United States Air and Radiation EPA420-P-99-009 Environmental Protection March 1999 Agency M6.EXH.007 vvEPA Determination of NOx and HC Basic Emission Rates, OBD and I/M Effects for Tier 1 and Later LDVs and LDTs DRAFT > Printed on Recycled Paper ------- EPA420-P-99-009 March 1999 of HC and I/M for 1 and and M6.EXH.007 John W. Koupal Edward L. Glover Assessment and Modeling Division Office of Mobile Sources U.S. Environmental Protection Agency NOTICE This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using data which are currently available. The purpose in the release of such reports is to facilitate the exchange of technical information and to inform the public of technical developments which may form, the basis for a final EPA decision, position, or regulatory action. ------- 1 Introduction This report discusses proposed MOBILE6 NOx and exhaustHC basic emission rates (BERs) for Tier 1 and later light-duty vehicles (LDVs) and light-duty trucks (LDTs), and the effects of On- Board Diagnostic (OBD) systems and Inspection/Maintenance (I/M) on these vehicles. The methodology discussed in this paper can be applied to generate BERs across all vehicle classes (LDV and LDT1 through 4) for all Tier 1 and later standards, including the TLEV, LEV, and ULEV standards under the California LEV I and Federal NLEV program, and LEV n standards recently adopted by California. For brevity, however, the results presented here focus on the Tier 1, LEV and ULEV standards for each vehicle class. Sufficient in-use data on LDVs or LDTs complying with Tier 1 or later standards were generally not available at the time of the analysis. Thus, the methodology used in the development of Tier 1 and later BERs is based on the differences in certification standards across standard level and vehicle class. For NOx, Tier 1 and later BERs were based on a sample of 186 LDVs certified to a 50,000 mile 0.4 gram/mile NOx standard (the Federal Tier 1 standard). For HC, Tier 1 and later emission rates were based on BERs developed for 1988 through 1993 Ported Fuel Injection (PFI) LDVs. Using certification standard as the base for Tier 1 and later BERs has two implicit ramifications. First, the BERs reflect the fuel which a vehicle is certified on the meet the standard: Indolene for Tier 1 standards, and California Phase IIRFGfor LEV and later standards. Second, the HC BERs are expressed as non-methane hydrocarbon (NMHC) for Tier 1 LDV/LDTs, and non- methane organic gas (NMOG) for LEV and later LDVs/LDTs. On-Board Diagnostics systems were required on all LDVs and LDTs sold outside California beginning in 1996. Tier 1 vehicles began entering the fleet in 1994, and for two years (1994 and 1995) were not equipped with OBD. For MOBILE6, it will be assumed that all 1996 and later LDVs and LDTs are equipped with OBD systems, which are designed to detect emission system malfunctions resulting in emissions at or above 1.5 times the applicable emission standard.1 If this criteria is met, a light on the vehicle's dashboard (the malfunction indicator light, or MIL) is illuminated to alert the driver that an emissions system repair is required. Thus, the rate of emission deterioration for Tier 1 and later vehicles must take into account the impact OBD systems will have overall in-use emissions, including a) the effectiveness of these systems in detecting emission malfunctions, b) the owner response rate to illuminated MILs, and c) the effectiveness of repair in addressing the detected problem. Beginning in 2001, all Inspection/Maintenance programs will require an OBD system check for OBD-equipped vehicles. In I/M areas, this will greatly increase the rate at which illuminated MILs are addressed, hence further improving the average rate of in-use deterioration for Tier 1 and lrThe "1.5 times the standard" criteria was initially required by ARB, while EPA adopted a different malfunction threshold approach. However, manufacturers were allowed to meet the federal program through compliance with ARB's requirement, and most chose this option. EPA's requirement has recently been amended to harmonize with ARB by requiring the "1.5 times the standard" criteria for vehicle sold federally. For MOBILE6, it will be assumed that all vehicle equipped with OBD since 1996 comply with the "1.5 times" malfunction criteria. DRAFT M6.EXH.007 1 March 1999 ------- later vehicles. Most Tier 1 and later vehicles will be equipped with an OBD system and, if in an I/M area, subject to OBD-based I/M rather than traditional exhaust I/M. However, some Tier 1 vehicles will not be equipped with OBD (model years 1994 and 1995), and some OBD-equipped vehicles will be subject to an exhaust I/M program (prior to calendar year 2001). To model emissions under these scenarios, a methodology for generating basic emission rates was developed for the following cases: No OBD/No EVI (Base) applies to pre-OBD Tier 1 vehicles (1994 and 1995 model years). It is also used as a basis for the computation of BERs under the OBD-only and OBD/EVI cases. OBD-Onlv (OBD/ No EVT) applies to 1996 and later OBD-equipped vehicles where an I/M program is not present. OBD/IM applies to 1996 and later OBD-equipped vehicles where an I/M program which conducts OBD checks is present. An exhaust test may or may not be performed; it is not differentiated from an I/M program with both OBD checks and exhaust testing, since additional I/M reductions are not given for exhaust testing if OBD checks are performed. Exhaust I/M represents a situation in which only an exhaust test is conducted in an I/M program (an EVI240, ASM, or Idle test). This will apply to 1994 and 1995 model year Tier 1 vehicles in all calender years, and all vehicles in calender years prior to 2001. This report gives an overview of the basic approach for generating BERs, then describes the specific details of BER development for NOx and exhaust HC for each case. 2 Basic Emission Rate Derivation Concept The basic concept underlying the generation of Tier 1 and later BERs is similar to the approach used to develop the I/M credits for 1981 through 1993 vehicles.2 For the No OBD/No EVI case, this concept segregates in-use vehicles into "normal" and "high" emitters. High emitters are those vehicles which have emission control systems which are malfunctioning in some way, and are producing average emission levels which are considerably higher than the overall mean emission levels; the threshold for defining a high emitter for NOx and HC is 2.0 times the intermediate life (50,000 mile) certification emission standard. The remainder of the fleet are considered to be properly functioning, and are considered normal emitters; by definition, these vehicles are below 2.0 Glover, E., and Brzezinksi, D., "MOBILE6 Inspection/Maintenance Benefits Methodology for 1981 through 1993 Model Year Light Vehicles", Draft MOBILE6 Report No. M6.IM.001, March 1999. Hereafter referred to as "Tier 0 I/M Report" DRAFT M6.EXH.007 2 March 1999 ------- times the intermediate life certification standard. It is important to note that both pollutants are considered independently when determining whether a vehicle is a high emitter. Thus, a vehicle could be a high NOx emitter, but a normal HC emitter. Although the segregation of vehicles into the "normal" and "high" categories (and their thresholds) is a somewhat arbitrary modeling method, the concept that average in-use emissions are driven by a group of vehicles emitting well above the applicable standard is supported by data from many years of EPA vehicle test and repair programs. This phenomena is contributed to by the "go/no go" nature of emission control technology components such as EGR valves and air pumps, and the high sensitivity of emissions to degradations in performance of other critical components such as the catalyst, oxygen sensor(s) and fuel injectors. Two important assumptions in the development of BERs for Tier 1 and later vehicles are a) the rate at which vehicles malfunction and become high emitters is independent of the certification standard level, and b) the average emission levels for high emitters becomes higher relative to the standard&s the certification standard becomes lower. Under this methodology, average in-use emissions are computed as a weighted average of high and normal emitters. Figure lisa general graphical view of the concept with the y-axis representing emissions in grams per mile, (or grams for start emissions) and the x-axis representing mileage.3 Three lines are shown in Figure 1 which show a) the average or basic emission rate, b) the normal emitter emission rate, and c) the high emitter emission level. Figure 1 GENERAL EMISSION FACTOR SCHEMATIC EMISSIONS MILEAGE MOBILE6 uses vehicle mileage as a surrogate for vehicle age. Age and mileage are used interchangeably throughout this document. DRAFT M6.EXH.007 March 1999 ------- The basic emission rate is shown as Line A. This line represents the average emissions of the fleet as a function of both normal emitters and high emitters. Line B represents the average emissions of normal-emitting vehicles. These are the vehicles which have emission control systems which are generally performing as designed. The line is shown as a linear function of mileage to reflect the gradual deterioration that normal vehicles experience, primarily due to catalyst degradation over the life of the vehicle. Normal emitter emissions are generally expressed by a least squares linear regression of emissions versus mileage. Line C represents the average emissions of high-emitting vehicles. This line is a flat horizontal line because emissions from these vehicles do not appear to be a strong function of mileage, based on previous analysis of Tier 0 data4 and born out by analyses of Tier 1 NOx data presented in the following section. The underlying phenomena expressed here is that emission control malfunction will drive high emissions regardless of vehicle mileage; as discussed in subsequent sections, what changes as the vehicle ages is the probability of malfunction, rather than the emission levels resulting from a malfunction. Line A represents the weighted average of lines B and C, based on appropriate weighting factors for normal and high emitters. On a fleet-wide basis, this weighting factor represents the fraction of high emitters in the fleet, as a function of vehicle age; on a per-vehicle basis, this weighting factor can be considered to be the probability the vehicle will be a high emitter at a given age. This weighting factor can be derived at any given vehicle age A (represented by vehicle mileage) by transforming Equations 1 and 2 into Equations 3 and 4 below. 4Tier 0 I/M report DRAFT M6.EXH.007 4 March 1999 ------- Highs + Normals = 1 Eqn 1 and AVE = High_ave * Highs + Norm_ave * Normals Eqn 2 Solving for the variables Highs and Normals produces: Highs = (AVE - Norm_ave) / (High_ave - Norm_ave) Eqn 3 Normals = 1 - Highs Eqn 4 Where: Highs = fraction of High emitters, age = A Normals = fraction of Normal emitters, age =A AVE is the average emission rate, age = A High_ave is the high emitter emission average (independent of age) Norm_ave is the normal emitter emission average, age = A 3 NOx BERs and Emitter Fractions: No IM/No OBD Case 3.1 TierlLDVs The No EVI / No OBD case was developed first because it did not require accounting for high emitters which underwent repair due to OBD MIL-on; hence, the methodology closely followed the basic emission rate derivation concept outlined in the previous section. Tier 1 LDVs served as the basis for developing BERs across vehicle class (LOT 1 through 4), and standard level (LEV, ULEV) for both NOx and NMHC. Thus, the derivation of these BERs is the first step in the generation of all BERs for all OBD/EVI cases, vehicles classes and standards. Tier 1 and later BERs were first generated in FTP space and subsequently split into running and start components, as discussed in Section 7; this deviates from the approach used 1981 through 1993 vehicles, for which start and running BERs were developed independently. The data used to generate No OBD/No EVI BERs for Tier 1 LDVs were 186 LDVs and LDTs tested by the California Air Resources Board (ARE) as part of Surveillance Programs 12 through 14; these data were provided to EPA by ARB specifically for this analysis. The model years of these vehicles ranged from 1989 through 1996, and all were certified to a 50,000 mile intermediate useful life NOx standard of 0.4 g/mi, the same as the Federal Tier 1 LDV/LDT1 standard. Figure 2 graphically presents the ARB data, in terms of FTP emissions versus mileage (in ten thousand mile units). DRAFT M6.EXH.007 5 March 1999 ------- Figure 2 - ARE Surveillance Data FTP NOx Results ARB Surveillance Data - FTP Results 0.4 NOx LDV/LDT 2.2 2.0 - 1.8 1.6 - = 1.4 - 1 1.2 5 o 1-° • 0.6 0.4 - 0.2 On * • • • High Emitters • « « • * * * • ** * 2,0 x Standard • » « * * * * * . *vv * * . • .* »i »„*** %»»IX<.V» .-••:. * * * * ^i* * «*i ^A **^ * * * Normal Emitters * * JM»* *^ ^ *. » *^ ** * * .^ A. * * * ^***»* *^ ^ ^ • ** 0 2 4 6 8 10 12 14 16 18 20 Mileage/10000 Using these data as a starting point, emissions for normal emitters, high emitters and the fraction of high emitters were derived, as outlined in the following sections. 3.1.1 Normal and High Emitter Emission Level The average FTP normal emitter emission level was obtained by separating the normal emitters from the high emitters in the ARB sample according to the "2.0 times the standard" criteria (i.e. all vehicles in the sample above 0.8 g/mi were defined as high emitters, as shown in Figure 2). Using this cutpoint, 19 vehicles were defined as high emitters, and the remainder (167) were defined as normal emitters. The BER for normal emitters was developed by fitting a linear regression of FTP emissions versus mileage; the result of this regression is shown in Equation 5 (the variable 'odom' is in units often thousand miles). Norm_Ave(g/mi) 0.265 + 0.0147 * odom Eqn 5 The high emitter NOX emission level was obtained by calculating the mean emission level of the 19 vehicles defined as High Emitters: High_Ave(g/mi) 1.278 Eqn 6 DRAFT M6.EXH.007 March 1999 ------- 3.1.2 High Emitter Fractions As discussed in Section 2, high emitter fractions allow the computation of average in-use emissions based on normal and high emitting BERs. These fractions increase with vehicle age, and for OBD-equipped vehicles will be used as the basis for computing OBD and/or I/M benefits. It is at this stage of the NOx BER computation that key assumptions regarding the representativeness of recruitment-based5 in-use emission testing program become relevant. In general, there is concern that recruitment-based in-use emission testing programs are less likely to procure dirtier vehicles, and hence do not gather data over the full range of the in-use fleet. These programs typically have low response rates (less than 25 percent), and in-use data from programs with higher rates of participation (e.g. I/M programs) show higher emissions than voluntary recruitment programs. As such, for the purpose of developing MOBILE6 basic emission rates for 1981 through 1993 vehicles, a "high emitter adjustment factor" was added to voluntary recruitment data to account the potential bias in recruitment.6 Since the ARE surveillance program is also a recruitment-based program, a similar methodology was applied for Tier 1 and later vehicles In addition to a high-emitter correction factor, a second adjustment was required to "remove" the impact of California's I/M program, which he vehicles included in the ARB sample were subjected to. This was necessary in order to develop basic emission factors which reflected no I/M program. Both the high-emitter and "no I/M" adjustments are relevant at this stage of the calculation because they only affected the weighting of high and normal emitters - in other words, they were used to increase only the number of high emitters estimated in the fleet. As discussed in Section 2 (and shown in Equations (1) through(4)), the high/normal weighting factor is a function of normal and high emitter emission levels and their combined average in-use emission level. Emissions for normal and high emitters as computed in Equations (5) and (6) were not affected by these adjustments. The high-emitter and "no I/M" adjustments were instead applied directly to the in-use average emission levels; based on normal and high emitter emissions computed from Equations (5) and (6), the high emitter fractions were then derived using Equations (1) through (4). This had the effect of increasing the weighting of high emitters in the fleet relative to the pre-adjustment level. This methodology is detailed in the following steps: (1) The unadjusted average in-use NOx emission level as a function of mileage was computed from the ARB dataset. This FTP NOx emission average was obtained by linear regression 5"Recruitment-based" in-use emission testing programs are defined here as programs in which vehicles are procured for testing from the general population in exchange for money and/or other incentives. Participants are initially contacted through mail or phone solicitation based on registration mailing lists, and participation is strictly voluntary. 6Enns, P. et al, "Determination of Running Emissions as a Function of Mileage for 1981 through 1993 Model Year Light-Duty Cars and Trucks", MOBILE6 Report No. M6.EXH.OO. Referred to as "Tier 0 Running BER Report" DRAFT M6.EXH.007 7 March 1999 ------- of the raw ARB data (all 186 vehicles) versus mileage. The intercept of the regression was set equal to the normal emitter intercept from Equation (5) (0.265 g/mi). The resulting regression equation is as shown: NOx FTP (g/mi) = 0.265 + 0.031 * Odom Eqn 7 (2) The second step was to generate an additive high emitter correction factor to account for the potential effects of recruitment bias on the ARB sample. The Tier 0 high emitter correction factor proposed for 1988 through 1993 PFILDVs (obtained from comparison of EVI240 data and data collected through voluntary recruitment programs)7 was calculated in FTP space, and fit with a linear regression as a function of mileage. The Tier 0 adjustment was reduced by 25 percent to account for reductions in high emitter emissions expected between Tier 0 and Tier 1 vehicles, and to prevent unduly large high emitter fractions. The resulting high emitter correction factor is shown in Equation 8. It is in units of FTP grams per mile, and is a function of mileage. This adjustment is added to the average NOx emission function from Step 1 above. HECF = 0.00795 * Odom Eqn 8a Where: HECF is the high emitter correction factor. Odom is the mileage, in ten thousands At zero miles, this adjustment adds 3 percent to the average in-use emission level; at 100,000 miles it adds 14 percent. These percent increases are comparable to those calculated from the data for 1988 through 1993 PFI LDVs with and without a high-emitter adjustment. (3) The third step was to generate a multiplicative "No I/M" correction factor to account for the I/M effect in the ARB sample. Most vehicles under California's Smog Check II program will undergo a basic idle I/M test,8 adjustment factors by vehicle age were developed by using MOBILES program with and without an idle I/M program (using by-model year output for a run in the year 2010). At each vehicle age, a correction factor based on the ratio of the "without I/M" and "with I/M" runs was computed; these factors were regressed with mileage to create a correction factor as a function of mileage. The resulting equations are shown as follows: 7Tier 0 Running BER Report 8"Smog Check Program Fact Sheet: A History of Inspection and Maintenance Programs in California", Smog Check Web Site (http://smogcheck.ca.gov), July 1997 DRAFT M6.EXH.007 8 March 1999 ------- NOIMC = 1.057- 0.0038 * Odom Eqn 8b (NOIMC > 1.00) NOIMC = 1.00 EqnSc (NOIMC < 1.00) NOIMC = 0.00 EqnSd (Odom = 0) (4) The corrected in-use average NOx FTP results (C_NOXFTP) were obtained by applying the high emitter and No I/M correction factors from Equations 8a through 8d to the raw NOx FTP value from Equation 7, as shown in Equation 9. C_NOXFTP = (NOx FTP + HECF) * NOIMC Eqn 9 (5) The fraction of high emitters in the fleet under NO I/M and NO OBD conditions (labeled HighBASE) is calculated by inserting the value of C_NOXFTP, Norm_ave, and High_ave (from Equations (5) and (6)) into Equation (3). Mathematically, this is shown in Equation 10. HignBASE = (C_NOXFTP - Norm_ave) / (High_ave - Norm_ave) Eqn 10 The resulting high emitter fractions for the No OBD/No EVI case for ages one through 25 are shown in Appendix A, Table A-l (mileage levels as a function of age are shown in Appendix A, Table A-3). The average in-use FTP-based NOx emission level for Tier 1 LDVs can be calculated at any vehicle age using Equation (2), based on the terms "High_ave" (Equation 6), "Norm_ave" (Equation 5), and "Highs" (Equation 10). According to Equation (4), "Normals" is simply 1 - Highs. 3.2 All Other Standard Levels and Vehicle Classes NOx BERs for No OBD / No EVI conditions are required for LDVs under post-Tier 1 standards, and all Tier 1 and later LDTs (LDT1 through 4). At the time of this analysis, EPA was not aware of any dataset which provided an adequate sample of in-use data for these combinations of vehicles class and standard level. BERs for these classes were derived from the Tier 1 LDVBERs developed above, using a set of specific assumptions about how average emissions for normal and high emitters, and high emitter fraction, would apply across standard level and class. DRAFT M6.EXH.007 9 March 1999 ------- 3.2.1 Normal Emitters It was assumed that for post-Tier 1 LDVs and Tier and later LDTs, normal emitter NOx emissions will on average maintain the same performance relative to the applicable 50,000 mile standard as Tier 1 LDVs. Thus, normal emitter BERs for all post-Tier 1 LDVs and Tier 1 and later LDTs were developed by calculating the ratio of the applicable standard level ("std") to the Tier 1 LDV standard, and applying this ratio to the Tier 1 BER (zero-mile level and deterioration rate), as follows: Norm_ave ZML(std) = Norm_ave ZML(tierlldv) * (Cert Std(std) / Cert Std(tierl Idv)) Eqn 1 la Norm_ave DR(std) = Norm_ave DR(tierl Idv) * (Cert Std(std) / Cert Std(tierl Idv)) Eqn 1 Ib For example, normal-emitting LDV LEV BERs were generated by multiplying the normal-emitting Tier 1 ZML and DR from Equation (5) by 0.5 (0.2 g/mi divided by 0.4 g/mi, the 50,000 miles standards). The rationale behind this approach is that basic emission levels for properly operating vehicles should receive the full benefit of reduced standards, including lower deterioration rates for lower standard levels. This approach presumes that normal emitters for all standards and vehicle classes will on average achieve the same compliance margin ("headroom") with the 50,000 mile certification standard as the normal emitters observed in the ARB data. With regards to trucks, this approach acknowledges that LOT emission performance relative to the standard is expected to be similar to LDVs because of increased similarities in a) emission control technology, b) manufacturer design practices, and c) driving and usage patterns. 3.2.2 High Emitters High Emitter BERs are meant to estimate emissions from vehicles that significantly exceed their certification standards due to malfunctioning emission control systems. A key assumption in the development of high-emitter BERs is that, as emission standards are lowered (or "raised" for truck BERs), emission levels for high emitters will not be changed in proportion to the standard change. Because emission control equipment on these vehicles will be degraded or completely malfunctioning, high emitter emissions would be less dependent of certification standard level. To illustrate, compliance with LEV NOx standards are expected to result from improved catalyst performance rather than reductions in engine-out emissions; as such, engine-out emissions from LEVs and Tier Is are likely more similar than the difference in standards, on a relative basis. A degraded catalyst on a LEV would thus result in emissions more similar to a Tier 1 with a comparably degraded catalyst. BERs for high emitters were developed for post-Tier 1 LDVs and Tier 1 and later LDTs by taking the average of the Tier 1 LDV high emitter NOx BER (1.278 g/mi) and the BER that would result if the ratio of 50,000 mile standards were applied to the Tier 1 BER, according to Equation 12: DRAFT M6.EXH.007 10 March 1999 ------- High_ave(std) = average[High_ave(tierlldv) * (Cert Std(std) / Cert Std(tierl Idv), High_ave(tierlldv)] Eqn 12 The result of this average is a high emitter BER which is 50 percent proportional to the change in standard, reflecting that malfunctioning vehicles will derive some benefit on average from lower emission standards, but not the full benefit as afforded to normal emitters. High emitter emissions are therefore tied closely to the Tier 1 LDV high emitter level, and the change in high emitter emissions is "muted" relative to changes in standard. Normal and high emitter BERs for LDVs and LDTs complying with Tier 1, LEV and ULEV standards are presented in Appendix B. 3.2.3 High Emitter Fractions The rate at which vehicles become high emitters under the No OBD / No EVI scenario was assumed constant for all vehicles and standard classes. Thus, the age-based high emitter fractions developed in Equation 10 and presented in Appendix A were applied to Tier 1 and later BERs for all classes. The rate of emission control technology malfunction was assumed the same between LDVs and LDTs, given that their emission technology and usage patterns are increasingly similar. Reduced certification standards are also not expected to influence the rate at which emission control technology malfunctions, because a) manufacturer's design and durability practices are not expected to differ between Tier 1 and later standards, and b) many cases of emission control degradation and/or malfunction are owner-induced, and hence outside the manufacturer's liability for in-use emission performance. It should be noted that the high-emitter fractions in Appendix A are shown to vary by class, due to differences in accumulated mileage at a given age. At the same mileage, the high emitter fractions are the same across all classes. The No OBD / NO I/M average in-use NOx emission rate for vehicle/standard = (V,S) can be calculated at any vehicle age using Equation (2), based on a) the Tier 1 LDV "High_ave" and "Norm_ave" terms from Equations (5) and (6) adjusted as described above based on the (V,S) standard level, and b) the base (No OBD / No EVI) high emitter fractions from Appendix A, Table A-l. 4 NMHC/NMOG BERs and Emitter Fractions: No IM/No OBD Case The development of NMHC/NMOG BERs shared many of the methodological assumptions outlined for NOx in Section 3. As with NOx, NMHC BERs for Tier 1 and NMOG BERs for LEV and later LDVs and LDTs were developed off of "base" LDV BERs; the primary difference between the methodologies for the two pollutants was the source of the base data. At the time of this analysis, sufficient in-use data on vehicles complying with EPA's Tierl NMHC standards (for any vehicle class) were not available. The ARB surveillance dataset used for the NOx analysis included 58 LDVs certified to a 0.25 g/mi 50,000 mile NMHC standard, the same as Federal Tier 1. However, these vehicles did not start entering the market until 1993; hence, most of the vehicles were of age three years or less at the time of testing (the average mileage of this sample was roughly 38,000 DRAFT M6.EXH.007 11 March 1999 ------- miles, versus 58,000 for the 0.40 NOx sample). As such, these data were judged to be inadequate for assessing overall in-use emission performance of Tier 1 LDVs. Tier 1 and later HC BERs were instead based on proposed MOBILE6 BERs for model year 1988 through 1993 Tier 0 LDVs with ported fuel inj ection (PFI). These BERs were developed based on several thousand vehicles tested by auto manufacturers, EPA, and through I/M programs. The Tier 0 emission rates were considered a good starting point for developing Tier 1 and later BERs because emission control technology used on later Tier 0 vehicles (e.g., 3-way catalysts and ported fuel injection) are generally similar to those used on Tier 1 and later vehicles. A comprehensive treatment of Tier 0 BERs and the datasets used to derive them are contained in the reports under Reference 7, and thus are not presented here. However, for this analysis a simplifying step was performed to generate a linear form of Tier 0 normal-emitter BERs, since as proposed for MOBILE6 these are expressed as nonlinear functions. The resulting normal and high emitting Tier 0 BERs (expressed as Total Hydrocarbon, or THC) used as a basis for Tier 1 and later BERs are shown in Equations (13) and (14). Norm_Ave(g/mi) = 0.16 + 0.0186 * odom Eqn 13 High_Ave(g/mi) = 2.076 Eqn 14 "odom" represents mileage in units often thousand miles. Using these values for normal and high emitters as a starting point, normal and high emitter NMHC/NMOG BERs for all Tier 1 and later LDVs and LDTs were developed using the identical methodology as for NOx (described in Sections 3.2.1 and 3.2.2) based on the ratio of the applicable 50,000 mile standard level to the Tier 0 level of 0.41 g/mi. Since Tier 1 standards are expressed as NMHC and LEV and later standards are expressed as NMOG, the shift from THC to these pollutants is accounted for in the standard ratios. The high emitter fractions developed for 1988 through 1993 Tier 0 PFI LDVs were used as the No OBD/No EVI emitter fractions for all Tier 1 and later vehicles, because of the expected similarity in vehicle malfunction rates across standard level and vehicle class as discussed in Section 3.2.3. These fractions are show in Appendix A, Table A-2. The No OBD / No EVI average in-use NMHC/NMOG emission rate for vehicle/standard = (V,S) can be calculated at any vehicle age using Equation (2), based on a) the Tier 0 "High_ave" and "Norm_ave" terms from Equations (13) and (14) adjusted based on the (V,S) standard level, and b) the base (No OBD / No EVI) high emitter fractions from Appendix A, Table A-2. 5 Effects of OBD and OBD-based I/M for NOx and HC Separate BERs were developed for all standard and vehicle classes to account for the effects of OBD and OBD-based I/M programs. The methodology used to account for these programs were identical for NOx and HC, based on reducing the fraction of high emitters in the fleet from the No DRAFT M6.EXH.007 12 March 1999 ------- OBD/No IM case. Thus, emission levels for normal and high emitters were not changed under these programs, only the fraction of highs in the fleet. This methodology introduces a new category of vehicle into the fleet: "Repaired" emitters. These vehicles are high emitters that are flagged by an OBD system and undergo successful repair. For the OBD-only and OBD/EVI cases, these vehicles are treated distinctly from normal and high emitters. 5.1 OBD Effectiveness OBD effectiveness is defined by three parameters: a) the probability the OBD system will detect a failure (MIL-on Rate), b) the probability an owner will respond to a MIL-on (Response Rate), and c) the average after-repair emission level for responding vehicles (Repair Level). At the time of this analysis, sufficient in-use data were not available to determine in-use patterns for these parameters. Thus, estimates were developed for the projected likelihood of malfunction detection, owner response and repair level; stakeholder comments on these estimates are encouraged. These are presented in Sections 5.1.1 through 5.1.3 for both the OBD-only and OBD/EVI cases. 5.1.1 MIL-on Rate For all vehicle classes and standards, it is assumed that OBD will detect (i.e., set an appropriate code and illuminate the MIL) 85 percent of high emitters. Because high emitters are defined independently for HC and NOx, this response rate is assumed to apply equally to both pollutants. The remaining 15 percent of fleet will not be identified, and thus will remain in the fleet as high emitters. No deterioration in the ability of the OBD system to correctly identify high emitters is assumed. Because this parameter is solely dependent on the vehicle's OBD system, it is the same for I/M and non-I/M areas. 5.1.2 Response Rate In order to obtain emission reductions from a vehicle equipped with an OBD system, not only must the system correctly identify the vehicle, but the motorist must also respond to the MIL and take corrective action in a timely manner. A key assumption for this analysis is that the response rate in OBD-based I/M areas (where repair is required) will be much higher than in non-I/M areas. In OBD-based I/M areas, the MIL-on response rate is assumed to be 90 percent over the entire life of the vehicle. Response was not set at 100 percent to account for waivers from, or evasion of, the I/M program. For non-IM areas, OBD response is assumed to be a function of vehicle warranty. It is assumed that an owner is much more likely to respond to a MIL-on when repairs will be paid for by the manufacturer. Three mileage bins were therefore developed: 1) 0 through 36,000, the standard bumper-to-bumper warranty period; 2) 36,001 to 80,000, for which federal law mandates that catalysts and electronic control modules (ECMs) remain under warranty; and 3) above 80,000, for which no warranty is in effect (extended warranties are not accounted for in this methodology). Under 36,000 miles, it is assumed that 90 percent of MIL-on vehicles will be repaired. This DRAFT M6.EXH.007 13 March 1999 ------- is based on the judgment that for new vehicles still under warranty, owners will have little hesitation in addressing a MIL-on. The 10 percent loss accounts for a small percentage of owners who will not respond to a MIL-on even with the warranty incentive. Between 36,000 miles and 80,000 miles, it is assumed that 10 percent of MIL-on vehicles will be repaired. This response rate is greatly reduced from the pre-36,000 mile level to account for the discontinuation of warranty coverage on several emission-related components (e.g. secondary air, EGR, oxygen sensors, fuel injectors), and reduced willingness of owners to make emission- related repairs on an aging vehicle in the absence of an I/M program. Above 80,000 miles, it is assumed that no MIL-on vehicles will be repaired. This assumption reflects the end of warranties, the lower economic value of the vehicle, and the (further) reduced willingness of owners to make emission-related repairs in the absence of an I/M program. 5.1.3 Repair Level Implementation of OBD requires a third emitter category - "repaired" emitters. Repaired emitters are assumed to have constant emissions at the after-repaired emission level, although a subset of these vehicles "migrate" back to the high emitter category. The emission level after an OBD-induced repair is assumed to be 1.5 times the applicable 50,000 mile certification standard. For Tier 1 LDVs and LDTs, this repair level is the required threshold for illuminating the MIL up to 50,000 miles; the repair level is lower than the OBD threshold above 50,000 miles, since the 1.5 times the standard criteria applies to the full useful life standard above this mileage. For LEVs, OBD detection is required at 1.75 times the full useful life standard at all mileages; hence, the estimated repair level is below the OBD threshold at all mileages. Because repaired emitters are assumed not to deteriorate for the remainder of their life (unless they migrate into the high emitter category), the benefit of repair increases over the life of the vehicle. Repaired emitter emissions levels are generally lower than average normal emitter emissions at higher mileages. The repaired emitter "BERs" are presented across standard and vehicle class in Appendix B. 5.2 High Emitter Fractions for OBD and OBD-based I/M Equations 15 through 17 were used to calculate the high emitter growth rate under the OBD and OBD-based I/M scenarios (HighOBD). Overall, the high emitter fraction in a given year is a function of a) the number of high emitters in the previous year, b) the base high emitter "growth rate" in the absence of OBD or I/M, and c) the OBD effectiveness assumptions outlined in Section 5.1. The subscript T is the vehicle age. High(O) is assumed to be zero. MOBILE6 will assign a value of 'odom' for each age 'i'. Nonhigh(i) = 1.0 - HighBASE(i) Eqn 15 Delta_High(i) = HighBASE(i)-HighBASE(i-l) Eqn 16 DRAFT M6.EXH.007 14 March 1999 ------- Growth_High(i) = Delta_High / Nonhigh(i) Eqn 17 HighOBD(i) = HighOBD(i-l) + [(l-OBD)*MIL*Growth_High(i)*(l-HighOBD(i-l))] + [(l-MIL)*Growth_High(i)*(l-HighOBD(i-l)] Eqn 18 Where: HighOBD(0) = 0.0 MIL = 0.85 Nonhigh = the fraction of normal and repaired vehicles Growth_High = the growth rate of high emitters (or, the rate at which "nonhighs" migrate into the high emitter category) 'OBD' is the OBD response rate; 0.90 for OBD-based I/M, and 0.90/0.10/0.0 for mileage bins (0 - 36K), (36K - 80K), and (80K+). An elaboration on Equations 15 through 18 is as follows: for a given vehicle age, the fraction of high emitters is a) the number of highs from the year before, plus b) the number of MIL-on highs added in that year due to OBD nonresponse (a function of "nonresponse" rate, MIL-on rate, and the high emitter growth rate applied to the available pool of normal and repaired vehicles), plus c) the number of highs added in that year that the OBD system did not detect (a function of MIL-"off' rate and the high emitter growth rate applied to the available pool of "non-highs"). The high emitter growth rate for a given year is the absolute increase in high emitters under the No OBD / No EVI case from the previous year divided by the fraction on nonhighs - i.e., the available pool of vehicles which can become high emitters. Once the high emitter fraction is calculated for the OBD or OBD/EVI cases, the fraction of repaired emitters can be calculated as the difference between the fraction of high emitters that would occur without OBD or I/M (HighBASE, from Equation (10)) and the fraction of high emitters with OBD and/or I/M from Equation 18. In equation form, Repaired(i) = HighBASE (i) - HighOBD (i) Eqn 19 The rate of normal emitters remains constant between the No OBD / No I/M, OBD-only and OBD/EVI case; only the number of high emitters decrease, directly replaced by repaired emitters. The emitter fractions for normal, high and repaired emitters for the OBD Only and OBD/EVI cases are shown in Appendix A (Tables A-l and A-2) for NOx and HC, by vehicle class. 5.3 BER Calculation for OBD and OBD-based I/M Calculation of average in-use FTP-based NOx and NMHC emission rates at a given vehicle age for the OBD-Only and OBD/EVI cases are similar to the methodology for No OBD / No I/M vehicles (Equation (2)); the primary differences are a) use of HighOBD rather than HighBASE emitter DRAFT M6.EXH.007 15 March 1999 ------- fractions, and b) addition of a term to account for repaired emitters. As mentioned, the normal and high emissions rates are unchanged from the No OBD / No I/M case. This computation is as follows: AVE = HighOBD * High_ave + Normal *Norm_ave + Repaired * Rep_ave Eqn 20 Rep_ave is 1.5 times the applicable 50,000 mile certification standard. 5.4 Results FTP-based BERs for the No OBD / No IM, OBD only and OBD/IM cases for Tier 1, LEV and ULEV standards across all vehicle classes are shown in Appendix C for NOx and Appendix D for NMHC/NMOG. The full range of standards that will be handled by MOBILE6 (primarily TLEV) are not shown, but can be derived using the methodology outlined in this paper. These charts highlight a key outcome of the "50 percent proportional" treatment of high emitters discussed in Section 3.2.2. BERs for standards below the base standard (Tier 1 LDV) are higher relative to the applicable standard, but have more OBD and/or I/M benefit (on a percent basis) relative to Tier 1 LDVs. BERs for standards above the base standard (Tier 1 LDV) are lower relative to the applicable standard, but have less OBD and/or I/M benefit (on a percent basis) relative to Tier 1 LDVs. This is rooted in the assumption that average high emitter emissions are not reduced in proportion to the standard change, thus increasing the disparity between normal and high emitter for lower standards. As standards are lowered, high emitters will have a more pronounced effect on average in-use emissions, and repair of these high emitters will result in greater relative emission benefit. 6 NOx and HC BERs for Exhaust-Only I/M Since an OBD check is currently an unproven concept in an I/M program, some I/M credit scenario must be developed for those areas that use traditional exhaust I/M test procedures. This scenario will be likely used frequently until calendar year 2001. By that time, it is assumed that I/M test procedures utilizing OBD checks on vehicles equipped with OBD will be in place. 6.1 No OBD with Exhaust I/M The "No OBD / Exhaust I/M" emission levels for this scenario are calculated using the methodology described in draft MOBILE6 document M6.EVI.001 ("MOBILE6 Inspection / Maintenance Benefits Methodology for 1981 through 1993 Model Year Light Vehicles"). This methodology utilizes I/M exhaust test identification rates and after repair effectiveness levels based on data collected from the Arizona I/M program. The "No I/M" and the "With Exhaust I/M" emission rates are used to calculate the I/M emission level and I/M credits for situations where exhaust-only I/M tests are being performed on Tier 1 vehicles without OBD. The only vehicles which will fall in this category are the 1994 and 1995 model years certified to Tierl standards. In this case, the structure of the I/M credits is identical to the Tier 0 I/M credits with the exception that the Norm_ave, High_ave, and fraction of highs in the fleet (HighBASE) are different from analogous DRAFT M6.EXH.007 16 March 1999 ------- Tier 0 parameters. Equation 21 defines this case mathematically. AVE = Norm_ave*(l-HighBASE) + High_ave*HighBASE*(l-IDR) + HighBASE*IDR*W*High_ave*RW + Norm_ave*R*HighBASE*IDR*FIX + High_ave*HighBASE*IDR*NC Eqn21 IDR is the identification rate of high emitters using an exhaust emission test. R is the after repair emission level of vehicles repaired to pass an exhaust I/M test. Fix is the fraction of vehicles which are repaired. NC is the fraction of vehicles which are in non-compliance following their I/M test W is the fraction of vehicles which receive a cost or other type of waiver. RW is the after repair level of the vehicles which get waived. It is shown as a fraction of the high emitter level. (see report M6.IM.001 for a full explanation of these terms). 6.2 OBD and Exhaust I/M In this scenario, the vehicles in the fleet are OBD compliant, but the state continues to conduct an exhaust I/M test; this is most likely scenario prior to calendar year 2001. For this scenario, the same I/M equations and assumptions used to model the 1981-93 Tier 0 vehicles are used. The primary difference is the fraction of high emitters is reduced somewhat due to the effects of OBD program (i.e., HighOBD is substituted for a higher rate of high emitters used in TierO. AVE = Norm_ave*(l-HighBASE) + Rep_ave* (Repaired) + High_ave*HighOBD*(l-IDR) + HighoBD*IDR*W*High_ave*RW + Norm_ave*R*HighOBD*IDR*FIX + High_ave*HighoBD*IDR*NC Eqn22 7 Derivation of Running and Start BERs for NOx and HC MOBILE6 requires a split between running and start emissions; because Tier 1 and later BERs were developed on an FTP-basis, factors will be required allow derivation of running and start BERs from FTP-based BERs. Running fractions were first derived by applying the methodology outlined in MOBILE6 report number M6. STE.002 ("The Determination of Hot Running Emissions From FTP Bag Emissions") to separate Tier 1 and LEV datasets. For Tier Is, a sample of 31 LDVs and LDTs vehicles tested by auto manufacturers as part of the SFTP rulemaking were used.9 For LEVs, a sample of 34 LDVs and LDTs tested by auto manufacturers, CRC and API and used for the development of proposed MOBILE6 fuel sulfur effects.10 Based on an analysis of separate factors for LDVs and LDTs, a single factor which applied to both LDVs and LDTs was judged to be adequate. For NOx, the resulting Tier 1 and LEV factors were found to be identical; the following equation was therefore used to generate running BERs for all standards and classes: 9SFTP Data CD 10Rao, V., "Fuel Sulfur Effects On Exhaust Emissions", EPA Draft MOBILE6 Report No. M6.FUL.001 DRAFT M6.EXH.007 17 March 1999 ------- Running NOx BER (g/mi) = 0.60 * FTP NOx BER Eqn 23 For NMHC/NMOG, separate factors were developed for Tier Is and LEVs: Tier 1 Running NMHC BER (g/mi) = 0.16 *FTPNMHCBER Eqn 24 LEV Running NMOG BER (g/mi) = 0.30 *FTPNMOGBER Eqn 25 In MOBILE6, start BERs are related to FTP and Running BERs as shown in Equation (26):11 FTP BER = (Running BER*7.5 +Start BER*0.43+Start BER*0.57*0.16) / 7.5 Eqn 26 Where: Running BERs = the results of Equations (23) through (25) 7.5 = total miles of the LA4 0.43/0.57 = Bag I/Bag 3 weighting across total FTP 0.16 = the ratio of Bag 3 emissions to Bag 1 emissions, based on HC 1988-1993 PFI BERs proposed for MOBILE6 (applied to both HC and NOx)12 Using this equation and the running and FTP BERs developed above, start factors were derived according to Equation (27): Tier I/LEV Start NOx BERs (grams) = 5.75 * FTP BER Eqn 27 Tier 1 Start NMHC BER (grams) = 12.1* FTP BER Eqn 27 LEV Start NMOG BER (grams) = 10.1* FTP BER Eqn 27 These factors were applied equally to each emitter class: normal, high and repaired. The resulting BERs for Tier 1, LEV and ULEV across all classes are shown in Appendix B. nGlover, E., and Carey, P., "Determination of Start Emissions as a Function of Mileage and Soak Time for 1981-1993 Model Year Light-Duty Vehicles", MOBILE6 report number M6.STE.003 12This factor varies somewhat across pollutant and vehicle grouping, but was held constant for the derivation of start factors. DRAFT M6.EXH.007 18 March 1999 ------- APPENDIX A: a) Emitter Fractions: No OBD/No IM, OBD Only, OBD/IM Cases b) Vehicle Mileage As A Function of Age ------- Table A-l: NOx Emitter Fractions Age (Years) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 LDV All Normal 1.000 0.946 0.909 0.872 0.836 0.801 0.767 0.734 0.701 0.670 0.639 0.609 0.580 0.552 0.524 0.494 0.465 0.436 0.409 0.382 0.356 0.330 0.305 0.281 0.258 0.236 Base High 0.000 0.054 0.091 0.128 0.164 0.199 0.233 0.266 0.299 0.330 0.361 0.391 0.420 0.448 0.476 0.506 0.535 0.564 0.591 0.618 0.644 0.670 0.695 0.719 0.742 0.764 OBD Only High 0.000 0.013 0.022 0.058 0.094 0.128 0.162 0.199 0.234 0.268 0.302 0.334 0.366 0.397 0.428 0.461 0.492 0.523 0.553 0.583 0.611 0.639 0.666 0.693 0.718 0.743 Repair 0.000 0.041 0.069 0.070 0.070 0.071 0.071 0.068 0.065 0.062 0.059 0.056 0.054 0.051 0.048 0.046 0.043 0.040 0.038 0.035 0.033 0.031 0.028 0.026 0.024 0.022 OBD/IM High 0.000 0.013 0.022 0.031 0.041 0.050 0.059 0.069 0.079 0.088 0.098 0.108 0.118 0.128 0.139 0.150 0.162 0.174 0.186 0.199 0.212 0.225 0.239 0.253 0.267 0.282 Repair 0.000 0.041 0.069 0.097 0.123 0.149 0.174 0.197 0.220 0.242 0.263 0.283 0.302 0.320 0.338 0.356 0.373 0.390 0.405 0.419 0.433 0.445 0.456 0.466 0.475 0.482 LOT 1/2 All Normal 1.000 0.934 0.885 0.837 0.790 0.744 0.700 0.657 0.617 0.577 0.540 0.500 0.462 0.426 0.391 0.359 0.328 0.300 0.273 0.249 0.227 0.206 0.188 0.171 0.156 0.142 Base High 0.000 0.066 0.115 0.163 0.210 0.256 0.300 0.343 0.383 0.423 0.460 0.500 0.538 0.574 0.609 0.641 0.672 0.700 0.727 0.751 0.773 0.794 0.812 0.829 0.844 0.858 OBD Only High 0.000 0.015 0.063 0.110 0.156 0.204 0.251 0.297 0.341 0.383 0.423 0.465 0.506 0.545 0.581 0.616 0.649 0.679 0.708 0.734 0.758 0.779 0.799 0.817 0.833 0.848 Repair 0.000 0.050 0.052 0.054 0.055 0.052 0.049 0.046 0.043 0.040 0.037 0.035 0.032 0.030 0.027 0.025 0.023 0.021 0.019 0.017 0.016 0.014 0.013 0.012 0.011 0.010 OBD/IM High 0.000 0.015 0.028 0.040 0.053 0.066 0.079 0.092 0.105 0.118 0.132 0.147 0.162 0.178 0.193 0.209 0.225 0.241 0.256 0.272 0.287 0.302 0.317 0.331 0.345 0.359 Repair 0.000 0.050 0.088 0.123 0.158 0.190 0.221 0.251 0.278 0.304 0.328 0.353 0.376 0.397 0.415 0.432 0.447 0.459 0.470 0.479 0.486 0.491 0.495 0.498 0.499 0.499 LDT3/4 All Normal 1.000 0.929 0.876 0.824 0.773 0.725 0.678 0.633 0.590 0.548 0.505 0.461 0.419 0.378 0.339 0.302 0.265 0.231 0.197 0.165 0.135 0.106 0.079 0.052 0.027 0.004 Base High 0.000 0.071 0.124 0.176 0.227 0.275 0.322 0.367 0.410 0.452 0.495 0.539 0.581 0.622 0.661 0.698 0.735 0.769 0.803 0.835 0.865 0.894 0.921 0.948 0.973 0.996 OBD Only High 0.000 0.017 0.069 0.119 0.168 0.221 0.271 0.319 0.366 0.411 0.457 0.504 0.549 0.593 0.635 0.676 0.715 0.752 0.788 0.822 0.855 0.886 0.916 0.944 0.971 0.996 Repair 0.000 0.054 0.056 0.057 0.058 0.055 0.051 0.048 0.044 0.041 0.038 0.035 0.032 0.028 0.026 0.023 0.020 0.017 0.015 0.012 0.010 0.008 0.006 0.004 0.002 0.000 OBD/IM High 0.000 0.017 0.030 0.043 0.057 0.071 0.085 0.099 0.114 0.129 0.145 0.162 0.180 0.199 0.218 0.239 0.260 0.283 0.307 0.334 0.362 0.394 0.431 0.476 0.535 0.629 Repair 0.000 0.054 0.094 0.133 0.169 0.204 0.237 0.267 0.296 0.323 0.351 0.377 0.401 0.423 0.442 0.460 0.474 0.486 0.495 0.501 0.503 0.499 0.490 0.472 0.438 0.367 ------- Table A-2: NMHC/NMOG Emitter Fractions Age (Years) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 LDV All Normal 0.982 0.975 0.958 0.934 0.910 0.887 0.865 0.843 0.822 0.801 0.782 0.763 0.744 0.727 0.710 0.693 0.677 0.662 0.647 0.633 0.620 0.606 0.594 0.582 0.570 0.559 Base High 0.018 0.025 0.042 0.066 0.090 0.113 0.135 0.157 0.178 0.199 0.218 0.237 0.256 0.273 0.290 0.307 0.323 0.338 0.353 0.367 0.380 0.394 0.406 0.418 0.430 0.441 OBD Only High 0.004 0.006 0.010 0.033 0.056 0.077 0.099 0.121 0.143 0.164 0.185 0.205 0.224 0.242 0.260 0.277 0.294 0.310 0.325 0.340 0.354 0.368 0.381 0.393 0.405 0.417 Repair 0.014 0.019 0.032 0.033 0.035 0.036 0.037 0.036 0.035 0.034 0.033 0.032 0.032 0.031 0.030 0.029 0.029 0.028 0.028 0.027 0.026 0.026 0.025 0.025 0.024 0.024 OBD/IM High 0.004 0.006 0.010 0.016 0.022 0.028 0.033 0.039 0.045 0.050 0.056 0.061 0.066 0.072 0.077 0.082 0.087 0.092 0.096 0.101 0.106 0.110 0.114 0.118 0.123 0.127 Repair 0.014 0.019 0.032 0.050 0.068 0.086 0.102 0.118 0.133 0.148 0.163 0.176 0.189 0.202 0.214 0.225 0.236 0.246 0.256 0.266 0.275 0.284 0.292 0.300 0.307 0.314 LOT 1/2 All Normal 0.982 0.969 0.942 0.910 0.879 0.850 0.821 0.794 0.767 0.743 0.719 0.697 0.676 0.656 0.638 0.621 0.606 0.591 0.578 0.566 0.555 0.545 0.536 0.528 0.521 0.515 Base High 0.018 0.031 0.058 0.090 0.121 0.150 0.179 0.206 0.233 0.257 0.281 0.303 0.324 0.344 0.362 0.379 0.394 0.409 0.422 0.434 0.445 0.455 0.464 0.472 0.479 0.485 OBD Only High 0.004 0.007 0.033 0.063 0.092 0.123 0.152 0.180 0.207 0.233 0.257 0.280 0.302 0.322 0.341 0.358 0.375 0.390 0.403 0.416 0.427 0.437 0.446 0.455 0.462 0.469 Repair 0.014 0.023 0.025 0.027 0.029 0.028 0.027 0.026 0.025 0.024 0.024 0.023 0.022 0.021 0.021 0.020 0.020 0.019 0.019 0.019 0.018 0.018 0.018 0.017 0.017 0.017 OBD/IM High 0.004 0.007 0.014 0.022 0.029 0.037 0.045 0.052 0.060 0.067 0.074 0.080 0.087 0.093 0.099 0.105 0.110 0.115 0.120 0.124 0.128 0.132 0.135 0.138 0.141 0.143 Repair 0.014 0.023 0.044 0.068 0.091 0.113 0.134 0.154 0.173 0.191 0.207 0.223 0.237 0.251 0.263 0.274 0.284 0.294 0.302 0.310 0.317 0.323 0.329 0.334 0.338 0.342 LDT3/4 All Normal 0.982 0.967 0.936 0.902 0.869 0.837 0.807 0.778 0.750 0.724 0.699 0.675 0.653 0.631 0.611 0.592 0.574 0.557 0.541 0.526 0.511 0.498 0.485 0.473 0.462 0.452 Base High 0.018 0.033 0.064 0.098 0.131 0.163 0.193 0.222 0.250 0.276 0.301 0.325 0.347 0.369 0.389 0.408 0.426 0.443 0.459 0.474 0.489 0.502 0.515 0.527 0.538 0.548 OBD Only High 0.004 0.008 0.037 0.069 0.100 0.133 0.165 0.195 0.223 0.250 0.276 0.301 0.324 0.346 0.367 0.387 0.406 0.423 0.440 0.456 0.471 0.485 0.498 0.510 0.522 0.532 Repair 0.014 0.025 0.027 0.029 0.031 0.030 0.029 0.028 0.027 0.026 0.025 0.024 0.023 0.022 0.022 0.021 0.020 0.020 0.019 0.019 0.018 0.018 0.017 0.017 0.016 0.016 OBD/IM High 0.004 0.008 0.015 0.024 0.032 0.040 0.049 0.057 0.064 0.072 0.080 0.087 0.094 0.101 0.108 0.115 0.121 0.127 0.133 0.139 0.144 0.149 0.155 0.159 0.164 0.168 Repair 0.014 0.025 0.049 0.074 0.099 0.122 0.145 0.165 0.185 0.204 0.221 0.238 0.253 0.267 0.281 0.294 0.305 0.316 0.326 0.336 0.345 0.353 0.360 0.367 0.374 0.380 ------- Table A-3: Draft MOBILE6 Cumulative Mileages (10,000 miles) Age (Years) 1 2 3 4 5 6 7 8 9 10 11 12 13 LDV 1.491 2.908 4.256 5.537 6.755 7.912 9.013 10.059 11.054 12.000 12.899 13.753 14.566 LDT1/2 1.950 3.788 5.519 7.146 8.672 10.100 11.436 12.681 13.839 14.914 15.910 16.829 17.676 LDT3/4 2.133 4.120 5.970 7.692 9.297 10.791 12.183 13.478 14.685 15.809 16.856 17.830 18.738 Age (Years) 14 15 16 17 18 19 20 21 22 23 24 25 LDV 15.338 16.072 16.770 17.434 18.064 18.664 19.234 19.776 20.291 20.781 21.247 21.690 LDT1/2 18.453 19.165 19.815 20.406 20.941 21.425 21.861 22.252 22.602 22.914 23.191 23.438 LDT3/4 19.583 20.371 21.104 21.786 22.422 23.014 23.566 24.079 24.557 25.003 25.418 25.804 ------- APPENDIX B: TIER 1, LEV and ULEV BERs By Emitter Category ------- Table B-l: NOx Basic Emission Rates Vehicle Class LDV/T1 LDT2/3 LDT4 Standard Class Tierl LEV/ULEV Tierl LEV/ULEV Tierl LEV/ULEV 50K Standard (g/mi) 0.4 0.2 0.7 0.4 1.1 0.6 Mode FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) "Normal" BER (g/mi) ZML 0.265 0.159 1.524 0.133 0.080 0.765 0.464 0.278 2.668 DR 0.0147 0.0088 0.0845 0.0074 0.0044 0.0426 0.0257 0.0154 0.1478 "High" BER (g/mi) 1.28 0.77 7.35 0.96 0.58 5.51 1.76 1.05 10.10 "Repaired" BER (g/mi) 0.600 0.360 3.450 0.300 0.180 1.725 1.050 0.630 6.038 SAME AS LDV/T1 TIER 1 FTP Running Start (grams) FTP Running Start (grams') 0.730 0.438 4.198 0.398 0.239 2.289 0.0404 0.0242 0.2323 0.0220 0.0132 0.1265 2.40 1.44 13.78 1.60 0.96 9.200 1.650 0.990 9.488 0.900 0.540 5.175 ------- Table B-2: NMHC/NMOG Vehicle Class LDV/T1 LDT2 LDT3 LDT4 Standard Class Tierl LEV ULEV Tierl LEV ULEV Tierl LEV ULEV Tierl LEV ULEV 50K Standard (g/mi) 0.25 0.075 0.04 0.32 0.10 0.05 0.32 0.16 0.10 0.39 0.195 0.117 Mode FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) FTP Running Start (grams) Basic Emission Rates "Normal" BER (g/mi) ZML 0.098 0.016 1.183 0.029 0.009 0.293 0.016 0.005 0.162 0.125 0.020 1.509 0.039 0.012 0.394 0.020 0.006 0.197 0.125 0.020 1.509 0.063 0.019 0.636 DR 0.0113 0.0018 0.1364 0.0034 0.0010 0.0343 0.0018 0.0005 0.0182 0.0145 0.0023 0.1750 0.0045 0.0013 0.0455 0.0023 0.0007 0.0229 0.0145 0.0023 0.1750 0.0073 0.0022 0.0737 SAME AS LDT2 FTP Running Start (grams) FTP Running Start (grams) FTP Running Start f crams') 0.152 0.024 1.835 0.076 0.023 0.770 0.046 0.014 0.462 0.0177 0.0029 0.2136 0.0088 0.0026 0.0889 0.0053 0.0016 0.0535 "High" BER (g/mi) 1.67 0.27 20.17 1.23 0.37 12.40 1.14 0.34 11.51 1.85 0.30 22.31 1.29 0.38 13.04 1.17 0.35 11.77 1.85 0.30 22.31 1.44 0.43 14.54 LEV 2.03 0.33 24.44 1.53 0.46 15.45 1.33 0.40 13.43 "Repaired" BER (g/mi) 0.375 0.060 4.526 0.113 0.034 1.141 0.060 0.018 0.606 0.480 0.077 5.794 0.150 0.045 1.515 0.075 0.022 0.758 0.480 0.077 5.794 0.240 0.072 2.424 0.585 0.094 7.061 0.293 0.087 2.954 0.176 0.052 1.773 ------- APPENDIX C: NOx FTP-Based Composite BERs for Tier 1, LEV and ULEV ------- LDV/LDT1 Tier 1 NOx o.o 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ No •OBD/No - - - OBD/ ------- LDV/LDT1 LEV/ULEV NOx o.o 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ No •OBD/No - - - OBD/ ------- LDT2/LDT3 Tier 1 NOx 1.6 1.4 1.2 « 1.0 E "55 | 0.8 O) o. t 0.6 0.4 0.2 0.0 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ No •OBD/No - - - OBD/ ------- LDT2/LDT3 LEV/ULEV NOx o.o 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ No •OBD/No - - - OBD/ ------- LDT4 Tier 1 NOx o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ No •OBD/No - - - OBD/ ------- LDT4 LEV/ULEV NOx 1.4 1.2 1.0 I 0.8 to CO 0.4 0.2 0.0 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ No •OBD/No - - - OBD/ ------- APPENDIX D (Corrected): NMHC/NMOG FTP-Based Composite BERs for Tier 1, LEV and ULEV ------- LDV/LDT1 Tierl NMHC o.o 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDV/LDT1 LEV NMOG o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDV/LDT1 ULEV NMOG o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDT2/LDT3 Tier 1 NMHC o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDT2/LDT3 LEV NMOG o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDT2/LDT3 ULEV NMOG o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDT4 Tier 1 NMHC o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDT4 LEV NMOG o.o 0.0 2.5 5.0 7.5 10.0 Mileage/10000 12.5 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- LDT4 ULEV NMOG o.o 0.0 2.5 5.0 7.5 10.0 12.5 Mileage/10000 15.0 17.5 20.0 •NoOBD/ NolM •OBD/No - - - OBD/ ------- |