7404 Federal Register /Vol. 60, No. 25/Tuesday, February 7 ... · PDF file7404 Federal Register/Vol. 60, ... speed and/or high acceleration) driving ... No. 25/Tuesday, February 7,

7404 Federal Register / Vol. 60, No. 25 / Tuesday, February 7, 1995 / Proposed Rules

1 Regulatory Impact Analysis for FTP Revisions,U.S. EPA, Office of Air and Radiation. Available inthe public docket for review.

ENVIRONMENTAL PROTECTIONAGENCY

40 CFR Part 86

[FRL–5150–1]

RIN 2060–AE27

Proposed Regulations for Revisions tothe Federal Test Procedure forEmissions From Motor Vehicles

AGENCY: Environmental ProtectionAgency (EPA).ACTION: Notice of proposed rulemaking(NRPM).

SUMMARY: This notice proposesadditions and revisions to the tailpipeemission portions of the Federal TestProcedure (FTP) for light-duty vehicles(LDVs) and light-duty trucks (LDTs).The primary new element of theproposal is a Supplemental Federal TestProcedure (SFTP) designed to addressshortcomings with the current FTP inthe representation of aggressive (highspeed and/or high acceleration) drivingbehavior, rapid speed fluctuations,driving behavior following startup, airconditioning, and intermediate-durationperiods where the engine is turned off.An element of the SFTP that also affectsthe conventional FTP is a new set ofrequirements designed to moreaccurately reflect real road forces on thetest dynamometer. The Agency is alsoproposing new emission standards forthe new control areas with a specifiedphase-in period for these standards.After complete fleet turnovers, thestandards proposed today are estimatedto reduce emissions from LDVs andLDTs by eight percent for non-methanehydrocarbons (NMHC), 18 percent forcarbon monoxide (CO), and 14 percentfor oxides of nitrogen (NOX).DATES: Written comments on this NPRMmust be submitted on or before 30 daysafter the public hearing date. TheAgency will conduct a public hearingon this NPRM approximately March 24,1995. The date of the public hearingwill be published in a future FederalRegister document.ADDRESSES: Interested parties maysubmit written comments (in duplicateif possible) to Public Docket No. A–92–64, at: Air Docket Section, U.S.Environmental Protection Agency, 401M Street SW, Washington, DC 20460.The public hearing will be held at alocation to be published in a futureFederal Register document.

Materials relevant to this proposedrulemaking have been placed in DocketNo. A–92–64. The docket is located atthe above address in Room M–1500,Waterside Mall, and may be inspected

weekdays between 8:30 a.m. and 5:30p.m. A reasonable fee may be chargedby EPA for copying docket materials.FOR FURTHER INFORMATION CONTACT: JohnGerman, Certification Division, U.S.Environmental Protection Agency,National Vehicle and Fuel EmissionsLaboratory, 2565 Plymouth Road, AnnArbor, Michigan, 48105. Telephone(313) 668–4214.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Obtaining Copies of the RegulatoryLanguage

II. IntroductionIII. Proposal Requirements and Alternative

ApproachesIV. Statutory Authority and Legal AnalysisV. The FTP Review Project and Areas of EPA

ConcernVI. In-Use BehaviorVII. Representative Driving CyclesVIII. Emission Inventory AssessmentsIX. Cause and Control of EmissionsX. Other Options and Information NeededXI. Environmental and Economic ImpactsXII. Public ParticipationXIII. Administrative DesignationXIV. Regulatory Flexibility ActXV. Reporting and Recordkeeping

Requirement

I. Obtaining Copies of the RegulatoryLanguage

Electronic copies (on 3.5′′ diskettes) ofboth the proposed regulatory languageand the Support Document to theProposed Regulations for Revisions tothe Federal Test Procedure: DetailedDiscussion and Analysis, RegulatoryImpact Analysis (RIA), and TechnicalReports may be obtained free of chargeby visiting, calling, or writing theEnvironmental Protection Agency,Certification Division, 2565 PlymouthRoad, Ann Arbor, MI 48105, (313) 668–4384. Refer to Docket A–92–64. A copyis available for inspection in the docket(See ADDRESSES).

The proposed regulatory language andthe Support Document to the ProposedRegulations for Revisions to the FederalTest Procedure: Detailed Discussion andAnalysis, RIA, and Technical Reportsare also available electronically on theTechnology Transfer Network (TTN).TTN is an electronic bulletin boardsystem (BBS) operated by EPA’s Officeof Air Quality Planning and Standards.Users are able to access and downloadTTN files on their first call. The stepsrequired to access information on thisrulemaking are listed below. The serviceis free, except for the cost of the phonecall.TTN BBS: 919–541–5742 (1,200–14,400

bps, no parity, eight data bits, onestop bit)

Voice help: 919–541–5384

Internet address: TELNETttnbbs.rtpnc.epa.gov Off-line:Mondays from 8:00–12:00 Noon ET

1. Technology Transfer Network TopMenu: <T> GATEWAY TO TTNTECHNICAL AREAS (BulletinBoards) (Command: T)

2. TTN TECHNICAL INFORMATIONAREAS: <M> OMS—Mobile SourcesInformation (Command: M)

3. OMS BBS === MAIN MENU FILETRANSFERS: <K> Rulemaking &Reporting (Command: K)

4. RULEMAKING PACKAGES: <1>[Light-Duty] (Command: 1)

5. Light-duty Rulemaking Area: File area#1 ... FTP Review (Command: 1)

At this stage, the system will list allavailable FTP Review files. Todownload a file, select a transferprotocol which will match the terminalsoftware on your computer, then setyour own software to receive the fileusing that same protocol.

If unfamiliar with handlingcompressed (that is, ZIP’d) files, go tothe TTN top menu, System Utilities(Command: 1) for information and thenecessary program to download in orderto unZIP the files of interest afterdownloading to your computer. Aftergetting the files you want onto yourcomputer, you can quit TTN BBS withthe <G>oodbye command.

II. Introduction

Automobiles are among the largestproducers of hydrocarbons (HC), carbonmonoxide (CO), and oxides of nitrogen(NOX), all of which have documentedimpacts on public health. Hydrocarbonsand oxides of nitrogen contribute to theformation of ozone, a powerful oxidantwhich irritates the respiratory systemand reduces lung function. Somestudies indicate that ozone maypermanently damage lung and othertissues. Elevated levels of CO decreasethe ability of blood to transport oxygenthroughout the body, which tends toexacerbate cardiovascular stress. Highambient levels of CO can also adverselyaffect the central nervous system, andthe presence of CO in even moderatelevels in the bloodstream may impactthe health of fetuses and newborns.1After complete turnover of the fleet, theAgency believes that the changesproposed today would result in an eightpercent reduction in non-methanehydrocarbons (NMHC), an 18 percentreduction in CO, and a 14 percentreduction in NOX emissions from

7405Federal Register / Vol. 60, No. 25 / Tuesday, February 7, 1995 / Proposed Rules

2 See the ‘‘Federal Test Procedure Review Project:Preliminary Technical Report,’’ EPA 420–R–93–007and the Technical Reports for this rulemaking, bothin the public docket, for descriptions of the surveysand data gathered.

automobiles during typical summertimeozone exceedance days.

The Agency has established a numberof emission standards for motor vehiclesand engines, designed to control airpollution by reducing in-use emissionsfrom motor vehicles. Compliance withthese standards is typically measuredusing a test procedure that simulates in-use driving. In 1990, Congress amendedthe Clean Air Act with passage of theClean Air Act Amendments (hereafter,CAAA or Amendments) and requiredthat EPA review these test proceduresand revise them as appropriate to reflectin-use conditions. The Agency’s reviewfocused on the procedures for light-dutymotor vehicles, especially the FederalTest Procedure (FTP), the procedureused to measure compliance with motorvehicle tailpipe and evaporativeemission standards.

The Agency, in conjunction withautomobile manufacturers andCalifornia’s Air Resources Board

(CARB), conducted an extensive reviewof in-use driving behavior, obtaining awealth of data on how cars are drivenduring trips, the length of trips, thelength of time between trips, and so on.2The Agency then generatedrepresentative driving cycles from thedata and conducted emission testing tocompare emissions over these cycleswith emissions over driving cycles usedin the FTP. These results confirmed thatrevisions to the FTP were needed, assignificant emissions were seen underconditions not represented by thecurrent FTP.

The Agency sought an approachwhich would extend the level of controlfound under current FTP conditionsacross all in-use driving behavior. Thus,

EPA developed various changes to theFTP, focusing on new driving cycles toadd to the current FTP. The Agency alsoinvestigated possible controltechnologies that could be used tocontrol emissions over these newcompliance cycles. Today’s proposalincludes these various changes in thetest procedure for tailpipe emissions, aswell as the emission standards related tothem.

In developing new compliance cycles,EPA did not re-evaluate the stringencyof current standards. Rather, EPA soughtparity between the types and extent ofcontrols that manufacturers currentlyemploy to comply with existing FTPstandards and those they wouldimplement to comply across all drivingbehavior. Thus, EPA believes thatmanufacturers for the most part willcomply by making simple changes totheir existing calibration strategies.

BILLING CODE 6560–50–P


BILLING CODE 6560–50–C


3 The Agency has historically relied on emissionperformance standards because they directly limitproduction of exhaust constituents that affectattainment of the National Ambient Air QualityStandards, while providing maximum flexibility tothe vehicle manufacturers in determining cost-effective compliance strategies. Other basiccompliance program approaches include systemperformance standards, which set bounds onmeasurable performance parameters of the engineor emission control system rather than actualemission levels, and design standards, whichprescribe primary design elements of the engine orcontrol system.

4 Road load forces refers to the force needed toovercome wind and tire resistance when driving atspecific speeds.

The FTP is the core procedure used tomeasure compliance with emissionstandards for light-duty vehicles (LDVs)and light-duty trucks (LDTs). Thecurrent version of the FTP (40 CFR86.130–96) consists of a series ofpreparatory steps to ensure the vehiclehas been properly preconditioned onthe test fuel, periods when the engine isoff between vehicle operation (called‘‘soaks’’), and emission tests whichmeasure tailpipe and evaporativeemissions. Tailpipe emissions aremeasured while the vehicle is operatedaccording to a specified driving cycle ona dynamometer. Figure 1 presents theUrban Dynamometer Driving Schedule,commonly referred to as the LA4. Withthe exception of running losses, whichare measured during dynamometeroperation, evaporative emissions aremeasured in a sealed enclosure whilethe vehicle is turned off. An additionalcold temperature CO test proceduremeasures tailpipe emissions at 20° Ffollowing a cold soak. By comparing theemission test results to emissionstandards applicable to a given vehicleclass, combustion cycle, and motor fuel,EPA determines if the vehicle meetsapplicable certification or in-userequirements.3

The current evaporative emissionprocedure, including refueling, and coldtemperature CO test procedures werepromulgated following passage of theAmendments. Thus, the test proceduresin these rules were recently developedto reflect the actual current drivingconditions under which motor vehiclesare used (57 FR 31888; 58 FR 16002).The Agency is not proposing to changethese test procedures and the remainderof this section and the subsequentproposal focuses on the light-dutytailpipe emission testing procedures ofthe FTP.

The FTP simulates on-road vehicleoperation using a dynamometer in alaboratory test cell held between 68° Fand 86° F. The vehicle is driven on thedynamometer over cycles that prescribethe vehicle operator’s speed as afunction of time. The method formeasuring tailpipe emissions of HC, CO,and NOx requires filling a bag with

exhaust drawn from the tailpipe anddiluted with background air while thevehicle is driven over the appropriatecycle. The bagged sample is analyzedfor the concentrations of exhaustconstituents, which serve as inputs tosubsequent emission compliancecalculations. Additional proceduresapply to the sampling of particulatematter from diesel-cycle vehicles andorganic gases from alternative-fuelvehicles.

III. Proposal Requirements andAlternative Approaches

Today’s proposal deals primarily withfive areas of driving behavior that havenot previously been represented in thetest procedure: aggressive drivingbehavior (such as high acceleration ratesand high speeds); rapid speedfluctuations (microtransient drivingbehavior); start driving behavior;intermediate soak times (engine-offtimes between 10 minutes and 2 hoursprior to vehicle start); and actual airconditioner (A/C) operation. TheAgency is proposing new requirementsfor these areas, separate from theexisting FTP requirements. Alsoincluded in this proposal arerequirements to improve the simulationof actual road load forces 4 across allspeed ranges and to revise the criteriafor allowable speed variation for a validtest, which would be applicable both tothe new provisions proposed in thisNPRM and the existing FTP.

As most of this proposal deals withareas that have not previously beenregulated, the Agency is considering abroad range of alternative approachesand requests. Comment on thealternative approaches, as well as thecentral proposal, are requested.Depending on comments and datareceived and analyses conductedsubsequent to today’s proposals, EPAmay include some of the alternatives, inwhole or in part, in the final rule.Interested parties may also submitcomments on alternatives notspecifically identified or analyzed byEPA for this proposal.

While both the central proposal andthe alternatives are EPA’s own design,they incorporate some concepts putforth both by the California AirResources Board (CARB) and the AdHoc Panel on Revisions to the FTP (AdHoc Panel), a joint committee of theAmerican Automobile ManufacturersAssociation (AAMA) and theAssociation of International AutomobileManufacturers (AIAM).

The proposed additions and revisionsto the tailpipe emission portions of theFTP would apply to all LDVs and LDTs,certifying on all current motor fuels.The proposed changes would apply totesting conducted during certification,Selective Enforcement Audits, and in-use enforcement (recall). Adjustmentsare included to accommodate certainvehicle types, transmission types, andperformance categories where theadditions are not representative of in-use driving. The Agency solicitscomments and data on the appropriatetreatment of vehicles for whichadjustments are allowed and themethods for making the adjustments.

A. Central ProposalThe central proposal relies on a new

Supplemental Federal Test Procedure(SFTP) that addresses variousconditions under which vehicles areactually driven and used, which are notin the FTP. The SFTP includes threenew driving cycles to represent (1)aggressive driving (as characterized byhigh speeds and/or high accelerations);(2) driving immediately followingvehicle startup; and (3) microtransientdriving (rapid speed fluctuations),which occur across the majority of thenormal ranges of operating speeds andaccelerations. The proposed SFTPincorporates conditions that aredesigned to more accurately reflectactual engine load due to A/C operationunder typical ozone exceedanceconditions. A new intermediate-duration (10- to 60-minute) soak periodis also included.

Two components of today’s proposalhave wider impacts than just the SFTP.The first is to more accurately simulatereal on-road loads at the tire/dynamometer interface, which is anelement of the proposal that affectsdynamometer operation throughoutboth the FTP and SFTP. The secondwould remove language specifying‘‘minimal throttle movement’’ whenconducting emission tests and replace itwith ‘‘appropriate throttle movement’’and require a specification of allowablespeed variation, which also impactsboth SFTP and FTP testing. The Agencyis also requesting comment on whetherthe increased sophistication of vehiclecomputers necessitates replacingexisting defeat device language with arequirement for proportional emissioncontrol under conditions not directlyrepresented by the FTP and the SFTP.

The proposed standards would applyfor full useful life under section 202 ofthe Clean Air Act. The warrantyprovisions under section 207 of theClean Air Act also apply to thisrulemaking.


5 5 Refers to Bag 2 of the LA4, preceding the 10-minute hot soak, lasting 866 seconds.

Supplementary Federal TestProcedure—The SFTP includes threesingle-bag emission test cycles: a hotstabilized 866 Cycle 5 run with a new

simulation of in-use A/C operation; anew Start Control Cycle (SC01, seefigure 2) simulating driving with thenew simulation of in-use A/C operationand proceeded by a soak period; and anew Aggressive Driving Cycle (US06,see figure 3) run in the hot stabilized

condition. The cycles of the SFTP canbe run as a sequence to save onpreconditioning and setup time;however, separate runs of the cycles arepermissible with the appropriate soak orpreconditioning steps appended.BILLING CODE 6560–50–P



BILLING CODE 6560–50–C


6 Light-duty trucks are divided into two weightcategories known as light light-duty trucks (rated upthrough 6000-pounds Gross Vehicle Weight Rating(GVWR)) and heavy light-duty trucks (rated greaterthan 6000-pounds GVWR).

7 Refer to the Final Technical Report onAggressive Driving Behavior for the Revised FederalTest Procedure Notice of Proposed Rulemaking fora detailed discussion of the points in the cyclewhere the proposed adjustments would be made.

Elements of the proposed A/Csimulation for certification testinginclude, a 95 °F ± 5 °F test cell ambienttemperature, A/C set to ‘‘maximum A/C’’ with interior air recirculation, highinterior fan setting, coldest setting onthe temperature slide, driver’s windowdown, and front-end supplemental fancooling. Although certification testingwould occur at 95°, the compliancerequirement would apply at lessdemanding temperatures as well. Thus,EPA confirmatory testing could takeplace at any point across the range 68 °Fto 95 °F. The compliance requirementwould would The Agency proposesthese conditions as a cost-effectivesurrogate for testing in a fully controlledenvironmental chamber set to simulateozone-exceedance conditions of ambienttemperature, humidity, solar load, andpavement temperature, although the useof a fully controlled environmentalchamber would be permitted.

The required elements for the SC01include the preconditioning, soakperiod, and compliance cyclerequirements. Prior to the soak period,the vehicle is to be preconditioned toallow engine and catalyst temperaturesto stabilize at typical warmed-upoperating temperatures. The Agencybelieves that running the vehicle overEPA’s Urban Dynamometer DrivingSchedule (LA4) is adequate to achieveengine and catalyst stabilizationregardless of the time period for whichthe vehicle was not operational prior topreconditioning. However, in the eventthe vehicle was shut off for less thantwo hours prior to preconditioning, theAgency believes that a 505 cycle isadequate for preconditioning thevehicle, although the 866 or the SC01 isalso acceptable.

Immediately following thepreconditioning cycle, the vehicle willenter the soak period. Manufacturertesting of engine families required tocomply with the intermediate soakrequirements for certification or SEAtesting must soak the vehicle for at least60 minutes. EPA will have the option oftesting any soak duration between 10and 60 minutes for certification, SEA,and in-use testing. If the engine familyis not required to meet the intermediatesoak requirements, a 10-minute soakperiod is proposed. During this period,cooling fans directed at the vehicle areto be shut off. The vehicle may beremoved from the dynamometer,provided the vehicle is not subjected tounrepresentative cooling of the engineor catalyst. Following the soak period,the vehicle will be run over the SC01cycle using the proposed A/Csimulation for proper representation of

engine and catalyst warm-up and startdriving.

The US06 driving cycle is designed tobe run in hot stabilized condition. High-volume exhaust flow for larger-displacement vehicles run on US06dictates use of a larger-capacity constantvolume sampler (CVS) than is neededfor current FTP testing. The proposedA/C simulation is not required for thistest cycle.

The Agency proposes thatmanufacturers determine theappropriate shift points for their manualtransmission applications and submitthe shift schedules for EPA approval. Ingeneral, EPA will allow manufacturersto specify upshift points, butdownshifting will not be permittedunless the vehicle is unable to staywithin the driving tolerance on thespeed trace in the existing gear.

Hot stabilized condition is achievedby including several preconditioningoptions as part of the formal procedureimmediately prior to the US06 Cycle. Ifthe vehicle has undergone a soak of 2hours or less, the preconditioning maybe a 505 Cycle, the 866 Cycle, US06, orthe SC01. Following longer soaks, theproposed preconditioning cycle is anLA4. For manufacturers who haveconcerns about fuel effects on adaptivememory systems, the proposal allowsmanufacturers, and upon manufacturerrequest, requires EPA to run the vehicleover the US06 Cycle on the certificationtest fuel before entering the formal testprocedure.

The Agency proposes adjustments tothe aggressive driving test cycle for allheavy light-duty trucks (HLDTs),6 andalso, for some low- and high-performance LDVs and LDTs. Theproposal calls for US06 Cycle testing ofHLDTs with the truck ballasted to curbweight plus 300 lbs and thedynamometer inertia weight determinedfrom this same basis, while FTP testingremains at Adjusted Loaded VehicleWeight. The proposed US06 Cycleadjustments based on performance levelare summarized in Table 1. For lowperformance vehicles, the inertia weightis adjusted by multiplying the originalinertia weight by the adjustment factorwhich is equal to the ratio of theapplicable performance cutoff and theW/P of the test vehicle. Where anadjustment factor is called for, it isapplied dynamically by thedynamometer only during thoseportions of the US06 Cycle that are the

most aggressive.7 No adjustment factorsare proposed for mid-performance(‘‘normal’’) vehicles. For highperformance vehicles, the manufacturermust demonstrate stoichiometric controlfor wide-open throttle events of twoseconds or less in order to ensure thatthese vehicles have aggressive drivingemission control over similar vehicleoperation as the rest of the fleet.

TABLE 1.—PERFORMANCE-BASEDADJUSTMENTS

Trans-mission type

Perform-ance (W/P

range)Adjustment

manual ....... lowW/P>34

dynamic dyna-mometer iner-tia weight re-duction.

normal18 W/P 34

none.

high W/P<18

2 second stoichcontrol.

automatic .... lowW/P>31

dynamic dyna-mometer iner-tia weight re-duction.

normal18 W/P 31

none.

high(W/P<18)

2 second stoichcontrol.

Determining compliance withstandards—With the exception ofchanges prompted by use of newdynamometers and an additional driverspeed variation tolerance, no changesare proposed for the driving cycle of theconventional FTP. Similarly, EPAproposes to retain unchanged themethod of calculating compliance withthe existing FTP. However, anadditional ‘‘composite’’ compliancecalculation is proposed that bringstogether elements of the conventionalFTP with results from the SFTP. In thecomposite calculation, emissions fromthe range of in-use driving areappropriately weighted, summed, andcompared to the proposed emissionperformance standards. For totalhydrocarbon (THC), non-methanehydrocarbons (NMHC), organic materialhydrocarbon equivalents (OMHCE),organic material non-methanehydrocarbon equivalents (OMNMHCE),and CO, the proposed standards are thesame as the standards applicable underthe conventional FTP; for NOX, anadjustment factor of 1.15 is applied tothat standard to account for theemission response of vehicles to thenew A/C test conditions. See the


8 The issue of what standards would apply in thecontext of a voluntary Federal low emission vehicleprogram will be determined in a separaterulemaking (60 FR 4712, January 24, 1995).

9 Both the Support Document to the ProposedRegulations for Revisions to the Federal TestProcedure: Detailed Discussion and Analysis andthe Technical Reports are in the public docket forreview.

10 Time required for the catalyst to reach thetemperature needed to sustain significant catalyticactivity.

Support Document to the ProposedRegulations for Revisions to the FederalTest Procedure: Detailed Discussion andAnalysis for the specific numericalstandards. Due to the absence ofrelevant test data on which to base adecision, no supplemental testprocedures or standards are proposedfor diesel particulate.

Included in the composite calculationare a cold start bag (based on Bag 1 ofthe conventional FTP) and the threebags of the SFTP (called Bag 4, 5, and6). The weighting factor for each of thefour bags is adjusted as appropriate toreflect the proposed level of control foreach type of driving in the SFTP.Because the exhaust constituentsrespond differently to the loads andspeeds of the new SFTP cycles, theproposed levels of control and, thus, theweighting factors of the compositecalculation differ somewhat for differentpollutants. The proposed weightingfactors are:

Percentages

THC/NMHC CO & NOX

Bag 1 (cold startfrom FTP) ...... 21 15

Bag 4 (866 cyclefrom SFTP) .... 24 37

Bag 5 (SC01from SFTP) .... 27 20

Bag 6 (US06from SFTP) .... 28 28

The Agency is proposing that changesin the achievable levels of control overthe SFTP tests would track changes inthe underlying FTP standards and, thus,adoption of the central proposal wouldhave the effect of automaticallyreducing the composite standards instep with any mandatory future declinesin the FTP standards.8

Flexibilities are proposed to allowmanufacturers to reduce their testingburden, particularly duringdevelopment testing. (See SupportDocument to the Proposed Regulationsfor Revisions to the Federal TestProcedure: Detailed Discussion andAnalysis and Technical Reports fordiscussion.) 9

Emissions Standards and Phase-in—The Agency is proposing to phase in theproposed requirements for aggressivedriving and air conditioning controlprior to implementing the intermediate

soak requirements. It is proposed thatthe standards apply to 40 percent ofeach manufacturer’s combinedproduction of LDVs and LDTs for the1998 model year, 80 percent in 1999,and 100 percent in 2000. Small volumemanufacturers would not have tocomply until the 2000 model year. Allthe proposed requirements would applyduring this phase-in period, except thatBag 5 could be conducted with a 10-minute soak instead of the proposed 60-minute soak for control of intermediatesoak emissions. The 60-minute soakwould be required for all vehiclesstarting with model year 2001,including small volume manufacturers.

The Agency is continuing to analyzethe impact of this phase-in schedule,particularly when considered inconjunction with other recentlypromulgated rules (such as revisions tothe evaporative test procedures) as wellas potential future programs (such asvoluntary Federal low emission vehiclestandards). Comments are specificallyrequested (1) on the impact of thisphase-in schedule when consideredwith other programs and (2) providingsuggestions for other schedules whichwill coordinate programs moreeffectively. The Agency will review thisinformation in developing the final ruleto determine if a more logicalcoordination schedule is possible whilemaximizing the cost/benefiteffectiveness of this rule.

The proposal recognizes that adoptionof emission standards more stringentthan current Federal Tier 1 standardswill likely result in emission controlstrategies that reduce catalyst light-offtimes.10 This could have a significantimpact on the costs and benefits of theintermediate soak requirement. As Tier1 standards are the current legalrequirement and the status of futurestandard changes is uncertain at thistime, this proposal presumes Tier 1applicability. The Agency invitescomments and data addressing the cost/benefit implications of the proposedsoak requirement under a Federal Tier2 (or equivalent) program.

Each of the test cycles is run on asystem providing accurate replication ofreal road load forces at the interfacebetween drive tires and thedynamometer over the full speed range.In addition, the new US06 cyclerequires significantly higher powerabsorption capacity, due to the higherpower requirements of this aggressivedriving cycle. While EPA intends to usea large-diameter single-roll

dynamometer with electronic control ofpower absorption to meet theserequirements for both the new SFTP andcurrent FTP testing, any system wouldbe allowed that yields equivalent orsuperior test results.

The improved road load simulationand the new criteria for allowable speedvariation for FTP compliancedetermination are proposed to beimplemented in the 1998 model year.Manufacturers could elect to useimproved road load simulations prior to1998, at their option.

The Agency is also proposing a minorprocedural change that would removethe current 5500-pound test weight cap,to be implemented in the 1998 modelyear with the improved road loadsimulations.

B. Alternative Approaches

As indicated, EPA is considering anumber of alternatives to criticalelements of the central proposal. Thefollowing provides a summary of themost important of these alternatives. Afull discussion of all the options andalternatives considered is found in theSupport Document to the ProposedRegulations for Revisions to the FederalTest Procedure: Detailed Discussion andAnalysis.

In determining compliance with theemission standards, EPA is consideringtwo alternatives to the proposed FTP/SFTP composite and the relatedstandards: (1) promulgating threeseparate sets of standards, one set eachfor aggressive driving, post-soak startupemissions, and A/C impacts; and (2)promulgating a single set of standards,based on a simple weighted average ofseparate standards for each control area.Both of these alternatives would use thesame cycles and test procedures as thecomposite approach of the centralproposal. However, instead of weighingthem with Bag 1 of the FTP and usingbag weights to help establishappropriate compliance procedures andstandards, the alternative approacheswould establish emission standardsspecifically for each new control area.

The Agency did not select either ofthese alternatives as the central proposalbecause of difficulties encountered indetermining the appropriate amount ofin-use compliance margin to allowwhen establishing emission standards.Also, the proposed concept of indexingthe SFTP standards to any futurechanges in FTP standards probablywould not work with either of the twoalternatives. If data are submitted thatcould help establish appropriate in-usemargins, EPA would reevaluate the mostappropriate compliance structure and, if


11 Ad Hoc Panel, ‘‘Industry Proposal on FTPRevisions,’’ October 20, 1994.

12 The Ad Hoc Panel has submitted a proposedmethodology for such a dynamometer simulation ofA/C load, dubbed ‘‘Nissan II.’’ Manufacturers are

pursuing additional refinements to addresspotential concerns with the approach, such as theability to simulate air compressor cycling and A/Cloads at idle, which cannot be simulated on adynamometer.

13 In this report, ‘‘driving behavior’’ refers to themeasurable consequences of the operator’s actionon the accelerator pedal, including vehicle speed,throttle variation, acceleration, and power.

14 Details about the development of the LA4driving cycle can be found in an SAE paper,‘‘Development of the Federal Urban DrivingSchedule,’’ Ronald E. Kruse and Thomas A. Huls,EPA, 1973, #730553.

appropriate, may select one of thealternatives in the final rule.

The Agency is also considering thealternative of establishing a singlestandard for NMHC+NOX, instead ofseparate standards, and invitescomments on the cost and emissionimpacts of this alternative.

One issue was identified too late forEPA to properly evaluate it. Concernwas raised that the proposed level of COcontrol may significantly interfere withthe ability for vehicles to comply withthe proposed level of NOX control.Should further data and analysessubstantiate that tradeoffs between COand NOX control would precludemeeting the proposed level of NOX

control, EPA would consider reducingthe stringency of the CO standards forthe new control areas in the final rule.

On October 20, 1994, EPArepresentatives received a joint vehiclemanufacturer proposal from the Ad HocPanel that addressed emissions arisingfrom aggressive driving and A/Coperation and proposed emissionstandards for each of these two areas.The Agency has not had sufficient timeto fully analyze the concepts offered bythe panel or to incorporate themanufacturer proposal as an explicit,complete alternative to the primaryAgency proposal presented today.Nevertheless, the manufacturers’specific proposals fall within the scopeof the options and alternativesdiscussed by EPA in today’s notice. TheAgency has submitted materialssupplied by the panel on October 20,1994, to the rulemaking docket.11

Analysis of these elements by theAgency, as well as any related materialsupplied in the future, will also bedocketed. In order that the Agency maymake the most informed andappropriate judgments in any finalrulemaking, EPA encourages interestedpersons and organizations to evaluateand comment upon these materials.

In the area of A/C emission control,EPA is considering an alternative to theproposed test simulation of A/Coperation, as well as the alternative ofrequiring A/C testing across the coldstart (that is, Bag 1 of the FTP). Thealternative A/C simulation would leavethe A/C off in the test cell, but wouldincrease the dynamometer load curveacross the range of vehicle speeds toreflect the additional load imposed byan A/C compressor during ozoneexceedance conditions.12

In the intermediate soak area, theeffect on in-use emissions of thealternatives depends on future changesto the stringency of the FTP standards,the control strategies manufacturerswould employ to meet such futurestandards, and the impacts thosestrategies might have on post-soakemissions. Because these are not known,alternatives might include exemptionfrom aspects of the soak requirement ortotal deletion of the soak requirement.

IV. Statutory Authority and LegalAnalysis

The promulgation of these regulationsis authorized by sections 202, 206, 208,and 301 of the Clean Air Act (CAA orthe Act) as amended by the Clean AirAct Amendments of 1990 (42 U.S.C.7521, 7525, 7542, and 7601). Section206(h) of the Act requires EPA to‘‘review and revise as necessary—thetesting of motor vehicles and motorvehicle engines to insure that vehiclesare tested under circumstances whichreflect the actual current drivingconditions under which motor vehiclesare used, including conditions relatingto fuel, temperature, acceleration, andaltitude.’’ Congress mandated that EPAexercise its authority under section206(a) of the Act, giving broad authorityto determine appropriate testprocedures, consistent with the broaddirection of section 206(h), to determineappropriate changes to reflect real worldconditions.

Although the text of the statute andthe legislative history do not provideexplicit criteria or intent for this review,EPA believes the primary concern ofCongress is having test procedures forLDVs and LDTs reflect in-use conditionsin order to obtain better in-use emissioncontrol. This flows from the basicpurpose of test procedures—to measurecompliance with the emissionstandards—and from standardsdesigned to obtain in-use emissionreductions. Therefore, EPA made thisthe primary concern and objective.

A more detailed analysis of thestatute, the scope of EPA’s authority,and interpretation of how best toexercise EPA’s discretion under section206(h) are found in the SupportDocument to the Proposed Regulationsfor Revisions to the Federal TestProcedure: Detailed Discussion andAnalysis.

V. The FTP Review Project and Areasof EPA Concern

In response to the review requirementof the CAAA, EPA initiated the FTPReview Project (the FTP Review) inNovember 1990. The first action of theproject team was to perform an initialreview of existing information toidentify elements of the current FTPthat might be of concern (justifyingadditional focus) and others that mightnot justify concern at this time.

Of immediate concern to EPA wasrepresentativeness of the driving cycleused in the current FTP, the ‘‘LA4’’ or‘‘Urban Dynamometer DrivingSchedule,’’ especially in the area ofaggressive driving behavior.13 It wasclear that the LA4 maximum speed of 57mph excluded a significant fraction ofhigher-speed, in-use operation.14

Similarly, EPA suspected that animportant fraction of in-useaccelerations were more severe thanthose found in the LA4. A 1990 CARBstudy found much higher emissions,particularly for CO, during operation athigh acceleration rates relative to thoseseen during FTP-level accelerations.

One possible explanation for theseemission increases is that the engineswere not calibrated for emission controlduring the higher engine loadsassociated with aggressive driving, asthese loads are not encountered duringcurrent FTP testing. However,insufficient data existed at the time toquantify the in-use frequency ofaggressive driving events or the actualemission impacts. There were alsoconcerns, based on engineeringjudgment, about other aspects of drivingbehavior that were not represented inthe current test procedures for which nodata existed. Thus, the Agencyconcluded that further information wasnecessary to properly represent actualdriving conditions. In collaborationwith key stakeholders, EPA beganextensive research into driving behaviorand conditions and their emissionimplications.

During the course of the research anumber of other concerns with thecurrent FTP were identified, includingtwo additional concerns with the LA4representation of in-use drivingbehavior. The first concern was startdriving behavior; that is, behaviorimmediately following vehicle startup


and initial idle. The LA4 cycle used inthe current FTP brings the mostaggressive operation close to thebeginning of the cycle; driving surveydata suggest this is atypical of in-useoperation. The second concern wasmicrotransient behavior (rapid speedfluctuations). In-use driving survey datacontains more frequent speedfluctuations than the FTP.

The Agency identified concerns aboutfour additional elements of the FTP: Theduration of the soaks; the representationof A/C load; representation of additionalloads on the engine due to factors suchas road grade, extra cargo, or trailertowing; and the adequacy of thedynamometer specification forrepresentation of real road load.

With respect to soaks, EPA sought todetermine if significant levels ofemissions are missed by the current FTPbecause only very short- and long-duration soaks are reflected in thecurrent structure. One relatedhypothesis was that the much fastercooling rate of catalysts compared toengines might lead to excessiveemissions during intermediate-durationsoaks.

Several aspects of the A/C loadsimulation were problematic. Thecurrent FTP adds load as a percentageof the base road load horsepower curve,which means the FTP A/C loaddecreases with decreasing speed, whilereal A/C system loads relative to roadload horsepower are highest at lowspeed. Also, vehicles with different basehorsepower curves end up withdifferent FTP A/C load simulations,even if they have identical A/C systems.Additionally, the Agency believes thatthe current method significantly under-represents the magnitude of in-use A/Cloads. As in the case of aggressivedriving behavior, incorrectrepresentation of A/C loads during theFTP risks incorrect simulation of theemissions these loads would generatefrom an engine in-use.

Road grade, vehicle towing, and cargoalso represent a load effect on theengine. The 300-pound passenger-plus-cargo allowance on the FTP is clearlyunrepresentative for some drivingsituations, especially for trucks, and theabsence of road grade or vehicle towingsimulations on the FTP means theseactual in-use loads are not a factordetermining emission standards orcompliance with those standards.

Three aspects of the current FTPdynamometer configuration have thepotential to misrepresent the actual roadload experienced by vehicles in-use.First, the shape of the speed/load curveon current certification dynamometersis fixed and cannot be changed; the

magnitude of the speed/load curve isadjusted by periodically calibrating thedynamometer at a single speed(currently, 50 mph). As a consequence,loads at speeds other than thecalibration point can be misrepresented.Second, current FTP dynamometerscradle the vehicle drive wheels betweentwo small (8.65-inch) rolls. Heatingeffects and pinching of the tire result inan unrepresentative simulation of road‘‘surface.’’ Third, the dynamometer rollsare currently uncoupled and the frontroll (which bears the power absorber)spins somewhat more slowly than therear (which provides the vehicle speedsignal); this tends to bias the systemtowards underloading the vehicle.

The Agency analyzed three otherelements of the FTP and believesrevising the current procedures isunnecessary at this time. The first sucharea was the altitude of testing. Giventhat EPA has the authority to performvehicle testing at any altitude, and itcurrently exercises that authority, theAgency is not proposing to supplementby further regulation the altitude testingflexibility in current law. While it ispossible that driving behavior maydiffer at high altitudes, EPA believesthat any emission controls required foraggressive driving will also be effectiveduring high altitude driving.

A second element which EPA did notpursue beyond the initial evaluationwas test fuels. In-use fuels have a widerange of properties. This specificationfor fast fuel allowance for a range offuels (40 CFR 86.113–94) appear toprovide EPA with the flexibility to usea variety of test fuels ranging from anaverage in-use fuel to some of the lesstypical in-use fuels with qualities thatcould effect emissions. Significantdifferences, with potentially largeemissions implications, do appear toexist between average in-use gasolineand the gasoline (indolene) typicallypurchased by both EPA and industry forcertification testing. After evaluatingapproaches to addressing this situation,EPA concluded that changes to theregulations are not necessarily required,since the current regulations provide theflexibility needed to address thosesituations where the use of indolenemay not be representative. In addition,various programs to address in-use fuelqualities are still under consideration. Ifa decision is ultimately made to changethe certification fuel regulations, it maybe best to do so along with changes tothe specifications for in-use fuels.

Finally, EPA believes that it isunnecessary to further address thedirect impacts of ambient temperatureon FTP tailpipe emissions in thisproposal. At the time the Amendments

were adopted, the FTP evaluatedtailpipe emissions performance in themidrange of temperature (68° F to 86°F), but omitted both cold and hottemperature testing. The emissionconcern following cold temperaturesoaks and during cold temperatureoperation is increased CO emissions.This concern was addressed throughEPA’s Cold Temperature CO rulemaking(57 FR 31888). The direct emissionimpact during hot temperatureoperation is increased fuel evaporation.Ambient temperature should nototherwise affect tailpipe emissions, asthe engine and combustion temperatureare not affected in any significant wayby temperatures hotter than 86° F. Thisconcern was addressed through theAgency’s Evaporative Emissionsrulemaking (58 FR 16002). Ambienttemperature also produces indirectemission effects through increasedoperation of the vehicle A/C, affectingthe load on the engine. This indirectaspect of temperature was addressed inEPA’s detailed review of the FTP and isreflected in today’s proposal.

The FTP Review project team foundthat existing information was clearlyinadequate for evaluating potentialrevisions to the test procedures.Consequently, a number of new datagathering and analytical efforts wereundertaken in connection with theproject. In several of these efforts, EPAresources were supplemented bysignificant cooperative investmentsfrom other sources, including theAmerican Automobile ManufacturersAssociation (AAMA), the Association ofInternational Automobile Manufacturers(AIAM), and the California AirResources Board (CARB). These studiesprovided EPA with unprecedented dataon which to base its comparative reviewwith the FTP and to construct theoptions presented in today’s proposal.

VI. In-Use BehaviorThe first critical need in reviewing the

FTP was a current database on in-usedriving and vehicle soak behavior. TheAgency collaborated with AAMA,AIAM, and CARB over the spring andsummer of 1992 to conduct surveys ofin-use driving and soak behavior in fourmajor U.S. cities.

A. In-Use Driving BehaviorInstrumented vehicle surveys and/or

chase car studies were conducted inBaltimore, Maryland; Spokane,Washington; Atlanta, Georgia; and LosAngeles, California. In May of 1993,EPA published its initial conclusionsregarding aggressive driving behavior inthe ‘‘Federal Test Procedure ReviewProject: Preliminary Technical


15 U.S. Environmental Protection Agency, FederalTest Procedure Review Project: PreliminaryTechnical Report, EPA 420–R–93–007, Office of Airand Radiation, May 1993.

16 The power needed from an engine to move avehicle is proportional to both the vehicle speed

and the acceleration rate. Neither variable, by itself,is a good measure of the load placed on the engine.The joint distribution of speed and acceleration isprobably the best measure, but it must be examinedin three dimensions, which is difficult to visualizeand comprehend. The concept of specific power

provides a two-dimensional measure which isroughly equal to 2*speed*acceleration and has theunits of mph2/sec.

17 See the Technical Reports for a full descriptionand analysis of this data.

Report.’’ 15 These conclusions werelargely based on the Baltimoreinstrumented vehicle survey data.Subsequent analysis has found thelarger three-city instrumented vehicleresults to be consistent with theBaltimore-only results. The three-cityanalysis showed that nearly 13 percentof vehicle operation, on a time-wrightedbasis, occurs at combinations of speedand acceleration that fall outside thematrix of speeds and accelerationsfound on the LA4 driving cycle. Themaximum observed in-use speed was95.5 mph, compared to the LA4maximum speed of 56.7 mph, andslightly more than seven percent of in-use vehicle operation time was spent atspeeds greater than 60 mph. Averagespeed from the three-city in-use datawas 25.9 mph compared to 19.6 mphover the LA4.

Specific power is also useful whenanalyzing aggressive driving behavior.16

Measures of power also indicated thatin-use driving behavior was moreaggressive than reflected in the LA4.Specific power in the three-city sampleranged up to 723 mph2/sec andaveraged 47.0 mph2/sec. The LA4 hasmaximum power of 192 mph2/sec andan average of 38.6 mph2/sec.

The Agency analyzed the in-usesurvey data to determine how the above

findings on speeds, accelerations, andpower measures were affected by otherfactors, including vehicle type (car/truck), transmission type, vehicleperformance level, time of day, and dayof the week. The first three vehicle-related factors are reflected in today’sproposal. The discussion of the analysisand findings are in the SupportDocument to the Proposed Regulationsfor Revisions to the Federal TestProcedure: Detailed Discussion andAnalysis and the Technical Reports.

The Agency also examined startdriving behavior as represented by theinstrumented vehicle survey data. TheAgency determined that the start driving(operation following the initial idle andbefore coolant temperature exceeded140° F) in the survey data generally didnot exceed 240 seconds. Furtheranalysis showed that the speeds of startdriving did not change substantiallyfollowing soaks of different durations,but they did differ from those found inhot stabilized driving. The results for in-use initial idle time and start driving aredifferent than the representation ofthese elements in the FTP. The LA4cycle has atypical high speeds over thefirst four minutes of a vehicle trip. Onthe other hand, the LA4 hassubstantially less aggressiveaccelerations than the first 80 seconds

or so of typical in-use start driving,while it is substantially over-aggressivewhen compared to the succeeding 160seconds. For initial idles, the FTPpresumes 20-second durations for bothcold and hot starts, whereas the in-useaverages from EPA’s data were 28seconds for cold starts and only 12seconds for hot starts.

The previous discussion of in-usespeeds and accelerations presents asnapshot of driving behavior. Althoughthe acceleration measure, which looks atthe change in speed from one second tothe next, partially characterizes thetransient nature of driving, othermeasures expand the time interval toexamine the rapid fluctuations in speed,or microtransients. One measure,referred to as jerk, is equal to the changein acceleration. A related measure is thesecond-to-second change in specificpower. Conceptually, this measurecaptures the change in the powerrequirement imposed by the drivingbehavior.

The Agency used the three-parameterinstrumented vehicle data fromBaltimore, Spokane, and Atlanta,17 tocalculate these microtransient measuresfor in-use driving behavior andcompared the results to the LA4’srepresentation. The measures of jerk andchange in power are shown in Table 2.

TABLE 2.—MEASURES OF MICROTRANSIENT DRIVING FROM INSTRUMENTED VEHICLE DATA/SEC

Source

Jerk Change in power

Mean of theabsolutevalues

(mph/sec)

Standarddeviation(mph/sec)

Mean of theabsolutevalues

(mph2/sec)

Standarddeviation

(mph2/sec)

In-use driving .................................................................................................................... 0.47 0.89 20.48 34.36LA4 ................................................................................................................................... 0.36 0.63 14.96 22.96

For both jerk and change in power, themean of the absolute values were usedin order to look at both the positive andnegative values (the mean of the signedvalues of jerk is always equal to zero).The in-use means were higher thanthose for the LA4, indicating larger in-use changes in acceleration and power,as well as reflecting, in part, the LA4’sacceleration rate cutoff of 3.3 mph/secand the maximum speed of 57 mph. Thestandard deviations of jerk and changein power are probably a better measureof microtransient behavior. Again, in-use data show larger values for both

measures. The greater variation aroundthe mean demonstrated by the in-usedata suggests that the LA4 does notadequately represent the microtransientnature of in-use driving behavior.

B. Soak BehaviorThe survey data were also analyzed to

determine the frequencies at whichsoaks of different durations occurred in-use. The Agency found that soaks of lessthan 10 minutes and greater than 8hours occur with the highestfrequencies in use. However, EPA alsofound that a significant portion of in-usesoaks are of intermediate duration. For

example, nearly 40 percent of all soaksin the Baltimore survey data werebetween 10 minutes and 2 hours. Giventhat the current FTP employs only twosoaks (the 10-minute hot soak and the12- to 36-hour cold soak) to representthe range of soaks in-use, EPA wasconcerned that the current FTP mightnot adequately control for emissionsfollowing these intermediate-durationsoaks.

C. Air ConditioningA number of variables affect the range

of A/C usage, particularly temperature,sun load, and humidity, all of which


18 18 U.S. Environmental Protection Agency,Passenger Car Fuel Economy: EPA and Road, EPA#460/3–80–010, September 1980, p. 119.

19 EPA’s assessment was limited to EPA’s andAAMA/AIAM data. Due to differences in testinghardware, CARB’s emission results were notdirectly comparable.

20 These estimates are only for the emissionunder-prediction related to driving behavior. Otherfactors such as soak are addressed in the sectionsto follow.

vary by season, time of day, andgeographic location. Given that theoverall goal of the Act is to help bringlocalities and regions into compliancewith the National Ambient Air QualityStandards (NAAQS), the Agency choseto focus attention on the contribution ofA/C to vehicle emissions during typicalhigh ozone situations. Analyses ofozone exceedances revealed that ozoneexceedances typically occur on dayswith a mean ambient temperature of 95°F, 30–40 percent relative humidity, andlimited cloud cover.

In August and September 1994, theAgency conducted an instrumentedvehicle study in Phoenix, Arizona.Preliminary analyses of the survey dataindicate that the average A/C usage was77 percent for days that reached a peaktemperature between 90 °F and 100 °F.The A/C compressor was actuallyengaged 61 percent of the total time (seeTechnical Reports and the SupportDocument to the Proposed Regulationsfor Revisions to the Federal TestProcedure: Detailed Discussion andAnalysis for full analysis). The high useof A/C in ozone exceedance conditionsmakes the accurate simulation of A/Cduring the FTP more important.

D. Additional Elements Affecting EngineLoad

A comprehensive evaluation ofadditional elements affecting engineload would require surveys of thefrequency of occurrence of the elementsin-use, as well as evaluation ofinteractive effects with driving behavior.For road grade, a 1980 EPA report 18

indicated that positive road gradesaverage 1.66 percent nationally and thatroughly six percent of national VMT isspent on grades of four percent orhigher. The Agency sought tosupplement this information withdriving behavior data over road grade,gathered during the chase car portion ofthe in-use driving surveys.Unfortunately, problems with noise andinsufficient resolution on the measure ofgrade rendered the data inadequate, andno alternative data source was available.In addition, EPA was unable to conductin-use surveys in the areas of passenger/cargo loading and trailer towing, due tothe scope and nature of the necessarysurvey instrument. As a consequence,EPA has insufficient data for use inevaluating the additional elementsaffecting engine load that wereoriginally identified as areas of concern.

VII. Representative Driving Cycles

In order to evaluate the emissionimpacts of in-use driving and soakbehavior, EPA designed three drivingcycles that were representative of the in-use survey results, using segments ofactual in-use driving survey data.Concurrently, EPA determinedweighting factors to reflect the fractionof in-use operation represented by eachcycle; these factors are used to properlyweight the emissions from the cycleswhen doing an emission assessment.

The Agency developed separatecycles for start driving and aggressivedriving. The Agency chose to developindividual cycles rather than a single‘‘representative’’ cycle in order toevaluate EPA’s areas of concernindependently. This is most critical inthe case of aggressive driving whereboth capturing the diversity ofaggressive driving behavior andrepresenting it proportionally in a singlecycle covering all in-use operationwould lead to a very long cycle.

The Start Cycle (ST01) representsthree successive 80-second segments ofin-use driving immediately followingthe initial idle. Testing using ST01allowed separate determination of startdriving emissions; ST01 was also usedto quantify the emissions effects ofvarying soak duration.

The second cycle, characterizingaggressive driving, was theRepresentative Non-LA4 Cycle (REP05).This cycle targeted speeds andaccelerations, as well as microtransienteffects, not covered by the current LA4.

To complete the representation of in-use driving behavior for emissionassessment purposes, a third cycle, theRemnant Cycle, was developed tocharacterize in-use driving behavior notrepresented by either the ST01 orREP05.

The Agency used the same basic cycledevelopment methodology for each ofthe three representative cycles. A fulldiscussion of the methodology used, thecomposition of each cycle, and how itcompares to the cycle in the FTP isfound in the Technical Reports.

It seemed clear from the in-use surveydata that rapid speed fluctuations,including ones not well represented onthe LA4, could be found in all types ofin-use vehicle operation. The Agency’suse of actual microtrips as the buildingblocks for the three representativecycles directly incorporated suchmicrotransient driving behavior into allthree cycles.

The Agency has assumed that drivingbehavior is not affected significantly byA/C operation and that therepresentative driving cycles developed

from the in-use driving survey data areequally applicable to testing with the A/C system on and off. In fact, eventhough the Atlanta driving survey wasthe only one of the three surveysconducted during the summer, that cityhad the most aggressive driving of thethree cities. Thus, it does not seemlikely that A/C operation could have asignificant impact on driving behavior.Nonetheless, the Agency welcomes dataand comments on the relationshipbetween A/C operation and drivingbehavior.

VIII. Emission Inventory Assessments

An assessment of emissions from fourareas for potential emission control wasconducted using the representative testcycles developed from the survey data.A full description of the test programsand the results can be found in theTechnical Reports. The followingsummarizes the conclusions for eacharea considered.

A. In-Use Driving Behavior

The FTP Review’s emissionassessment of in-use driving behaviorwas based on a vehicle emission testprogram conducted cooperatively byEPA, CARB, AAMA, and AIAM during1993 and early 1994 (referred tosubsequently as the Non-LA4 EmissionsTest Program).19

On the basis of the EPA data, theproject team concluded that the LA4under-predicts actual in-use hotstabilized emissions by 0.043 g/miNMHC, 2.8 g/mi CO, and 0.083 g/miNOX on current technology, properlyoperating vehicles.20 These numbers donot have any direct bearing on the FTPstandards; they are simply an estimateof the additional amount such vehiclesactually emit in-use, compared to theFTP test results.

Table 3 shows the percentagecontribution to the in-use emissionincrease from the Start (ST01), Remnant,and aggressive (REP05) driving cycles,weighted by their respective proportionof in-use driving. As expected, theaggressive driving of REP05 contributedsignificantly to the difference. Moresurprisingly, however, significantcontributions to the increase also camefrom the Start and Remnant Cycles,particularly for NMHC and NOX.


21 In fact, the Agency believes that the effect onemission values of the additional ten percentdynamometer road load horsepower is negligibleand unobservable within the range of current test-to-test variability.

22 This program was developed as a cooperativeeffort between EPA and manufacturers with fundingfrom manufacturers.

TABLE 3.—CONTRIBUTIONS TO THE IN-USE G/MI INCREASE BY THREETYPES OF DRIVING

Driving NMHC CO NOX

All (In-UseIn-crease).

0.043 g/mi 2.784 g/mi.

0.083 g/mi.

Start ........ 30.2% ....... 17.1% .. 23.0%.Remnant . 33.8% ....... 25.0% .. 45.6%.Aggres-

sive.36.0% ....... 57.8% .. 31.4%.

The AAMA/AIAM portion of theprogram was conducted in late 1993 andearly 1994. This 26-vehicle, 8-manufacturer program included hotstabilized testing with REP05, the 505,and the 866, but none with the Remnantor Start Cycles; thus, a completeassessment of in-use hot stabilizeddriving could not be conducted with themanufacturers’ data. Nevertheless,comparisons were made between theEPA and manufacturer program resultsfor REP05 as well as the differencebetween REP05 emissions and hotstabilized LA4 emissions. In looking atthe emission difference between REP05and hot LA4, the LDV and light light-duty truck (LLDT) average for the EPAtested vehicles was 0.04 g/mi while itwas 0.06 g/mi for the vehicles tested bythe manufacturers. The CO emissionstracked better, with the REP05 and hotLA4 difference of 5.71 g/mi for EPA and5.32 g/mi for the manufacturer tests.The manufacturer testing showed amuch larger NOX differential. The NOX

difference between REP05 and hot LA4was 0.25 g/mi for the manufacturers’testing while only 0.09 g/mi for EPAtesting. The NMHC and CO differencesare primarily among the LLDTs whilethe NOX difference was found in LDVsand LLDTs. The Agency did not test anyheavy light-duty trucks (HLDTs);however, the manufacturers’ resultsshowed these vehicles as having thelargest grams per mile increases fromhot LA4 to REP05. This comparisonsuggests that EPA’s emission assessmentshould provide a reasonable, if notconservative, estimate of in-useemissions.

B. Intermediate Soaks

The Agency conducted theassessment of in-use emissionsfollowing intermediate soaks using datafrom EPA’s Soak/Start Test Program,conducted in two phases between July1993 and June 1994. The testingrepresented the soaks observed in thedriving survey data. The primary cyclesused to measure post-soak emissionlevels for the emission assessment were

variations of EPA’s representative StartCycle (ST01).

Post-soak emissions in the Soak/StartTest Program, measured over the ST01cycle, increased steadily and sharply assoak duration was incremented between10 minutes and 60 minutes. The averageST01 emissions for all vehicles testedfor NMHC, CO, and NOX were higherfollowing the 60-minute soak than theywere for the 10-minute soak by factorsof seven, two, and four, respectively.The increases were significant inabsolute terms as well; for example, theaverage NMHC emissions on three Tier1 vehicles went from about 0.05 g/mifollowing the 10-minute soak to over0.50 g/mi following the 60-minute soak.The rate of increase moderated withsoaks longer than 60 minutes, such thatemissions of all constituents following a2-hour soak were within 50 percent ofcold soak levels. The subset of Tier 1vehicles in the EPA program showedsimilar percentage increases as afunction of soak duration relative to theTier 0 vehicles, although the averageemission levels of these vehicles werelower than the Tier 0 vehicles.

C. In-Use Air Conditioner Operation

The Agency conducted three testprograms and participated cooperativelywith AIAM and AAMA in an additionaltest program during late 1993 and early1994 with the purpose of assessing in-use emissions due to A/C operation.Detailed descriptions of all of theseprograms and the results are containedin the Support Document to theProposed Regulations for Revisions tothe Federal Test Procedure: DetailedDiscussion and Analysis.

The first test program comparedemissions during the current FTP A/Csimulation to emissions obtained withthe A/C actually operating andconfirmed that the current A/Csimulation method significantly under-represents the actual load of the A/C onthe engine.21

The second test program went beyondthe current FTP by testing A/C impactsover the three representative cycles(REP05, ST01, Remnant) as well as overthe LA4. As in the first program, resultsfrom this testing demonstrated anoverall increase in actual emissionswith the A/C operating. In particular,the magnitude of the NOX increase inboth programs was much larger thanexpected and caused the Agency tofocus further research and analysis on

the effects of A/C operation on NOX

emissions.The third test program was very

similar to the second but was designedto collect second-by-second emissionsand vehicle operating data. Analysis ofthese data indicated that the significantA/C-related emission impacts wereoccurring during idles andaccelerations; on the LA4, ST01, andRemnant cycles the combination of idlesand accelerations accounted for morethan 80 percent of the total observedNOX increase. As was the case in theprevious program, the overall increasesin NOX were heavily weighted towardsthe moderate and lower speed driving ofthe ST01, Remnant, and LA4 cycles,although some increases were seen onthe REP05 cycle.

A detriment of these test programs isthat they did not adequately or fullyrepresent the actual conditions underwhich A/C systems are likely to beoperated. To test vehicles under anaccurate simulation of environmentalconditions and vehicle speed, anemission testing program (referred to asthe AC Rochester [ACR] test program)was conducted by vehiclemanufacturers in a sophisticatedenvironmental test facility.22 TheAgency and manufacturerscooperatively defined for the testing aset of environmental and meteorologicalparameters to represent a typical ozonenonattainment day.

Eight vehicles certified to the EPA’sTier 1 emission standards with HFC–134a A/C refrigerant systems weretested in the program. Once again, theeffects of A/C operation were mostpronounced on the moderate-to-lowerspeed cycles. On a hot, stabilized LA4,the average increases were 0.011 g/mifor NMHC, 0.3 g/mi for CO, and 0.205g/mi for NOX. The increases observedon the REP05 cycle were smaller thanon the LA4, but still noteworthy due tothe performance of several of thevehicles, causing the Agency someconcern about the impact of A/Coperation during aggressive drivingbehavior. Fuel economy decreased byabout 13 percent on the REP05 with theA/C operating, substantially less thanthe 20 percent reduction on the LA4,further indicating that the A/C load asa proportion of total load tends todiminish as speeds and accelerationsincrease.


23 AAMA/AIAM spotlighted commandedenrichment by retesting a portion of the vehicles intheir test program in a stoichiometric configuration,as well as in the ‘‘production’’ configuration andprovided second-by-second data acquisitioncapability for emissions and a variety of engine andemission control parameters, allowing fine scrutinyof individual driving events.

D. Additional Elements Contributing toEngine Load

As part of the Non-LA4 Emission TestProgram, EPA conducted an evaluationof emission impacts from road grade bysimulating a two percent grade throughincreased inertia weight at thedynamometer during testing of threevehicles over the three representativecycles. The road grade effect, weightedby the percentages of the driving typesin-use, showed a consistent HC increaseof 0.04 g/mile, a highly variable COincrease averaging 3.2 g/mile, and aNOX increase (due largely to onevehicle) of 0.19 g/mi. Due to the absenceof comprehensive in-use surveyinformation, EPA did not calculateadjustments to these numbers to reflectin-use frequency of grade ormodifications to driving behavior overgrades.

IX. Cause and Control of EmissionsThree candidate areas for emission

control are aggressive driving behavior,intermediate soak periods, and A/Coperation. Microtransient drivingbehavior carries over and is addressedwithing these candidate areas. Thefollowing discusses each of these areas,the causes of emission, and potentialstrategies for controlling the emission.

A. Aggressive Driving EmissionsBoth agencies and the vehicle

manufacturers anticipated that aprimary cause of higher emissionsduring aggressive operation would be‘‘commanded enrichment,’’ which isdone by programming the vehicle’scomputer to change the air/fuel ratio tothe rich side (more fuel for the same air)of stoichiometric operation, typically inresponse to high loads on the engine.Aggressive driving, positive road grade,increased vehicle loading, and airconditioning operation all generateincreased load on the engine. Further,the effect of these factors arecumulative. Manufacturers currentlyemploy commanded enrichment inessentially all applications when highload at the engine (regardless of thesource) is detected, both to provideincreased power and to cool the engineor catalyst.

Using data from EPA’s Non-LA4 TestProgram, supplemented by AAMA/AIAM data,23 the Agency concludedelevated HC and CO emissions during

aggressive driving are due primarily toenrichment, both commanded andtransient. High NOX emissions duringaggressive driving, EPA believes, aredue both to an increase in engine outNOX (from higher temperatures) and torelatively poor catalytic conversion.Poor catalytic conversion is due to leanevents resulting from erratic A/F controland to an A/F control strategy which isnot biased rich. The Agency alsorecognizes that catalyst breakthrough isa potential contributor to CO and NOX

emissions during aggressive driving.The Agency considered five strategies

that manufacturers might employ foraddressing the causes of high emissionsfrom aggressive driving: improvedcontrol of the A/F ratio (fuel control)through calibration; improved fuelcontrol by upgrading fuel injectionsystems to sequential firing; upgradingto electronic throttle control;improvements to catalyst design; andreapplication or refinement ofconventional NOX emission controlsystems. These strategies are discussedin detail in the Technical Reports.

Of these strategies, the variousrecalibration options appeared to be theleast costly, because each of theremaining strategies involved per-vehicle hardware modifications. Inaddition, data from the Non-LA4 testprogram indicated that recalibrationswould probably control the vastmajority of aggressive driving emissions.

B. Intermediate Soak PeriodsThe Agency examined the causes of

post-soak emissions using data from theEPA Soak/Start Test Program and apreliminary program called the AlbanyCooldown Study that gathered real-world engine and catalyst cooldownprofiles. The data from these programsindicated that increased emissionsfollowing intermediate soaks arise inthree ways:

• Rapid catalyst cooldown followingkeyoff,

• Slow catalyst thermal recoveryfollowing a restart, and

• Manufacturer calibration strategiesin response to the startup condition.

The Agency data indicate the catalystcools to below the temperature neededto sustain significant catalytic activity(‘‘light-off’’ temperature) within 20–30minutes of vehicle shutoff, while theengine is still near its normal operatingtemperature. Data also indicated asignificant delay in achieving light-offtemperature upon restart, apparentlydue to the cool initial temperature of theengine-out exhaust. Because tailpipeemissions increase dramatically whenthe catalyst is below light-offtemperatures, the relatively long delay

in achieving light-off results indisproportionately high emissionincreases over intermediate soaks.

The current FTP provides noincentive for manufacturers to retard therapid cooldown of the catalyst duringintermediate soaks. In addition, testingfound differences in engine-outemissions determined by themanufacturer’s calibration strategy uponrestart. Following intermediate-durationsoaks, one vehicle had a lean calibrationstrategy which increased NOX

emissions. Here again, the test resultsindicate that significant emissions maybe occurring in-use because of a lack ofincentive for manufacturers to optimizestartup calibrations followingintermediate soaks.

In general, strategies for reducingpost-intermediate soak emissions arecatalyst-based and either focus on theretarding of catalyst cooldown throughinsulation after the vehicle is shut off orthe enhancement of catalyst light-offupon restart.

Of the potential approachesconsidered for control of intermediatesoaks, EPA is focusing on catalystinsulation as the primary controlstrategy. Use of insulation results ingreater emission reductions overintermediate soaks than strategies whichfocus on improving catalyst light-offthrough conventional means andprovides more cost-effective emissionbenefits than advanced cold startapproaches. Although intermediate soakemissions will likely be reduced tosome extent due to directionalimprovements in cold startperformance, EPA believes that on Tier1 vehicles intermediate soak emissionswill continue to be relatively significantbecause the primary cause ofintermediate soak emissions—rapidcooling of the catalyst—will remainunaddressed. Because insulationdirectly addresses catalyst cooldown,EPA anticipates that this approach willincur significant emission reductionsover intermediate soaks on Tier 1vehicles, including those which willincidentally reduce intermediate soakemissions through improved cold startperformance.

C. Air Conditioner OperationThe Agency focused on the NOX

impacts from A/C use because of thelarge observed increases. The increasesin tailpipe NOX with the A/C operatingseen in the ACR Test Program couldclearly be linked to large increasesobserved in engine out NOX, which areprobably caused primarily by highercombustion temperatures due to theadditional load of the A/C system.Tailpipe NOX can be improved by


24 The relationship between curtailingcommanded enrichment and catalyst deteriorationis addressed in the discussion of feasibility in theTechnical Report.

increasing NOX conversion efficiency inthe catalyst or decreasing engine outNOX. Control strategies includeimproving control of the A/F ratio,eliminating the lean-on-cruisecalibration strategy, adjusting sparktiming, adding or enhancing EGRsystems strategic cycling of the A/Ccompressor, and improving catalysts toenhance NOX conversion efficiency.

The testing at ACR confirmed that HCand CO were also impacted by A/Coperation. The Agency believes thatthese HC and CO increases are relatedto the increased load on the enginetriggering additional periods ofcommanded enrichment when the A/Cis on. The Agency believes that thecontrol strategies for HC and COdiscussed in ‘‘IX.A. Aggressive DrivingEmissions’’ will eliminate HC and COemissions increases due to A/Coperation as well as during aggressivedriving.

X. Options Considered and InformationNeeded

The following outlines the optionswhich EPA has considered indeveloping today’s proposal and issueson which more information is needed.As has been indicated, EPA and otherstakeholders conducted extensiveresearch and examined many options.While today’s proposal selects theapproach EPA felt would provide themost emission benefits feasible, indeveloping the final rule EPA willreconsider each of the options in termsof new research and data submitted. TheAgency welcomes comments andadditional data on these and any otherpoints. A full discussion of these issuesand a detailed analysis of each option isfound in the Support Document to theProposed Regulations for Revisions tothe Federal Test Procedure: DetailedDiscussion and Analysis.

A. Affecting Aggressive Driving CycleThe Agency evaluated three basic

options for establishing standards andvehicle testing aimed at controllingemissions from aggressive driving. Twooptions were based on emissionperformance standards with compliancemeasured using a test cycle, and oneoption was based on a performancestandard using the A/F ratio with arelated test procedure.

The Agency was guided by sevencriteria in evaluating the options. First,EPA sought an option that would leadto control of emissions over the broadrange of aggressive driving behaviorfound in the in-use driving survey data.Second, due to the non-linear nature ofHC and CO emission increase duringenrichment, a high priority was to

ensure sufficient content from thehighest-emission operating modes toprompt manufacturers to employappropriate control strategies, includingcurtailing commanded enrichment.Third, the Agency sought consensuswith CARB, to avoid duplicate orincompatible test requirements. Fourth,EPA sought to reasonably account fortechnical concerns raised by vehiclemanufacturers, particularlymanufacturer comment on the necessityof some commanded enrichment eventsto avoid elevated catalyst temperaturelevels from in-use operation leading tocatalyst deterioration.24 Fifth, EPAsought to pursue cost saving elementslike reduction in test time wherepractical. Sixth, the Agency soughtpractical control of microtransientbehavior, a candidate area of controlthat spans all driving. Finally, EPAfavored strategies to control aggressivedriving emissions that would alsoaddress the potentially significant (butunquantified) emissions from otherengine load factors like road grade.

A full analysis of each option, how itwas evaluated, how the level ofemission control was determined, andthe feasibility of the approach is in theSupport Document to the ProposedRegulations for Revisions to the FederalTest Procedure: Detailed Discussion andAnalysis and Technical Reports andcomment on the analysis is welcome.Comment is specifically solicited on thefollowing items:

• Comment is requested on the needto allow some commanded enrichmentevents during the USO6 Cycle to avoidelevated catalyst temperature levelsfrom in-use operation leading to catalystdeterioration.

• The Agency is proposing that US06HC and NMHC emissions be controlledto the same gram-per-mile emissionlevels currently achieved on the secondbag of the FTP. US06 CO and NOX

emissions are proposed to be controlledto overall FTP emission levels. Theseproposals are based upon the Agency’sanalyses of the potential controltechnology and their related costs andemission reductions, which aredescribed in detail in the TechnicalReports. Comments and additional dataaddressing these proposed levels ofcontrol are solicited. Additionalinformation and data are also requestedabout the potential tradeoffs betweenNOX and CO control during aggressivedriving, and on the impact such

tradeoffs could have on the appropriatelevel of CO control.

• Although concern has beenexpressed that removal of commandedenrichment could impose a 2 percent to10 percent power penalty, EPA believespower enrichment would not beprecluded outright by this proposal, butrather curtailed only within thedurations and speed-accelerationcombinations found in the US06 cycle.Thus, the Agency has concluded on thebasis of available data that compliancewith the US06 standard should have anegligible effect on vehicle performance.Additional data on the effect on vehicleperformance under this proposal isrequested.

• The Agency has proposedadjustments to the US06 for all HLDTsand some LDVs and LDTs. Theseinclude a change in determing inertiaweight for HLDTs, dynamic loadadjustment for low-performancevehicles, and demonstration ofstoichiometric control for wide-openthrottle events for high-performancevehicles. Comments and data aresolicited on the appropriateness of theseadjustments and of the weight-to-powercutpoints. Of special concern is thepossible unfair advantage the proposedhigh performance cut-off may provide tovehicles in the 18–21 W/P range.

• The Agency has proposed a W/P-based measure for the performancecutoffs after also considering thealternative performance criteria basedon a vehicle’s acceleration time fromzero to 60 mph. The Agency rejected thezero to 60 time approach on the basis ofpractical problems related toestablishing appropriate cutoff pointsand a standardized procedure fordetermining zero to 60 times. TheAgency solicits comments on theproposed method for making vehicleperformance adjustments, as well asinput on alternatives, including the onediscussed above.

B. Affecting Start Driving Cycle andIntermediate Soak

The compliance program approachevaluated for intermediate soaks andstart driving was an emissionperformance standard applied to theresults of testing over an emissioncontrol cycle following a soak period ofintermediate duration. As with controlprogram approaches for aggressivedriving emissions, EPA believes that anemission performance standardprovides the most direct method ofcontrolling the emissions arising duringthe particular type of vehicle operation.Given the particular causes of highemissions in this case, use of designstandards or system performance


25 The severity of one SC01 acceleration wasartificially modified to be less severe than in theoriginal microtrip. This preserved the designobjectives of matching the 505 trip distance andreflecting moderate, rather than aggressive driving.The representative level of microtransient behaviorin the cycle was unaffected by this change.

26 Analysis of the two microtrips used to completeSC01 shows higher power levels than thecomparable portion of the 505. The Agency plansto replace these microtrips with those which matchpower levels of the 505 more closely. Thecompleted cycle, known as SC02, will replace SC01and serve the same purpose.

27 Per vehicles with both under-body and close-coupled catalysts, EPA anticipates that only theunderfloor catalyst would need to be insulated.

28 The three options—bench testing,dynamometer simulation, and running test with A/C on—are discussed in more detail in the SupportDocument to the Proposed Regulations forRevisions to the Federal Test Procedure: DetailedDiscussion and Analysis and the Final TechnicalReport on Air Conditioning for the Federal TestProcedure Revisions Notice of ProposedRulemaking U. S. Environmental ProtectionAgency.

standards would be particularlycomplex and restrictive of themanufacturers’ options.

The Agency developed a new StartControl Cycle (SC01) to be used forcontrolling emissions followingintermediate soaks. Initial idles andstart driving are addressed in SC01 byincorporating the EPA Start Cycle(ST01) in its entirety. The balance ofSC01 is composed of two microtrips ofmoderate driving, selected from the in-use survey database in order to bring thetotal distance of the new control cycleup to match the 3.6-mile distance of the505 Cycle; the resulting cycle is 568seconds long.25, 26

A full analysis of the approaches andissues considered, how each wasevaluated, how the level of emissioncontrol was determined, and thefeasibility of the proposed approach isin the Support Document to theProposed Regulations for Revisions tothe Federal Test Procedure: DetailedDiscussion and Analysis and TechnicalReports and comment on the analysis iswelcome. Comment is specificallysolicited on the following items:

• The Agency believes thatmanufacturers should be able to controlemissions on the SC01 cycle followinga soak of from 10- to 60-minutes to thesame gram-per-mile emission levelscurrently achieved on the third bag ofthe FTP. Comment on theappropriateness of this level of standardand method for determining complianceis requested.

• The Agency believes that internalcatalyst insulation does not pose atemperature-based feasibility problemfor underbody catalysts.27 However,EPA had insufficient data to reach afirm view on this issue for the smallnumber of Tier 1 vehicles which mightneed to insulate close-coupled catalysts.Thus, EPA solicits comments or data onthe temperature-based feasibility ofinsulation for close-coupled catalysts.

• The Agency believes thatapplication of catalyst insulation as astrategy for control of emissionsfollowing intermediate soaks is feasible.

Data and comments are solicited on thefeasibility of catalyst insulation and itsimpact on catalyst operation anddurability.

• Comments are solicited onstrategies to mitigate temperatureincreases in the catalyst brought aboutby insulation (such as moving thecatalyst further downstream andsubsequently conserving exhaust heatahead of the catalyst to not impair coldstart performance, or switching to moretemperature-resistant noble metals likepalladium), as well as spinoff effects ofsuch strategies.

• The Agency believes it is necessaryto move forward with an intermediatesoak standard either if a significantproportion of vehicles are certified toTier 1 standards for a significant timeperiod following implementation or if itis cost effective and feasible to pursuecontrol over intermediate soaks onvehicles certified to the lower standards.The Agency requests comment on theissues of cost-effectiveness andfeasibility of an intermediate soakrequirement on vehicles certified tolower emission standards.

• Criteria are being considered topermit manufacturers to forego the datasubmittal requirement for SC01 testingfollowing a 60-minute soak on an enginefamily basis, allowing manufacturers toreduce the SFTP soak duration to 10minutes. Under this option,manufacturers would be allowed tosubmit a technical justificationdemonstrating that an engine familywould clearly pass the intermediatesoak requirement. The Agency solicitscomment on this option and potentialcriteria for granting such a waiver.

C. Affecting Air Conditioner OperationThe Agency analyzed several possible

approaches to compliance testingdesigned to control emissions due toA/C operation. These options hinged ondetermination of two importantelements—the choice of a control cycleand the choice of a methodology forsimulating A/C operation over thatcycle. The Agency pursued a controlprogram for A/C-on emissions thatutilized an emission performancestandard rather than other controloptions.

A full analysis of each optionconsidered, how it was evaluated, howthe level of emission control wasdetermined, and the feasibility of theapproach is in the Support Document tothe Proposed Regulations for Revisionsto the Federal Test Procedure: DetailedDiscussion and Analysis and TechnicalReports and comment on the analysis iswelcome. Comment is specificallysolicited on the following items:

• The control cycle for A/C-relatedemissions being proposed is the 866plus SC01. While the Agency believesthese are the best cycles for A/C control,comments are solicited on thepossibility of substituting the 505component of the LA4 for SC01.Comments are also solicited on whetherfull A/C simulation should be added tothe US06 cycle.

• A cold start test is not included intoday’s proposal, but the Agency doesbelieve that it may be appropriate toreturn to this issue with respect tofuture technologies and future testprocedures and emission standards. TheAgency specifically solicits commentson this issue.

• Independent from determining theappropriate control cycles for testing,the Agency evaluated three principleoptions for simulating A/C operation ona given test cycle.28 The Agencyrequests comment on the potentialapplicability of each option and thevarious methods of implementing eachoption, whether any specific methodshould be retained as an option in thefinal rule to allow for its futuredevelopment and use by petitioning forAgency approval. The ‘‘Nissan-II’’approach is currently being evaluatedby a consortium of auto manufacturersand the Agency expects to review andevaluate the data as soon as it becomesavailable. The Agency specificallyrequests comments and data that wouldallow a better evaluation of thisapproach and its viability, as well assuggested improvements that wouldalleviate the Agency’s concerns, asdetailed in the Support Document to theProposed Regulations for Revisions tothe Federal Test Procedure: DetailedDiscussion and Analysis.

• The Agency has estimated thatvehicles can maintain existing NMHCand CO emission levels with the A/Cturned on. For NOX, the Agencybelieves that 25 percent of the NOX

increase with the A/C engaed is likelyto be unavoidable without increasingthe stringency of the current NOX

standard, but is proposing to control theother 75 percent. The Agency requestscomments on the feasibility of thisproposed level of control and thetechnology implications of controllingto this level.


29 Tier 1 standards were set for two points in theuseful life of a vehicle—50,000 miles (intermediate)and 100,000 miles (full).

D. Affecting Whole Proposal

The Agency evaluated four differentoptions to translate the proposed levelof emission control for US06,intermediate soak, and A/C intocompliance procedures and appropriateemission standards. The first optionwould set stand-alone standards foreach control area. Complianceprocedures and standards would beestablished individually for aggressiveand microtransient driving behavior, A/C, and intermediate soaks. The secondoption would combine the three non-FTP areas of control into a singlestandard. The third option wouldestablish a composite standard based onresults drawn from both the SFTP andthe FTP. While the basic concept issimilar to the second option, theapproach is specifically structured todirectly implement the proposed levelof control for each area using bagweights and to preserve the existing FTPcompliance margins. The fourth optionconsidered by EPA would replace thecurrent FTP with an entirely new FTPthat reflects, as accurately as possible,actual driving behavior.

A full analysis of each option, how itwas evaluated, and the feasibility ofeach approach is in the SupportDocument to the Proposed Regulationsfor Revisions to the Federal TestProcedure: Detailed Discussion andAnalysis. Comment on the analysis iswelcome. Comments are specificallysolicited on the following issues whichrelate to all cycles in the SFTP orchanges to the FTP.

• Use of a composite non-FTPemission standard was chosen as thecentral approach chosen instead ofusing individual stand-alone standards,a single combined stand-alone standard,a replacement FTP, or some otheroption not considered. Stand-alonestandards or a single combined stand-alone standard were not chosenprimarily because of the lack of data todetermine appropriate compliancemargins and the difficulty indetermining a single emission levelgiven the disparity in emission levelsfrom vehicle to vehicle. Replacing thecurrent FTP at this time was not chosenprimarily because revising the existingFTP would potentially impact thestringency of more stringent emissionstandards currently being considered fordifferent parts of the country, such asthe California LEV and ULEV standards,efforts by the Northeast states to adoptCalifornia requirements, and voluntary49-state emissions standards(‘‘FEDLEV’’). Additional informationand data are requested on the use of anyof these approaches. Comments

concerning stand-alone standards, or thesimple average of the compositestandards, should include considerationof how to set appropriate standards forboth intermediate and full useful life.29

Durability procedures for new stand-alone standards should also beaddressed.

• Because replacing the FTP wouldoffer better assurances of in-useemission control and would simplifythe test procedure, EPA believes itmakes sense in the long term toconsolidate all the test requirementsinto a revised FTP. However, to avoidjeopardizing work on more stringentemission standards and to avoiddelaying implementation of today’sproposal, EPA believes it is better toincorporate consolidation of the FTPwith future consideration of tighterfederal standards. Comments aresolicited on when consolidation shouldoccur.

• Under the non-FTP compositeapproach, the bag weights for each cycleare selected to mirror the proposed levelof control determined using the in-usedriving survey data. (A discussion of theproposed level of control for eachpollutant and how it was determinedcan be found in the Final TechnicalReport on Aggressive Driving Behaviorfor the Revised Federal Test ProcedureNotice of Proposed Rulemaking).Comments are requested on the methodused to select each weighting factor orthe weighting given to each bag whendetermining compliance with thecomposite non-FTP emission standards.

• Emission standards are proposed tobe set at current Tier 1 FTP levels, withan adjustment made for NOX, and aretied to future changes in the FTPstandards. Comments on tying the non-FTP composite standards to FTPstandards, the method used fordetermining the standards, the NoX

adjustment provided, or the need forother adjustments are requested.

• The Agency considered separatingLDVs and LDTs but determined drivingbehavior was similar between theseclasses. Some adjustments are providedin the proposal for specific vehicletypes, transmission types, andperformance rating. Comments on themethod used for determining theseadjustments, the need for otheradjustments, or other related issues arewelcome.

• Very little emission data currentlyexists on emission impacts using fuelsother than gasoline during the SFTP.Because of this, EPA considered

exempting alternative- and/or diesel-fueled vehicles from the SFTPrequirements, but decided such vehicleswould be able to comply. Informationand data related to applying today’sproposal to alternative- and diesel-fueled vehicles are welcome.

• The Agency is asking for commentson whether or not it would beappropriate to establish a singleNMHC+NOX standard for stand-aloneA/C or soak/start requirements or for theproposed composite standards.Comments are also solicited on both thepotential emission impacts and costimplications of this proposedalternative.

• Comments are requested on thebenefits and feasibility of the proposedphase-in schedule from MY1998 to2001. The Agency is particularlyinterested in data and comments onhow potential concerns with highercatalyst temperatures should influencelead time, as well as how these concernsshould be balanced with the objective toobtain the emission benefits under thisrulemaking as quickly as possible. If itappears that wholesale elimination ofcommanded enrichment with short leadtime could introduce unanticipatedproblems with catalyst deterioration,EPA may elect to spread theimplementation of the requirementsover a longer period in the final rule.Another option might be to set anintermediate standard level for theinitial phase-in. Comments are solicitedon the relative benefits and costs of anintermediate standard compared to aphase-in directly to the final standards.

• Today’s proposal provides twoblanket, automatic substitutions fromthe SFTP to the FTP to reduce testingcosts and time for manufacturers. Nosubstitution of FTP bags into the SFTPcalculation is allowed. Flexibility inpreconditioning is also provided in theproposal. If stand-alone standards arepromulgated, EPA is considering anexemption from the intermediate soakrequirements. Comments on any ofthese aspects or related matters arerequested.

• Today’s proposal will improve theaccuracy of the dynamometersimulation of actual on-road operationduring vehicle testing. In addition, thechange in dynamometers to improveaccuracy also allows modifying theequivalent test weight requirements toremove the cap. Comments are solicitedon these changes.

• Comments are specifically solicitedon the need for additional lead time toimplement the new road loadrequirements in terms of thedynamometer changes. If data andadditional information submitted


30 DPWRSUM is the sum in the change in power,a statistic which is derived from the vehicle speed.

supports longer lead times, the Agencymay elect to phase in the FTP under thesame phase-in schedule used for thenew SFTP requirements. Under thisalternative, any engine family includedin the SFTP phase-in would also use theimproved road load simulations for FTPtesting. To minimize the laboratoryburden of maintaining two different setsof dynamometers, EPA would like tocouple any phase-in of the new roadload requirements with proceduresallowing an electric dynamometer tosimulate the existing dynamometerload. Comments addressing new roadload lead time should also comment onhow such a simulation could beincorporated.

• Changes to allow ‘‘appropriate’’throttle action and new speed tolerancecriteria are included in today’s proposal.For each test cycle, a range of acceptablespeed variation is created using theDPWRSUM 30 variable. Each drivingcycle has a unique value of DPWRSUM,which is compared to the DPWRSUMcalculated from the driver’s trace (whatthe vehicle actually drove) to determinea valid test. Comments are solicited onthese aspects of today’s proposal,specifically on the proper method forsetting the lower DPWRSUM thresholdfor a valid test.

XI. Environmental and EconomicImpacts

To estimate the emission reductionsassociated with the proposal, theexpected lifetime emission reductionswere determined per vehicle sold afterimplementation of the proposedregulations. Baseline emissions aretaken from the extensive test programsconducted by the Agency and theoriginal equipment manufacturers insupport of the FTP Review Project, asdiscussed earlier. The weighted

averages of the emission results of thesetest vehicles over the various new testprocedures constitute the baselineemissions used in this analysis.

A. Emission Reductions

The emission reductions used in thisanalysis were calculated by subtractingthe proposed level of control for eachcontrol area from the baseline testvehicle emissions. These test vehiclereductions were then weight averaged tosimulate the reductions associated withthe actual in-use vehicle fleet mix. Itshould be noted that the test resultswere derived for an average vehicle witha 50,000 mile catalyst and do notinclude any allowance for in-usecompliance margins. Thus, the emissionbenefits calculated here are likely to beunderstated.

The average emission factor impactsper vehicle associated with theproposed regulations are shown inTable 4. The calculated results for A/Ccontrol listed in Table 4 include a factorto account for driving with the A/C‘‘on’’ versus driving with it ‘‘off.’’ Arecent survey of actual A/C operation inPhoenix, AZ found that the compressorwas engaged about 61 percent of thetime during typical ozone exceedancedays. Thus, the estimated g/mireduction from A/C control wasmultiplied by 0.61 for inclusion inTable 4.

TABLE 4.—AVERAGE EMISSIONFACTOR REDUCTION PER VEHICLE

Control area NMHC(g/mi)

CO(g/mi)

NOX(g/mi)

High speed/accel 0.055 2.39 0.062Soak/start .......... 0.022 0.02 0.037Air conditioning . 0.000 0.00 0.91

These emission reduction numbersconstitute the emission reductionsassociated with the proposedrequirements in g/mi. These g/mi valueswere converted into the estimatedlifetime emission reduction per vehicleusing assumptions about average annualmileage accumulation rates, a discountrate of seven percent, and estimatedsurvival rates. The results are listed inTable 5; a detailed discussion of themethodology can be found in theRegulatory Impact Analysis.

TABLE 5.—DISCOUNTED LIFETIMEEMISSION REDUCTIONS POUNDSPER VEHICLE

Control area NMHC CO NOX

US06 ................. 10.1 441 11.4Soak/start .......... 4.1 4 6.8Air conditioning . 0.0 0 16.9

Total ........... 14.2 445 35.1

The tons per summer day emissionreductions in various years as a result ofthe proposed test proceduremodifications were estimated usingvehicle miles traveled (VMT) fordifferent model year vehicles duringeach year of interest, the emission factorreductions shown in Table 4, and theproposed phase-in schedule. Thesecalculations are show in Appendix B ofthe Regulatory Impact Analysis (RIA)and are summarized in Table 5. Thepercent reduction columns in Table 6compare these estimated tons persummer day (tpsd) emission reductionsto the baseline emissions for the light-duty fleet (cars and trucks). Calculationsfor these percentage reductions areshown in Appendix C of the RIA.

TABLE 6.—FLEET EMISSION REDUCTIONS IN TONS/SUMMER DAY AND PERCENT OF LIGHT-DUTY FLEET

NMHC CO NOX

tpsd % tpsd % tpsd %

2005 .................................................................................................................................................... 404 4 12655 11 1000 92010 .................................................................................................................................................... 577 6 18047 15 1427 122015 .................................................................................................................................................... 694 7 21717 17 1717 142020 .................................................................................................................................................... 765 8 23938 18 1892 14

B. Economic Impact

The proposed additions to emissiontest procedures will impose severalcosts on the original equipmentmanufacturers. These costs includeadded hardware for improved emission

control and associated development andredesign costs, improved engine controlcalibrations, and increased costsassociated with the certification processincluding durability data vehicle testingand reporting.

The cost estimates correspond to costsincurred by the manufacturer incomplying with the proposedrequirements. These costs can bedivided into fixed and variable costs.Fixed costs are those costs made priorto vehicle production and are relatively


31 From Cost Projections, FFA, 1992, updatedfrom DOE/EIA Monthly Energy Review, May 1994,and DOT/FHA. According to FHA, average sales-weighted state taxes for gasoline were 18.54¢ inJune 1994. Federal tax is 18.4¢.

independent of production volumes.The fixed costs considered in thisanalysis are those for engine controlrecalibration, vehicle redesign,mechanical integrity testing onredesigned engine families, certificationdurability demonstration, annualcertification costs, and test facilityupgrades and construction. Variablecosts are costs for the necessaryemission control hardware and are, bynature, directly dependent onproduction volume. Table 7 presents asummary of the cost estimatescalculated by the Agency. Discussion of

the assumptions and data included inthese estimates can be found in the RIA.

TABLE 7.—REGULATORY COSTESTIMATES

Annual cost($ million)

Cost/vehicle($)

US06 ............. 16.8 1.12Soak/start ...... 139.4–187.0 9.30–12.47A/C ................ 18.3 1.22

Totals ..... 174.5–222.1 11.63–14.81

C. Cost-EffectivenessThe cost-effectiveness estimate

represents the expected cost per ton of

pollutant reduced. The costs presentedin Table 7 are not necessarily equallyspread among the three pollutantemissions (NMHC, CO, and NOX). Sincethe requirements associated with A/Care targeted for NOX control, all costsassociated with A/C have been allocatedto NOX. For US06, the costs associatedwith each area have been allocatedequally across each pollutant. As the COreduction from soak/start is minimal,the costs associated with soak/start havebeen split equally between NMHC andNOX. Table 8 contains the per vehiclecost allocation to each pollutant withineach control area.

TABLE 8.—COST ALLOCATION ($/VEHICLE)

NMHC CO NOX Total

US06 costs ................................................................................................................... 0.37 0.37 0.37 1.12Soak/start costs ............................................................................................................ 4.65–6.23 0.00 4.65–6.23 9.30–12.47A/C Costs ..................................................................................................................... 0.00 0.00 1.22 1.22

Total ...................................................................................................................... 5.02–6.61 0.37 6.24–7.83 11.63–14.81

Dividing the costs shown in Table 8 by the lifetime emission reductions shown in Table 5, gives the cost-effectivenessestimates shown in Table 9.

TABLE 9.—COST-EFFECTIVENESS ESTIMATES ($/TON)

Control area NMHC CO NOx

US06 ...................................................................................................................................................................... 74 2 65Soak/start ............................................................................................................................................................... 2291–3072 NA 1362–1827A/C ......................................................................................................................................................................... NA NA 153

Total ................................................................................................................................................................ 707–930 2 355–445

D. Consumer Impacts

Two impacts on value to theconsumer not included in the aboveestimates are potential savingsassociated with reduced fuelconsumption and impact on thehorsepower output of some vehicleengines. As previously discussed, EPAexpects manufacturers to eliminate orgreatly reduce the amount ofcommanded enrichment currently usedin order to meet the NMHC and COstandards for the US06 control cycle.Due to the lower fuel consumptionassociated with stoichiometric air/fuelcontrol as compared to commandedenrichment, this action will result bothin a small improvement in fuel economyand a small loss in horsepower output.The Agency approximated the fueleconomy benefit by determining howmuch extra fuel is used duringcommanded enrichment operatingmodes and the in-use incidence of thesecommanded enrichment operatingmodes. The result was an estimated 0.51percent reduction in fuel consumption.Using this fuel consumption reductionand multiplying it by the miles drivenin a given year, the appropriate survival

rate and a seven percent discount factor,results in an estimated lifetime fueleconomy savings of $16.56, based on agasoline cost of $0.80 per gallon,excluding state and federal taxes.31 Amore detailed discussion of fueleconomy cost savings can be found inthe RIA for this rule.

Accompanying the lost horsepoweroutput will be the potential for someconsumers to consider such affectedvehicles as having less value. TheAgency does not believe that this lostvalue will be noticed by mostconsumers, as the horsepower loss isquite small, but acknowledges itspotential effect nonetheless. Due to thedifficult nature of trying to quantify acost associated with reduced poweroutput, or reduced 0 to 60 mphacceleration time, etc., the Agency hasnot been able to quantify the loss inconsumer value. However, the Agencybelieves that this cost should be roughlynegated by the associated savings in fuel

expenses. Comments and data aresolicited on ways to quantify theconsumer value of the power loss.

The Agency does not anticipate thattoday’s proposal will have any impacton Inspection/Maintenance programs.

XII. Public Participation

A. Comments and the Public Docket

The Agency welcomes comments onall aspects of this proposed rulemaking.All comments, with the exception ofproprietary information, should bedirected to the EPA Air Docket Section,Docket No. A–92–64 (see ADDRESSES).Commenters who wish to submitproprietary information forconsideration should clearly separatesuch information from other commentsby:

• Labeling proprietary information‘‘Confidential Business Information’’and

• Sending proprietary informationdirectly to the contact person listed (seeFOR FURTHER INFORMATION CONTACT) andnot to the public docket.

This will help ensure that proprietaryinformation is not inadvertently placedin the docket. If a commenter wants


EPA to use a submission labeled asconfidential business information aspart of the basis for the final rule, thena nonconfidential version of thedocument, which summarizes the keydata or information, should be sent tothe docket.

Information covered by a claim ofconfidentiality will be disclosed by EPAonly to the extent allowed and by theprocedures set forth in 40 CFR part 2.If no claim of confidentialityaccompanies the submission when it isreceived by EPA, the submission may bemade available to the public withoutnotifying the commenters.

B. Public HearingAnyone wishing to present testimony

about this proposal at the public hearing(see DATES) should, if possible, notifythe contact person (see FOR FURTHERINFORMATION CONTACT) at least sevendays prior to the day of the hearing. Thecontact person should be given anestimate of the time required for thepresentation of testimony andnotification of any need for audio/visualequipment. A sign-up sheet will beavailable at the registration table themorning of the hearing for schedulingthose who have not notified the contactearlier. This testimony will bescheduled on a first-come, first-servedbasis, and will follow the testimony thatis arranged in advance.

The Agency recommends thatapproximately 50 copies of thestatement or material to be presented bebrought to the hearing for distribution tothe audience. In addition, EPA wouldfind it helpful to receive an advancecopy of any statement or material to bepresented at the hearing at least oneweek before the scheduled hearing date.This is to give EPA staff adequate timeto review such material before thehearing. Such advance copies should besubmitted to the contact person listed.

The official records of the hearing willbe kept open for 30 days following thehearing to allow submissions of rebuttaland supplementary testimony. All suchsubmittals should be directed to the AirDocket, Docket No. A–92–64 (seeADDRESSES).

The hearing will be conductedinformally, and technical rules ofevidence will not apply. Writtentranscripts of the hearing will be madeand a copy thereof placed in the docket.Anyone desiring to purchase a copy ofthe transcript should make individualarrangements with the court reporterrecording the proceeding.

XIII. Administrative Designation

Under Executive Order 12866 (58 FR51735), the Agency must determinewhether the regulatory action is‘‘significant’’ and therefore subject toOMB review and the requirements ofthe Executive Order. The Order definesa ‘‘significant regulatory action’’ as onethat is likely to result in a rule that may:

(1) Have an annual effect on theeconomy of $100 million or more oradversely affect in a material way theeconomy, a sector of the economy,productivity, competition, jobs, theenvironment, public health or safety, orState, local, or tribal governments orcommunities;

(2) Create a serious inconsistency orotherwise interfere with an action takenor planned by another agency;

(3) Materially alter the budgetaryimpact of entitlements, grants, user fees,or loan programs or the rights andobligations of recipients thereof; or

(4) Raise novel legal or policy issuesarising out of legal mandates, thePresident’s priorities, or the principlesset forth in the Executive Order.

Pursuant to the terms of ExecutiveOrder 12866, it has been determinedthat this rule is a ‘‘significant regulatoryaction’’ because of annual impacts onthe economy that are likely to exceed$100 million. As such, this action wassubmitted to OMB for review. Changesmade in response to OMB suggestions orrecommendations will be documentedin the public record.

XIV. Regulatory Flexibility Act

The Regulatory Flexibility Act of 1990requires federal agencies to identifypotentially adverse impacts of federalregulations upon small entities. Ininstances where significant impacts arepossible on a substantial number ofthese entities, agencies are required toperform a Regulatory FlexibilityAnalysis (RFA).

The Agency has determined that thisaction will not have a significant impacton a substantial number of smallentities. This regulation will affect onlymanufacturers of motor vehicles, agroup which does not contain asubstantial number of small entities.

Therefore, as required under section605 of the Regulatory Flexibility Act, 5U.S.C. 601 et. seq., I certify that thisregulation does not have a significantimpact on a substantial number of smallentities.

XV. Reporting and RecordkeepingRequirement

The information collectionrequirements in this proposed rule havebeen submitted for approval to theOffice of Management and Budget(OMB) under the Paperwork ReductionAct, 44 U.S.C. 3501 et seq. AnInformation Collection Requestdocument has been prepared by EPA(ICR No. 2060–0104) and a copy may beobtained from Sandy Farmer,Information Policy Branch, EPA, 401 MSt., SW (Mail Code 2136), Washington,DC 20460 or by calling (202) 260–2740.

The information collection burdenassociated with this rule (testing, recordkeeping and reporting requirements) isestimated to average 566 hours annuallyfor a typical manufacturer. However, thehours spent annually on informationcollection activities by a givenmanufacturer depends uponmanufacturer-specific variables, such asthe number of engine families,production changes, emissions defects,and so forth. The burden estimateincludes such things as reviewinginstructions, searching existing datasources, setting up and maintainingequipment, performing emission testing,gathering and maintaining data,performing analyses, and reviewing andsubmitting information.

Send comments regarding the burdenestimate or any other aspect of thiscollection of information, includingsuggestions for reducing this burden toChief, Information Policy Branch, EPA,401 M St., SW (Mail Code 2136),Washington, DC 20460 and to the Officeof Information and Regulatory Affairs,Office of Management and Budget,Washington, DC 20503, and marked‘‘Attention: Desk Officer for EPA.’’ Thefinal rule will respond to any OMB orpublic comments on the informationcollection requirements contained inthis proposal.

List of Subjects in 40 CFR Part 86

Administrative practice andprocedure, Air pollution control,Confidential business information,Environmental protection, Gasoline,Imports, Labelling, Motor vehicles,Motor vehicle pollution, Reporting andrecordkeeping requirements.

Dated: January 31, 1995.Carol M. Browner,Administrator.[FR Doc. 95–2833 Filed 2–6–95; 8:45 am]BILLING CODE 6560–50–P

Documents

7404 Federal Register /Vol. 60, No. 25/Tuesday, February 7 ... · PDF file7404 Federal Register/Vol. 60, ... speed and/or high acceleration) driving ... No. 25/Tuesday, February 7,