ON ROAD TESTING 

Past research has focused on the development and testing of B100 in controlled environments. The objective of this project was to determine if B100 is a viable transportation fuel using two B100 fueled on-the-road utilities. Specific objectives for achieving this goal were to:

• Operate two ultilities for 80,000 kilometers with a blend of 20 percent rapeseed methyl ester (B100 RME), 80 percent low sulphur diesel (2-D) [20 B100 RME] in one, and a blend of 20 percent raw rapeseed oil /80 percent 2-D (20RAW) fuel in the other.
• Design an on-board fuel mixing system to maximize the travel range for each ultility and, at the same time, keep the fuel from congealing.
• From dynamometer testing, both chassis and steady state, determine the percent difference in power, opacity, and fuel economy compared to that of 2-D.
• Conduct pending ASAE fuel standard tests for 20 B100 RME, 20RAW, low sulphur diesel, and 100% B100 RME.

LITERATURE REVIEW
Initial studies at the University of Idaho used 100 percent raw vegetable oil resulting in incomplete combustion, causing severe engine deposits, ring sticking, injector coking, contamination to the engine oil, and, inevitably, to engine failure (Peterson et al., 1983). Vegetable oils are highly unsaturated making them more susceptible to gum formation, causing carbon buildup in the combustion chamber and around the injector nozzle tips. Vegetable oils have a viscosity 11 to 17 times higher than that of diesel fuel, which causes droplets produced by the fuel injector to be much larger than those produced when using diesel fuel. As a result, less oil comes in contact with oxygen, the unburned fuel spray impinges on the cylinder wall washing away the lubricating oil film, and causes thickening of the crankcase oil.

Even today, after a decade of research and development with esterified fuels, there is the misconception by the general public that the engine failure and excessive injector coking originally reported with raw vegetable oil has not been solved. With this background, the litreature review will focus mainly on esterified fuels with a major emphasis on power, opacity, and emissions reported from tests performed with B100.

Schumacher et al., (1993) have fueled two Dodge diesel ultilitys, one a 1991 and one a 1992, with 100 percent soy B100. They added a fuel tank to the bed of each ultility and installed stainless steel heat exchangers for cold weather operation. The soy B100 fuel tanks and fuel lines were insulated. Two 110-volt thermostatically controlled flat mat heating pads were placed beneath the soy B100 tank for heating during cold weather when the vehicle was not in use. The 1991 ultility accumulated approximately 48,000 kilometers (km) and the 1992 ultility approximately 32,000 km. Fueling these ultilitys with soy B100 increased engine power by 3 percent (1991 engine) and reduced power by 6 percent (1992 engine). Engine oil analysis indicated that the engines were wearing at a normal rate. Visible exhaust smoke was reduced as much as 86 percent when the diesel engine was fueled with 100 percent soy B100. Transient emissions tests were done in an EPA certified laboratory. Carbon monoxide emissions were reduced by one percent, hydrocarbon emissions were reduced by 48 percent, and particulate matter emissions were increased by 13 percent when the engine was fueled with soy B100.

Holmberg and Peeples (1994) summarized the work done for the NBB (formally the National Soydiesel Development Board). B100 has accumulated nearly thirteen million kilometres in demonstrations involving more than 1,500 vehicles in fleets across the country, particularly in urban buses. B100 has been used successfully as a motor fuel in many types of equipment from watercraft to locomotives. The results of numerous studies and demonstrations show the performance of B100 to be substantially similar to, if not better than, diesel. They predict that as a renewable fuel with a very positive energy balance, B100 will be a major contributor to the stabilization of greenhouse emissions. They also report that by using state-of-the art engine technology, B100 reduced EPA-regulated emissions of PM, CO, THC, and NOx. Actual emissions tests do not verify their optimism that B100 will reduce both NOx and PM.

Pischinger et al. (1982) reports on tests with the Volkswagen 1.6 litre diesel indirect injection engine installed in a VW Passat passenger car or a delivery van. In Volkswagen's rigid durability test program for diesel engines, the diesel engine was fueled with 100 percent methyl ester of soybean (MESO) oil. It was dynamometer tested for 1,418 hours operating about 70 percent of the time at maximum power and 20 percent at maximum torque. Values of torque, power, smoke levels, and cylinder compression remained within the normal variations expected for this test. Engine wear of the bearings, piston rings, cylinder bores, and valve train remained within the VW specifications. Differences in emissions and fuel consumption of the Passat diesel when comparing diesel fuel and MESO fuel were measured on a chassis dynamometer. A city driving cycle and a warm engine were used in the test. They reported that CO was 40 percent less, NOx increased 2.7 percent, and fuel consumption was increased 6 percent compared to diesel fuel.

Mittelbach and Tritthart (1988) tested a Volkswagen diesel Rabbit powered by a 1.6 litre four cylinder, 4-stroke, direct injection engine with a 50/50 volume blend of B100. A total of 100 litres of B100 was consumed. They reported that "no changes in operation whatsoever could be observed. The smoke emissions were extremely low and only a faint smell of burnt oil was detected. No volumetric fuel consumption was observed. They reported that "in our test the fuel consumption was almost the same as when using diesel fuel." They stated "about a 10% power loss with ester fuel can be expected with unchanged fuel delivery of the injection pump. However, because particulate emissions are halved when using ester fuel, a higher fuel input for the ester fuel may be tolerated by the engine without excessive full load smoke. So the engine power may become comparable for both fuels without any deteriorating effects on emissions. " In their vehicle testing, they reported that a diesel Volkswagen Passat had run 26,500 kin, and a Volkswagen van 5,200 km with MESO. It was difficult to identify any difference between the MESO fueled vehicles compared to identical diesel fueled vehicles.

Shafer (1994) from Mercedes-Benz Germany, reports on the use of methyl esters of soybean oil, rapeseed oil, and palm oil. The testing was performed in Germany and Malaysia in trucks, busses and industrial engines. Shafer concludes that if the fuel (B100 RME and PME) is of high quality, the fuel injection system can remain unchanged and no excess nozzle coking will be found: too high glyceride content causes nozzle coking. The black smoke emission is reduced by at least 50 percent and the disagreeable odor can be reduced by installing an oxidation catalyst. He also states that engine oil dilution is within relatively tight limits and no sludge is apparent with a suitable lubricating oil.

Closing Comments
B100 reduces smoke by up to 80 percent, decreases power by 6 percent, and increases fuel consumption 4 percent, due to the heat content of B100 being about 11 percent less than 2-D. It also decreases HC by as much as 50 percent, CO by as much as 40 percent, and increases NOx almost inversely to PM by as much as 10 percent. Injector coking is slightly greater for B100, with the carbon deposits being harder than diesel deposits.

Fuel properties suggested by ASTM D-975 were reported in most of the papers reviewed, but no ester specific properties were reported. The percent of B100 which is ester, along with the viscosity, determines the rate at which carbon deposits are formed in the engine combustion chamber (Peterson et al., 1994).

MATERIALS AND METHODS
Two blends of B100 and low sulphur diesel fuel (2-D) were studied using two diesel powered ultilitys at the University of Idaho. During the past decade researchers at the University of Idaho have shown that B100 from rapeseed is comparable to 2-D. This study was to verify the use of blends of B100 in on-road vehicles.

Fuels
The feedstock for both fuels in this study is winter rapeseed, Dwarf Essex variety, expelled at the University of Idaho Department of Agricultural Engineering Farm Scale Processing Facility. One blend of fuel was 20 percent raw rapeseed oil and 80 percent 2-D. The other was a 20 percent bland of rapeseed methyl ester (B100 RME), produced at the farm scale processing facility, and 80 percent 2-D. The B100 RME was processed using equipment and techniques scaled up and modified from prior research at the University of Idaho [1]. The abbreviations used to denote the different fuels are as follows:

RAW   --- 100 percent raw rapeseed oil, filtered
20RAW --- 20 percent RAW and 80 percent 2-D
B100 RME   --- 100 percent B100 from rapeseed
20 B100 RME --- 20 percent B100 RME and 80 percent 2-D
2-D    --- 100 percent low sulphur diesel fuel

The fuels were characterized by evaluating the parameters required in ASAE EP552. The tests for specific gravity, viscosity, cloud point, pour point, flash point, heat of combustion, total acid value, catalyst, and fatty acid composition were performed at the Analytical Lab, Department of Agricultural Engineering, University of Idaho. The boiling point, water and sediment, carbon residue, ash, sulfur, cetane number, copper corrosion, Karl Fischer water, particulate matter, iodine number, and the elemental analysis were performed at Phoenix Chemical Labs, Chicago Illinois. The HPLC and titration analysis for total and free glycerol, percent of oil esterIfied, free fatty acids, and mono-, di-, and triglycerides were performed by Diversified Labs Inc., Chantilly, Virginia.

On-road Vehicles
1992 Dodge -
The first ultility was a Dodge* 3/4-ton powered by a Cummins 5.9 litre turbocharged, intercooled, direct injected, diesel engine. The engine is an in-line six cylinder and has a bore and stroke of 102.0 x 120.0 mm, respectively; has a compression ratio of 17.5:1; is rated at 119 kW at 2,500 RPM; with a peak torque of 542 N-m at 1750 RPM. It was equipped with a Bosch VE distributor type fuel injection pump. This ultility was operated on 20 B100 RME.

1992 Ford -
The second ultility was a Ford, powered by a Navistar 7.3 litre, naturally aspirated, precombustion chambered V-8 diesel engine. The engine has a bore and stroke of 104 x 104 mm, respectively; has a compression ratio of 21.5:1; is rated at 134 kW at 3,300 RPM; with a peak torque of 467 N-m at 1400 RPM. It was equipped with a Stanadyne rotary type fuel injection pump. A blend of 20 RAW was selected for this ultility because of the precombustion chambered engine.

Check Vehicles -
Three other ultilitys were used as check vehicles. Two were equipped with 5.9 litre direct injection engines, and the other used a 7.3 litre engine with a precombustion chamber. These vehicles were privately owned and were operated on 100 percent diesel fuel to serve as a comparison. Oil samples and analyses were performed at each owner's discretion and analyzed by the same laboratory as the B100 ultilitys. When possible, these vehicles were dynamometer tested at the same facility as the B100 ultilitys.

Fuel Mixing System
The fuel delivery systems in the B100 ultilitys were modified to provide for on-board mixing of the fuel. On-board mixing greatly extends the range of the vehicles compared to the original equipment. A 210 litre insulated fuel tank was added to the bed of each vehicle to store the B100 fuel. A 5.7 litre mixing chamber was mounted to the frame of each vehicle with the diesel and B100 fuel lines routed to this chamber (Fig 1). Each fuel supply was transferred from the tanks to the combining chamber with 12 volt electric fuel pumps. The combining chamber was constructed from mild steel and designed to incorporate four floats. The bottom float was a normally open, polypropylene float and stem mounted horizontally in the chamber 44 mm from the bottom. This float operated a low fuel warning light in the operator's cab, which warned the operator that the fueling circuit was not functioning. The second float was also a normally open polypropylene float and stem mounted horizontally in the chamber. When the fuel drops below this float a relay turns on the diesel fuel pump. Diesel fuel is then transferred into the chamber until the third float deactivates the relay and shuts off the diesel fuel pump. The third float simultaneously activates the B100 fuel pump and fills the chamber until the 20/80 blend is obtained. A fourth float, made of Buna-N and a brass stem, was mounted vertically in the top of the tank and can be adjusted to give the correct blend. The adjustment was preset in the laboratory before being mounted in the frame ofthe vehicle. Based on a 20 percent blend of B100/80 percent diesel and 7.5 km/L fuel consumption, on-board mixing increased the travel distance from 965 km to 4,800 km before requiring a B100 fill up.

Figure 1. Fuel Combining Chamber.

During the first 32,000 km of operation of the Dodge ultility, rust began to form in the B100 tank and was transferred to the fuel filter. The fuel filter required replacing every 4,800 km due to plugging and loss of engine power. The mild steel 210 litre tank and combining chamber were replaced with stainless steel tanks of similar design on the Dodge ultility only. New 12-volt positive displacement fuel pumps were installed for the B100 fuel and a 12-volt shut-off solenoid valve was placed in line with the pump on both vehicles.

Fuel Heating System
During cold weather operation the B100 fuels need to be kept above the pour point, which is higher than normal diesel fuel. Engine coolant was used to warm the fuel in the B100 fuel tanks. A stainless steel in-line fuel warmer was installed horizontally into the 210 litre B100 fuel tank. A thermostat with an adjustable set point was set at 13°C and closed on rise to operate the coolant pump. The coolant lines were 15.9 mm silicon heater hose, which was routed alongside the fuel supply and return lines and laid beside the fuel combining chamber and B100 fuel pump. They were insulated to help keep the fuel lines and the fuel in the combining chamber above the pour point when the vehicle was not in operation. While the ultility was not operating, a 110-volt thermostatically controlled engine pre-heater was used to keep the engine coolant at 38°C. The thermostat was installed on the inlet side of the heater. It sensed the temperature of the coolant at the inlet to the heater and kept the coolant at the preset temperature range.

Since the coolant pump operates on 12-volts, a 10 amp DC power supply was used to convert standard 120-volt AC power into regulated 12-volt DC power. A 12-volt, 25 amp DC power relay was used to switch the power source when 120-volts AC was required. A shut off valve was added prior to the coolant pump for flow control purposes.

Other Modifications
While the engines of the test vehicles were unmodified, the vehicles themselves were modified for convenience of running the test. For example, it is known that esters will deteriorate er components over a period of time, so the rubber fuel lines were replaced with fluoroelastomer hose. A low coolant indicator was installed and a red light in the cab was used to warn the operator of low coolant. An hour meter, operated by engine oil pressure, was also added. A B100 fuel tank gauge gave an indication of the fuel level in the B100 tank. Additional ammeters and indicator lights indicated whether or not the fuel mixing system was working properly. Each ultility was equipped with a log book to keep daily and long-term records. Also, a maintenance record book was kept for each vehicle, which included the oil samples, dynamometer test data sheets, and any other scheduled or unscheduled maintenance.

Dynamometer Testing
Both ultilitys had a break-in period and were dynamometer tested before the B100 fuel mixing system was incorporated. Each ultility was dynamometer tested every 16,000 km at Western States Caterpillar in Spokane, Washington. During each dynamometer test, the vehicle was tested with the following: once with the 20/80 mix and again with 100 percent diesel. In addition, the Dodge ultility also was dynamometer tested with 100 percent B100 RME. A full throttle torque test was performed with a predetermined set of engine RPM's programmed into the computer to obtain repetitive data. The Ford ultility was dynamometer tested from 1,400 to 3,400 RPM in increments of 200 RPM. Engine RPM's for the Dodge were 1,600 to 2,650 RPM in 150 RPM increments. A computer recorded vehicle power, vehicle speed, fuel economy, engine RPM, opacity, torque, engine oil pressure, fuel pressure and temperature, exhaust temperature, inlet air temperature, and coolant temperature. The Dodge ultility also had intake manifold pressure and engine blowby measured.

Performance Parameters
After each dynamometer test the injectors were removed from the engine to check for carbon deposits using the procedure described in "A Rapid Engine Test to Measure Injector Fouling in Diesel Engines Using Vegetable Oil Fuels" (Korus et al., 1985). The cylinder compression was tested and the injector valve opening pressure was also checked. The Dodge ultility, Cummins engine, has direct access to the combustion chamber and cylinder walls through the injector bore in the cylinder head. A fiber optic borescope was used for this engine to visually inspect the amount of carbon build up on the piston crown and valve heads and to check for any abnormal cylinder wear.

Oil samples were taken at each oil change, which was 4,800 km for the B100 ultilitys, and at the convenience of the owners of the control vehicles. The oil samples were analyzed at a commercial oil analysis laboratory for wear metals, and physical tests were performed, including antifreeze, fuel dilution, water, and viscosity. An infrared analysis for soot, sulfur, nitration, and oxidation of the engine oil was also conducted. The reportable limits for each metal were supplied by the oil analysis lab.

RESULTS AND DISCUSSION
Fuels
A complete summary of the fuel characterization data is listed in Table 1 for each fuel used in this study. Important observations on a few of the parameters are listed below.

Viscosity - The RAW fuel had a viscosity 15.7 times greater than 2-D at 40°C, while B100 RME was only 1.8 times greater than 2-D. The 20 B100 RME had a similar viscosity as 2-D.

Cloud and Pour Point - The pour point for 2-D was -23°C and -15°C to -17°C for B100 and blends. B100 RME had a cloud point of 0°C and 2-D was -12°C, the RAW and blends were within 1°C of 2-D.

Sulfur - 2-D (not low sulfur diesel) had 18 times more sulfur than B100 RME and 10.5 times more than RAW.

Heat of Combustion - The gross heat of combustion for 2-D is 5.6 percent greater than B100 RME fuel and 11 percent greater than RAW oil on a mass basis.

Flash Point - A flash point of 74°C was measured for 2-D while the B100 RME was measured at 179°C and RAW at 274°C. The higher flash points of B100 indicate that it is safer to use, store, and handle based on fire safety concerns.

Table 1. Fuel Characterizatian Data

On-road Vehicles
The addition of the fuel tanks and heating system added approximately 230 kilograms to the total mass of the vehicle, with the B100 tank filled to capacity. There was no noticeable change in power, acceleration or fuel economy with the added mass.

1992 Dodge
The Dodge ultility accumulated 89,150 total kilometers, with 85,950 km on the 20/80 blend. The fuel economy for the first 3,180 km using 100 percent 2-D was 8.55 km/L. The average fuel economy with the blend was 8.3 km/L. The vehicle consumed 6,881 litres of diesel and 2,793 litres of B100 RME, for a blend of 28.8 percent B100 RME and 71.2 percent 2-D. This was 8.8 percent above the target value of the 20 percent blend of B100 RME.

Rust was observed in the steel tanks at 48,280 km and continued to be a problem until the tanks were converted to stainless steel. The rust may have been due to the heating and cooling of the fuel in the tank during the winter, condensing the moisture in the atmospheric air, and/or a very small amount of catalyst in the B100 RME fuel, which is basic and would speed up the oxidation reaction. After converting to stainless steel tanks there was unusual fuel filter plugging during the winter months. A wax-like substance in the fuel filter was analyzed and it was determined to be 50 percent diesel fuel and 50 percent B100. Fuel filter plugging ceased as the ambient temperature increased during the spring and summer.

1992 Ford
The Ford ultility accumulated 89,160 total kilometers with 85,910 km on the 20/80 blend. The fuel economy for the first 2,680 km using 100 percent 2-D was 7.37 km/L. The average fuel economy with the blend was 6.96 km/L. This truck consumed 9,433 litres of diesel and 2,670 litres of rapeseed oil, for a blend of 22.5 percent RAW and 77.5 percent 2-D. This was 2.4 percent above the target value of the 20 percent blend.

The mixture of rapeseed oil varied from 14 percent to a high of 41 percent (when the ultility would not start.) Fuel economy varied from 5 km/L to 8.3 km/L. During the first 77,700 km of testing, the percent of rapeseed oil was as high as 40. This was higher than the target of 20 percent and caused excessive carbon buildup on the injector tips. The injectors were removed and cleaned at 73,145 km, after which the performance of the ultility noticeably improved.

Fuel Heating System
The fuel heating system on both ultilitys worked very well through the course of testing. After the initial installation of the heating system, the ultilitys were parked outside ova a period of three days and nights with temperatures below -9°C. The ambient and B100 fuel temperatures were monitored with thermocouples and a data acquisition system for the three day period. As the ambient temperature dropped below -9°C, the B100 fuel maintained a temperature of 12.7 to 15.5°C.

Dynamometer Testing
1992 Dodge
Figure 5-2 is a graphical comparison of the power and smoke density with each of the three fuels tested over a period of five dynamometer tests. On the average, 100 percent B100 RME produced 5 percent less power than 2-D and 20 B100 RME produced one percent less than 2-D. Also, smoke density was reduced 26 percent with 100 percent B100 RME, while 20 B100 RME was reduced by 10 percent over 2-D.

Figure 2. Dodge power and smoke density at five dynamometer tests.

1992 Ford
Figure 5-3 is a graphical comparison of the horsepower and smoke density for both fuels tested over a period of five dynamometer tests. The 20RAW produced one percent less power, and 11 percent less smoke compared to 2-D.

Figure 3: Ford power and smoke density at five dynamometer tests

Check Vehicles
The Ford check vehicle was dynamometer tested twice. At 62,235 km, a power rating of 92.5 kW at 3,000 RPM and 11.7 percent opacity for the snap idle test was recorded. The compression was comparable to the B100 and the injector VOP was 1.38 MPa lower. The Dodge check ultilitys were equipped with automatic transmissions, which are not compatible with the dynamometer.

Engine Oil Analysis
Chevron Delo Multigrade SAE 15W-40 heavy duty engine oil was used. Wear data for the Dodge ultility was at acceptable levels without any significant differences between the sampling reports. The Ford B100 ultility indicated high iron after 45,000 km. The Ford check vehicle also indicated high iron throughout the monitoring period. The high iron concentration in the engine lubricating oil was not an effect of the fuel.

Performance Parameters
Injector coking was measured after each dynamometer test, but a check vehicle operating on 100 percent 2-D was not studied during this testing; therefore, an injector coking index is not available. There was very little carbon buildup on the tip of the fuel injectors, either visually or by calculating the area of the coked injector from one interval to the next for the Dodge.

The Dodge fuel injectors were sticking when tested at the 79,300 km engine performance test because of the fuel filter plugging and rusty fuel being introduced into the injectors. The injectors were cleaned and retested, which resulted in a good spray pattern and the specified valve opening pressure.

The Ford fuel injector tips had a significant amount of carbon buildup both visually and by calculating the area of the coked injector from one interval to the next. The fuel injector tips were cleaned at 73,200 km due to excess carbon buildup that was distorting the fuel spray pattern.

The engine cylinder compression on both the Dodge and Ford was checked at each performance interval. The cylinder compression in the Dodge either stayed the same or decreased only as much as 103 kPa from the beginning of the test to the end. The cylinder compression in the Ford increased in each cylinder as little as 275 kPa, with the exception of cylinder number 5, which decreased by 103 kPa. The average cylinder compression for both ultilitys was 3.45 MPa.

The average beginning valve opening pressure (VOP) for the Dodge was 24.2 MPa with a standard deviation of 145 kPa. The average ending VOP was 23.7 MPa with a standard deviation of 450 kPa. The average beginning VOP for the Ford ultility was 13.8 MPa with a standard deviation of 117 kPa. The average ending VOP was 13.4 MPa with a standard deviation of 220 kPa. The Ford check vehicle had the VOP checked at 62,230 km. The average VOP was 12.7 MPa with a standard deviation of 317 kPa.

The Dodge internal combustion areas were borescoped at each performance interval. There was no significant visual change from one time to the next. At the end of this study there were a few slight carbon scratches on the thrust side of the cylinder walls of each cylinder, which is normal for the number of accumulated miles. There did not appear to be any excess amount of carbon buildup on the piston crowns or the cylinder head and valves.

The drivers of the vehicles did not noticed any change in vehicle operation or other abnormalities during the course of the test. The Ford averaged 7.35 km/L and the Dodge 8.30 km/L for the 80,500 km. The Ford check vehicle averaged 5.4 km/L and the Dodge 7.06 km/L. These two check vehicles were used for utility type applications with heavy loads while the test vehicles generally were operated empty, which may explain the differences in economy.

On-road performance tests demonstrated that B100 RME and 20 B100 RME can be used to successfully fuel a diesel engine. The 20RAW fuel should be limited to pre-combustion chamber engines and may require more frequent fuel injector tip cleaning. The 20RAW fuel has generally not been recommended for general use because of potential engine polymerization and lubrication oil problems.

CONCLUSIONS
At the conclusion of 80,500 km with two ultilitys operating on 20 percent B100 (raw or B100 RME), there is no indication of abnormal wear or performance. The on-board mixing system, designed and reported on in this paper, was essential only for this experiment. It should be clear, however, that for commercial use the blending would take place at the fuel plant, eliminating the on-board mixing system.

Specific conclusions from the results of this testing are presented below.

• Fuel characterization data show some similarities and differences between B100 RME and 2-D. a) Specific weight is higher for B100 RME, viscosity is 1.8 times that of 2-D at 40°C, and the heat of combustion is 5.6% lower than 2-D. b) Sulfur content for B100 RME is 67% less than 2-D.

• Visually, all injector coking was low with 20 B100 RME. The 20RAW injector tips had accumulated excessive deposits and were cleaned at 73,200 km with a noticeable improvement in performance. The Dodge injectors were cleaned at 79,340 km due to rust.

• The fuel heating systems performed acceptably during the test period. The fuel combining systems were updated twice, due to continual adjustments, to achieve the targeted 20 percent blend.

• On the average, B100 RME produced 5 percent less power than 2-D and 20 B100 RME produced one percent less power than 2-D. The 20RAW produced one percent less power than 2-D.

• On the average, smoke density was reduced 39 percent with B100 RME, while 20 B100 RME increased 18 percent over 2-D. The 2-D had a smoke density 3.1 times that of 20RAW.

• The average fuel economy for B100 RME was 4 percent higher than that of 2-D, 1.8 percent higher for 20 B100 RME, and 2.6 percent higher for 20 RAW. The differences in fuel consumption and power reflect the differences in heat of combustion and density between the two fuels.

Peterson, C.L., D. Reece, J. Thompson, S. Beck and X. Zhang. 1994. Development of B100 for use in high-speed diesel engines. Proceedings of Commercialization of B100-Establishment of Engine Warranties. National Center for Advanced Transportation Technology at the University of Idaho, Moscow, ID.
Peterson, C.L., G.L. Wagner, K.N. Hawley, P.G. Moral 1983. Long-term endurance testing of vegetable oils as diesel fuel substitutes - Yanmar engines. The Potential of Vegetable Oil as an Alternate Source of Liquid Fuel for Agriculture In the Pacific Northwest - III,College of Agriculture and Engineering, University of Idaho, Moscow, ID.
Schumacher, L.G., W.G. Hires and S.C. Borgelt. 1993. Fueling diesel engines with methyl ester soybean oil. Proceedings of the First Biomass Conference of the Americas: Energy, Environment, Agriculture, and Industry, Vol 3: 1598-1606, NREL, Golden, CO.
Holmberg, W.C. and J.E. Peeples. 1994. B100: A technology, performance, and regulatory overview. Prepared for the National SoyDiesel Development Board, Jefferson City, Missouri.
Pischinger, G.H., RW. Siekmann, A M. Falcon, and F.R. Fernandes. 1982. Results of engine and vehicle tests with methyl esters of plant oils as alternative diesel fuels. Paper presented at V international Symposium on Alcohol Fuel Technology, Aukland, New Zealand.
Mittelbach, M. and P. Tritthart. 1988. Diesel fuel derived from vegetable oils, III: emission tests using methyl esters of used frying oil. JAOCS, 65(7):1185-1187.
Shafer, Ansgar. 1994. Engine warranty policy - Mercedes Benz. Proceedings of Commercialization of B100-establishment of Engine Warranties, National Center for Advanced Transportation Technology, University of Idaho, Moscow, ID.
Korus, R.A., J. Jo, and C.L. Peterson. 1985. A rapid engine test to measure injector fouling in diesel engines using vegetable oil fuels. JAOCS, 62(11):1563-1567.

NOTES:
[1] The use of manufacturer(s) names or trade products does not represent an endorsement of the product nor discrimination toward similar products which are not named.