2
Development of Vehicle Subsystems

CANDIDATE SYSTEMS

The ultimate success of the PNGV program will be measured by its ability to integrate R&D programs that collectively improve the fuel efficiency of automobiles within the very stringent boundary conditions of size, reliability, durability, safety, and affordability of today's cars. At the same time, the vehicles must meet even more stringent emission and recycling levels and must use components that can be mass produced and maintained in a manner similar to current automotive products.

In order to achieve a Goal 3 fuel economy that approaches the 80 mpg target (80 mpg is about three times the fuel efficiency of today's comparable vehicles), the energy conversion efficiency of the chemical conversion system (e.g., a power plant, such as a CIDI engine, a gas turbine, a Stirling engine, or a fuel cell) averaged over a driving cycle will have to be at least 40 percent, approximately double today's efficiency. This is an extremely challenging goal and will require assessing many possible concepts for improving efficiency. For example, the PNGV high fuel economy level of 80 mpg will require the integration of the primary power plant with energy storage devices, as well as the use of lightweight materials for the vehicle structure to reduce vehicle weight.

Despite concerted efforts in the last year to develop and evaluate the various candidate systems, none of the energy conversion power trains being considered meets all of the constraints. Therefore, R&D programs on both the selected and nonselected candidate systems have to be continued after the 1997 technology selection for the first concept vehicles, with the objective of attaining the breakthroughs that would make one or more of the technologies viable for meeting Goal 3 requirements. In 1997, the PNGV identified a stretch research objective of



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 20
2 Development of Vehicle Subsystems CANDIDATE SYSTEMS The ultimate success of the PNGV program will be measured by its ability to integrate R&D programs that collectively improve the fuel efficiency of automobiles within the very stringent boundary conditions of size, reliability, durability, safety, and affordability of today's cars. At the same time, the vehicles must meet even more stringent emission and recycling levels and must use components that can be mass produced and maintained in a manner similar to current automotive products. In order to achieve a Goal 3 fuel economy that approaches the 80 mpg target (80 mpg is about three times the fuel efficiency of today's comparable vehicles), the energy conversion efficiency of the chemical conversion system (e.g., a power plant, such as a CIDI engine, a gas turbine, a Stirling engine, or a fuel cell) averaged over a driving cycle will have to be at least 40 percent, approximately double today's efficiency. This is an extremely challenging goal and will require assessing many possible concepts for improving efficiency. For example, the PNGV high fuel economy level of 80 mpg will require the integration of the primary power plant with energy storage devices, as well as the use of lightweight materials for the vehicle structure to reduce vehicle weight. Despite concerted efforts in the last year to develop and evaluate the various candidate systems, none of the energy conversion power trains being considered meets all of the constraints. Therefore, R&D programs on both the selected and nonselected candidate systems have to be continued after the 1997 technology selection for the first concept vehicles, with the objective of attaining the breakthroughs that would make one or more of the technologies viable for meeting Goal 3 requirements. In 1997, the PNGV identified a stretch research objective of

OCR for page 20
0.01 g/mile for emissions of particulate matter for the CIDI engine. The current target is 0.04 g/mile. Meeting the stretch research objective presents new challenges to the candidate CIDI engine, which would require expanded technology development to meet the PNGV goals. PNGV would also need to reevaluate other power plants relative to CIDI engines. The hybrid electric vehicle (HEV), which is the PNGV power train of choice, uses an energy storage device to decrease the fluctuations in the demands on the primary power plant. This reduction allows for a decrease in the peak power output required from the primary energy conversion system and an opportunity to improve efficiency both by restricting the power fluctuations and by recovering a significant fraction of the vehicle's kinetic energy during braking operations. The PNGV is sponsoring research on batteries, flywheels, and ultracapacitors as energy storage devices. Achieving the high fuel economy levels for the Goal 3 vehicle will require more than improving the energy conversion efficiency of the power train (including energy converters and transmissions) and reducing other energy losses in the vehicle. Vehicle weight reduction through the use of new vehicle designs and lightweight materials will be extremely important in achieving the very ambitious fuel economy targets. The committee re-evaluated the candidate energy conversion and energy storage technologies, as well as candidate electrical and electronic systems, that were considered last year and addresses them in this chapter. This chapter also reviews progress on advanced structural materials for the vehicle body, a subject that was not addressed by the committee in its third review. The technologies evaluated in this chapter are listed below: four-stroke CIDI engines continuous combustion systems fuel cells electrochemical storage systems electro-mechanical storage systems electrical and electronic power-conversion devices materials The committee reviewed R&D programs on each of these technologies to assess the progress that has been made and the developments required for the future. The PNGV Technical Roadmap, which has been updated for most of these technologies, provided a good summary of the program goals (PNGV, 1997). In the committee's opinion, the PNGV has made substantial progress in assessing the potential of most candidate systems and identifying critical technologies that must be addressed to make each system viable. A few exceptions are noted in the sections describing specific technologies. The committee has also described some international developments in the various technology areas, based both on its own knowledge and experience and

OCR for page 20
on selected information gathering activities, but an extensive review of worldwide developments was not part of its task. Nevertheless, the issue of global competitiveness of the U.S. automotive industry is a key consideration in the development of advanced automotive technologies. INTERNAL COMBUSTION RECIPROCATING ENGINES The research team on the four-stroke direct injection (4SDI) engine has evaluated four engine configurations as candidate power plants: the CIDI engine, the homogeneous charge compression ignition engine, the gasoline direct injection (GDI) engine, and the homogeneous charge spark ignition engine. The PNGV has indicated that at this time the CIDI engine has the potential for the highest fuel conversion efficiency. Furthermore, because of the increased penetration of automotive diesel engines into the European market and a technical and manufacturing maturity that falls within the PNGV program schedule, there is a high level of confidence in the assessments of future improvements in CIDI engine performance. In addition to better fuel economy, the performance of the CIDI engine is superior to other engine types in terms of evaporative, cold start, and hydrocarbon and carbon monoxide (CO) emissions. However, there are still significant challenges facing the development of CIDI engines that can meet the PNGV targets. The challenges include reducing the emissions of nitrogen oxides (NOx) and particulates, reducing the weight of the power plant, and reducing costs. The CIDI engine is being considered as a possible stand-alone power plant, as well as part of either a series or parallel HEV. The trade-off of fuel economy and weight involved in adding energy conversion devices with a hybrid vehicle design must be carefully evaluated because an increase in vehicle weight results in a decrease in fuel economy. Program Status and Progress The 4SDI team was very active this past year. A five-year comprehensive plan was developed, and the technical developments required for each component of a CIDI engine for a PNGV vehicle were identified. Technologies that would enable an advanced CIDI engine to meet Goal 3 objectives would include four valves per cylinder; a common rail, electronically controlled fuel injection system; a variable-geometry turbocharger; exhaust-gas after-treatment for NOx and particulates; electric actuators; and an aluminum block. It would be fueled with a very low-sulfur diesel or alternative fuel. This engine would be significantly different from current diesel engines, which typically have two valves per cylinder; rotary pump fuel injection systems; fixed-geometry turbochargers; oxidation catalysts in the exhaust; pneumatic actuators; cast-iron structures; and use conventional diesel fuel. Technical developments are expected to reduce NOx

OCR for page 20
and particulate emissions; reduce noise vibration and harshness (NVH); improve power density; and improve fuel economy. Last year, the 4SDI team identified five high-priority areas for research: lightweight engine architectures; dimethyl ether (DME) as an alternative fuel for CIDI engines; combustion-related processes; lean NOx catalysis; and alternative fuels for CIDI engines. The 4SDI team has been active in all five areas in the last year. Lightweight engine structures are being investigated to reduce vehicle weight. Ford, for example, is testing the DIATA (direct-injection, aluminum-block, through-bolt assembly) engine, a 1.2-liter displacement engine designed to produce 45 kW/l. The design is a lightweight engine that achieves state of the art NVH. Under a contract with the U.S. Department of Defense Tank Automotive Command, Ricardo, Inc., has developed a preliminary design for an all new, three-cylinder, lightweight, high speed, direct injection engine. The main objective of this program was to establish an engine architecture compatible with lightweight materials. The key challenge is placing the lightweight materials under compression, and through-bolt assembly was considered the most promising way to accomplish this. Key aspects of this engine architecture are expected to be used in the Chrysler-U.S. Department of Energy HEV program. One of the critical challenges to the CIDI engine is the so-called trade-off between emissions of soot (particulates) and NOx. In current diesel engines, methods used to reduce NOx (typically increased exhaust-gas recirculation and retarded injection timing) result in increased soot emissions and vise versa. It is usual to display the diesel emission characteristics on a graph of soot versus NOx. Obtaining a net benefit in emissions requires decreasing overall emissions toward the origin of the operating curve, rather than along the soot-NO x trade-off curve. The CIDI engine technology under development is targeted to achieve a 0.04 g/mile particulate emissions level or better by (1) limiting the application of engine controls that reduce NOx (e.g., exhaust gas recirculation) in order to minimize energy-out particulate emissions and (2) lowering NOx emissions to target levels using catalytic after-treatment. There is some leeway in implementing in-cylinder NOx reduction strategies, at the expense of increasing particulate emissions, while still meeting the total emission design target. Meeting the stretch research objective of 0.01 g/mile particulate emissions will require simultaneous reductions of soot and NOx and cannot be met by manipulating the soot-NOx trade-off relationship. Therefore, breakthrough improvements in engine controls to reduce emissions and for exhaust-gas after-treatment for both NOx and particulates will be required, as well as significant changes in fuels. The research objective for particulate emissions, therefore, will require fundamental investigations of the in-cylinder combustion process with the objective of altering the interaction between soot and NOx emissions. To this end, PNGV has begun research on combustion fundamentals, such as combustion control through electronic control of the fuel-injection process and assessing the interaction

OCR for page 20
between the combustion chamber geometry, the fuel injection, and fuel-air mixing. Ford and FEV have also been pursuing new technologies in fuel injection rate-shaping using piezoelectric techniques. Many collaborative programs have been put in place in the past year. The 4SDI technical team participated in Vice President Gore's Technical Symposium Number 6 on 4SDI engines, which consisted of five sessions held over two days in the summer of 1997. The PNGV has established a Fuels Working Group and an Aftertreatment Working Group to develop PNGV strategy and plans. Cross-cutting teams were established to promote interchanges between the light-duty CIDI engine researchers and the heavy-duty diesel engine industry. For example, Chrysler is working cooperatively with Detroit Diesel Corporation to integrate a three-cylinder, 1.5-liter displacement, direct-injection, turbocharged, intercooled engine with hot and cold exhaust gas recirculation into an HEV. A DME fuel system has been designed and a follow-on program established. Efforts to develop combustion systems are being augmented by experimental work at Sandia National Laboratories (SNL) and Wayne State University with computational support from the University of Wisconsin, Madison. SNL, Oak Ridge National Laboratory (ORNL), and Los Alamos National Laboratory (LANL) are pursuing improved lean NOx catalysts, and Pacific Northwest National Laboratory (PNL) and Lawrence Livermore National Laboratory (LLNL) are investigating plasma-assisted NOx reduction catalysis. A reformulated diesel fuel testing program is under way at EPA, and a USCAR-supported auto/energy fuel testing plan is being developed. Technical Targets The critical characteristics of a CIDI engine that can meet the PNGV performance targets are shown as a function of milestone targets in Table 2-1. All of the USCAR partners have made good progress towards meeting these targets. A comprehensive evaluation of the Chrysler Generation I 1.46-liter, three-cylinder engine has revealed general conformance with the 1997 targets with respect to part load brake thermal efficiency, exhaust emissions of NOx and particulates, based on a 14-mode test protocol. The one-meter noise assessment is in conformance with the 1997 target. Peak thermal efficiency is 2.5 percentage points below the 1997 target; however, improvements are expected in both the Generation I and II versions. Both displacement and weight-specific power are below target for 1997 by 5 and 13 percent, respectively. The latter shortfalls will be addressed by the Generation II design, which will incorporate more lightweight materials than the Generation I version. Mount vibration measurements have not yet been made. Initial engine dynamometer tests have been made on the Ford Research 1.2-liter, four-cylinder engine. Displacement-specific power and engine noise results compare favorably to the 1997 PNGV targets. The part-load thermal

OCR for page 20
TABLE 2-1 Critical Characteristics of the CIDI Engine vs. PNGV Milestone Targets     1995 1997 2000 2004 Characteristic Units Target Target Target Target Best brake thermal efficiency % 41.5 43 44 45 Displacement specific power kW/L 35 40 42 45 Power specific weight kW/kg 0.50 0.53 0.59 0.63 Cost per kW $/kW 30 30 30 30 Durability 1,000 miles 150 150 150 150 NVH (one meter noise) dBA 100 97 94 90 Engine-out NOx emissionsa g/kW-hr 3.4 2.7 2.0 1.4 Engine-out particulatesa g/kW-hr 0.3 0.25 0.20 0.15 FTP 75 NOx emissions in 2,500 lb ETW vehicle g/mile 0.6 0.4 0.3 0.2 FTP 75 particulate emissions in 2,500 lb ETW vehicle g/mile 0.08 0.06 0.04 0.04b Source: Based on Table III. F-1 in PNGV (1997). Acronyms: NVH = noise, vibration, and harshness; FTP = federal test procedure; ETW = emissions test weight. aRepresentative values for operation over the FTP cycle bIn 1997, PNGV identified a stretch research objective for particulate emissions of 0.01 g/mile. efficiency and emissions levels are very calibration-specific and are under development, making comparisons with existing engines difficult. Peak thermal efficiency is below the 1997 target but is also under development. The weight-specific power and package volume PNGV targets were established assuming a three-cylinder engine. This four-cylinder engine is still below the weight-specific power targets and the package volume targets. Based on testing of GM's single-cylinder CIDI research engine, projected emissions data meet the 1997 PNGV targets. Other targets cannot be assessed from tests on a single-cylinder engine. Because of the soot-NOx trade-off inherent in CIDI engines, the more restrictive research objective for particulate emissions would alter the basis on which the CIDI engine has been evaluated as a potential PNGV power plant. To achieve the stretch objective, technological breakthroughs will be necessary for the CIDI engine to meet the PNGV milestones. A consequence of the more ambitious 0.01 g/mile research objective for particulate emissions is that fuel characteristics are now more important for meeting the PNGV goals. For example, the 0.01 g/mile particulate research objective corresponds to an emission of sulfates for a fuel with approximately 50 ppm of sulfur and a vehicle with a fuel economy of 80 mpg. Therefore, at a minimum, sulfur levels in the fuel will have to be drastically reduced from the current limit of approximately 500 ppm to about 50 ppm.

OCR for page 20
The stretch objective would make the CIDI engine a high-risk candidate for meeting the PNGV goals. Meeting the stretch research objective for particulate emissions increases the overall challenge of meeting the exhaust emission standards. Reducing NOx emissions is still one of the biggest challenges for the CIDI engine. Engine-out emissions appear to be 0.5 g/mile or greater, whereas the current emission target is the Tier 2 federal NOx limit of 0.2 g/mile. Meeting these limits will require after-treatment of NOx, with a NOx conversion efficiency of 60 to 90 percent. Demonstrated after-treatment efficiencies are currently less than 40 percent, and many technologies under development require near sulfur-free fuel. Clearly, a high-efficiency, sulfur-tolerant after-treatment device for NOx must be developed for the CIDI engine to be a viable option. In addition to the higher priority of fuels technology, which should include investigating alternative fuels, such as DME and Fischer-Tropsch diesel fuel, the development of exhaust-gas after-treatment technologies must also be expanded to include methods for reducing particulate emissions. Current Program Elements Advances in engine combustion, exhaust-gas after-treatment and fuels technology will be necessary to meet the stringent PNGV emission requirements. The current program includes work on some aspects of all three of these technologies. High-Pressure Fuel Injection Systems and Combustion Fundamentals Significant advancements in combustion control will be necessary to meet the low emission targets without sacrificing fuel economy. The in-house research programs of all the USCAR partners on the fundamentals of combustion and emissions formation are being augmented by work at the national laboratories and universities. The use of electronically controlled, high-pressure fuel injection systems as a means of combustion control is being investigated by the heavy-duty diesel engine industry. Next-generation electronic fuel injectors may allow for dynamic rate-of-injection profiling, in addition to multiple injections per cycle. NOx, particulates, and fuel economy can be significantly affected in heavy-duty diesel engines by manipulating fuel injection. The extent to which these advanced technologies can be used to improve combustion in small CIDI engines, such as those that would be used in a PNGV concept car, is not known. The smaller CIDI engine will be operating at a higher speed than the typical heavy-duty diesel engine, the quantity of fuel injected will be smaller, and the combustion chamber will be smaller, so that surfaces will be closer together and fuel jets from the injectors will impinge on those surfaces; the injector holes, however, will be approximately the same size. As a consequence, the smaller high-speed CIDI

OCR for page 20
engines have a shorter injection duration and less time available for mixing, so that controlling combustion through injection manipulation is uncertain. This is a serious technical challenge for the 4SDI team. Interaction between the PNGV and the heavy-duty diesel engine industry via the crosscutting team is an appropriate way for PNGV to address this issue. Exhaust-Gas After-treatment Exhaust-gas after-treatment represents one of the most challenging aspects of the 4SDI program. The treatment of exhaust gas, both of NOx and of particulates, will be required to meet the program goals. Cooperative programs with Argonne National Laboratory (ANL), SNL, LANL, LLNL, ORNL, and PNL are already established. Both lean-NOx catalysis and plasma-assisted after-treatment approaches are being investigated. Yet progress in this area has been slow. In addition, the performance of those technologies at the present state of development is adversely affected by sulfur in the fuel. The best ''full brick" catalyst1 with diesel engine exhaust, with diesel fuel added as a reductant, reduces NOx by up to 37 percent at steady state over a narrow temperature range. Other reductants provide a higher efficiency but would require an auxiliary source of reductants onboard the vehicle. The fuel economy penalty of using a reductant is approximately 1 percent. Plasma-assisted catalysis, a process in which an electrical potential difference generates nitrogen ions that combine with NO to form molecular nitrogen and atomic oxygen, shows promise of reducing both NOx and particulate emissions. The current state of the art of plasma exhaust treatment technology requires that a catalyst also be used to maximize the reduction of the NOx emissions. The results of early laboratory tests have been encouraging, but the technology must still be demonstrated on an engine. Estimates of the fuel economy penalty for plasma systems are in the range of 2 to 5 percent. Particulate and NOx traps are also being considered. Issues of cost, durability, and regeneration capacity remain for particulate traps, and the engine control systems for momentary fuel enrichment to release the NOx from the trap and subsequent catalytic reduction are complex challenges that must still be met. Despite the challenges, the emphasis on exhaust-gas after-treatment of NOx and particulate matter will continue. Breakthroughs will be necessary for the development of a sulfur-tolerant, long-life, effective, passive NOx removal device. In September 1997, PNGV representatives met with four major catalyst suppliers to discuss cooperative development. 1   In the vast majority of laboratory catalyst tests, the catalyst is in the form of a powder. However, in current commercial catalytic converters, most of the catalysts are in the form of a coating on a monolithic structure. "Full brick" means that the catalyst test was conducted with a monolithic structure.

OCR for page 20
Fuels Technology The relationship between the physical and chemical characteristics of a fuel and the emissions from an engine is the basis for government regulations on fuel properties. In diesel fuel, for example, the cetane number, aromatic content, and sulfur levels are all subject to regulation. However, the relationships between the chemical characteristics of a fuel and its physical properties, such as viscosity, lubricity, and cetane number, are extremely complex. For example, the cetane number of a fuel can be increased either by decreasing the aromatic content in favor of longer-chain paraffins or by adding cetane improvers. Both fuels will have similar overall combustion and emissions characteristics as assessed by today's metrics. In an engine designed to meet PNGV goals, slight differences in the composition and properties of fuels may also have a significant effect on the performance of after-treatment devices, such as NOx catalysts. Fuels containing oxygen generally produce less soot, which might be a basis for reducing emissions. However, the incremental costs and the effects on the infrastructure of this change in fuel composition have to be considered. Certain components in fuel, such as sulfur, can affect both combustion and exhaust-gas after-treatment systems. Sulfur in the fuel can contribute to soot emissions by forming sulfates; sulfur can also deactivate the exhaust catalyst. Indeed the PNGV recognizes the importance of the interactions between fuel and engine performance and is involved in programs to reduce emissions through fuel changes. These programs are all based on the target of 0.04 g/mile for particulate emissions. The stretch research objective of 0.01 g/mile requires PNGV to change its research goals accordingly and concentrate on alternative fuels. PNGV should establish a cooperative program with the U.S. transportation fuels industry (see Chapter 5). The PNGV already has some programs in place to evaluate alternative fuels. Under a contract issued in cooperation with the of U.S. Department of Defense Tank Automotive Command, AVL List GmbH conducted an assessment of DME as an alternative fuel for diesel engines, and DME is a candidate for further investigation. Because the physical characteristics of DME are much like those of propane, significant infrastructure changes would have to be made if DME were chosen. EPA is also evaluating reformulated and alternative fuels. The committee believes that all of these programs should be continued and that all alternative fuels should be investigated. However, it appears that none of the operating regimes for any of the candidate engines will meet the design targets with the research objective (0.01 g/mile) for particulate emissions. Therefore, the potential for altering the soot-NOx trade-off by manipulating the fuel formulation and the subsequent impact of fuel composition on exhaust-gas after-treatment devices is a critical issue for the 4SDI program, which must now investigate the engine and fuel as an integrated system. The fuels industry should be involved in this important area of development and

OCR for page 20
research, just as it is in the European Auto Oil program and the Japan Clean Air Program (see Chapter 5 and Jones, 1997). International Developments Automotive Diesel Engines The small direct-injection diesel engine is widely used in automobiles in Europe. Currently about half of all new cars sold in Europe are diesel powered, and it is estimated that as much as 30 percent of the entire fleet could be diesel powered by the year 2000. This market shift is being motivated by a combination of tax policies favoring the use of diesel fuel over gasoline, the high cost of fuel, and the consequent consumer demand for fuel-efficient vehicles. Diesel-powered automobiles are not as prevalent in Japan as they are in Europe, but they have a higher market penetration than in the United States, where less than 1 percent of the automobiles are diesel powered. Because there is little market demand for automotive diesel engines in the United States, domestic industries have little incentive to pursue critical technologies in this area. Therefore, it is not surprising that technical leadership in the critical areas of fuel injection and electronic control is outside the United States. The world leaders in automotive diesel engine injection and control technologies are probably Bosch (in Germany), Denso (in Japan), and Lucas CAV (in the United Kingdom). However, some important work is being done in the United States by Caterpillar and Navistar on automotive applications of electronically controlled, common rail, hydraulically amplified injection systems. The partners in PNGV are well aware of the advancing state of the art in automotive diesel engines. In fact, through their foreign affiliates, they are participating in the development of these engines. GM has developed the Ecotec engine, and Ford has developed the DIATA engine in their respective European operations. Chrysler is involved in developing a state-of-the-art automotive-size diesel engine through working agreements with both a domestic and an international company. Although the United States cannot claim technical leadership in the general area of automotive-size diesel engines, the PNGV is well aware of the current state of the art and directions in development of this power plant. The USCAR partners are fully capable of utilizing this technology worldwide through their foreign affiliates and international agreements. Gasoline Direct Injection Engines In Japan, Mitsubishi, Toyota, and Nissan have introduced GDI engines to their domestic markets, claiming fuel efficiency improvements of 20 to 30 percent over the conventional spark-ignition engine vehicles. The PNGV partners have been following developments closely but have concluded that GDI engines

OCR for page 20
would have a lower fuel conversion efficiency than CIDI engines and, like CIDI engines, they would not meet the U.S. emission standards (see Appendix B).2 The belief that the maximum fuel economy of the GDI engine would be less than that of a CIDI engine was reported to the committee during the Phase 3 review by Dr. Peter Herzog of AVL (Herzog, 1996). Hence, the GDI spark-ignition engine was not listed as a candidate power plant for the PNGV concept vehicles. As part of the Phase 4 review, Dr. Ando of Mitsubishi Motors presented impressive results of recent developments on their GDI engines (Ando, 1997 a,b; Iwamoto et al., 1997; Kume et al., 1996). Mitsubishi claims that significant improvements have been made in the total performance of the engine by changing the intake manifold design, altering the in-cylinder flow pattern, maximizing the distance between the fuel injector and the spark plug, carefully matching the fuel injector characteristics to the cylinder flow at different loads, and taking full advantage of the capabilities of advanced electronic controls. At this time Mitsubishi Motors believes the new design of the GDI engines will be able to meet the stringent European and U.S. low-emission vehicle (LEV) standards with fuel conversion efficiencies within 1 percent of CIDI engines. Mitsubishi estimates that by the year 2000 85 percent of the engines they produce will be GDI engines. The advancements of the GDI engine claimed by Mitsubishi represent technical strides for this power plant. If these claims of improved performance can be realized, the GDI engine would be a viable competitor to the CIDI engine. However, even if the GDI engine meets the LEV standards at a fuel conversion efficiency within 1 percent of the CIDI engine, it is not known if it will meet the PNGV emission targets, which are the ultra low emission vehicle (ULEV) standards. The PNGV partners are aware of the GDI programs in Japan and are assessing the potential of the GDI as a power plant for a PNGV vehicle. The committee feels that this assessment should continue. Assessment of the Program Excellent progress has been made in the past year in all aspects of the 4SDI program. However, the identification of a stretch research objective for particulate emissions of 0.01 g/mile presents significant additional challenges to the 4SDI program in developing the CIDI engine as a PNGV power plant. The prospect of developing a CIDI engine that can meet this research objective is high risk and would make a reevaluation of other candidate engines and system configurations necessary. To maximize the probability of success, the 4SDI program may have to be augmented and redirected. The 4SDI team must determine if the new stretch research objective can be met by operating the engine in previously 2   This view is detailed in Figure III.F-1 of the 1997 update of the PNGV Technical Roadmap (PNGV, 1997).

OCR for page 20
excel in accepting the high power generated during engine and vehicle braking and delivering it to meet vehicle system peak power needs. The issues of safety, cost, and size are still serious but are yielding to development programs. Work continues at ORNL to develop more specific flywheel design guidelines to support fast-response power-plant systems. A fast-response power plant, which has a similar response time to a conventional engine, places a much lower requirement on the power output of the flywheel than a slow-response power plant. A vehicle system model is now available that can be used for simulations to optimize the performance of the flywheel and to update the PNGV technical targets, which have been unchanged since 1996 (see Table 2-2). Flywheel designs will not be pursued for slow-response power plants because the much greater energy demands would require a larger and more costly flywheel system. Program Status, Progress, and Plans The PNGV flywheel technical term is now confident that it is possible to design and build a practical prototype energy storage flywheel system for automotive applications. A significant amount of work on failure containment has provided more confidence that the system can be designed to comply with the established safety criteria. Furthermore, improved containment designs have reduced cost and weight, which lends further support to the expectation that a practical system can be designed. A key technology development is the design of an adequate containment mechanism in case of failure. The flywheel technical team has followed several strategies and has essentially overcome this significant barrier. Perhaps the most important advance is the growing evidence that flywheels (or portions thereof) that fail at low stress-to-strength ratios do not "burst" but remain intact. This knowledge dictates that the rotating parts have a high ultimate strength-to-maximum operating stress ratio (about 4:1). Retaining "loose flywheels" is significantly easier than containing fragments because of the increased time for energy dissipation. The new design strategy for flywheel housings are designed to retain loose flywheels and contain fragments from partial flywheel failures instead of containing a complete burst and disintegration of a flywheel. This design strategy also attempts to manage energy as it dissipates. Limiting the use of flywheels to fast-response power plants reduces the energy storage requirement and permits the design to meet the safety goal for strength-to-stress ratio with a manageable increase in weight. A variety of flywheel containment tests have been run in the Trinity/LLNL project, which led to a lightweight stainless steel/aluminum honeycomb structure that has so far been tested 41 times in sample form and has demonstrated consistent retention of test projectiles. The current overall state of development for a flywheel system with a power level of 30 kW and an energy storage of 300 Wh indicates that, although the

OCR for page 20
projected cost and weight are still well above PNGV targets, substantial improvements have been made in the last year. The assumptions in the projections include a cost of $5/lb for carbon fiber material, a containment-to-rotor weight ratio of 2:1, and a cost of $4/kW for power electronics. The first two projections appear to be achievable by 2004, but the $4/kW target is $3/kW lower than the $7/kW target set for the power electronics overall, which is already considered an extremely ambitious cost target. The relative simplicity of the flywheel electronic controls might justify these lower cost estimates and target, but the committee is not convinced of this. The cost for the flywheel, including the containment housing and motor generator, is substantially higher than the target; the weight is also significantly over its target. The amount of effort that will be required to achieve the desired targets is clearly much less than was projected a year ago but is still sizable. The flywheel technical team noted that if the power plant system required a flywheel with a capability of 15 kW power and 200 Wh energy-storage, current estimates of cost and weight would practically meet the targets. A failure mode and effects analysis for the vehicle system incorporating a flywheel was conducted at ORNL, and a flywheel simulation model has been provided to the systems analysis team by LLNL. Assumptions for the current flywheel system are that life-cycle tests will demonstrate that the vacuum in the flywheel chamber will be maintained and that the sealed bearings will continue to operate with acceptable characteristics. The flywheel technical team and ORNL have yet to confirm the basic shape of the flywheel system, which may have implications for gyroscopic effects and containment costs. The development of analytical models will continue at ORNL and LLNL in 1998 to support design assumptions and containment design and development. The flywheel design assumptions will be written to support performance of a design level failure model and effects analysis, which in turn will be used to support requests for quotes for the flywheel system in 1999. With further refinements of the vehicle system model with iterations of flywheel system capabilities, estimates for the targets, which are currently considered rough projections, will be improved. The Trinity/LLNL project will continue testing the containment with a full-size housing subjected to overspeed burst. Assessment of the Program The flywheel technical team is confident that it is now possible to design and build practical energy storage flywheel systems for automotive applications. The team, working with a number of outside agencies, all of whom are providing corroborating data of various kinds, has a firm grasp on the key technology barrier, namely, containment of failure. The committee suggests, however, that an out-of-balance sensor be considered to shut off electrical power during flywheel run-up if a significant amount of out-of-balance vibration is detected. The committee believes the flywheel system should be included in second-generation

OCR for page 20
concept vehicles to help clarify how the vehicle system and power plant will incorporate the energy captured by the flywheel and enable the development of a delivery system that optimizes overall cost, size, performance, and efficiency. Lowering the cost and reducing the weight of the flywheel system are still necessary, but with a smaller flywheel system, the tasks will be easier to accomplish because the power and energy storage requirements will be lower. The committee believes that the relatively large cost penalty of $300 for the flywheel system will be extremely difficult to offset by reductions elsewhere in the vehicle system. Recommendation Recommendation. If vehicle systems modeling indicates an acceptable level of performance and cost for the flywheel, the PNGV should plan for the physical installation of flywheel hardware in a post-2000 concept vehicle. POWER ELECTRONICS AND ELECTRICAL SYSTEMS All three USCAR partners have elected to pursue HEV designs for the 2000 concept vehicle (Malcolm, 1997). Fuel cells for energy generation and flywheels for energy storage could very well be practical within the PNGV time frame. Electrification of major auxiliary functions, e.g., air conditioning and power steering, is attractive for ease of control and improved energy management and efficiency. Success depends on the development of efficient and economically acceptable actuators, motors, and power electronic converters. Program Status and Progress The electrical and electronics power conversion devices team (EE technical team) has made considerable progress in organizing and coordinating its efforts. The committee had noted in the third review that this team lacked leadership and had recommended that a full-time leader be appointed (NRC, 1997). Not only has this been done, but two technical subteams have also been established to address power electronics enablers and electric motor enablers. The subteams appear to be effectively setting priorities and addressing the important issues in their areas. Both have made designing and manufacturing for low cost their highest priority. The progress of the EE technical team since the committee's last review is evident in the team's performance as measured against the targets for specific power, volumetric power density, cost, and efficiency. Except for motor efficiency, the 1997 targets for both the motor and electronics are reported to have been either met or exceeded (Malcolm, 1997). The reported cost of $15/kW for the power electronics module is particularly impressive in comparison to the 1997 target of $25/kW and is extraordinary in comparison to the interim technical

OCR for page 20
target of $100/kW by 1997 that was indicated in the PNGV Technical Roadmap (PNGV, 1997). The EE technical team has also made progress in leveraging the activities of other organizations, especially the Office of Naval Research (ONR), Wright-Patterson Air Force Base, and the National Renewable Energy Laboratory. The success of the PNGV power electronics developments appears to rely heavily on ONR's Power Electronic Building Block (PEBB) program. The EE technical team has established a liaison with the PEBB researchers and has been working with the PEBB program management to focus attention on PEBB specifications that are applicable to the PNGV program. Although the EE team has made considerable progress in working more closely with the systems analysis team, some necessary models have still not been provided to the systems analysis team, including models for motor/generators, power electronic converters, and control algorithms for both the series and parallel hybrid drive configurations. The EE technical team is currently working on providing them. Accessory loads, heating, ventilation, and air conditioning (HVAC), and regenerative braking were identified by the EE technical team as priorities for technology development. Work is being done on starting/charging and accessory loads by both the USCAR partners and their suppliers. In the committee's third review, regenerative braking efficiency was identified as an area of concern (NRC, 1997). Because this topic was not addressed in presentations for this (fourth) review, the committee has concluded that regenerative braking efficiency is still an outstanding issue. The EE technical team is aware of HVAC-related work being done at the national laboratories and is coordinating the electrical requirements of the HVAC system with developments as they are evaluated and adopted by the systems analysis team and the systems engineering team. The USCAR partners have independently chosen HEV designs for the year 2000 concept vehicle. This means that each company may have different electrical system architectures and requirements. Assessment of the Program Both motor and power electronics technology will meet PNGV functional requirements. The remaining challenge for the EE technical team, as the team has clearly stated, is meeting the cost targets for these components. The 2004 target cost for motors is $4/kW, which the team recognizes as the approximate cost of materials in state-of-the-art motors. Although the team apparently already has achieved the very low cost of $15/kW for power electronics, the PNGV Technical Roadmap target for 2004 requires a further reduction of more than 50 percent to $7/kW. In the committee's opinion, the impact of the cost of power electronics on other vehicle systems has either not been recognized or has been considerably underestimated. Although the EE technical team has tried to provide models to

OCR for page 20
the systems analysis team, coordination with other program teams could still be improved. The committee was not made aware of assessments of foreign technology by the EE technical team. However, given the introduction of commercial HEVs by foreign manufacturers (e.g., the Toyota Prius) in the past year, the committee believes that the USCAR partners are aware of these developments and have been conducting independent evaluations of foreign technology. The hybrid designs chosen independently by the USCAR partners will make a more comprehensive evaluation of competing technologies possible in an application where no single correct approach is obvious, especially for the electrical and electronic subsystems. However, the committee has not seen any evidence that the cost target of $7/kW for electronic power modules can be achieved in any of the designs. Given that the cost of less sophisticated computer power supplies produced in high volume is about $100/kW, the committee is concerned that the cost goals established by the EE technical team are too aggressive. The committee is also concerned that the EE technical team's reliance on the ONR's PEBB program will jeopardize the cost targets if the PEBB program changes direction, loses funding, or cannot meet its goals. Recommendations Recommendation. The PNGV should perform a thorough and convincing verification of the reported 1997 cost of power electronics in conjunction with a reevaluation of the 2004 cost goal. The reevaluation should include identifying developments that support the assumption that the cost target can be met. Recommendation. The PNGV electronics and electrical systems team's reliance on the Navy's Power Electronic Building Block (PEBB) program should be mitigated by the initiation of a PNGV-specific cost reduction program. Recommendation. Given the critical importance of electrical and electronic systems to the success of the PNGV program, the electrical and electronic systems technical team should provide cost models to the systems analysis team as soon as possible to ensure that cost assumptions for other subsystems that rely on power electronics are consistent with the projections and targets. MATERIALS The reduction of vehicle mass through the use of lightweight materials is one of the key elements in meeting the fuel economy goal of the PNGV program. Because the leading candidate materials are currently more costly than the steel used today, the requirement that there should be no increase in the overall cost of vehicle ownership presents a major hurdle to meeting the fuel economy goal.

OCR for page 20
PNGV personnel are working closely with materials suppliers to develop less costly manufacturing processes and new design practices that utilize materials more efficiently. Each lightweight material alternative also offers a different weight reduction, has different costs, and raises different issues in terms of manufacturing feasibility, design experience and confidence, infrastructure needs, new failure modes, repairability, and recyclability. To meet the 80 mpg vehicle fuel economy objective of Goal 3 while maintaining vehicle performance, size, utility, and cost of ownership, vehicle curb weight will have to be reduced by 40 percent from 3,240 to 1,960 lb. Table 2–3 shows the breakdown of the weight reduction targets by major vehicle subsystem. Program Status The major alternatives under consideration in 1997 for reducing vehicle weight were: more efficient design of the current steel-intensive vehicle, which is being led by the American Iron and Steel Institute; aluminum sheet and castings; fiber-reinforced composites; magnesium; matrix composites; titanium; and lightweight glazing (thinner glass and polymers). More Efficient Steel Design The approach of the American Iron and Steel Institute program has the potential of a 20 percent weight reduction for the body-in-white (BIW) structure (bolt-on panels, such as the hood, doors, front fenders, and deck lids are not included in the BIW). Based on efficient steel design technology, the weight of the baseline PNGV vehicle BIW could be reduced from 598 to 478 lbs; the cost could be reduced by $154. The PNGV material team also studied the possibility of using a stainless steel space frame but found that it would generate a cost penalty of $200 for a weight saving of only 22 percent, which is half of the savings needed for the Goal 3 vehicle. Aluminum and Magnesium The total potential vehicle weight savings with aluminum sheets and castings is 600 lbs; the total potential weight savings with magnesium castings alone is 150 lbs. Note that the weight saving potentials for the alternative materials are not additive because certain parts, wheels for example, have been targeted for both materials in the individual computations. On a part-by-part basis, the weight saved by substituting steel sheet with aluminum sheet is typically more than 50 percent. This number does not include secondary weight savings that result from reducing the size of other components, such as brakes, wheels, and suspensions. The total vehicle must be redesigned around the primary weight saving to assess accurately potential secondary savings.

OCR for page 20
TABLE 2–3 Vehicle Weight Reduction Targets for the Goal 3 Vehicle Subsystem Current Vehicle (lbs) PNGV Vehicle Target (lbs) % Mass Reduction Body 1,134 566 50 Chassis 1,101 550 50 Power train 868 781 10 Fuel/other 137 63 55 Curb Weight 3240 1960 40   Source: Stuef (1997). The automotive industry has considerable production experience with aluminum in the form of stamped body panels. The procedures and processes for recycling aluminum are also in place today, and much of the material is returned to high value automotive applications. As a matter of public record, several aluminum-intensive prototype vehicles have been built outside the PNGV program by the USCAR partners and evaluated for ride, handling, NVH, crashworthiness, joining, and painting (Jewett, 1997). Thus, the change to an aluminum-intensive vehicle would not be a major technology challenge because the USCAR partners already have extensive design and manufacturing expertise with this material. The challenge is to develop new processing methods so that an aluminum-intensive vehicle can be made as inexpensively as a steel vehicle. Based on a price for aluminum of $1.60/lb, the cost penalty of an aluminum BIW is estimated at $400. Cast aluminum and magnesium would be used in the chassis, the body, and the power train subsystems. Major efforts to reduce the costs of feedstock and improve the casting processes of both materials are under way. Studies to improve the machinability of magnesium castings and to develop a lower cost high-temperature alloy are also under way. Improved processes for recycling magnesium will have to be developed. Fiber-Reinforced Plastic In 1997, a number of serious obstacles were identified that will limit the use of graphite fiber-reinforced plastic composite material (GrFRP) in meeting Goal 3. First, not enough is known about the consistency of the mechanical properties of GrFRP when produced in large volumes. Second, the criteria and methods for reliably designing GrFRPs for fatigue resistance and crashworthiness in the complex loading environment presented by the automotive body structure are far from production-ready. Third, methods for recycling the material into high-value applications to take advantage of its intrinsic properties, as opposed to using it only as filler, must still be developed.

OCR for page 20
According to the latest PNGV/USCAR studies, the potential BIW weight reductions with GrFRPs are only a few percentage points better than for aluminum, 59 percent versus 55 percent. The committee was surprised at the reportedly small margin of improvement of GrFRPs over aluminum. PNGV/USCAR explained that the small margin reflected a combination of two factors. Manufacturing considerations and the need for sufficient structural strength to accommodate multi-axial loads required more material than the committee had expected. The major barriers to the intensive use of GrFRPs for a Goal 3 vehicle, however, are the high cost of the graphite fibers and the lack of a suitable high-volume manufacturing process for the material. Currently, the cost penalty of a GrFRP BIW is higher than a steel BIW, and there is no feasible high-volume manufacturing process for the thin sections. Glass fiber-reinforced plastic composites (FRPs) offer weight savings in the 25 to 35 percent range. FRPs that incorporate thermoset resins have been used extensively in noncritical stressed structures, such as hoods, deck lids, doors, and fenders. One drawback of using thermoset materials is the substantial investment in tooling because of the relatively slow cycle times. Chrysler is investigating low-cycle-time injection molding for glass-reinforced thermoplastic resins, which has additives for improving crashworthiness and weather resistance. Chrysler is considering using them in both body panels and body structures. Computer simulation and hardware tests are being used to test the crashworthiness of these materials. Metal Matrix Composites The applications for metal matrix composites are in the chassis and power train subsystems. The total potential weight savings of using metal matrix composites is only 30 to 50 lbs. The major hurdles to developing applications of this material are feedstock costs and the development of a reliable process for compositing the materials. Titanium The applications for titanium are in the chassis (40 lbs potential savings) and power train (10 lbs potential savings). The components of interest are springs, piston pins, connecting rods, and valves. The major barrier to the use of this material is its high feedstock cost. Lightweight Glazing New glazing materials—thin glass and polymers—are being considered for weight reduction. The potential weight saving is 50 lbs. The major concerns for polymers are abrasion/scratch resistance and cost.

OCR for page 20
Program Progress and Plans Steel The American Iron and Steel Institute efficient steel vehicle design program, which will be completed in 1998, offers near-term weight and cost savings that can be implemented in goal 1 and 2 applications, once the USCAR partners have verified the findings. However, the intensive use of steel is not feasible for meeting the Goal 3 weight reduction targets unless there is a major breakthrough in power train efficiency. Aluminum Sheet In the past year, aluminum has become a leading structural material candidate for the Goal 3 vehicle technology selection process because (1) it offers a much larger percentage weight reduction than steel, (2) it is much less costly than GrFRP, (3) the knowledge base for the design and manufacture of this material is extensive, and (4) existing stamping facilities for steel can be used for aluminum without major modifications. Because of aluminum's importance to weight reduction and high fuel economy, a major cooperative research and development agreement (CRADA) has been initiated between Reynolds Metals, LANL, and the USCAR United States Automotive Materials Partnership to reduce the cost of aluminum sheet through the development of a thin-slab (less than 1 in thick) continuous-casting process to replace the more costly ingot-based process used today. The preliminary results of this program are very promising. Another program to develop low-cost, non-heat-treatable alloys competitive with the 6000 series aluminum alloys is also under way, as well as studies on improving the formability of aluminum stampings and making the walls of aluminum extrusions thinner. The goal of all of these programs is to reduce the cost of an aluminum-intensive vehicle. Graphite-Fiber-Reinforced Plastic Composite Material The weight saving of GrFRPs compared to aluminum is not sufficiently attractive for this material to be the leading candidate in the Goal 3 technology selection process. Cost and thin-section manufacturing are major issues that cannot be resolved in time to support Goal 3 year 2000 concept vehicles. The development of GrFRPs should be continued with applications targeted beyond Goal 3. R&D should concentrate on reducing the fiber cost to $ 3/lb, developing a process for mass producing components in thin sections, acquiring a deeper understanding of reliable design for complex loading conditions, and developing high-value applications for the recycled material. The injection molding of FRPs (glass-reinforced thermoplastics) being

OCR for page 20
investigated by Chrysler looks much more promising at this juncture. Although the potential weight saving of 25 to 35 percent is not as great as it would be with GrFRPs, FRP may compete effectively with aluminum for body panels and certain structural applications because the processing technology could integrate two or more parts that are currently made with steel or aluminum into a single part. Nevertheless, the extensive design and production experience with aluminum for body panels and structural applications gives FRPs an edge in the near term. Aluminum and Magnesium Castings Several important R&D studies were conducted in 1997 to reduce the costs and improve the properties of aluminum and magnesium castings, including sand casting, semi-permanent mold casting, squeeze casting, and high-pressure die casting. Simulation models of the casting flow and solidification processes are being developed to predict microporosity as a function of the part and casting process design. Improved nondestructive evaluation techniques are being developed to support the use of aluminum and magnesium in more demanding applications. The development of rapid tooling processes for die-casting applications is under way. Methods and materials for improving the die-casting dies are also being studied to reduce the overall cost of using these materials. Plans for 1998 The plans for 1998 are to continue the major material initiatives already under way. The committee agrees that the focus should be on following through on the cost-reduction initiatives begun in 1997, especially for the continuous casting of aluminum sheet. Assessment of the Program In 1997, the PNGV materials technology team and the vehicle engineering team made a thorough joint evaluation of the lightweight material candidates for the Goal 3 technology selection process. The criteria included potential weight savings, feedstock cost, manufacturing cost and feasibility, design and manufacturing experience, and the ability to recycle the material into high-value applications. The committee agrees with the criteria for selection used by the PNGV teams and with their conclusion that aluminum is the lightweight material of choice for intensive use in support of Goal 3 objectives, along with the selective use of FRPs, magnesium, GrFRPs, and titanium. The committee also agrees with PNGV/USCAR's assessment that intensive use of GrFRPs should be a longer range goal. The manufacture of an aluminum-intensive vehicle would not meet the Goal 3 cost objectives if the aluminum were produced with today's mill practice

OCR for page 20
technology. Consequently, the CRADA to develop continuous casting of aluminum sheet for automotive applications is critical to eliminating the cost penalty and, therefore, critical to meeting the overall objectives of Goal 3. Finding ways to improve the physical properties and reduce the cost of cast aluminum and magnesium is also important to meeting Goal 3 objectives. PNGV/USCAR should monitor these projects closely and ensure that they have adequate resources to meet these critical objectives. Recommendations Recommendation. The development of the continuous casting process of aluminum sheet should be given the highest priority in terms of resources and technical support. This includes support of the work already under way to characterize the material in terms of its microstructure, strength, ductility, formability, and weldability in parallel with the development of mill processing techniques. Recommendation. The development program for low-cost graphite fiber should be continued for longer-term applications beyond Goal 3. A new program for manufacturing graphite-fiber-reinforced plastic composite materials in thin sections should be initiated to take advantage of the unique properties of this material.