400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760 Web: www.sae.org
SAE TECHNICALPAPER SERIES 2003-01-0029
Two-Step Variable Valve Actuation for FuelEconomy, Emissions, and Performance
Mark Sellnau and Eric RaskDelphi Research Labs
Reprinted From: Variable Valve Actuation 2003(SP-1752 / SP-1752CD)
2003 SAE World CongressDetroit, Michigan
March 3-6, 2003
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, ortransmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,without the prior written permission of SAE.
For permission and licensing requests contact:
SAE Permissions400 Commonwealth DriveWarrendale, PA 15096-0001-USAEmail: [email protected]: 724-772-4028Tel: 724-772-4891
For multiple print copies contact:
SAE Customer ServiceTel: 877-606-7323 (inside USA and Canada)Tel: 724-776-4970 (outside USA)Fax: 724-776-1615Email: [email protected]
ISSN 0148-7191Copyright © 2003 SAE International
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE.The author is solely responsible for the content of the paper. A process is available by which discussionswill be printed with the paper if it is published in SAE Transactions.
Persons wishing to submit papers to be considered for presentation or publication by SAE should send themanuscript or a 300 word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.
Printed in USA
2003-01-0029
Two-Step Variable Valve Actuation for Fuel Economy, Emissions, and Performance
Mark Sellnau and Eric Rask Delphi Research Labs
Copyright © 2003 SAE International
ABSTRACT
Variable-Valve Actuation (VVA) provides improvements in engine efficiency, emissions, and performance by changing the valve lift and timing as a function of engine operating conditions. Two-Step VVA systems utilize two discrete valve-lift profiles and may be combined with continuously variable cam phasing. Two-Step VVA systems are relatively simple, low cost and easy to package on new and existing engines, and therefore, are attractive to engine manufacturers. The objective of this work was to optimize Two-Step system design and operation for maximum system benefits. An Early-Intake-Valve-Closing (EIVC) strategy was selected for warmed-up operating conditions, and a Late-Intake-Valve-Opening (LIVO) strategy was selected for the cold start. Engine modeling tools were used to fundamentally understand the thermodynamic and fluid mechanical processes involved. A procedure was developed to rapidly and automatically process engine simulations for a wide range of engine operating conditions and valve train parameters including valve lift, duration, and timing. Modeling results indicate that substantial improvements in fuel economy, NOx emissions, and performance can be achieved. Reduced cold-start HC emissions are also expected. A comparison to fully flexible VVA, Three-Step VVA, and dual-independent cam phasing (only) is presented. INTRODUCTION
Compromises inherent with fixed valve lift and timing have prompted engine manufacturers to consider Variable Valve Actuation (VVA) systems on spark-ignited engines for many decades. These systems have good potential to meet future powertrain requirements for improved fuel economy, emissions, and performance [1,2,3,4,5,6]. Unlike lean-burn Direct Injection Gasoline (DIG) technology, VVA does not require lean aftertreatment technology and is therefore significantly
lower risk in the face of increasingly stringent emissions regulations [7]. Continuously variable VVA systems vary the valve lift and timing continuously over the operating range and may be electro-hydraulically actuated [8-12], electro-magnetically actuated [13-20], or electro-mechanically actuated [21-30]. Most continuously variable VVA systems use Early-Intake-Valve-Closing (EIVC) as the primary strategy to reduce pumping work and improve fuel economy. However, these systems are relatively complex, costly, and difficult to package for high-volume production applications. Control of valve seating velocity, high power consumption, and speed limitations continue to present significant challenges for many continuously variable VVA systems [20,31]. Discrete Two-Step VVA systems represent a practical alternative to continuously variable systems due to the relative ease of application for a variety of valvetrain types. Two-Step VVA systems incorporate both a low lift cam (LLC) and a high lift cam (HLC). In general, the LLC is optimized for low speed operation and may provide some improvement in both part-load fuel economy and low-to-medium speed torque. The HLC may be independently optimized for peak power at rated engine speed. This can improve the classic tradeoff between low-speed low-load operation, and high-speed full-load operation. Overall, the optimization process must yield a balanced system that satisfies powertrain requirements for fuel economy, emissions, and performance. Two-Step systems are attractive to manufacturers because they may be configured for a variety of VVA strategies (eg; cylinder deactivation or valve deactivation, etc) using a common system architecture and component set. The ability to reconfigure Two-Step systems provides substantial flexibility to tune engine characteristics for high fuel economy, reduced cold start emissions, or high performance for sport/luxury applications.
The objectives of the current work are to identify leading VVA strategies for maximum benefits, understand how to optimize Two-Step VVA systems, and realistically quantify system benefits. Figure 1 shows the general approach to the problem used by Delphi. Through this process, the best mechanisms may be matched to the leading strategies to determine the most cost effective solutions.
Figure 1. Delphi Approach to VVA System Development
The primary issues surrounding Two-Step VVA system optimization include the following:
1. What are the optimum valve lift profiles and timing to improve fuel economy, performance, and emissions?
2. What are the benefits of cam phasing when combined with Two-Step VVA?
3. What are the benefits of optimized Two-Step VVA systems compared to continuously variable VVA systems?
These issues, among others, are addressed in this work. Modeling results for both part-load and full-load operating conditions, and the drive cycle are presented. A comparison to fully flexible VVA, 3-Step VVA, and dual-independent cam phasing (only) is also shown. ALTERNATIVE TWO-STEP VVA STRATEGIES
A variety of VVA strategies exist and many have been well documented in the literature [1,2,6]. Figure 2 shows intake valve lift profiles for four basic VVA strategies including Early-Intake-Valve-Closing (EIVC)[32], Late-Intake-Valve-Closing (LIVC)[33], Late-Intake-Valve-Opening (LIVO), and Variable-Max-Valve-Lift (VMVL)[34]. These strategies represent a different way to control the airflow into the engine and reduce pumping losses during gas exchange. By modulating the lift and timing of the intake valves using these strategies, non-throttled load control (NTLC) may be achieved with varying effects on engine operating characteristics. A summary of these effects is shown in Figure 3. Overall, Early-Intake-Valve-Closing is the preferred VVA strategy for reduced pumping losses.
Figure 2. Fundamental VVA Strategies
Figure 3. Attributes of Ideal VVA Strategies
A variety of Two-Step VVA strategies are also available. Table 1 lists these VVA strategies and indicates primary benefits and applications.
Two-Step VVA systems were first used in passenger cars in 1989 when Honda introduced the first VTEC engine [35,36,37,38]. These early Two-Step applications were engineered to increase engine output for performance applications, such as the Honda Civic CRX [35], the Acura NSX [39], and the Honda S2000 [40]. To achieve this, the low lift cam (LLC) profile was chosen to maximize low to medium speed torque, and the high lift cam (HLC) profile was chosen to maximize peak power. When a higher specific output is desired, a higher engine speed than that required by the conventional engine is required. In addition to performance improvements, Two-Step VVA may be applied on passenger car engines to reduce engine displacement at equal performance (downsizing) for improved fuel economy [36]. This approach was implemented at Subaru [41] and Mitsubishi [42]. While early systems did not include cam phasing, more recent versions of the systems do [43-47].
Identify Fundamental VVA Strategies
Determine the Quantified Benefits of Leading VVA Strategy/Mechanism Combinations (ie, Technology Assessment)
Match VT Mechanisms to Leading VVA Strategies
Value Analysis of Leading VVA Strategy/Mechanism Combinations (incld. Cost Studies)
Valvetrain Technology
Plan
Cost-EffectiveSolutions
Identify Fundamental VVA Strategies
Determine the Quantified Benefits of Leading VVA Strategy/Mechanism Combinations (ie, Technology Assessment)
Match VT Mechanisms to Leading VVA Strategies
Value Analysis of Leading VVA Strategy/Mechanism Combinations (incld. Cost Studies)
Valvetrain Technology
Plan
Cost-EffectiveSolutions
0 2 4 6 8
10
120 240 360 480 600
Crank Position (CAD)
Valve Lift (mm)
(TDC) BDC BDC
Increasing Load
0
2
4
6
8
10
120 240 360 480 600 Crank Position (CAD)
Valve Lift (mm)
(TDC) BDCBDC
Increasing Load
0 2 4 6 8
10
120 240 360 480 600 Crank Position (CAD)
Valve Lift (mm)
(TDC) BDC BDC
Increasing Load
0
2
4
6
8
10
120 240 360 480 600 Crank Position (CAD)
Valve Lift (mm)
(TDC) BDC BDC
Increasing Load
Late Intake Valve Opening (LIVO) Variable Max Valve Lift (VMVL)
Late Intake Valve Closing (LIVC)Early Intake Valve Closing (EIVC)
0 2 4 6 8
10
120 240 360 480 600
Crank Position (CAD)
Valve Lift (mm)
(TDC) BDC BDC
Increasing Load
0
2
4
6
8
10
120 240 360 480 600 Crank Position (CAD)
Valve Lift (mm)
(TDC) BDCBDC
Increasing Load
0 2 4 6 8
10
120 240 360 480 600 Crank Position (CAD)
Valve Lift (mm)
(TDC) BDC BDC
Increasing Load
0
2
4
6
8
10
120 240 360 480 600 Crank Position (CAD)
Valve Lift (mm)
(TDC) BDC BDC
Increasing Load
Late Intake Valve Opening (LIVO) Variable Max Valve Lift (VMVL)
Late Intake Valve Closing (LIVC)Early Intake Valve Closing (EIVC)
EIVC LIVC LIVO VMVLPumping Work ++ + -- - Mixture Preparation (intake kinetic energy) + -- ++ ++ VT Friction ++ - ++ + Effective Comp. Ratio Effects - - 0 0Mixture Motion @ Ignition - - ++ + Volumetric Effic. @ Low RPM ++ ++ 0 0
Viability at Light Loads 0 -- 0 -- HC/NOx Emissions +/+ 0/+ +/0 +/0 -- - 0 + ++ Improvement over Conv. Throttled Engine
Table 1. Alternative Two-Step VVA Strategies Cylinder deactivation (CDA) is another Two-Step VVA strategy [48-50] that has recently seen renewed interest (see Table 1). CDA is a well-known technique to unthrottle the active cylinders of an engine operating at part throttle and reduce overall pumping losses. In this strategy, the Two-Step valvetrain is applied to both intake and exhaust valves of the deactivated cylinders and the LLC profile is simply the base circle (zero valve lift). Due to higher load factors for the active cylinders, this strategy may result in increased NOx emissions. Valve deactivation (VDA) is a third Two-Step VVA strategy used on more recent Honda VTEC engines (see Table 1). In this strategy, the lift of one intake valve is greatly reduced while the other intake valve operates at full lift or moderate lift. Both Two-Step versions [51,52] and 3-Step versions [53] have been implemented with the objective of inducing in-cylinder swirl for extended dilution limits and reduced pumping losses. For cold start operating conditions, in-cylinder swirl results in enhanced combustion rates enabling leaner air-fuel ratios for reduced HC emissions [52]. An alternative to the Two-Step VVA strategies listed in Table 1 is a new strategy that combines Two-Step cam profile switching with Early-Intake-Valve-Closing (EIVC). Two-Step EIVC is a preferred strategy for improved fuel economy with the goal of providing most of the fuel economy benefit of continuously variable systems at a fraction of the cost. Two-Step EIVC is the focus of the current investigation. A summary of EIVC and a system description are presented in the following sections. EIVC OVERVIEW
For the current investigation, Two-Step EIVC has been chosen as the preferred strategy to improve part-load fuel economy. In this section, the EIVC process and the impact on the engine thermodynamic cycle will be reviewed in more detail.
A typical family of valve-lift profiles for EIVC is shown in Figure 4. As engine load and airflow is decreased, valve lift and duration are reduced and the timing of intake valve closing (IVC) is advanced. Note that for the classic EIVC strategy, the timing of intake valve opening (IVO) remains constant. For four-valve-per-cylinder applications, generally, both intake valves would have identical lift and timing; however, similar approaches with differential lift and/or timing may be used to generate some in-cylinder swirl.
Crank Position (CAD)Va
lve
Lift
(mm
)
112 132152 172188 210226 234282
HLC Profile
Figure 4. Valve Lift Profiles for EIVC Strategy
Figure 5. Pressure-Volume Diagram for Conventional Throttled Engine and EIVC Engine. Figure 5 shows the pressure-volume (PV) diagrams for a conventional throttled engine and an EIVC engine operating at 2000 rpm, 200 kPa BMEP. At the beginning of the intake stroke, the intake pressure of the throttled engine drops to low levels, resulting in a large pumping loop. Alternately, the EIVC engine operating at reduced intake valve lift and duration is nearly completely unthrottled. After early closing of the intake valve, the cylinder undergoes isentropic decompression and recompression at a lower effective compression ratio (ECR). The result is greatly reduced pumping work as shown in Figure 5. Due to the lower ECR for EIVC,
Log Cylinder Volume
2-Step Strategy Primary Benefit Application2-Step I&E without ICP Performance Honda[36-40], Subaru[41], Nissan[43]2-Step I&E with ICP Performance Toyota[46,47], Honda2-Step Intake with ICP Performance Porche Vario Cam Plus[44,45]
CDA Fuel Economy GM[50], Daimler Chrysler[48]CDA and 2-Step Fuel Economy Mitsubishi MIVEC[39]
VDA (1-Valve) Fuel Econ., Emis. Honda [51]VDA (2-Valve) Fuel Econ., Emis. Honda [52]VDA (2-Valve) w/ICP Fuel Econ., Emis, Perf. Honda [54]VDA (3-Step) Fuel Econ., Emis, Perf. Honda [53]
LegendCDA Cylinder DeactivationE ExhaustI IntakeICP Intake Cam PhasingVDA Valve Deactivation
average cycle pressure and temperature are reduced, and this contributes to lower heat losses and lower NOx emissions. Due to lower heat losses and other factors, EIVC may also favorably impact the indicated thermal efficiency of engines. This will be treated in a following section. The EIVC strategy can be very effective at reducing pumping losses of engines. Considering the simple ideal EIVC process, air is inducted into the cylinder just after top dead center (TDC) using a short duration, and infinitely high lift intake event. The engine is completely unthrottled and there is no pressure drop across the intake valves during induction. In the limit, all pumping losses are eliminated.
SYSTEM DESCRIPTION: TWO-STEP EIVC
Typical valve lift profiles for a Two-Step EIVC system are shown in Figure 6. By properly switching between two intake lift profiles, the functionality of continuously variable EIVC systems can be approximated. In order to enhance benefits of the system, cam profile switching may be combined with intake and/or exhaust cam phasing.
0
2
4
6
8
10
270 360 450 540 630Crank Position (CAD)
Valv
e Li
ft (m
m)
TDC
BDC
Exhaust
Low-LiftCam
High-Lift Cam
Figure 6. Typical Valve-Lift Profiles for Two-Step EIVC Components used to implement Two-Step VVA systems include a Two-Step roller finger follower (RFF), a control valve for actuating the Two-Step RFF, intake and exhaust cam phasers, and the control valves for actuating the cam phasers. These components are shown in Figures 7, 8, 9, and 10, respectively. Each of these components is designed to integrate into modern four-valve-per-cylinder cylinder head layouts with minimal packaging changes (Type II valvetrain). The Two-Step RFF is a “drop in” replacement for existing roller finger followers. Because of these packaging advantages, the cost of integration of this Two-Step VVA system is low compared to other more complex VVA systems.
Figures 7 and 8. Two-Step Valvetrain Mechanism and Hydraulic Control Valve for Overhead Cam Engines
Figures 9 and 10. Vane-Type Cam Phaser and Hydraulic Control Valve An overall schematic of the system mechanization is shown in Figure 11. The addition of intake and/or exhaust cam phasers in the system enables removal of the EGR valve. Also, electronic throttle control (ETC), as shown in Figure 11, may be required for smooth mode transitions during transients.
Figure 11. Schematic of System Mechanization for Two-Step EIVC with Cam Phasing
ETCEngine with
2-Step Valvetrain & Intake Cam Phasing
AcceleratorPedal Module
Engine Control Module
2-Step VT Control Lines
MainThrottle
MAFSensor
TPS
2-Step VT Drivers
Cam Position
IntakeCam Phaser
Coolant Temp.
ETCEngine with
2-Step Valvetrain & Intake Cam Phasing
AcceleratorPedal Module
Engine Control Module
2-Step VT Control Lines
MainThrottle
MAFSensor
TPS
2-Step VT Drivers
Cam Position
IntakeCam Phaser
Coolant Temp.
Components of the Two-Step EIVC system are used differently depending on engine operating conditions. The various operating modes of the system are shown in Figure 12. For light to medium loads, the system would operate on the LLC and the intake phaser would be advanced for maximum dilution, minimum pumping losses, and low NOx emissions. Some friction benefit could be expected from operation on the LLC.
Figure 12. Operating Strategies for Two-Step EIVC For warm idle operating conditions, the system would also operate on the LLC but the intake phaser would be somewhat retarded. Improved idle stability and efficiency are expected relative to the conventional throttled engine. For cold start operating conditions, the system would operate on the LLC but would utilize a Late-Intake-Valve-Opening (LIVO) strategy with full intake phaser retard. This provides minimum dilution, high ECR, and high intake velocities. This approach should enable improved mixture preparation and leaner air-fuel ratios for potentially large reductions in HC emissions. For full load operation, the cam profile and cam timing for best torque are selected. MODELING AND SIMULATION
A modern four-valve, spark-ignited DOHC V6 engine was chosen for simulation of Two-Step EIVC in this investigation. The engine has highly tuned, and free breathing intake and exhaust systems, and features significant advanced technology including Dual-Independent-Cam Phasing (DICP), dual-plenum Variable Intake Manifold (VIM), and low friction components. An exhaust gas recirculation (EGR) system is not included. The geometric compression ratio is 10.3.
A full multicylinder engine model with complete intake and exhaust systems was used. This was necessary to
capture tuning and gas exchange effects on dilution and engine efficiency as a function of cam profile and cam timing. A full multicylinder model was also necessary to simulate full-load engine performance. The complete engine model was validated using engine dynamometer data.
A potential benefit of Two-Step VVA systems is the reduction of valvetrain friction by operation at reduced valve lift. It is also possible to reduce the cost of Two-Step valvetrain systems by selective use of “slider elements” in place of “roller elements”, however, this may result in increased valvetrain friction. In order to comprehend the effects of valvetrain friction on brake thermal efficiency (BTE), analysis of simulation results was performed on a brake basis. This required that a detailed friction model of the valvetrain and engine be developed.
Simulation was performed using GT Power engine simulation software [55] that is commercially available from Gamma Technologies, Inc. A separate “front end” to the simulation was developed to enhance the simulation process using Microsoft Visual Basic for Applications (VBA) and Microsoft Excel. The “front end” provided the following benefits:
1.) Efficient processing of many simulations for a variety of valve lift profiles and timings over the operating map.
2.) Additional calculations using simulation data such as:
Net Pumping Mean Effective Pressure (NPMEP) Net Gross Indicated Mean Effective Pressure (NGIMEP) Tumble Ratio (TR) at 90 CAD BTDC Combustion Dilute Limits (estimated) Engine Friction Each of these parameters will be described below. 3.) Optimization of Results using a sorting procedure to determine the best valve profile and phasing when subject to constraints.
4.) Automatic plotting of simulation results.
A flow chart showing the overall simulation process is shown in Figure 13. The following paragraphs briefly describe the additional calculations used in this study.
Net Pumping Work When using an EIVC strategy, the conventional definition of pumping work does not accurately reflect the reduction in actual pumping losses. This is because pumping work is “recovered” when the piston moves upward from BDC during early compression. Positive work is done by atmospheric pressure acting on the piston and is labeled “Area C” in Figure 14.
Valve Lift Profiles
0.0
2.0
4.0
6.0
8.0
10.0
270 360 450 540 630Crank Position (CAD)
Valv
e Li
ft (m
m)
TDC
BDC
Low-SpeedIVO = 310Exhaust
Full Cam Advance: Max Overlap Maximum ResidualsMaximum Efficiency Moderate Cam Advance:
Moderate OverlapModerate Dilution
Full Cam Retard:Cold StartZero OverlapMinimum ResidualsHigh Effective CRHigh Intake Velocity
Low Lift Cam
High-SpeedIVO = 340
High-Lift Cam
Net Pumping Mean Effective Pressure (NPMEP) and Net Gross Indicated Mean Effective Pressure (NGIMEP) were defined to represent the actual pumping work and indicated work for engines operating with EIVC. Clear definitions for these parameters are shown in Figure 14 for an extreme case of very low load where the pressure during expansion is below atmospheric pressure. NPMEP was computed by subtracting Area C from PMEP; and NGIMEP was computed by subtracting Area C from IMEP. Area C was computed in the “front end” using trapezoidal integration of PV data.
Figure 14. Pressure-Volume Diagram Showing Revised Definitions for Pumping Work and Gross Indicated Work
Tumble Ratio at 90 CAD BTDC Tumble Ratio (TR) is defined as the ratio of tumble rotation rate to engine crankshaft rotation rate. The tumble ratio (TR) at TDC and BDC are available as standard outputs directly from the simulation [55]. However, the intake valves may still be open at BDC, and TR at BDC may not reflect the total integrated annular momentum flux in the cylinder. In order to approximate the TR prior to ignition at a single crank position, the TR at 90 CAD BTDC was computed. Tumble Ratio reported in this work did not include allowance for tumble decay.
Combustion Model A Wiebe function was used to model heat release in this study [58]. Based on measured durations for modern four-valve-per-cylinder engines, the 0-10 CAD burn duration and the 10-90 CAD burn duration were prescribed as a function of total charge dilution. These burn durations are shown in Figure 15 and are representative of a modern, fast burn combustion system. It was assumed that EIVC effects did not degrade combustion characteristics. However, as will be shown, operation at lower valve lifts may reduce in-cylinder mixture motion, and therefore some form of combustion enhancement may be required.
Ignition timing in the simulation was set such that the 50 percent mass burned point (CA50) occurred at 10 CAD ATDC. In order to facilitate idle speed control it is common practice to retard spark timing during engine idle to enable rapid engine torque variation by varying
Figure 13. Flow Chart for Engine Simulation and Data Processing
Sort Results
Quantity Sorter Advisor/Matlab
SelectedCar/Drive Cycle
MPG &Emissions
Generate Engine Mapsfor various L, IVO.Includes idle and full-load
Process Maps in Optimizer using practical constraints.Generate “Combined” engine maps for 2-Step EIVC
Process “Combined” Maps in Advisor.Calc Fuel Economy & Performance benefits
“Front End” “Optimizer” “Vehicle Simulation”
At DesiredNMEP?
Adjust Throttle ( MAP)
Call GT Power
Simulation StartInput: RPM, NMEP, IVO, EVO, EGR, VL Profile, Friction Model
Adjust EGR
EGR at Limit
Generate Output
All PointsRun?
End
No
Yes
Yes
Yes
No
No
EGR Optimizer
Load Finder
EGR On/Off
On
Off
Additional Calculations using Simulation Output
Plot Results
Friction FinderDetermine
Engine Friction
At DesiredNMEP?
At DesiredNMEP?
Adjust Throttle ( MAP)
Call GT Power
Simulation StartInput: RPM, NMEP, IVO, EVO, EGR, VL Profile, Friction Model
Adjust EGR
EGR at LimitEGR at Limit
Generate Output
All PointsRun?
All PointsRun?
End
No
Yes
Yes
Yes
No
No
EGR Optimizer
Load Finder
EGR On/OffEGR On/Off
On
Off
Additional Calculations using Simulation Output
Plot Results
Friction FinderDetermine
Engine Friction
Sort Results
Quantity Sorter Advisor/Matlab
SelectedCar/Drive Cycle
MPG &Emissions
Generate Engine Mapsfor various L, IVO.Includes idle and full-load
Process Maps in Optimizer using practical constraints.Generate “Combined” engine maps for 2-Step EIVC
Process “Combined” Maps in Advisor.Calc Fuel Economy & Performance benefits
“Front End” “Optimizer” “Vehicle Simulation”
At DesiredNMEP?
Adjust Throttle ( MAP)
Call GT Power
Simulation StartInput: RPM, NMEP, IVO, EVO, EGR, VL Profile, Friction Model
Adjust EGR
EGR at Limit
Generate Output
All PointsRun?
End
No
Yes
Yes
Yes
No
No
EGR Optimizer
Load Finder
EGR On/Off
On
Off
Additional Calculations using Simulation Output
Plot Results
Friction FinderDetermine
Engine Friction
At DesiredNMEP?
At DesiredNMEP?
Adjust Throttle ( MAP)
Call GT Power
Simulation StartInput: RPM, NMEP, IVO, EVO, EGR, VL Profile, Friction Model
Adjust EGR
EGR at LimitEGR at Limit
Generate Output
All PointsRun?
All PointsRun?
End
No
Yes
Yes
Yes
No
No
EGR Optimizer
Load Finder
EGR On/OffEGR On/Off
On
Off
Additional Calculations using Simulation Output
Plot Results
Friction FinderDetermine
Engine Friction
- -+ +
PMEP Compression
PMEP Expansion
IMEP Corrected
+
Actual Work Produced ………………..Conventional Pumping Work …………Net Pumping Work …………………….Conventional Gross Indicated Work … Net Gross Indicated Work…………….
Area “C”
Cylinder Volume (L)
Pres
sure
(b
ar)
- -+ +
PMEP Compression
PMEP Expansion
IMEP Corrected
+
Actual Work Produced ………………..Conventional Pumping Work …………Net Pumping Work …………………….Conventional Gross Indicated Work … Net Gross Indicated Work…………….
Area “C”
Cylinder Volume (L)
Pres
sure
(b
ar)
spark advance. CA50 was retarded 10 CAD (CA50=20) for simulations performed at idle conditions.
Figure 15. 0-10 and 10-90 Burn Durations as a Function of Total Dilution
Combustion Dilution Limits Interpretation of simulation results may be difficult without a consistent way of estimating combustion dilution limits. In a separate study, it was found that combustion dilution limits could be correlated with laminar flame speed (SL) at ignition [56]. Using engine test data for modern four-valve spark-ignited engines, it was determined that SL of about 0.45 m/s conservatively indicated combustion dilution limits. In this study, the front end was used to compute laminar flame speeds [57] at ignition from simulation results for cylinder pressure, temperature, and dilution. All simulation results were constrained by combustion dilute limits using this method.
Engine Friction Model The “front end” was also used to compute variations in engine friction as a function of valve lift profile and valvetrain design characteristics. To develop the friction model, bench tests were conducted on a cylinder head fitted with Type II roller and slider valvetrain components [59]. Valvetrain FMEP as a function of valve lift and engine speed is shown in Figure 16.
Figure 16. Valvetrain Friction Mean Effective Pressure
SIMULATION RESULTS AT 2000 RPM 200 KPA BMEP
Simulation results for 2000 rpm and 200 kPa BMEP are shown in Figures 17 through 26. The baseline in this comparison is a conventional engine without EGR and without cam phasing. The baseline used a roller valvetrain with 280 CAD intake duration and IVO of 340 CAD. “Black stars” are used to denote both the baseline and best efficiency points in each of the figures. Brake Thermal Efficiency (BTE) as a function of IVO at fixed exhaust valve opening (EVO) for the range of valve profiles shown in Figure 4 is presented in Figure 17. In general, as IVO was advanced for each profile, BTE increases up to some maximum. Both manifold absolute pressure (MAP) and dilution increase as IVO is advanced. Corresponding MAP and dilution data is shown in Figures 18 and 19, respectively.
Figure 17. Brake Thermal Efficiency as a Function of IVO for the Profile Family For profiles with short durations of 112 and 132 CAD, the engine was completely unthrottled at relatively low IVO advance and the BTE maximum for these profiles was constrained by engine airflow requirements (airflow limited). Most of the increase in BTE was due to reduced pumping work that results from the EIVC effect. Net PMEP is shown in Figure 20 For profiles with longer durations of 152, 172, and 202 CAD, more lift area was available to satisfy airflow requirements while also requiring more IVO advance and more overlap (increased exhaust residuals) to achieve the BTE maximum. For these longer profiles, the BTE maximum was increased relative to the short duration profiles and corresponded to the dilution peak as IVO was advanced. Therefore, BTE for these longer profiles is “dilution limited”.
0.00
0.10
0.20
0.30
0.40
0 1000 2000 3000 4000 5000 6000 7000Engine Speed (rev.min)
VT F
rictio
n FM
EP (b
ar)
Slider 10.95 mmSlider 9mmSlider 6mmSlider 4mmSlider 3mmSlider 2mmRoller at 10.95mmRoller 3mm
Intake: Slider (various lift)Exhaust: Roller at 10.95 mm
Roller 3mm
Roller 10.95mm
Intake: Roller (various lift)
23
24
25
26
27
280 300 320 340 360IVO (CAD)
BTE
(%)
112 132152 172202 222242 280
Baseline
+13.5 %
CombustionLimited
0
20
40
60
80
100
120
0 10 20 30 40Total Dilution (percent)
Bur
n D
urat
ion
(CA
D)
90 % Mass Burn Fraction
10 % Mass Burn Fraction
Combustion Efficiency: 0.97
The remaining profiles with the longest durations of 222, 242, and 280 CAD had the greatest lift area and required the greatest IVO advance to achieve the BTE maximums. However, this produced very high residuals that exceeded combustion dilution limits. BTE for these longest profiles is “combustion limited”. A white circle in the figures indicates the operating points for which the system is combustion limited.
0
20
40
60
80
100
280 300 320 340 360IVO (CAD)
MA
P (k
Pa)
112 132152 172202 222242 280
Baseline
-54 kPa
Figure 18. MAP as a Function of IVO for the Profile Family
0
10
20
30
40
280 300 320 340 360IVO (CAD)
Tota
l Dilu
tion
(mas
s %
)
112 132152 172202 222242 280
Baseline
+14 %Dilution
Figure 19. Total Dilution as a Function of IVO for the Profile Family
Of all the profiles, the 202 profile provided the largest BTE increase of 13.5 percent relative to the baseline. Most of this increase corresponds to the 85 percent reduction of Net PMEP as shown in Figure 20. NPMEP was reduced to as low as 6 kPa at the IVO corresponding to best efficiency.
Net gross indicated thermal efficiency (NGITE) also improved somewhat, as shown in Figure 21 and accounted for about 2 percent of the 13.5 percent total
improvement. NGITE increased due to both reduced heat losses (reduced cycle temperatures) and improved ratio of specific heats when dilution was increased.
0
10
20
30
40
50
60
280 300 320 340 360IVO (CAD)
Net
PM
EP (k
Pa)
112 132152 172202 222242 280 Baseline
-85 %
Figure 20. Net PMEP as a Function of IVO for the Profile Family Depending on profile duration, NGITE was also impacted somewhat by compression ratio effects as IVO advanced. Shorter profiles exhibited decreasing ECR and longer profiles exhibited increasing ECR as IVO was advanced. The effective compression ratio as a function of profile duration and IVO is shown in Figure 22
31.5
32.5
33.5
34.5
35.5
280 300 320 340 360IVO (CAD)
GIT
E(%
)
112 132152 172202 222242 280
Baseline
+0.6 %
Figure 21. Net Gross Indicated Thermal Efficiency as a Function of IVO for the Profile Family It was assumed that the LLC operated with a roller valvetrain for the 2000rpm, 200 kPa operating condition. Reduced valve duration and lift did reduce engine friction relative to the roller HLC, but this was a negligible factor for improved BTE. For the family of profiles, engine FMEP varied less than 1 kPa, as shown in Figure 23.
4
5
6
7
8
9
10
280 300 320 340 360 380IVO (CAD)
Effe
ct. C
ompr
ess.
Rat
io
112 CAD 132 CAD152 CAD 172 CAD188 CAD 202 CAD210 CAD 226 CAD234 CAD 280 CAD
Figure 22. Effective Compression Ratio as a Function of IVO for the Profile Family
51
51.5
52
52.5
280 300 320 340 360IVO (CAD)
Engi
ne F
MEP
(kPa
)
112 132152 172202 222242 280 Baseline
-0.43 kPa
Figure 23. Engine FMEP as a Function of IVO for the Profile Family
0
10
20
30
280 300 320 340 360IVO (CAD)
BSN
Ox
(g/k
W-h
r)
112 132152 172202 222242 280
Baseline
-83%
Figure 24. BSNOx as a Function of IVO for the Profile Family
Simulation results showed that emissions of oxides of nitrogen were strongly dependent on both valve lift profile and IVO advance as shown in Figure 24. BSNOx levels followed trends for charge dilution, with the lowest BSNOx corresponding to the highest dilution. For the 202 profile, BSNOx was reduced 83 percent relative to the baseline. Clearly, the shorter profiles, which could not attain high levels of dilution, carried a large BSNOx penalty. In conclusion, proper selection of both profile duration and IVO advance is necessary to maximize dilution levels to achieve the highest BTE and lowest BSNOx.
EIVC Effects on Mixture Motion In-cylinder mixture motion is an important consideration for Two-Step EIVC systems and can significantly affect combustion quality and knock propensity in real systems. Tumble Index (TI) versus valve lift over diameter (L/D), as measured on a steady flow bench are shown in Figure 25. This data shows that TI is substantially reduced for the low valve lifts used for Two-Step EIVC strategies.
0
0.5
1
1.5
2
0 2 4 6 8 10 12Valve Lift (mm)
Tum
ble
Inde
x
Figure 25. Tumble Index as a Function of IVO for the Profile Family
Using the data of Figure 25, the simulation was used to estimate the Tumble Ratio at 90 CAD BTDC. Results are shown in Figure 26 and show that TR decreases for LLC profiles as the timing of those profiles is advanced. This is because a greater portion of the total airflow is inducted at lower lifts as IVO is advanced. For the 202 profile at its peak BTE timing, TR is reduced about 63% relative to the baseline, yet this estimate of TR does not include tumble decay effects. In general, EIVC systems have more time available for tumble decay since induction is completed earlier in the cycle.
From inspection of estimated tumble levels, some compromise of combustion quality can be expected for operation on the LLC profiles. For low speed, light load operating conditions, this deterioration of combustion quality would reduce the estimates of BTE predicted in this study. Excessively low mixture motion (tumble) can also adversely affect knock limits at high loads.
0
0.5
1
1.5
2
280 300 320 340 360IVO (CAD)
Tum
ble
Rat
io a
t 90
Btd
c112 132 152 172202 222 242 280
Baseline
-63 %
Figure 26. Tumble Ratio as a Function of IVO for the Profile Family Consequently, to obtain the full fuel economy potential of EIVC systems, several methods to enhance combustion rates are listed below. These include:
- Increased tumble at low lifts using a chamber mask [23, 27]
- Increased swirl via valve deactivation, port deactivation, or differential valve lift and timing
- Dual ignition Use of these methods will depend on powertrain requirements for respective powertrain applications. SIMULATION RESULTS FOR THE DRIVE CYCLE
In this section, the fuel economy and emissions benefits for the drive cycle are addressed. Results are presented for a typical vehicle with representative vehicle load factors (VLF). Results are expected to vary somewhat depending on VLF. The vehicle used in the drive cycle analysis had the following attributes: Engine Displacement: 1.6L Transmission: 5 -Speed Manual Vehicle Mass: 1477 kg A four-point approximation was used to estimate fuel economy for the New European Drive Cycle (NEDC) drive cycle. Table 2 below shows the engine speed, engine load (NMEP), and relative fuel consumption for the four test points. These were derived from analysis of data collected from a vehicle driven on the NEDC test cycle.
FuelSpeed NMEP Consumed(rpm) (kPa) (%)800 200 8.91200 200 10.12200 350 64.53000 450 16.5
Table 2. Engine Speed, Load, and Relative Fuel Consumption for Four-Point Approximation
Fuel Economy Benefit The fuel economy benefit of Two-Step EIVC with Cam Phasing as a function of LLC duration is shown in Figure 27. Results for DICP only and ICP only are also shown. The baseline for this analysis is a fixed-cam engine without EGR.
Results show that the fuel economy benefit of Two-Step EIVC with DICP is about 8 percent and is approximately invariant for LLC durations from 152 to 226. The constancy of fuel economy benefit with LLC duration results from a tradeoff between low-speed, light-load and medium-speed, medium-load operating conditions. The shorter duration profiles had the best fuel economy at the light loads and the longer duration profiles had the best fuel economy at the medium loads. For durations less than 152, engine airflow requirements could not be satisfied on the LLC for the higher speed-load operating conditions during the drive cycle, and the results reflect partial operation on the HLC.
0
2
4
6
8
10
100 150 200 250 300Low-Lift Cam Duration (CAD)
Fuel
Eco
n. B
enef
it (%
)
Baseline: Fixed Cams, w/o EGR
DICP
ICP
3-Step(152 & 226)
8-Step
2-Step w/DICP ~1%
Can Not Run WholeDrive Cycle
2-Step w/ICP
2-Step w/o CP
Figure 27. Fuel Economy Benefit for the NEDC Drive Cycle
Two-Step EIVC with DICP showed about 1 percent fuel economy benefit over Two-Step EIVC with ICP only. Most of this difference is attributed to reduced blowdown loss and increased dilution available with exhaust cam phasing (ECP). However, for light-load operating conditions, the ECP could be advanced which enabled increased IVO advance for reduced NPMEP (EIVC effect). Generally, addition of ECP reduced advance requirements for the intake cam phaser (ICP).
Figure 27 also shows results for Two-Step EIVC without intake or exhaust cam phasing. For this system, the timing of IVO (fixed) was determined by conditions at idle. For short duration profiles (i.e. 152 CAD), significant efficiency improvement could be achieved. However, as will be shown below, NOx emissions for this system were high and an EGR valve would likely be required for NOx control. For longer duration profiles, IVO must be substantially retarded from peak efficiency values to avoid excessive dilution at idle. This greatly reduced the fuel economy benefit for the longer duration profiles.
Data showing the estimated fuel economy benefit from Three-Step EIVC with DICP, and Eight-Step EIVC with DICP are also shown in Figure 27. The Three-Step system utilized the 152 and 226 duration profiles, while the Eight-Step system utilized all the valve lift profiles available in this study. Results indicate that the Three-Step system and the Eight-Step system have comparable fuel economy benefit, and the additional benefit over Two-Step EIVC is only about 1 percent. This suggests that Two-Step can provide most of the benefit of continuously variable systems (using the same valve profiles) and that no more than three steps are needed to achieve nearly all of the fuel economy potential of EIVC systems. For DICP only and ICP only, as shown in Figure 27 the fuel economy benefit was about 2 percent relative to the baseline. BSNOx Emissions Simulation predictions for warmed-up BSNOx emissions on the drive cycle are shown in Figure 28. These results correspond to the data in Figure 27 and represent NOx levels for EIVC systems optimized for best fuel economy. Somewhat lower NOx levels could be achieved if the system was optimized for minimum NOx.
0
20
40
60
80
100
100 150 200 250 300Low Lift Cam Duration (CAD)
BSN
Ox
Red
uctio
n (%
)
Baseline: Fixed Cams w/o EGR
DICP
ICP
3-Step(152 & 226)
8-Step
2-Step w/DICP
~21%
Can Not Run WholeDrive Cycle
2-Step w/ICP
2-Step w/o CP
Figure 28. Brake Specific NOx Reduction for the NEDC Drive Cycle As shown in Figure 28 valve lift profiles with longer duration provide greater NOx reduction than do shorter profiles. This is because more valve overlap and more dilution were available at the medium to high load points in the drive cycle where most of the mass NOx emissions are generated. Addition of ECP greatly improved NOx emissions for the same reason. For Two-Step EIVC with DICP and a profile duration of 188 CAD, BSNOx was reduced about 62 percent relative to the baseline. This NOx level was still about 21 percent higher than for DICP only. For the shortest profiles of 132 and 152 CAD, to meet airflow requirements, the engine was operated on the HLC for some of the high load operating conditions, and this helped reduce NOx emissions. Figure 28 also shows results for Two-Step
EIVC without intake or exhaust cam phasing. Due to limited IVO advance and reduced dilution for the range of profile durations, NOx emissions were severely compromised.
Realizing Benefits in Practical Systems Achievement of the fuel economy and NOx emissions benefits using Two-Step EIVC on practical engines depends on the combustion system and ICP system characteristics. As already stated, mixture motion enhancements may be needed to achieve full fuel economy potential of EIVC systems. For the ICP system, increased phasing authority and increased phasing rate may be required depending on duration of the LLC profile. Also, for any engine application, ICP advance characteristics may be limited by valve-to-piston interference and oil pressure characteristics of the engine’s lubrication system. For the ECP system, phasing authority of 50 CAD should be adequate. In any case, realized benefits will be very application specific and will depend somewhat on vehicle load factors for each application. See the last section of this paper for additional discussion on this topic.
SIMULATION RESULTS AT FULL LOAD
Two-Step EIVC systems may be tuned for increased specific output and dynamic range. Two-Step valvetrain alone does not enable increased peak power but Two-Step valvetrain in combination with increased peak engine speed does [36,41,42].
In the current study, improved performance using Two-Step at equal peak engine speed was investigated. The LLC profile may be optimized for improved fuel economy, while also potentially providing some improvement in low-speed BMEP (depending on the LLC profile). Independently, the HLC can be optimized for medium-speed BMEP and peak power with little or no compromise in fuel economy. In the course of this work, a “performance option” was created as an example of improved performance using a total system approach.
Effect of LLC Profile on Low-Speed BMEP The effect of LLC profile on full-load BMEP over the engine speed range is shown in Figure 29. Results reflect full-load output for optimized IVO and EVO for each profile. The baseline in this comparison is the same engine with DICP only, as shown by the solid black line in the figure. Full-load BMEP for a fixed-cam version of the engine (no cam phasers) is shown by the dashed black line. A knock prediction model was not used in this simulation, and all cases were assumed to be “non-knock limited.”
Results show that low-speed BMEP can be significantly improved by using longer LLC profiles. The profile with duration of 210 CAD provides about 6 percent BMEP increase at 2000 rpm, and falls below the baseline at 2000 rpm and above. The profile with duration of 226 CAD provides about 12 percent increase at 1200 rpm, and falls below the baseline at speeds above 2400 rpm.
Judging from the results for a profile duration of 234 CAD, without engine tuning changes, further improvements in low-speed BMEP will be difficult to achieve with any profile. For the profiles with 188 and 172 durations, BMEP decreased up to 5 percent for speeds below 1200 rpm. For these profiles, the valvetrain would have to be switched to the HLC profile at about 1200 rpm to avoid more substantial reduction of full-load output.
0
200
400
600
800
1000
1200
1400
0 2000 4000 6000 8000
Engine Speed (rev/min)
BM
EP (k
Pa)
Assumed Not Knock Limited
~12%1200 rpm
172
DICP Baseline
226
188
210
Dur=234 Fixed-Cam
Figure 29. Effect of Low-Lift Cam Profile on Full-Load BMEP
0
200
400
600
800
1000
1200
1400
0 2000 4000 6000 8000
Engine Speed (rev/min)
BM
EP (k
Pa)
Assumed Not Knock Limited
~12%1200 rpm
DICP Baseline
Dur=226 CAD
PerformanceOption 10%
6500 rpm
~5%
Fixed-CamBaseline
Figure 30. Full-Load BMEP for “Performance Option” Referring to the fuel economy and NOx benefits reported in Figures 27 and 28, it is apparent that there is a tradeoff among part-load fuel economy, part-load NOx emissions, and full-load low-speed output for the Two-Step EIVC system. This will be further addressed in the section concerning system optimization. Performance Option A version of the engine model was developed as a performance option for selected sport or luxury applications. For these applications, it was assumed that additional powertrain performance has more value than additional fuel economy benefit. Applications with
higher peak engine speed were not the objective of the current work.
To create this performance engine model, a total system approach was elected. The following changes to the model were made:
1. The intake valve diameter was increased such that the ratio of valve diameter to cylinder bore was 0.354.
2. The intake port and valve flow coefficient was increased to a maximum of 0.73 at a L/D of 0.345.
3. The exhaust valve diameter was increased such that the ratio of valve diameter to cylinder bore was 0.312.
4. The exhaust port and valve flow coefficient was increased to a maximum of 0.7 at a L/D of .345.
5. The diameters of intake and exhaust pipes and runners were increased 20 percent relative to the base model.
No other changes to pipe lengths, plenum volumes, or tuning characteristics were made.
Brake Mean Effective Pressure and Brake Power (BP) for the performance option are shown in Figures 30 and 31, respectively. For the LLC, the profile with duration of 226 CAD was chosen for best low-speed torque, as already shown in Figure 29. The fuel economy benefit of this LLC was slightly decreased to 7.7 percent relative to the “fixed-cam, No EGR” baseline.
0
50
100
150
200
250
0 2000 4000 6000 8000Engine Speed (rev/min)
Bra
ke P
ower
(kW
)
Assumed Not Knock Limited
DICP Baseline
Dur=226 CAD
PerformanceOption
10% 13%
6900 rpm56 kW/L(75 HP/L)
7300 rpm64 kW/L(86 HP/L)
Fixed-Cam
Figure 31. Full-Load Brake Power for the “Performance Option” For the HLC, a BMEP improvement of approximately 5 percent at medium speed was indicated. At rated maximum engine speed of 6500 rpm, a BMEP improvement of about 10 percent was indicated. Most of the benefit at medium and high speeds resulted from improved flow efficiency of the intake ports and valves. For the HLC, a family of valve lift profiles with increased lift and duration were also investigated. However, simulation results indicated that, at maximum rated
engine speed (6500 rpm), comparable peak power was obtained for the standard profile and the higher lift, longer duration profiles. Therefore, the output increase shown in Figures 30 and 31 cannot be attributed to the Two-Step valvetrain.
UNDERSTANDING SYSTEM OPTIMIZATION
A number of EIVC systems with varying system benefits, complexity, and system on-cost have been introduced in the previous sections. In this section, the systems are compared and discussed from a systems benefits and requirements perspective. Finally, some recommendations are made as to preferred system selections. A comparison of the systems is presented in Figure 32, where potential fuel economy benefit is plotted as a function of the estimated performance benefit. The baseline in the comparison is a fixed-cam engine without EGR.
0
13
0 15Performance Benefit
Fuel
Eco
nom
y B
enef
it
Fixed-Camw/o EGR
2-Step EIVC w/ICP
2-Step EIVC w/DICP
2-Step EIVC w/DICPw/Perf Option
IncreasingComplexity
DICP Alone
ICP Alone
2-Step EIVC w/DICPw/Perf Optionw/DownSizing
1/5 Slope
3-Step EIVC w/DICP
2-Step EIVC w/o CP
Figure 32. Comparison of Fuel Economy and Performance Benefits of EIVC Systems Beginning with the base Two-Step EIVC system for which no cam phaser is used, generally poor results are indicated with limited fuel economy benefit and no performance benefit. Addition of ICP to the system provides significant improvement in fuel economy over the non-phaser system, while also matching or exceeding the performance of ICP only. Further addition of ECP (Two-Step EIVC w/DICP) to the system appears to be a good value since it improves fuel economy, warm NOx emissions, and performance, with an expected improvement of cold starting characteristics. This system (Two-Step EIVC with DICP) achieves most of the benefit of the Eight-Step system and holds good promise for high volume production passenger car applications. The system with Three Steps and DICP, while more complex, shows further improvement in fuel economy and NOx emissions without any compromise of low-speed torque. Simulation results suggest that Three-Step EIVC with DICP achieves almost all of the benefits of Eight-Step. If sufficiently low cost, this system also holds good promise for engines with certain valvetrain types.
For sport and luxury applications, the Two-Step with DICP system, and the Three-Step with DICP system may be combined with the “Performance Option” (described earlier) for significant performance increases. This combination exhibits both high fuel economy and good performance. However, it is noted that the performance benefit, which results at equal peak engine speed, is not enabled by multi-step valvetrain.
Finally, the dashed line in Figure 32 shows that vehicle performance may be traded off for improved fuel economy. The upper point in the figure is not the result of simulation but rather derives from the rule of thumb that 1 percent improvement in fuel economy can be achieved by trading off about 5 percent in vehicle performance. Similarly, by downsizing (DSZ), vehicle performance may be maintained while improving fuel economy. The upper point in Figure 32 also reflects this approach.
To highlight some of the tradeoffs among LLC duration and system benefits for Two-Step EIVC with DICP, and to select preferred LLC durations, a compendium of results for EIVC systems with DICP is shown in Table 3. System benefits are shown in columns 3, 4, and 5; while TR, the need for enhanced combustion, required ICP authority, and required ICP rate are shown in columns 6, 7, 8, and 9, respectively. Results for Three-Step, Eight-Step, DICP only, and ICP only are also listed.
For Two-Step EIVC with DICP, system benefits depend strongly on LLC duration and there is a tradeoff among part-load fuel economy, part-load NOx emissions, and full-load low-speed output (Table 3). The duration with the best compromise may be 210 CAD that provides almost all of the fuel economy benefit, good NOx emissions, and about half of the available low-speed BMEP improvement. Referring to Table 3, the TR for this duration is about half of that of the baseline HLC profile, and with a duration this long (210 CAD), the system is able to generate high levels of dilution. This implies that the system will be combustion limited over a greater portion of the light load operating range than systems with shorter duration profiles. This explains the greater need for enhanced combustion rates as shown in column 7 of the table. The required ICP authority is also higher for longer duration LLC profiles, as shown in column 8. The estimated required ICP rate for mode switching is shown in column 9 of the table. For the shortest duration LLC, a switch event will produce a large change in airflow, which must be accompanied by a relatively large change in throttle position and ICP position. Switching will also occur more frequently during the drive cycle. For the longer duration profiles, mode switching entails smaller changes in throttle the position and ICP position, and may require very few switch events on the drive cycle. Therefore, the challenges for successful mode switching, as faced by the engine management system (EMS) and vehicle calibrators, may possibly be significantly reduced.
For Three-Step EIVC with DICP, the results shown in Table 3 indicate that the tradeoff among fuel economy, NOx emissions, and low-speed torque is completely resolved. This system, with LLC duration of 152 CAD and a medium-lift cam (MLC) duration of 226 CAD, obtains all the available low-speed BMEP, while achieving comparable fuel economy and warm NOx emissions to the Eight-Step System. Importantly, the Three-Step system can also reduce requirements on the combustion system and ICP system, as projected in Table 3. Overall, Three-Step EIVC with DICP remains a very attractive strategy for future passenger car applications.
SUMMARY AND CONCLUSIONS
A simulation study has been conducted to investigate the benefits of Two-Step EIVC and Three-Step EIVC strategies on spark-ignited engines. By combining these strategies with cam phasing, significantly improved fuel economy, NOx emissions, and low-speed BMEP has been demonstrated. These strategies reduce pumping losses and provide increased internal dilution (subject to combustion limits) for improved indicated thermal efficiency and reduced NOx emissions. While results are expected to be application dependent, the following conclusions may be made:
• Longer duration low-lift cam profiles of 210 to 226 CAD are preferred for high fuel economy, low NOx emissions, and increased low-speed BMEP. However, more advanced intake valve timing is needed compared to shorter duration profiles.
• Two-Step EIVC with DICP improves fuel economy about 8% relative to a fixed-cam baseline without
EGR, or about 6% relative to the same engine with DICP alone. This system provides most of the fuel economy benefit and NOx reduction of a hypothetical Eight-Step EIVC system.
• Three-Step EIVC with Dual Independent Cam Phasing improves fuel economy about 9% relative to a fixed-cam baseline without EGR, or about 7% relative to the same engine with DICP alone. This system provides comparable fuel economy and NOx emissions to that of a hypothetical Eight-Step EIVC system.
• Low-speed engine volumetric efficiency was improved about 12% for low-lift cam duration of 234 CAD and about 6% for duration of 226 CAD. To achieve this potential BMEP improvement (subject to knock limitations), a small tradeoff with fuel economy was observed.
• Exhaust cam phasing improves fuel economy about 1% and reduces NOx emissions relative to Two-Step EIVC with intake cam phasing alone.
• Engine operation at reduced valve lift reduces in-cylinder tumble and may degrade combustion under dilution combustion conditions. Some form of combustion enhancement may be needed to achieve the full potential of EIVC systems.
• For Type II roller valve trains, operation at reduced valve lift reduces engine friction but the impact of this reduced friction on fuel economy is small.
Overall, both Two-Step EIVC with Cam Phasing and Three-Step EIVC with Cam Phasing hold good promise for high volume production applications on spark-ignited engines.
Strategy System Benefits System CharacteristicsProfile Potential Warm Low-Spd Tumble Need for ICP ICP Rate***
Duration Fuel Econ. BSNOx BMEP Ratio Enhanced Authority Reqt forBenefit Reduct. Benefit** w/o Mask Comb Rate Reqt Switching
(CAD) (%) (%) (%) (+/-) (+/-) (+/-)
Two-Step EIVC* 132 7.1 66 loss 0.4-0.5 0 0 +++ 152 8 57 loss 0.3-0.5 0 0 +++ 172 8 56 0 0.3-0.6 + + ++ 188 8 62 0 0.5-0.7 ++ + + 210 7.9 65 6 0.8-1.1 ++ ++ + 226 7.7 70 12 1-1.2 ++ +++ +
Three-Step EIVC* 152 & 226 9 71 12 0.3-1.2 + 0 0
Eight-Step EIVC* all 9.1 71 12 0.3-1.6 + 0 0DICP only 284 2.3 74 0 1.5-1.6 0 0 n/aICP only 284 2 81 loss 1.5-1.6 0 0 n/a
Baseline: Fixed-Cam Engine without EGR
* Strategies include ICP and ECP** Benefit relative to an engine with DICP only; assumed non-knock limited*** Estimated ICP Rate Requirement
Table 3. System Benefits and Requirements for EIVC Systems
ACKNOWLEDGEMENTS
The authors would like to thank Dr. James Sinnamon and Mr. Timothy Kunz for their expertise and contributions provided in the course of this work. The authors also gratefully acknowledge the following persons who also supported this work:
James Burkhard, Aaron Domuracki, Hermes Fernandez, Ryan Fogarty, Wayne Harris, Mark Krage, John Krieg, Jongmin Lee, and James Niemeier, at Delphi Corporation.
Annette Cusenza, Richard Davis, Rick Hart, Ron Herrin, Bob Jacques, David Lancaster, Frederic Matekunas, Bill Miller, Kenneth Patton, and Rodney Rask; at General Motors Corporation.
Mark Christie at Ricardo Consulting Engineers.
Thomas Morel, Brad Tillock, Tom Wanat, Matthew Warner; at Gamma Technologies, Inc.
CONTACT
For additional information, Mark Sellnau [email protected] Delphi Research Labs 51786 Shelby Parkway Shelby Twp., MI 48315
REFERENCES
1. Kreuter, P., et al., “Strategies to Improve SI Engine Performance by Means of Variable Intake Lift, Timing, and Duration,” SAE 920449, 1992.
2. Ahmad, T., and Theobald, M., “A Survey of Variable-Valve-Actuation Technology”, SAE Paper 891674, 1989.
3. Moro, D., et al., “Thermodynamic Analysis of Variable Valve Timing Influence on SI Engine Efficiency,” SAE Paper 2001-01-0667, 2001.
4. Bozza, F., et al., “A Theoretical Comparison of Various VVA Systems for Performance and Emissions Improvements of SI-Engines”, SAE Paper 2001-01-0670, 2001.
5. Klein, F. et al, “The Influence of the Valve Stroke Design in Variable Valve Timing Systems on Load Cycle, Mixture Formation and the Combustion Process in Conjunction with Throttle-free Load Governing,” SAE Paper 981030, 1998.
6. Stone, R., and Kwan, E., “Variable Valve Actuation Mechanisms and the Potential for their Application,” SAE Paper 890673, 1989.
7. Grebe, U., Konigstein, A., Friedrich, C., and Schaffner, P., “System Comparison between Gasoline Engines with Direct Injection and
Throttleless Load Control using Variable Valve Trains,” Wien 1998.
8. Shindoh, et al., “Hydraulically-Actuated Valvetrain with No Camshaft,” Inter-Tech Energy Progress, Inc., San Antonio, TX, 2000.
9. Schechter, M., and Levin, M., “Camless Engine”, SAE Paper 960581, 1996.
10. Miller, R., et al., “Unthrottled Camless Valvetrain Strategy for Spark-Ignited Engines”, Paper No. 97-ICE-43; ICE Vol. 29-1, 1997 Fall Technical Conference, ASME, 1997.
11. Ashhab, M., “Camless Engine Control for a Robust Unthrottled Operation”, SAE Paper 981031, 1998.
12. Misovec, K., “Digital Valve Technology Applied to the Control of an Hydraulic Valve Actuator”, SAE Paper 1999-01-0825, 1999.
13. Theobald, M., et al., “Control of Engine Load via Electro-magnetic Valve Actuators,” SAE Paper 940816, 1994.
14. Henry, R., and Lequesne, B., “A Novel, Fully-Flexible, Electro-Mechanical Engine Valve Actuation System”, SAE Paper 970249, 1997.
15. Hammoud, M., Haghgooie, M., “Benefits of Variable Valve Timing to Reduce Fuel Consumption and Emissions”, Essen Conference, March 28, 2000.
16. Pischinger, M., et al., “Benefits of the Electromechanical Valve Train in Vehicle Operation”, SAE Paper 2000-01-1223, 2000.
17. Flierl, R., and Klüting, M., “The Third Generation of Valvetrains – New Fully Variable Valvetrains for Throttle-Free Load Control”, SAE Paper 2000-01-1227, 2000.
18. Cosfeld, R., et al., “Technological Approaches to Realize an Electromechanical Valve Train”, Aachem Colloquium 1999.
19. Roux, I., et al., “Adaptation of Electromechanical Valve Control on a 2.0L 4-Cylinder Engine”, Essen Conference 2000.
20. Warnecke, V., et al., “The Electrical and Mechanical Properties of the System Components of the Electrical Mechanic Valve Train”, Aachem Colloquium 1999.
21. Torazza, G., “A Variable Lift and Event Control Device for Piston Engine Valve Operation”, Paper 2/10, pp 59-67, 14th Fisita Congress, 1972.
22. Hara, S., et al., “Application of a Valve Lift and Timing Control System to an Automotive Engine,” SAE 890681, 1989.
23. Heuser, P., Kreuter, P., Reinicke-Murmann, J., Erz, R., and Peter, U., “The META VVH System – The Advantages of Continuously Mechanical Variable Valve Timing,” SAE Paper 1999-01-0329,1999.
24. Nakamura, M., et al., “A Continuous Variable Valve Event and Lift Control Device (VEL) for Automotive Engines,” SAE Paper 2001-01-0244, 2001.
25. Takemura, S., et al., “A Study of a Continuous Variable Valve Event and Lift (VEL) System,” SAE Paper 2001-01-243, 2001.
26. Titolo, A., “The Variable Valve Timing System – Application on a V8 Engine,” SAE Paper 910009, 1991.
27. Burkhard, J., et al., “Benefits of a Mechanical Variable Valve Actuation System,” Aachem Kolloquium, Fajrzeig-und Motorentechnik, Aachem, Germany, 2000.
28. Pierik, R., and Burkhard, J., “Design and Development of a Mechanical Variable Valve Actuation System,” SAE 2000-01-1221, 2000.
29. Yagi, S., et al., “A New Variable Valve Engine Applying Shuttle Cam Mechanism,” SAE 920450, 1992.
30. Klüting, M., et al., “Throttle-Free Load Control by Means of Fully Variable Valve Train,” MTZ Motortechnische Zeitschrift 60 (1998) 7/8.
31. Kilger, M., et al., “Noise Reduction of an Electromagnetic Valve Train through Intelligent Control”, Essen Conference, March 28, 2000.
32. Tuttle, J., “Controlling Engine Load by Means of Early Intake Valve Closing”, SAE Paper 820408, 1982.
33. Tuttle, J., “Controlling Engine Load by Means of Late Intake Valve Closing”, SAE Paper 800794, 1980.
34. Stivender, D., “Intake Valve Throttling (IVT) – A Sonic Throttling Intake Valve Engine”, SAE Paper 680339, 1968.
35. Variable Valve Actuation”, Automotive Engineering, October 1991.
36. Inoue, K., Nagahiro, K., Ajiki, Y., and Kishi, N. “A High Power, Wide Torque Range Efficient Engine with a Newly Developed Variable Valve Lift and Timing Mechanism,” SAE Paper 890675,1989.
37. Inoue, K., et al., “A High Power, Wide Torque Range Efficient Engine with a Newly Developed Variable Valve Lift and Timing Mechanism; Part 1 Development of Variable Valving System” JSAE Paper 891004, 1989.
38. Inoue, K., et al., “A High Power, Wide Torque Range Efficient Engine with a Newly Developed Variable Valve Lift and Timing Mechanism; Part 2 Engine Performance and Car Performance,” JSAE Paper 891005,1989.
39. Hosaka, T., “Development of the Variable Valve Timing and Lift (VTEC) Engine for the Honda NSX”, SAE Paper 910008, 1991.
40. Otobe, Y., et al., “Development of the High-Power, Low-Emission Engine for the “Honda S2000”, SAE Paper 2000-01-0670, 2000.
41. Shibano, K., Kamimaru, S., Yamada, T., Watanabe, K., “A Newly Developed Variable Valving Mechanism with Low-Mechanical Friction,” SAE 920451, 1992.
42. Hatano, K., et al., “Development of a New Multi-Mode Variable Valve Timing Engine,” SAE 930878, 1993.
43. Noguchi, M., Honda, H., Matsui, K., and Katsuki, T., “Introduction to Newly Developed Nissan VVL (Variable Valve Lift & Timing) Engine,” JSAE-97A056, Japanese Society of Automotive Engineers, 1997.
44. Brüstle, C., and Schwarzenthal, D., “VarioCam Plus – A Highlight of the Porsche 911 Turbo Engine,” SAE Paper 2001-01-0245, 2001.
45. Spiegel, L., et al., “Potentials of High Performance Engines with Variable Valve Systems,” Vienna Motor Symposium, April 2002.
46. Kawase, H., et al., “Development of the VVTL-I: A Variable Valve Train System for High Performance Engines,” Toyota Technical Review, Vol. 50, No. 2, March 2001.
47. Shikida, T., et al., “Development of the High Speed 2ZZ-GE Engine,” SAE Paper 2000-01-0671, 2000.
48. Fortnagel, M., Doll, G., Kollmann, K., and Weining, H., “Four Made of Eight – The New 4.3L and 5.0L V8 Engines”, MTZ Special Edition, July, 1999, pp 58-63.
49. Allen, J., and Dopson, C, “The Development and Application of Two Camshaft Profile Switching Systems, “Institution of Mechanical Engineers. Seminar “Variable Geometry Engines,” London, June 1991. Paper 8, pgs 81-93.
50. Mueller, R.S., and Uitvlugt, M.W., “Valve Selector Hardware,” SAE Paper 780146, 1978.
51. Horie, K., “The Development of a High Fuel Economy and High Performance Four-Valve Lean Burn Engine,” SAE Paper 920455, 1992.
52. Fukuo, K., et al., “Honda 3.0L, New V6 Engine,” SAE 970916, 1997.
53. Matsuki, M., Nakano, K., Amemiya, T., Tanabe, Y., Shimizu, D., and Ohmura, I., “Development of a Lean Burn Engine with a Variable Valve Timing Mechanism,” SAE Paper 960583, 1996.
54. Honda Website www.Honda.com 55. GT Power Software, Version 5.1, Gamma
Technologies, Inc, Westmont, Illinois, June, 2001.
56. Sinnamon, J.F., Personal Communication, April, 2001.
57. Metghalchi, M., and Keck, J.C., “Laminar Burning Velocities of Mixtures of Air with Methanol, Iso-Octane, and Indolene at High Pressure and Temperature,” Combustion and Flame, Vol. 48, pp. 191-210, 1982.
58. Vibe, I.I., “Brennverlauf und Kreisprozeb von Verbrennungs-Motoren,” VEB Technik, Berlin, 1970.
59. Personal Communication, Fogarty, R., Kunz, T., Lee, J., Delphi Corporation, April, 2002.
NOMENCLATURE AND UNITS
A/F Air-Fuel Ratio ATDC After Top Dead Center BDC Bottom Dead Center BMEP Brake Mean Effective Pressure (kPa) BTE Brake Thermal Efficiency (%) C Centigrade CA50 Crank Angle for 50% Mass Burned (CAD) CAD Crank Angle Degrees CDA Cylinder Deactivation CR Compression Ratio CSHC Cold-Start Hydrocarbon Emissions CVI Closed Valve Injection DICP Dual Independent Cam Phasing DOHC Double Overhead Cam Shaft DSZ Downsizing ECP Exhaust Cam Phasing ECR Effective Compression Ratio EGR Exhaust Gas Recirculation EIVC Early Intake Valve Closing EVO Exhaust Valve Opening (CAD ATDCf) FE Fuel Economy FMEP Friction Mean Effective Pressure (kPa) FTP Federal Test Procedure GCR Geometric Compression Ratio GITE Gross Indicated Thermal Efficiency (%) GTP GT Power (Engine Simulation Software) HC Hydrocarbon Emissions (g/kW-hr) HCCI Homogeneous Charge Compression Ignition HLA Hydraulic Lash Assembly (for Type II RFF) HLC High-Lift Cam ICP Intake Cam Phasing IMEP Indicated Mean Effective Pressure (kPa) IVO Intake Valve Opening (CAD ATDCf) L Lift (mm) LIVO Late Intake Valve Opening
LLC Low-Lift Cam LPVT Laboratory Programmable Valve Train MAP Manifold Absolute Pressure (kPa) MBT Minimum Spark Advance for Best Torque (CAD) NEDC New European Drive Cycle NMEP Net Mean Effective Pressure (kPa) NOx Oxides of Nitrogen Emissions (g/kW-hr) NGIMEP Net Gross Indicated Mean Effect Pressure
(kPa) NPMEP Net Pumping Mean Effective Pressure (kPa) NGITE Net Gross Indicated Thermal Efficiency (%) NVH Noise, Vibration, and Harshness OVI Open-Valve Injection P Cylinder Pressure (bar) PF Coef. for Cyl Pres Effect in Eng. Friction Model PMEP Pumping Mean Effective Pressure (kPa) PR Push Rod RFF Roller Finger Follower ROC Radius of Curvature (mm) RPM Revolutions per Minute SL Laminar Flame Speed (m/s) SP Mean Piston Speed (m/s) SRFF Switchable Roller Finger Follower SUV Sport Utility Vehicle TC Tumble Coefficient TDC Top Dead Center TI Tumble Index TR Tumble Ratio VDA Valve Deactivation VIM Variable Intake Manifold VL Valve Lift (mm) VLC Valve Load Control VLF Vehicle Load Factor VT Valve Train VVA Variable Valve Actuation
