0.45P [bar] 400 600S [deg. bTDC] 70/20 70/40L 2.7 3.7
![Page 10: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/10.jpg)
FACTORIAL STUDY - CONSTANT LAMBDA TESTS SHOWS SIGNIFICANT BOOST EFFECT ON COMBUSTION PHASING, NOISE AND EMISSIONS
10
EffectP S B PS PB SB PSB
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5 CA10
CA50CA90
EffectP S B PS PB SB PSB
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5 NOx
HCCO
EffectP S B PS PB SB PSB
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5 CA10
CA50CA90
EffectP S B PS PB SB PSB
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5
IMEPCOVNoiseBSFC
EffectP S B PS PB SB PSB
Nor
mal
ized
[a.u
.]
-1.5-1.0-0.50.00.51.01.5
NOxHCCO
1000 RPM
2000 RPM
Advanced ignition at higher boost
Reduced COV, BSFC, Increased Noise at higher boost
Overmixed at high P_inj(high HC, CO)
P Injection PressureS Start of injectionL LambdaB Boost
L 4.2P [bar] 400 600S [deg. bTDC] 15 141B [bar] 0.15 0.3
L 3.1P [bar] 400 600S [deg. bTDC] 70/20 70/40B [bar] 0.35 0.55
![Page 11: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/11.jpg)
0
20
40
60
80
100
120
140
0
5
10
15
20
25
30
‐150 ‐100 ‐50 0 50 100 150
HRR
[J/CAD
]
Curren
t [A]
CA [aTDC]
Injector Current and HRR of 5 bar and 8 bar at 2000 RPM with ~30% EGR (NOx<0.45 g/kWhr, CO<1 g/kWhr, HC<3g/kWhr, FSN < 0.03)
20160404_000
20160404_001
20160404_002
20160405_002
20160405_003
HRR_0404_000
HRR_0404_001
HRR_0404_002
HRR_0405_002
HRR_0405_003
SOI OF MULTIPLE INJECTION AND HRR: 5 BAR AND 8 BAR BMEP
11
SO
I 1st
= -1
00 a
TDC
SO
I 2nd
= -7
0 aT
DC
SO
I 3rd
= -2
5 aT
DC
Pinj = 600 bar5 bar
(48 C, 1.37 bar)
8 bar(48 C, 1.63
bar)
![Page 12: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/12.jpg)
CAD
0 10 20 30 40 50
HR
R [J
/CA
D]
-20
0
20
40
60
80
100800 bar SOI (47/29/16)600 bar SOI (100/70/25)
INJECTION PRESSURE & SOI EFFECTIVELY CONTROL COMBUSTION PHASING
12
CA10/50/90 delayed by 4
deg
Retarded SOI further delay ignition and combustion phasing: this results in slight reduction of noise, but most of all relax the constraint of EGR
Higher peak pressure near TDC for late SOI case implies higher heat loss. This resulted in lower performance of BSFC (increase by 22 g/kW-hr). Adjustment in 1st
and 2nd injection and/or increase in EGR would possibly help
![Page 13: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/13.jpg)
CHALLENGE IN FUNDAMENTAL STUDY OF SOI EFFECT
13
Fueling rate varied when changing SOI Injector per cylinder needs to adjusted SOI duration to
match load, combustion phasing
Slight change in 3rd
injection timing (0.2 deg) could make a difference in noise level (~4 dB level)
CAD
-5 0 5 10 15 20 25 30 35N
orm
aliz
ed M
FB
0.00.10.20.30.40.50.60.70.80.91.01.1
Cyl 1Cyl 2Cyl 3Cyl 4
![Page 14: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/14.jpg)
LP EGR SETUP & TESTING CONDITIONLP EGR Adjustment by means
ExhaustValve
Throttle valve on overall exhaust discharge
Inlet ValveThrottle on fresh intake air upstream of turbo, to drive LP-EGR
LP EGR Valve
Throttle valve between post DPF exhaust and turbocharger intake
Test condition: EGR% sweep at constant load
(BMEP ~ 3 bar at 2000 RPM) EGR% adjusted by separate valves
(most effective exhaust valve) Triple injection (SOI of each: 100-70-
25 deg. bTDC) Supercharger (ON) for P intake = 0.6
bar• Allows for precise control of
intake pressure!
Inj 1 Inj 2 Inj 3
Pinj = 400 bar
14
Smoke meter
DPF
AMA2000
Lambda sensor
![Page 15: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/15.jpg)
RECENT RESULTS WITH LP-EGR: APPROACHING AN OPTIMAL INJECTION STRATEGY FOR LOW NOISE, BSFC AT FULL LOAD RANGE
15
S/C OFF (new test) Since start of injection is very in
controlling the combustion phasing; noise is also reduced
EGR was even reduced from 30% to 21% even for late injection
Higher EGR may be required to maintain location of combustion phasing
Advantage of using higher injection pressure for fuel atomization and promote mixing to achieve lean combustion (lambda ~ 1.6-1.9); more stable combustion
Speed Rail P EGR P_intk T_intk Lambda BMEP BSFC CA10 CA50 CA90 IMEP COV Noise NOX HC CO FSN T_exhaust[rpm] [bar] [%] [bar] [deg. C] [a.u.] [bar] [g/kw‐hr] [bar] [%] [dB] [a.u.] [deg. C]2000 600 29.8 1.4 45.6 1.4 4.9 312.0 7.8 10.8 17.6 6.1 3.9 93.9 0.4 1.2 3.2 0.017 316.92000 800 21.1 1.5 48.1 1.6 4.8 337.5 12.4 16.1 25.7 6.0 3.3 91.8 0.5 0.1 0.4 0.024 452.8
[g/kw‐hr][aTDC]
Low emission
![Page 16: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/16.jpg)
LP-EGR IMPROVES PERFORMANCE AT HIGHER LOADS AS WELL
Emissions (NOx, HC, CO) and FSN are reduced significantly with LP- EGR
Exhaust temperature remains high with EGR• Improved aftertreatment expected with higher
exhaust T Supercharger needed to maintain boost stability
• Higher BSFC than intended• LP-EGR modifications to allow use of turbo
only
GM 1.9 L 17.8:1 (CR)Engine speed 2000 rpm
Injection pressure 400 bar
Injector-Bosch
7 hole 120 degcone angle
Fuel E1016
EGR Intake T T_exh BMEP BSFC Noise NOX HC CO FSN[%] [deg. C] [deg C] [bar] [g/kw-hr] [dB] [g/kW-hr] [g/kW-hr] [g/kW-hr] [a.u.]
29.41 53.64 309.5 4.32 346.13 90.13 0.45 0.91 2.10 0.03029.84 45.56 316.9 4.91 312.04 93.90 0.42 1.16 3.20 0.01729.43 47.11 327.5 5.16 317.10 91.58 0.33 0.57 1.34 0.02030.50 47.58 419.9 8.27 280.25 93.88 0.04 0.37 1.27 0.02730.48 47.91 412.2 8.33 278.78 91.01 0.05 0.37 1.44 0.024
![Page 17: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/17.jpg)
17
Study at UCB (Vuillemier) shows that LTHR has significant effect on gasoline HCCI ignition
His conclusions indicate:
A fuel’s Octane Index is a good indicator of its GCI Low-Load Performance
LTHR Onset Pressure in an HCCI engine correlates very well with GCI Low-Load Performance.
Increased intake pressure increases low-temperature heat release, enabling lower loads in a GCI engine
COLLABORATION - UCB WORK SHOWS SIGNIFICANT LTHR DEPENDENCE FOR GCI
Courtesy of David Vuilleumier (August 2015 AEC meeting)
![Page 18: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/18.jpg)
SUMMARY New approach to study the ignition of E-10 in GCI engine using “constant lambda” to
investigate ignition behavior Combustion modes were seen distinctively from CA10/50/90 vs SOI plot: “quasi”
HCCI, transition, and GCI modes Design of experiment (DOE) provide overall view of different parametric study using
minimal number of test runs: constant lambda, constant boost– Boost, lambda, and SOI affect strongly on ignition, combustion phasing, and emission– Injection pressure has minimal effect at low speed; but contribute significantly at high speed
in controlling NOx/HC/CO, and combustion stability
Installation of the new LP-EGR has been evaluated with benefit of using smoke-filtered and cooled EGR to effectively control the combustion phasing, therefore noise and emission– Higher EGR decrease combustion noise but fuel consumption is still high due to incomplete
combustion– Reduced EGR required for 5 bar BMEP was achieved by retarding SOIs. This was
performed with higher injection pressure with the same emission output (low emission, and ultra-low smoke)
– Combustion phasing was further delayed to facilitate with combustion noise– The SOI of the 3rd injection (determine load and combustion phasing the most) is very
sensitive to both noise and BSFC. Further investigation is needed to identify the flexibility when changing higher load/speed
18
![Page 19: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/19.jpg)
FUTURE WORK Continue to explore/understand effect of injection strategy
upon GCI operation E10 is sensitive to both boost and EGR
Explore more conditions with LP-EGR Provide more boost at low speeds/loads with EGR
Examine influence of CR upon combustion noise Alter IVC relative to exhaust cam Use lower CR piston crowns (we have 16, 15 and 14:1)
Continue to develop strategy for transient operation with injection, boost and EGR
Continue to track and account for USCAR guidelines combustion noise – later injections at higher pressure show significant promise! Target <90 dB for high load, <85 dB for low load
Continue to characterize GCI particulate emissions TEM sampling and analysis19
![Page 20: GASOLINE COMPRESSION IGNITION – A PROMISING …](https://reader030.fdocuments.in/reader030/viewer/2022012621/61a1731f90da37060c1fde4d/html5/thumbnails/20.jpg)
www.anl.gov
THANK YOU FOR YOUR ATTENTION!
QUESTIONS?