alternative fuel light-duty vehicles - Energy.govAlternative Motor Fuels Act (AMFA) was enacted in 1988 to encourage the development and use of alternative transportation fuels (including

Alternative Fuel Light-Duty Vehicles

T OF ENERGYD

EPA

RTMEN

U

E

NIT

ED

STAT S OFA

ER

ICA

M

SUMMARY OF RESULTS FROM THENATIONAL RENEWABLE ENERGY LABORATORY’S

VEHICLE EVALUATION DATA COLLECTION EFFORTS

Alternative Fuel Light-Duty Vehicles

SUMMARY OF RESULTS FROM THENATIONAL RENEWABLE ENERGY LABORATORY’S

VEHICLE EVALUATION DATA COLLECTION EFFORTS

PEG WHALENKENNETH KELLYROB MOTTA

JOHN BRODERICK

MAY 1996

NT

Y

AU

E OF ENERGD

EPA

RTM

E

NIT

ED

STAT S OFA

ER

IC

M

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2Light-Duty Vehicles in the Program . . . . . . . . . . . . . . . . . . . . . . . . . . . .2Data Collected . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

Performance and Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

Fuel Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

Emissions Testing Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13OEM Vehicle Emission Test Results . . . . . . . . . . . . . . . . . . . . . . . . . .14Conversion Vehicle Emissions Test Results . . . . . . . . . . . . . . . . . . . . .16

Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

Additional Acquisition Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18Fuel and Oil Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19Maintenance Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21Sample Operating Cost Comparison . . . . . . . . . . . . . . . . . . . . . . . . . .21

Other Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

Data Collection Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23Other Information Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25

For Additional Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27

Contents

iii

Summary

The U.S. Department of Energy’sNational Renewable Energy Laboratory conducted a data collec-tion project for light-duty, alternativefuel vehicles (AFVs) for about 4 years. The project has collecteddata on 10 vehicle models (from theoriginal equipment manufacturers[OEM]), spanning model years 1991through 1995. Emissions data havealso been collected from a number of vehicles that were converted tooperate on compressed natural gas(CNG) and liquefied petroleum gas.Most of the vehicles involved in the data collection and evaluation are part of the General ServicesAdministration’s fleet of AFVs. This evaluation effort addressed the performance and reliability,fuel economy, and emissions oflight-duty AFVs, with comparisonsto similar gasoline vehicles whenpossible.

Driver-reported complaints andunscheduled vehicle repairs wereused to assess the performance andreliability of the AFVs compared tocomparable gasoline vehicles. Drivercomplaints and unscheduled repairswere highest for early model AFVs,and both have decreased with eachnew model year. The driver com-plaint and vehicle repair trends indicate that AFV performance andreliability are approaching the levelsof similar gasoline-fueled vehicles.

Two sources of fuel economy datawere available, one from testing of vehicles on a chassis dynamo-meter, and the other from records

of in-service fuel use. Fuel economycan vary over a wide range, depend-ing on type of service and individual driving styles. In general, on anequivalent energy basis, the fueleconomy of the AFVs in this evaluation was comparable to that of standard gasoline vehicles.

This report includes results fromemissions testing completed on 169 AFVs and 161 gasoline controlvehicles. Alcohol vehicles in generalindicated equivalent or lower regu-lated emissions compared to refor-mulated gasoline. CNG vehicles didshow significantly lower emissionsthan gasoline vehicles. Preliminaryemissions results from vehicles thathave undergone aftermarket conver-sion are not as promising as forOEM AFVs. Conversion emissionsin many cases were higher than the vehicle emissions were beforeconversion.

The costs associated with owningand operating vehicles significantlyaffect the likelihood that fleet own-ers or individuals will buy them.Acquisition costs of AFVs have generally been higher than those of comparable gasoline vehicles,with CNG vehicles priced as muchas 25% higher. Fuel and oil costs tend to be higher for alcohol-fueled vehicles, but lower for CNG-fueledvehicles. Maintenance costs of AFVsare expected to be marginally higherthan those for similar gasoline vehi-cles. The included operating costanalysis allows first-order compar-isons between AFV and gasolinevehicle costs.

1

2

Light-Duty Vehicle Summary

Introduction

Overview

More than 60% of the oil consumedin the United States is used for trans-portation, and imported oil accountsfor more than 50% of the oil sup-plies. In response to widespread concern about the impact of fuelconsumption on our economy andthe environment, Congress recog-nized the need to reduce our nation’sdependence on imported oil. TheAlternative Motor Fuels Act(AMFA) was enacted in 1988 toencourage the development and useof alternative transportation fuels(including methanol, ethanol, andcompressed natural gas [CNG]),and to encourage the production ofmethanol-, ethanol-, and CNG-fueledvehicles. AMFA also requires theU.S. Department of Energy (DOE) to collect data to evaluate the perfor-mance, fuel economy, emissions, andoperating and maintenance recordsof alternative fuel vehicles (AFVs).DOE designated the NationalRenewable Energy Laboratory(NREL) as the program manager forthe light-duty vehicle data collectionand evaluation projects. The Federalgoverment’s General ServicesAdministration (GSA) has beenrequired to purchase AFVs from theoriginal equipment manufacturers(OEM) for use throughout its fleet.NREL worked with GSA to identifyagencies that were using the AFVsand were good candidates for thedata collection efforts.

NREL has been collecting opera-tional, performance, maintenance,and emissions data from variouslight-duty vehicles, including

passenger cars, vans, and pickuptrucks, for about 4 years. These data have been publicly accessiblethrough the Alternative Fuels DataCenter (AFDC) electronic database,and through the National AlternativeFuels Hotline.

The light-duty vehicle program wasdesigned to evaluate AFVs availablefrom automobile manufacturers thatoperate on ethanol, methanol, orCNG. The program has recentlybeen involved in the aftermarket conversion of a number of vehiclesin the Federal fleet.

The evaluation focuses on perfor-mance and reliability, fuel economy,and emissions of the vehicles operat-ing on the various alternative fuels.Analyses of the data collected pro-vide useful information to fleet oper-ators who are using or consideringadding AFVs to their fleet.

Light-Duty Vehicles in the Program

The program has been collecting datafrom drivers of 10 light-duty vehiclemodels for about 4 years, spanningmodel years 1991 through 1995 (seeTable 1). The vehicles involved in thedata collection and evaluation are asubset of the AFV portion of the GSAfleet, which has grown to more than15,000 vehicles in 1995. The AFVsstudied included three CNG-fueledmodels, two ethanol-fueled models,five methanol-fueled models, and sixgasoline-fueled models as controls.Although 1992 CNG Chevrolet pickups were recalled by the manu-facturer for safety reasons, the datacollected on those vehicles are included in the results presented here.

Federal agencies at various locationsacross the United States have partici-pated in the data collection efforts(Figure 1). Vehicles have been oper-ated in a wide range of “real-world”applications under varying climaticconditions and at various altitudes.The size of the data collection fleethas varied during the program asmore AFVs became available and asfunding for the project changed. The data collection fleet contained about80 vehicles at the beginning and grewto a maximum of about 800 vehicles.In 1995, data were collected from337 vehicles that operated on alterna-tive fuels, and 146 vehicles that operated on gasoline and served ascontrol vehicles. The mix of AFVs inthe program has also varied, becauseGSA generally removes passengervehicles from service after 3 years,and vans and trucks after 6 years.

The automobile manufacturersoffered various passenger cars, vans,and pickup trucks that could operateon one of the alternative fuels. Thisdata collection and evaluation efforthas primarily focused on AFVsavailable from the OEMs. NREL,as directed by DOE, has also assistedin aftermarket conversions to enablevehicles to operate on alternativefuels (see sidebar on OEM/QVM/conversion on next page).

The vehicles in this evaluationincluded dedicated CNG vehicles,and flexible-fuel vehicles (FFVs)that can operate on 85% methanol(M85) or 85% ethanol (E85). Inaddition, the evaluation included anumber of vehicles that have beenconverted to operate on CNG andLPG; these vehicles have been sub-jected to limited emissions testing.

3


Table 1. Light-Duty Vehicle Models in Data Collection Program

Type of Fuel Model Year Total Number

Chevrolet pickup* 1992 210

Compressed Dodge Caravan 1994 8Natural Gas 1995 1

Dodge Ram van 1992 711994 40

Ethanol Chevrolet Lumina 1992 211993 34

Ford Taurus 1994 16

Chevrolet Lumina 1991 211993 29

Dodge Spirit 1993 2791994 13

Methanol Dodge Intrepid 1995 8

Ford Econoline van** 1992 141993 3

Ford Taurus 1991** 361993 16

Chevrolet pickup 1993 67

Chevrolet Lumina 1991 81993 39

Dodge Spirit 1993 88

Gasoline Dodge Ram van 1992 261994 41

Dodge Intrepid 1995 10

Ford Econoline van 1993 20

Ford Taurus 1991 81993 26

* Recalled for safety reasons; available data included in analysis** Not a production vehicle; part of a vehicle demonstration fleet


The OEM dedicated-CNG vehicleswere specifically designed to operateonly on CNG. A number of vehicledesign modifications were requiredto enable operation on a gaseousfuel. These include installing

high-pressure gas cylinders in placeof the fuel tank, a new fuel injectionsystem, internal engine modificationsto accommodate the gaseous fuel,suspension changes to accommodate the additional weight of fuel tanks,and the addition of other fuel-management-related hardware such as pressure-relief devices.

The alcohol-fueled vehicles arecalled FFVs. This type of vehicle is capable of operating on gasolinealone or on various blends of alco-hol and gasoline, to a maximum of 85% ethanol or methanol. Thedesign changes required to accom-modate alcohol-fueled vehicles areless extensive than those required toaccommodate operation on a gaseousfuel. The FFV design modificationsincluded incorporation of alcohol-resistant materials in the fuel system,a different fuel sensor, and a

4

What’s the difference between an OEM, a QVM, and a converted vehicle?

There are three principal types of AFVs available to the fleet manager.

OEM Vehicles: These vehicles are designed and built by the OEMs, primarily Chrysler, Ford, and GM. All alcohol-fueledvehicles, and some CNG and liquefied petroleum gas (LPG) vehicles fall into this category. OEM vehicles are designed tohave all of the engine, suspension, and chassis upgrades that an AFV needs for optimum performance and durability. Theyare designed to meet Federal safety and emissions standards, including crash worthiness. They have single comprehensivewarranties that cover all components, including the alternative fuel ones. These vehicles tend to be most optimized for thealternative fuel.

Qualified Vehicle Modifier (QVM) Vehicles: These vehicles are similar to the OEM vehicles except that the OEM haslinked up with a “qualified” conversion company to complete the final assembly of the vehicle. These vehicles generallyhave the same upgrades to the engine and chassis as the OEM vehicle, meet the same safety and emissions standards, andoffer a single comprehensive warranty. They may have less sophisticated fuel systems than the OEM models. The finalvehicle may be dedicated or bi-fuel depending on owner preference.

Aftermarket Conversions: These are gasoline vehicles that are converted by an independent company after a vehicle issold. They do not have the engine and chassis upgrades that the other two types of vehicles offer. The conversion companyprovides a warranty that is separate from the OEM warranty, and generally the OEM warranty will not cover any damagescaused by the installation or operation of the vehicle on the alternative fuel. Available aftermarket conversions enable vehicles to operate on CNG or LPG.

Figure 1. Locations of light-dutytest vehicles


5

reprogrammed engine micro-processor designed to compensatefor varying fuel blends.

Another option available to fleetoperators is aftermarket conversionof vehicles. Aftermarket conversionsinvolve adding equipment, after theoriginal purchase of the vehicle, toenable operation on a fuel other thanthe original design fuel. Light-dutyvehicles are commonly converted tooperate on CNG and propane. MostCNG conversions result in bi-fuelvehicles, which means they can oper-ate on gasoline or on the designatedalternative fuel. The NREL projectwill convert about 900 Federal fleetvehicles, which are operating at vari-ous locations throughout the country.

Data Collected

Information related to vehicle perfor-mance and reliability, fuel economy,and emissions was collected. Twotypes of data were collected to evalu-ate performance and reliability of theAFVs and gasoline control vehicles.The first type was driver feedback on various aspects of vehicle perfor-mance. Each driver of a study vehiclewas asked to fill out a card at eachvehicle refueling. The card containedcheck-off boxes that allowed the dri-vers to report whether any of severalproblems occurred while they oper-ated the vehicle; the cards also pro-vided space for the drivers to makeother comments about the vehicle.

The second type of performance andreliability data was vehicle mainte-nance and repair records. These datawere initially collected by NRELfrom automobile dealers and repairshops that maintained the GSA

vehicles. For a time, as the size of thedata collection fleet increased, NRELused subcontractors to collect themaintenance and repair records forthe various participating sites. NRELhas also obtained GSA’s maintenanceand repair records for the vehiclesparticipating in this study.

Drivers were also asked to recordvehicle mileage and fuel added ateach refueling. This information wasused to evaluate in-use vehicle fueleconomy.

Fuels Being Evaluated

Alternative Fuels

Methanol: Methanol is an alcohol produced primarily from natural gas,but it can also be derived from biomass or coal. Thus the potential domesticresource base for methanol is vast. The M85 FFVs in this program can oper-ate on a mixture of 85% methanol and 15% gasoline.

Ethanol: Ethanol is an alcohol derived from biomass such as corn, sugarcane, grasses, trees, and agricultural waste. The domestic resource base forethanol production is vast. The E85 FFVs in this program can operate on amixture of 85% ethanol and 15% gasoline.

Compressed Natural Gas: Natural gas is composed primarily of methane(approximately 93%) with a mixture of other hydrocarbons. It is derived fromgas wells or in conjunction with crude oil production. CNG is stored in high-pressure cylinders in a gaseous form.

Liquefied Petroleum Gas: LPG is a mixture of petroleum and natural gases,whose primary constituent is propane. It is produced as a by-product of nat-ural gas processing and petroleum refining. It is a gas at ambient temperatureand pressure, but with moderate pressure (less than 200 psi) it condenses to aliquid, making it easy to store and transport.

Gasoline

Reformulated Gasoline (RFG): RFG is gasoline developed to have lowersulfur content, reduced aromatic concentrations, and a lower vapor pressure.Oxygenates (such as ethanol, methyl tertiary butyl ether, ethyl tertiary butylether, or tertiary amyl methyl ether), are added to reduce carbon monoxide(CO) emissions while improving the octane quality of the gasoline. California-certified RFG was used for the emissions tests reported here.


6

At regularly scheduled mileage inter-vals, a large sample of the study fleetvehicles was temporarily removedfrom service to undergo emissionstesting. Standard Federal emissionstest procedures and qualified test lab-oratories were used to compare theemissions of vehicles operating onalternative fuels to those of vehiclesoperating on RFG.

Data were generally collected onvehicles as long as they remained in service. The quantity of data avail-able for each vehicle depended onhow the vehicles were used. GSAvehicles tended to accumulatemileage at a relatively low rate—usually less than 10,000 miles annu-ally. Thus, many GSA vehicles weretaken out of service before the vehi-cles were out of warranty, which lim-ited the evaluation efforts. The datacollection effort thus far has focusedon the OEM AFVs, with only limitedemissions data collected on conver-sion vehicles. The goal was to collectand provide a variety of data duringthe service life of the various AFVsand like gasoline vehicles in this project. This report summarizes thedata collected and provides an evalua-tion of the AFVs in this study, andwhere possible provides comparisonsof comparable gasoline vehicles.

The sections that follow address performance and reliability, fueleconomy, and emissions for the light-duty vehicles in this program. Wehave included a discussion on AFVcost-related issues, although much of this information did not comedirectly from this evaluation effort.The final section discusses some ofour data collection experiences, and

provides sources for additional information on light-duty AFVs.

Performance and ReliabilityThe performance and reliability oflight-duty vehicles can be assessed in several ways. In our program, twotypes of information were collectedfor this purpose. Drivers of the vari-ous AFVs and gasoline control vehi-cles were asked to provide feedbackon the performance of the vehiclesthey drove. The drivers reported any of a number of common performance-related problemsencountered while operating the vehicle. Data were also collected onthe repairs required by the variousvehicles to help assess their reliabili-ty. Maintenance and repair recordswere often difficult to obtain fromrepair shops and GSA. GSA does not generally track all repairs that arecovered under warranty because oftenno cost is associated with them. Theanalysis of vehicle reliability is basedon the data available, but we areaware that our records are incomplete.

The drivers of the study vehicles were asked to fill out a card each timethey refueled the vehicle. The cardincluded a series of check-off boxes,in which the drivers could indicatewhether they had experienced any of a number of performance-relatedproblems while driving the vehicle.The performance-related complaintsincluded the vehicle being hard tostart, the check engine light comingon, idle quality, hesitation, lack ofpower, engine ping, and the vehiclestalling after starting or in traffic.Table 2 summarizes the numbers ofdriver-reported complaints for each


7

vehicle model in this project. Driver-reported complaints were significantly more common for theearlier model AFVs, which have moretime in service, and which in generalhave accumulated more miles than

newer models. In addition, duringapproximately the first year of datacollection, the study fleet was smalland it was easier to obtain more complete participation in the data collection effort.

Table 2. Summary of Driver-Reported Complaints* for All Vehicles in LDV Data Collection Program

Year Fuel Model Vehicles in Total Reported Total Reported Complaints/VehicleProgram Complaints (< 10,000 mile) (< 10,000 mile)

1991 M85 Chevy Lumina 21 435 107 5.1M85 Ford Taurus 36 752 395 10.9Gas Chevy Lumina 8 174 62 7.8Gas Ford Taurus 8 45 22 2.8

1992 CNG Chevy pickup 210 1726 1513 7.2CNG Dodge Ram van 71 962 498 7.0E85 Chevy Lumina 21 133 39 1.8M85 Ford Econoline van 14 33 8 0.57Gas Dodge Ram van 26 0 0 0

1993 E85 Chevy Lumina 34 10 4 0.12M85 Ford Econoline van 3 0 0 0M85 Chevy Lumina 29 37 38 1.3M85 Dodge Spirit 279 221 102 0.36M85 Ford Taurus 16 28 19 1.2Gas Chevy pickup 67 12 12 0.18Gas Ford Econoline van 20 10 9 0.45Gas Chevy Lumina 39 9 7 0.18Gas Dodge Spirit 88 127 22 0.25Gas Ford Taurus 26 25 13 0.5

1994 CNG Dodge Caravan 8 14 15 1.9CNG Dodge Ram van 40 47 43 1.1E85 Ford Taurus 16 6 4 0.25M85 Dodge Spirit 13 0 0 0Gas Dodge Ram van 41 10 9 0.22

1995 CNG Dodge Caravan 1 1 1 **M85 Dodge Intrepid 8 0 0 **Gas Dodge Intrepid 10 2 2 **

* Includes the following types of complaints: ** Insufficient data accumulated to dateHesitation Lack of powerCheck engine light PingingHard to start Stalling after startPoor idling Stalling in traffic


8

To ensure a more valid comparisonbetween the different model yearvehicles, only driver complaintsreported in the first 10,000 miles ofvehicle operation were included inthe analysis presented here. Figure 2provides an overall view of the aver-age number of complaints per vehicleby fuel type (1995 vehicles were

excluded because of insufficientdata). This figure includes all thevehicles and model years for eachfuel type. Table 3 gives an idea of themost commonly reported complaints.Table 2 and Figure 2 data indicatethat drivers experience the least difficulty with gasoline vehicles and experience the most difficultywith CNG-fueled vehicles.

When the number of complaints is examined by model year (see Figure 3) for each fuel, however, theresults paint a somewhat differentpicture. Again, only complaintsreported in the first 10,000 miles ofvehicle operation are included. Morethan five complaints per vehicle werereported for M85 and gasoline ver-sions of 1991 vehicles. The numberof complaints on the early modelyear vehicles may be high partiallybecause the number of vehicles wassmall, and the project was new, sodata collection and driver participa-tion were easier to obtain. Even so,the number of complaints per vehiclehas decreased with newer model yearvehicles. The data for the 1993 and1994 vehicles indicate that AFV per-formance, based on driver feedback,is approaching that of gasoline-fueled vehicles. In general, driversatisfaction with AFVs has increasedsignificantly since the vehicles werefirst introduced to the GSA fleet.

Figure 4 provides a more detailedillustration of how complaints pervehicle have changed with newermodel vehicles. This figure showsthe average number of various per-formance complaints for two modelyears of CNG Dodge Ram vans. For1992 vehicles, the average number of

Table 3. Most Commonly Reported Driver Complaints by Fuel*

Complaint CNG M85 E85 Gasoline

Hesitation 18% 13% 32%

Check engine light 9%

Hard to start 17% 9% 20%

Poor idling 23% 45% 14% 47%

Lack of power 11% 10% 12% 14%

Stalling after start 13% 13%

Stalling in traffic 11%

* Values are percent of total complaints for all program vehicles operating on each fuel

Figure 2. Average driver-reportedvehicle complaints for each fuel (all program vehicles, all reports in first 10,000 miles of vehicle operation)

Number of Complaints/ Vehicle

0 7654321

Gasoline

E85

M85

CNG


each type of complaint was less thantwo, with vehicle hesitation and lackof power being the most commoncomplaints. Drivers of 1994 CNGvans reported fewer than 0.5 com-plaints per vehicle for all categories.The experience gained from the ear-lier vehicle model has apparentlyresulted in significantly improvedperformance in the newer CNGmodel.

One common performance-relatedcomplaint that CNG-fueled vehicledrivers reported (beyond choices onthe data card) was a lack of range.Some drivers reportedly limited theiruse of these vehicles for fear of run-ning out of fuel. This continues to bea factor when considering dedicatedCNG vehicles, particularly in areaswith a limited fueling infrastructure.

Figure 5 shows the complaints per vehicle for the 1993 ChevroletLuminas operated on gasoline, M85,and E85. All performance areasreceived fewer than one complaintper vehicle (note x-axis scale com-pared to Figures 2 and 4), regardlessof fuel. This indicates that in generalthe drivers of this vehicle were satis-fied with performance. Similarresults were seen for other newermodel vehicles when a direct com-parison could be made between thealternative fuel and gasoline.

The assessment of vehicle reliabilityhas been based on the maintenanceand repair records available for thevarious vehicles in our program.Maintenance and repair characteris-tics of any vehicle are significantlyinfluenced by factors such as the duty cycle and the rate of mileageaccumulation. Also, in this program

9

Figure 3. Driver-reported complaints byfuel type (in first 10,000 miles of vehicleoperation)

1 No CNG vehicles in 1991 or 1993 and no E85vehicles in 1991

2 No gasoline vehicle complaints reported in 1992

3 No M85 vehicle complaints reported in 1994

Figure 4. Driver-reported complaints per vehicle for CNGDodge Ram van (in first 10,000miles of vehicle operation)

Aver

age

of C

ompl

aint

s/Ve

hicl

e

1994199319921991

10

8

6

4

2

0211

E85M85CNGGasoline

1 3

1992 1994

Number of Complaints/Vehicle

0 21

Stall in Traffic

Stalled after Start

Pinging

Lack of Power

Poor Idle Quality

Hard to Start

Hesitation

1.50.5


10

some early model AFVs were proto-types or included components thatwere still developmental. With thesefactors in mind, to evaluate the relia-bility of the vehicles, we assessed thenumber and type of unscheduledrepairs. The term unscheduled repairsimplies breakdowns or problems that

require servicing and affect vehicleavailability. Most unscheduled repairsfor the AFVs involved the fuel or fuelsystem, and, as with the performanceproblems, they were more numerousin the early model year AFVs (1995vehicles were excluded because ofinsufficient data).

The unscheduled repairs per vehicle are shown in Figure 6,which includes data for all vehiclemodels and years by fuel. The CNGand M85 vehicles have averagedabout 2.5 times the number ofunscheduled repairs as gasoline vehicles. Although shown in this figure, the repair data for the ethanolvehicles have been very limited, andare not sufficient to enable a validassessment of their reliability.

When the analysis focused onunscheduled repairs in vehicle sys-tems affected by alternative fuels(such as fuel system, engine, andexhaust system), the early modelCNG and M85 vehicles were foundto have significantly more repairs pervehicle than the gasoline vehicles(see Figure 7). For the 1992 CNGvehicles, the most common fuel-related repairs included fuel tank(replacement or additions), fuel linesand valves, fuel injectors, and fuelpressure regulators. Reports of injec-tor repairs are nearly gone, and thenumber of fuel tank and valve andline repairs has decreased for 1994CNG vehicles. The most commonrepairs for the 1991 M85 vehiclesincluded fuel pumps, fuel injectors,fuel filters, and fuel senders andgauges. Reports of similar repairshave decreased with each new model

Figure 5. Driver-reported complaints per vehicle for 1993 Chevrolet Luminas (withfewer than 10,000 miles on vehicles)

Gasoline M85

Number of Complaints/Vehicle

0

Stall in Traffic

Stalled after Start

Pinging

Lack of Power

Poor Idle Quality

Hard to Start

Hesitation

0.60.1 0.2 0.3 0.4 0.5

E85

Figure 6. Average number ofunscheduled repairs by fuel type(repair data are very limited for E85 vehicles)

Num

ber o

f Rep

airs

/Veh

icle

2.5

2.0

1.5

1.0

0.5

0Gasoline E85 M85 CNG


year and are significantly lower thanfor the 1991 M85 vehicles.

In evaluating Figure 7, it is importantto keep in mind that the earlier vehi-cles have accumulated more mileage,and are likely to have reported morerepairs, than the 1994 vehicles. How-ever, when comparing the samemodel year, the AFVs tend to havemore fuel- and fuel-system-relatedrepairs.

The mileage and in-service time vary for vehicles of different modelyears, but unscheduled repairsappear to be decreasing for newerAFVs. Figure 8 displays the averagenumber of unscheduled repairs that have been reported in the first10,000 miles of vehicle operation.The gasoline vehicles reported about1.25 unscheduled repairs per 1992model vehicles to about 0.6 per 1993and 1994 models. The 1992 CNGvehicles reported about 50% moreunscheduled repairs than the gasolinevehicles, and the 1992 M85 vehiclesreported about 40% more. However,the average number of unscheduledrepairs to the 1994 CNG and the1993 and 1994 M85 vehiclesdropped significantly, to be slightlylower than that for gasoline vehiclesof the same model year.

The decline in unscheduled repairsfor vehicles reported with fewer than 10,000 vehicle miles probablyresults both from vehicle designchanges and from repair technicians’increased familiarity with AFVs. Inmost cases unscheduled fuel-relatedand other repairs were covered bywarranty, so excess downtime wasthe real cost to fleet operators. But

the repair trends indicate thatunscheduled repairs of AFVs havedecreased and are approaching the levels seen in similar gasoline-fueled vehicles.

The data collected on light-dutyAFVs in this program indicate thatboth the performance and reliabilityof the AFVs have improved witheach new model year.

11

Figure 7. Average number of fuel-related unscheduled repairs bymodel year

Repa

irs p

er V

ehic

le

7

6

5

4

3

2

1

01992 1994 1991 1992 1993 1994 1991 1992 1993 1994

M85 GasolineCNG

Figure 8. Average number ofunscheduled repairs reported (in first 10,000 miles of vehicle operation)

Aver

age

Num

ber o

f Rep

airs

/Veh

icle

2

1

0

M85 GasolineCNG

19941993199219941993199219941992


Fuel Economy

Two sources of fuel economy datawere available, one from testingvehicles on a chassis dynamometer(during emissions tests), and theother from analyzing refuelingrecords. The alternative fuel con-sumption was converted to gallons ofgasoline equivalent (GGE) for bothsets of data so they could then becompared to fuel economy data forgasoline. The conversion is per-formed by comparing the potentialenergy contained in each of the alter-native fuels. Table 4 lists the energycontent (in Btu) of each typical com-mercially available fuel, and theGGE correction factor for each fuel.The energy content for CNG is listedin Btu/lb, because CNG is a gas andcannot be directly related to theBtu/gal measure for gasoline.

Fuel economy information wasobtained on each vehicle duringemissions testing on a chassisdynamometer. Each vehicle followedthe Federal Test Procedure (FTP)

using the Urban DrivingDynamometer Schedule (city cycle)and specially blended emissions testfuels. A number of similar vehiclesare tested and measurements are thenaveraged to obtain a single value.Average values from vehicles whoseemissions were tested are providedin the column labeled “mpg” inTable 5. Each alternative fuel’s aver-age mpg was within 1 or 2 mpg ofthe average mpg obtained with RFG.

In-service fuel economy data werecalculated by using refueling recordsmaintained in DOE’s AFDC atNREL. Vehicle use varies consider-ably because of individual drivingstyle and other factors such as typeof driving (stop-and-go city, high-way, deliveries, or some combinationof all three), and climate.

Calculations of in-use fuel economywere further complicated in the caseof the alcohol-fueled vehicles. TheFFVs could be fueled with eitheralcohol or gasoline at any refueling.When this occurred, the gasolinewould mix with the alcohol fuelremaining in the tank, creating aunique blend. To account for this, thein-use fuel economy for the alcohol-fueled vehicles was based only oninstances where the vehicles wererefueled three consecutive times with alcohol fuel.

In-service fuel economy is alsoshown in Table 5. The in-service lowand high mpg values are presented foreach AFV and comparable gasolinecontrol vehicle. These values repre-sent the ranges of fuel economyobtained by the various vehiclesinvolved in this light-duty vehiclestudy. Some vehicles have a wide

12

Table 4. Typical Fuel Energy Content and Conversions Data

Energy Content

Fuel Lower Heating Gallon of Gasoline Value (Btu/gal) Equivalent (GGE)*

Gasoline 115,400 1

Ethanol (E85) 81,500 1.42

Methanol (M85) 65,100 1.77

Btu/lb

CNG 20,356 N/A

* Divide gallons of fuel by this factor to get GGE


mpg range, which is most likely theresult of the variability in type of driving and driving styles. In mostcases, on an equivalent energy basis,the vehicles operating on alternativefuels achieve fuel economy levelssimilar to standard gasoline vehicles.

Emissions

Emissions measurement is the singlemost comprehensive part of the testprogram for light-duty vehicles. Theeffort undertaken by NREL is alsothe most extensive and carefully controlled study of AFV emissionsof which we are aware. The study

focused primarily on vehicles fromthe OEMs, but a small-scale investi-gation of emissions from convertedvehicles was also undertaken. The preliminary results of bothefforts is discussed below.

Emissions Testing Procedures

To evaluate the emissions perfor-mance of AFVs, a large number ofrandomly selected OEM vehicleswere tested using standard FTPs,the same procedures used by theEnvironmental Protection Agency(EPA) to certify vehicle emissionsperformance. These procedures are

13

Table 5. Fuel Economy Results from Emissions Tests and In-Use Fueling Records (All Alternative Fuel Values in Miles per Gallon of Gasoline Equivalent)

Vehicle Model Fuel Type* Model Year MPG** In-Use MPG

Low High

Chevrolet pickup CNG 1992 12.0 7 14Gasoline 1993 14.0 10 16

Chevrolet Lumina E85 1992, 1993 20.2 9 29M85 1993 19.5 14 30Gasoline 1993 19.1 14 28

Dodge Caravan CNG 1994 not available 8 13

Dodge Ram van CNG 1992, 1994 12.5 8 15Gasoline 1992, 1994 13.5 6 17

Dodge Spirit M85 1993, 1994 22.3 15 31Gasoline 1993 24.0 21 32

Ford Econoline*** M85 1992, 1993 13.9 8 19Gasoline 1993 15.0 9 18

Ford Taurus E85 1994 22.0 11 28M85 1993 20.7 18 31Gasoline 1993 21.4 21 34

* Reformulated gasoline was used for all gasoline emissions tests.** Average fuel economy measurements during emissions tests.

*** Not a production vehicle, part of a vehicle demonstration fleet.

far more complex and representativeof real-world conditions than a typicaldealership or state emissions inspec-tion station can perform, and includeemissions taken while the car is driven over a simulated city route.The primary regulated emissionscompounds, carbon monoxide (CO),oxides of nitrogen (NOx), hydrocar-bons (HC), and non-methane hydro-carbons (NMHC) were measured.Carbon dioxide (CO2), which is currently not regulated, but which is a“greenhouse” gas, was also measured.Because HC emissions are really aclass of compounds, rather than a single compound, the HC emissionsfrom a representative sample of vehi-cles were broken down into theirmore than 300 constituent compoundsto evaluate their toxicity and ozone-(smog-) forming potential.

Vehicles were taken from GSA andother government fleets. In all cases,vehicles were tested by one of threelaboratories in the country. Theselaboratories were carefully selectedbased on their proven ability to perform high-quality testing thatconforms to EPA’s procedures. Specially blended fuels were usedfor all testing, including for the gaso-line “baseline” fuel. For the gasolinefuel, California Phase 2 reformulatedgasoline (RFG) was chosen becauseit represents the state of the art ofgasoline fuel today. If alternativefuels are to compete successfully,they must compete with the bestgasoline fuels.

The goal of the ongoing program isto test vehicles at regular mileageintervals and to test the latest AFVtechnologies as they become avail-able. The results presented here arebased on analysis of the first roundof testing on the vehicles.

OEM Vehicle Emission TestResults

The OEM vehicles tested wereethanol and methanol FFVs, dedi-cated CNG vehicles, and similar con-trol vehicles running on RFG. Tocompare emission levels at variousfuel blends, FFVs were tested onthree fuels: a mixture containing85% alcohol and 15% RFG, a mix-ture containing 50% alcohol and 50%RFG, and 100% RFG. Vehicles thatoperate on CNG are tested on CNG,and the results are compared to simi-lar gasoline vehicles that run on RFG.All the test fuels used in the emis-sions testing were specifically blended for this program. The resultspresented in this report are based

14


Table 6. Vehicles with Extensive Emissions Testing Completed

Vehicle Model Year Model Type Number Number of Vehicles of Tests

Chevrolet Lumina 1991 Standard 8 25

1992 E85 flexible-fuel 13 59

1993 E85 flexible-fuel 12 42Standard 16 23

Dodge B250 van 1991 Dedicated CNG 2 2

1992 Dedicated CNG 36 50Standard 22 25

1994 Dedicated CNG 14 14Standard 25 35

Dodge Spirit 1993 M85 flexible-fuel 76 319Standard 72 126

Ford Econoline van 1992 M85 flexible-fuel 13 53

1993 M85 flexible-fuel 3 9Standard 18 23

on 548 emissions tests conducted on 169 AFVs, and 257 emissions testsconducted on 161 gasoline controlvehicles. These totals include tests on multiple fuels, and repeat tests on individual vehicles.

Emissions testing has been con-ducted on several vehicle models,but the results presented in thisreport are only for the models withfairly extensive testing completed.Table 6 lists the vehicles included inthis analysis and the number of emis-sions tests completed. The vehiclesinclude Dodge Spirits (M85),Chevrolet Luminas (E85), FordEconoline vans (M85) and Dodge B250 vans (CNG). The averageexhaust emissions for NMHC,CO, NOx, and CO2 are shown in Figures 9–12 for each model.

The results indicate that the alcohol-fueled vehicles in general showreduced or equivalent emissions ofregulated compounds compared to RFG, although they do have otherenvironmental benefits discussedlater in this section (see Figures9–11). The ethanol Lumina had the largest reductions of the alcoholvehicles, averaging about 15%–20%lower regulated emissions than RFG.

The CNG vehicles showed signifi-cantly lower emissions of the regu-lated exhaust pollutants than did the standard gasoline vehicles (see Figure 12). NMHC emissions wereapproximately 64% lower, COapproximately 43% lower, and NOxabout 31% lower than the levelsobserved for the gasoline vehicles.

A more thorough analysis of the HCemissions to determine the toxicityand smog-forming potential of the

various constituent compounds yielded some important results. TheHC profiles for the alternative fuelswere much different than those forRFG. For example, HC toxins, suchas benzene and 1,3-butadiene wereas much as 80% lower for alcoholfuel tests and approximately 95% lower for CNG tests than forthe same vehicle on RFG. On theother hand, formaldehyde andacetaldehyde emissions wereincreased for methanol and ethanol,respectively, and were generallylower for CNG. Again, these results are all compared to RFG.

15


Figure 9. Emissions results fromDodge Spirits

420

400

380

360

340

320

300CO2NMHCNOxCO

CO2

Emis

sion

s ra

te (g

/mile

)

Figure 10. Emissions results fromFord Econoline vans

620

600

580

560

540

520

500CO2NMHCNOxCO

CO2

Emis

sion

s ra

te (g

/mile

)

HC analysis also allows a relativecomparison between fuels on thepotential for exhaust HCs to react in the atmosphere to form ozone,more commonly known as smog. Thepreliminary results indicate that theozone-forming potential was reducedfor each alternative fuel compared tovehicles running on RFG. The resultsare shown in Figure 13. Ozone-forming potential was reducedapproximately 25% for the ethanol

vehicles, up to approximately 50% forthe methanol vehicles, and approxi-mately 80% for the CNG vehicles.

Carbon dioxide is not a regulatedexhaust emission, but studies havelinked CO2 emissions to the green-house effect and global climatechange. Figures 9–12 show that allalternative fuels tested producedlower CO2 emissions.

Conversion Vehicle EmissionsTest Results

Limited emissions testing was com-pleted on gasoline vehicles convertedto operate on CNG and LPG. Thir-teen CNG conversions and threeLPG conversions were tested. Thesame kind of testing as describedabove for the OEM vehicles wasconducted on the conversions. Eachvehicle was tested on RFG beforeconversion, then emissions tested on RFG and either CNG or LPG following conversion.

Emissions levels of all three regulat-ed emissions (NMHC, CO, and NOx)were either improved or unchangedon only two of the 16 vehicles whenrunning on the alternative fuel. CNGconversions generally showed a sig-nificant reduction in NMHC emis-sions, but an increase in CO or NOx,or both. The emissions results on the CNG conversions are shown inTable 7. These results contrast withthose from the OEM CNG modeldiscussed above, in which substantialacross-the-board emissions benefitswere realized. The three LPG conver-sions tested showed increased emis-sions on gasoline after conversion,and mixed results on LPG. Theincreased emissions on gasoline after

16


Figure 11. Emissions results fromChevrolet Luminas

480

460

440

420

400

380

360CO2NMHCNOxCO

CO2

Emis

sion

s ra

te (g

/mile

)

Figure 12. Emissions results fromDodge B250 vans

680

660

640

620

600

580

560

540

520

500

480CO2NMHCNOxCO

CO2

Emis

sion

s ra

te (g

/mile

)

conversion are likely a result of thekit design or installation rather thanof the fuel. No comparison betweenCNG and LPG is possible becausethe conversion kits used for CNG andLPG were dramatically different indesign and operating principles.

Carbon dioxide emissions decreasedby approximately 20% for CNG and10% to 15% for LPG, similar to the OEM vehicles. Ozone-formingpotential was not calculated for thesevehicles because of insufficient data.

Finally, the emissions testing, and theresultant effect on the environment, isa complex and evolving science. EPArecently implemented or proposedvarious changes to its standard testprocedures to add cold temperature

17


Figure 13. Ozone-forming potential of the various AFVs

CNG

Dodg

eB2

50 V

an

E85

Chev

role

t

Lum

ina

Chan

ge in

Ozo

ne-F

orm

ing

Pote

ntia

lCo

mpa

red

to R

FG

0%

-20%

-40%

-60%

-80%

-100%

M85

Dod

geSp

irit

M85

For

dEc

onol

ine

Table 7. Emissions Test Results from CNG Conversions

Vehicle Model After Conversion (CNG)Model Year NOx CO NMHC

Acclaim 1992

Acclaim 1992

Astro 1992

Caravan 1992

Caravan 1992

Safari 1993 NC

Safari 1993

Taurus 1994 NC

Taurus 1994 NC

B250 1994

B250 1994

C1500 1994

C1500 1994

Moderate emissions decrease (10%–50%)

NC = No change (i.e., less than 10%)

Large emissions increase (>50%)

Moderate emissions increase (10%–50%)

Large emissions decrease (>50%)

testing and higher acceleration test-ing, two areas where gaseous-fueledvehicles are expected to excel. Test-ing is now being done to evaluate theperformance of all the alternativefuels using these new procedures.

In summary, the test results showthat AFVs have the potential to sig-nificantly reduce exhaust emissionsand their impact on the environment,but this is not guaranteed. The tech-nology used is as important as thefuel used, and fleets should be care-ful in their choice of vehicles if amain objective is to improve theenvironment.

Because of the disappointing perfor-mance of the aftermarket conversionkits cited, fleets considering conver-sions should require that they meetEPA’s new standards. (Standards forEmissions From Natural Gas-Fueled,and Liquefied Petroleum Gas-FueledMotor Vehicle Engines and Certifica-tion Procedures for AftermarketConversions, Federal Register,September 1994.)

Cost

Major vehicle costs associ-ated with owning andoperating any light-dutyvehicle include the depre-ciation, fuel and oil, inter-est and fees, maintenanceand tires, and insurance. The distribution of costsfor operating a gasoline-fueled fleet vehicle aredepicted in Figure 14. Thisdistribution represents cost

during the life of a vehicle, and is a

composite average of fleets acrossthe United States. These costs arebased on a vehicle being in servicean average of 27 months and havingan average acquisition cost of$15,000. Differences such as self-insurance for government and someprivate fleets, and the methods oflogging some types of maintenanceand repairs, will affect the distribu-tion of vehicle costs.

Costs associated with owning andoperating vehicles significantly affectthe likelihood that fleets or individu-als will buy them. Distributions ofcost are likely to be different for afleet of AFVs than for a fleet ofgasoline vehicles. Differences in ini-tial acquisition costs (and therefore,depreciation), fuel costs, and mainte-nance are expected to affect theoperation cost distribution for AFVs.Information on differences in AFVoperation costs, including acquisi-tion, fuel, and maintenance, are dis-cussed in the sections that follow. A sample analysis that compares anAFV and gasoline vehicle operatingcosts is also included.

Additional Acquisition Cost

The information available to date onthe AFVs in the Federal light-dutyvehicle evaluation program indicatesthat AFV initial acquisition costranges up to 25% more than compa-rable gasoline vehicles. The differ-ence is primarily due to modificationsrequired to enable a gasoline modelvehicle to operate on an alternativefuel, and the limited production ofthese vehicles. If more AFVs wereproduced, their initial cost wouldlikely drop.

18


Figure 14. Distribution of operatingcosts for composite U.S. fleet

Source: PHH Vehicle Management Services

Price increases range from $0 to$800 for the FFVs, depending on the manufacturer. Most of the costincrease is due to the special fuelsystem materials required for thealcohol fuels.

For CNG vehicles, the acquisitioncost has been as much as $5,000more than a similar gasoline modelsince the Federal government startedincorporating AFVs into its fleet.The bulk of the price increase is theresult of the different fuel storageand delivery systems necessary toaccommodate a gaseous fuel. As thedemand for CNG vehicles increasesproduction numbers, and manufac-turers continue to optimize theirdesigns, the price differential isexpected to decrease. For example,one automobile manufacturer hasannounced a reduction of $750 in theCNG option price on its most popu-lar CNG models in 1996.

The cost of aftermarket conversionsof light-duty vehicles depends on thealternative fuel, the level of conver-sion technology, and the number andsize of fuel tanks. The average totalcost for each CNG conversion in ourprogram has been about $4,500.Each propane conversion has runabout $2,800. The CNG conversioncost is comparable to the premiumcharged for the CNG option packageon an OEM vehicle. The biggest costfactor in CNG and propane conver-sions is the number and size of thefuel tanks that are installed. The fuel tanks used in each conversiondepend on the vehicle type and thedesired vehicle range.

As with standard gasoline models,acquisition price for any AFV willvary depending on the vehicle modeland options. Although an unknownat this point, the resale market isexpected to grow as the AFV marketincreases, thus enabling recovery ofsome of the additional acquisitioncosts.

Fuel and Oil Costs

Fluctuations in wholesale prices and differences in state and local taxstructures result in a wide variationin retail fuel costs across the country.Wholesale fuel prices vary as muchas $0.20 per equivalent gallon in var-ious cities across the country. Stateand local taxes increase the variationin retail fuel prices to as much $0.40per equivalent gallon of fuel acrossthe United States. Some states andcommunities offer incentives to pur-chasers of alternative fuels thatinclude reduced tax rates, taxingonly the gasoline portion of an alcohol fuel, and taxing all fuels at the same rate as gasoline.

In early 1996, retail (pump) priceswere approximately $1.62 to $1.72per gallon of 85% ethanol (averageprice: $1.66/GGE); $1.73 to $2.69per gallon of 85% methanol (averageprice: $2.00/GGE); $0.58 to $1.05per gallon of CNG (average price:$0.86/GGE); and $1.06 to $1.18 forregular unleaded gasoline (averageprice: $1.13/gal.). All prices reflectconversion to GGE (or for an amountwith the same energy content as onegallon of gasoline). A number ofindustry sources, including Oxy-FuelNews, 21st Century Fuels, and

19


New Fuels Report, track the whole-sale and retail prices of the variousalternative motor fuels. DOE’s Energy Information Administrationalso produces weekly and monthlypublications with information ongasoline prices across the country.

Variations in fuel costs, combinedwith variations in fuel economy forthe various vehicles, result in a widerange of fuel costs per mile driven.The ranges of fuel costs per mile forthe AFVs and gasoline vehicles inthe Federal test fleet are shown inFigure 15. The fuel cost per mileranges were calculated using thevariations in fuel price per equivalentgallon and range of fuel economy ofeach vehicle in the program. Theaverage range of fuel cost per milefor ethanol and methanol sedans wassomewhat higher than that for gaso-line. Chevrolet Lumina and FordTaurus sedans operated on ethanol;Luminas, Taurus, Dodge Spirit, and

Dodge Intrepid sedans operated onmethanol; and Lumina, Taurus, andSpirit sedans operated on gasoline.For vans, CNG is comparable togasoline in range of fuel costs permile, but methanol is somewhathigher. Vans in this analysis included Ford Econolines operating onmethanol, Dodge Caravans and Ramvans operating on CNG, and FordEconolines and Dodge Ram vansoperating on gasoline. The actualcost per mile for any specific vehiclewill be affected by factors such asthe driving cycle of the vehicle, thedriving style of the operator, theactual mpg level achieved by a vehicle, and the local cost of fuel.

Oil cost depends on the price and fre-quency of oil changes. Automobilemanufacturers typically recommendoil change intervals of 7,500 miles or 6 months (whichever occurs first)for gasoline-fueled vehicles. Becausealcohol fuels are corrosive, manufac-turers recommend an oil changeinterval of 5,000 miles or 6 months(whichever occurs first) for FFVsoperated on alcohol fuels. Because of these more frequent oil changes(requiring a special oil that costsmore), typical oil costs are approxi-mately 0.5 cents per mile for alcohol-fueled vehicles. This is higher thanthe 0.3 cents per mile cost for gaso-line vehicles, based average oilchange prices. Manufacturers recom-mend the same oil change interval forCNG-fueled vehicles as they recom-mend for gasoline-fueled vehicles.Therefore, assuming comparableprices, these vehicles will have similar oil costs. All vehicles in theFederal light-duty evaluation program

20

Figure 15. Range of fuel costs per mile (alternate fuel costs arebased on GGE)


have followed the manufacturer-recommended oil change intervals.

Maintenance Costs

The availability of maintenance costinformation has been affected by anumber of factors. First, much of themaintenance on the Federal test vehi-cles has been done under warranty atno cost to the fleet operator (exceptfor lost time in service). Manyunscheduled repairs on early AFVswere due to the developmental statusof the fuel and control systems.Many fuel injection, fuel filter, andfuel pump problems in the 1991 M85Luminas and Tauruses, and fuelinjection, pressure regulator, line,and valve problems in the 1991 and1992 CNG-fueled vehicles attest tothis status. In addition, the increasingnumber and variety of vehicles in thestudy fleet made it difficult to collectdetailed maintenance cost informa-tion. Finally, it is generally desirableto have large populations to calculatemeaningful averages for maintenancecosts. Because of these factors, dataon the actual cost of maintenance,and a summary of maintenance costper mile, are not available.

However, based on the experiences of this study, maintenance costs areexpected to be marginally higher forAFVs than for gasoline vehicles forseveral reasons. Some parts costmore for vehicles in limited produc-tion, which is still the case for mostAFV models. Some maintenanceproblems, and therefore costs, areunique to AFVs. For instance,although fuel pump and injectorproblems were common in the earlymodels, they have decreased as man-ufacturers gained experience and

improved AFV designs. Scheduledmaintenance costs, specifically for oilchanges, will continue to cost some-what more for alcohol-fueled vehi-cles, because of the more frequentrecommended oil change schedule.

The analysis of available mainte-nance data focused on the CNG-and methanol-fueled vehicles,because we lacked adequate data to analyze ethanol vehicles. Theanalysis focused on the number of unscheduled repairs as opposed to cost of repairs, for the reasonsdescribed above. The repair trendsfor CNG and M85 vehicles (Figure 8, page 11) clearly indicatethat unscheduled maintenance hasdecreased with the newer modelyears. These trends and the growingexperience with AFVs increase confidence that their long-term maintenance costs should, in time,approach the levels of gasoline vehicles.

Sample Operating Cost Comparison

At this point, we do not have enoughcost information to do a detailed costanalysis based on the study vehiclesin the Federal fleet. However, wehave included the results of a samplecalculation that includes the majorvehicle operation cost factors andprovides a first-order comparison ofoperating costs between an AFV anda gasoline-fueled vehicle. This sam-ple analysis was provided by PHHVehicle Management Services,whose expertise includes fleet datacollection and analysis.

The analysis provides a simple com-parison between annual operating

21


costs of an AFV and a gasoline vehi-cle, including the primary economicfactors that a fleet operator wouldneed to consider. CNG is used as thealternative fuel in this sample analy-sis. The assumptions used in thisanalysis are listed in Table 8 (insurance, interest, fees, taxes, andincidentals are not included). Nocosts are included for a site requiringa refueling station. Table 9 shows theresulting annual operating cost for agasoline and a CNG vehicle withmonthly mileage accumulations of2,100; 3,000; and 3,500 miles. Themaintenance costs were assumed to be the same for both fuels. Basedon the performance and reliabilitytrends in the Federal fleet, thisassumption appears reasonable.Table 10 shows the results when the price of gasoline is increased to $1.25 per gallon. The economicsare highly dependent on the price ofgasoline, and can be affected by thevehicle’s fuel economy. In general,as the mileage increases, CNG-fueled vehicles become more eco-nomical. Similar comparisons couldbe done for other alternative fuels bysubstituting the appropriate vehicleand fuel cost information andassumptions.

As described, this sample analysisprovides a simple first-order compar-ison of AFV and gasoline vehiclecosts, including the primary factorsthat affect fleet economics. Clearlythere are many other factors to con-sider when adding AFVs to a fleet.For instance, economies of scale,which would affect vehicle purchaseprice and other assumptions, may beavailable to a fleet. The fueling infra-structure can significantly affect cost.

22

Table 8. Assumptions Used in Sample Analysis

Vehicle Acquisition Information

Capital costs (gasoline) $14,000

Alternative fuel—CNG option cost $4,000

Alternative fuel capital costs $18,000

Assumptions

Gasoline price (per gallon) $1.19

Miles per gallon rating (mpg) 23

Alternative fuel—equivalent gallon price $0.80

Expected resale value ~40%*

Vehicle life 36 months

* Resale value decreases with increased vehicle mileage, approximately $300 per additional 5,000 miles

Table 9. Annual Operating Costs (with gasoline at $1.19 per gallon)

2,100 mi/mo 3,000 mi/mo 3,500 mi/mo

Gasoline vehicle

Fuel $1,303.83 $1,862.61 $2,173.04

Maintenance $550.00 $550.00 $550.00

Depreciation $2,800.00 $3,000.00 $3,100.00

Annual Total $4,653.83 $5,412.61 $5,823.04

Cost per Mile 18.5 cents 15.0 cents 13.9 cents

CNG vehicle

Fuel $876.52 $1252.17 $1460.86

Maintenance $550.00 $550.00 $550.00

Depreciation $3,600.00 $3,800.00 $3,900.00

Annual Total $5,026.52 $5,602.17 $5,910.86

Cost per Mile 19.9 cents 15.6 cents 14.1 cents

Differential Cost per Mile of a CNG Vehicle 1.4 cents 0.6 cents 0.2 cents


Infrastructure costs include every-thing from building a fueling stationto personnel and operational readi-ness for alternative fuels. The fuelingfacility costs will vary with the fuelused and how it is provided (e.g.,on-site central fuel station). A smallportable ethanol or methanol stationmay cost a few thousand dollars; alarge permanent CNG refueling sta-tion may cost hundreds of thousandsof dollars. Personnel costs are gener-ally associated with driver training,maintenance, and security, and staffmember safety. Operational readi-ness includes vehicle storage. Indoorstorage of CNG vehicles may re-quire building modifications such as upgrades to ventilation systems.These are all examples of other considerations that are part of theeconomics of incorporating AFVsinto a fleet. A more detailed study of fleet economics, based on resultsof the CleanFleet project (see ForAdditional Information on page 27),presents a fairly comprehensive con-sideration of the factors that affectAFV costs for a fleet.

Government incentives can signifi-cantly reduce the incremental costsof AFVs. Under the Energy PolicyAct of 1992, the Federal governmentallows a maximum tax deduction of$2,000 for the incremental costs ofAFVs up to 10,000 lb gross vehicleweight. Many states offer incentivesfor purchasing dedicated or con-verted AFVs, including income taxcredits, special or reduced fuel excisetax on the alternative fuel, reducedsales tax on fuel and conversionequipment, rebates on costs of con-verting vehicles, and low-interestloans for purchasing AFVs. Fleet

operators should contact their stateenergy office for information onincentives available in their state.

Other Information

Data Collection Experience

This data collection project has beenextremely challenging, and has pro-vided many lessons about how futureprograms should be designed. It wasdifficult to keep drivers interested in filling out data cards, particularlyfor the extended time periods werequired, and some organizations andmanufacturers have resisted disclos-ing information. Federal employeeswere burdened with completing datacollection cards, with no incentivefor the extra effort. On the other

23

Table 10. Annual Operating Costs (with gasoline at $1.25 per gallon)

2,100 mi/mo 3,000 mi/mo 3,500 mi/mo

Gasoline vehicle

Fuel $ 1,369.57 $1,956.52 $2,282.61

Maintenance $550.00 $550.00 $550.00

Depreciation $2,800.00 $3,000.00 $3,100.00

Annual Total $4,719.57 $5,506.52 $5,932.61

Cost per mile 18.7 cents 15.3 cents 14.1 cents

CNG vehicle

Fuel $876.52 $1252.17 $1460.86

Maintenance $550.00 $550.00 $550.00

Depreciation $3,600.00 $3,800.00 $3,900.00

Annual Total $5,026.52 $5,602.17 $5,910.86

Cost per mile 19.9 cents 15.6 cents 14.1 cents

Differential Cost per 1.2 cents 0.3 cents 0.0 centsMile of a CNG Vehicle


hand, we experienced extraordinaryparticipation from a number of sites,including some that provided datavoluntarily. The project size and vari-ety of vehicles made focusing on acontrolled set of data collection vehi-cles and a consistent data collectionmethod difficult. Vehicles were oper-ated under random driving patterns,in various areas across the country.This dramatically increased the effortrequired to achieve continuity amongdetails of data collection.

Some early prototype AFVs wereplaced in service while still under-going development. These vehicleshad reliability problems in service:even though successive models weregreatly improved, the earliest impres-sions of AFVs lingered. Failure inservice ranged from vehicles runningout of fuel in the California desertbecause of faulty fuel gauges, tovehicles not starting in less than idealurban neighborhoods. Some agenciesare reluctant to operate AFVsbecause of these early experiences.

Finding FFVs that were actuallybeing operated on an alternative fuel was difficult, partly because of the lack of infrastructure, whichresulted in the alternative fuel refill-ing stations being less convenientlylocated than gasoline stations. And,as mentioned earlier, many vehiclesin this program accumulated mileagevery slowly.

Our experiences in these data collec-tion efforts have led to a number ofideas for changing future data collec-tion efforts. We believe the followingmodifications would improve futuredata collection efforts:

• Smaller fleet size (< 50 vehicles,especially for detailed data collec-tion)

• Match vehicle range and utility tofleet needs (where possible)

• Establish agreements with mainte-nance and repair shops to obtaindetailed repair records and costinformation

• Work with sites that have conve-nient refueling

• Work with fleets that accumulatemileage rapidly, to reduce the timeneeded to collect data.

These are some of the criteria weplan to include in our future data collection efforts. The ultimate goalcontinues to be high-quality data onAFV performance and characteris-tics, with evaluation results availablein a timely manner.

Other Information Sources

To enable access to data being collected in the light-duty AFV program, and other alternative fuel demonstration programs,DOE established the AFDC atNREL. The AFDC stores all thelight-duty vehicle data in a database,and information is available to any-one, free of charge. The AFDC canbe accessed via the World Wide Webon the Internet, using browser soft-ware such as Mosaic or Netscape (at http://www.afdc.doe.gov).

The National Alternative Fuels Hotline is a DOE source of informa-tion concerning alternative fuels and related issues. The Hotline staff can answer questions aboutalternative fuels, process requests

24


for information, or refer callers toother data sources or organizations.The Hotline can be reached at 1-800-423-1DOE (1363).

Additional, more detailed reports areavailable on the OEM and conversionemissions results. Three Society ofAutomotive Engineers (SAE) paperson OEM emissions results, plus amore detailed report on the conver-sion experiences, are available. Callthe Hotline at 1-800-423-1DOE, oraccess our World Wide Web site athttp://www.afdc.doe.gov.

Acknowledgment

NREL wishes to extend thanks to themany Federal government agencies,and particularly the drivers of thevehicles in our study fleet, for theirparticipation in this vehicle evalua-tion effort. NREL would also like to acknowledge Patrick McConnelland Bill Rivers of GSA in Washington, D.C., and other regionalGSA personnel who have provided awide range of information on theAFVs in the GSA fleet, includingsharing vehicle maintenance records.

25


27

For Additional Information. . .

Visit our World Wide Web site at http://www.afdc.nrel.gov or call the National Alternative Fuels Hotline at 1-800-423-1DOE. Selected citations are shown below as examples of the types of information you can find on the Web or through the Hotline (refer to AFDC accession numbers, NREL numbers, or SAE numbers when call-ing the Hotline; Web searches can be performed by key word, author, or title, for example).

• Motta, R., Kelly, K., and Warnock, W., April 1996, Compressed Natural Gas and Liquefied Petroleum Gas Conversions: The National Renewable Energy Laboratory's Experience, NREL/SP-425-20514, Golden, CO:prepared by NREL for DOE.

• Federal Alternative Motor Fuels Programs: Fourth Annual Report to Congress, July 1995, DOE/GO-10095-150, AFDC accession number 2892, Golden, CO: published by NREL for DOE.

• Reports on the Federal Express CleanFleet demonstration (series), November 1993–December 1995, AFDCaccession numbers 2555, 2560, 2562, 2837, 2864, and 3035, published by Battelle Memorial Institute.

• Kelly, K., Bailey, B., Coburn, T., Clark, W., Eudy, L., and Lissiuk, P., May 1996, FTP Emissions Test Resultsfrom Flexible-Fuel Methanol Dodge Spirits and Ford Econoline Vans, SAE Technical Paper Series #961090,Warrendale, PA: SAE.

• Kelly, K., Bailey, B., Coburn, T., Eudy, L., and Lissiuk, P., May 1996, Round 1 Emissions Test Results from Compressed Natural Gas Vans and Gasoline Controls Operating in the U.S. Federal Fleet, SAE TechnicalPaper Series #961091, Warrendale, PA: SAE.

• Kelly, K., Bailey, B., Coburn, T., Clark, W., and Lissiuk, P., May 1996, Federal Test Procedure Emissions Test Results from Ethanol Variable-Fuel Vehicle Chevrolet Luminas, SAE Technical Paper Series #961092,Warrendale, PA: SAE.

• AFDC Update: News of the Alternative Fuels Data Center, Volume 4, Issue 4, March 1996, AFDC accessionnumber 3041, Golden, CO: published by NREL and Information Resources, Inc., for DOE.

Produced for the Office of Transportation Technologiesof the U.S. Department of Energy (DOE)

by the National Renewable Energy Laboratory, a DOE national laboratory

May 1996NREL/SP-425-20821

DE96000542

T OF ENERGYD

EPA

RTMEN

U

E

NIT

ED

STAT S OFA

ER

ICA

M

Printed with a renewable-source ink on paper containing at least 50% wastepaper, including 20% postconsumer waste

Documents

alternative fuel light-duty vehicles - Energy.govAlternative Motor Fuels Act (AMFA) was enacted in 1988 to encourage the development and use of alternative transportation fuels (including