Search Results

Beginning in 2007, heavy-duty engine manufacturers in the U.S. have been responsible for verifying the compliance of in-use vehicles with Not-to-Exceed (NTE) standards under the Heavy-Duty In-Use Testing Program (HDIUT). This in-use testing is conducted using Portable Emission Measurement Systems (PEMS) which are installed on the vehicles to measure emissions during field operation. A key component of the HDIUT program is the generation of measurement allowances which account for the relative accuracy of PEMS as compared to conventional laboratory-based measurement techniques. A program to determine these measurement allowances for gaseous emissions was jointly funded by the U.S. Environmental Protection Agency (EPA), the California Air Resources Board (CARB), and various member companies of the Engine Manufacturer's Association (EMA). The gaseous pollutants examined in the program were carbon monoxide (CO), non-methane hydrocarbons (NMHC), and oxides of nitrogen (NOx).

Beginning June 1, 2006, 80% of the highway diesel fuel produced in the United States had to contain 15 ppm sulfur or less. To account for sulfur test method variability, the United States Environmental Protection Agency (US EPA) allowed a 2 ppm compliance margin, meaning that in an EPA enforcement action fuel measuring 17 ppm or less would still be deemed compliant since the true sulfur level could still be 15 ppm. Concern was voiced over the appropriateness of the 2 ppm compliance margin, citing recent American Society for Testing and Materials (ASTM) round-robin and crosscheck test program results that showed sulfur test lab-to-lab variability (reproducibility) on the order of 4 to 5 ppm depending on test method.

The U.S. Environmental Protection Agency (EPA) formed a Heavy-Duty Engine Working Group (HDEWG) in the Mobile Sources Technical Advisory Subcommittee in 1995. The goal of the HDEWG was to help define the role of the fuel in meeting the future emissions standards in advanced technology engines (beyond 2004 regulated emissions levels). A three-phase program was developed. This paper presents the results of the statistical analysis of the data collected in the Phase II program. Included is a description of the design of the fuel test matrix, and a listing of the regression equations developed to predict emissions as a function of fuel density, cetane number, monoaromatics, and polyaromatics. Also included is a description of selected analyses of the emissions from a smaller set of fuel data that allowed direct comparison of the effects of natural and boosted cetane number.

In 1995, US Environmental Protection Agency (EPA) formed the Heavy-Duty Engine Working Group (HDEWG). The objective of the group was to assess the role diesel fuel could play in meeting exhaust emission standards proposed for model year 2004+ heavy-duty diesel engines. The group developed a three-phase program to achieve this objective. This paper describes the development of test fuels used in Phase 2 of the EPA HDEWG Program to investigate the effect of fuel properties on heavy-duty diesel engine emissions. It discusses the design of the fuel matrix, reviews the process of test fuel preparation and presents the results of a multi-laboratory fuel analysis program. Fuel properties selected for investigation included density, cetane number, mono- and polyaromatic hydrocarbon content.

A Coordinating Research Council sponsored test program was conducted to determine the effects of diesel fuel properties on emissions from two heavy-duty diesel engines designed to meet EPA emission requirements for 1994. Results for a prototype 1994 DDC Series 60 were reported in SAE Paper 941020. This paper reports the results from a prototype 1994 Navistar DTA-466 engine equipped with an exhaust catalyst. A set of ten fuels having specific variations in cetane number, aromatics, and oxygen were used to study effects of these fuel properties on emissions. Using glycol diether compounds as an oxygenated additive, selected diesel fuels were treated to obtain 2 and 4 mass percent oxygen. Cetane number was increased for selected fuels using a cetane improver. Emissions were measured during transient FTP operation of the Navistar engine tuned for a nominal 5 g/hp-hr NOx, then repeated using a 4 g/hp-hr NOx calibration.

As stringent emission regulations further constrain engine manufacturers by tightening both NOx and particulate emission limits, a knowledge of fuel effects becomes more important than ever. Among the fuel properties that affect heavy-duty diesel engine emissions, cetane number can be very important. Part of the CRC-APRAC VE-10 Project was developed to quantify the effects of cetane number on NOx and other emissions from a prototype 1998 Detroit Diesel Series 60. Three fuels with different natural cetane number (41, 45, 52) were treated with several levels and types of cetane improvers to study a range of cetane number from 40 to 60. Statistical analysis was used to define how regulated emissions from this prototype 1998 engine decreased with chemically-induced cetane number increase. Variation of HC, CO, NOx, and PM were modeled using a combination of a fuel's naturally-occurring cetane number and its total cetane number obtained with cetane improver.

Five different diesel fuel feedstocks were processed to two levels of aromatic (0.05 sulfur, and then 10 percent) content. These materials were distilled into 6 to 8 narrow boiling range fractions that were each characterized in terms of the properties and composition. The fractions were also tested at five different speed load conditions in a single cylinder engine where high speed combustion data and emissions measurements were obtained. Linear regression analysis was used to develop relationships between the properties and composition, and the combustion and emissions characteristics as determined in the engine. The results are presented in the form of the regression equations and discussed in terms of the relative importance of the various properties in controlling the combustion and emissions characteristics. The results of these analysis confirm the importance of aromatic content on the cetane number, the smoke and the NOx emissions.

Several studies have investigated the effects of diesel fuel properties on heavy-duty engine emissions. The objective of this CRC-sponsored test program was to determine the effects of oxygenated diesel fuel, and to further investigate the effects of cetane number and aromatic content on emissions from a heavy-duty engine set to obtain transient NOx emissions below 5 and then 4 g/hp-hr. A fuel set was developed with controlled variations in cetane number, aromatics, and oxygen to superette their effects on emissions. Ignition improver was used to increase cetane number of several fuels. Oxygenated diesel fuel was achieved by adding a “glyme” compound to selected fuels to obtain 2 and 4 mass percent oxygen concentrations. With these fuels, emissions were measured over the EPA transient FTP using a prototype 1994 DDC Series 60 tuned for 5 and then 4 g/hp-hr NOx. No exhaust aftertreatment device was used on this engine.

The SwRI/BMW N.A. Intake Valve Deposit Test procedure was the first performance-based test procedure adopted for fuels qualification in the United States. The initial fuel evaluations were begun in January 1988 with six 1985 BMW 318i vehicles. Since that time, the fleet has grown to include over 60 BMW cars, and more than 2000 tests have been performed. This paper gives a statistical summary of approximately 1800 tests performed over a four-year period. Performance data and possible sources of test variation are discussed. Data and analyses offered represent results of tests by all clients. However, data is presented such that no individual test or client is identified.

Several diesel fuel properties have been identified as having significant effects on diesel engine emissions. For heavy-duty diesel engines, fuel properties of aromatics, back end volatility (represented by the 90 percent boiling point), and sulfur were examined in a previous CRC VE-1 study in which reductions in all three properties decreased regulated emissions to varying degrees. Aromatic levels and cetane numbers were generally correlated in the previous study, so variation in emissions due to “aromatics” could not clearly be assigned to variation in aromatic levels alone. To separate the effects of aromatics and cetane number, a fuel set with controlled variation in aromatics and cetane number was developed, including the use of ignition improver to increase the cetane number of selected fuels. The fuel set was used in a 1991 Prototype DDC Series 60 heavy-duty diesel engine to examine regulated emissions over EPA transient cycle operation.

Exhaust emission data from several fuel effects studies were normalized and subjected to statistical analyses. The goal of this work was to determine whether emission effects of property variation in alternate-source fuels were similar, less pronounced, or more pronounced than the effects of property variation in petroleum fuels. A literature search was conducted, reviewing hundreds of studies and finally selecting nine which dealt with fuel property effects on emissions. From these studies, 15 test cases were reported. Due to the wide variety of vehicles, fuels, test cycles, and measurement techniques used in the studies, a method to relate them all in terms of general trends was developed. Statistics and methods used included bivariate correlation coefficients, regression analysis, scattergrams and goodness-of-fit determinations.

Many challenging statistical problems arise in the research and development of fuels and lubricants. These occur as a result of both economic and environmental constraints since fuels and lubricants are very costly to engine test and must be blended in such a way as to achieve maximum energy efficiency. Engineers are continually faced with devising statistical experiments and analyzing the resultant data to determine how fuel economy or engine performance is affected by changes in fuels and lubricants. Examples of such efforts include predicting how alternative fuels improve mileage, determining the effect of fuel additives on engine performance, and determining the fuel efficiency encountered when using multi-viscosity oils. In this paper some of the statistical techniques that are useful in designing such experiments and in developing useful prediction equations are presented.