Appendix A

Emission and Energy Characteristics of Heavy-Duty Diesel-Powered Trucks and Buses

K. G. DULEEP

Energy and Environmental Analysis, Inc.

Heavy-duty diesel-powered trucks are major contributors to oxides of nitrogen (NOx) emissions and combustion-derived particulate emissions in many urban areas. This appendix provides a brief review of the energy use and emissions characteristics of heavy-duty diesel vehicles (HDDVs) and reviews the effects of expansions of highway capacity on emissions.

The structure of the HDDV fleet, which encompasses a wide range of vehicles [from 8,500 lb gross vehicle weight (GVW) to more than 80,000 lb GVW], is discussed. Data on sales, populations, and use of the HDDV fleet are presented.

Historical and future emissions regulations for HDDVs are reviewed. Since California has been the leader in new emission standards and inuse controls, particular attention is given to the California standards and the proposed low-emission truck standards. Fuel standards and in-use requirements are also discussed in detail.

The data that have been used to construct emission factors and speed correction factors for HDDVs are reviewed. In particular, U.S. Environmental Protection Agency (EPA) emission factors and speed



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



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

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

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use Appendix A Emission and Energy Characteristics of Heavy-Duty Diesel-Powered Trucks and Buses K. G. DULEEP Energy and Environmental Analysis, Inc. Heavy-duty diesel-powered trucks are major contributors to oxides of nitrogen (NOx) emissions and combustion-derived particulate emissions in many urban areas. This appendix provides a brief review of the energy use and emissions characteristics of heavy-duty diesel vehicles (HDDVs) and reviews the effects of expansions of highway capacity on emissions. The structure of the HDDV fleet, which encompasses a wide range of vehicles [from 8,500 lb gross vehicle weight (GVW) to more than 80,000 lb GVW], is discussed. Data on sales, populations, and use of the HDDV fleet are presented. Historical and future emissions regulations for HDDVs are reviewed. Since California has been the leader in new emission standards and inuse controls, particular attention is given to the California standards and the proposed low-emission truck standards. Fuel standards and in-use requirements are also discussed in detail. The data that have been used to construct emission factors and speed correction factors for HDDVs are reviewed. In particular, U.S. Environmental Protection Agency (EPA) emission factors and speed

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use correction factors are contrasted with the findings on these issues from other data or engineering analyses. HDDV fuel economy data are reviewed, with emphasis on average fuel economy derived from surveys. Data on the change of fuel economy with speed derived from simulation models or on-road tests are presented. These data and their relationship to the conversion factor used to convert emissions expressed in units of work to the more familiar units of grams per mile are explored. Finally, the findings are summarized in the context of the National Research Council's project goals of estimating the effects of expansions of highway capacity. FLEET CHARACTERIZATION Truck Classification The term “heavy-duty vehicles” as defined by EPA covers trucks and buses ranging in weight from 8,501 to 80,000 lb GVW. The GVW is the total weight of vehicle with its maximum payload and is sometimes referred to as gross combination weight (GCW) for truck and trailer combinations. Trucks exceeding 80,000 lb GVW are not allowed on the Interstate highway system, although individual states permit their operation on highways. Their use in off-highway applications such as mining or logging is common. This appendix addresses only those vehicles certified for on-highway use between 14,000 and 80,000 lb GVW, largely because the 8,500- to 14,000-lb GVW class includes vehicles more similar to light-duty trucks. These trucks and buses are also classified more commonly on the basis of a system used by industry that divides the fleet into eight classes. Classes I and II cover the 0- to 10,000-lb GVW range and include all the light-duty pickup, van, and utility vehicles that are used for personal transportation as well as in light commercial applications. Class III covers the 10,000- to 14,000-lb range and incorporates the “heavy” version of pickup trucks and vans that are used in delivery service or as motor homes. This class also incorporates some imported delivery trucks manufactured by companies such as Iveco and Isuzu (Iveco no longer sells trucks in the United States).

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use Class IV spans the 14,000- to 16,000-lb GVW range, and Class V spans the 16,000- to 19,500-lb GVW range. In the past, few trucks have been sold in these classes. Class IV trucks have generally been imports that are similar to the Class III imports. Class V trucks are generally the lightest versions of the Class VII trucks and are typically powered by gasoline engines. Sales in these classes account for less than 5 percent of all trucks exceeding 14,000 lb GVW. However, Ford introduced a version of its pickup truck rated at slightly more than 14,000 lb GVW in 1989–1990, which served to inflate the sales totals for this class in 1990, although the truck is really closer to a Class III truck. Classes VI and VII span the 19,500- to 26,000-lb GVW and the 26,000-to 33,000-lb GVW ranges, respectively, and include most trucks that are referred to as medium-duty trucks. Over the last 15 years, the use of diesel engines in these trucks has increased. The extra weight of the diesel engine and related drivetrain components has caused a sales shift from Class VI to Class VII during the last decade, in parallel with increasing use of diesel engines in this market. Table A-1 gives the changing sales composition over the period from 1980 to 1990. Many import models from German and Japanese manufacturers are available in these two segments, and import penetration ranges from 10 to 15 percent. Class VIII trucks span the range from 33,000 to 80,000 lb GVW and are sometimes incorrectly referred to as heavy-heavy duty trucks. In fact, this class is better represented as two classes, VIIIA and VIIIB. Class VIIIA spans the 33,000- to 55,000/60,000-lb range and includes trucks referred to as super-mediums and trucks used in rough applications such as construction, mining, gravel and concrete delivery, garbage hauling, and so forth. Most of these applications involve local or short-haul use. Retail sales data do not generally provide the distinction between trucks in Classes VIIIA and VIIIB. Only a few import models are sold as Class VIIIA trucks, but there are no imports in Class VIIIB. Class VIIIB trucks, which cover the 60,000- to 80,000-lb range, are primarily for intercity or long-haul use. These trucks are correctly referred to as heavy-heavy. Table A-2 gives the factory sales by class for the United States in 1980, 1985, and 1990 and breaks out the Class VIII trucks into the two subclasses. Class VIIIB truck sales are the single largest segment of total sales and are approximately equal to half

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use TABLE A-1 U.S. Retail Sales by Class, All Trucks (MVMA 1981, 1986, 1991) GVW CLASS 1980 1985 1990 PERCENT NUMBER PERCENT NUMBER PERCENT NUMBER PERCENT IMPORT PENETRATION IV ~0 195 0 0 9.9 27,453a 4.6 V 0.9 2,309 1.8 5,081 1.8 5,055 66.6 VI 33.5 89,764 17.0 48,358 13.8 38,209 14.2 VII 21.8 58,434 34.2 96,973 30.8 85,343 10.6 VIII 43.8 117,270 47.0 133,581 43.7 121,324 0.15 Total 100 267,972 100 283,993 100 277,384 6.93 NOTE: Data on import penetration are not available for 1980 and 1985. GVW = gross vehicle weight. a   Includes a large number of Ford Super Duty Pickup Trucks.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use TABLE A-2 U.S. Factory Sales and Dieselization (Excludes Imports) (MVMA 1981, 1986, 1991)   1980 1985 1990 GVW CLASS NUMBER PERCENT DIESEL NUMBER PERCENT DIESEL NUMBER PERCENT DIESEL IV 24 0 0 0 21,512 64.50 V 1,859 0 5,123 0 1,644 2.95 VI 51,171 21.94 27,419 59.38 16,288 71.05 VI Bus 26,084 5.90 20,530 44.26 17,853 65.17 VII 54,363 65.45 82,225 63.84 66,990 81.55 VII Bus 6,285a 79.70 8,756a 99.10 11,563 99.85 VIIIA 11,396 74.60 8,962 93.20 10,739 99.70 VIIIB 92,493 100 120,342 100 109,848 100 Total 243,675   273,357   256,437   NOTE: GVW = gross vehicle weight. a Approximately 35 percent classified as Class VIII. of total sales. Table A-2 also indicates the increased diesel penetration in the medium-duty classes. The data in Table A-2 exclude imports, all of which are diesel. Buses are generally classified into two types: school buses and heavy-duty transit buses. School buses are generally in the 18,000- to 21,000-lb GVW category (Class V or VI), whereas transit buses are in the 33,000-to 35,000-lb GVW range (Class VIII). Typically, school buses are used very little, whereas transit buses and intercity buses are used intensively. Before 1979, most school buses were gasoline powered, but the use of diesel engines in school buses increased in the 1980s. Transit and intercity buses have been powered mostly by diesel engines for nearly two decades. Engine Types Unlike car and light truck manufacturers, heavy-duty truck manufacturers do not necessarily manufacture their own engines. Even the ones who manufacture engines allow users to specify alternative

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use engine makes in their trucks in many cases. Engines are classified into three types: light-heavy, medium-heavy, and heavy-heavy. Manufacturers compete in the market for each segment. Table A-3 gives the U.S. factory sales by engine manufacturer in 1990, excluding imports. Data on imported engines and in domestic trucks are included in this table. The light-heavy engine market is dominated by three models: the Navistar 7.3-L, the GM 6.2-L, and the Cummins B series engine. These engines are characterized by their relatively high rated speed of 3,000 to 3,600 revolutions per minute (RPM) and their light weight (750 to 900 lb). EPA requires that their emissions be certified to a useful life of 110,000 mi. Until 1986 the medium-duty diesel engine market was dominated by the Navistar DT-360 and DT-466, the GM 8.2-L, and the Caterpillar 3208. Ford introduced its Brazilian 6.6/7.8-L engines in the late 1980s, Caterpillar has replaced the 3208 with the 3116, and GM has discontinued the 8.2-L as of 1991. EPA rates the useful life of these engines at 185,000 mi. The heavy-heavy duty engine market for heavy-heavy duty trucks (Class VIIIB) also evolved in the 1980s. Engines are classified into those below 300 HP and those above 300 HP. Four manufacturers (Cummins, Caterpillar, Detroit Diesel, and Mack) have essentially shared these two markets. EPA rates the useful life of these engines at 285,000 mi. Most engine models in the medium and heavy segments are offered with a variety of horsepower and RPM ratings. Typically, horsepower is changed by recalibrating the fuel system and restricting or increasing maximum fuel flow and by changing the governed speed or rated RPM. (These changes induce changes in the turbocharger size, valve timing, and injector size.) Most medium-duty engines are rated in the 2,400-to 2,600-RPM range, although ratings as low as 2,100 RPM are offered. Most heavy-duty engines are rated in the 1,800- to 2,100-RPM range, although ratings as low as 1,600 RPM are offered. In general, lower-rated RPM engines result in lower maximum truck speeds, and, as a result, low-RPM engines are less popular with truck drivers. Diesel Truck Population, Use, and Scrappage Details on truck use and scrappage are derived primarily from survey data. The most comprehensive survey is one conducted by the Bureau

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use TABLE A-3 Diesel Engines Used in Domestic Trucks, 1990 (MVMA 1991) ENGINE MANUFACTURER IV V VI VII VIIIA VIIIB BUS VI BUS VII AND BUS VIII Caterpillar 0 0 1,184 15,575 391 33,706 0 0 Cummins 0 51 294 0 343 55,854 4 326 DDC 0 0 181 214 5 15,207 5,143 834 Ford 5 0 2,998 18,997 5,158 0 2,646 0 Mack 0 0 0 0 0 16,842 0 0 Navistar 14,820 0 7,519 25,147 5,947 0 5,526 11,117 Import Enginea 0 0 536 1,074 102 572 0 0 Total 14,825 51 12,712 61,007 11,946 122,181 13,319 12,277 a Mostly Mercedes or Volvo.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use of the Census called the Truck Inventory and Use Survey (TIUS). The survey is conducted every 5 years; the most recent one (the 1987 TIUS) was conducted in early 1988. The data collected are from a sample of 104,000 trucks. Energy and Environmental Analysis, Inc. (EEA) has obtained and cleaned the 1987 TIUS data and has derived a useable sample of approximately 98,000 records. Of course, most of the trucks are light duty, since the sampling was generally according to the population distribution of trucks. Of the 98,000 trucks, 35,689 exceeded 14,000 lb GVW (that is, they were in Class IV or higher). Truck usage is defined in TIUS on the basis of radius of operation. Local use is use within 50 mi of home base, short haul is that between 50 and 200 mi, and long haul is that exceeding 200 mi. As indicated in Table A-4, most of the Class VI through VIIA trucks are used on short-haul or local operation, whereas only Class VIIIB trucks are used in long-haul operation. An analysis of diesel truck annual vehicle miles of travel (VMT) by vintage was also performed at the national level. The sample of diesel trucks in Classes IV and V is inadequate to provide statistically significant results. Table A-5 indicates that trucks in Classes VI, VII, and TABLE A-4 Percent of Trucks by Area of Operation by GVW Class and Fuel Type [Based on EEA Analysis of 1987 TIUS Data (Bureau of the Census 1987)]   AREA OF OPERATIONa GVW CLASS /ENGINE TYPE LOCAL SHORT HAUL LONG HAUL Class VI Gasoline 84.4 13.8 1.8 Class VI Diesel 66.1 27.8 6.1 Class VII Gasoline 86.2 12.5 1.3 Class VII Diesel 66.4 29.0 4.7 Class VIIIA Gasoline 86.3 11.9 1.8 Class VIIIA Diesel 63.4 24.4 12.1 Class VIIIB Diesel 25.4 32.9 41.7 NOTE: GVW = gross vehicle weight. a Local if greatest percentage of annual miles was accrued within 50 mi of home base. Short haul if greatest percentage of miles was accrued between 50 and 200 mi of home base. Long haul if greatest percentage of miles was accrued beyond 200 mi from home base.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use VIIIA have very similar annual VMT by vintage. Annual VMT appears to be 24,000 to 26,000 mi/year for the first 4 years with a steady decline thereafter for trucks in Classes VI and VII. The same trend is apparent for Class VIIIA trucks but at slightly higher annual VMT. The VMT of Class VIIIB trucks is substantially higher, with average VMT exceeding 80,000 mi/year for the first 4 years. In an informal survey of national fleet users, EEA found that many intensive users retain a truck for about 4 years before selling it, confirming the validity of the TIUS analysis. The data in Table A-5 are an average for trucks in all types of operation including local, short, and long haul; data on long-haul Class VIIIB trucks indicate annual VMT of about 100,000 mi/year for the first 4 years. Truck usage and scrappage are strongly interrelated. Although EPA has set useful life by engine type, engine manufacturers and truck owners both confirm that well-maintained on-road trucks not subject to severe duty cycles last substantially longer. Typically, heavy-heavy duty engines used in Class VIIIB trucks have a useful life of about 400,000 to 500,000 mi before rebuild. In long-haul use, such engines are rebuilt for the first time at the end of 4 or 5 years and undergo a second rebuild in many cases after another 300,000 to 350,000 mi. TABLE A-5 Annual Mileage by Vintage, Diesel Trucks Only [Based on EEA Analysis of 1987 TIUS Data (Bureau of the Census 1987)] VINTAGE (YEARS) VI VII VIIIA VIIIB 1a 24,049 22,183 29,819 77,957 2 23,681 26,462 33,302 86,684 3 22,308 36,186 34,974 84,475 4 24,392 26,850 39,080 82,722 5 22,541 21,220 32,937 76,912 6 19,216 21,034 28,658 66,270 7 26,082 20,051 26,503 60,169 8 20,219 20,314 27,092 55,015 9 18,028 19,478 26,238 50,600 10 16,261 17,222 24,909 45,136 11 and older 8,434 11,749 14,396 31,232 a Includes trucks with less than 1 year of service.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use Thus, the total lifetime mileage can be 1 million to 1.2 million mi with two rebuilds, and engines and trucks can last up to 15 years. However, in many “rough use” applications such as construction and mining, the useful life can be considerably shorter. Medium-duty engines have a typical useful life (under normal use) of 230,000 to 250,000 mi. In general, these engines are rebuilt only once, after 8 to 9 years, when used in short- or long-haul operations. Light-heavy duty diesels are not rebuilt at all, and their use in Class IV and V trucks and school buses is generally in low annual VMT applications. Light-heavy duty diesels can last about 150,000 mi in normal service, corresponding to 12 to 15 years of use, on the average. Detailed data on registration by vintage for each of the three groups of heavy-duty trucks are not well defined because of the difficulty in classifying such trucks on the basis of available registration data. An analysis of California truck registrations by the California Air Resources Board (CARB) indicates the following approximate percentages as of 1990: AGE (YEARS) LIGHT AND MEDIUM-HEAVY HEAVY-HEAVY 5 or less 25.6 64.6 6 to 10 39.3 30.0 11 to 15 25.6 5.4 16 to 20 6.5 ~0 21 to 24 2.5 ~0 25 or more 0.6 ~0 These data indicate that virtually no heavy-heavy duty diesel trucks are older than 15 years and that about 3 percent of light/medium diesel trucks are older than 20 years. EEA believes that these data may not be correct, but the percentage of diesel trucks older than 20 years (i.e., pre-1974) that were not subject to any emission controls is small, probably less than 4 percent of all diesel trucks. The TIUS survey also provides data on the average operating weight of trucks and the percentage of miles operated without any load (because of empty backhaul, for instance). The average loaded weights are shown in Figure A-1 by truck class. Figure A-2 shows the empty operation fraction. Surprisingly, virtually all diesel trucks report empty VMT at close to 30 percent of total annual miles.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use FIGURE A-1 Average percentage of annual mileage when no load was carried, by gross vehicle weight class [based on EEA analysis of 1987 TIUS data (Bureau of the Census 1987)].

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use The data in TIUS do not permit a detailed disaggregation of the effect of weight or driving cycle on fuel economy because only a reported average over the previous year is available. TIUS also reports on average payload and “area of operation” (divided into local, short haul, or long haul, depending on where the greatest percentage of annual miles was accrued). Regression analysis of the data indicated that the ratio of average payload to maximum weight and the area of operation had significant influence on the fuel economy of medium-duty trucks, whereas horsepower and area of operation were most significant for Class VIIIB trucks (average payloads did not vary as much across the sample for Class VIIIB trucks). On the basis of these regressions, for medium-duty truck (Classes VI, VII, and VIIIA), predominantly local operation reduced fuel economy by 5 ± 1.5 percent compared with trucks in short-or long-haul operation, whereas a 10 percent change in operating weight resulted in a 2.2 ± 0.2 percent change in fuel economy. For a Class VIIIB truck, predominantly local operation causes a 2.5 ± 0.3 percent change in fuel consumption compared with trucks in predominantly long-haul use, whereas a 10 percent increase in horsepower results in a 1.1 ± 0.1 percent decrease in fuel economy. However, these broad averages may mask differences in truck types for those in local use and in average payload over the entire cycle of operation. Fuel Economy Dependence on Speed Few data are publicly available on fuel economy from tests of HDDVs conducted under controlled conditions on a chassis dynamometer. Data from on-road tests are principally at highway speeds. However, truck manufacturers have developed simulation models for fuel economy that are used as a marketing tool to educate their customers about the effect of truck component specification on fuel economy. In particular, Navistar's TCAPE model is well documented (1990), and its representatives claim that the model has been validated to within ±5 percent in field tests. The model has three built-in cycles to represent city, suburban, and highway driving conditions with average speeds of 20, 40, and 55 mph, respectively.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use Figure A-8 shows the change in fuel economy as computed by TCAPE for two trucks at the three speed cycles. The first is a Class VIIIA super medium truck of 48,000-lb GVW loaded to 40,000 lb. The second is a 78,000-1b Class VIIIB truck loaded to maximum weight. These cases are illustrative, because most medium-duty vehicles are rarely loaded to maximum weight. Figure A-8 shows that both trucks display relatively small fuel economy changes with speed between the 40 and 55 mph cycles, but fuel economy falls off in the city cycle. The Class VIIIB truck experiences a larger percentage loss in fuel economy in the city partly because of its higher payload and partly because Class VIIIB trucks are more highly optimized for highway use. At speeds higher than 55 mph, fuel economy drops, largely because of the rapid rise of aerodynamic drag. Studies conducted by the U.S. Department of Transportation (DOT) in the mid-1970s in response to the enactment of the 55-mph speed limit indicated the horsepower required to move a Class VIIIB truck at the following steady speeds (DOT 1974):   HORSEPOWER SPEED (MPH) ROLLING RESISTANCE AERODYNAMIC DRAG TOTAL 50 144 57 201 60 172 97 269 70 200 157 357 Truck engines of that era were typically in the 360- to 400-HP range to allow speeds of up to 70 mph in addition to powering accessory drive loads. Low-rolling-resistance radial tires and improved truck aerodynamics have reduced both loads by about 20 percent, so a modern truck would require only 285 HP to cruise at 70 mph; today's trucks are often powered by 300- to 325-HP engines. Using these data, it can be seen that driving at 70 mph requires 5.1 bhp-hr/mi,4 whereas driving at 50 mph requires only 4.02 bhp-hr/mi, so the speed increase from 50 to 70 mph increases power consumption by 27 to 28 percent. (The increase is similar for modern trucks.) Engine efficiency in terms of brake-specific fuel consumption also can

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use FIGURE A-8 Fuel economy of heavy-duty trucks as a function of driving cycle (Navistar, TCAPE outputs).

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use decrease with increased engine speed, so a net decrease in fuel economy of more than 30 percent can be expected for a speed increase from 50 to 70 mph. Table A-12 indicates the fuel economy decrease observed from actual trucks in the 50- to 65-mph range. Although there is considerable variation between trucks due to the choice of engines, axle ratios, and gear ratios, the fuel economy loss with speeds over 50 mph is directionally consistent on all trucks tested. Emission Conversion Factor The emission conversion factor provides a link between engine emissions per unit of work (g/bhp-hr) to vehicle emissions per mile. The conversion factor is based on the following identity: g/mi = g/bhp-hr × bhp-hr/mi The conversion factor, bhp-hr/mi, is the work required to move a vehicle 1 mi; obviously the larger and heavier the truck, the greater the work and the larger the conversion factor. The conversion factor can be further decomposed into the following: where bsfc is the engine brake-specific fuel consumption expressed in pounds of fuel consumed per brake horsepower-hour produced, fuel density is a constant, and MPG is the fuel economy of the truck. Detailed knowledge about MPG is available from the TIUS data, as presented earlier. The bsfc of HDD engines is available from the emissions test. Once these variables are known, the conversion factor can be obtained with good accuracy. Only one aspect of the estimating procedure could lead to inconsistencies: the fuel economy is an on-road average and is not tied to the specific driving cycle and speed of the base emission factor. The average speed of many medium-duty vehi-

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use TABLE A-12 Effect of Speed on Fuel Consumption Rates (DOT 1974)   FUEL CONSUMPTION RATE (MILES PER GALLON) FOR INDICATED SPEED (MPH) PERCENTAGE INCREASE IN FUEL CONSUMPTION CAUSED BY INDICATED INCREASE IN SPEED (MPH) VEHICLE NO. 50 55 60 65 50 TO 55 55 TO 60 60 TO 65 50 TO 60 50 TO 65 1 5.12 5.06 4.71 a 1.2 7.4 a 8.7 a 2 5.41 5.02 4.59 4.08 7.8 9.4 12.5 17.9 32.6 3 5.45 4.97 4.52 a 9.7 10.0 a 20.6 a 4 5.21 4.90 4.88 4.47 6.3 — 9.2 6.8 16.6 5 4.49 4.40 4.14 3.72 2.0 6.3 11.3 8.5 20.7 6 4.97 4.51 4.42 a 10.2 0.2 a 12.4 a NOTE: Each figure is the average of more than 20 timed runs on roadways with less than 1 percent grade. a The governor setting that controls fuel injection did not permit this vehicle to be operated at 65 mph.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use cles in urban and short-haul routes may, indeed, be close to 20 mph, making the conversion factor derivation approximately consistent with the base emission factor. The average fuel economy of the heavy-duty Class VIIIB truck is likely to be more heavily weighted to highway speeds, leading to some degree of inconsistency in the derived conversion factor. EPA has derived a composite conversion factor for all heavy-duty trucks that is a sales- and travel-weighted average for all diesel trucks exceeding 8,500 lb GVW. This average was found to be 2.60 in 1982. Projections of the conversion factor require the MPG and bsfc of future vehicles and engines. Engine efficiency improvements translate into bsfc improvements that are inversely proportional to MPG improvements. Hence, the term bsfc × MPG does not change with increased engine efficiency. Changes in the conversion factor depend only on the vehicle technology improvements that improve fuel economy, including Weight reduction, Aerodynamic improvements, Reduction in tire rolling resistance, Improved lubricants, Transmission and axle improvements, and Use of electronic speed control. EPA has developed a forecast of these technology improvements. On the basis of the expected changes in MPG, the conversion factor will decrease from 2.60 in 1982 to 2.30 in 1997, a 14 percent reduction. This translates into a 0.88 percent per year MPG improvement from non-engine-related improvements. The TIUS data indicated that HDDV fuel economy has increased between 1.2 and 1.5 percent per year depending on class. Given that engine bsfc has decreased by about 0.5 percent per year over the last decade, the EPA forecast appears reasonable, since the total forecast change is 0.88 + 0.5, or about 1.4 percent per year, consistent with the observed MPG change. Studies conducted by EEA for EPA suggest that annual improvements of 1.2 to 1.5 percent in fuel economy are likely to continue over the next decade, with the conversion factor also continuing to decrease at approximately 0.9 percent per year to 2000.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use In conclusion, the conversion factor estimates are well grounded in data, and the only source of error is the absence of a direct link of MPG and bsfc to vehicle speed. The total estimation error is not likely to be large (about 5 percent of the estimate), and the conversion factor is not a source of major concern in estimating HDDV emissions. Conversion factors for buses are included in the EPA weighted average but can easily be calculated separately. SUMMARY The discussion in the preceding four sections is the basis for the following conclusions. In addressing HDDVs, it is useful to divide transit buses and trucks into two weight classes. For heavy-duty trucks the categories are medium-heavy and heavy-heavy, and for buses the categories are transit buses and school buses. The annual use, operating radius, useful life, and fuel economy vary dramatically across these classes. Light-heavy duty diesels, which are more similar to light-duty vehicles, are not covered in this appendix. Unfortunately, EPA does not maintain these distinctions in modeling emissions from heavy-duty diesels, but aggregates emissions from all subcategories of diesels. These distinctions are probably important in the analysis of emissions associated with increases in urban highway capacity, since light-heavy and medium-heavy trucks are more likely to be used in urban areas than are heavy-heavy duty trucks. Emissions data on heavy-duty diesel engines suggest that emissions over a transient driving cycle can be correlated with emissions from a steady state engine “map” for VOC, CO, and NOx and even possibly for particulates. EPA and engine manufacturers differ in their assessment of the correlation for particulates. Although the development of “modal” models to calculate emissions from any arbitrary driving cycle has not been pursued, there are no significant technical obstacles to their development to estimate VOC, CO, and NOx emissions. The data also indicate good correspondence between emissions measured on an engine test and emissions measured from a vehicle test. The representativeness of an engine dynamometer test has been

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use questioned in some quarters, but EEA believes that the errors introduced by this type of test are not large. Part of the reason is that trucks and buses are so power limited that accelerations are usually conducted at wide-open throttle, so that errors in acceleration rates between engine and vehicle tests have no significant effect on engine load and emissions. EPA's approach to emission factors has been to group all subclasses of heavy-duty diesel trucks and buses into one class, making it difficult to model the differential deterioration rates in emissions between subclasses. EPA's approach to deriving heavy-duty diesel emission factors are simplistic and are based on assumed certification levels. In particular, EPA's assumption of zero deterioration rates cannot be justified. Hence, the EPA estimates of absolute emissions are almost certainly in error. Improved models, such as the one developed by Radian for CARB, appear to give more defensible estimates of emission factors for older trucks. Both the EPA and CARB estimates of VOC emissions from newer HDDVs are probably in error since they presuppose a relationship between VOC standards and certification levels, whereas actual VOC certification levels have declined to values far below standards as a result of technology improvements. Urban buses have received little attention, but limited data suggest that they are particularly high emitters. SCFs derived by EPA, on the other hand, appear to be reasonable up to 55 mph and conform to engineering expectations. At speeds above 55 mph, there are no data available on emissions for HDDVs, and engineering analysis must be used to gauge the correctness of the SCFs. The analysis indicates that NOx emissions at 70 mph are probably overstated by 50 to 60 percent, whereas VOC emissions at 70 mph are understated by nearly 100 percent. Although the SCFs were derived from limited data on older trucks, engineering analysis suggests the low speed factors are unlikely to change significantly even if newer data become available. The emission conversion factor to convert g/bhp-hr to g/mi is estimated from relatively good data and is not likely to be in significant error. (This conversion factor does not affect the SCF.) In the case of both factors, there are some issues that could change their absolute magnitude by 20 to 30 percent if the issues are resolved correctly. However, trends with speed up to 55 mph in the

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use case of the SCF or with time in the case of the conversion factor are not likely to be affected much. In summary, there are probably significant errors in the absolute magnitude of emissions predicted by current methods, but not in the trends with time or speed up to 55 mph. In general, average speed is a good surrogate for the drive cycle encountered for heavy-duty trucks simply because acceleration rates of trucks are so limited. As noted, virtually all accelerations of heavy-duty diesels are conducted at wide-open throttle, and the acceleration rate (an important determinant of emissions for light-duty vehicles) is not a significant source of error in this analysis. On the basis of EEA's analysis, it appears that emissions of all pollutants decline with speed up to 35 to 40 mph. Beyond that speed, NOx and VOC emissions increase, with VOC emissions increasing more sharply at speeds exceeding 50 mph. Anecdotal information from industry experts suggests that particulate emissions follow the same trend as VOC emissions up to about 50 mph, but the behavior at higher speeds is not well understood. These trends should provide some guidelines for most traffic conditions except those where high speeds (over 50 mph) are encountered for heavy-duty trucks. The lack of good particulate emissions data as a function of speed is a major drawback, since HDDVs are a major source of combustion particulate, and particulate emissions are known to increase the most of all regulated pollutants under in-use conditions. The use of the VOC SCF for estimating the change of particulate emissions with speed can be considered a very rough approximation, which may not hold good as the engine approaches the “smoke limit” at full load. More research in this topic could be fruitful. Fuel economy is a function of truck size, vintage, load, and speed. Good data are available on average truck fuel economy by weight class and by area of operation from surveys conducted by the Bureau of the Census. Fuel economy as a function of speed is dependent on truck design and load; light-heavy and medium-heavy vehicle fuel economy is less sensitive to speed change than over-the-road heavy-heavy truck fuel economy. Nevertheless, it appears that speed increases to about 50 mph have favorable effects on fuel economy for all vehicles. Beyond 50 mph, fuel economy declines sharply, falling by as much as 30 percent from 50 to 70 mph.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use NOTES 1. Most CNG/LPG engines are of the spark ignition type. 2. The method relied on engineering analysis to develop a model to link the dynamometer inertia weight and power absorption setting to cycle bhp-hr using field consumption data. 3. The percentages can be calculated from the deterioration rate and intercept in Table 4-1 as follows for VOC: 0.0185 * 28.5/0.732. 4. From the table at 70 mph, bhp-hr/mi REFERENCES ABBREVIATIONS CARB California Air Resources Board DOT U.S. Department of Transportation EEA Energy and Environmental Analysis, Inc. EPA Environmental Protection Agency MVMA Motor Vehicle Manufacturers Association Acurex Environmental Corp. 1992. Technical Feasibility of Reducing NOx and Particulate Emissions from Heavy Duty Engines. Draft Final Report. California Air Resources Board. Bureau of the Census. 1987. Truck Inventory and Use Survey. CARB. 1990. Technical Support Document: Proposed Roadside Smoke Test Procedures and Opacity Standards for Heavy Duty Diesel Trucks. June. CARB. 1993. Mobile Source Emission Standards Summary. June. Curran, T., T. Fitz-Simons, W. Freas, J. Hemby, D. Mintz, S. Nizich, B. Parzygnat, and M. Wayland. 1994. National Air Quality and Emissions Trends Report, 1993. 454-R-94-026. U.S. Environmental Protection Agency. Research Triangle Park, N.C., Oct., 157 pp. Dietzmann, H.E., et al. 1983. Emissions from In-Use Trucks by Chassis Version of 1983 Transient Procedure. U.S Environmental Protection Agency, June. DOT. 1974. The Effect of Speed on Truck Fuel Consumption Rates. Aug. EEA. 1985. Mobile Source Emissions Analysis for California: Vol. II. June. EEA. 1993. Feasibility Study for an I/M Program for Heavy Duty Vehicles. California Bureau of Automotive Repair, Jan. EPA. 1981. Emissions from Heavy Duty Engines Using the 1984 Transient Test Procedure: Vol. II. South West Research Institute report to EPA. PB83-142067, July.

OCR for page 237
EXPANDING METROPOLITAN HIGHWAYS: Implications for Air Quality and Energy Use EPA. 1992. MOBILE5 Emission Factor Model. Washington, D.C. MVMA. 1981, 1986, 1991. Factor Sales of Trucks and Buses. FS-3 and FS-5 monthly reports, Jan. Navistar. 1990. Truck Computer Analysis of Performance and Fuel Economy (TCAPE). Brochure PSM 15200. Radian. 1988. Heavy-Duty Vehicle I/M Study. CARB, Sacramento, Calif., May. Ullman, T.L., et al. 1989. Investigation of the Effects of Fuel Composition on Heavy-Duty Diesel Engine Emissions. Paper 892072. Society of Automotive Engineers.