February 28, 1973
Page 5830
NATIONAL ACADEMY OF SCIENCES REPORT ON MOTOR VEHICLE EMISSIONS
Mr. MUSKIE. Mr. President, there has been much discussion recently of the possibility or impossibility of attaining the levels of auto emissions control required by the 1970 clean air amendments.
The 1970 Clean Air Act required the National Academy of Sciences to conduct an annual review of the technical capability of the auto industry to meet 1975-76 clean car standards. The second major report of the National Academy of Sciences Committee on Motor Vehicle Emissions has been released. Their conclusion is, and I quote:
Achievement of the 1975 standards may be technologically feasible and that achievement of the 1976 standards is likely but may not be attainable on the established schedule.
In perhaps the most important statement of the report, however, the National Academy of Sciences states that of the several technological alternatives available to meet the standards, the alternative currently being pursued by the auto companies is potentially the most expensive and least dependable of all the alternatives.
The material in this report is an important contribution to the current discussion about the automotive emission standards set by the Clean Air Act. For this reason, Mr. President, and because it may be weeks before the document is generally available to the public, I ask that the report of the National Academy of Sciences Committee on Motor Vehicle Emission be included in the RECORD at this point.
There being, no objection, the report was ordered to be printed in the RECORD, as follows:
NATIONAL ACADEMY OF SCIENCES
February 15, 1973.
THE PRESIDENT OF THE SENATE,
THE SPEAKER OF THE HOUSE OF REPRESENTATIVES,
THE ADMINISTRATOR OF THE ENVIRONMENTAL PROTECTION AGENCY.
SIRS: I have the honor to transmit a report summarizing the work and findings of our Committee on Motor Vehicle Emissions in accord with the provisions of Section 6 of Public Law 91-604, the Clean Air Amendments of 1970. We trust that this report will be of assistance to the Administrator of the Environmental Protection Agency in discharging his responsibilities under that Act and that it will inform the Congress of the progress which has been made, to date, toward achieving some of the goals of that Act.
The report constitutes a description, as of 1 February, of the "technological feasibility," on the part of the automobile and related industries, of achieving the automotive emissions control standards established by the Act. As the report reveals, that Act has stimulated an almost worldwide effort to develop effective emissions control systems. Of necessity, however, this report is presented at a time when the pace of development can readily overtake categorical conclusions based on the information available today; it is, therefore, a review of the current "state-of-the-art," presented while that state is changing rapidly, and not a summary of a stabilized situation. It is for that reason, inter alia, that the report presents an analysis but offers no recommendations concerning enforcement, on schedule, of the relevant provisions of the Act.
The Committee defined "technological feasibility" to mean that an emissions control system capable of meeting the standards set for the three major pollutants can be developed, designed, produced in large numbers, and maintained in service, all at reasonable cost. By these criteria, the Committee's analysis indicates that achievement of the 1975 standards may be technologically feasible and that achievement of the 1976 standards is likely but may not be attainable on the established schedule.
However, these seemingly definitive conclusions are offered with several reservations which are held in varying degrees of gravity by individual members of the Committee. The nature of these reservations will be found in the report. They are concerned, variously, with the durability in customer use of catalyst-dependent control systems, the requirement for a network of inspection and maintenance stations, the actual likelihood of sufficiently early development of a dual- catalyst system capable of achieving the 1976 standards, and the likelihood of manufacture for Model Year 1976, on a scale commensurate with projected total national production, of a sufficient number of vehicles actually capable of meeting the 1976 standards in customer use.
The Committee is seriously concerned that the certification procedure may not prove to be an adequate indicator of the continuing reliability of catalyst-dependent emissions control systems under the more stressful, varied conditions of consumer use. Data in this regard are not yet available, even for systems intended to meet the 1975 standards. To assure that vehicle classes certified for production actually do continue to meet the prescribed standards, the Committee considers it advisable to develop a network of inspection and maintenance stations and to train a corps of mechanics sufficient to that task. Some of the Committee, however, suggest that no more need be done than to enforce the recall provision of the Act, when so indicated by defective behavior of a reasonable sample of vehicles. It should be noted, however, that whereas that provision is binding upon the manufacturer, it is not mandatory for the vehicle owner to respond.
In view of the low response to recalls for defects relating to passenger safety (30 to 50%), simple use of the recall provision under these circumstances would not suffice to meet the goals of the Act. In this regard also, it should be noted that there is not available, for such national use, a relatively simple, foolproof, reliable, diagnostic instrument for assessment of the automotive emission of the three pollutants with which the Act is concerned. It may be necessary for the Environmental Protection Agency to stimulate the research and development required to make such instrumentation available on the schedule necessitated by the Act.
The Committee found it unnecessary and inadvisable to recommend a set of interim standards for 1975 or 1976 model year vehicles. But, while contemplating its responsibility for such a recommendation, under the terms of the contract, the Committee became aware of controversies surrounding many aspects of the problem of standard setting, e.g., the nature and magnitude of the hazards to health posed by the pollutants released in automotive emissions, the relationships among the various pollutants and their ambient concentrations with respect to their health effects, the relative contributions of mobile and stationary power sources, etc. Resolution of these controversies appears imperative to long-term policy with respect to the protection of air quality.
Hence, on page 127, the Committee urges that Congress and the Environmental Protection Agency initiate a comprehensive study of these and related matters. This Academy would he pleased to be of assistance in such an effort. That recommendation should not be interpreted as taking exception to the standards established by the Clean Air Act of 1970. Most of the Committee believes that only if such an examination were to reveal compelling evidence and arguments to the contrary should the effort to achieve the emissions control standards established by the Act be relaxed; indeed, the Committee is particularly concerned that continued progress be made with respect to improvement of air quality in those urban centers where, patently, automotive emissions have contributed significantly to the deterioration of the local environment.
A major quandary which the Committee wishes to place before the Congress and the Environmental Protection Agency (page 5) arises from awareness of the relatively recent development, largely in the hands of a Japanese manufacturer, of a dual-carbureted, stratified charge engine. Although the general principle is not new, the particular design in question, incorporated into small size engines, has met the 1975 certification standards and bids fair to meet the 1976 standards. As compared with the catalyst-dependent systems now being emphasized by the major manufacturers this system offers the promise of lower initial purchase costs, greater durability in service and significantly greater fuel economy. The Committee is concerned that mass production of what are presently deemed to be relatively fragile, catalyst- dependent systems, of unproved reliability in actual service, may engender an episode of considerable national turmoil. It is further concerned that, once committed to the manufacture of catalyst-dependent control systems, rather than switch to some more generally acceptable system such as a version of the stratified charge engine that now offers great promise, the relatively ponderous automobile industry will continue to manufacture catalyst-dependent systems for some years, albeit, presumably, while also seeking more durable catalysts and mechanisms to reduce the severe fuel penalty of current catalyst-dependent systems with their associated mechanical features. The dilemma, then, is to determine what course of action by government, would assure the earliest possible optimal outcome while scrupulously avoiding dictation, by government, of the technology to be used. The Committee offers no recommendations in this regard.
Relevant to this situation are the costs, per vehicle, associated with the initial purchase, maintenance and operation (including the effects on fuel consumption) of the various emissions control systems under consideration. The annualized incremental costs, viz., the cost per car/year for a standard engine, relative to a 1970 standard engine, due to the emission control system, for operation and maintenance of the vehicle with the purchase cost of the system amortized over the first five years of operation, were found, by the Committee to be as follows: 1973 engine, about $100; single catalyst system (1975 standards), about $225; dual catalyst system (1976 standards), about $270; and the dual-carbureted stratified charge engine, about $70. The high annualized costs of the catalyst systems reflect the serious associated fuel penalties.
For the nation, these costs represent a concrete example of the principle that the costs of environmental protection can be met only if they are internalized. The magnitude of this process derives from the great numbers involved, viz., about 107 (10 million) vehicles produced per year and almost 108 (100 million) registered automobiles. Thus, a one-year production of 107 vehicles equipped with the presently proposed dual-catalyst system would result in additional expenditures – as compared with 1970 automobiles – of $2.7 billion per year for each of the first five years of the life of that model year of cars, assuming constant fuel prices, and about $2 billion per year thereafter.
In due course, all vehicles will be equipped with emission controls capable of meeting the 1976 standards. Were all cars equipped with the dual-catalyst system, at current costs this would result in a national annual total expenditure for emissions control of the order of $23.5 billion (assuming a mean life of ten years/car with purchase cost amortized over the first five years, and current fuel prices).
Such figures are to be taken as no more than an indication of their orders of magnitude. The increased sticker price would tend to cause consumers to buy smaller cars of greater fuel economy and the fuel penalty would tend to reduce mileage. The initial costs will probably decline as dual-catalyst systems and their manufacture are improved. On the other hand these gains could be offset by the foreseeable rise in fuel prices. Unless satisfactory feedback control and fuel injection systems, for catalyst-dependent systems with the associated mechanical features, become available, the total costs will be dominated by the fuel penalty associated with such systems. These costs, in dollars and in depletion of fuel reserves, are so great that they should serve as a national incentive to hasten the development of reliable lower-cost alternatives to the dual-catalyst system as a solution to the problem of emissions control.
In contrast, several of the promising alternatives, such as the carbureted stratified-charge engine, carry with them costs of the order of those already associated with the 1973 engine, viz., an annualized cost for emissions control of about $100/car during the first five years of service, for an annualized total of perhaps $7.5 billion for the full fleet of 108 cars.
Costs of these magnitudes suggest, of themselves, the need for attention to a series of considerations which lie outside the scope of the present report. Among such questions are: What is the effect of this enterprise on the GNP? Is it, in effect, a stimulus to the overall economy or are the funds utilized to defray these expenditures removed from alternate uses in the economy, e.g., for improvement in the health care delivery system? If the answer to the latter is affirmative in significant measure, is this the wisest use of such funds for protection of the public health?
What effects would large-scale employment of the most promising emissions control systems have on the international balance of payments? If emissions control can only be satisfactorily accomplished by acceptance of a large fuel penalty, e.g., the dual-catalyst system or the Wankel engine with thermal reactor, what judgment should be made under such circumstances? Should it turn out that noble metal catalysts are more effective and, seemingly, economic than other catalysts, how should one weigh this raid on the very limited supply of this resource in the skin of the planet?
Whereas pollutant emission is undesirable anywhere, emission control does not appear, today, to be essential on the basis of either essentially aesthetic or health considerations in large areas of the nation. Indeed, overall, natural production of hydrocarbons, carbon monoxide and perhaps NOx far exceeds that from man-made sources. In view of the costs to the nation, in dollars and in
fuel consumption, of early implementation of the 1975-76 standards, attention seems warranted to the possibility of temporarily enforcing the established emission standards only in those specific urban areas where air quality is known to be adversely affected by automotive emissions, reserving national implementation for the day when there are available reliable, relatively inexpensive emissions control systems which exact no fuel penalty.
Emissions control is but one aspect of "the problem of the automobile" in our society. This device has given Americans an "automobility" unknown in previous human history, enriched the personal experience of each of us, broadened our horizons and helped to make the large expanse of American geography into one nation. But this aspect of our society has begun to be defeated by its very success.
The automobile has also accelerated depletion of several critical natural resources, including the petroleum which fuels it. It has scarred the land and choked the city, contributing seriously to deterioration of the quality of urban life. In the long run, the truly effective mechanisms for emission control must include a significant reduction in the number of cars operated in the city, a solution dependent upon acceptable public mass transit systems, and a substantial reduction in the mean size (weight, volume, and horsepower) of those automobiles which do function in the city, as well as, perhaps, redistribution of the pattern of physical relationships among dwelling and working areas. Patently, these are relatively long-term goals, achievement of which will require extensive, meticulous study and planning with subsequent large public expenditures and careful public intervention into the behavior of the private sector.
For the short term, however, automotive emissions control can be accomplished by a relatively simpler, technological "quick fix," and, perhaps, on the schedule established by the Clean Air Amendments of 1970. The attached report summarizes the status of the alternatives currently offered as means whereby to achieve the earliest acceptable technological solution to this problem.
Respectfully yours,
PHILIP HANDLER,
President.
NATIONAL RESEARCH COUNCIL, NATIONAL ACADEMY OF SCIENCES, NATIONAL ACADEMY OF ENGINEERING,
Washington, D.C., February 12, 1973.
Dr. PHILIP HANDLER,
President, National Academy of Sciences, Washington, D.C.
DEAR PRESIDENT HANDLER: I am herewith transmitting for submission by the Academy to the Congress and to the Environmental Protection Agency the Report of the Committee on Motor Vehicle Emissions dated February 12, 1973.
Sincerely yours,
E. L. GINZTON,
Chairman.
REPORT BY THE COMMITTEE ON MOTOR VEHICLE EMISSIONS, DIVISION OF ENGINEERING, NATIONAL RESEARCH COUNCIL IN ACCORDANCE WITH SECTION 202(c) OF THE CLEAN AIR AMENDMENTS OF 1970 AND IN PARTIAL
FULFILLMENT OF CONTRACT No. 68-01-0402, BETWEEN THE ENVIRONMENTAL PROTECTION AGENCY AND THE NATIONAL ACADEMY OF SCIENCES
NOTICE
The study reported herein was undertaken under the aegis of the National Academy of Sciences and with the express approval of the Governing Board of the National Research Council. Such approval indicated that the Board considered that the problem is of national significance; that elucidation and/or solution of the problem required scientific and technical competence and that the resources of the National Research Council were particularly suitable to the conduct of
the project.
The members of the committee were selected for their individual scholarly and technical competence and judgment with due consideration for the balance and breadth of disciplines. Responsibility for the detailed aspects of this report rests with the committee, to whom we express our sincere appreciation.
Reports of our study committees are not submitted for approval to the Academy membership.
The report was reviewed by a panel of Academy members according to procedures established and monitored by the Academy's Report Review Committee. Such reviews are intended to determine, inter alia, whether the major questions and relevant points of view have been addressed and whether the reported findings, conclusions and recommendations arose from the available data and information. Distribution of the report was approved, by the President, only after completion of this review process.
SUMMARY AND CONCLUSIONS
In legislating the Clean Air Amendments of 1970, the Congress asked the Environmental Protection Agency (EPA) to contract with the National Academy of Sciences (NAS) to conduct a comprehensive study and investigation of the technological feasibility of meeting the motor vehicle emissions standards prescribed in accordance with the law. In responding to this request, pursuant to a contract with the EPA, the Academy established a Committee on Motor Vehicle Emissions (CMVE) and charged it with the conduct of this study.
In its investigation of "technological feasibility," the CMVE addressed the following issues:
1. Determination of the feasibility of developing and designing an emissions control system that would enable compliance with the legally established emissions standards as judged by the certification procedures prescribed by the EPA.
2. The feasibility of mass producing those systems of promising design.
3. The projected performance of such emissions control systems in customer usage, including the requirements for maintenance necessary to assure continuing reliability.
4. The costs, per vehicle, associated with acquisition, maintenance, and operation of the emissions control system.
In the course of its work, the Committee has examined the variety of approaches of manufacturers and others to the problems relating to emissions control. At the time of this report, progress toward resolution of these problems in the four aspects listed above, although rapid, is uneven and uncertain, and the outlook toward 1975 and 1976 is not yet clear. Moreover, the rapid pace of that progress complicates judgment concerning the most appropriate course of action for attainment of the standards required by the law.
For 1975 model year light-duty motor vehicles, the Committee concludes that–
1. Four types of systems will meet the prescribed emissions standards during certification testing. These are: the modified conventional engine equipped with an oxidation catalyst, the carbureted stratified-charge engine, the Wankel engine equipped with an exhaust thermal reactor, and the diesel engine. For the catalyst system, one catalyst change must be permitted during the 50,000-mile durability testing for certification, and fuel with a suitably low level of catalyst poisons must be allowed. In determining whether vehicles mass-produced comply with an outstanding certificate of conformity under Section 206 of the Clean Air Act, provisions must be made for averaging of emission test results within a vehicle and engine class.
2. Vehicles incorporating these systems can be mass-produced in great enough volume to satisfy, in aggregate, the expected demand for vehicles in model year 1975.
3. It is important for two reasons that a suitable maintenance and inspection system be established for vehicles in use by the public.
First, there are no data concerning the deterioration of emission-control systems under conditions of customer use, and the Committee believes that the certification procedure alone is not a sufficient indicator of system durability. Even if it is demonstrated that properly maintained vehicles can comply with the standards under conditions of customer use, an adequate vehicle maintenance and inspection system will be required to assure that most vehicles will meet the standards when used by the general public; this is especially important for catalyst equipped vehicles.
Second, if it is determined that a substantial number of any class of vehicles or engines, although properly maintained and used, is not meeting the standards in use, Section 207(c) of the Clean Air Amendments empowers the Administrator of EPA to require the manufacturer to submit a plan for remedying the nonconformity. Under such a plan, the manufacturer is required to correct only those vehicles or engines which have been properly maintained and used.
4. The average increase in sticker price due to the emissions-control system of a catalyst equipped vehicle is estimated to be $160 above a current (1973) vehicle and $230 above a 1970 model year vehicle. Except for the diesel engine, lesser increases are expected for the other emission-control systems, when comparing vehicles of similar size and type.
Model year 1975 vehicles using Wankel engines or catalyst-equipped spark-ignition piston engines will use significantly more fuel than their 1973 counterparts. Carbureted stratified-charge engines will suffer only a slight fuel penalty; and the diesel engine will offer improved fuel economy, enough to compensate for its high initial cost within a few years of driving.
For 1976 model year light-duty motor vehicles, the Committee concludes that
1. Five control systems now in early stages of development have met the 1976 emission standards at low mileage. These are: the modified conventional engine equipped with dual catalysts, or with dual catalysts plus thermal reactor, or with two thermal reactors and a reduction catalyst, or with a three-way catalyst and electronic fuel injection, and the stratified-charge engine employing fuel injection and equipped with an oxidation catalyst. It is possible, but not certain, that some of these systems may prove to be certifiable for 1976, contingent upon the acceptance of the same provisos previously mentioned for 1975 model year vehicles.
More importantly, the recently developed carbureted stratified-charge engine, after 50,000 miles of durability testing on a compact car, has achieved well over the 90 percent reduction in hydrocarbon and carbon monoxide emissions called for in the Act and about 83 percent reduction in NOx. The Committee believes that this engine will be certifiable for 1976, at least in smaller engine sizes.
2. If certifiable, vehicles incorporating any of these systems can be mass-produced, but not necessarily in great enough volume to satisfy, in aggregate, the expected demand for vehicles in model year 1976.
3. The Committee holds the same concerns for performance of 1976 vehicles in use as discussed above for 1975 systems.
4. The average increase in sticker price of a dual-catalyst-equipped vehicle is expected to be $290 above a current (1973) vehicle, and $370 above a 1970 model year vehicle. Average annual costs of a dual-catalyst emissions-control system, including maintenance and fuel, with the increase in sticker price amortized over five years, is estimated to be $260 per year, compared with a 1970 model year vehicle. In contrast, the annualized costs for several other systems are estimated to he less than $100.
The Committee is greatly concerned about the trend of development of the 1976 control systems. The system most likely to be available in 1976 in the greatest numbers – the dual-catalyst system – is the most disadvantageous with respect to first cost, fuel economy, maintainability, and durability. On the other hand, the most promising system – the carbureted stratified-charge engine – which may not be available in very large numbers in 1976, is superior in all these categories. The Committee wishes to alert both EPA and the Congress to this development and believes that it warrants immediate attention.
INTRODUCTION
The Clean Air Amendments of 1970, which established exhaust emission standards for 1975 and 1976 light-duty vehicles (henceforth called vehicles) and light-duty vehicle engines, called on the Administrator of the Environmental Protection Agency (EPA) "to enter into appropriate arrangements with the National Academy of Sciences (NAS) to conduct a comprehensive study and investigation of the technological feasibility of meeting the emission standards" promulgated by the Clean Air Amendments. Meetings held between the NAS and EPA early in 1971 resulted in the establishment of a mutually agreeable work statement for the Committee on Motor Vehicle Emissions of the National Academy of Sciences. An extract from the work statement follows:
Statement of work
A. The Contractor shall conduct a many-faceted study of the technological feasibility of meeting the motor vehicle emission standards prescribed by the Administrator of the Environmental Protection Agency, as required by Section 202(b) of the Clean Air Act, as amended.
B. For the purposes of this study the term "technological feasibility" includes the ability within the automobile industry or elsewhere to
1. Design an engine, control system, or device capable of meeting the statutory emission standards using fuels which are or could be available
2. Mass produce such an engine, control system, or device
3. Maintain such an engine, control system, or device so that it will continue to meet the statutory emission standards with safety for a period of five years or 50,000 miles of operation, whichever is shorter.
C. The study of technological feasibility as defined shall include a study emphasizing technical aspects of the reported costs expected to be incurred in and the estimated time for the design, development, and mass production of an engine, control system, or device capable of meeting the statutory emission standards.
D. The study of technological feasibility shall include a study emphasizing the technical aspects of the reported estimates of extra cost incurred in maintaining such an engine, control system, or device so that it will meet the statutory emission standards for a period of five years or 50,000 miles, whichever is shorter.
E. Should the contractor conclude that the attainment of emission standards on the schedule provided by Section 202(b) (1) of the Clean Air Act is not technologically feasible, the contractor shall specifically determine technologically feasible interim emission levels to assist the Administrator in exercising his responsibilities under Section 202(b) (5) of the Act.
Past work of the Committee on Motor Vehicle Emissions
Membership of the Committee, shown in Appendix A, was selected entirely by the National Academy of Sciences.
The first meeting of the Committee took place on June 16, 1971, with subsequent meetings held approximately once each month. The Clean Air Amendments called for the Committee to submit semiannual progress reports to the Administrator and to Congress. One of the primary functions of such reports was to provide advice to the Administrator of EPA with respect to his decision whether or not to postpone for one year the applicable deadlines of the standards called for by the Clean Air Amendments. Under the legislation, anytime after January 1, 1972, any manufacturer may file with the Administrator an application requesting a one year suspension of the regulations applicable to emissions from 1975 model year vehicles. Anytime after January 1, 1973, any manufacturer may file an application requesting a one year suspension of the regulations applicable to emissions from 1976 model year vehicles. The Administrator must make his determination of each request for suspension within 60 days.
To provide maximum assistance to the Administrator in the formulation of his decision, and with due consideration of the timing required for such a decision, the Committee issued its first substantive report on January 1, 1972, containing a comprehensive study of the technological feasibility of the standards applicable to 1975 model year vehicles. In April 1972, in response to a direct request from EPA, the Committee prepared a report with respect to possible interim standards, in the event the Administrator were to grant a suspension of the 1975 standards. A brief progress report was submitted July 1, 1972, discussing the various areas of investigation of the Committee at that time.
This report of the Committee emphasizes the question of technological feasibility of the 1976 standards. In August 1972, the Administrator denied the requests of Volvo, International Harvester, Chrysler, Ford, and General Motors for a suspension of the 1975 standards. Requests for suspension of the 1975 standards may, however, be filed again by the above manufacturers or by others. A portion of this report is thus addressed to the technological feasibility of the 1975 standards.
Panels of consultants
The Committee has recognized the importance of having available to it the most recent and complete technical data and information upon which to make its judgments. Much of the information has been provided by eight panels of consultants, each panels [sic] were in operation during 1971. The of importance in the Committee deliberations [sic]. Panel members were selected by the Committee on the basis of recognized competence in specific areas. Membership of the panels is shown in Appendix B. Seven of these panels were in operation during 1971. The Catalyst Panel was added early in 1972 after the Committee became aware of the many controversial and critical factors associated with the operational characteristics of the automotive catalyst. The work of each of the panels was as follows.
Testing, Inspection, and Maintenance
The Panel on Testing, Inspection, and Maintenance was organized to assess the feasibility of ensuring that automobiles manufactured for 1975-1976 model years continue to meet the specified emission standards in actual customer use over the required period.
The panel evaluated each method as a system, from certification testing through assembly line control, surveillance, inspection, and maintenance in use. This study also considered the necessary training and licensing of mechanics, enforcement action required, short emission tests suitable for inspection or diagnosis, surveillance testing, feasibility of required maintenance procedures, and costs of maintaining emission-control systems for 1975-1976 vehicles.
Emission-Control Systems
The Panel on Emission-Control Systems was to investigate the potential of experimental 1975- 1976 emission-control systems, including consideration of the durability of these systems. The activities of this panel were restricted to studies of emission control for the spark-ignition internal-combustion engine including the Wankel and stratified charge types of engines. The use of different fuels, such as liquified petroleum gas (LPG), as well as dual-fuel concepts were also evaluated.
Alternate Power Sources
This panel was responsible for evaluating all automobile power source concepts except the conventional Otto-cycle engine, the internal-combustion Wankel engine, and the stratified-charge engine. The panel thus considered diesel engines, Rankine-cycle engines, Brayton-cycle engines, Stirling engines, electric systems, hybrid systems, and several systems that fall into no broad category.
The panel was to determine if it would be possible for any of the candidate engine systems to meet the 1975-1976 emission standards in production and the 50,000-mile (or five-year) life standard. For each promising system, the panel estimated the earliest possible date that mass production could be achieved. Major technical problem areas were identified for each system, and the probability of solving these problems was estimated.
Acceptability of each system, to the customer and to the industry, was predicted by the panel on the basis of driveability, safety, starting characteristics, maintainability, noise, cost, fuel economy, and many other factors. Some of these determinations were made in cooperation with other panels.
Manufacturing and Producibility
This panel was concerned with the manufacturability of low-emission systems and their components. The effort was not limited to the technical possibility of building one or a few systems; the technological feasibility of producing millions of systems in 1975 and 1976, was determined. This study included such considerations as producibility, tooling, lead time, and costs.
The work of this panel was directed toward helping the Committee determine, as specified in paragraph B2 of the Statement of Work, whether, within the automobile industry or elsewhere, there was a capability to mass-produce an engine, control system, or device capable of meeting the emission standards.
Driveability
The mission of this panel was to appraise the driveability of vehicles powered by candidate engine systems. Good driveability is loosely defined as the ability of a vehicle to start, operate, and stop smoothly under all environmental and operating conditions, without stalls, surges, hesitations, after-firing, and other undesirable characteristics.
There has been considerable testimony expressing opinions that some of the emission-control systems, especially if not properly maintained, would seriously affect the safety of the car, not only relative to its occupants, but also relative to other vehicles in traffic. Thus, assessing driveability is an important aspect of determining the feasibility of using a given system or engine. The work of this panel was done in conjunction with that of the Panel on Emission- Control Systems.
CATALYSTS
The CMVE organized this panel when it became apparent that the durability of many proposed emission-control systems is closely tied to catalyst performance. This panel analyzed activity and durability of both oxidation and reduction catalysts for emission-control systems. The major causes of catalyst failure during vehicle operation were examined. The effect on catalyst deterioration of the level of poisons in gasoline such as lead, sulfur, and phosphorus was studied, as was the effect of over-temperature on catalyst activity. Availability of catalytic materials and possible toxicity problems associated with the use of certain catalysts were also investigated.
Emission Standards and Atmospheric Chemistry
The work of the Panels on Emission Standards and Atmospheric Chemistry was associated with the requirement in the original work statement concerning recommendation by the Academy of technologically feasible interim emission levels.
The major concern was with interim levels for the 1976 standards, in the event that achievement of such standards was to be delayed a year. For the 1976 standards, oxides of nitrogen (NOx) must be controlled in addition to hydrocarbons (HC) and carbon monoxide (CO); procedures that reduce NOx do not necessarily reduce HC and CO, and may increase them. There are many sets of technologically feasible levels of the three pollutants that the Committee recommend. Thus, these two panels have been the various possibilities and tradeoffs.
The Panel on Atmospheric Chemistry determined, from the latest available data, the relationship between ambient concentrations of HC and NOx necessary to cause undesirable levels of oxidant production. The Panel on Emission Standards used these data, along with desirable air-quality goals for CO and NOx, and developed corresponding motor vehicle emission levels.
Each of the panels has devoted considerable time and effort to the work of the Committee. Some of the panel members have given virtually full-time effort to Committee work. These panels have traveled extensively and probed deeply in their attempts to bring before the Committee the material and information needed for the Committee to reach the judgments called for in the legislation. Panel visits have been made to domestic and foreign automobile manufacturers, to domestic and foreign catalyst suppliers, to the EPA and other government laboratories, to independent research laboratories, to state and local agencies concerned with the problems of enforcing emission standards, to those carrying out research and development on many types of alternate power plants, to oil companies, and to many others. A list of companies and individuals visited or otherwise contacted by CMVE personnel is given in Appendix C.
In each visit, panel members have endeavored to ensure the timeliness and validity of the data furnished to them. Panel visits have involved discussions with personnel ranging from top management to working technicians and engineers.
The panels have reported periodically to the parent Committee on their progress. Close contact has been maintained between the panels and the Committee, to ensure that the panels were stressing the necessary topical areas in their investigations. Panel activities terminated with the submission of final written panel reports to the Committee.
Other means of obtaining information
The Committee has attempted to solicit pertinent information from the general public.
Announcements have been placed in the Federal Register requesting information with respect to technological feasibility. Descriptions of these announcements are included as Appendix D.
Finally, the Committee as a whole visited General Motors and Ford in Detroit on May 18 and 19, 1972, to get a first-hand view of the efforts of two of the larger manufacturers toward meeting the emission standards. Visits by selected members of the Committee were made to other manufacturers.
The judgments of the Committee to be presented in this report necessarily rely upon the information received using the various sources mentioned above. The Committee believes that it has had presented to it sufficient information upon which to base its judgments.
NOTE
Final reports of the CMVE panels are being prepared as technical publications and will be made available to the public by the National Research Council. Other pertinent information will be maintained as a public record in the files of the CMVE.
(Figures referred to in this text are omitted in the RECORD).
THE STANDARDS, CERTIFICATION AND TESTING
Numerical values of standards
Section 202 of the Clean Air Amendments of 1970 requires the Administrator of the EPA to prescribe emission standards for light duty motor vehicles together with measurement techniques on which such standards are to be based [The Federal Register of November 15, 1972, contains a complete description of the regulations concerning the standards, test procedures, allowable maintenance, etc.] Such standards require that the emissions of carbon monoxide and unburned hydrocarbons from light-duty vehicles and engines manufactured during or after modal year 1975 be reduced by at least 90 percent from these required of 1970 vehicles; also, emissions of oxides of nitrogen from light-duty motor vehicles and engines manufactured during or after model year 1976 are to be at least 90 percent below the average of those actually measured from light-duty vehicles manufactured during model year 1971 which are not subject to any federal or state emission standards. The 1975 model year standards are:
0.41 grams per vehicle mile for hydrocarbons (HC).
3.4 grams per vehicle mile for carbon monoxide (CO) and
3.1 grams per vehicle mile for oxides of nitrogen (NOx).
Standards for 1976 model year vehicles are:
0.41 grams par vehicle mile for hydrocarbons.
3.4 grams par vehicle mile for carbon monoxide and
0.4 grams par vehicle mile for oxides of nitrogen.
The Clean Air Amendments call for vehicle compliance with the above standards for five years or 50,000 miles, whichever occurs first.
Procedures for certification, CVS-CH test
The numerical values of the standards must be defined in terms of a specific method of measurement and a specific driving cycle. The EPA Administrator is required by Section 206 to test any motor vehicle or class of motor vehicles to determine whether they meet the standards set forth in Section 202(b) of the Clean Air Amendments. A certificate of conformity is to be issued for classes of motor vehicles that comply with the standards, thus permitting the manufacture and sale of these classes of vehicles. The emissions test to be used for certification of 1975 and 1976 vehicles, referred to as the CVS-CH test, consists of a 12-hour wait at a temperature between 60' and 86º F, a cold-engine startup, a continuous sequence of different driving modes simulating an average trip over a 23-minute route in an urban area, a ten-minute shutdown followed by a hot-engine restart, and a repeat of the first 505 seconds of the 23-minute cycle. This test is performed with the vehicle on a chassis dynamometer. Exhaust-gas sampling begins immediately after the key is turned on (whether the engine starts or not). Diluted exhaust emissions are collected during the first 505 seconds in one bag, those during the remainder of the 23-minute cycle in a second bag, and those from the hot-restart phase in a third bag. Contents of the three bags are than analyzed and weighed in accordance with the EPA test procedure to be the final mass emissions, in grams par mile, of HC, CO, and Nox.
To obtain a certificate of conformity for a class of vehicles, the automobile manufacturer must also demonstrate the effectiveness of the vehicles' emission-control system over the "useful life" of a vehicle. The regulations require a manufacturer to test two separate fleets of prototype vehicles representing models to be sold to the public. The "emission data" fleet is intended to determine the emissions of relatively new vehicles. The vehicles in this fleet are driven 4,000 miles to break in the engine and stabilize emissions. The emissions are then measured, using the CVS-CH test procedure. Allowable maintenance on emission-data vehicles is limited to the adjustment of engine idle speed at the 4,000-mile test point.
The second fleet, the "durability-data" fleet, is designed to determine the capability of the emission-control system to keep emissions below the standards over the expected useful life of the vehicle. The vehicles in this fleet are driven for 50,000 miles and tested for emissions every 4,000 miles. The procedure for mileage accumulation is the Durability Driving Schedule over a modified AMA route. The maximum speed is 70 mph, and the average is 30 mph. One major engine tuneup to manufacturer's specifications may be performed on durability vehicles at 24,000 miles (on vehicles with 051-CID or less, at 12,000, 24,000, and 36,000 miles). The replacements and adjustments allowed are detailed in the regulations. Emissions tests must be run before and after any vehicle maintenance that may be reasonably expected to affect emissions. As the first step in determining compliance of a new light-duty vehicle, emission-deterioration factors are determined from the durability-data fleet emission test results. Separate factors are determined for HC, CO, and NOx and for each engine/control-system combination. A straight line is fitted, by the method of least squares, to each of the plots of emissions versus mileage for the endurance fleet. For each of the three pollutants, deterioration factors are determined from these curves as the ratio of emission at 50,000 miles to those at 4,000 miles. The emission test results, at 4,000 miles, for each emission-data fleet vehicle are then multiplied by the appropriate deterioration factor to give adjusted emissions for each vehicle. These adjusted emissions are then compared to the standards. Every test vehicle from an engine family must comply with the standards before any vehicle in that family can be certified.
Production-line testing
To ascertain whether vehicles are being manufactured in accordance with the regulations with respect to which a certificate of conformity was issued, Section 206(b) authorizes EPA to test new vehicles and engines. Such tests can be conducted by EPA or by the manufacturer in accordance with conditions specified by EPA. According to Section 206(b) (2) (A), if:
"Based on tests ... on a sample of new vehicles or engines covered by a certificate of conformity, the Administrator determines that all or part of the vehicles or engines so covered do not conform with the regulations with respect to which the certificate of conformity was issued, he may suspend or revoke such certificate in whole or in part ...”
The Act sets forth a procedure for reissuance of the certificate and for hearings on its suspension or revocation.
It is impractical to determine the emissions from a large fraction of production vehicles by the CVS-CH procedure. Each test involves a twelve-hour wait at room temperature followed by a 41-minute test, with expensive and complex instrumentation required. In addition, measurements of exhaust emissions show poor repeatability. Data from the automobile manufacturers and
several independent laboratories taken on 1972 model year vehicles using the 1975 Federal CVS-CH test show coefficients of variation of 10 to 20 percent. When the CVS-CH test is applied to 1975-76 prototype vehicles, the probable percentage error increases, since emission levels have decreased from 1972. Only limited data were available of test-reproducibility using the CVS-CH test with dual-catalyst-equipped vehicles near the 1976 emission levels.
Coefficients of variation for these limited data, consisting of 16 repetitive tests on a vehicle, were as great as 50 percent for CO. Test on three-valve carbureted stratified-charge engines, which meet 1975 standards, have yielded much lower coefficients of variation.
Some of the reasons for the lack of repeatability of measurements are the difficulty of following the speed-time curve specified in the CVS-CH procedure, lack of repeatability of some of the engine functions having an effect on emission control (such as the choke time), and the fact that the 1975-76 pollutant levels give such low concentrations in the sample bags that the resolution capabilities of the instruments are approached. The variation of test results during the cold portion of the test is the largest part of total test variation, with a large percentage of CO and HC emissions occurring during the first two minutes of the CVS-CH test. An individual measurement using the CVS-CH test on a 1975-76 car can vary from the true value by 50 percent and thus little significance should be attached to a single test. Since a hot-start test does not include the emissions that dominate the results obtained with the CVS-CH test, no short hot-start test will pass and fail exactly the same vehicles as the CVS-CH test. It thus is evident that 100 percent production-line testing with any procedure except CVS-CH test (impractical on production line) cannot determine if all vehicles are being manufactured in accordance with the regulations with respect to which a certificate of conformity was issued.
In terms of overall air quality, it is the average emissions from a group of cars that determines the automotive contribution to air pollution. This fact and the variability of test results mentioned above shows a need for regulations that control the manufacture and operation of vehicles to ensure that the emissions on the average meet the standards. It is only necessary to test a quality audit sample of production-line vehicles to demonstrate that the average emissions of production vehicles compare satisfactorily with the certification standard, taking into account a prescribed useful life deterioration factor and a tolerance factor reflecting the difference between production vehicles and pre-production prototypes. The logical test to choose for this quality audit is the CVS-CH procedure.
The Commission recommends, as in the CMVE January 1, 1972 report, that only the emissions of the average of each engine-vehicle combination on the production line be required to meet the standards. A requirement that all vehicles meet the emission standards is too restrictive and is unnecessary to meet air quality needs.
Compliance after sale, warranty
The recommendation covering production-line averaging has an impact on the warranty provisions of the law, contained in Sections 207(a) and (b) of the Clean Air Amendments.
Section 207(a) requires that the manufacturer of each new motor vehicle and engine shall warrant to the purchaser that it is (1) designed, built, and equipped so as to conform at the time of sale with (the applicable standards), and (2) free from defects in materials and workmanship which cause such vehicle or engine to fail to conform with (the standards) for its "useful life."
Section 207(b) deals with warranties for vehicles throughout their useful life. This section states that manufacturers could ultimately be required to warrant compliance of the emission control system of a vehicle throughout its useful life if it is maintained and operated in accordance with the manufacturer's instructions and if the nonconformity results in the car owner "having to bear any penalty or other sanction ... under State or Federal law." Before such a useful-life warranty can be imposed however, the Administrator must first determine that there are available testing methods and procedures to ascertain whether a vehicle, when in actual use, complies with the emission standards and that such methods and procedures are reasonably capable of being correlated with tests conducted by EPA preparatory to issuance of a certificate of conformity (meaning the full CVS-CH procedure). These warranty provisions apply to each vehicle.
However, if it is concluded that there is no short test or procedure that reasonably correlates with the CVS-CH test, the problem of possible conflict in implementing the warranty requirement is eliminated. After consideration of this point, the Committee concludes that no short test is available now, or likely to be available in the near future, that will pass and fail the same vehicles as the full CVS-CH test.
Section 207(c) of the Clean Air Act deals specifically with recall of vehicles which, although properly maintained, do not conform with the standards. According to Section 207(c) (1), "If the Administrator determines that a substantial number of any class or category of vehicles or engines, although properly maintained and used, do not conform to the regulations prescribed under Section 202, when in actual use throughout their useful life (as determined under Section 202(d)), he shall immediately notify the manufacturer thereof of such nonconformity, and he shall require the manufacturer to submit a plan for remedying the nonconformity of the vehicles or engines with respect to which such notification is given."
Since the purpose of the Clean Air Amendments is to control the automotive contribution to air pollution, the Committee believes that a surveillance audit of a statistical sample of vehicles is adequate to determine if each family of cars meets the standards. The full CVS-CH test would be run for the surveillance audit. Such surveillance tests could be used not only to determine emissions from each car family, but also to evaluate deterioration of emission control systems in actual use, to evaluate the effectiveness of prescribed maintenance procedures, to develop information on failure modes, and to establish the need for recall of a class of vehicles.
POTENTIAL OF SPARK-IGNITION INTERNAL-COMBUSTION ENGINES PASSING EMISSION CERTIFICATION FOR 1975 AND 1978
Introduction
The investigations of the Committee to date have shown that, according to current planning of the automobile manufacturers, the great majority of the engines to be used in 1975-76 model year vehicles will be conventional, reciprocating, spark-ignition engines. A smaller fraction of the 1975-76 vehicles will use Wankel rotary engines, and one manufacturer (Honda) has plans to produce a new type of carbureted stratified-charge engine. Some passenger-car diesel engines will still be produced, but these will differ only slightly from the diesel engines currently available. Diesel engines are discussed more fully in Section 6.1.
This section of the report will present an analysis and evaluation of the prospects of spark- ignition engines passing the emissions certification test for 1975 and 1976 model year vehicles.
CURRENT STATUS OF 1975 SYSTEMS
The January 1, 1972, report of the CMVE dealt at considerable length with the technological feasibility of meeting the 1975 standards. At the present time, most automobile manufacturers have developed somewhat similar prototype emission-control systems for their 1975 model year vehicles. The major U.S. and foreign manufacturers are currently assembling and testing fleets of vehicles equipped with the complete system to evaluate different promising catalyst materials and to obtain data on system durability before final production designs are frozen.
These 1975 emission-control systems typically consist of:
(i) An improved carburetor to provide more accurate fuel metering, with compensation for air-density changes, and with an electrically powered choke that comes off quickly at ambient temperatures of about 70º F.
(ii) A quick-heat intake manifold designed to promote rapid fuel evaporation after engine start-up.
(iii) An electronic ignition system to eliminate the wear and other problems of current distributor assemblies and to allow easier spark-timing control. (Inadequate maintenance of present distributors commonly results in increased engine emissions.)
(iv) An exhaust-gas recycle (EGR) line and control valve designed to recycle about 10 percent of the exhaust flow to hold NOx emissions below 3 grams per mile (g/mile).
(v) An air pump to inject air into the exhaust ports to oxidize carbon monoxide and hydrocarbons.
(vi) A catalytic converter in the exhaust system to promote further oxidation of the HC and CO emissions from the engine.
For some manufacturers, the current fleet tests represent the first extensive evaluation of the complete engine emission controls with the best oxidation catalyst materials now available. Data obtained from some of these manufacturers' fleets are shown in Table 3-1. (Most of the data in Section 3 of this report were received in reply to a questionnaire dated July 13, 1972 or were presented during recent panel visits.) These tests follow the durability driving cycle and maintenance procedures used in the emissions certification of vehicles.
Progress in emission control for 1975 systems using catalytic converters has been made since the CMVE report of January 1, 1972. It is highly probable that most manufacturers will be able to produce vehicles that will pass the 1975 certification test procedure, providing allowance is made for one catalyst replacement during the 50,000-mile durability tests, that fuel containing sufficiently low levels of lead and other catalyst poisons is used, and that averaging of emissions within automobile and engine classes is allowed.
Emission-control systems are also being developed that do not use catalysts and therefore have improved durability over the catalytic systems. The three-valve stratified-charge carbureted engine, under development by Honda, has achieved emissions below the 1975 standards at low mileage on compact cars (for details) see Section 3.9.2). Three vehicles equipped with a 2-liter engine (122 CID) have completed 50,000-mile durability testing and met the standards at every test throughout the test period.
The Wankel engine with a thermal reactor has achieved emission levels below the 1975 standards in a compact car. This system has already demonstrated improved durability over a catalytic system. Data from the few cars tested over extended mileage indicate that deterioration will be relatively low.
Engine emissions for 1976 systems
Introduction
In developing systems to meet the 1975 standards, most automobile manufacturers have emphasized that such systems must be compatible with 1976 requirements. There has thus been a concentration on 1975 control systems that can be modified to achieve the greater NOx emission control called for in 1976. Additional NOx control can be achieved by increasing the amount of exhaust gas recycle, by adding an NOx-reducing catalytic converter to the exhaust system, or by a combination of both techniques.
To approach the 0.4 g/mile NOx level with a conventional spark-ignition engine, a combination of both techniques appears to be required. The use of large amounts of EGR (20 percent or more) results in a large fuel-economy penalty, severe driveability problems with attendant safety hazards, and an increase in engine HC and CO emissions. It is not practical at this stage to achieve NOx emission levels approaching 0.4 g/mile with EGR alone in a conventional engine.
Without EGR, engine NOx emissions vary between about 3 and 8 g/mile, depending on the air-fuel ratio, spark timing, engine size, and other details of engine design and operation. In the dual-catalyst system, a separate NOx-reduction catalyst is added to the exhaust system, between the engine and the oxidation catalyst. In the three-way catalyst system, a single catalytic converter simultaneously reduces the concentration of all three pollutants (HC, CO, and NOx) in the exhaust stream. Air-fuel ratio must be closely regulated in such system. The NOx-reduction catalysts currently available, as will be described below, are not able to retain sufficient activity over extended mileage to reduce these engine emissions below the 1976 standards. Thus, unless further improvements in NOx catalyst durability occur over the next year, only systems for conventional engines with increased EGR and an NOx-reduction catalyst show any promise of approaching the 0.4 g/mile level. To minimize demands on catalyst size, cost, and durability, there is a continuing emphasis on achieving low and stable engine emissions. Techniques for controlling emissions during the first part of the test when the engine is still cold, fuel-metering requirements, EGR systems, and the potential for improved engine emissions control are examined next.
COLD-START EMISSION CONTROLS
With 1975-76 catalyst-based emission-control systems, a large portion of the carbon monoxide and hydrocarbon emissions occur during the cold-start and engine warm-up phase of the drive schedule in the CVS-CH test. To compensate for the low volatility of cold gasoline, a rich mixture must be provided during cold starts. The excess fuel is not fully burned in the combustion chamber, and the cold engine thus emits high levels of hydrocarbons and carbon monoxide. Because the oxidizing catalyst is inefficient while cold, large amounts of these contaminants are discharged to the atmosphere. Although the NOx catalyst is also cold and ineffective during start-up, cold-start NOx emissions tend to be lower because rich mixtures and cold cylinder walls reduce the formation of NOx.
Because of the high HC and CO emissions during the cold start, considerable development effort has been spent in a number of cold-start controls and procedures. In the prototypes of their 1975 and 1976 systems, most manufacturers have elected to modify the cold-start process.
Specifically, they have achieved start-up with leaner air-fuel mixtures by preheating the air and fuel, by improving the mixture control in the carburetor, and by shortening the choking period without seriously impairing cold engine operation and driveability. Most prototypes include air and fuel preheat systems and modified choke operation.
The purpose of preheating air and fuel is to achieve higher volatility with cold fuel, which, in turn, allows leaner engine operation and shorter choking period during warm-up. The air preheat system that has been incorporated in most 1970 and subsequent American-make cars, and has proven reliable, will be used in most 1975-76 models.
In addition to heating the intake air, several proposed emission-control systems promote further evaporation of the fuel by supplying heat to the base of the carburetor. This is accomplished by using a heat exchanger between the carburetor and the exhaust-manifold crossover, causing the fuel droplets to make contact with a hot surface and to flash into vapor. Several problems remain to be solved to ensure that production units can attain the emission reduction predicted by experimental designs. While durability of the system has not yet been evaluated, there appear to be no major technical difficulties.
A quick-acting choke, employing electrical or mechanical timing devices, will be used to lean out the mixture as early as possible after start-up. For systems incorporating air and mixture preheating, choking times have been reduced from several minutes to less than 30 seconds, while maintaining adequate driveability.
Carburetors
The precise metering of the fuel and air to automotive engines has become much more important in recent years because the mixture ratio is a critical parameter affecting the exhaust composition and the functioning of exhaust-treating devices. Most 1975-76 model carburetors have been redesigned to achieve better air-fuel ratio control and maintain good cold-start performance of the engine.
Except for the demands during extreme accelerations and decelerations, the newly designed carburetors are capable of maintaining tolerances of + 3 percent of the set air-fuel ratio. This approaches the fuel-metering accuracy required for the dual-catalyst 1976 control systems in which the air-fuel ratio must be held between about 13.8 and 14.5 to achieve adequate NOx reduction in the first catalytic converter.
Considerable design work remains to be done to ensure durability with these finely adjusted carburetors. Most manufacturers are considering factory-sealed, tamper-proof settings because it is believed impossible for a typical mechanic to make the required adjustments. The dependability of these factory-set adjustments is unknown.
Electronic Fuel Injection
Several companies are considering Electronic Fuel Injection (EFI) as an alternative to the carburetor. In such systems, an electronic module controls the amount of fuel provided to the engine.
An advantage of electronic control is that, by using appropriate transducers, air-fuel ratio can be compensated for variations in such operating parameters as engine speed, manifold vacuum, ambient conditions, various engine temperatures, exhaust composition, etc. Thus, there is potential for adequate control of mixture ratio over a wide range of operating conditions. EFI systems can respond quickly to changes in operating conditions and are therefore able to provide satisfactory control of air-fuel ratio under transient, conditions. However, contacts made with carburetor manufacturers, automobile manufacturers, and producers of electronic fuel injection equipment indicate that current EFI systems do not provide substantial improvement in air-fuel control over the advanced-design carburetors operated under steady conditions.
EFI systems have been and are in production on several European cars. Field experience with these systems initially showed a high component-failure rate, although performance is improving. The advantage of EFI over current carburetors in small cars is in performance characteristics and fuel economy, i.e.:
Increased power output, particularly for high-rpm high-performance engines
Better fuel economy for high-speed driving
Improved driveability, particularly with manual shift engines
At least one manufacturer is introducing a mechanically controlled fuel-injection system which may show performance comparable with the improved EFI system and at savings in cost.
Exhaust-Gas Recycle (EGR)
The most extensively developed technique for reducing engine NOx emissions is the recycling of a fraction of the exhaust to the engine intake. The recycled exhaust gases dilute the fresh mixture, thus reducing peak combustion temperatures and Nox formation rates. The disadvantages of EGR are the loss in engine power and the reduction in tolerable air-fuel ratio variations consistent with smooth engine operation. The use of EGR requires some mixture enrichment to maintain adequate driveability, which results in a fuel-economy penalty. In most systems, EGR is cut out at wide open throttle and idle operation.
EGR was introduced in most 1973 model year vehicles to bring NOx below 3 g/mile.
Experience from the durability testing of these EGR systems indicates that plugging of the recycle line and control valve with leaded fuels is a significant problem. But with unleaded fuels, and with regular inspection and cleaning of the system, these problems are not expected to be severe.
As the amount of EGR is increased to reduce NOx, engine emissions below 3 g/mile, there is a need for more precise matching of the recycle flow to fresh mixture flow, and for more uniform mixing of the recycled exhaust in the intake. Engine combustion-chamber redesign with higher turbulence levels to promote more rapid combustion also improves the tolerance of the engines to EGR.
Potential for Engine Emission Reduction
The methods of emission reduction discussed so far have been engine modifications that reduce emissions from the bare engine, i.e., before after-treatment devices such as catalysts and thermal reactors. The first two rows of Table 3-2 give typical engine emissions from a General Motors Corporation 1972 production audit. Both mean emissions and the standard deviation are given.
The magnitude of the standard deviation indicates the spread in emissions about the mean value. This spread is due to differences in items such as brake setting, variations in transmissions, engine friction, carburetor settings, and stacking up of engine tolerances. The best GM division has been able to reduce both mean engine emissions and the spread in emissions through improved production control, as indicated in the second row of the table.
With the addition of a quick-heat manifold and an improved carburetor with a quick-acting choke these HC and CO engine emissions can be improved. However, use of EGR to reduce NOx emissions requires some mixture enrichment to compensate for the decreased flame speed, and engine HC and CO emissions rise. The last two rows in Table 3-2 are estimates of achievable engine emissions goals at low mileage for standard-size engines in standard-size vehicles. The third row corresponds to a lean and the fourth row to a richer carburetor setting. The further reduction of emissions in conventional engines must be achieved with exhaust treatment, such as catalysts or thermal reactors.
Catalysts
The control system for 1976 on which most development effort has been concentrated uses two catalyst beds to clean up the engine emissions before exhaust to the atmosphere. A typical system layout is shown in Figure 3-1. The bed closest to the engine is used to remove NOx. It is operated under net reducing exhaust-gas conditions (between 1 and 2 percent carbon monoxide in the exhaust gas, corresponding to a slightly rich carburetor calibration). Air is then added to the exhaust stream between the catalyst beds, and the remaining HC and CO emissions are removed in the second catalyst, the oxidation bed. The two catalytic beds may be in separate containers as shown in the figure, or they may be packaged in a single container. The system is a logical development of the 1975 control system described previously.
Because the NOx catalyst bed must be placed ahead of the oxidation bed, the oxidation catalyst warms up more slowly. Control of HC and CO emissions during start-up would thus be delayed if air were always injected between the catalyst beds. To maintain control over 1976 HC and CO levels, air is diverted to upstream of the NOx-reduction bed during the engine-warm-up phase.
Thus the NOx bed can act initially as an oxidation catalyst. Once the oxidation catalyst is
warmed up, the air is diverted to between the two beds and the first catalyst acts primarily as an NOx-reduction catalyst. (As an alternative, a "start" catalyst can be used ahead of the NOox catalyst and bypassed after the engine is warm.)
The oxidation catalysts used in these dual-catalyst systems are the catalysts now being developed for 1975 model year vehicles. These consist of noble metals (platinum and/or palladium) or base metals promoted with noble metals (a small amount of noble metal required to initiate activity) deposited on both monolithic and pellet substrates. Except for precious metals, there appears to be no problem of supply of raw materials for the active ingredient and the support in the catalysts contemplated. The early state of development of oxidation catalysts was described in the previous Committee report.
Several companies now have products that have demonstrated adequate initial activity, and some of these have reasonable durability.
The NOx catalysts tested to date include noble metals (platinum, ruthenium, palladium), base metals, or base metals promoted with noble metals, deposited on both monolithic and pelleted ceramic substrates. Nickel-copper and Inconel metallic monolithic NOx catalysts are also being tested. Reduction catalysts are generally less well developed than oxidation catalysts. Whereas
several NOx catalysts have demonstrated sufficient initial activity, the limited amount of durability data currently available is not encouraging. Several examples of the best low-mileage emission data available are shown in Table 3-3.
Only a few 1976 experimental vehicles have been tested to evaluate NOx catalyst durability. Before attempting extensive durability tests, most manufacturers are working to optimize the performance of the system to reduce low-mileage emissions to values at least 50-60 percent below the 1976 standards levels, to improve vehicle driveability, and to hold performance losses to a minimum. Since NOx-catalyst durability is substantially inferior to that of the best oxidation catalysts, a production goal of .25 g/mile does not seem unreasonably stringent.
The results of the most promising durability tests of some dual-catalyst vehicles are summarized in Table 3-4. Emissions are shown as a function of mileage. Where engine-emissions data are available, the average catalyst-conversion efficiencies over the entire CVS-CH driving cycle can be estimated and are shown in the table. Conversion efficiency is the percentage of entering emissions removed in the converter. These data show that the initial high conversion efficiency rapidly deteriorates.
The causes of this rapid deterioration in NOx-catalyst efficiency are not yet quantitatively understood. Catalysts have a long history of success in the petroleum and chemical industries; the major processes in these industries employ steady-state conditions of temperature, pressure, and flow rate of gases, with careful exclusion of poisons. Many automobile catalysts developed today would work similarly well and probably last 50,000 miles if they could work within narrowly defined operating ranges (or "windows") in each of four variables: temperature, gas competition, gas flow, and poison concentrations. Catalysts can tolerate occasional excursions from these windows, but prolonged excursions invariably lead to slow chronic aging or quick massive failures.
In actual practice, an automobile is always in a transient condition: the catalyst is too cold during start-up and too hot during a long down-hill cruise; the air-fuel ratio is too rich on idle and too lean during high speed; the exhaust-gas flow is slow during idle and fast during upgrade cruise.
The catalysts are also exposed to repeated cycles of heating and cooling, evaporation and condensation of water, pulsating flow from exhaust gases, vigorous shaking on the road, and a variety of poisons including lead and sulfur. Under these excursions from the windows, catalysts deteriorate rapidly.
The rate of the deterioration due to inadequate control of air-fuel ratio in the engine and the temperature fluctuation in the catalyst bed has not been quantified. Thermal degradation of the catalyst can occur due to damage to the surface structure caused by overheating. The presence of an oxidizing atmosphere even for short periods of time when the catalyst is hot is known to be detrimental, especially to the nickel-copper alloy metallic catalysts. The loss of catalytic material both from noble metal and base metal NOx catalysts has also been observed to occur, probably due to oxidation.
It is clear that poisoning of the active catalyst material by contaminants in the fuel is the cause of some deterioration. During the last two years, it has become evident that the oxidation catalysts being tested in 1975 prototype vehicles deteriorate as a consequence of the trace quantities of lead, phosphorus, and other elements in lead-free fuels and lubricants. It is anticipated that, as a result of EPA regulations, the lead-free fuel available in 1975 and 1976 will have average contaminant levels of about 0.03 grams/gallon lead, less than 0.005 grams/gallon phosphorus, and about 0.04 percent sulfur by weight. It is not known how severely these contaminant levels will affect the activity of the different NOx catalysts now being evaluated. Laboratory tests on a noble metal NOx catalyst which contained platinum and other metals showed lead poisoning of magnitude comparable to that observed with platinum oxidation catalysts. Sulfur also affected the activity of this NOx catalyst at levels of 0.04 percent by weight or more. However, a ruthenium NOx catalyst appeared in bench tests to be much more resistant to lead poisoning.
There are NOx catalysts of both noble metal and copper-nickel that are remarkably resistant to lead levels up to 0.5 gram/gallon. NOx catalysts may have the ability to partially recover from sulfur poisoning if operated in an oxidizing atmosphere at high temperatures.
The ability to control air-fuel ratio within narrow limits with dual-catalyst systems is critical. It is well known that, under certain operating conditions, nitric oxide, NOx is reduced in the NOx catalyst bed to ammonia, most of which is then oxidized back to NOx in the oxidation catalyst. Ammonia formation increases with carbon monoxide level in exhaust mixtures. Thus, whereas the air-fuel mixture must be slightly fuel-rich to provide net reducing conditions at the NOx -catalyst bed, too great a CO level leads to excess ammonia formation and results in increased concentrations of NOx in the exhaust. It appears that the air-fuel ratio must be controlled to give about 1 to 2 percent CO in the exhaust, corresponding to an air-fuel ratio of 13.8 to 14.5.
Progress is being made in the development of NOx catalysts for automotive use. However, to retain its effectiveness, it is necessary that the catalyst be integrated into the engine emission- control system to protect the catalyst from long excursions from its operating windows.
Unfortunately, coordination of research by the automobile manufacturers and catalyst suppliers is far from ideal. The composition of the catalyst supplied to the auto manufacturer for testing is proprietary to the catalyst supplier; when failures occur during durability testing, the catalyst must be returned to the catalyst supplier for further analysis. 50,000-mile durability testing requires three to four months, 25,000-mile testing takes six to eight weeks. The process is a slow one. Many months may elapse from the time a catalyst is supplied to the time data may be available that could lead to an improved catalyst composition.
Three-Way Catalysts
Several catalyst manufacturers are developing single-bed catalysts that, under carefully controlled operating conditions, will simultaneously promote oxidation of HC and CO and reduction of NOx.
Successful operation of such a three-way catalyst has been found to occur in a narrow window of air-fuel ratio, slightly on the rich side of stoichiometry. The width of this window has been found to be only +.1 air-fuel ratio, thus requiring an overall control of air-fuel ratio to within the + 1 percent. Lean-side deviations from this window result in a drastic loss of NOx conversions; rich-side deviations lead to considerable loss of HC and CO oxidation efficiencies and increased ammonia formation. The difficulties of maintaining such a close control of air-fuel ratio during all the transient conditions that occur during typical operation of an automobile have been cited in Section 3.4.
If a three-way catalyst could be successfully incorporated into a vehicle and proved to possess adequate durability a simpler system than the dual-catalyst system would result. Since there is enough oxygen present in the exhaust gas, the air pump would no longer be required. One catalyst bed would be eliminated, and the difficulties of heating up the catalyst beds rapidly to control cold start emissions would be simplified. Vehicle driveability is known to be good with a stoichiometric air-fuel mixture, and the fuel economy for a given engine would be reasonably close to optimum.
Feedback Control for Air-Fuel Ratio
The need for more close control of exhaust gas composition both in the dual-catalyst and the 3-way catalyst system has led to the investigation of methods of achieving more precise fuel and air metering. A method that has met with some success and is under development in several European and American companies is a feedback, or closed-loop, system of fuel metering. In such a system, an oxygen sensor is used to detect the level of oxygen in the exhaust stream and to supply an error feedback signal either to an electronic fuel-injection (EFI) control module or to a specially constructed carburetor. This signal causes adjustments in the fuel or air supply, thereby maintaining close control of air-fuel ratio. Unfortunately, these oxygen sensors function only when hot and are thus ineffective in controlling cold-start emissions.
Because the oxygen sensor is an electrical transducer, it is particularly well suited for adaptation to the electronic circuitry of the EFI control module. Also, fuel-injection systems provide quicker response to error signals than do carburetors. The EFI system or carburetor used must be able to control air-fuel ratio to within +5 percent of stoichimetry without feedback; the sensor feedback system at present has limited control authority so that it cannot correct for deviations outside this range. Current EFI systems and advanced carburetors can achieve this degree of air-fuel ratio control during steady state operation, but deviate significantly outside the +5 percent range during transient vehicle operation such as acceleration and deceleration. To obtain full benefit from the sensor-feedback system, improvements in the performance of the basic EFI or carburetor will be required.
The sensor is expected to be inexpensive and a life of about 12,000 miles is the development target. The sensor could then be exchanged as a spark plug.
The durability of oxygen sensors now available is inadequate, and their life in a vehicle is only a few hours. The major problems are thermal shock, erosion of the electrodes, and maintaining good electrical contacts with the sensor. In addition, bench tests have shown that the sensor can be poisoned by lead, sulfur, phosphorus, and other impurities. Considerable development will therefore be required before the sensor and feedback system is ready for mass production.
If the durability and poisoning problems are solved, the oxygen sensor could make an important contribution to lowering emission levels and improve durability of the catalysts in the exhaust system by avoiding large variations in exhaust composition and temperature during operation of the engine.
Developments on feedback systems have been proceeding most intensively in Europe, where the performance gains of fuel-injection systems on smaller, higher-speed engines are more significant. Test results on six different European vehicles are summarized in Table 3-5. Also shown are data from Bendix, also with a compact car. These results are at low mileage; durability data are not yet available.
The sensor-feedback concept also can be used to control air-fuel ratio, and thus exhaust-gas composition, with the dual-catalyst emission-control system. This system is more complex than the three-way catalyst approach since an air pump is required to supply secondary air upstream of the oxidation catalyst, and air must be diverted to upstream of the NOx reduction catalyst during the cold-start portion of the test. Again, the elimination of exhaust-gas composition excursions outside the desirable operating window of the catalysts would be expected to improve catalyst life, and there is a better chance to develop separate oxidation and reduction catalysts by a given time than to find a three-way catalyst adequate to the job.
Several parallel developments are necessary to make the oxygen-sensor feedback control catalyst system more than a promising concept. First, the oxygen sensor must be shown to be durable in the exhaust environment of an operating vehicle with commercially available lead-free fuels.
Second, the EFI system or carburetor must be improved to provide control of air-fuel ratio to within +5 percent of stoichiometric for all engine operating modes, or the effective range of the feedback system must be extended. Finally, either the durability of the catalyst in the three-way system must be demonstrated, or a durable NOx-reduction catalyst must be developed for the dual-catalyst system. The effects of fuel contaminants and temperature fluctuations on the catalyst activity and size of the required air-fuel ratio operating window are unknown at this time.
Thermal Reactors
A substantial amount of work has been done on emission-control systems that incorporate a thermal reactor, a chamber in which HC and CO emissions are burned after leaving the engine. Typically the reactor is bolted to the cylinder head in place of the normal exhaust manifold, although some systems employ thermal reactors further back in the exhaust system. Some systems combine a thermal reactor with catalysts; some use a thermal reactor alone.
There have been two basic versions of systems using a thermal reactor only: fuel-rich and fuel-lean systems. The fuel-rich system results in less NOx formation, but only at the expense of substantially poorer fuel economy. The lean system does not require air injection, so it has the advantage of being simpler. Fuel-rich reactor cars typically operate in the range of air-fuel ratios from 11:1 to 13.1, while lean reactor cars operate at air-fuel ratios of 17 to 19:1, depending upon the degree of exhaust-gas re-circulation. These ranges of operation result in acceptable but not always good driveability.
A major difficulty in thermal-reactor systems has been achieving high enough gas temperature inside the reactor to burn up the engine HC and CO emissions. In the rich reactor approach, the chemical energy in the exhaust is used to obtain core gas temperatures of up to 1800ºF. Rich reactor systems achieve better emissions control than lean reactor systems, but have a much higher fuel-economy penalty and have more severe durability problems because of the higher temperatures.
Systems that combine one or more thermal reactors with catalytic converters are also being developed. In most of these systems, the thermal reactor bolted to the cylinder head is used to achieve partial burn-up of the engine HC and CO emissions to reduce the load on the oxidation catalyst downstream.
General Motors is developing a reactor-catalyst combination called a triple-mode system. The aim of the system is to avoid damaging the NOx and oxidation catalyst beds by overheating during engine operation at high load. At about 55 mph vehicle speed, the dual-catalyst system is bypassed through a thermal reactor. HC, CO, and NOx emissions from the vehicle are all higher during the bypass mode. At lower speeds, the bypass is sealed with a valve. This system has met the 1976 standards at low mileage; its durability has yet to be established. The system is more complicated than the dual-catalyst system; the development of an effective valve is a formidable problem since the bypass must be sealed tight when not in use; and the emissions at high vehicle speed are higher than values obtained with the dual-catalyst system alone. The claimed, but yet to be demonstrated, advantage is that catalyst life would be extended by the elimination of prolonged high-temperature catalyst operation.
Another reactor-plus-catalyst approach is being developed by Questor. Their system consists of a small-volume thermal reactor bolted onto the cylinder head, in which partial oxidation of engine HC and CO emissions occurs, followed by an Inconel 601 screen NOx-reduction catalyst, followed by a final oxidizing thermal reactor. Air is injected into the exhaust ports and downstream of the NOx catalyst. The engine is operated fuel-rich, so there is a fuel economy penalty relative to the 1971 production test vehicle of 25 percent in stop-and-go driving.
Low-mileage emissions are below the 1976 standards, and some durability has been demonstrated. Mileage accumulation is being done with highly leaded fuels, and lead poisoning of the catalyst appears not to be a problem. Catalyst overheating is controlled by water injection.
Wankel Engine
The Wankel rotary engine is being developed by some manufacturers because it offers the potential of being a small, smooth-running, lightweight, relatively inexpensive power plant. Fuel economy of the bare engine is generally poorer than that of a piston engine of comparable power, and there have been durability problems, but development is continuing.
At the present state of development, emissions from an uncontrolled Wankel engine compare with those of an uncontrolled piston engine of equivalent power approximately as follows: hydrocarbons 2-5 times higher, carbon monoxide 1-3 times higher, and oxides of nitrogen 25-75 percent lower.
The effect of EGR on emissions from a Wankel engine without a catalyst or thermal reactor is shown in Table 3-6. With penalties in fuel economy and driveability, EGR can be quite effective in reducing NOx emissions. However, HC and CO emissions are high, and these can be reduced to approach the 1975/76 standards only by using external emission-control devices. The small engine size provides an advantage in packaging these devices.
With its inherently high exhaust temperatures and its exhaust ports all adjacent, the Wankel engine is particularly well suited to emission control by a thermal reactor. Current production rotary engines on compact cars operated with rich carburetor settings and thermal reactors have been developed to meet the 1975 standards with NOx levels of about 1 gram per mile. However, the fuel-economy penalty compared with a current equivalent piston engine is about 30 percent.
The use of EGR and richer carburetion with the thermal reactor reduces NOx-emission levels, but not yet to 0.4 grams per mile with adequate driveability. The fuel-economy penalty increases by a further 5 percent. Table 3-7 lists the best emission levels obtained with thermal-reactor Wankel-engine systems. Several of these cars have met the levels of the 1975 standards, and one has nearly met the 1976 levels.
Toyo Kogyo has demonstrated durability of the rich thermal-reactor system by testing one car successfully to 50,000 miles, and with another still under test at 28,000 miles. The emissions from these cars have remained below the 1975 standards, and work to achieve the 1976 NOx requirement is continuing.
The best results achieved with oxidation catalysts and rotary engines are shown in Table 3-8. Levels approaching the 1975 standards of HC and CO have been achieved at low mileage with a compact car.
Other control approaches with the Wankel engine have been less extensively developed. At leaner carburetor settings, oxidation catalysts have been shown to reduce HC emissions at low mileage in a compact car to levels about 50 percent above the 1976 standards, and CO to well below the standard. However, the durability of the rotary-engine systems with catalysts has not been evaluated. EGR has been shown to reduce NOx levels below 0.4 grams per mile, but engine HC and CO emissions were comparable to uncontrolled piston-engine levels.
Stratified-Charge Engines
Stratified-charge spark-ignition engines are being developed to remove some of the performance and emissions limitations of conventional spark-ignition engines by controlling the air-fuel mixing and combustion process occurring inside the engine cylinder. The basic concept is not new and was first suggested in the early 1920's. Two stratified-charge engine types have been developed to the multi-cylinder engine stage.
Fuel-Injected Stratified-Charge Engines
This stratified-charge engine concept employs a combination of air inlet port swirl and high- pressure timed combustion-chamber fuel injection to achieve a local fuel-rich ignitable mixture near the point of ignition while the overall air-fuel ratio supplied to the engine is fuel-lean for most operating conditions.
Research and development on this concept, the open-chamber stratified-charge engine, has been sponsored by the U.S. Army Tank and Automotive Command (TACOM) for a number of years. This work has involved development and testing of two engine designs, one based on the Ford Programmed Combustion Process (PROCO) and the other based on the Texaco Combustion Process (TCP). The present TACOM vehicles employ exhaust-gas recirculation for NOx
control and oxidizing catalysts for control of carbon monoxide and particularly unburned hydrocarbons. As a consequence, the usual problems with oxidizing catalyst durability have been experienced. These engines are four-cylinder engines nominally rated at 70 horsepower for the military jeep.
Military vehicles equipped with four-cylinder L-141 engines modified by both Ford and Texaco are currently undergoing emissions durability tests. Emissions levels for these vehicles, which are equipped with oxidizing catalysts and exhaust-gas re-circulation, are presented in Table 3-9 for the Ford and Texaco vehicles. Results indicate that with low vehicle mileage these systems are capable of meeting the 1976 federal emissions standards. Durability results are also presented.
The most important conclusion from the durability tests is that the basic combustion process is stable. Difficulties with plugging of EGR systems with particulates were experienced during this mileage accumulation. The Ford engine required two catalyst changes and frequent maintenance. The Texaco engine used three oxidation catalysts in series to achieve the required HC and CO emission control. These engines are still in the research stage and are not production prototypes.
Durability tests with these vehicles are continuing at the present time. In addition to work performed under contract to TACOM, Ford has conducted experiments involving this type of engine in passenger cars. Both four-cylinder and eight-cylinder engine conversions have been used. In addition, stratified-charge engine installations have been made on two commercial-type vehicles. It is evident from Table 3-10 that several of the vehicles under test in this program are capable of meeting the 1976 federal emission standards at low vehicle mileage.
One of the advantages of the stratified-charge engine is excellent fuel economy relative to conventional engines, particularly when emissions controls are applied. The original version of the stratified-charge L-141 engine developed for optimum fuel economy showed a 30 percent fuel economy gain over the conventional carbureted engine. However, as in a conventional engine, when EGR is used to reduce NOx emissions, the fuel economy is reduced. With NOx emissions at 0.33 grams per mile, the emissions-controlled stratified-charge engine fuel economy is comparable to that of the original L-141 conventional engine. With less EGR, at 0.7 grams per mile NOx, about a 10 percent fuel economy gain is obtained.
Carbureted Stratified-Charge Engine
An alternative approach, the CVCC system, now being developed by Honda, achieves charge stratification with a prechamber and dual carburetor. The engine uses a conventional engine block, pistons, and spark plugs; only the cylinder head, intake and exhaust manifolds, and carburetor are modified. The cylinder head contains a small precombustion chamber in addition to the main combustion chamber. The spark plug is located in the prechamber, which is fed through a separate carburetor and intake system with a fuel-rich mixture through a small third valve. The main carburetor and intake system feeds a fuel-lean mixture to the normal intake valve.
The fuel-rich mixture ensures good ignition; the approximately stoichiometric mixture at the prechamber exit propagates the flame into the fuel-lean mixture in the main. chamber. A slow-burning flame is required to reduce NOx formation and allow HC and CO burn-up inside the engine. Emissions of Nox, CO, and HC are all lower than those of a conventional engine at the same lean air-fuel ratios.
In February 1971, emissions data with this system on engine dynamometer tests indicated the engine could meet 1975 standards; the first successful car test that met the standards was in Spring 1972. In addition to developing a 2-liter, 4-cylinder engine for their own vehicle, Honda has applied the same techniques to modify two Chevrolet Vega 4-cylinder engines.
The Honda system is the most developed stratified-charge engine to date and has the lowest bare-engine emissions. Low-mileage emissions data are given in Table 3-11 for 54 Honda vehicles and two modified GM Vegas. All these cars met the 1975 standards without EGR or exhaust treatment, and Honda has expressed confidence that larger engines using the CVCC approach could also be made to meet 1975 standards without a catalyst. Especially impressive is the standard deviation of the low-mileage emissions of these vehicles. The standard deviation is 10 to 15 percent of the mean emissions. In comparison, mass-produced conventional-engine vehicles show standard deviations of 30 percent of the mean at higher emission levels.
Three Honda cars have completed 50,000-mile durability testing and met the 1975 standards with ease at every 4,000 miles. Data for these tests are given in Table 3-12. The Federal Test Procedure 11-lap mode was followed in these tests. Maintenance required was minor.
In a recent series of three tests at low mileage, the average emissions measured were 0.25 grams per mile HC, 2.5 grams per mile CO, and 0.43 grams per mile NOx. These levels were achieved by improving the configuration of the auxiliary combustion chamber and the air-fuel control pattern. No EGR or exhaust-treatment devices were used.
The emissions are not especially sensitive to variations in air-fuel ratio. Thus the required performance of the double carburetor system is no more demanding than current requirements. The two throttle plates are linked mechanically. The mean air-fuel ratio varies with operating mode.
The new cylinder head is about the same height as a conventional head. The new head, intake, and carburetor on the modified Vega fit comfortably into the engine compartment. The engine can operate on regular leaded gasoline; durability testing has been on unleaded gasoline to simulate fuel anticipated in the United States in 1975.
The effects on vehicle performance of the CVCC system are small. There is a slight loss in power for the same engine displacement due to leaner operation and decreased volumetric efficiency. Fuel economy is essentially unchanged. There are no driveabllity penalties.
Development of the Honda CVCC engine to achieve lower NOx emissions is continuing. The effects of EGR and modifications to the basic combustion process are being examined.
Effect of Emission-Control Devices on Vehicle Performance, Driveability, Fuel Economy, and Safety
Some of the emission-control devices and techniques required to meet the 1976 emission standards have a profound effect on at least three areas of vehicle performance: acceleration capability, fuel economy, and driveabllity. There is also some concern that poor performance of such cars will make them unsafe in certain circumstances, for example, if the vehicle stalls when accelerating into fast-moving traffic. The customer is sensitive to these characteristics which affect both his pocketbook and his attitude toward any particular vehicle. Traditionally this area has been one in which customer complaints and warranty returns have been especially prevalent.
It is therefore not surprising that manufacturers have registered great concern in the past about the adverse effects of emission control devices. By the same token, however, the market place imposes considerable inherent motivation for manufacturers to devote great attention to product improvement in these areas.
The comments that follow in this section refer primarily to vehicles equipped with the dual- catalyst emission-control system.
In general, vehicle acceleration capability is reduced by control measures applied for control of all three pollutants (HC, CO, and NOx); however, NOx control measures which reduce combustion temperature have the most serious deleterious effects. Reductions in compression ratio to enable use of lower-octane gasoline resulted in acceleration penalties, as did the minimization of enrichment techniques formerly provided specifically for rapid acceleration capability. In addition, the use of EGR to reduce combustion temperatures and thereby inhibit NOx production imposes a severe acceleration penalty.
Losses in fuel economy accompany most of these losses in acceleration capability and are aggravated by countermeasures taken to overcome deficiencies in acceleration capability and driveability. Many of the smaller engines have been dropped in the various car lines. The use of a larger displacement engine results in a fuel economy penalty for both city and open-highway driving. When EGR is used to control NOx emissions, the mixture must be enriched to retain adequate driveability, causing drastic reductions in fuel economy.
The most troublesome of numerous driveability problems is the cold-start problem. The quick choke action and subsequent lean mixtures required to minimize HC and CO emissions introduce problems with engine stalls and unsatisfactory drive-away during warm-up. EGR and spark retard cause such problems as lack of response, die-outs, and hesitation on acceleration.
In its January 1, 1972, report, the CMVE concluded that all three areas of vehicle performance discussed above would be adversely affected by the 1975 emission-control systems. Information received from manufacturers indicated losses in acceleration capability ranging from a minimum of 5 percent to a maximum of 20 percent over 1971 levels. All manufacturers anticipated losses in driveability, in some cases indicated to be severe. Anticipated increases in fuel consumption ranged from 5 to 15 percent for standard sized cars up to 20 to 30 percent for small cars, again over 1971 levels. Much of the deterioration in performance was anticipated to come with the introduction of NOx requirements in 1973, and early reports on performance of the new models have confirmed this.
During 1972, the CMVE has received reports on both the 1975 and 1976 emission-control system progress. While manufacturers are still concerned with performance, particularly fuel consumption, the concern over vehicle driveability has diminished.
No substantial new acceleration, fuel economy, or driveability problems are introduced with the 1976 emission-control systems compared with the 1975 systems. At the same time, considerable progress has been made in finding solutions to problems that appeared to be very serious one year ago. It seems likely that competitive pressures will result in further improvements and improved reliability in these performance areas. The effort required is essentially engineering development based on extensive field experience with these new systems. The major long-term concern should be the continuing fuel economy penalty which results from the decreased compression ratio to allow the use of unleaded fuels, compounded by the use of EGR to control NOx emissions to very low levels, and aggravated by the increased engine sizes introduced to compensate for the loss in performance.
Alternative fuels
One approach to reduce emissions from conventional engines is the use of alternative fuels. The use of liquefied natural gas (LNG), liquefied petroleum gas (LPG), hydrogen, and alcohols have been considered by the Committee.
Liquefied natural gas and liquefied petroleum gas
Both industry and governmental groups have evaluated natural gas and propane (LPG) to determine their capability in reducing emissions from automobiles. One engine manufacturer showed that emission levels approaching the 1975-76 standards can be achieved, but exhaust gas re-circulation is still required to reduce NOx formation to the 1975-76 standard. There is an 8 percent loss in peak engine power (350 cu. in. 1970 engine) from gasoline when using LPG and a 15 percent loss using natural gas. There is a substantial loss in fuel economy (30 percent), and driveabllity is impaired. The use of LPG for starting and warm-up in a dual-fuel car using gasoline for conventional operation was attempted. Cold-start emissions are decreased.
On an experimental natural-gas 6-cylinder engine sized for but operation [sic], another manufacturer showed that the use of compressed or liquefied natural gas would produce emissions which would meet 1975 standards. The 1976 NOx standard could be met only with EGR, a catalytic after-burner, and a great reduction in performance. The emissions were odorless and there was no particulate matter present.
There are over 5,000 cars converted to run on gaseous fuels in the Los Angeles basin where gas supplies and liquid systems have been joined together to provide the gaseous fuels to the car operators. Emissions are cleaner, maintenance is reduced, but a heavy bulky tank is required to hold the gaseous fuel.
Status of liquefied-gas substitutes for gasoline
The CMVE has investigated the technical problems and economic factors involved in supplying natural gas and LPG. It is possible to modify the petroleum refining process so that LPG can be substituted for gasoline for motor vehicles. The original capital costs would be in the $50 billion range. The fuel costs to the customer would be about twice as much as gasoline presently costs.
Also, there is a serious net loss of energy in changing from gasoline to LPG. The percentage of crude oil consumed in the processing operations would increase from about 4 percent to about 14 percent. This would be an unrecoverable waste of natural resources.
There is not enough LPG, LNG, or Synthetic Natural Gas (SNG) currently available to be significant if conversion were desired now. A three-year lead-time for making changes for supplying these alternative fuels is a minimum.
Hydrogen
Hydrogen gas has three properties which when taken together, give it a unique potential as a vehicular fuel. First, since there is no carbon in the fuel, the problems of unburned hydrocarbons and of carbon monoxide do not exist. No after-burner, catalyst, or other secondary reaction vessels are needed.
Second, the flammability limits of hydrogen are extremely wide. The volume percentage of hydrogen in air can range over a factor of 19 and still be ignited by a spark. This contrasts with the factor of 5 for gasoline vapor. Because of this high flammability range, very lean mixtures of hydrogen gas may be used, thereby insuring that NOx will stay within acceptable standards. With hydrogen as a fuel, no EGR is needed to reduce NOx.
Third, the supply of hydrogen gas is virtually inexhaustible, although plants for its mass production are not yet available. Currently, the cheapest way of making hydrogen gas is to use natural gas as a base material. When natural gas approaches exhaustion, the cheapest way of making hydrogen gas will be to use coal as the base material. When the price of coal becomes too high, hydrogen can be made by heating or electrolyzing water. A source of energy is required to produce hydrogen by any of these methods.
An engine burning hydrogen gas at stoichiometric ratio emits no measurable hydrocarbons, organic or sulfur compounds, and only one-tenth the NOx as when burning gasoline vapor at its stoichiometric ratio. Furthermore, at an air-fuel ratio of 1.75 times stoichiometric, the NOx composition of the hydrogen exhaust is reduced by a further factor of 20, well below the 1976 standards. Several experimenters have reported satisfactory performance from internal combustion engines converted to hydrogen fuel.
The cryogenic fuel tank plus its hydrogen fuel would weigh 40 percent less than the conventional tank plus its gasoline having the same cruising radius, but would occupy five times the volume. Other storage methods are being sought.
One company is attempting to produce H2 and CO2 from unleaded gasoline in the car gas tank using a small reformer located in the trunk of the car. The H2 and CO2, produced in small quantities to avoid safety problems, could be burned cleanly in the slightly modified Otto-cycle engine. Questions remain on the ability of the reformer to carry out this reaction and on its efficiency, size and cost. Sound experimental work and socio-economic impact studies on the use of hydrogen as a vehicular fuel are required before unqualified success could be claimed for the approach. In any case large-scale use of hydrogen as an automotive fuel is not possible by 1976.
Alcohols
Alcohol has been proposed and used as a fuel for the internal-combustion engine; e.g., methyl alcohol is widely used as a racing fuel. Methanol has the advantage of providing a lower combustion temperature, reducing the NOx emissions, and it also has lower lean misfire limits than gasoline, thus reducing HC, CO, and NOx emissions while maintaining a satisfactory driveability. Emissions tests have been run on a 1970 American Motors Gremlin, using pure methanol as fuel, with a platinum catalyst converter in the exhaust. Emissions of HC, CO, and NOx, using the 1972 CVS Federal Test Procedure, were below the 1976 standards.
Methanol has a lower heating value than gasoline, so yields correspondingly fewer miles per gallon. Starting at low temperatures with methanol is a problem; volatile compounds have to be added to assure starting.
Similar data on ethanol are not available. Tests on gasoline with up to 30 percent ethanol as fuel show no substantial improvement in emissions over pure gasoline.
POTENTIAL OF SPARK-IGNITION INTERNAL-COMBUSTION ENGINES FOR MEETING STANDARDS IN USE
Introduction
Section 3 presented an evaluation of the feasibility of spark-ignition internal-combustion engines passing the certification test for 1975 and 1976. This section assesses the feasibility of such engines continuing to meet the standards in customers' hands.
In this assessment, the first question to be answered is the adequacy of the certification test to evaluate the emissions performance of vehicles in customers' hands. The next question is the maintenance required on prospective 1975-76 control systems to achieve compliance in use. This calls for discussion of the procedures necessary to ensure proper maintenance, namely the nature and feasibility of required testing and maintenance. The latter depends in turn on the adequacy of the service industry and the interest of state governmental bodies enacting required legislation.
Perhaps the overriding question is whether adequate consideration has been given to maintenance in the design of planned 1976 emission-control systems, most of which involve the use of catalytic converters.
Differences Between Certification Test and In-Use Operation
It is relevant to discuss here some of the more significant differences between the stresses on the emission-control system experienced during certification and during normal customer operation.
The driving modes specified in the mileage-accumulation schedule of the certification procedure do not represent all the possible modes encountered in real life. There are some not included, such as sustained operation at high engine power, and long decelerations, that will provide severe tests for emission-control systems, especially those using catalytic converters. Ford believes the certification test driving does not provide enough mechanical stress on the catalyst, especially if the driving is carried out on an automatically operated dynamometer, which is the usual procedure for the accumulation of mileage. The vehicles are stopped only for tests every 4,000 miles; in normal use, of course, vehicles stop much more frequently. With catalytic systems, catalysts will heat up and cool down several hundred more times in 50,000 miles of normal use than in certification. The New York City Department of Air Resources has also pointed out that the certification procedure does not represent actual driving conditions because of insufficient allowance for the effect of accessories.
Sufficient data are not available to fully assess the effects of low-temperature operation on catalyst durability. However, increased loading on the catalysts due to low ambient temperatures, as well as occasional bouts of freezing and thawing, appear to offer in-use conditions that would lead to the necessity for more frequent catalyst replacement than during the certification procedure.
The durability phase of the certification test should be sufficiently demanding to establish that the emission-control systems will perform in the hands of customers. The allowable maintenance in the durability test – one major tune-up in 50,000 miles – was selected to make the test tough and realistic. In real life, much more frequent maintenance will probably be necessary to keep 1975-1976 systems within the specified emissions levels, and the Act (Section 207(c) (3) ) requires the manufacturers to furnish written maintenance instructions with each new vehicle.
Manufacturers are in agreement that more maintenance than is allowed in the certification test will be necessary in actual use. This is supported by the fact that manufacturers are now requiring more maintenance as a condition of warranty than they were allowed in the certification procedure.
In summary, the Committee recognizes that vehicles in customers' hands will not be driven according to the CVS-1975 test procedure, will not be driven according to the durability driving schedule, and will not receive the maintenance specified by the manufacturer without rigid enforcement procedures. Therefore, stresses on the systems may be substantially greater in customer usage than in certification, and in-use emission levels may thus be correspondingly higher.
Maintenance Procedures Required for 1975-76 Systems
Although there are no data on the deterioration of the projected 1975-76 control systems in customer use, there are data on the typical deterioration of emission levels from the cars now being driven. These data provide some limited indication of the deterioration as a function of mileage that can be expected to occur with the new systems. Information provided by EPA, California, and ARCO on 1971 and prior model year cars indicates a substantial increase in emissions in customer use. Emissions were found to exceed the applicable standards at relatively low mileage.
The most comprehensive surveillance data on the emissions of cars in use have been taken by the California Air Resources Board. Data taken between January and March 1972 showed that 1970 model year cars, with an average accumulation of 32,000 miles, exceed the applicable California standards for all three pollutants by amounts ranging from 10 to 60 percent. 1971 model year cars, with an average mileage accumulation of only 13,000 miles, exceeded the applicable standards for at least one of the pollutants. Cars for these model years contain neither catalysts nor much of the other complex hardware proposed by most manufacturers for 1975-76 model years. Further, the applicable standards for California for these model years are many times higher than the federal standards for 1975 and 1976.
The dual-catalyst emission-control system proposed by most manufacturers for 1976 model year vehicles is a far more complex system than that used on current vehicles. Involved are a multitude of control valves, quick-warm-up systems, control circuits, etc., as shown in Figure 3-1. Of all these components, the catalysts themselves appear to be the least durable items. Spark plug misfire, sustained operation at high engine power, and descent down long hills are examples of situations that would result in catalyst overheating and possible failure. Such vehicle operation and driving modes would not occur in the mileage accumulation specified for the certification test.
In addition, there appears to be little incentive for the car owner to maintain the emission-control system. To the contrary, the engine will run more efficiently and smoothly with some elements of the emission-control system inoperative. For example, plugging of the EGR system would improve gas mileage, although also increased NOx emissions. Increase of choking time would improve vehicle starting characteristics, yet also increase cold start emissions.
The importance of adequate maintenance is recognized in Section 207(b) (2) (A) of the Clean Air Act, which requires manufacturers to warrant their emission-control systems to the purchaser if the vehicle or engine is maintained and operated in accordance with the manufacturer's instructions, and, in the recall provisions of Section 207(c) (1), which empowers the Administrator to recall a class of vehicles or engines if a substantial number of vehicles in each class, although properly maintained and used, do not conform with the standards. The recall provisions could be enforced by relatively frequent analysis of emissions from a sample fleet, carefully chosen for appropriate statistical representation. Should these give evidence of rapid deterioration of the control system, the recall power provided by Section 207(c) of the Act may then be invoked, with the manufacturer specifically enjoined to replace defective parts, and to defray the associated labor costs at his expense. In this situation, the burden falls not only on the manufacturer for recall and repair, but also on the car owner, for performing the required routine maintenance and for responding to notification of recall.
In order to achieve the reduction in automotive emissions anticipated by the Clean Air Act, it is apparent that methods must be provided for ensuring proper maintenance of the 1975 and 1976 emission-control systems in public use. Methods of ensuring the required maintenance include:
1. Requiring the service industry to adjust each car to manufacturer's specifications when performing any maintenance.
2. Periodically testing all cars and designating for adjustment or repair those not meeting pre-selected standards.
3. Periodically subjecting all cars to adjustment or repair.
The first method is based on the probability that the manufacturer's specifications for 1975-76 vehicles will represent adjustment to minimum emissions. Mechanics currently adjust cars for high performance. If they continue to do this when manufacturers' specifications are for low-emission adjustments, the cars will emit above the standards.
The principal variations in the second method are related to how much of the work is done in state-owned and how much in privately owned facilities, the testing procedure used, testing frequency, pass/fail standards, provision for retesting after repair/adjustment, and disposition of vehicles that cannot meet the standards.
The significant variations in the third method are related to whether the cars are adjusted to some pre-selected standards and whether preventive maintenance is included. Preventive maintenance may be the best feature of this method. Other methods for ensuring the maintenance of cars in use are probably feasible only if engineering changes, which do not seem likely by 1976, are made. They are:
4. Repair at the time of failure of any important emission-control device based on the presence of devices that signal the failure not only to the driver but also to the traffic officer.
5. Repair at the time of failure of any important emission-control device based on the manufacture of control systems that noticeably degrade the vehicle performance when an important component fails.
6. Prescribed maintenance at predetermined intervals. This method would require strict quality control of the manufacturing process so that essentially every car was held to a configuration proved to give low emissions in actual use.
Adequacy of the Service Industry Training
The service industry at the present time is not adequate to service 1975-76 cars from an emission control standpoint. Knowledge of the devices, the diagnostic equipment, and the number of mechanics are inadequate. The number of vehicles per mechanic in the country has risen from 75 in 1950 to 145 in 1970. During the same time, cars have become more complex and less repair-oriented in design. The states that have studied the problem all feel that training on emission-control devices is needed and that the states should be responsible for recommending suitable equipment. New York plans to certify garages as properly equipped for emission-control work. California licenses mechanics, and New Jersey will depend on the promise of large- volume business to motivate the private sector to establish its own training and licensing programs. The service-industry mechanic will have to be trained to understand and perform repairs and adjustments whether or not he performs the complete or partial diagnosis to isolate problems causing excessive emissions. The amount of training will vary slightly with the degree of state control on mechanics, but will generally have to be extensive.
Number of mechanics
The number of mechanics required to maintain 1975-76 emission-control systems will depend upon the interpretation of the 50,000-mile warranty provision of the 1970 Clean Air Amendments. If the new-car dealerships assume the responsibility and owners are required to bring their cars into the dealers' garages for periodic inspection and maintenance, a new force of about 12,000 mechanics per year will need to be trained for the dealers, on the assumption that 1975-76 control systems will require about two hours more per year than now spent to perform routine inspection and maintenance. This number of men will need to be added each year for about five years if new-car dealers maintain new cars during warranties: i.e., a new work force of about 60,000 men will be needed by 1980. When warranties expire, experience has shown that most owners will take their cars to garages other than new car dealers; hence, after 1980, an indefinite number of additional mechanics in garages other than new car dealers will need to be trained. The number of mechanics needed in the service industry is difficult to estimate because most of them will probably work only part time on emission-control.
California is the only state that licenses mechanics to install and repair emission-control devices at this time. Only a few other states have plans to license mechanics. A Northrup study for the State of California and a similar study by the State of New Jersey have shown that a simple indoctrination of mechanics is not sufficient to obtain cost-effective emission tune-ups.
Equipment
The garages in the service industry will need to be significantly upgraded with new equipment to perform diagnoses and tests to determine if vehicles need adjustment or repair and to show that the adjustments and repairs were accomplished. The amount of equipment needed will depend partially on whether or not the state operates inspection stations and what kind of inspection test the state performs. However, the state inspection system that would significantly reduce service- industry equipment requirements would be one in which the state would perform the complete diagnosis and instruct mechanics on what parts to replace.
State Action
Inspection and maintenance systems
State governments have been interested in inspection and maintenance of motor vehicles as a means of reducing exhaust emissions for many years. For example, the New Jersey system, put in operation on July 5, 1972, is the outgrowth of an investigation that started in 1966. It should be noted that even in the New Jersey system, which is the farthest advanced, the features of compulsory maintenance will not be instituted until July 5, 1973. The first year of inspections is being used only to educate the public and eliminate the difficulties.
California has required a certificate of compliance from licensed installation stations since PCV valves were first required in 1963. Idle-exhaust measurement for HC and CO is also now done as part of California's roadside safety inspection. Cars exceeding inspection standards must be taken to a licensed mechanic for adjustment, and a notice that the work was done must be returned by mail.
This interest in inspection/maintenance systems undoubtedly arose from the knowledge that a well-maintained car emits less pollutants. A large impetus toward such systems has been added by the realization that the manufacture of cars meeting the 1975 and 1976 federal standards is not sufficient unless some system can be found to keep the complicated emission-control devices operating properly.
A survey of the present status of the state efforts to establish inspection/maintenance systems and an investigation of the reasons for the long-time delay in even the most active programs are, therefore, relevant parts of the Committee's investigation of methods of ensuring that the 1975- 1976 cars meet the federal standards in actual use.
Certain federal action or lack thereof has had a noticeable effect on state action in this field: The Clean Air Act generally preempted motor vehicle emission control for the federal government. This raised several problems connected with the design of state systems. The first is a tendency toward delay; in the few cases in which a state had already started the design, revision was necessary and the states which had not started tended to wait for federal action.
Also pertinent are Sections 110 and 207 of the Clean Air Act. Section 110 requires the states to submit a plan for the implementation of the national ambient-air quality standards, and Section 207 deals specifically with motor vehicle compliance. Under the latter, once EPA determines that adequate inspection procedures are available, they are to be established by regulation. Since implementation plans are not yet final for all the states, and since the determination called for by 209(h) has not been made, resulting uncertainty inevitably leads to delays in program planning by the states.
The selection of the most suitable method for a state system depends not only on the engineering approach finally adopted by the manufacturers and on the test procedure designated by EPA, but also on whether the method is to be used only to minimize emissions or also to enforce the warranty on individual vehicles.
If the purpose is only to minimize emissions, periodic repair/adjustment of all cars including preventive maintenance is a possible choice. This approach would require no special inspection facilities owned by the state, but it would require careful surveillance of privately owned garages and additional equipment in these garages.
Requiring the garages to adjust each car to manufacturer's specifications when performing any maintenance also does not require state-owned facilities but does require close supervision. Preventive maintenance could be part of this method and, with this addition, this method only differs from the one first discussed by being voluntary instead of mandatory.
Periodic inspection of all cars with measurement of the exhaust emissions and compulsory adjustment or repair of those cars that have emissions exceeding pre-set standards is the method usually meant by an inspection/maintenance system. However, tests other than emissions measurements can be used for inspection in this method. It is normally thought of as occurring annually at the time of license renewal. This system can be operated on three bases: inspection and repair/adjustment in state facilities, inspection in state facilities and repair/adjustment in privately owned garages, and inspection and repair/adjustment in privately owned garages licensed or franchised by the state. The second of the three choices is the usual one principally because of the public distrust of the service industry, which causes the public to prefer inspection by the state. The first one is not chosen because of the reluctance of the state to compete with private enterprise and because of the many complications connected with building state-owned repair facilities.
If the method selected must also include enforcement of the individual car warranty, it will be built around a test yet to be specified by EPA. It is usually assumed that this will be some type of short emissions test, but Section 207(b) of the Act would allow the Administrator to decide that functional tests on the components correlated reasonably well with the results of the CVS-CH test. It is conceivable that these tests could be done in privately owned facilities, but the quasiofficial nature of the warranty test and its consequences make a state-owned inspection lane the more obvious choice.
If the method includes enforcement of the warranty, EPA will provide the appropriate test; if not, there is wide latitude. Diagnostic tests will be part of any system, since such tests and repair/ adjustment cannot be separated, and repair by mechanics is the only operation that provides a direct emission reduction. Inspection lanes select the vehicles needing adjustment or repair but otherwise do nothing to reduce emissions unless the results assist or control the mechanics making the repairs. Inspection lanes can assist and control the mechanics in one or more of the following ways:
1. Detect vehicles with excessive emissions (needing repairs).
2. Detect vehicles with excessive emissions and give a partial diagnosis to help the mechanics get started.
3. Provide a complete diagnosis of repairs needed on vehicles with excessive emissions; and
4. Insure that repairs are complete and correct.
A study by Northrop Corporation for the State of California found that a schedule of diagnostic tests was not a cost-effective approach to the emission control of used cars. The study showed that exhaust-emission tests by a short dynamometer test (Key Mode) or an idle test gave partial diagnostic information and was more cost-effective. Partial diagnostic information was given to the repair mechanic to assist him in the final diagnosis.
The Key Mode and idle approach were reasonably effective for correcting the major emissions problems in used cars. They may not be adequate for controlling 1975-1976 cars to much lower emission levels because there is not yet available a short test that is precise enough to give a pass or fail that is meaningful in terms of meeting 1975-1976 standards. 1975-1976 cars will require a much more thorough diagnosis of the complete emission control system.
Diagnostic tests could be useful in inspection lanes if they provided complete information on engine and control-system failures and operation. This could be accomplished by an automatic and computerized diagnostic console, programmed to accomplish quickly and inexpensively one or both of the following:
1. Functional tests showing that the engine and/or the control system are not within specifications where it is known that the combined system will meet the standards.
2. Diagnostic tests showing what parts need to be checked and/or replaced.
The mechanics in the service industry could be trained to understand and use the information supplied by the inspection-lane diagnoses.
Diagnostic Tests at Garages
The facilities and equipment of garages can be upgraded to perform diagnostic tests at periodic intervals. The advantage of this approach is that the mechanics accomplishing the repairs would have first-hand knowledge of the diagnostic test results. Disadvantages of this approach for 1976 vehicles are:
1. Functional NOx catalyst activity tests and NOx analyzers have not yet been developed to diagnose NOx controls. It may be possible to check the activity of a reduction catalyst in the oxidizing mode.
2. The engines may need to be loaded in order to produce enough NOx for a meaningful test.
Selection of Repair/Adjustment Standards
Since the object of an inspection/maintenance system is to reduce the total amount of pollutants emitted to the atmosphere, there is a strong incentive to require repair and adjustment for a higher percentage of the cars found to be over standards. However, as more and more cars are adjusted/ repaired, the gain in air quality per dollar spent decreases; i.e., the cost effectiveness decreases.
With practical and cost considerations thus limiting the number of cars sent for adjustment and repair, the maximum emissions reduction is to be achieved by adjusting only those cars whose emissions are clearly high and leaving alone those below or near the satisfactory level.
The percentage of cars sent for adjustment/repair must be considered with great care also because it increases the load on both the service industry and the inspection lanes and because a high percentage of re-rejections will destroy public support, which is so important.
With the present state of the service industry, a sizable percentage of cars will not meet the standards after the first repair/adjustment if the levels are strict. In addition there is a shortage of mechanics of even reasonable training. If the standards are set to send a high percentage of cars for repair/ adjustment, the number of cars that cannot meet the standards without costly repairs will be so large that it will again affect public support.
Timing and Cost of Inspection Facilities
The time and cost required to set up inspection facilities depend to a large extent on the amount and type of related facilities that are already available. Three cases will be considered:
1. Safety-inspection facilities are already available and emissions testing can be added to such facilities.
2. Properly controlled, privately owned service facilities are available, with emissions testing done at such facilities.
3. Neither condition 1 or condition 2 exists and inspection facilities must be built.
New Jersey is an example of the first situation and it has proved relatively easy from a physical standpoint to add emissions testing to the state-owned safety-inspection lanes. For the idle test that they are using, equipment costs are about $2,000 per lane. On the assumption that legislative authority already exists, it should be possible to put emissions testing in operation in one year.
Extra manpower required would be one per lane. No meaningful estimate of operating or capital costs chargeable to the emissions testing is possible because of shared costs. Time and cost would both increase if the testing were also intended to enforce the federal individual car warranty.
California could be an example of the second situation since they license Grade A mechanics for various specialties including emission-control devices. The time required in this case should also not exceed one year. Costs for added equipment would again be about $2,000 per station for an idle test. Operating costs would be mixed with adjustment and repair costs and, consequently, a separate estimate is probably of questionable meaning.
The third situation has been studied in considerable detail by Northrop-Olson Laboratories and also by TRW. Because of the conditions assumed in this study, the cost results must be qualified although the results do give a good indication of the range to be expected. The land, structure, and equipment will cost from $23,000 to $60,000 per inspection lane, with a major portion f the difference in cost caused by the presence or absence of dynamometer equipment. These numbers are approximately confirmed by the TRW study which estimated $44,000 to $52,000 per lane for dynamometer-equipped facilities. Different tests not only use different equipment but they also have different throughputs per lane.
Based on these factors, the cost of land, structure, and equipment on a one-inspection-per-year basis is between $1.30 and $8.80 per car when calculated for California's population distribution and 10 million cars.
Operating costs in 1976 would be between $1.20 and $4.00 per car per year, again under California conditions. The original capital costs are a small fraction of this and they are included with structures amortized over 20 years and equipment over 5 to 10 years.
Training time for personnel would be between 90 and 180 hours per man, which includes 40 hours classroom training.
Again on the assumption that legislative authority already existed, it would probably take 1.5 to 2 years to acquire land, erect and equip the buildings, and train personnel. At least one year must be added to any of the above time schedules if legislative authority does not already exist. Even more time must be added if an operational plan does not exist; witness the New Jersey and New York experiences. A state just starting would probably be fortunate to have a fully operational inspection system in 4 years.
In summary, only few states have any semblance of a testing/inspection system that would be adequate to ensure compliance in use. Most states do not even have plans for such systems. The present service industry is inadequate to maintain the complex emission-control hardware called for with the dual-catalyst system planned for use in 1975-76. With this pessimistic appraisal of feasibility, it is well to consider alternate approaches.
Incorporation of Maintenance Considerations in Emission-Control System Design
The pessimistic appraisal of the feasibility of vehicles equipped with dual-catalyst control systems meeting the standards in customer use is indicative of a lack of consideration of maintenance in the design of such systems. From the data presented in Section 3, it appears that several systems offer maintenance advantages over the dual-catalyst system, although the low- mileage emissions of such systems on experimental vehicles, may not currently be as low as those of the dual-catalyst system.
The three-valve carbureted stratified-charge engine and the Wankel engine with thermal reactor show potential for low emissions without the use of catalysts. HC and CO deterioration factors for the former, at 1975 levels and as measured on the federal driving cycle, are considerably less than those from catalyst-equipped vehicles.
Development work is required on the engine to reduce NOx emissions to 0.4 grams per mile; however, such a development effort would seem well worthwhile due to the potential of the engine for reduced maintenance and improve performance in use over the dual-catalyst system.
Systems employing precise control of air-fuel ratio with a feedback loop, discussed in Section 3.6, have several possible maintenance advantages. Since an air-fuel ratio near stoichiometry results in almost optimum performance, the serious performance and fuel penalties inherent in other NOx-control methods would be eliminated; the advantage, from the owner's viewpoint, of an inoperative control system would be removed. In fact, any malfunction of this system might easily degrade vehicle performance so that the owner would be encouraged to get the emission-control system fixed.
Since such a feedback loop makes the engine essentially self-tuning, this approach should also eliminate a large fraction of the inherent variability between individual vehicles that results from manufacturing tolerances. Possibly also, operational variabilities that result from variations in driving habits, fuel consumption, atmospheric parameters, and induction-system deterioration would be largely eliminated. Thus a larger fraction of cars would operate as designed and emit less pollutants.
Excessive catalyst temperature caused by the simultaneous presence of excess oxygen and large amounts of combustibles would be eliminated since neither rich mixtures nor secondary air is required. Finally, since the system includes an electronic control circuit, installation of signals for malfunctions should be relatively easy.
Summary
Emissions of 1975-76 vehicles in customer usage can be expected to be greater than those measured during certification. Because of the added emission controls, most vehicle configurations proposed for these years will require more maintenance than at present. For all systems, some additional inspection and maintenance will be necessary to assure that the vehicles are meeting standards in use. Some legal enforcement procedures will be required to assure that necessary inspection and maintenance are performed; otherwise, vehicles will very likely exceed the emission standards in use.
The service industry at the present time is not adequate to service the 1975-76 cars from an emissions standpoint. Only few states have a semblance of a testing/inspection system for emissions that would be adequate to ensure compliance in use.
A basic problem in establishing technological feasibility is that maintenance considerations have not been given adequate attention in design. The three-valve carbureted stratified-charge engine, Wankel with thermal reactor, and catalytic system with exhaust sensors and feedback control seem to have far more potential for achieving low emissions in use than the dual-catalyst system currently being proposed by most manufacturers for the 1976 model year.
MANUFACTURING, COSTS, AND PRODUCIBILITY
Manufacturing plans of the major automobile companies for 1975-1976 systems are not firm at this stage. Changes will almost certainly be made between now and the start of production.
However, each company has taken positive steps toward implementation of their best estimate of the components that might be introduced for 1975 and 1976. Schedules are compressed and significant risks are involved. Consequently most companies have more than one alternative plan for the emission-control system for these model years. In some cases, manufacturers have designed and/or made tooling for alternative configurations.
Manufacturability of several proposed engine systems
Several types of engines that might be produced in the 1976 model year have been evaluated from the view point of manufacturability and costs. These engines are: 1) the dual-catalyst system proposed by most manufacturers, 2) the diesel, 3) the Wankel, 4) the three-valve stratified-charge, and 5) a feedback-controlled system with electronic fuel injection and a three-way catalyst.
The dual-catalyst system
In response to California and federal regulations over the years, the automotive industry has progressively added to the emissions-control devices on automobiles. Due in part to a determined effort to preserve as much of the technology of the carbureted internal-combustion engine as possible, the approaches to emissions control have consisted of add-ons and relatively minor engine modifications. Although the various companies have worked independently, there have been many similarities in approach, and the typical pattern of hardware addition is presented in Table 5-1. Beginning with the 1976 model year, this system will include both oxidation and reduction catalysts; thus it is termed the dual-catalyst system. This system is shown schematically in Figure 3-1.
The corresponding increases in sticker price associated with these hardware additions are detailed in Table 5-2 and summarized in Table 5-3. According to these estimates, the additional price increase of 1976 models over those of 1975 is about $134.00, or nearly the same as the increment for the preceding year.
At this time it still appears possible for the manufacturers to mass-produce systems similar to that shown in Figure 3-1 for their 1976 models. However, until the systems show more likelihood of meeting certification for 1976, the manufacturers are reluctant to make major commitments, particularly for catalysts, and much more delay will make these systems technologically impossible for 1976 because of insufficient lead time. If this type of system is to be mass- produced in the 1976 model year, the following must have been accomplished by mid-1973:
Freeze design for production.
Build catalytic converter plant and line.
Commit to plant and equipment for substrate.
Commit to new carburetor production design.
Freeze design of early fuel-evaporation system.
Diesel Engine
Some light-weight diesels are currently being produced for passenger cars, mainly in Europe and Japan. However, because it is difficult to make a diesel engine meet the 1976 NOx standards, and, for other reasons discussed in Section 6.1, there is no serious effort to develop the diesel engine for large scale mass-production as a passenger-car engine. If diesel engines are developed to meet the 1976 emission levels, the emission control will probably be largely achieved by engine modifications and possibly turbocharging. Thus, even though the exact configuration is undefined, the manufacturability would not differ greatly from that of current diesel engines, and the major manufacturing problems can be identified.
The engines themselves are quite similar to Otto-cycle piston engines, but necessarily heavier to withstand higher operating pressures. The transfer and assembly lines for these engines are similar to those used for existing gasoline engines. Fuel-injection pumps and injection nozzles are now being produced on very modern mass-production equipment in England and Germany. Turbochargers have been produced in low volume for larger engines, and adaptation to mass production for smaller engines is quite feasible. Most of the technology for mass production of light-weight diesel engines is available but scattered, mostly in Europe. This wide dispersion of technology is a major barrier to the coordinated development of a low-emission diesel engine.
In addition to changes in the engine and its auxiliaries, conversion of automobiles to diesel power would require relatively major modifications of the frame, suspension, and body in order to accommodate the larger, heavier engine. If a diesel engine that can meet the 1976 emission standards is developed, and if, as assumed here, it is generally similar to present diesel engines, it should be possible to mass-produce them for the 1976 model year if the following have been accomplished by mid-1973:
Freeze design for production.
Arrange for transfer of European light-duty diesel technology.
Build low-volume production tooling.
Plan for conversion of gasoline engine lines for diesel engine production.
Plan body changes.
Arrange for supply of turbochargers (if used).
Wankel Engine
The Wankel engine is being mass produced in Japan and sold in the United States at competitive prices. The engine is in mass production in Japan at Toyo Kogyo with American sales of the Mazda in the United States projected at 350,000 units in 1975. A recent announcement indicates a production commitment to the Wankel engine by General Motors. There is every indication that a substantial number of Wankel-powered automobiles will be driven on United States roads in 1976.
The engine has a cost advantage due to its low weight per horsepower – about 1.5 pounds per horsepower compared to 4 to 6 pounds for a piston-type gasoline engine. The manufacturing advantages of the Wankel engine are that it can be manufactured and assembled on fully automatic production lines. The engine design will eventually allow a new frame and body design that will have many safety, space, and weight advantages. The implementation plan for the General Motors Wankel engine has it introduced into the low end of the line, possibly replacing both the 4- and 6-cylinder engines in turn. The optimum-cost volume per year of the Wankel engine will be between 450,000 and 600,000 engines per year. The small V-8s might also find a larger-diameter 2-rotor Wankel engine as a competitor. A 4-rotor Wankel engine is a more complex design with longer crank shaft. Two to four more years will be required on its development before it can be considered a competitor to the larger V-8.
The manufacturing requirements for the Wankel engine are concentrated around the following significant equipment: a trochoid grinder for the rotor housing, a rotary grinder for finishing of the end housings, an eccentric grinder for the rotor, some special plating equipment combined with surface preparation equipment, and special equipment for pressing and sintering the apex seals. These machines are available today from several machine-tool concerns and can be delivered within one or two years. Mass production conversions of these will require between one and two years of tooling design. An automatic assembly line and machining line combined will probably take anywhere from three to five years to develop and install.
The cost of a future Wankel-powered car will be $140.00 to $800.00 less per car than the corresponding 1976 dual-catalyst configuration; of this amount, $25.00 to $77.00 is due to the engine, and the remainder of the saving would come from design of a lighter, shorter car.
The Carbureted Three-Valve Stratified Charge Engine
Because the three-valve stratified-charge engine is basically an existing carbureted spark-ignition piston engine except for modifications to the cylinder head, carburetor, and manifolds, it presents relatively few production problems. Manufacture of all components is based on known and proven technology. Honda Motor Company plans to produce this type of system for their 1974 models in Japan, and they will introduce it in the United States in 1975. For another manufacturer to mass produce this system in model year 1978 would require the following accomplishments by mid-1973:
Transfer technology from Honda Motors.
Freeze design for production.
Decisions made and orders placed for new transfer lines for cylinder heads, manifold systems, and carburetors.
Design new camshaft-production line.
A Typical Feedback-Controlled System
Because of the apparent potential for emission reduction and ease of maintenance, which might result with further development of some of the feedback-controlled systems, manufacturability and costs of one of these systems were evaluated. The configuration studied included electronic fuel injection and a three-way catalyst. As with the dual-catalyst system discussed in Section 5.1.1, this approach requires relatively minor changes to existing engines, with the conversion from carburetion to fuel injection being the most significant. The mini-computer that controls the injection timing and duration is based on known technology, and manufacture of the catalyst is similar to that for the dual-catalyst system. Once a satisfactorily durable oxygen sensor is developed, its manufacture should be relatively simple. Production of this system for the 1976 model year is quite feasible, provided the following have been accomplished by mid-1973:
Freeze design for production.
Commit to pump and nozzle plants.
Build low-volume production tooling and vehicles.
Field test low-volume production vehicles.
Commit to electronic emissions control unit plant and tooling.
Manufacturability and Costs of Automotive Exhaust Catalysts
As discussed previously, most manufacturers plan to use a dual-catalytic system for 1976 model year vehicles. From a manufacturing standpoint, the problems of producing oxidizing and reducing catalysts are the same. The catalyst manufacturers who propose pelletized catalysts already have the sources for a substantial portion of the carrier materials and some capacity for coating with the active material. This type of catalyst is used extensively in the petroleum industry. The manufacturing facilities need only to be increased or additional similar type of equipment provided.
Many companies are active in the development of catalysts and substrates. In addition to the long-established catalyst and substrate manufacturers, General Motors has recently disclosed that they have developed an extrusion method for making monolith catalyst carriers. They have plans for constructing these facilities and have indicated their intention to become major emission- control catalyst manufacturers, including the carrier containers, and possibly the active material that is coated on the carrier.
It has become increasingly apparent that 1976 catalysts will require the use of large quantities of noble metals. The two noble metals of greatest promise are platinum and palladium; for oxidation alone, a car of 350-cubic-inch displacement would need up to 0.15 ounces of either metal. This figure would be doubled if the requirement for the NOx catalyst is similar. Thus, there would be a demand of as much as 3 million ounces for the initial installation of the catalytic converters required, a figure comparable to the world production in 1970. Ruthenium is the most promising NOx catalyst, although it is in short supply. The recovery of platinum contained in spent catalyst delivered to the door of precious metal refiners should be above 99 percent. The efficiency of scavengers in collecting spent noble-metal catalysts is difficult to estimate. Since the value of the recovered metal is of the order of $15-20 per car, efficiency of scavenging should be high. For comparison, copper is 500 per pound and 61 percent of scrap copper is recycled in the United States. Most base-metal catalysts are promoted with precious metals at less than 0.01 ounce per car. In this case, there is less incentive for scavengers to collect resources.
It appears that the required amounts of noble metal can be made available to meet production schedules if decisions are made early enough; postponement would cause increasing difficulties with delivery. Some companies have delayed decisions because of the very large commitments for opening mines and having new plants built.
Summary of costs of various proposed systems
The relevant cost concept is the total cost to the American people of meeting the emission standards, which must be weighed against the cost of air pollution by present automobiles with their attendant human discomforts and illnesses. This includes not only increases in automobile purchase prices, but also increased cost of fuel, maintenance, repair, and driveability that result from pollution-control devices. Of these considerations, it is especially difficult to relate poorer driveability to a cost in dollars, but the customer pays in other ways, e.g., through frustrations and delays. Dollar estimates of the other costs can be made, although these are necessarily imprecise because of uncertainties at this stage.
A summary of the estimated increments in annual costs due to emissions-control systems for several possible 1976 car and engine combinations is given in Table 5-4. The engines are those that have been discussed, and price increments have been calculated for those car-engine combinations that appear feasible. The stratified-charge 3-valve engine may eventually be developed for larger cars, but so far its potential for low emissions has been demonstrated only in small cars. The cost increments are measured from equivalent 1970 model cars as a baseline, and these annual costs are amortized over a five-year period. These figures include not only the direct cost of emissions hardware, but also associated costs of redesign of the rest of the car to accommodate the new systems. These associated costs include weight penalties, which can be quite significant in either direction; e.g., diesel-powered cars will be relatively heavy, whereas an automobile designed around the compact Wankel engine can be appreciably lighter than present cars.
Estimates of increased costs of fuel consumption and maintenance due to emission controls are also included in the figures in Table 5-4. Of the five engines listed, the emission-controlled diesel and the stratified-charge engines show promise for fuel economy competitive with 1970 gasoline engines. The feedback-controlled spark-ignition engine with electronic fuel injection promises reasonable fuel mileage, because of its operation near stoichiometry, but will still suffer a 10-15 percent fuel penalty over 1970 engines. The dual-catalyst system proposed by most manufacturers will use about 25 percent more fuel than its 1970 counterpart; and the Wankel configuration, which seems most likely to meet the 1976 standards, will probably pay a fuel penalty of approximately 30 percent, due to its rich mixture ratio.
Exercise to illustrate the impact of possible use of a mix of engines and control systems
As mentioned earlier, the American automobile producers are by and large seeking to meet the 1976 requirements with a dual-catalyst modified carbureted piston engine across their car lines.
However, it is quite unlikely that any single engine type or control system will prove suitable for all sizes and types of 1976 automobiles. Furthermore, several new low-emission engine configurations may well phase in to replace some of the carbureted piston engines. Clearly, phasing in of these various new engines and control systems and phasing out of the engines they replace will have an effect on sticker price due to the capital costs incurred. A computer simulation of the dynamics of such a process was carried out to determine the magnitude of this effect. Although any set of assumptions could have applied in this simulation, a set was chosen which leads to a relatively high impact on the industry, i.e., it phases out the present type engines very quickly. (It should be emphasized that the Committee does not consider such a drastic change to be probable.) The following are the assumptions used:
1. The modified carbureted piston engine equipped with an oxidation catalyst will be produced only in model year 1975 and no modified carbureted piston engine using catalytic control of emissions will be produced in model year 1976.
2. The Wankel engine will be introduced initially in the small cars (subcompact and compact) and subsequently will be developed in higher-horsepower versions for larger vehicles.
3. Diesel engines (4 and 6 cylinders) will be introduced for fleet-car usage by 1975. A V-B Diesel will be introduced subsequently.
4. A limited number of stratified-charge engines (3-valve) will be introduced in 4- and 6-cylinder versions for small cars.
5. Gasoline engines with electronic fuel injection will be introduced by 1976 in 4-, 6-, and 8-cylinder versions in very large quantities.
Applying these hypothetical assumptions to the simulation model, the capital-investment impact on manufacturing facilities was then developed, as a sticker price increase. In the model, the aggregate American production was considered without identifying the specific producer. The car configurations were detailed down to the major components and subassemblies. These units were then scheduled in production in the proper sequence and at the proper time to yield the desired schedule using standard industry lead times. These numbers were developed giving due consideration to expected product life and normal industry amortization practices.
The expected sticker-price increases to return the capital investment in new production lines, old production-line tear-up, assembly-line change, and new facilities were found to range (even with such a drastic change in engines and control systems in such a short time) from $8 to $150 per car.
ALTERNATIVE SYSTEMS FOR LOW-EMISSION AUTOMOBILES
The Committee also considered power systems other than Otto-cycle gasoline engines. It became apparent quite early in this study that no alternative power system could be produced in sufficient numbers by 1975 or 1976 to displace an appreciable part of present engine-production quantities.
Several power systems (e.g., Rankine, Stirling, batteries, fuel cells) show promise for eventually meeting 1976 standards, but development time and cost reduction are necessary before these can become competitive. Two engines (diesel and gas turbine) show promise of meeting 1975 emission standards. However, even though such engines have already been adapted to passenger cars, little development is being done on them for 1975 and 1976 because they are costly and have other detractive characteristics. The present diesel is heavy, tends to smoke, and its exhaust is odorous. The gas turbine has poor fuel economy at part load, and the NOx emissions are not presently controllable to low enough levels.
Although it is unlikely that any alternative engine will be in appreciable mass production by 1975 or 1976, some of them will be phased in within the next decade. Thus, summaries of the findings concerning the various systems are given below.
Diesel engines
Recent data show that several current four-stroke, and one two-stroke, diesel engines can meet 1975 standards for carbon monoxide and unburned hydrocarbons. A typical NOx value for a current Mercedes Benz 220D under the CVS-CH test is 1.65 g/ mile. There have been no results obtained on diesel engines showing ability to meet the 1976 NOx standard of 0.4 g/mile. Daimler-Benz estimates that the lowest NOx levels achievable for diesels at the present state of the art would be about 0.8 g/mile.
New developments in diesel engines, such as a two-stroke engine with a new, low-emission combustion method, and the use of positive-displacement rotary prime movers, such as the Wankel-engine configuration, offer the future possibility of meeting, or nearly meeting, 1976 standards with an engine that is smaller and cheaper than the present (1970) gasoline engine.
Much work must still be done to prepare even suitable prototypes of these concepts.
There is a good possibility that a diesel engine of sufficient power density, light enough weight, and emissions nearly satisfactory for 1976 automobiles can be built. But much engineering work must still be done before there can be a proven concept. Potential problems of smoke, white smoke, odor, and noise still remain. It appears that good single prototypes of the advanced engine will not be available before 1975. Limited production might be possible by 1980.
A passenger-car diesel engine designed according to existing technology may have a possible disadvantage in slightly greater weight and larger size over a spark-ignition engine of comparable output. It may cost more basically, but the difference shrinks when the emission controls for gasoline engines are added in, since the add-ons for diesels to meet 1975 standards are minimal.
It will give better fuel economy and require less maintenance, which should quickly make
up any first-cost difference. The efficient diesel will tolerate a wide range of fuels and becomes of greater interest as our concerns with energy conservation increase. Because fuel of lower volatility is used, diesel engines have an additional safety factor, and also there would be less fuel-vapor emissions at the filling station.
Gas turbines
Gas turbines are a feasible method of propulsion for standard-size U.S. passenger cars. In prototype form, they have demonstrated acceptable or superior weight, size, fuel consumption, driveability, maintainability, resistance to abuse and neglect, and safety. Carbon monoxide and hydrocarbon emissions are below the 1976 standards; NOx emissions are presently above the 1976 limits, but several approaches have shown that it is technically feasible to lower NOx to 1976 requirements especially for low-pressure-ratio engines. The concepts can probably be incorporated in a prototype by 1976. The added controls or costs of reaching 1976 NOx standards are not yet known.
Gas turbines to date have all shown poor fuel consumption at low design power and while operating at low fractions of the design power. Highly regenerated units tend to limit the effect, but the possibility of economic gas turbines having design power below 150 horsepower and operating under lightly loaded conditions is still a controversial matter.
The retail costs of future gas turbines installed in automobiles are highly uncertain. Estimates made by various highly qualified individuals or organizations run from a price below that of the cleaned-up spark-ignition engine to one three or four times higher. These estimates are based on the use of materials similar to those in today's engines.
Future possibilities for gas turbines improve as the use of ceramics for many parts is proven. If ceramics become widely available for the hot parts of gas turbines, it is generally agreed that the engines would eventually cost less than the spark-ignition alternative. In addition, the employment of critical resources would be greatly reduced.
A realistic schedule for advanced gas turbines to be produced in quantity would be for advanced limited-production engines by 1982, followed by mass production by 1984.
Stirling engines
At the present state of development, Stirling engines are very efficient engines that could allow high-performance full-size automobiles to meet the 1976 emission standards. Any form of heat energy or fuel source can be used to operate it. The engineering problems that remain to be solved before it would be possible to adopt them as practical engines for limited application relate to the reliability of sealing the working fluid inside the engine, to the cost and reliability of the heater assembly, and to the development of a simple, versatile power output control system.
Considerably more engineering is necessary to allow the engine to be considered as an entirely suitable automobile power plant. Additional developments necessary to make this possible relate to cost, operation in the hands of the customer, and integration into the automobile. The two sets of problems are best attacked simultaneously and may involve changes in the present form of the engine.
The potential of the engine goes well beyond its present state. Size, weight, producibility, safety, response to abuse and neglect, starting ease, driveability and versatility, control ease, fuel economy, noise, emissions, and cost potential all show indications of being competitive with or better than diesels in the present generation of development, and equal to or better than gasoline engines in the next generation of development. Thus, the engine could fit into the auto industry, truck industry, and other segments f the transportation industry, independent of the eventual outcome of the energy crisis or the fuel controversy. Approximately 4 to 10 years of additional development will be reduced to solve the outstanding engineering problems and produce a prototype advanced Stirling engine suitable for present-type automobiles.
Electrically Driven Vehicles
Electrically driven vehicles in principle provide freedom from pollution and are characterized by high energy efficiency, flexibility of performance, good durability, and low maintenance requirements. At present, the limiting factor relating to the technical and economic feasibility of electric vehicles is the vehicular power source. Electric drive systems (motor and controls) having excellent characteristics have been demonstrated; development of optimal drive systems is not considered to be limiting in the ultimate realization of electric automobiles.
Fuel-cell-powered electric vehicles in which the free energy of fossil fuels is directly converted into electrical energy for motive power do not emit CO or NOx; unused hydrocarbons can be easily removed from the exhaust. Fuel cells are not heat engines and are not subject to the Carnot limitation. For this reason they may operate at very high energy-conversion efficiency, resulting in superior fuel economy.
Although some fuel-cell systems have been successfully deployed in space missions, these are not adaptable for applications where low cost is important. Current advanced developments directed toward stationary applications in commercial and consumer markets are in the field-test stage. These represent important cost reduction and performance improvements relative to the aerospace units. With further significant cost and performance improvements, vehicular applications in small quantities may become feasible within 10 to 15 years.
Vehicles that employ rechargeable batteries as a power source do not have emissions resulting from the combustion of fuels; the site of emission is transferred to central power stations where such emissions are understood to be more effectively controlled, and at a lower cost. Because of the high efficiency of batteries and of electric drives, the net fuel economy of such vehicles promises to be better than that of present automobiles. Furthermore, if we move toward an electric economy, batteries may assume a unique role in the transportation system.
In contrast to fuel cells, extensive experience exists with respect to the performance characteristics of at least one battery system – lead/acid. This battery is rugged, efficient, reliable, and can respond instanteously to large changes in load. Presently available special-purpose vehicles can provide ranges of up to 50 miles and modest acceleration marginally acceptable under urban driving conditions, at a high cost. Other currently available rechargeable batteries, such as zinc/silver-oxide and cadmium/nickel-oxide, while superior in some respects to the lead/ acid system, are inherently unsuitable for vehicular applications because of cost and/or limited availability of materials. Still other battery systems concurrently in various stages of development offer significant performance improvements, and may meet the cost and materials requirements for vehicular applications.
The zinc/nickel-oxide battery is expected to allow a vehicle design with acceptable acceleration and a range of about 80 miles between recharges.
The most promising of the advanced battery systems are sodium/sulfur and lithium/sulfur batteries, which operate at temperatures in the range 300-400ºC, and are maintained at operating temperature by their reject heat and appropriate thermal insulation. These batteries are expected to have specific energies of 100 watt-hours/pound and specific powers of 100-200 watts/pound, permitting the design and construction of electric vehicles with excellent acceleration capabilities and a range of about 200 miles between recharges. About 7 or 8 years of optimum effort will probably be required for the development of pilot quantities of these batteries for vehicle test purposes. Still other promising non-aqueous systems are in early stages of exploration.
Hybrid electric/heat-engine powerplants are claimed to enable reduction of the emission of air pollutants. The expected improvement in driveability by using the electric motor for power surges should allow the heat engine to operate cleanly and economically at one setting or with a slowly varying setting over a range. There are significant penalties in the areas of cost and complexity that must be overcome before the hybrid can be considered a viable contender. Even if the technical and economic criteria can be met, it is doubtful whether introduction of this new and complex power-plant scheme will represent any more than an interim solution with respect to pollution abatement and effective use of natural resources.
Rankine engines
Tests made on Rankine-engine components have shown that the 1976 standards could probably be met with Rankine-engine-powered, standard-size automobiles. Various approaches to the design indicate that Rankine engines can be made to fit into full-size automobiles. These findings are to be demonstrated with working units in real automobiles by 1975.
Engine noise promises to be low except for the condenser fans, which could be troublesome due to large air-flow requirements. Starting should be easy, although time-consuming (one minute being a practical estimate). The driveability of Rankine-powered automobiles should be satisfactory if a sufficiently high power-to-weight ratio can be achieved.
One full-size automobile has been fitted with a 150-horsepower steam engine. Emissions did not meet 1976 standards and there were other detracting features, which can be traced partly to the underdeveloped nature of the engine. Lower-power steam engines have been fitted into compact-size automobiles and demonstrated. Low power density is a general characteristic of these engines, traceable to poor efficiency.
Newer forms of Rankine engines that use organic fluids flowing through either reciprocating or turbine machinery offer the possibility of trouble-free operation (no freezing, easy starting) at the expense of poorer fuel economy as compared with steam. These units will be larger and more difficult to integrate than will steam engines.
The Rankine cycle in any version will tend to have relatively uniform specific fuel consumption over the operating range. This leads to reasonable fuel economy (but less than that of gasoline- powered automobiles of similar size) over typical driving schedules when steam, or the best organic-fluid, engines are considered.
To achieve an engine with reasonable fuel economy, the controls have to be complex and the engine has to be as large as possible within that allowable envelope. Thus, any Rankine engine will be pushed to the allowable limits on size, weight, and cost for a given application, and the automobile will be considerably underpowered and overpriced as compared with a gasoline engine in the same application. Despite its potentially good emissions, driveability, and low noise, most of the other realistic evaluation features for automobile engines (such as size, weight, cost, fuel economy, and starting time) are missed by the Rankine engine, independent of type.
It is problematic whether even limited production of full-power engines could be feasible before 1980. Limited production of existing designs for low-power applications could begin by 1976-77.
Major questions remain to be answered affirmatively with respect to safety, operability, reliability, and overall driving versatility in the hands of the public. Unit cost and the ability to be phased into production present even larger questions for which affirmative answers are lacking.
A suitable full-size, prototype Rankine engine will not be available until 1975 (EPA schedule). Development f a manufacturable prototype must follow this by several years, which must in turn be followed by normal development.
Other Engines
A wide variety of other engines with some potential advantage over the gasoline engine or diesel engine have been considered over the years. Most of these have not been developed even as far as the automobile gas turbine, Rankine engine, or Stirling engines. None of them seem to offer a clear-cut advantage in emissions over the other types, and they all offer some increase in complexity, weight, volume, and probably cost.
Systems using positive displacement machinery but with combustion taking place outside the cylinder (out-of-cylinder combustion systems) have been studied for engines operating on the diesel cycle, the Otto cycle, the Brayton cycle, and many variations. They all suffer from lowered efficiency, larger size, and probable high NOx values. None of these systems appear to offer any basic advantage that cannot be achieved ultimately by diesels, gas turbines, and Stirling engines, all of which show promise of lower cost.
DISCUSSION
Introduction
As a result of the Clean Air Amendments of 1970, automotive and related manufacturers – both within and outside the United States – have embarked upon research, development, and manufacturing programs designed to meet the newly established emission standards for light-duty motor vehicles. As observed in the January 1972 report of this Committee, it is unfortunate that the automobile industry did not seriously undertake such a program on its own volition until it was subjected to governmental pressure. A relatively modest investment, over the past decade, in developmental programs related to emission control could have precluded the crisis that now prevails in the industry and the nation. The current crash programs of the major manufacturers have turned out to be expensive and, in retrospect, not well planned.
Nevertheless, the almost world-wide effort to achieve the federal emission standards set for the light-duty motor vehicles in the United States has produced a significant rate of progress toward meeting the requirements of the Clean Air Amendments of 1970. It is the very pace of that progress that complicates judgment today concerning the most appropriate course of action with respect to attainment of the standards required by that law.
As discussed in earlier parts of this report, several systems have been shown capable of attaining emission certification in 1975 model year cars. Among these are the diesel (discussed in Section 6.1) and the three systems discussed in Section 3.2 (the conventional engine with modification and oxidation catalyst, the Wankel with a thermal reactor, and the carbureted three-valve stratified-charge engine). While continued progress can be expected in development of all these systems, they do not possess equally desirable characteristics.
Several control systems in early states of development have met the 1976 standards at low mileage. Some of these represent further development of systems designed for certification and manufacture in model year 1975. Others are relatively new and their ultimate manufacture will require energetic commitment by the industry to further develop approaches that have been pursued only in smaller companies and at relatively low levels of effort. One system promises to be acceptable in use for the full 50,000 miles. Durability and other performance data are already available for that system. The future performance and acceptability of other systems – especially those currently being developed by the principal manufacturers – remain in doubt. In the following discussion, we shall briefly compare those systems that warrant consideration for certification and production in model year 1976.
Dual-Catalyst System
To date, the belated research and development programs of the major automobile manufacturers have been devoted almost entirely to the development and incorporation of such minimal modifications to the basic spark-ignition, internal-combustion engine as may be required to achieve certification in 1975 and 1976. This situation is a result of the short time between passage of the Act and the scheduled date of its enforcement, and the desire of the manufacturers both to protect their investments in the internal combustion engine and to utilize their vast experience with this engine. The modifications made to achieve emission levels required by the 1973 federal standards represent just such continued development of the conventional engines of previous years.
To achieve the further reductions called for by the 1975 and 1976 standards, most major manufacturers currently plan to use catalysts in the exhaust stream to promote both oxidation of carbon monoxide and hydrocarbons and chemical reduction of NOx. The CMVE believes that engines equipped with oxidation catalysts will be able to meet the certification requirements for model year 1975. At this time, no experimental engine modified to include the dual-catalyst system has exhibited the durability required to achieve compliance with the 1976 standards.
Nevertheless, assuming a continuation of the intensity of the current effort, extrapolation of the rate of recent progress suggests that catalysts with the durability required by the 1976 standards will be developed. But it cannot be stated with certainty that such developments will occur in time for 1976 production of automobiles.
Although American manufacturers and others evidently will be able to produce catalyst-equipped vehicles capable of certification for the 1975 model year, and even possibly capable of 1976 certification, compliance with the certification procedure, of itself, may not constitute indication of satisfactory performance of catalyst-equipped vehicles in actual customer use. As discussed in Section 4.2, the diverse conditions to be undergone by the engine and control systems during 50,000 miles of customer use are far more strenuous than those undergone during certification.
These more strenuous conditions may result in significant damage to a catalyst. In view of the performance history of catalytic systems observed to date on experimental vehicles, under laboratory conditions, there is concern that there may be frequent catalyst failure under conditions of actual use well before a scheduled 25,000mile replacement.
Admittedly, there has not been actual customer-like experience with catalytic systems that have met the 1975 or 1976 certification requirements, and these concerns may be overdrawn.
Furthermore, failure in service of cars properly maintained and used will call into operation Section 207(c) of the Act, by which the manufacturer can be forced by EPA to remedy the deficiency at his own expense. Obviously, this concern would be relieved by either the expected early development of catalysts demonstrably more rugged and durable than those tested to date.
or by demonstrably satisfactory performance, in conditions similar to customer use, of those catalysts now under investigation. Only one manufacturer has commenced such tests with a few cars equipped with a single-catalyst system that have met with 1975 standards. Final judgment of the actual performance of such systems must await experience.
Alternatives to the Dual-Catalyst Approach
In view of the fact that the dual-catalyst approach to a non-polluting automobile power plant may not lead to a truly satisfactory long-term solution to the environmental problem, it is encouraging to note that promising alternative systems are under intensive investigation. Although some are only in the earliest stages of development, others are more advanced and promise to achieve 1975 emissions certification when utilized on smaller engines. These include the carbureted three-valve stratified-charge engine, the modified diesel, and the Wankel with thermal reactor. Each of these alternative systems is described below.
Carbureted three-valve stratified-charge engine
Prototype compact cars equipped with the carbureted three-valve stratified-charge engine have met the 1975 standards for 50,000 miles. Three tests on vehicles equipped with an advanced version f this system show average low-mileage emissions of 0.25 grams per mile HC, 2.50 grams per mile CO, and 0.43 grams per mile NOx (see Table 3-11). This system should be capable of certification on small cars in time for model year 1976 production, and with adequate margin of safety for each of the three contaminants. This approach should also be applicable to larger engines, but sufficient experience is not yet available for evaluation.
A substantial degree of confidence can be placed in the estimation that the emissions performance of this engine in use will be quite close to its performance during certification. The maintenance required on the carbureted stratified-charge engine should be no greater than that required on a conventional 1973 engine. In fuel economy, this engine is comparable with a 1972 engine and much superior to a dual-catalyst-equipped 1976 engine.
Diesel engines
Emissions achieved by a current diesel powered vehicle are 0.15. 2.5, and 1.65 g/mile for HC, CO, and NOx, respectively, and this engine is certifiable for 1975 production. Further improvements are possible, but much innovative engineering work must still be done before the diesel can meet the 1976 standards. Limited production of adequately improved vehicles might be possible by 1980. Since the diesel would provide a significant fuel economy, even compared with 1972 engines, further development of the diesel warrants encouragement.
Wankel engines
As shown in Table 3-7, the Wankel engine with thermal reactor on a compact car has met the 1975 standards with NOx levels of about 1 g/mile for 50,000 miles, but with a fuel penalty of about 30 percent compared with a 1973 equivalent piston engine. The use of EGR and richer carburetion can probably further reduce NOx levels, but at the cost of even greater fuel consumption, and even so it is not yet certain that the 1976 standard for NOx can be achieved.
Durability performance of the Wankel engine with thermal reactor on a compact car has been shown to be superior to that of the dual-catalyst system. However, temperatures experienced by the reactor during operation in the hands of the public should be somewhat higher during certain driving modes, and durability under such conditions has not been established.
Catalytic systems with feedback control
A system with three-way catalyst and feedback control (see Section 3.6) promises improvement over the dual-catalyst system. However, adequate durability data with respect to both the catalyst and the oxygen sensor are not available to make meaningful estimates of the performance of such systems either during certification or in use.
Feedback control of a dual-catalyst system would be expected to increase the life of the catalyst, reduce emissions, and significantly improve fuel economy. At this writing, such a system is not available but may be capable of development, though perhaps not in time for production in quantity in 1976.
Interim standards
According to the work statement agreed to by the EPA and the National Academy of Sciences, "Should the Contractor conclude that the attainment of emission standards on the schedule provided by Section 202(b) (1) f the Clean Air Act is not technologically feasible, the Contractor shall specifically determine technologically feasible interim emission levels to assist the Administrator in exercising his responsibilities under Section 202(b) (5) of the Act."
However, the considerations that must enter into the determination of optimal technologically feasible interim standards are so complex and carry so many implications that, as explained below, it is inadvisable and inappropriate for this Committee to recommend a specific set of interim levels at this time.
It is not yet possible to make a definitive prediction with respect to which engine systems will achieve certification for 1976. The most likely candidate is the carbureted stratified-charge system on smaller engines. It is probable that others, particularly the dual-catalyst system, will also qualify at that time. It is conceivable that the projected automobile production for 1976 can be achieved only by a mix of engines, some certifiable and some (probably larger engines) not quite certifiable. However, while it is premature to judge the issue at this time, a rationale may later be required for upward adjustment of one or more f the standards to permit production of a sufficient number of vehicles of various sizes in 1976.
Examination of possible interim standards for the three pollutants is complicated by the fact that the technologically feasible levels of the three pollutants are interdependent. For several of the systems discussed, further decreases in NOx can be achieved, for example by greater reliance upon EGR, but only by accepting higher levels of CO and HC. Thus, before selection of a particular set of interim levels as achievable, answers will be required to such questions as: Is it more important to reduce NOx emissions than CO or HC? Or vice versa? Further, compact cars are capable of lower emissions than are standard or large cars with similar control systems, while consuming less fuel. What emphasis should be placed on significantly different levels of fuel consumption that are associated with the various control systems and vehicle sizes and the substantial possible impact on total petroleum requirements?
The Committee made no attempt to resolve these and related questions, as judgments regarding these matters were deemed to be beyond the scope of the study commissioned to the Academy and delineated by the EPA-Academy contract. Thus, at this time, the Committee finds it inadvisable to recommend a specific set of interim standards.
Effects of a Delay in Enforcement on Total Automobile Emissions
To illustrate the effects of various delays in implementing the emissions standards, should this be found necessary, a computer model was used to calculate total automotive emissions in a typical metropolitan area for the years from 1960 to 2000. This model accounted for factors such as vehicle age distribution among all automobiles, the decrease in vehicle miles driven per year per car as vehicle age increases, the predicted nationwide growth in vehicle population each year, the emission reduction achieved through crankcase blow-by and evaporative-loss control, the effect of federal exhaust-emission standards, and deterioration of emission controls with mileage.
Vehicle age distribution was taken from a national average automobile population, which is a reasonable distribution for many large urban areas. Urban driving was assumed in the model, and average emissions for urban driving were used. These emissions values were obtained from records for 1972 and older model-year cars. For cars built or to be built after the 1972 model year, the emissions values were based on various implementation plans.
Figures 7.1, 7.2, and 7.3 show the variations in emissions of HC, CO, and NOx, respectively; these curves are normalized against the maximum for each contaminant. Four cases are represented in each set of curves:
1. Standards maintained at the 1973 levels indefinitely.
2. 1975 and 1976 standards implemented and met on schedule.
3. 1975 and 1976 standards each delayed one year – the maximum allowable under the law.
4. 1973 standards maintained through 1976 model year and 1976 standards implemented in 1977 model year.
The implementation of emissions controls since 1968 has already caused an appreciable reduction in annual emissions f HC and CO, but little reduction in NOx. Federal standards for model year 1973 cars call for decreases of approximately 80 percent for hydrocarbons, 70 percent for carbon monoxide, and 50 percent for oxides of nitrogen, all measured in relation to the uncontrolled emissions of pre-1968 vehicles. As seen in the curves, were 1973 standards to remain in force, total emissions of hydrocarbons and carbon monoxide would continue to decline for some years, as would that of NOx. Preponderantly, these effects reflect the removal from service of older uncontrolled, or less-well-controlled automobiles.
Implementation of 1975 and 1976 Standards and Related Matters
Of two promising candidates for certification and production in 1975 and 1976 – the dual- catalyst system and the carbureted stratified-charge engine – only the former is planned for manufacture on a scale commensurate with expected requirements in those years. Even if durable catalysts became available, the dual-catalyst system would still have several undesirable characteristics, the more important of which are listed below.
1. The dual-catalyst system is expected to have poor fuel economy. Improvements in fuel economy could be obtained by the use of proper feedback control mechanisms, but these are unlikely to become available for production in 1975 or even 1976.
2. Dual-catalyst systems will have a higher initial cost, be more difficult to maintain, and be less durable.
3. Manufacture of vehicles equipped with single- or dual-catalyst systems in large numbers before sufficient experience with these devices under actual diverse consumer use is precarious.
Independent of whether each car must periodically pass inspection or whether the manufacturer is repeatedly compelled to exercise the recall provisions of the Act; if a large fraction of all cars markedly exceeds the emission standards, the entire rationale of this procedure becomes suspect.
4. The 1973 class vehicles when converted to 1975-76 systems can be expected to be more difficult to start, thus wasting some fuel and increasing emission of pollutants (although it should be possible to mitigate this situation by future technical improvements)
The circumstances recounted above – the probable certifiability of the carbureted stratified- charge engine under both 1975 and 1976 standards but its relatively limited planned production, particularly in 1975, and the considerable promise of other, as yet incompletely developed systems – make judgment concerning an optimal national approach to decision concerning the scheduled implementation of the 1975/1976 standards extraordinarily complex – precisely because the entire research and development aspect of this situation is very much in flux and changing rapidly.
Some members of CMVE are concerned that strict enforcement of the provisions of the Act might, by forcing adoption of the control system first to be developed and certified, defeat the goal of the earliest possible attainment of compliance by the most generally desirable means.
These members of CMVE believe that, once having embarked upon large-scale production of the catalyst-dependent control systems, several years would elapse before major manufacturers would alter course in favor of producing more generally satisfactory vehicles. This would happen, it is thought, because it would be consistent with the tradition of the industry of slowly improving technology already in use rather than switch to a significantly new and different technology not yet tried on a mass scale. Further, there is concern that existing market mechanisms would not suffice to accelerate conversion to a substantially different technology at a pace consistent with the overall national interest.
A minority view within the CMVE states that: (a) only rigorous enforcement of the Act will assure the pace of continued progress toward the goals of the Act; (b) by the time 1975 cars are placed in production, the catalysts used in catalyst-dependent systems may prove decidedly more reliable than are those now available; (c) there is no assurance that the additional development time would not simply be employed by the major manufacturers for further development of the present systems; and (d) the presence on the market of even a small number of alternative control systems that are more reliable, cheaper, and accompanied by a lesser fuel penalty, if any, would constitute an effective market device, which, without other intervention, would assure changeover by the major manufacturers at an acceptable pace, particularly if the recall provisions of the Act are enforced as warranted.
The majority view of CMVE suggests that, on balance, it may be prudent for EPA to consider a delay in the imposition of 1975 and 1976 standards, but no longer than that provided for in the Act. It is thought that this would provide the manufacturers an opportunity to consider and implement alternative and, quite possibly, more generally satisfactory technologies with which to attain the goals of the Act. In this view, as shown in Section 7.5, such an action would not result in an unacceptable deceleration in reduction of automotive emissions.
In its work, CMVE became aware of a continuing controversy concerning the stringency of existing emission standards. Strongly held differences of interests and views surround all the major factors that affect the selection of automotive emission standards: the health effects of individual pollutants, their relation to ambient concentrations, the relationship of total emissions to primary and secondary ambient pollutant levels, the contribution of automobile usage to total emissions, and the possible relative reductions in emissions from stationary and mobile sources.
Some of the issues posed by these considerations are resolvable only by further scientific research; all will require the attention of officials concerned with pollution control.
These matters are so complex and important that the Committee strongly urges an early and thorough reexamination by Congress, EPA, and the Academy of all aspects of motor vehicle pollution standards established in the Clean Air Amendments of 1970 – their premises, underlying assumptions, the goals that were set, and the interplay among the three pollutants dealt with specifically in the Act. In the light of the material developed in its study, CMVE believes that such a reexamination would be extremely valuable in relating motor vehicle emission control to the many issues relevant to a sound national environmental policy.