2014-01-1300
Published 04/01/2014
Copyright © 2014 SAE International
doi:10.4271/2014-01-1300
saeeng.saejournals.org
Development of a Gasoline Direct Injection Compression Ignition
(GDCI) Engine
Mark Sellnau, Matthew Foster, Kevin Hoyer, Wayne Moore, James Sinnamon, and
Harry Husted
Delphi Automotive
ABSTRACT
In previous work, Gasoline Direct Injection Compression Ignition (GDCI) has demonstrated good potential for high fuel
efficiency, low NOx, and low PM over the speed-load range using RON91 gasoline. In the current work, a four-cylinder,
1.8L engine was designed and built based on extensive simulations and single-cylinder engine tests. The engine features
a pent roof combustion chamber, central-mounted injector, 15:1 compression ratio, and zero swirl and squish. A new piston
was developed and matched with the injection system. The fuel injection, valvetrain, and boost systems were key
technology enablers.
Engine dynamometer tests were conducted at idle, part-load, and full-load operating conditions. For all operating
conditions, the engine was operated with partially premixed compression ignition without mode switching or diffusion
controlled combustion. At idle and low load, rebreathing of hot exhaust gases provided stable combustion with NOx and
PM emissions below targets of 0.2g/kWh and FSN 0.1, respectively. The coefficient of variation of IMEP was less than 3
percent and the exhaust temperature at turbocharger inlet was greater than 250 C. BSFC of 280 g/kWh was measured at
2000 rpm-2bar BMEP. At medium-to-higher loads, rebreathing was not used and cooled EGR provided NOx, PM, and
combustion noise below targets. MAP was reduced to minimize boost parasitics. At full load operating conditions, near
stoichiometric mixtures were used with up to 45 percent EGR. Maximum BMEP was about 20 bar at 3000 rpm. For all
operating conditions, injection quantities and timings were used to control mixture stratificaton and combustion phasing.
Transient co-simulations of the engine system were conducted to develop control strategies for boost, EGR, and intake air
temperature control. Preliminary transient tests on a real engine with high rate of load increase demonstrated potential for
very good control. Cold start simulations were also conducted using an intake air heating strategy. Preliminary cold start
tests on a real engine at room temperature demonstrated potential for very good cold starting. More work is needed to
calibrate the engine over the full operating map and to further develop the engine control system.
CITATION: Sellnau, M., Foster, M., Hoyer, K., Moore, W. et al., "Development of a Gasoline Direct Injection Compression
Ignition (GDCI) Engine," SAE Int. J. Engines 7(2):2014, doi:10.4271/2014-01-1300.
INTRODUCTION
Throughout the world, many efforts are being made to improve
the thermal efficiency of internal combustion engines. One
relatively new approach is gasoline partially-premixed
compression ignition (PPCI) that was introduced by Kalghatgi
[1] and first tested by Johansson [2]. A high octane fuel was
injected late on the compression stroke of a boosted diesel
engine operating with high EGR. The injection process was
complete prior to the start of combustion enabling partial
mixing of the fuel and air prior to heat release. Very low fuel
consumption, NOx, and PM emissions were measured. This
early work established that gasoline-like fuels with high
resistance to autoignition are preferred for PPCI. These fuels
have greater ignition delay relative to high cetane fuels, which
gives them more time for mixing. The high volatility of gasoline
also helps mix the fuel and air quickly after injection. Kalghatgi
taught that PPCI requires some stratification of the fuel-air
mixture (“partially premixed but not fully mixed”) to achieve
ignition and controlled burn duration.
Since this early work, many researchers have performed
additional engine experiments and simulations, which have
contributed to a significant body of PPCI test data and
experience. Tests have been performed on diesel engines
using various gasoline-like fuels at Shell [3, 4, 5, 6], Lund [7, 8,
9, 10], Cambridge [11], UW-Madison [12, 13, 14, 15, 16],
Argonne [17, 18, 19, 20], Tsinghua [21], and RWTH [22]. In
practically all cases, diesel-like efficiency was measured.
Various injector designs and piston designs have also been
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
developed with significant improvements. Delphi [23, 24, 25,
26, 27] reported single-cylinder and multi-cylinder engine test
results with various injectors using single, double, and triple
injection strategies.
Finally, aggressive transients with high rate of load increase
were simulated and then tested on the real engine. Cold starts
at various ambient temperatures were also simulated and
tested on the real engine.
Engine tests have also been performed using naphtha fuels on
both modified spark-ignited engines [28] and modified diesel
engines [29]. Naphtha has significantly lower octane number
(RON 70 to 84) and significantly lower processing cost
compared to current market gasoline. This work has shown
compatibility of the PPCI combustion process with lower
octane fuels in a longer term perspective.
Table 1. Preliminary Targets for Engine Testing.
PPCI has demonstrated very good potential for very high fuel
efficiency with low engine-out NOx and PM emissions using a
range of gasoline-like fuels. However, towards a production
solution, significant issues remain. Due to the lower exhaust
enthalpy of low temperature engines using PPCI, it is difficult to
produce intake boost with acceptable boost system parasitics.
A practical powertrain system with robust PPCI combustion is
needed, including injection, valvetrain, boost, and exhaust
subsystems. The engine must also meet vehicle packaging
requirements under hood and satisfy cold start and transient
response requirements.
As part of a US Department of Energy funded program, Delphi
has been developing a multi-cylinder engine concept for PPCI
combustion with current US market gasoline (RON91). The
engine has four cylinders with a 1.8L displacement and was
designed based on extensive simulations and single-cylinder
engine tests. A multiple-late-injection (MLI) strategy with
GDi-like injection pressures was selected without use of a
premixed charge. The absence of classic knock and preignition limits in this process enabled a higher compression
ratio of 15. The engine operates “full time” [25] over the entire
operating map with partially premixed compression ignition. No
combustion mode switching, diffusion controlled combustion, or
spark plugs were used. Delphi uses the term Gasoline
Direction Injection Compression Ignition (GDCI) in reference to
this combustion process.
One program objective was to build a practical GDCI engine
that achieves diesel-like fuel efficiency using current market
gasoline (RON91) with engine-out NOx and PM emissions
below that needed for NOx or PM aftertreatment. Table 1 lists
initial targets for engine testing and development. Combustion
noise level (CNL) limits are shown in Figure 1. Other program
objectives include demonstration of good transient load
response and room temperature cold starts.
In the current work, analysis and simulation tools were used to
design and fabricate a new multi-cylinder GDCI engine. Design
tools were used to package the powertrain in a D-class vehicle.
Engine dynamometer tests were conducted over a range of
operating conditions and included preliminary calibration
mapping. With this data, a competitive assessment of brake
specific fuel consumption (BSFC) was performed using
published data for gasoline, diesel, and hybrid vehicle engines.
Figure 1. Combustion Noise Level (CNL) Limits used for Testing.
GDCI CONCEPT AND INJECTION
STRATEGY
The GDCI engine concept combines the best of diesel and
spark-ignited engine technology. Like diesel engines, the
compression ratio is high, there is no intake throttling, and the
mixture is lean for improved ratio of specific heats. GDCI uses
a new low-temperature combustion process for partiallypremixed compression-ignition. Multiple-late-injections of
gasoline (RON91) vaporize and mix very quickly at low
injection pressure typical of direct injected gasoline engines.
Low combustion temperatures combined with low mixture
motion and reduced chamber surface area result in reduced
heat losses.
A schematic of the GDCI combustion chamber concept is
shown in Figure 2. The engine features a shallow pent roof
combustion chamber, central-mounted injector, and 15:1
compression ratio. A quiescent, open chamber design was
chosen to support injection-controlled mixture stratification.
Swirl, tumble, and squish were minimized since excessive
mixture motion may destroy mixture stratification created
during the injection process. The piston and bowl shape were
matched with the injection system and spray characteristics.
The bowl is a symmetrical shape and was centered on the
cylinder and injector axes.
The GDCI injection strategy is central to the overall GDCI
concept and is depicted in the Phi-T (equivalence ratiotemperature) diagram shown in Figure 3. The color contours in
Figure 3 show simulated CO emissions concentration. The
injection process involves one, two, or three injections during
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
the compression stroke and are shown as quantities Q1, Q2,
and Q3. Each injection begins in the upper left of the Phi-T
diagram (liquid) and quickly vaporizes and mixes to Phi less
than 2 by the start of combustion (SOC). At SOC the fuel-air
mixture is stratified to achieve stable ignition and controlled
heat release. Wall wetting is minimized and fuel is kept away
from cold zones such as the piston topland and cylinder liner
that may impede complete oxidation.
phenomena. Results for a double injection strategy at 2000
rpm and 11 bar indicated mean effective pressure (IMEP) are
shown in Figure 4a,b,c,d,e at various crank positions.
Figure 4a. Phi-T diagram at −50 CAD atdc (during first injection).
Figure 2. GDCI combustion chamber concept.
The ideal “stratification line” is shown in Figure 3 and
represents the ideal injection process. To achieve low NOx and
low PM emissions simultaneously, combustion must occur “in
the box” shown in Figure 3 (avoiding soot and NOx formation
regions). To also minimize CO emissions, which can
compromise efficiency, combustion must occur in the region
0<Phi<1.2 with 1300<T<2200 degrees K. Therefore at SOC,
which is approximately top-dead-center (TDC), all parcels in
the combustion chamber should be no richer than Phi of
approximately 1.2. This corresponds to the top of the
stratification line in Figure 3. The stratification line is inclined
due to charge cooling effects with the richest parcels cooling
more than the leanest parcels.
Figure 4b. Phi-T diagram at −34 CAD atdc (start of second injection).
Figure 3. GDCI Injection Strategy depicted on Phi-T Diagram with NOx,
Soot, and CO Contours.
KIVA simulations were performed with this combustion system
and an injector spray model developed using spray chamber
test data. A full 360-degree mesh with a fine grid was
necessary to capture the injection, mixing, and combustion
Figure 4c. Phi-T diagram at −10 CAD atdc (prior to start of
combustion).
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
This sequence of KIVA simulations represents the intended
injection, mixing, and combustion processes for GDCI
combustion. Because combustion occurs at relatively low peak
gas temperatures, it is referred to as low-temperature
combustion. The lower combustion temperatures are a main
factor in reducing heat transfer through the cylinder head, liner,
and piston and contribute to very low indicated specific fuel
consumption and emissions.
ENGINE DESCRIPTION
Figure 4d. Phi-T diagram at 0 CAD atdc (near start of combustion).
The absence of classic combustion knock and preignition for
GDCI means that a GDCI engine can be operated on RON91
gasoline at high compression ratio for high efficiency. For this
reason, GDCI is also a good candidate for aggressive
downsizing, down-speeding, and boosting, which is a proven
strategy for high vehicle-level fuel economy. These were two
main considerations in base engine design.
Table 2. Base Engine Specifications.
Figure 4e. Phi-T diagram at 21 CAD atdc (near end of combustion).
Figure 4a at −50 crank angle degrees (CAD) after-top-deadcenter (atdc) corresponds to the first injection. The fuel spray is
entraining air rapidly with very fast atomization and
vaporization in the bulk gas. Each data point plotted in Figure
4a corresponds to a different parcel in the combustion chamber
with color indicating the fuel mass fraction. Figure 4b at −34
CAD atdc corresponds to the beginning of the second injection.
Like the first injection, fuel remains in the bulk gas without
contacting cool wall zones. This second and last injection
controls the ignition process. Figure 4c at −10 CAD atdc shows
the mixture stratification prior to start of low-temperature
reactions. As fuel vaporizes and mixes, the equivalence ratio of
the richest part of the charge is decreasing. If too lean, the
charge will be overmixed and ignition compromised. If too rich,
soot and CO emissions will increase. The stratification line is
inclined to lower temperatures for the richer elements of the
charge due to charge cooling. Figure 4d at TDC, is
approximately start of combustion with some temperature rise
visible in the figure. Parcels with Phi near stoichiometry are the
first to release heat. Figure 4e at 21 CAD atdc corresponds to
near end-of-combustion (EOC). CO has burned out across the
chamber and small amounts of NOx were produced in the
hottest portions of the chamber. The temperature in the Phi-T
plot has now shifted to the right due to heat release and
temperature rise for each parcel. Richer parcels realize higher
temperature rise than the leanest parcels.
Figure 5. GDCI Multi-cylinder Engine.
Table 2 lists the base engine specifications. The engine is a
four-cylinder 1.8L engine with four-valves-per-cylinder. Bore,
stroke, and connecting rod length are 82 mm, 85.2 mm, and
144.7 mm, respectively, for B/S of 0.96 and r/l of 0.29. With
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
geometric compression ratio of 15, this engine has geometry
similar to modern light duty diesel engines. A photograph of the
first assembled GDCI engine is shown in Figure 5.
the GDCI and baseline diesel pistons. Significant fuel
consumption improvements were measured at several
speed-load points as shown in Figure 6 [27].
Cylinder Head
The cylinder head is an all new design with four-valve-percylinder rated at 200 bar peak cylinder pressure (PCP) for
development purposes. Extensive analysis work was
conducted to develop the cylinder head using conventional
aluminum alloys to meet cylinder sealing requirements. A new
multilayer steel head gasket was developed for this application.
The injector is central-mounted without offset or inclination
relative to the cylinder axis. The combustion chamber features
a shallow pent roof with valves flush with the pent roof surface.
The low pent roof angle was selected to achieve high
compression ratio, while maintaining low chamber surface area
and reasonable valve sizes. Intake and exhaust ports were
developed for good flow characteristics but without the need
for swirl, tumble, or squish. This was a significant advantage
and produced good intake flow efficiency as a function of valve
lift as measured on a steady flow bench. A double overhead
camshaft, Type II valvetrain with roller finger followers was
selected due to low friction and compact packaging. No spark
plug exists in the cylinder head but there is provision for
cylinder pressure sensors. In addition to cylinder pressure
sensing, Kistler cylinder pressure transducers were mounted
vertically at the cylinder periphery. Overall, the cylinder head is
very compact and employs conventional aluminum casting
methods, while meeting demanding structural requirements.
Piston
A new piston was designed for GDCI based on extensive KIVA
simulations. The piston bowl shape and injector spray
characteristics were matched for typical GDCI injection timings.
This resulted in significantly reduced piston surface area and
less propensity for wall wetting relative to typical diesel pistons.
The bowl is symmetrical and is located on the cylinder axis.
Table 3. Main Features of GDCI Piston relative to typical LD diesel
piston [27].
Figure 6. Fuel consumption and combustion noise for GDCI piston
compared to diesel baseline from Delphi [24].
Injection System
Previous work by the authors has demonstrated improved
GDCI combustion performance through injection system
developments [23, 24, 25, 26, 27]. Simulation tools were used
to develop injector spray characteristics that provide very fast
vaporization, low spray penetration, while also providing the
necessary injection rate (flow rate). Designed experiments on a
single-cylinder engine were used to determine best injection
timings, quantities, and injection pressures for various injector
designs. Injection pressures are in the range used by
production gasoline direct injection (GDi) engines. Lower
injection pressures are desirable to reduce pump parasitics
and fuel system cost.
The fuel pump was mounted at the end of the cylinder head
and was driven by the exhaust camshaft. The pump shaft has
four lobes and exhibits excellent rail pressure control via
modulation of the spill valve. The fuel rail has inlet and outlet
orifices for dampening of pressure pulsations in the system.
Injector drive waveforms were developed to support the
multiple late injection strategies used for GDCI. Importantly, the
system delivers accurate fuel quantities for multiple injections
in all cylinders.
Valvetrain System
The main features of the GDCI piston are listed in Table 3 [27].
The piston, pin, and ring pack were rated at 200 bar peak
cylinder pressure. Surface area, squish area and blowby area
(defined by ring end gap and second land radial clearance)
were significantly reduced relative to typical light-duty diesel
piston designs (BMW M57 piston used as reference). Back-toback tests were conducted on a single-cylinder engine using
Full-time GDCI combustion over the operating map is
achievable by using exhaust rebreathing (RB) at low loads and
cooled EGR at medium to higher loads. Rebreathing (RB) is
accomplished using the exhaust valvetrain system, which
provides a secondary exhaust lift event during the intake
stroke. It is an effective method to recuperate heat from hot
exhaust gases in order to raise mixture temperatures. This
heat promotes autoignition at low loads when boost pressure is
zero or low. Rebreathing also collapses the pumping loop
(intake and exhaust valves are open at same time) and raises
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
exhaust temperatures significantly. In this way, a close-coupled
oxidation catalyst will remain warm over a wide range of low
loads without special maintenance heating.
The valvetrain system is a Type II, continuously-variable
mechanical system that provides the rebreath functionality. The
valvetrain has good packaging characteristics with low friction.
A Delphi electric camshaft phaser is used to actuate the
exhaust valvetrain and control the secondary valve lift with very
fast response. Figure 7 shows a photograph of the front of the
engine with electric camshaft phasers.
Figure 8. Schematic of boost system with charge air coolers and
LPL-EGR.
Exhaust System and Aftertreatment
Figure 9 shows a photograph of the exhaust system used on
multicylinder GDCI engines. The manifold has a 4-into-1
configuration with compact, symmetrical design, and tubular
construction. The runner lengths are nearly identical for each
cylinder for uniform flow and pressure dynamics. This insures
that rebreathing of exhaust gases will also be uniform across
the cylinders. The exhaust system is insulated to maintain
sufficient exhaust temperatures for aftertreatment.
Figure 7. Photograph of front of engine with electric phasers.
Boost System
Intake boosting is difficult for low-temperature engines because
the exhaust enthalpy to drive a turbocharger is very low. For
medium-and-high-load operation, GDCI requires increased
cooled EGR to dilute the charge and maintain proper
combustion phasing. This puts greater demand on the boost
system. Extensive engine simulations were performed [26] to
develop a system that provides the necessary intake boost with
low boost parasitics. Several boost architectures with various
boost devices were studied including a 1-stage turbocharger
(TC), a 2-stage TC, a 1-stage supercharger (SC), and a
2-stage TC-SC system. The 2-stage system with one variable
geometry TC, one SC, and two charge coolers was selected
for the most efficient operation over the operating range.
The aftertreatment system is comprised of only an oxidation
catalyst. The intent is to satisfy tailpipe NOx and PM emissions
targets using clean GDCI combustion technology. The
oxidation catalyst is close-coupled to the turbocharger outlet
for heat conservation. The catalyst brick is a round, alumina
monolith with high loading of platinum and palladium for low
temperature light off of CO and HC species. The entire exhaust
system including flex pipes and sensors is packaged for the
GDCI vehicle application. An exhaust pipe and muffler were
used for dynamometer testing to emulate the restriction of the
vehicle exhaust system.
Figure 8 shows a schematic of the boost system. No intake
throttle is used. The overall system including liquid-cooled
charge air coolers was developed and packaged for minimum
system volume, good cylinder-to-cylinder flow uniformity, and
low flow restriction on the vehicle. Since the supercharger is a
positive displacement device driven by the engine, transient
response of the boost system is very fast.
The LPL-EGR system is also shown in Figure 8. The exhaust
gases exit the turbine and pass through a close-coupled
oxidation catalyst to reduce HC and CO emissions. Exhaust
gases at this point are already relatively cool. The EGR flow
enters an EGR cooler located upstream of the EGR valve. The
EGR cooler operates at engine coolant temperature to prevent
water condensate at the TC compressor inlet. The EGR system
is designed for high flow in a compact layout for fast response
on the vehicle.
Figure 9. Photograph of GDCI exhaust system on a dynamometer
engine.
EXPERIMENTAL METHODS
Experiments were performed on an engine dynamometer at
Delphi. The test fuel was Shell regular unleaded gasoline with
approximately 10 percent ethanol and research octane number
(RON) of 91, as is representative of commercial gasoline fuel
in the United States. Fuel property data are listed in Table 4.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
Table 4. Properties of Shell Test Fuel.
Engine air flow was measured by a Meriam laminar flow element
with a 55 gallon surge tank located upstream of the engine.
Combustion air was conditioned for temperature and humidity.
Fuel flow was measured with an AVL 735 Fuel Mass Flow Meter
in combination with an AVL 753 Fuel Temperature Controller.
Fuel flow data for operation on E10 fuel was corrected to E00
based on the lower heating value of the test fuel.
Exhaust emissions were measured conventionally. Nondispersive infrared analysers were used for CO2, CO, and
intake CO2 species. Intake CO2 data was used to calculate
EGR levels. A chemiluminescence analyser was used for NOx
species, a heated flame-ionization detector was used for total
HC, and a paramagnetic analyser was used for O2. Exhaust
smoke was measured with an AVL 415S smoke meter. Exhaust
particulate size distribution was measured with a TSI 3090
Engine Exhaust Particle Size Spectrometer [30, 31]. The
sampling system for exhaust particulate is described in [24].
Cylinder pressure was measured with a flush-mounted Kistler
6125CU20 pressure transducer. These transducers were
inserted from the top of the cylinder head and located flush
with the chamber near the cylinder bore. A Kistler 2613B
crankshaft encoder provided crank position data and was
dynamically aligned with engine TDC using a Kistler 2629B
TDC probe. Cylinder pressure data was sampled every 0.5
CAD. Indicated mean effective pressure (IMEP) was reported
on a gross basis. Combustion noise level was measured using
an AVL FLEXIFEM Noise Meter [32].
Engine Controller
The engine controller used for GDCI dynamometer engine
testing is based on National Instruments hardware and
LabView software [33], and was built by National InstrumentsSan Antonio. The system schematic is shown in Figure 10. The
controller features real-time combustion analysis using NI
Combustion Analysis System (NI-CAS) and allows rapid
prototyping of test cell engine control software using LabView.
The controller is comprised of a National Instruments 8110
Real-Time processor, two PXi 7813R 3M gate field
programmable gate arrays (FPGA), two PXi 6123 modules for
fast data acquisition, a PXi 8512 for CAN communications, and
numerous cRIO modules for interfacing to engine sensors,
actuators, and other test cell instrumentation.
Figure 10. Schematic of Test Cell Engine Control System for GDCI
Multicylinder Engine Testing.
Engine control software for dynamometer testing was
developed by Delphi. The software balances IMEP for each
cylinder based on cylinder pressure data. A multiple-injectioncontrol utility [23] was developed to control the fuel injection
quantity, Q, for each injection event. The Q for each injection is
determined using instantaneous rail pressure, cylinder
pressure, and an embedded calibration map for each injector.
Fuel rail pressure is tightly controlled using the rail pressure
sensor and the GDi fuel pump spill valve. Intake boost is
controlled by the supercharger and turbocharger rack position
for minimum boost parasitics. Engine coolant temperature and
liquid-cooled, charge air coolers are controlled via temperature
feedback to maintain prescribed set points. The engine
controller provides various other controls functions for the EGR
valve, SC clutch, 2-stage oil pump, etc.
A CAN bus was used to communicate among the main
controller, actuator smart controllers, and the PUMA test
system in the test cell. The CAN bus passes information for
site safeties, emissions, fuel flow, as well as pressure
transducers and thermocouple values (Figure 10). Data
acquisition of all high-speed combustion data and low-speed
data is performed by the engine controller and saved in a
single, binary file.
PRELIMINARY CALIBRATION TESTS
Engine calibration tests were performed over a wide range of
steady-state, warmed-up, operating conditions. The engine
included all subsystems including all accessories (water pump,
air conditioner compressor, alternator, accessory drive), and
complete vehicle exhaust system. Typical operating settings for
these tests are listed in Table 5.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
The engine underwent a break-in schedule and motoring
friction was measured using the “motoring method”. Motoring
friction mean effective pressure (M-FMEP) is defined as brake
mean effective pressure (BMEP) minus net mean effective
pressure (NMEP), where NMEP is carefully measured using
cylinder pressure transducers. M-FMEP is shown in Figure 11
and is reasonably low for the first build of this new engine.
Table 5. Engine temperatures, air humidity, fuel and lubricant used for
calibration mapping tests.
Figure 12b. Log PV diagram for GDCI combustion at 1000 rpm and 3
bar IMEP.
Calibration tests were performed in two test sessions. The first
session was for low loads (2-4.5 bar BMEP) and low speeds
(800-1500 rpm) for which exhaust rebreathing was used.
Results for these tests are shown in Figure 13. The second test
session was for medium-to-higher loads (8-17 bar BMEP) in
the medium speed range (1500-2500 rpm) for which exhaust
rebreathing is not used and cooled EGR is used. Results for
these tests are shown in Figure 14. Injection strategies vary as
a function of operating condition. For all tests, “design of
experiments” methods were not used due to the high number
of control factors and limited time for testing. Test results are
considered preliminary and not yet optimized.
Figure 11. Motoring Friction Mean Effective Pressure (M-FMEP) for
first multicylinder GDCI engine.
Figures 12a and 12b show GDCI combustion characteristics at
1000 rpm and 3 bar IMEP load. Start of combustion is near
TDC with short combustion duration for high expansion of all
burned fuel for high thermodynamic efficiency. The heat
release profile does not have a diffusion tail, however, is long
enough for quiet combustion. The PV diagram in Figure 12b
shows very low pumping loop of 3 kPa.
For all operating conditions, targets for NOx, PM, CNL, and
coefficient of variation of IMEP (COV IMEP) were fully met.
Brake specific fuel consumption at all conditions was very good
with indicated specific fuel consumption (ISFC) below 180 g/
kWh over very wide operating ranges. Minimum BSFC of 214
g/kWh was observed at 2000 rpm-13.5 bar BMEP. GDCI BSFC
data is compared to other competitive engines in the
following section.
For low loads, smoke was practically absent, combustion
noise was well below targets, and combustion stability was
typically less than 3 percent. Exhaust rebreathing was very
effective to maintain exhaust temperatures at the
turbocharger inlet typically above 250 C for the lowest loads
tested. Combustion efficiency ranged from 92 to 96 percent.
Further improvements in combustion efficiency are expected
with improved calibrations.
For medium-to-high loads, NOx emissions exhibited a unique
decreasing trend with increasing load. This is contrary to
typical NOx trends for spark-ignited and diesel engines and is
attributed to increasing EGR levels with load. This is a very
important outcome and will be a key enabler to meet tailpipe
NOx targets in upcoming vehicle testing.
Figure 12a. Cylinder pressure and heat release for GDCI combustion
at 1000 rpm and 3 bar IMEP.
For all tests, peak exhaust temperatures at the turbocharger
inlet were below about 500 C, which is desirable for durability
of exhaust system components. For all tests, peak cylinder
pressure was less than 160 bar, inlet manifold pressure was
less than 3.5 bar absolute, and EGR was less than 45 percent.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
Figure 13. Measured fuel consumption, emissions, and combustion analysis results for low-load calibration tests.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
Figure 14. Measured fuel consumption, emissions, and combustion analysis results for med-high load calibration tests.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
Figure 15. Competitive engines selected for comparison.
Table 6. Estimated BSFC data (g/kWh) for competitive engines at various test conditions.
BSFC COMPETITIVE ASSESSMENT
The primary objective of this program is to achieve diesel-like
fuel economy in a practical gasoline multi-cylinder GDCI
engine. Toward this goal, a competitive assessment was
performed using recent published data for leading engines in
the same displacement and power class. Competitive engines
selected for comparison included two diesel engines [34, 35],
two hybrid vehicle engines [36, 37], and two turbo GDI engines
equipped with variable valvetrain systems [38, 39]. Figure 15
shows photos of these engines.
Engines designed for different markets in the world use different
fuels and are compliant with different emissions regulations. In
Europe, gasoline octane number and diesel cetane number are
significantly higher than in the US. As well, in the US, emissions
standards are significantly more stringent than in Europe. There
may also be differences in accessories used for testing, and this
can affect fuel consumption. Therefore, it is difficult to make
direct comparisons of fuel consumption among some engines
based on published data alone.
Table 6 lists the engines in the assessment with their
respective model year, fuel used, emission compliance, and
reference. On the right side of Table 6, brake specific fuel
consumption (g/kWh) is listed for various operating conditions
including the “best efficiency point” on the map. The data
reported for the GDCI engine is obtained directly from the
calibration mapping results presented in the last section. No
corrections for small differences in fuel lower heating value
(LHV) were made.
For the “best efficiency point” on the map, the BSFC of the
GDCI engine matches the hybrid vehicle engines, is about 5
percent higher than the two diesels, and is significantly better
than the spark ignited engines for which data is published. For
the low speed/load range, BSFC for GDCI is better than most
of the diesel engine data. BSFC for GDCI is also significantly
better than the SI engine data and comparable to the very
efficient hybrid vehicle engines. At the 2000 rpm-2 bar BMEP
world test point, BSFC for GDCI was measured at 280 g/kWh,
which is lower than all gasoline and diesel engines for which
data could be found. For the medium-to-high speed/load range,
BSFC for GDCI was within 1 to 5 percent of diesel data, and
significantly better than the spark-ignited engines. The hybrid
vehicle engines did not produce enough torque at these loads
for a comparison and are labelled as “off map”.
In summary, the first multi-cylinder GDCI engine produced very
good BSFC in preliminary testing. The results are not
considered optimized. Further BSFC improvements are
expected through near term developments of the combustion
system, improved firing friction, and refined engine calibrations.
FULL-LOAD PERFORMANCE
CHARACTERISTICS
Preliminary engine dynamometer tests were performed to
characterize full-load performance of the GDCI multicylinder
engine. For full load, the air-fuel ratio was typically at or near
stoichiometry and cooled EGR was used up to about 45
percent. Intake air temperature was controlled to 55 C and inlet
manifold absolute pressure (MAP) was limited to 3.5 bar.
Results are shown in Figure 16. For engine speeds of 1500,
2000, and 2500 rpm, BMEP was measured up to about 18 bar.
For these tests, NOx, smoke, CNL, and COV IMEP met
stringent program targets. Even at full load, the engine emitted
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
NOx less than 0.02 g/kWh and smoke less than 0.08 FSN. This
is expected to be lower than engine-out emissions of most if
not all production gasoline and diesel engines in the world.
Figure 16. Measured full-load BMEP 1500-2500 rpm and simulated
(GT Power) full-load BMEP for higher engine speeds.
For engine speeds above 2500 rpm, tests were interrupted due
to a component failure external to the engine, thus making this
data unavailable for this paper. As an estimate of expected
performance, simulations were conducted using GT Power
[40]. The model was correlated with measured full load data at
1500, 2000, and 2500 rpm, and is described in a previous
publication [26]. The model predicted maximum BMEP of about
20 bar at about 3000 rpm with 45 percent EGR. Engine test
data will be obtained to confirm these simulations.
The GDCI engine and boost system are capable of providing
greatly increased full-load performance if lower EGR levels are
permissible. Figure 16 shows simulation results for lower EGR,
with up to 35 bar BMEP predicted with zero EGR. To operate at
these lower EGR levels, a late injection strategy for diffusioncontrolled combustion is expected to be necessary. While this
could be developed, Delphi is pursuing the use of clean GDCI
combustion over the entire operating map. Overall, the
preliminary full-load results for the first GDCI multicylinder
engine meet preliminary expectations.
The combustion process also depends on fuel injection
parameters (number of injection pulses, timing and quantity of
each pulse), which are also optimized in dynamometer testing.
Injection parameters are fast parameters and can be
accurately controlled on an individual-cylinder, individual-cycle
basis with very fast control capability. However, MAP, IAT, and
EGR responses are relatively slow due to transport delays,
thermal inertia and actuator slew rate limits, which are intrinsic
to the system. For the best possible control, the intake,
exhaust, and EGR systems were designed and packaged very
compactly with minimal pipe lengths and plenum volumes.
To aid in controls developments, a detailed and comprehensive
GT-Suite [40] model of the engine system was created and
coupled in co-simulation to a Simulink model [41] of the control
algorithms. Figure 17 shows the upper level of the system
model. Figure 18 shows the upper level of the engine model
consisting of the base GDCI engine, an engine thermal model,
an engine cooling system model, and a charge air cooling
system model. The base engine model includes the lowpressure EGR system with EGR cooler, turbocharger, the first
charge air cooler (CAC-1), supercharger, and the second
charge air cooler (CAC-2). The engine cooling system model
includes cylinder heat transfer to head, liner and piston, coolant
pump, engine radiator, and thermostat. The charge air cooling
system model includes the variable-speed coolant pump,
low-temperature radiator mounted ahead of the engine
radiator, and a blend valve to maintain CAC coolant
temperature in a desired range at low ambient temperatures by
blending a controlled amount of engine coolant. Combustion
was not the focus of this particular simulation and was
modelled simply with a wiebe function. Wiebe parameters were
fit to typical combustion test data from the engine.
TRANSIENT CO-SIMULATION AND TEST
Accurate combustion control during speed/load transients is an
important requirement for GDCI multicylinder engines. Robust
auto-ignition with proper combustion phasing and heat release
rates must be maintained. The combustion process depends
on many interactive operating variables such as intake
manifold pressure (MAP), intake air temperature (IAT), exhaust
gas recirculation (EGR) and exhaust rebreath quantity (RB); all
of which affect the state of the trapped cylinder charge
(temperature, pressure, composition). Optimal steady-state
values for these parameters are determined in dynamometer
testing and used as target values during transients. The overall
goal for transient control is to maintain MAP, IAT, and EGR as
close as possible to the optimal targets
Figure 17. System model, GT-Suite Simulink co-simulation.
Simulation results are shown in Figure 19 for a tip-in tip-out
transient starting from a low-speed, light-load cruise condition
of 1000 rpm and 3 bar IMEP, to a moderate load condition of
2000 rpm and 10 bar IMEP. This moderate load condition is
typical of the heaviest acceleration transients on the FTP test
for this vehicle. The rate of IMEP rise during tip-in is 7 bar/sec.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
The supercharger and turbocharger are used to control MAP
during the transient. Boost control is very good with maximum
deviation from target of about 4 kPa. The boost system is
capable of significantly faster load transients than simulated in
this case. The simulated MAP lags the desired by
approximately 50 milliseconds (1st order time constant).
Figure 18. Engine system model using GT-Suite.
IAT is controlled by modulating low-temperature loop coolant
flow rate. IAT is generally well controlled with a small positive
excursion on tip-in and a small negative excursion on tip-out.
When high rates of boost are applied, the air in the manifold
between the CAC-2 outlet and the intake valves is
compression heated during tip-in (or expanded during tip-out).
There is no opportunity to cool (or heat) this air. In this
simulation, the IAT maximum deviation from target is +/− 3 deg
C. The system is able to maintain IAT to within 1.0 C of the
desired value in 1.3 seconds.
Figure 20. Preliminary transient response on a real GDCI multi-cylinder
engine.
Figure 19. Simulated transient response for the system.
EGR control produces a smooth EGR response with a lag of
about 0.5 second between the EGR valve, where it is
controlled, and the intake valves. The lag is due partially to gas
transport and partially to mixing in the heat exchanger inlet and
outlet headers. This is a relatively low lag and results from
compact design of the EGR and intake systems. The EGR flow
path includes an oxidation catalyst located directly at the
turbocharger outlet, an EGR cooler, and a short pipe to the
EGR valve. EGR is introduced close to the turbo-compressor
inlet, where EGR gases are mixed with incoming fresh air. The
flow must also pass through CAC-1, which is sized for good
high load thermal control at elevated ambient temperatures.
The EGR response lag will cause engine EGR to significantly
deviate from targets during transients. Since GDCI combustion
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
depends on EGR, compensation using other operating
variables during transients is used to maintain robust
combustion.
Preliminary transient tests were performed on a real GDCI
multi-cylinder engine and results are shown in Figure 20. This
test demonstrates excellent boost system response for a very
fast load transient (∼15 bar/sec). The pressure ratios across
the supercharger and turbocharger are plotted separately,
which shows that fast SC boost control can compensate for the
slower TC response. This preliminary work demonstrates very
fast response capability for precise boost control.
COLD START SIMULATION AND TEST
A robust cold start with low emissions is another important
requirement for GDCI multi-cylinder engines. The goal is to
obtain robust starts without a spark plug to avoid the additional
cost, system complexity and the controls difficulties associated
with combustion mode switching. The GDCI cold start system
includes intake air heating and leverages the pressure and
heat produced by the supercharger.
Co-simulation of the engine system was performed to
understand the effect of intake air heating and intake boost on
in-cylinder gas pressure and temperature. Cold cranking was
simulated for starter motor acceleration up to 200 rpm in one
second. Initial temperatures of all engine components and
gases were set equal to ambient temperature. The heater
power delivered to the intake air was varied as a parameter,
but held constant during each cranking run.
Figure 21 shows results for ambient temperatures of 25 C and 0
C. Intake air heat increases MAP significantly due to heat addition
downstream of the charge air cooler. However, at cranking
speeds, intake heating causes a one to two bar decrease in peak
cylinder pressure due to increased in-cylinder heat transfer. Note
that without heating, the intake air remains at ambient
temperature, even though the supercharger delivers significant
boost and temperature rise. This is because the charge air cooler
absorbs nearly all the heat generated by the supercharger.
However, as shown in Figure 21, heat addition of 100 and 200
watts per cylinder is sufficient to significantly increase both intake
air temperature and peak cylinder gas temperature.
Figure 22 summarizes the effect of heat addition on peak
cylinder temperature for the full range of simulated ambient
temperatures (25 to −25 deg C) and heating rates (0 to 300
W). It is estimated from KIVA simulations that a peak gas
temperature of 700 to 800 K is sufficient to achieve robust
auto-ignition. These results suggest that room temperature (25
C) starts may require a small amount of heater power (∼50W).
These results also suggest that intake air heating may enable
cold starts at a low ambient temperature of −25 C if 175 W per
cylinder can be delivered to the air.
Figure 21. Simulation results for cold cranking.
Figure 22. Effect of intake air heat on peak cylinder temperature.
Figure 23 shows measured engine speed and cylinder
pressure for each cylinder during a preliminary room
temperature cold start (∼ 20 C) on a real multi-cylinder GDCI
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
engine. Cranking speed of 300 rpm was quickly achieved and
one motoring engine cycle occurs prior to the first fuel injection.
A small amount of intake air heating enabled combustion on
the first cycle, which is consistent with the simulations
described above. Stable combustion is quickly achieved at the
target idle speed of 800 rpm. This preliminary test
demonstrates a good cold start at 20 C with good potential for
robust cold starting at lower temperatures.
Figure 23. Preliminary cold start test at room temperature on a
multicylinder GDCI engine.
SUMMARY
Building on previous developments and extensive simulation
work, a 1.8L four-cylinder GDCI engine was developed and
tested using market gasoline (RON91) at Delphi. The engine
uses a new low-temperature combustion process for gasoline
partially-premixed compression ignition. Central to these
advancements was a fuel injection system and injection
strategy combined with a new piston design. Using multiple
late injections and GDi-like fuel pressure, the fuel-air mixture
could be stratified but sufficiently mixed. This produced very
clean and efficient combustion as simulated using KIVA code.
Preliminary steady-state, calibration tests were performed. Test
results showed that targets for NOx (0.2 g/kWh), smoke (0.1
FSN), combustion noise (load dependent), and stability (3%
COV IMEP) were met. Injection parameters could be used at
all operating conditions to control combustion phasing.
BSFC was measured and compared to competitive gasoline
and diesel engines over a wide range of operating conditions.
BSFC for GDCI was significantly better than advanced
production SI engines, and comparable to the very efficient
hybrid vehicle engines at their best efficiency conditions (214
g/kWh). Compared to new diesel engines, BSFC for GDCI at
light loads was comparable or better, and at high loads was
about 5 percent higher. At all loads, GDCI was remarkably
clean with the potential for no aftertreatment for NOx and
particulate emissions. Further BSFC improvements are
expected through planned development work.
Preliminary transient simulations and tests were performed for
warmed-up operating conditions. Very good transient response
was observed in preliminary tests. The high boost rates of the
supercharger could offset the slower response of the
turbocharger. Preliminary cold start simulations and tests were
also performed using intake air heating. This work showed the
potential for practical GDCI cold starts. More work is needed to
optimize the engine calibration over the operating map and
develop engine controls for the vehicle.
REFERENCES
1. Kalghatgi, G., “Auto-Ignition Quality of Practical Fuels and
Implications for Fuel Requirements of Future SI and HCCI
Engines,” SAE Technical Paper 2005-01-0239, 2005,
doi:10.4271/2005-01-0239.
2. Johansson, B., “High-Load Partially Premixed Combustion in a
Heavy-Duty Diesel Engine, Diesel Engine Emissions Reduction
(DEER) Conference Presentations, 2005, Chicago, IL.
3. Risberg, P., Kalghatgi, G., Ångstrom, H., and Wåhlin, F., “AutoIgnition Quality of Diesel-Like Fuels in HCCI Engines,” SAE
Technical Paper 2005-01-2127, 2005, doi:10.4271/2005-01-2127.
4. Kalghatgi, G., Risberg, P., and Ångström, H., “Advantages of
Fuels with High Resistance to Auto-ignition in Late-injection, Lowtemperature, Compression Ignition Combustion,” SAE Technical
Paper 2006-01-3385, 2006, doi:10.4271/2006-01-3385.
5. Kalghatgi, G., Risberg, P., and Ångström, H., “Partially Pre-Mixed
Auto-Ignition of Gasoline to Attain Low Smoke and Low NOx at
High Load in a Compression Ignition Engine and Comparison
with a Diesel Fuel,” SAE Technical Paper 2007-01-0006, 2007,
doi:10.4271/2007-01-0006.
6. Kalghatgi, G., “Low NOx and Low Smoke Operation of a Diesel
Engine using Gasoline-Like Fuels”, ASME 2009 International
Combustion Engine Division Spring Technical Conference,
ICES2009-76034, 2009.
7. Manente, V., Johansson, B., and Tunestal, P., “Partially Premixed
Combustion at High Load using Gasoline and Ethanol, a
Comparison with Diesel,” SAE Technical Paper 2009-01-0944,
2009, doi:10.4271/2009-01-0944.
8. Manente, V., Johansson, B., and Tunestal, P., “Half Load Partially
Premixed Combustion, PPC, with High Octane Number Fuels.
Gasoline and Ethanol Compared with Diesel”, SIAT 2009 295,
2009.
9. Manente, V., Johansson, B., and Tunestal, P., “Characterization of
Partially Premixed Combustion with Ethanol: EGR Sweeps, Low
and Maximum Loads”, ASME International Combustion Engine
Division 2009 Technical Conference, ICES2009-76165, 2009.
10.Manente, V., Tunestal, P., Johansson, B., and Cannella, W.,
“Effects of Ethanol and Different Type of Gasoline Fuels on
Partially Premixed Combustion from Low to High Load,” SAE
Technical Paper 2010-01-0871, 2010, doi:10.4271/2010-01-0871.
11. Weall, A. and Collings, N., “Gasoline Fuelled Partially Premixed
Compression Ignition in a Light Duty Multi Cylinder Engine: A
Study of Low Load and Low Speed Operation,” SAE Int. J. Engines
2(1):1574-1586, 2009, doi:10.4271/2009-01-1791.
12.Ra, Y., Yun, J.E., and Reitz, R., “Numerical Simulation of Diesel
and Gasoline-fueled Compression Ignition Combustion with HighPressure Late Direct Injection,” Int. J. Vehicle Design, Vol. 50, Nos.
1,2,3,4. pp.3-34, 2009.
13.Ra, Y., Yun, J.E., and Reitz, R., “Numerical Parametric Study of
Diesel Engine Operation with Gasoline,” Comb. Sci. and Tech.,
181:350-378, 2009.
14.Dempsey, A. and Reitz, R., “Computational Optimization
of a Heavy-Duty Compression Ignition Engine Fueled with
Conventional Gasoline,” SAE Int. J. Engines 4(1):338-359, 2011,
doi:10.4271/2011-01-0356.
15.Ra, Y., Loeper, P., Reitz, R., Andrie, M. et al., “Study of High Speed
Gasoline Direct Injection Compression Ignition (GDICI) Engine
Operation in the LTC Regime,” SAE Int. J. Engines 4(1):14121430, 2011, doi:10.4271/2011-01-1182.
16.Ra, Y., Loeper, P., Andrie, M., Krieger, R. et al., “Gasoline DICI
Engine Operation in the LTC Regime Using Triple- Pulse Injection,”
SAE Int. J. Engines 5(3):1109-1132, 2012, doi:10.4271/2012-011131.
17.Ciatti, S., and Subramanian, S., “An Experimental Investigation
of Low Octane Gasoline in Diesel Engines,” ASME International
Combustion Engine Division 2010 Technical Conference,
ICEF2010-35056, 2010.
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
18.Das Adhikary, B., Ra, Y., Reitz, R., and Ciatti, S., “Numerical
Optimization of a Light-Duty Compression Ignition Engine Fuelled
With Low-Octane Gasoline,” SAE Technical Paper 2012-01-1336,
2012, doi:10.4271/2012-01-1336.
19.Ciatti, S., Johnson, M., Das Adhikary, B., Reitz, R. et al., “Efficiency
and Emissions Performance of Multizone Stratified Compression
Ignition Using Different Octane Fuels,” SAE Technical Paper 201301-0263, 2013, doi:10.4271/2013-01-0263.
20.Das Adhikary, B., Reitz, R., and Ciatti, S., “Study of In-Cylinder
Combustion and Multi-Cylinder Light Duty Compression Ignition
Engine Performance Using Different RON Fuels at Light
Load Conditions,” SAE Technical Paper 2013-01-0900, 2013,
doi:10.4271/2013-01-0900.
21.Yang, H., Shuai, S., Wang, Z., and Wang, J., “High Efficiency
and Low Pollutants Combustion: Gasoline Multiple Premixed
Compression Ignition (MPCI),” SAE Technical Paper 2012-010382, 2012, doi:10.4271/2012-01-0382.
22.Won, H., Peters, N., Pitsch, H., Tait, N. et al., “Partially Premixed
Combustion of Gasoline Type Fuels Using Larger Size Nozzle and
Higher Compression Ratio in a Diesel Engine,” SAE Technical
Paper 2013-01-2539, 2013, doi:10.4271/2013-01-2539.
23.Sellnau, M., Sinnamon, J., Hoyer, K., and Husted, H., “Gasoline
Direct Injection Compression Ignition (GDCI) - Diesel-like Efficiency
with Low CO2 Emissions,” SAE Int. J. Engines 4(1):2010-2022,
2011, doi:10.4271/2011-01-1386.
24.Sellnau, M., Sinnamon, J., Hoyer, K., and Husted, H.,
“Development of Full-Time Gasoline Direct-Injection Compression
Ignition (GDCI) for High Efficiency and Low CO2, NOx and PM,”
Aachen Colloquium 2011, 2011.
25.Sellnau, M., Sinnamon, J., Hoyer, K., and Husted, H., “Full-Time
Gasoline Direct-Injection Compression Ignition (GDCI) for High
Efficiency and Low NOx and PM,” SAE Int. J. Engines 5(2):300314, 2012, doi:10.4271/2012-01-0384.
26.Hoyer, K., Sellnau, M., Sinnamon, J., and Husted, H., “Boost
System Development for Gasoline Direct-Injection CompressionIgnition (GDCI),” SAE Int. J. Engines 6(2):815-826, 2013,
doi:10.4271/2013-01-0928.
27.Sellnau, M., Sinnamon, J., Hoyer, K., Kim, J. et al., “Part-Load
Operation of Gasoline Direct-Injection Compression Ignition
(GDCI) Engine,” SAE Technical Paper 2013-01-0272, 2013,
doi:10.4271/2013-01-0272.
28.Chang, J., Kalghatgi, G., Amer, A., Adomeit, P. et al., “Vehicle
Demonstration of Naphtha Fuel Achieving Both High Efficiency and
Drivability with EURO6 Engine-Out NOx Emission,” SAE Int. J.
Engines 6(1):101-119, 2013, doi:10.4271/2013-01-0267.
29.Chang, J., Viollet, Y., Amer, A., and Kalghatgi, G., “Fuel Economy
Potential of Partially Premixed Compression Ignition (PPCI)
Combustion with Naphtha Fuel,” SAE Technical Paper 2013-012701, 2013, doi:10.4271/2013-01-2701.
30.MODEL 379020 ROTATING DISK THERMODILUTER, Operation
Manual, TSI Incorporated, June, 2009.
31.MODEL 3090 ENGINE EXHAUST PARTICLE SIZE
SPECTROMETER, Operation Manual, TSI Incorporated, October,
2010.
32.AVL List, “Product Description - FlexIFEM Indi,” http://www.avl.
com/c/document_library, Mar. 2013.
33.Labview RT Software, Release 2009, National Instruments, Austin
TX, 2009.
34.Hadler, J., Rudolph, F., Dorenkamp, R., Stehr, H., et al.,
“Volkswagen's New 2.0I TDI Engine Fulfills the Most Stringent
Emissions Standards,” Vienna Symposium 2008.
35.Murata, Y., Tajiri, K., Sasaki, Y., Fukushima, H., et al.,
“Development of the New 2.2I Fuel-Efficient Diesel Engine for the
Honda Civic,” Aachen Colloquium 2012.
36.Yonekawa, A., Ueno, M., Watanabe, O., and Ishikawa, N.,
“Development of New Gasoline Engine for ACCORD Plugin Hybrid,” SAE Technical Paper 2013-01-1738, 2013,
doi:10.4271/2013-01-1738.
37.Adachi, S., and Hagihara, H., “The Renewed 4-Cylinder Engine
Series for Toyota Hybrid System,” Vienna Symposium 2012.
38.Schopp, J., Kiesgen, G., Kilias, H., Lechner, M., et al., “The New
1.6-Litre Turbocharged Engines with Direct Injection and Fully
Variable Valve Gear for the New BMW 1 Series Car,” Aachen
Colloquium 2011.
39.Vent, G., Enderle, C., Merdes, N., Kreitmann, F., and Weller, R.,
“The New 2.0I Turbo Engine from the Mercedes-Benz 4-Cylinder
Engine Family,” Aachen 2012.
40.GT Power Software, Version 7, Gamma Technologies, Inc.,
Westmont, IL, 2010.
41.Matlab Simulink Software, Mathworks, Inc., Natick, MA, 2013.
CONTACT INFORMATION
Mark Sellnau
Engineering Manager
Delphi Advanced Powertrain
3000 University Drive
Auburn Hills, MI 48326
mark.sellnau@delphi.com
ACKNOWLEDGMENTS
This work was supported by the US Dept of Energy, Office of
Vehicle Technology, and administered by Gurpreet Singh under
contract DOE DE-EE0003258.
The authors gratefully acknowledge collaboration with and
support from Hyundai Motor Company. The authors also
appreciate contributions to this work from engineers and
technicians at Delphi Powertrain, Delphi Diesel Systems;
Delphi Thermal Systems, Delphi Electronics and Safety, and
Delphi Korean Automotive Components. At Wisconsin Engine
Research Consultants (WERC), simulation work by Professor
Rolf Reitz, Professor Chris Rutland, Harmit Juneja, Chris
Wright, and Mike Bergin is gratefully acknowledged. At the
Engine Research Center (ERC) at University of WisconsinMadison, contributions by Professor Jaal Ghandhi and Dr.
Jason Oakley are gratefully acknowledged.
DEFINITIONS/ABBREVIATIONS
atdc - After Top Dead Center
B - Bore
BDC - Bottom Dead Center
BMEP - Brake Mean Effective Pressure
BSFC - Brake Specific Fuel Consumption
C - Centigrade
CAC - Charge Air Cooler
CAN - Controller Area Network
CE - Combustion efficiency
CFD - Computational Fluid Dynamics
CNL - Combustion Noise Level
CO - Carbon monoxide emissions
COVIMEP - Coefficient of Variation of IMEP
DoE - Design of Experiments
E00 - Gasoline without Ethanol
E10 - Gasoline with 10% Ethanol
EEPS - Eng. Exh Particle Size Spectra
EGR - Exhaust gas recirculation
Sellnau et al / SAE Int. J. Engines / Volume 7, Issue 2 (July 2014)
FIE - Fuel injection equipment
MLI - Multiple Late Injection
FSN - Filtered Smoke Number
NMEP - Net Mean Effective Pressure
GCR - Geometric Compression Ratio
NOx - Oxides of Nitrogen
GDi - Gasoline Direct Injection
PCP - Peak Cylinder Pressure
GDCI - Gasoline Direct Inj. Comp. Ignition
PHI - Equivalence Ratio
HC - Hydrocarbon emissions
PID - Proport., Integ., Deriv. Controller
HCCI - Homo. Charge Comp. Ignition
Pinj - Injection Pressure
HRR - Heat Release Rate
PM - Particulate Matter
IAT - Intake Air Temperature
PPCI - Partially Premixed Comp. Ignition
IMEP - Indicated Mean Effective Pressure
Q - Quantity injected
ISCO - Indicated specific carbon monoxide
r - Crank throw length
ISFC - Indicated specific fuel consumption
RB - Rebreath
ISHC - Indicated specific hydrocarbon
RON - Research Octane Number
ISNOx - Indicated specific nitrous oxide
S - Stroke
KIVA - Comput. Fluid Dyn. Code developed at Los Alamos
National Labs
SGDI - Stratified Gasoline Direct Injection
I - Connecting Rod LengthLTC - Low Temperature Combustion
SOI - Start of Injection
LHV - Lower Heating Value
MAP - Manifold Absolute Pressure
M-FMEP - Motoring Friction Mean Eff. Pres.
SOC - Start of Combustion
SC - Supercharging
TC - Turbocharging
TDC - Top dead center
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written permission of SAE International.
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE International. The author is solely responsible for the content of the
paper.