10PFL-0711
CVT-Based Full Hybrid Powertrain Offering High Efficiency at Lower Cost Patrick Debal, Saphir Faid, Laurent Tricoche and Steven Bervoets Punch Powertrain
Brecht Pauwels and Kevin Verhaeghe PsiControl Mechatronics Copyright © 2010 SAE International
ABSTRACT In 2007 Punch Powertrain started the development of a full hybrid powertrain concept based on its CVT. A performance and efficiency analysis proved that a post-configuration offered the best solution. In parallel to the mechanical and electrical development an advanced, Matlab/Simulink® simulation system was established. A robust powertrain strategy was developed and implemented into the simulation system. Results show a potential of 30% to 70% fuel efficiency improvement depending on the cycle (type approval and real world cycles). A higher saving potential is possible as a plug-in. The fuel efficiency improvement is reached while meeting other important targets. First of all, the powertrain cost premium needs to match the saving. Next to keeping the transmission cost under control the electric drive technology and the batteries are cornerstones of the powertrain development. A dedicated switched reluctance electric motor/generator is developed at a partner. Switched reluctance combines high efficiency and dynamic behaviour with a low cost potential. Special care has been taken to iron out some drawbacks like torque ripple and noise. LiFePO4 is the preferred battery chemistry. It offers the best combination of performance and cost without the safety risk of classic lithium ion or polymer cells. Additionally, the powertrain size is very restricted. The development team at Punch Powertrain managed to keep the powertrain length, width and height within the size of the conventional counterpart. This enables a straightforward integration into most engine bays. As such, Punch Powertrain offers a fairly easy hybridization path for conventional cars. During the development project the search for auxiliary components was a continuous effort to find affordable components that do not jeopardize the development targets for mass, space and efficiency. For most components a “hybrid compatible” solution was found. The hydraulic pump for the transmission was one exception. A parallel development was initiated during the project. A demonstrator vehicle was built to drive as EV mid 2009 and as full hybrid by the end of 2009. In parallel a second powertrain is built to undergo a series of tests on bench to validate and optimize the powertrain strategy. By exchanging powertrain control modifications between the demonstrator and the test bench driveability will be guaranteed while further optimizing the fuel economy. Experience from both test platforms will be used in Page 1 of 14
the industrialization project. This project will result in a production ready powertrain design as well as a flexible production system for small to medium production series.
INTRODUCTION In 2006 Punch International took over the Belgian CVT transmission plant of ZF with a clear target: the development of hybrid powertrains for small and medium passenger cars. The company was renamed Punch Powertrain. As Tier-1 supplier Punch Powertrain set ambitious targets to assure the hybrid powertrain is competitive and appealing to OEMs at the time of introduction and during the following years. Next to fuel efficiency improvements the development project also targets low cost and easy vehicle integration. This paper focuses on the chosen parallel topology, the general optimization strategy, the technology and components selection and the control system development. Simulations for different target vehicles are performed with detailed component maps. The fuel consumption target is well within reach. Functional hardware tests have started in 2009. Performance tests will be performed in 2010.
DEVELOPMENT TARGETS The strategic view of Punch Powertrain is to develop a next generation hybrid powertrain. This powertrain needs to lower the current barriers for OEMs to offer hybrid versions of their vehicles. To reach this goal ambitious targets were set at the beginning of the project.
FUEL SAVING The main target is a fuel saving of minimum 25%/15% on the NEDC-cycle with gasoline/diesel cars while realizing similar savings in real traffic. These savings need to be realized by the powertrain only without recharging the battery. This implies no other changes are applied to the vehicle than those required for accommodating the hybrid powertrain. Additional measures taken by the OEM like engine efficiency improvement, mass reduction and streamlining allow realizing further fuel consumption reduction. The similar savings of greenhouse gases can help to meet the CO2 emission target of 130 g/km set forward by the European Commission or equivalent standards in other regions.
EV-RANGE An EV-range of at least 13 km (8 miles) at city traffic speeds fits into policies of some major European cities to reduce harmful emissions in city centers. When benefits from incentives can be gained or taxes (e.g. the congestion tax in London) can be avoided the cost premium for the hybrid powertrain can be partially or totally compensated in a short period.
INTEGRATION WITHOUT COMPROMISES As Punch Powertrain intends to supply the hybrid powertrains to OEMs the impact of the powertrain on the vehicle should be minimal. Next to the adaptations to make the vehicle “hybrid ready” other required changes need to have the smallest possible impact. Consequently, the hybrid version of a vehicle must have the same functionalities as the conventional counterpart, with respect to luggage space, the full use of folding seats and model choice. Also the powertrain excluding the battery system needs fit in the existing engine compartment. Meeting this target allows a reduced application development cost and time at the OEM side. Also market adoption will not Page 2 of 14
be hindered by hybrid versions of passenger cars with reduced handiness or restricted availability over the model range.
TARGET SEGMENTS Punch Powertrain targets the most popular vehicle segments of small and medium passenger cars and small vans in Europe. The hybrid powertrain can replace conventional powertrains with naturally aspirated gasoline engines up to 2.8l. Vehicle segments using such conventional powertrains can be provided with the hybrid powertrain under development. Both Toyota and Honda have proven that in the compact vehicle segments a considerable number of hybrid vehicles can be sold.
COST TARGET Punch Powertrain has set strict cost targets to lower the barrier for OEMs and their customers to buy vehicles with this powertrain. Buyers in the target segments are very cost sensitive. Therefore the cost premium for the hybrid must clearly allow a return on investment.
Figure 1: Hybrid Powertrain by Punch
HYBRID STRATEGY DEVELOPMENT The fuel saving target immediately ruled out a mild hybrid powertrain. Another choice was less obvious. As developer of transmissions Punch Powertrain had the freedom to choose between two configurations. Page 3 of 14
HYBRID CONFIGURATION The first option is to link the electric motor/generator with the powertrain before the CVT variator. This configuration is applied in mild hybrids as applied in the Honda Insight/Civic and Hyundai Accent (aka Kia Rio). Hyundai applied the VT2 CVT from Punch Powertrain in its hybrid. Lotus in the UK developed the EVE, a full hybrid demonstrator also using the VT2 CVT from Punch Powertrain. These examples use a thin, sandwiched flywheel motor. Alternatively, a transmission developer can also integrate the electric motor elsewhere on the transmission for a similar configuration. At Punch Powertrain the configuration with the motor before the CVT is called the “PRE” configuration. The other configuration, hence called the “POST” configuration, links the electric motor/generator behind the variator. pre-transmission Post-transmission
post-transmission Pre-transmission
battery charge mode
battery charge mode
electric drive mode
electric drive mode
Invertor Battery System
Invertor Battery System Electric Motor/ Generator Engine
Engine Electric Motor/ Generator
Figure 2: Post and Pre Configuration For both configurations a hybrid strategy was elaborated in depth. A first step in this strategy development is to create a holistic view on the optimization principles. To reach the ambitious fuel reduction targets a system optimization rather than a component optimization(1) is required. New mathematical models were developed to derive the operating areas by applying the optimization principles for the different hybrid modes for both configurations. After optimizing the strategy for both configurations simulations were run on different type approval as well as real world drive cycles. The POST configuration yields 8 to 10% more fuel consumption reduction over the NEDC cycle as well as real traffic cycles. Additionally, the EV-range for the POST configuration is substantially longer than for the PRE-configuration. Hence Punch Powertrain opted for the POST-configuration although this requires a more powerful electric motor/generator.
BASIC SIMULATIONS Initially, a backwards(2) calculation scheme was used to implement and refine the strategies. This scheme was first based on simple component characteristics as a first validation of the optimization principles. The required torque at the wheels is calculated backwards to engine and motor torque. This allowed an initial sizing of the electric motor/generator and the battery system.
Page 4 of 14
Figure 3: Combined Efficiency of Engine with CVT Gradually, the calculation scheme was extended with more complex component maps. This required a migration from a spreadsheet based tool to a Matlab® based application. Eventually, the calculation scheme contained very detailed component maps. Although it was unable to simulate transient events like clutch closing the calculation scheme yielded realistic fuel consumption results for non-hybrid vehicles. Therefore it was decided it could be used for comparative calculations. Efficiency Engine + CVT - 1000 cc 3 cyl - 3300 rpm 35% 30%
Efficiency [%]
25%
EV mode
20% Generate mode
15%
Assist mode
Conventional mode
10% 5% 0% 0
25
50
75
100
125
150
175
200
225
Torque [Nm]
Figure 4: Use of the Hybrid Modes depending on Torque Levels at a Given Speed The calculation scheme was used for comparing both the PRE and POST configuration. Overall the POST configuration resulted in higher fuel savings on both type approval cycles and real world cycles. The main Page 5 of 14
advantage of the PRE configuration is higher launch acceleration from stand still due the torque multiplication by the variator. Due the higher fuel saving potential Punch Powertrain opted for the POST configuration.
HIGH-END, DYNAMIC SIMULATIONS In parallel with the development and use of the calculation scheme, a highly detailed and dynamic hybrid powertrain simulation was developed in Matlab®/Simulink® by using the SimDriveLine® toolbox. The POST configuration strategy was carried over from the calculation scheme and further refined. The powertrain simulation tool uses a forward approach, i.e. a driver action causes the powertrain to change its operation like in real vehicles. Furthermore, the inertia of all powertrain components as well as highly detailed component models are included. This allows fully simulating the transient behavior inside the powertrain and adopting the strategy to obtain a good driveability (low jerk level). Tritec 1000 3 cylinder + CVT - MOL cycle 200 Max
31%
30%
29%
28%
27%
25%
23%
20%
15%
10%
5%
BYD
HEV
180
160
Secondary Torque [Nm]
140
120
100
80
60
40
20
0 0
500
1000
1500
2000
2500
3000 3500 4000 Secondary Speed [rpm]
4500
5000
5500
6000
6500
7000
Tritec 1000 3 cylinder + CVT - MOL cycle 200 Max
31%
30%
29%
28%
27%
25%
23%
20%
15%
10%
5%
BYD
HEV
180
160
Secondary Torque [Nm]
140
120
Assist
100
EV 80
Conv
60
40
20
Generate
0 0
500
1000
1500
2000
2500
3000 3500 4000 Secondary Speed [rpm]
4500
5000
5500
6000
6500
7000
Figure 5: Operating Points and Actual Modes during a Real World Cycle Page 6 of 14
The powertrain control logic for the high-end simulation was developed in Stateflow®. This tool allows an easy migration of the control logic into embedded software by autocoding.
COMPONENT AND SUBSYSTEM SELECTION The selection of the batteries and the electric motor/generator are two important cornerstones in how the hybrid strategy can realize the fuel saving target. It is mandatory that the benefits realized at the conventional side (engine plus variator) largely outweigh the losses at the electrical side. Consequently, efficiency was the major criterion in the selection process while cost, mass and size were also important.
ELECTRIC MOTOR/GENERATOR DRIVE The combination of the electric motor/generator and its power electronics need to realize a high efficiency over a wide speed and torque range. At the same time the drive needs to have a high power density and a low cost. An investigation of available technologies, products and suppliers resulted in the choice for switched reluctance (SR) motor/generator. This type of electric motor/generator best fits the above requirements. Punch Powertrain decided to partner with PsiControl mechatronics for the development and production the SR motor/generator. PsiControl mechatronics has more than 10 years experience in industrial drivelines with SR, mainly for weaving machines. PsiControl mechatronics has over 300.000 axes installed worldwide.Thisd makes them an authority in SR
Figure 6: SR Motor - Sample Stator, Rotor and Actual Prototype Page 7 of 14
SR motors have the inherent advantage of their robust design, both mechanically and electrically. Their straightforward mechanical structure is also advantageous for cost compared to other motor topologies when produced in similar quantities. An SR motor has no natural boundaries, neither in torque nor speed compared to conventional motors In contrast with most industrial applications, a motor/generator in an automotive hybrid powertrain does not stay in a nominal point of operation for most of the time, but varies continuously over the full speed and torque range. Therefore PsiControl mechatronics needed to implement a novel development approach, especially concerning efficiency evaluation and comparison, and the thermal modeling. For both, a reference cycle was used. PsiControl mechatronics developed a dedicated SR motor/generator capable of 200 Nm and 30 kW (both peak ratings, available for 30 s). This torque and power is provided in a package with diameter 225 mm and length 251 mm (external dimensions). The motor has a continuous power rating of 15kW and can deliver 90Nm continuously at standstill. From other hybrid projects Punch Powertrain learned that SR motor/generators are generally known for high torque ripple at very low speeds and for noise. Both these issues were identified as serious barriers for market acceptance due to reduced driveability at vehicle launch and driving comfort. Torque ripple is mainly caused by the salient poles and are therefore inherent to the design of the motor, but can be greatly reduced through the application of advanced control techniques as shown in Figure 6 for low speed operation. The resulting torque ripple is low enough such that is does not induce vibrations and noise in the system, or is felt by the passengers.
Figure 7: Torque Ripple Reduction in SR Motor/Generator Page 8 of 14
Although no specification on maximum dB rating was set in advance, because meeting such a limit can be done while still producing a noise quality that will not be accepted by OEMs and the public, the motor/generator is designed with low noise generation in mind. From previous studies was known that the noise is highly influenced by the mechanical structure and by the control method. Signifying improvements on both aspects with respect to previous designs were incorporated in the design of the SR motor/generator.
BATTERY SYSTEM The efficiency and the EV-range target rule out the use of NiMH batteries. A NiMH battery system would be too large, too heavy and too expensive for the application. The common Lithium chemistries as currently used for laptops and cell phones pose a potential but serious safety risk especially when these cells are scaled to capacities as required for hybrid vehicles. Therefore Punch Powertrain opts for the LiFePO4 chemistry. This emerging chemistry combines sufficient efficiency, usable SoC range, power and energy density, cycle life and safety.(3)
Figure 8: Prototype Battery System An evaluation of different suppliers of a combination of LiFePO4 cells and battery management systems yielded a preliminary short list of possible battery system suppliers. One battery system was ordered early in the project. Later on two more arrived for different vehicle and test bench tests as well as for assessments of battery cells and battery management systems. Meanwhile different interested companies in Flanders have joined Punch Powertrain in a project for testing and assessing different cells from a selected group of cell manufacturers. In this additional project also information and knowledge about batteries and battery management is gathered by and shared between the different partners. As such it provides an undeniable support for the hybrid powertrain development. During the project the short list of battery system suppliers evolves based on new insights and information. It is considered as preliminary also because LiFePO4 technology is still immature and important players are still expected to emerge. Page 9 of 14
TRANSMISSION DESIGN The transmission design is based on the current VT2 CVT. Because this transmission is a thoroughly optimized for conventional powertrains its adaptation for the POST configuration was not straightforward. The prototype hybrid transmission currently used on test benches and in a demonstrator vehicle is therefore considered as an interim solution, mainly for functional testing and demonstration purposes. A high volute chain is used to connect the electric motor/generator to the secondary shaft of the CVT transmission. This connection, at the engine side of the transmission, causes minimal changes to the rest of the transmission and allows a fairly long electric motor/generator. Consequently a large number of parts is carried over from the conventional CVT.
Figure 9: Hybrid Transmission with Electric Motor/Generator and Chain Drive. New parts in exploded view. Further changes are related to removing the conventional reverse drive. All reverse driving will be done in EVmode. To provide oil in EV-mode to the transmission, the internal, engine driven oil pump is replaced by an external, electrically driven oil pump. During the whole development process priority was given to getting a functional hybrid transmission running as soon as possible. Design changes to apply later versions for performance improvements and cost reductions were recorded. They will be applied together with the lessons learned from the functional prototype. By meeting the tight sizing constraints the hybrid powertrain has the same powertrain length as the conventional version. Therefore it will fit in nearly any engine bay that can accommodate a conventional powertrain with Punch Powertrain’s CVT. This is illustrated Figure 10.
Page 10 of 14
Figure 10: Comparison between Conventional and Hybrid Transmission Once the hybrid powertrain has proven its potential a production intent design will be made. This new transmission will feature several efficiency improving components and designs. Today most building blocks of this next design are known.
CONTROL SYSTEM The simulation tool mentioned earlier does not stand by itself. It is the first stepstone in the development of an embedded hybrid powertrain control system. The logic developed in the Matlab®/Simulink® + Stateflow® environment can be converted into embedded control software with a minimum of hand coding. Punch Powertrain has acquired different dSpace® rapid control prototype hardware and software tools to implement the hybrid control logic from the simulation in prototype controllers. The hybrid development project is the first project at Punch Powertrain where the complete software of an embedded controller will be generated by the autocoding process. The hybrid control unit (HCU) is the master controller in the powertrain. The throttle pedal actuation by the driver is converted by the HCU into a torque demand for both the engine and the motor/generator and a ratio setting for the transmission. The hybrid powertrain control continuously sets the most efficient operation parameters. The strategy is to a large extent robust. This implies that in a variety of operating conditions the powertrain can operate at an efficiency that is substantially higher than the conventional powertrain. Consequently a substantial fuel consumption reduction is achieved for different cycles in charge sustaining mode.
Page 11 of 14
POWERTRAIN MANAGEMENT
HCU
BCU
CAN
ECU
TCU
MCU
BMS
EMG ICE
+ CVT
PowEl
Bat
Figure 11: Powertrain Control Architecture
RESULTS AND SHORT TERM PLANS The calculations as well as the simulations have shown that the target fuel consumption saving is within reach. Currently results better than the 25% saving on the NEDC-cycle are obtained while other cycles also yield high savings. The first tests of the SR motor are also promising with respect to torque and efficiency.
Figure 12: Hybrid Powertrain in Smart ForFour on Chassis Dynamometer Two prototype powertrains are built. One powertrain will undergo functional testing as well as strategy validation tests on a dedicated powertrain test bench. Due to the economic climate the investment in this test Page 12 of 14
bench has been delayed. In parallel the second powertrain is built into a demonstrator. To show the potential of the hybrid powertrain with the focus on its compact size a Smart ForFour was selected. This vehicle was built on the Mitsubishi Colt platform. Its size is similar to the Ford Fiesta or Toyota Yaris. Currently the powertrain in the Smart is being calibrated for driveability. When the driveability has been achieved the calibrated control system will be implemented on the powertrain on the test bench. Then the powertrain strategy validation and efficiency testing will start. Tests so far with the Smart Forfour as well as a transmission check have revealed no major issues with the transmission. The SR motor/generator low speed torque ripple improvement as described in the section "Electric Motor/Generator Drive" was applied on the vehicle after the initial assessments revealed an improvement was required. Similarly, the vehicle has been driven in pure electric mode to evaluate the SR motor/generator noise. Even though the vehicle had no doors and a completely open engine compartment as shown in Figure 12 the SR motor/generator noise was exceeded by other noises in the vehicle during tests on normal road surface.
PRELIMINARY CONCLUSION The hybrid powertrain under development at Punch Powertrain is made to be built into standard vehicles without too much modifications and compromises at the OEM side. It will provide a substantial fuel saving compared to similar conventional powertrains under most to all circumstances. The functional transmission design is a very good basis for the later final design. Only minor modifications are required. The SR electric motor/generator performs as expected or better with respect to vehicle acceleration, torque ripple and noise. Both the transmission and the electric motor/generator design are in line with the cost target. The hybrid powertrain is also compact When the strategy validation proves that the fuel saving can be realized Punch Powertrain and PsiControl mechatronics have developed an efficient hybrid powertrain at a lower cost than systems available today.
REFERENCES 1. Pasquier, M., “Continuously Variable Transmission Modifications and Control for a Diesel Hybrid Electric Powertrain”, SAE-Paper, CVT 2004-34-2896, 2004 2. Van Mierlo, J. and G. Magetto, “Innovative Iteration Algorithm for a Vehicle Simulation Program”, IEEE Transactions on Vehicular Technology, Vol. 53, No. 2, March 2004. 3. Bauer, S., "New Lithium Ion Technologies, Performance Characteristics for Different Fields of Applications", Advanced Battery Technologies, July 2008
CONTACT INFORMATION ir. Patrick Debal, Project Manager R&D - Development Hybrid Powertrain PUNCH Powertrain Industriezone Schurhovenveld 4 125, BE-3800 Sint-Truiden, Belgium Tel. +32 11 679 266 - Fax +32 11 679 230 e-mail
[email protected] ir. Brecht Pauwels, R&D Engineer PsiControl mechatronics Karel Steverlyncklaan 13, B-8900 Ieper, Belgium Page 13 of 14
Tel: +32 57 409 635 - Fax: +32 57 409 697 e-mail:
[email protected]
ACKNOWLEDGMENTS The development of the hybrid powertrain at Punch Powertrain is supported by the Flemish Government as an IWT industrial research and development projects. The IWT is the Institute for the promotion of Innovation by Science and Technology in Flanders.
DEFINITIONS/ABBREVIATIONS Bat BCU BMS CO2 CVT ECU EMG EV HCU ICE LiFePO4 MCU NEDC NiMH OEM PowEl SoC SR TCU
Page 14 of 14
Battery system Brake control unit Battery management system Carbon dioxide Continuously variable transmission Engine control unit Electric motor/generator Electric vehicle Hybrid control unit Internal combustion engine Lithium iron phosphate Motor control unit New European drive cycle Nickel metal hydride Original equipment manufacturer Power electronics State of charge Switched reluctance Transmission control unit
