Evaluation of the Effectiveness of
On-Board Diagnostic (OBD) Systems
in Identifying Fuel Vapor Losses
from Light-Duty Vehicles
Final Report
&EPA
United States
Environmental Protection
Agency
-------
Evaluation of the Effectiveness of
On-Board Diagnostic (OBD) Systems
in Identifying Fuel Vapor Losses
from Light-Duty Vehicles
Final Report
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
Prepared for EPA by
Eastern Research Group, Inc.
EPA Contract No. EP-C-06-0-80
Work Assignment No. 5-10
NOTICE
This technical report does not necessarily represent final EPA decisions or
positions. It is intended to present technical analysis of issues using data
that 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.
United States
Environmental Protection
Agency
EPA-420-R-14-010
March 2014
-------
Evaluation of the
Effectiveness of On-Board
Diagnostic (OBD) Systems in
Identifying Fuel Vapor Losses
from Light-Duty Vehicles
FINAL REPORT
Version 11
Contract No. EP-C-06-0-80
Work Assignment 5-10
Prepared by:
Eastern Research Group, Inc.
Prepared for:
U.S. Environmental Protection Agency
Revised by:
Eastern Research Group, Inc. and
U.S. Environmental Protection Agency in response to
External Peer Review comments provided in RTI
Technical Memo entitled "Peer Review of LDV/LOT
OBD and High Evaporative Emission Report, Work
Assignment 4-01, (RTI 005)", May 2012
Aprils, 2013
-------
. .< ,/vw.erq.com
ERG No.: 0218.05.010.001
Evaluation of the Effectiveness of On-Board Diagnostic (OBD) Systems in
Identifying Fuel Vapor Losses from Light-Duty Vehicles
FINAL REPORT
Version 11
Contract No. EP-C-06-0-80
Work Assignment 5-10
Prepared by:
Eastern Research Group, Inc.
3508 Far West Blvd, Suite 210
Austin, TX 78731
Prepared for:
Ms. Connie Hart
OTAQ Emissions Program Group
U.S. Environmental Protection Agency
2000 Traverwood Dr.
Ann Arbor MI 48105
April 5, 2013
3508 FarWest Blvd., Suite 210, Austin, TX 78731 Phone: 512-407-1820 Fax: 512-419-0
Arlington, VA Atlanta, GA Austin, TX • Boston, MA Chantilly.VA Chicago, II Cincinnati, OH Hershey, PA
Prairie Village, KS Lexington, MA : Nashua, NH Research Triangle Park, NC Sacramento, CA
Equal Opportunity Employer Printed on 100% Post-Consumer Recycled Paper
-------
Table of Contents
1.0 INTRODUCTION 1-1
2.0 BACKGROUND 2-1
2.1 OBD Overview 2-1
2.2 OBD Monitors and Readiness 2-2
2.3 OBDDTCs 2-3
2.4 Evaporative Emissions Control System Overview 2-4
2.5 Recent Evaporative Emissions Real World Test Programs 2-5
3.0 ANALYSIS 3-1
3.1 Data Preparation 3-2
3.2 Data Analysis Results 3-2
3.3 Comparison of SHED Values to Presence of OBD Evaporative System
DTCs 3-3
4.0 CONCLUSIONS 4-1
List of Tables
TABLE 1-1. OVERVIEW OF DATASETS USED IN ANALYSIS 1-2
TABLE3-1. SUMMARY OF MATCHED SHED AND I/M RECORDS USED IN ANALYSIS 3-2
TABLE 3-2. PAIR COUNTS FOR MY96 LIPAN, KEN CARYL, AND DENVER HIGH EVAP
STUDY PARTICIPANTS COMBINED 3-3
TABLE 3-3. NUMBER OF PAIRS WITH EVAP DTCs SET VERSUS HIGH AND Low SHEDs (1.0
G/QHR CUTPOINT) 3-4
TABLE 3-4. NUMBER OF PAIRS WITH EVAP DTCs SET VERSUS HIGH AND Low SHEDs (0.3
G/QHR CUTPOINT) 3-4
TABLE 3-5. DETAILS FOR VEHICLES WITH SHED RESULTS > 1 G/QHR (EXCLUDING LIQUID
AND UNMONITORED VAPOR LEAKS) 3-6
TABLE 3-6. MODEL YEAR COUNTS FOR HIGH AND Low SHED PAIRS IN THE DATA 3-9
TABLE 4-1. COMPARISON OF SHED AND OBD RESULTS FOR VEHICLES WITH SHED
EMISSIONS > 1 G/QHR (SUMMARY OF INFORMATION FOR VEHICLES LISTED IN
TABLE 3-5) 4-2
-------
List of Figures
Figure 1 -1. Electronic HC detector used for MCM Inspections 1-2
Figure 1-2. Portable Sealed Housing for Evaporative Determination (PSHED) used in
Denver High Evaporative Emission Field Studies 1-3
11
-------
1.0 Introduction
This document is Eastern Research Group's (ERG's) final report on evaluating the
effectiveness of Second Generation On-Board Diagnostic (OBD) systems through a comparative
analysis of evaporative emissions field data collected in the Denver area during the past four
years. These field studies were performed through Cooperative Research and Development
Agreements (CRADAs 471-08 and 471-A-l 1) between the Assessment and Standards Division
(ASD) within the Office of Transportation and Air Quality (OTAQ) of the United States
Environmental Protection Agency (EPA) and the Colorado Department of Public Health and
Environment (CDPHE), with support provided by ERG under Work Assignments 1-2, 2-2, 2-4,
3-4 and 4-4 of Contract EP-C-06-080 with the EPA. A wealth of data has been collected at the
Denver Lipan1 and Ken Caryl2 Inspection and Maintenance (I/M) lanes using a Portable Sealed
Housing for Evaporative Determination (PSHED) and remote sensing devices. Because the
studies took place at I/M lanes in Colorado, OBD information is also available for each vehicle.
CDPHE independently performed additional work during the summer of 2010 in a repair
effectiveness study conducted at CDPHE's Aurora and West Tech test centers. During this
"Denver 2010" study, vehicles were recruited and brought to one of the two CDPHE facilities
where laboratory-grade Sealed Housing for Evaporative Determination (SHED) testing was
performed, in addition to other evaporative control system diagnostic tests. Repairs were
performed on vehicles with elevated evaporative emission levels. Repeat tests were conducted to
assess the effectiveness of the evaporative emissions repairs performed by CDPHE.
The focus of this report is on the analysis of evaporative emissions and related data
collected on model year 1996 and newer light-duty vehicles equipped with OBD systems.
Results for the Lipan, Ken Caryl and Denver 2010 studies are analyzed in this report, and Table
1-1 lists the number of 1996 and later OBD equipped vehicles which participated in each of the
three studies.
1 DeFries, T; Lindner, J; Kishan, S; Palacios, C (2011) "Investigation of Techniques for High Evaporative
Emissions Vehicle Detection: Denver Summer 2008 Pilot Study at Lipan Street Station"
2 DeFries, T; Palacios, C; Kishan, S (2012) "Estimated Summer Hot Soak Distributions for the Denver's Ken Caryl
IM Station Fleet"
1-1
-------
Table 1-1. Overview of Datasets used in Analysis
Dataset Source
Lipan Participant (LP)
Ken Caryl Participant (CP)
Denver 2010 Participant (DP)
Total participants (vehicles) in 3 studies
Number of 1996+
Vehicles
24
89
68
181
Data Available in Dataset
OBD, MCM*, PSHED**
OBD, MCM, PSHED
OBD, MCM, PSHED or LSHED
OBD, MCM, PSHED or LSHED
*MCM- Modified California Method
**PSHED - Portable Sealed Housing for Evaporative Determination
The purpose of this analysis is to evaluate OBD's ability to identify vehicles with
elevated evaporative emissions (due to a leak in the fuel/vapor control system). OBD systems
are not designed for and are generally not capable of determining and setting a code for a leak in
the fuel system which would result in a liquid leak.3 Three types of data are used in the analysis
in this report:
OBD - OBD codes which indicate system or component malfunctions that could result in
elevated exhaust or evaporative emission levels. During routine I/M inspections, vehicle OBD
system data (including diagnostic trouble codes (DTCs)) is collected for 1996 and newer
vehicles.
MCM - Vehicles in all three studies received a Modified California Method (MCM)
inspection, which is an under-hood and under-body inspection for evaporative emissions using
olfactory, visual, and electronic hydrocarbon (HC) detector checks. MCM variables are generally
discrete. MCM inspections provide several kinds of information including smell/no-smell noted
by the inspector, visual condition of various fuel system and evaporative emissions control
system components, and detect/no-detect by the electronic HC detector (shown in Figure 1-1).
Figure 1-1. Electronic HC detector used for MCM Inspections
3 40 CFR 86.099-17
1-2
-------
PSHED and LSHED - Hot-soak evaporative emission estimates were gathered for each
vehicle in the three studies by placing the vehicle in a Laboratory Sealed Housing for
Evaporative Determination (LSHED) or a Portable Sealed Housing for Evaporative
Determination (PSHED). LSHED values are reported as continuous variables with units of
grams HC/15minutes. In the Lipan and Ken Caryl studies, a PSHED was used, while in the
Denver 2010 study a PSHED was used for the vehicles tested at the CDPHE's West Tech
Center, and a laboratory (conventional) SHED was used for vehicles tested at CDPHE's Aurora
laboratory testing facility. Performance of the PSHED has been evaluated in comparison to a
laboratory SHED in a separate report4 which shows that for fifteen minutes the PSHED had an
average Recovery of 97.6% and an average Retention of 95.7%, which are within the
requirements of the laboratory SHED for this time period. Figure 1-2 shows the PSHED used in
the Denver studies.
Figure 1-2. Portable Sealed Housing for Evaporative Determination (PSHED) used
in Denver High Evaporative Emission Field Studies
The analysis in this report is broken out by vehicle model year, OBD evaporative system
readiness monitor status, and the presence of enhanced evaporative emission controls for 1996-
1998 vehicles. Vehicles with "Not Ready" systems (as described in Section 2.0, Background)
4 Lindner, J; DeFries, T (2012) "Evaluation of Portable SHED Characteristics"
1-3
-------
associated with the relevant OBD code categories are delineated in the analysis. The break-outs
were made to the finest level of detail possible; however, this was dependent on the sample size
for each category. OTAQ assisted ERG in identifying those 1996-1998 vehicles that were
equipped with enhanced-evaporative emissions systems, as well as whether a vehicle was
certified to meet the federal or California emission standards.5
5 Note that US EPA enhanced evaporative emission controls for passenger cars and light trucks phased in over the
1996-1999 model years.
1-4
-------
2.0 Background
2.1 OBD Overview
Second-Generation Onboard Diagnostics (OBDII, typically referred to simply as OBD) is
a system used to monitor the status and performance of a vehicle's powertrain and emissions
control system components in order to detect malfunctions which could result in an emissions
increase above a specified threshold. For tailpipe emissions, this threshold is 1.5 times the
federal certification emission standard.6 For evaporative emissions control systems, the OBD
system monitors for proper operation of components, adequate purge flow of the evaporative
control system and vapor leaks equivalent to certain orifice diameters (such as 0.020" or 0.040",
depending on which of the applicable OBD standards were followed for certification). All
model year 1996 and newer light-duty non-diesel on-road vehicles sold in the United States are
federally required to be equipped with OBD.7
The OBD system consists of the powertrain control module (PCM), various engine and
transmission input sensors and components, emission control components, a malfunction
indicator light (MIL) and a diagnostic link connector (DLC). These systems work together to
monitor a vehicle's powertrain and emission control components during vehicle operation and to
store diagnostic information and alert the driver if a malfunction is identified. Although the
emission standards to which OBD must comply are federally defined, development of the
systems and algorithms used to identify malfunctions is the responsibility of each vehicle
manufacturer. Therefore, monitoring criteria and the amount of time required for a monitor to
achieve readiness (as defined in the next section) can vary among vehicle manufacturers and
even vehicle models and model years. In addition, some manufacturers choose to comply with
more strict standards than federal standards in order to ensure all their vehicles are 50-state legal.
For example, many manufacturers choose to have their OBD systems certified to detection of a
0.020" diameter vapor leak required by the California Air Resources Board rather than the less
strict federal 0.040" requirement in order to maintain model consistency and avoid the 49-state
certification issue.
Diagnostic information is stored in the system using a five digit code which provides
information about the source of the malfunction, and the driver is alerted when the MIL
(commonly referred to as a check engine light) is illuminated. By connecting an OBD scan tool
(or an inspection and maintenance analyzer) to the vehicle's DLC, certain information including
6 Center for Automotive Science & Technology at Weber State University. (2000). On-Board Diagnostics
Conference 2000 On-Board Diagnostics Manual, (pp. 2-5). Ogden, Utah: Weber State University.
7 40 CFR 86.094-17
2-1
-------
the vehicle's MIL command status (whether the MIL is commanded on or off) and any stored
diagnostic trouble codes (DTCs) can be downloaded from the vehicle's PCM. These DTCs are
five-digit alpha numeric codes of the format P####. The codes of primary interest in this report
are the generic (set in federal regulations) PO series codes (i.e., P0###) defined in SAE J1979,
since these are common among all vehicle manufacturers and are the codes primarily used in
vehicle inspection and maintenance programs.
2.2 OBD Monitors and Readiness
Before a vehicle's OBD system can make a determination about whether or not
malfunctions are occurring, a series of tests are required to evaluate the operational status of the
various OBD subsystems. These operational tests are categorized into meaningful sets of
subsystems referred to as monitors. Each monitor has a readiness code that identifies whether or
not the related subsystem has been evaluated. If a monitor is "ready", that simply means that
testing on that particular subsystem has been completed, and results are available (a DTC will be
stored if a malfunction was identified). On the other hand, if a monitor is "not ready", testing of
that particular subsystem has not been completed, so no information is yet available on whether
or not malfunctions exist in that subsystem. Therefore, a lack of DTCs for a "not ready" monitor
does not indicate that no malfunctions (i.e., DTCs) are present; it simply means the system
evaluation is not complete.
Monitors may be classified as those that are always "ready" (continuous monitors), and
those that become ready at different times (non-continuous monitors). The three continuous
monitors include the Comprehensive Component, Misfire and Fuel System monitors, while the
typical non-continuous monitors are the Oxygen Sensor, Catalyst, Exhaust Gas Recirculation
(EGR), Evaporative Emissions Control System (evaporative system), Air Injection System
(AIR), Thermostat and Positive Crankcase Ventilation System (PCV) monitors.8 As the name
implies, the continuous monitors are always running (these are "critical" monitors which
primarily protect the catalytic converter). The non-continuous monitors are only evaluated under
certain powertrain operating conditions (the vehicle's PCM uses complex algorithms to
determine when these non-continuous monitors may be run). For this reason, at any point in
time, a vehicle will have a number of "ready" monitors, and also may have a number of "not
ready" monitors. Normally, non-continuous monitors which are "not ready" are set to assess
readiness after each key-on operating event.
8 Birnbaum, R, & Truglia, J (2000). Getting to Know OBDIL asttraining.com, ISBN 0-9706711-0-5, pg, 32.
2-2
-------
The amount of time necessary for monitors to achieve readiness varies on a number of
factors. OBD systems on older model-year vehicles (i.e., model year 1996-2000 vehicles)
typically take longer to achieve "readiness" than newer vehicles (as monitoring algorithms have
been refined and improved over the years). In addition, different types of systems require
different conditions (referred to as enabling criteria) before they can be tested. For example, the
operating conditions required to test an oxygen sensor are more frequently encountered than the
operating conditions necessary to evaluate the evaporative system monitor (the enabling criteria
are met more frequently for an oxygen sensor monitor than for an evaporative system monitor).
Therefore, after a PCM reset, the oxygen sensor monitor status is usually "ready" before the
evaporative system monitor status is "ready", and consequently, more vehicles on the road likely
have "ready" oxygen sensor monitors than vehicles with "ready" evaporative system monitors.
Once a monitor's system evaluation has been completed, the status changes from "not ready" to
"ready", and at that time, if any malfunctions are detected, a DTC is stored. However, until the
monitor is "ready", no information can be known about whether or not malfunctions are present.
This assessment cannot be made until the monitor is "ready".
2.3 OBD DTCs
As previously described, once a fault is detected, a 5-digit DTC is stored. This is a
standardized code which provides information about the type of system or fault, which conforms
to the following format:9
• Position 1: Device Code: B=Body; C=Chassis; P=Powertrain; U=Network or data link
• Position 2: Type: 0 = generic (legislated and common among all vehicle manufacturers);
1 = manufacturer specific
• Position 3: Specific vehicle system:
— 1,2 = Fuel and Air Metering
— 3 = Ignition System or Misfire
— 4 = Auxiliary Emission Controls
— 5 = Vehicle Speed Control and Idle Control System
— 6 = Computer Output Circuit
— 7, 8 = Transmission
• Positions 4 and 5: These last two digits indicate which component or system has a fault
As an example, P0455 is a powertrain (P) generic (0) code which involves an auxiliary
emission control system or component (4), and the last two digits indicate this is an "Evaporative
Emission Control System Leak Detected (gross leak)"10 (this code is typical for a missing, faulty
9 SAE International, J1979 Surface Vehicle Standard, Issued 1991-12, Revised 2004-04, page 33.
10 Society of Automotive Engineers. (1999, March). On-Board Diagnostics for Light and Medium Duty Vehicles
Standards Manual, pg. 123.
2-3
-------
or incorrectly installed gas cap, among other things). Aftermarket scan tools and I/M analyzers
can access and download all generic powertrain (PO) codes, and other codes (such as
manufacturer-specific PI codes) may be accessible by some scanners and systems, including
specialized and manufacturer-specific diagnostic equipment.
Logic for setting a code varies depending on what system or component is being
monitored. Some codes are immediately set and the MIL is commanded "on" after only one
occurrence of a fault. However, PCM algorithms may require repeat failures on two successive
trips before the code is set and the MIL is commanded on. For these algorithms, if a fault is
identified on the first of two tests, a "pending" DTC is stored. If the fault is again detected on a
subsequent trip (generally the next consecutive trip), then the "pending" DTC becomes a
"confirmed" DTC, and the MIL is commanded on. On the other hand, if a fault is not identified
on the next trip, the "pending" DTC is erased (except for fuel and misfire monitoring codes,
which aren't immediately deleted). Since "pending" DTCs indicate that there may be a problem
but this is not yet confirmed, the MIL is not commanded on for a "pending" DTC. The MIL is
only commanded on for a confirmed DTC, whether it be a single trip DTC or a two-trip DTC.
Once a confirmed DTC is stored, a snapshot of engine operating conditions at the time the code
was set is stored in the PCM's memory. This snapshot of conditions is called freeze frame data
and offers clues that can help a repair technician identify and effectively repair the failure.
As previously described, the PCM can be "reset" (codes, freeze frame data and MIL
command "on" cleared) by disconnecting the vehicle's battery or by using the reset feature of a
scan tool. This also resets all monitors to "not ready". However, the PCM will also
automatically erase codes and freeze frame data and turn the MIL off if the fault which triggered
the DTC is not encountered during three consecutive tests11 (again, excluding fuel and misfire
monitors, which are tested in a different manner whose description is beyond the scope of this
general description).
As a side note, a flashing MIL (as opposed to constant MIL illumination) indicates the
fault which has been detected poses an immediate threat to the catalytic converter.12 However,
an immediate threat to the catalytic converter isn't always represented by a flashing MIL.
2.4 Evaporative Emissions Control System Overview
The focus of this report is on OBD monitoring of the evaporative emissions control
system. The general purpose of the evaporative emissions control system is to prevent release of
11 Birnbaum, R, & Truglia, J (2000). Getting to Know OBDIL asttraining.com, ISBN 0-9706711-0-5, pg, 26.
12 Birnbaum, R, & Truglia, J (2000). Getting to Know OBDIL asttraining.com, ISBN 0-9706711-0-5, pg, 24.
2-4
-------
fuel vapor (raw hydrocarbons) into the atmosphere during vehicle operation, hot soaks and
diurnal soaks. This is accomplished by a system design which reduces permeation and vapor
leaks from various materials and connections. Most of the vapor is routed to a carbon canister
where it is stored and then released into the engine to be burned at certain times during operation
of the engine.
Two overall types of evaporative system OBD monitoring exist, non-enhanced and
enhanced systems. In general, non-enhanced systems only monitor canister purge flow, while
enhanced systems monitor both purge flow and test for leaks in the vapor recovery system.13
Consequently, enhanced systems require leak detection sensors and valves which aren't required
in non-enhanced systems. In general, hardware for evaporative emissions control systems may
include vapor transport lines, control solenoids and valves, pumps, pressure sensors, and the
evaporative carbon canister.
2.5 Recent Evaporative Emissions Real World Test Programs
The Coordinating Research Council (CRC) Real World Group has been conducting
permeation evaporative emissions testing in the E-77 and E-77-2 programs.14 The permeation
test procedure was refined in the pilot E-77 program before using it in the larger E-77-2 program.
Most of the vehicles were specifically recruited as aging-enhanced evaporative emission
vehicles, model years 1996-2000. Three pre-enhanced evaporative emission vehicles and one
newer near-zero emissions vehicle which met California PZEV evaporative requirements were
also tested. The focus was on non-ethanol fuel with both 7 and 9 RVP. The program also
evaluated 24-hour diurnals at the traditional temperature range of 65-105 °F and a higher
temperature diurnal of 85-120°F. A subset of the vehicles was tested for 72-hour diurnals. A leak
was implanted in the gas cap of one of the aging enhanced vehicles, and the series of tests were
run both before and after implanting the leak. The size of the leak was the minimum size
necessary for detection by OBD, 0.020" effective diameter as previously described, (although
federal OBD monitoring requirements specify detection of a 0.040" leak, most manufacturers
certify to California's 0.020" in order to avoid the 49-state certification issue). The E-77 study
demonstrated that hydrocarbon emissions can be up to several orders of magnitude larger with an
13 Birnbaum, R, & Truglia, J (2000). Getting to Know OBDII. asttraining.com, ISBN 0-9706711-0-5, pg, 44.
14 CRC E-77 reports: Haskew, H; Liberty, T (2008) "Vehicle Evaporative Emission Mechanisms: A Pilot Study,
CRC Project E-77", Haskew, H; Liberty, T (2010) "Enhanced Evaporative Emission Vehicles (CRC E-77-2)",
Haskew, H; Liberty, T (2010) "Evaporative Emissions from In-Use Vehicles: Test Fleet Expansion (CRC E-77-
2b)", Haskew, H; Liberty, T (2010) "Study to Determine Evaporative Emission Breakdown, Including Permeation
Effects and Diurnal Emissions Using E20 Fuels on Aging Enhanced Evaporative Emissions Certified Vehicles, CRC
E-77-2c"
2-5
-------
implanted leak. This established the need to define the rate of occurrence of "leakers" in the fleet
in order to determine the overall impact on the inventory.
During additional laboratory testing in the E-77-2 and E-77-2c programs, similarly-sized
leaks were implanted not only in the gas cap, but in other common locations identified during a
special study conducted in 2008 by CDPHE at the Lipan I/M station (described later in this
section). The results indicated evaporative emissions leak rates are dependent on location of the
leak, in particular for vehicles not equipped with On-Board Refueling Vapor Recovery (ORVR).
Regardless of ORVR, higher leak rates were measured for leaks implanted at the top of the fuel
tank and at the inlet connection to evaporative control canister as compared to the gas cap.
The E-77 testing programs confirmed that both liquid and vapor leaks can be a significant
part of any hydrocarbon inventory. The missing piece of information is the prevalence of these
leaks in the fleet. A comprehensive program for quantifying these vehicles in the existing fleet
has not been attempted since the American Petroleum Institute's "Raw Fuel Leak Survey in I/M
Lanes" study in 199815 or the California Bureau of Automotive Repair's "Evaporative Emissions
Impact of Smog Check" study in 2000.16 The current fleet consists of aging enhanced
evaporative emissions vehicles which were not part of or were quite new in these earlier studies.
During the summer of 2008, CDPHE provided access to the Lipan I/M station as well as
staff and remote sensing equipment for a study to determine the prevalence of vehicles with
evaporative leaks in the fleet. Certain model year 1981 and newer vehicles were selected (as
described in Reference 1) and offered participation in evaporative emissions testing, which
consisted of a 15-minute hot-soak test in a portable SHED (PSHED), a visual, olfactory, and
electronic HC detector examination of vehicle fuel metering and evaporative emissions control
system components. Measurements for a comparison of PSHED results with laboratory SHED
were also made on selected participating vehicles.
During the following summer of 2009, CDPHE and the Regional Air Quality Council
(RAQC) performed a follow-on study with EPA reverting to an advisory role. This study was
performed at the Ken Caryl I/M station in Denver and was developed based on the Lipan
experience. Vehicles were selected (as described in Reference 2) and offered participation in
intensive evaporative emissions testing, which consisted of the PSHED hot-soak test, visual,
olfactory, and electronic HC detector examination of the vehicle.
15 McClement, D. (1998) "Raw Fuel Leak Survey in I/M Lanes,"
16 Amlin, D.; Carlisle, R.; Kishan, S.; Klausmeier, R.F.; Haskew, H. (2000) "Evaporative Emissions Impact of
Smog Check,"
2-6
-------
Further work was continued in the summer of 2010 by the CDPHE and the RAQC in a
repair effectiveness study in which vehicles were brought to and tested at CDPHE's Aurora and
West Tech laboratories. In these efforts, SHED tests were performed followed by extensive
diagnostics and repairs. Repeat SHED tests were then performed on vehicles which had received
repairs.
Results from these studies suggest vapor leaks can have an impact on the inventory.
Traditional evaporative emissions testing, as well as the recent E-77 testing, have shown a range
of evaporative emissions from test vehicles, particularly with aging vehicles. The higher
emissions levels (but lower than the "gross liquid leakers") have been associated with slow
growing vapor leaks which occur as the evaporative emissions control systems age. Vehicles
with current (Tier 2 and Enhanced) evaporative emissions control technology are now starting to
age, and information on the prevalence of these higher emissions vehicles in the fleet is still
being collected.
2-7
-------
3.0 Analysis
This analysis involves the comparison of information on OBD evaporative system DTCs
with SHED results for the same vehicles. It is not intended to find prevalence rates with OBD
evaporative codes. The dataset was drawn from three sets of data: the participants in the High
Evaporative Emission studies from 1) the summer of 2008 at the Lipan I/M testing station in
Denver, Colorado, 2) the summer of 2009 at the Ken Caryl I/M testing station in Denver,
Colorado, and 3) the summer and fall of 2010 at the CDPHE Aurora and West Tech laboratories
in Denver, Colorado. Each of these three datasets include vehicles that were tested for
evaporative emissions using a PSHED for 15 minutes (at the Aurora station the vehicles were
tested using an LSHED for 1 hour, but only the first 15 minutes of continuous measurement were
used in this analysis to compare with the PSHEDs).
The I/M dataset included 1996 and newer vehicles. Records for vehicles whose I/M
inspections were aborted or incomplete due to OBD communication problems were removed
from the dataset. The excluded vehicles are described in more detail in Sections 3.1 (Data
Preparation) and 3.2 (Data Analysis Results). Each vehicle's SHED results were then matched
with the OBD test that was performed immediately before the SHED test.
In the case of the Lipan and Ken Caryl high evaporative field studies, the vehicles
received an OBD test as part of their regular I/M inspection prior to solicitation for participation
in the study. Therefore, the OBD test usually happened several hours or even a day before the
SHED test occurred. Therefore, because the vehicle was solicited during the I/M inspection
(immediately before or just after the I/M test was performed, before the motorist left the facility)
and was then retained by the staff performing the high evaporative emissions study, it was still in
the same repair state for both the SHED test and the OBD test.
In the case of the summer 2010 high evaporative emissions study performed at CDPHE's
West Tech and Aurora laboratories, vehicles were solicited by letter and then brought to the
vehicle test station by the owner. The vehicle was kept for several days by the CDPHE staff and
SHED tested several times, before and after repair. Each time a SHED test was performed, a
series of six IM240 tests were performed just prior to the SHED test in order to condition the
vehicle for the SHED testing. The last IM240 with OBD test that was performed of the six prior
to the SHED test was used to match with the SHED testing for this analysis. Only the before
repair test for each of the vehicles in this dataset was used for each vehicle to limit the analysis to
"as received" condition.
3-1
-------
The specific number of OBD/SHED data pairs available is described for each of the three
datasets below. The overall results for the three datasets combined are described in the Data
Analysis Results section after that.
3.1 Data Preparation
Twenty-four PSHED tests were conducted on 24 vehicles during the Lipan High
Evaporative Emissions Field Study conducted in the summer of 2008. ERG successfully
matched 17 of these 24 PSHED results with each vehicle's I/M OBD results (the I/M test was
either aborted or did not contain OBD data for the other 7 tests - OBD is advisory only in the
Colorado I/M program, meaning no pass or fail is assigned based on OBDII results). Similarly,
89 PSHED tests were conducted on 89 vehicles during the Ken Caryl High Evaporative
Emissions Field Study, and ERG successfully matched 76 of these 89 records with
corresponding OBD results (as before, the I/M test was either aborted or did not contain OBD
data for the other 13 tests).
Eighty-seven PSHED/SHED tests were conducted on 68 vehicles during the Denver 2010
study (some of the 68 vehicles in the Denver 2010 study received two tests, a "before repair" and
an "after repair" test). ERG was able to successfully match all 87 of the Denver 2010 study
PSHED/SHED results with valid OBD results. Table 3-1 provides a summary of the final
datasets used in this analysis. As shown in this table, in total, ERG successfully matched results
from 180 PHED/SHED tests with OBD data. Results for comparison of OBD evaporative codes
with PSHED/SHED results are provided in the following section.
Table 3-1. Summary of Matched SHED and I/M Records used in Analysis
Study
Lipan
Ken Caryl
Denver 20 10
Total
Total # of
Vehicles
24
89
68
181
Total # of
PSHED/SHED
Tests
24
89
87
200
"Dropped" PSHED/SHED
Tests (those with no
corresponding OBD record)
7
13
0
20
Remaining PSHED/SHED
Tests matched with OBD
records
17
76
87
180
3.2 Data Analysis Results
As shown in the last column of Table 3-1, when the participant vehicle tests from each of
the three studies were combined, there were a total of 180 PSHED/SHED tests with valid OBD
matches (for the remainder of this report, both PSHED and SHED results will be generically
referred to as "SHED" results for clarity). However, some of these SHED tests were performed
3-2
-------
after the vehicle was repaired. These after-repair SHED results have been removed from the
analysis. Also, some of the SHED and OBD test pairs had an evaporative OBD monitor status of
"Not Ready" at the time of the OBD test, so these test pairs have also been removed from this
analysis. Table 3-2 summarizes the remaining vehicle SHED and OBD pairs after application of
each of these filters.
In this analysis, only test results with "confirmed" evaporative-related DTCs (DTCs in
the range P0440 to P0469 with a MIL command status of "on") were considered. No "Pending"
evaporative DTCs (DTCs in the range P0440 to P0469 but which were "pending" as described in
Section 2.3, OBD DTCs) were identified in the data used for this analysis.
Table 3-2. Pair Counts for MY96 Lipan, Ken Caryl, and Denver High Evap Study
Participants Combined
Total matched SHED/OBD test pairs
Before repair SHED/OBD test pairs
Before repair test pairs with evaporative system monitor status of "Ready"
Before repair test pairs with evaporative monitor status of "Ready" and
with no confirmed DTCs
Before repair test pairs with evaporative monitor status of "Ready" and
with confirmed DTCs and MIL commanded "on"
SHED/OBD Test Pairs
180
157
149
145
4
3.3 Comparison of SHED Values to Presence of OBD Evaporative System DTCs
To compare the OBD and SHED results for each matched pair of records, each SHED
result was defined as either high or low using two SHED cut point values; either 0.3 grams of
total HC emissions per 15 minutes (0.3 g/Qhr), or 1 gram of total HC emissions per 15 minutes
(1 g/Qhr). Separate analyses were performed by defining "high" SHED test results above each
of these cut points or below each of these cut points. The SHED cut point of 1 g/Qhr was used
because it was found to be the approximate average emission level for the first fifteen minutes of
a laboratory Hot Soak of implanted leaks with 0.02" diameter (the minimum detectable diameter
of the evaporative OBD systems) in the Coordinating Research Council (CRC) E-77 test
programs.17 The 0.3 g/Qhr cut point is an estimate of a 15-minute hot-soak portion of the 2 g/hr
Diurnal + Hot Soak enhanced evaporative emission standard. This value was derived from the 2
g/hr Diurnal + Hot Soak enhanced evaporative emission standard by assuming 20% of the 2 g/hr
emissions are due to Hot Soak, and then assuming 75% of that would be in the first 15 minutes.
17 It was found that when leaks were implanted at the gas cap on an ORVR equipped vehicle the emission levels
were contained. This effect will be system dependent, and it is therefore likely that some designs of ORVR vehicles
would show uncontrolled levels of emissions for a leak. The average discussed here is approximate for the
implanted leak locations at the canister connection and at the top of the tank.
3-3
-------
The SHED values were compared for pairs with and without confirmed evaporative
system DTCs. It can be seen from Table 3-2 that only 4 of the 145 records had confirmed
evaporative DTCs stored. Tables Table 3-3 and Table 3-4 compare the number of paired
records with confirmed evaporative DTCs and high/low SHED classification for 1 g/Qhr and 0.3
g/Qhr, respectively. As stated earlier, these tables are provide a comparison of OBD-equipped
vehicles that were measured with a SHED test, but due to the recruitment methods (see
references (1) and (2)), the percentages in these tables are not intended to be representative of the
fleet. The tables also list the percentages of pairs within the four quadrants for each row and
column. All 4 paired records with confirmed evaporative DTCs had high corresponding SHED
values (when defining "high" as 0.3 g/Qhr) and 75% had confirmed evaporative DTCs when
defining "high" as 1.0 g/Qhr. In contrast, 12% of the 145 pairs with no confirmed evaporative
DTCs had high SHED values (with a 1.0 g/Qhr cut point) and 16% were "high" at the 0.3 g/Qhr
cut point. In both Table 3-3 and Table 3-4, 85% of the vehicles with high SHED values did not
have a confirmed evaporative DTC. This could be due to a number of reasons, such as the fact
that not all high evaporative emissions are due to vapor leaks which the OBD system is designed
to identify, but the high percentages does indicate additional investigation is warranted.
Table 3-3. Number of Pairs with Evap DTCs Set Versus High and
Low SHEDs (1.0 g/Qhr cutpoint)
HIGH SHED
(> l.Og/ 15 minutes)
LOW SHED (<=
1.0 g/15minutes)
Total
OBD Evap DTCs not set
17 Vehicles
(85% of 20)
(12% of 145)
128 Vehicles
(99% of 129)
(88% of 145)
145 Vehicles
OBD Evap DTCs set
3 Vehicles
(15% of 20)
(75% of 4)
1 Vehicles
(1% of 129)
(25% of 4)
4 Vehicles
Total
20 Vehicles
129 Vehicles
149 Vehicles
Table 3-4. Number of Pairs with Evap DTCs Set Versus High and
Low SHEDs (0.3 g/Qhr cutpoint)
HIGH SHED
(> 0.3 g/15/minutes)
LOW SHED
(<= 0.3 g/15 minutes)
Total
OBD Evap DTCs
not set
23 Vehicles
(85% of 27)
(16% of 145)
122 Vehicles
(1 00% of 122)
(84% of 145)
145 Vehicles
OBD Evap DTCs
set
4 Vehicles
(15% of 27)
(100% of 4)
0 Vehicles
(0% of 122)
(0%of4)
4 Vehicles
Total
27 Vehicles
122 Vehicles
149 Vehicles
3-4
-------
Table 3-5 provides additional vehicle and test information for vehicles with SHED results
above the 1 g/Qhr designated threshold, including the packet ID, vehicle make, model year,
model and the evaporative certification family for the vehicle. The "evap_mon_ready" column
describes the readiness status of the vehicle's evaporative system monitor. Table 3-5 also lists
DTCs that were downloaded at the time of the OBD test and indicates whether or not the DTCs
were categorized as evaporative DTCs (confirmed DTCs in the range of P0440 to P0469). A
footnote in the table provides definitions of the evaporative DTCs. Additional information on
these vehicles and all vehicles with PSHEDs > .3 g/Qhr, including non-ready OBD evaporative
monitor vehicles, is provided in the appendix of this report.
Table 3-5 also lists comments provided by the vehicle inspectors as they tried to identify
the source of the vapor leaks using a hand held HC vapor detector. In many instances, the
vehicle vapor lines were inaccessible. Some were behind sheet metal shrouds and no intrusive
process was undertaken to get behind these coverings. Based on the inspector comments, the
make and model year of the vehicle, and any other visual evidence, ERG and OTAQ made a
determination regarding whether or not a malfunction should be detectable by the vehicle's OBD
system. This determination is listed in the column labeled "Should Problem be Detectable by
OBD?" If the vehicle inspector was not able to identify the source of the leak, then a
determination was not made regarding whether or not the OBD system should have been able to
detect the malfunction, and a "?" is listed in the table. The areas of the fuel system inaccessible
to the HC detector, as mentioned above, were behind shrouds, above the fuel tank, or within
bundles of hoses packed too tightly to be reached by the HC detector probe. Due to their
locations and lack of evidence of a liquid leak, leaks located in these inaccessible areas were
likely to be a vapor leaks, rather than liquid leaks. Liquid leaks would likely occur at connection
points in the fuel line or the fuel rail and would likely be detectable by the handheld detector. It
is unlikely these liquid leaks would be detectable by the OBD systems for most vehicles, since
OBD systems are not designed to detect liquid leaks.
3-5
-------
Table 3-5. Details for Vehicles with SHED Results > 1 g/Qhr (excluding liquid and unmonitored vapor leaks)
Table 3-5A
PacketID
LIP-22
CRL-540
LIP-256
CRL-475
HE-6725
LIP- 193
HE-6763
HE-6011
HE-6799
HE-6018
Evap Family
2FJXR01251BB
XGMXR0124912
YGMXE0111911
WTYXE0095AE1
WFMXE0045AAA
VFM1057BYMA
2FJXR01251CC
TFM2045AYPBA
WFMXE0115BAE
Dataset
LIP AN
CARYL
LIP AN
CARYL
DENVER
LIP AN
DENVER
DENVER
DENVER
DENVER
Model
Year
2002
1998
1999
2000
1998
1998
1997
2002
1996
1998
Make
SUBARU
HONDA
OLDS
CHEV
TOYOTA
FORD
MERCURY
SUBA
FORD
FORD
Model
IMPREZA
ACCORD
ALERO
ASTRO VAN
CAMRY
MUSTANG
VILLAGER
LEGACY
RANGER
TAURUS
SHED g/15mi
n
1.30
2.10
2.18
2.45
6.30
11.22
20.81
5.21
12.21
19.93
PSHED over IgHC
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
evap DTC
No
No
No
No
No
No
No
Yes
Yes
Yes
Table 3-5B
PacketID
LIP-22
CRL-540
LIP-256
CRL-475
HE-6725
LIP- 193
HE-6763
HE-6011
HE-6799
HE-6018
Evap Family
2FJXR01251BB
XGMXR0124912
YGMXE01 11911
WTYXE0095AE1
WFMXE0045AAA
VFM1057BYMA
2FJXR01251CC
TFM2045AYPBA
WFMXE0115BAE
Dataset
LIP AN
CARYL
LIP AN
CARYL
DENVER
LIP AN
DENVER
DENVER
DENVER
DENVER
evap mon ready
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Yes
Yes
DTC codes
-
-
-
-
-
-
-
451|457
443
443
Should Malfunction be
Detectable by OBD?
Not ascertainable*, **
Not ascertainable*, **
Not ascertainable*, **
Yes**
Yes***
Not ascertainable*, ***
yes - purge solenoid
Yes***
Yes -purge solenoid
Yes -purge solenoid
Vapor/ Liquid/ No Leak
Identified
No Leak Identified
Vapor
No Leak Identified
Vapor
Vapor
No Leak Identified
Vapor
Vapor
Vapor
Vapor
*The source of the high SHED emissions could not be determined from the technician comments in the field studies.
**Manufacturer and EPA certification representatives have stated that these manufacturers are certified to California enhanced OBD requirement of a 0.020"
diameter orifice detection capability.
***The magnitude of the emissions indicates that this would have been equivalent to at least a 0.040" diameter orifice and therefore should have been detected
by Federal OBD.
3-6
-------
Table 3-5C
PacketID
LIP-22
CRL-540
LIP-256
CRL-475
HE-6725
LIP- 193
HE-6763
HE-6011
HE-6799
HE-6018
Evap Family
2FJXR01251BB
XGMXR0124912
YGMXE0111911
WTYXE0095AE1
WFMXE0045AAA
VFM1057BYMA
2FJXR01251CC
TFM2045AYPBA
WFMXE0115BAE
Dataset
LIP AN
CARYL
LIP AN
CARYL
DENVER
LIP AN
DENVER
DENVER
DENVER
DENVER
Inspector Comments
Nothing Found
Sniffer leak detected at vapor canister. No non-OEM installations. Did not find fuel filter.
Canister leak = vapor not liquid.
Nothing Found
Suspect that fuel pump replacement damaged gasket. Areas on top of tank and near tank to fill
pipe connect set sniffer off. Also evap canister set sniffer off. No access to engine and that part
of fuel system.
Gas cap was cause of leak.
Nothing Found
Vapor/purge hose cracked.
Vapor lines at the top of filler neck not connected.
Found wires corroded/broken that activated the purge valve. Supplying power and ground
showed the purge valve did not operate. Removed canister and evacuated over night.
Extreme HC odor at right rear. Exhaust collector is right behind this area.
Table 3-5 Legend
• P0440 Evaporative Emission
• P0441 Evaporative Emission
• P0442 Evaporative Emission
• P0443 Evaporative Emission
• P0444 Evaporative Emission
Open
• P0445 Evaporative Emission
Shorted
• P0446 Evaporative Emission
Malfunction
• P0447 Evaporative Emission
• P0448 Evaporative Emission
• P0449 Evaporative Emission
Malfunction
• P0450 Evaporative Emission
Control System Malfunction
Control System Incorrect Purge Flow
Control System Leak Detected (small leak)
Control System Purge Control Valve Circuit
Control System Purge Control Valve Circuit
Control System Purge Control Valve Circuit
Control System Vent Control Circuit
Control System Vent Control Circuit Open
Control System Vent Control Circuit Shorted
Control System Vent Valve/Solenoid Circuit
Control System Pressure Sensor Malfunction
• P0451 Evaporative
Range/Performance
• P0452 Evaporative
• P0453 Evaporative
• P0454 Evaporative
• P0455 Evaporative
• P0456 Evaporative
• P0457 Evaporative
• P0458 Evaporative
• P0459 Evaporative
• P0465 Purge Flow
• P0466 Purge Flow
• P0467 Purge Flow
• P0468 Purge Flow
• P0469 Purge Flow
Emission Control System Pressure Sensor
Emission Control System Pressure Sensor Low Input
Emission Control System Pressure Sensor High Input
Emission Control System Pressure Sensor Intermittent
Emission Control System Leak Detected (gross leak)
Emissions System Small Leak Detected
Emission Control System Leak Detected
Emission System Purge Control Valve Circuit Low
Emission System Purge Control Valve Circuit High
Sensor Circuit Malfunction
Sensor Circuit Range/Performance
Sensor Circuit Low Input
Sensor Circuit High Input
Sensor Circuit Intermittent
3-7
-------
As shown in Table 3-3, twenty vehicles with an evaporative system monitor status of
"ready" had SHED measurements greater than 1 g/Qhr. Details of these 20 vehicles are listed in
Appendix A. However, it can be seen that only 10 vehicles with SHED measurements greater
than 1 g/Qhr are listed in Table 3-5, and of these only 8 have evaporative monitors with a
"ready" status (as opposed to all 20 vehicles in Table 3-3). The following reasons explain
differences between the "ready" vehicles emitting greater than 1 g/Qhr in Table 3-5 and the
vehicles listed in Table 3-3:
• Table 3-5 includes two vehicles (vehicles associated with packet IDs LIP-256 and
CRL-475) whose evaporative system monitor status was "not ready" (these are
not included in the counts provided in Table 3-3). Appendix A provides details of
vehicles with both "ready" and "not ready" evaporative system monitors.
• OBD evaporative system monitoring is not designed to identify liquid leaks;
since Table 3-5 was developed to assess OBD's ability to identify monitored
vapor leaks, these vehicles with (unmonitored) liquid leaks which are included in
Table 3-3 are not listed in Table 3-5.
• Since many 1996 and 1997 vehicles which conform to Federal (non-California)
standards were only designed to monitor purge solenoid checks but not vapor
leaks18, those model year 1996 and 1997 vehicles included in Table 3-3 for which
the inspection indicated the malfunction was anything other than a purge solenoid
were excluded from Table 3-5, as these would be malfunctions not monitored by
OBD. Again, details on each of these vehicles are provided in Appendix A.
• As OBD requirements developed and phased in, it became common practice for
auto companies, such as those listed in Table 3-5, to certify to the California OBD
leak detection standard of 0.020" all of their vehicles sold federally. Laboratory
data has shown a 0.020" diameter orifice to be roughly equivalent to an emission
rate of 1 g/Qhr. Using the formula for the area of a circle (rcr2), it can be seen that
the magnitude of emissions quadruples with the doubling of the diameter size to
0.040", and this has been found to be the case with laboratory testing comparing
0.020" and 0.040" leaks. Consequently, based on the magnitude of emissions, for
the vehicles in Table 3-5 with SHED emissions of 6.30 g/Qhr and greater, it is
reasonable to assume that they have leaks of diameter 0.040".
Based on these two datasets, this preliminary analysis suggests that OBD systems did not
identify 50-70% of the potential high evaporative emitters in these studies. The lower range,
50% (3 out of 6), is based on the occurrences of when the OBD system should probably have
detected the leak from a known source based on the technician comments, disregarding the "?"
vehicles in the "Should Malfunction be Detectable by OBD?" column. The higher range, 70% (7
18 40 CFR 86.094-17
3-8
-------
out of 10), includes the unknown leak sources ("?") for which no leak was identified or the vapor
leak was likely located in the area of the evaporative control system inaccessible to the HC
detector. Liquid leaks (which would generally not be monitored by OBD) were not included in
this estimate.
It should be noted that this preliminary analysis was based on a qualitative review of the
technician comments, rather than a quantitative assessment of OBD monitoring criteria (such as
a pressure drop associated with a 0.020" or 0.040" diameter orifice leak). In addition, since 2 of
these 10 vehicles had a "not ready" evaporative system monitor, it is possible that the vehicle's
OBD system would have detected the leak and stored a DTC once the evaporative system
monitor achieved readiness. However, as most I/M programs in the United States typically allow
1 to 2 "not ready" monitors, these vehicles could potentially pass an I/M OBD inspection with
these types of evaporative system failures before the monitor achieved readiness.
Some of the vehicles in these studies were model year 1996 through 1998 vehicles and
therefore may not have been equipped with OBD enhanced evaporative system monitoring. For
these older vehicles, some of the evaporative system failures could be undetectable by the OBD
systems installed on those vehicles. ERG compared the model year distributions with high/low
SHED categories to determine if a correlation existed between SHED values and vehicle model
year. The high SHED pairs included vehicles with model years from 1996-2003, and the low
SHED pairs included vehicles with model years from 1996-2010. This suggests that high SHED
results tended to be associated with older model year vehicles. Table 3-6 lists model year counts
for high and low SHED pairs vehicles of model year 1996-2010 in the dataset.
Table 3-6. Model year Counts for High and Low SHED Pairs in the Data
Model
Year
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
SHED Cut Point of 1.0 g/Qhr
# of Pairs with
SHED Value High
5
7
4
-
2
-
2
-
-
-
-
-
# of Pairs with
SHED Value Low
9
13
10
10
12
15
17
12
9
13
-
1
SHED Cut Point of 0.3 g/Qhr
# of Pairs with
SHED Value High
7
9
4
1
2
1
2
1
-
-
-
-
# of Pairs with
SHED Value Low
7
11
10
9
12
14
17
11
9
13
-
1
3-9
-------
Table 3-6. Model year Counts for High and Low SHED Pairs in the Data
Model
Year
2008
2009
2010
Total
Vehicles
SHED Cut Point of 1.0 g/Qhr
# of Pairs with
SHED Value High
-
-
-
20
# of Pairs with
SHED Value Low
2
4
2
129
SHED Cut Point of 0.3 g/Qhr
# of Pairs with
SHED Value High
-
-
-
27
# of Pairs with
SHED Value Low
2
4
2
122
3-10
-------
4.0 Conclusions
Three different datasets were used in this study to determine if high evaporative emitters,
as defined by LSHED or PSHED tests, were appropriately identified by each vehicle's OBD
system. Although the total number of observations was limited, this analysis indicates that many
vehicles with high SHED values (hot-soak emissions equal to or greater than 1 g/Qhr as
measured by a SHED test) do not have stored evaporative DTCs or MILs. For vehicles with
high SHED HC measurement results, an attempt was made to determine the source of the vapor
leak. Results are shown in Table 4-1, which summarizes pertinent information for all vehicles
listed in Table 3-5. Due to the configuration of the vehicle chassis the source or sources of each
leak could not always be isolated. Table 4-1 shows that OBD systems did not identify 50-70% of
the high evaporative emitters in these studies (for various reasons including "not ready"
evaporative system monitors). The lower range, 50% (3 out of 6), is based on evaporative
system failures that should probably have been detected by the OBD system. The higher range,
70% (7 out of 10), includes vehicles for which no leak source was identified or the vapor leak
was likely located in an area of the evaporative control system inaccessible to the HC detector
("Not ascertained" in Table 4-1). Liquid leaks (which generally are not monitored by OBD)
were not included in this estimate.
From an overall perspective, this study identified 29 vehicles (27 listed in Table 3-4 plus
2 with "not ready" evaporative monitors) with PSHED values greater than 0.3 g/Qhr (roughly
equivalent to the hot-soak component of the 2 g/hr Diurnal + Hot Soak standard), and only 4 of
those 29 vehicles had stored evaporative DTCs. Two vehicles had "not ready" evaporative
monitors. The source of elevated evaporative emissions from the remaining 23 "ready" vehicles
which had no evaporative DTCs was either undetectable by current OBD design (for example, a
hidden liquid leak not observed by field technicians), was from a part of the system not
monitored by OBD, or was a vapor leak which should have been identified but was not (i.e., the
evaporative emissions monitoring system was malfunctioning in some way).
No "OBD false positive" vehicles were identified in this study (those with stored
evaporative system DTCs with SHED emissions under 0.3 g/Qhr). The preferential recruiting of
vehicles with high evaporative emissions would tend to reduce the likelihood of this category of
vehicles from participation in these studies, so no conclusions can be drawn from the lack of
these vehicles in these results. However, the key finding of this report is the high number of
vehicles with high evaporative emissions which were not identified by the OBD evaporative
system.
4-1
-------
Since the sample size used for this analysis was limited to ten vehicles which were
expected to set DTCs, EPA sought to enhance our understanding with larger OBD datasets.
Analysis was performed using inspection and maintenance data from Texas and California.
Preliminary conclusions from detailed analysis of data from these states complement the
conclusions of this report and can be found in Appendix B.
Table 4-1. Comparison of SHED and OBD Results for Vehicles with SHED
Emissions > 1 g/Qhr (Summary of Information for Vehicles Listed in Table 3-5)
Data Set
Lip an
Ken Caryl
Lip an
Ken Caryl
Denver
Lip an
Denver
Denver
Denver
Denver
Model
Year
2002
1998
1999
2000
1998
1998
1997
2002
1996
1998
SHED
g/15
min
1.30
2.10
2.18
2.45
6.30
11.22
20.81
5.21
12.21
19.93
Evap
DTC
No
No
No
No
No
No
No
Yes
Yes
Yes
Evap
Monitor
Ready
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Evaporative
DTC code
-
-
-
-
-
-
-
451/457
443
443
Should Malfunction
be detectable by
OBD?
Not ascertainable*, **
Not ascertainable*, **
Not ascertainable*, **
Yes**
Yes***
Not ascertainable*, ***
Yes - purge solenoid
Yes***
Yes - purge solenoid
Yes - purge solenoid
*The source of the high SHED emissions could not be determined from the technician comments in the field studies.
**Manufacuter and EPA certification representatives have stated that these manufacturers certified to California
enhanced OBD requirement of a 0.020" diameter orifice detection capability.
*** the magnitude of the emissions indicates that this would have been equivalent to at least a 0.040" diameter
orifice and therefore should have been detected by OBD on a federally certified vehicle
4-2
-------
Appendix A
-------
Appendix A: Details for All Study Vehicles with SHED Results > 0.3 g/Qhr
PacketID
LIP-254
CRL-568
CRL-393
CRL-546
HE-6454
CRL-395
CRL-271
LIP-22
HE-6418
CRL-540
LIP-256
CRL-357
CRL-475
HE-6331
HE-6609
CRL-85
HE-6725
HE-6647
HE-6759
LIP- 193
CRL-321
HE-6620
HE-6519
HE-6702
HE-6763
CRL-316
HE-6011
HE-6799
HE-6018
Evap Family
TCR1073AYPBP
VGM1035AYPAA
TNT1047DYMAO
XFMXE0120BAE
VGM1095AYMEA
VGM1098AYMBA
2FJXR01251BB
VFM1065BYMAA
XGMXR0124912
VNS1110AYMEA
YGMXE0111911
TFM1045AYPAB
YGMXE0095905
TCR1073AYPOB
WTYXE0095AE1
YSZXT0095MEO
VGM1070AYMAA
WFMXE0045AAA
VGM1095AYMEA
TNS1057BYMBB
YTY1047DYMAO
TGM1089AYMEA
VFM1057BYMA
3HYXR0105PEA
2FJXR01251CC
TFM2045AYPBA
WFMXE0115BAE
Dataset
LIP AN
CARYL
CARYL
CARYL
DENVER
CARYL
CARYL
LIP AN
DENVER
CARYL
LIP AN
CARYL
CARYL
DENVER
DENVER
CARYL
DENVER
DENVER
DENVER
LIP AN
CARYL
DENVER
DENVER
DENVER
DENVER
CARYL
DENVER
DENVER
DENVER
Model Year
1996
1997
2001
1996
1999
1997
1997
2002
1997
1998
1999
1997
2000
1996
2000
1996
1998
2000
1997
1998
1997
1996
1997
1996
1997
2003
2002
1996
1998
Make
JEEP
STRN
TOYOTA
GEO
MERCURY
CHEV
CMC
SUBA
FORD
HOND
OLDS
NISS
CHEV
FORD
OLDS
DODG
TOYOTA
ISUZU
OLDS
FORD
CHEV
NISS
TOYT
CHEV
MERCURY
HYUN
SUBA
FORD
FORD
Model
CHEROKEE
SL
SOLARA
PRIZM
MOUNTAINEER
S10
SIERRA
IMPREZA
ASPIRE
ACCORD
ALERO
PATHFINDER
ASTRO VAN
CONTOUR
BRAVADA
RAM 1500
CAMRY
RODEO
DELTA 88
MUSTANG
S10
QUEST
COROLLA
CAVALIER
VILLAGER
ACCENT
LEGACY
RANGER
TAURUS
SHED
g/15min
0.33
0.39
0.46
0.62
0.67
0.72
1.28
1.30
1.47
2.10
2.18
2.32
2.45
2.79
5.77
6.08
6.30
8.65
9.10
11.22
12.33
12.79
16.41
19.53
20.81
0.69
5.21
12.21
19.93
PSHED
>1
No
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
PSHED
>.3
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
evap
DTC
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Yes
evap
monitor
ready
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
DTC codes
•1-
•I-
733|734|. .
1135|1155
•I-
•1-
...
•I-
•I-
172|175|. .
•I-
•I-
•I-
•I-
•I-
•I-
•I-
•I-
•I-
•I-
325|1447|.
•1-
..
•I-
133|304|. .
•I-
•1-
•I-
•I-
420|.|.
•I-
401|463|171|172|174
•I-
•I-
•1-
•I-
•I-
•I-
400|325|. .
•I-
•1-
•I-
•I-
325|300|. .
442|650|46
31
451|457|420|.|.
443|.|.
•I-
1518|443.
..
A-l
-------
Appendix A: Details for All Study Vehicles with SHED Results > 0.3 g/Qhr
PacketID
LIP-254
CRL-568
CRL-393
CRL-546
HE-6454
CRL-395
CRL-271
LIP-22
Evap Family
TCR1073AYPBP
VGM1035AYPAA
TNT1047DYMAO
XFMXE0120BAE
VGM1095AYMEA
VGM1098AYMBA
2FJXR01251BB
Dataset
LIP AN
CARYL
CARYL
CARYL
DENVER
CARYL
CARYL
LIP AN
Model Year
1996
1997
2001
1996
1999
1997
1997
2002
Make
JEEP
STRN
TOYOTA
GEO
MERCURY
CHEV
CMC
SUBA
Model
CHEROKEE
SL
SOLARA
PRIZM
MOUNTAINEER
S10
SIERRA
IMPREZA
Should
Malfunction
be
detectable
by OBD?
no evap
codes yet*
no evap
codes yet*
?
no evap
codes yet*
?
no evap
codes yet*
no evap
codes yet*
?
Vapor/
Liquid/
No
Leak
Found
Vapor
Vapor
Vapor
Vapor
No
Leak
Found
Vapor
Vapor
No
Leak
Found
Comments
Gas cap IM failure. MCM Comments:
Nothing found.
Gas cap IM failure. MCM Comments:
Sniffer goes off inconsistently in area
of fuel tank especially around top.
Odor of gasoline under vehicle in fuel
tank area.
Underbody-fuel lines not visible but
accessible to sniffer. Fuel injectors
pretty much hidden below intake
manifold.
While lying below fuel tank area I
thought I got a faint whiff of gasoline.
Bruce said same as we were closing
PSHED door.
No comments
Tank is severely dented. Cannot see
liquid fuel but top of fuel tank
activates sniffer. Can hear fuel pump
whining while driving.
Cannot see filler to tank connection.
Fuel tank skid plate has some damage.
Around top of tank on fill pipe side
(right) the sniffer went off particularly
near the front of tank.
Nothing Found
A-2
-------
Appendix A: Details for All Study Vehicles with SHED Results > 0.3 g/Qhr
PacketID
HE-6418
CRL-540
LIP-256
CRL-357
CRL-475
HE-6331
HE-6609
CRL-85
Evap Family
VFM1065BYMAA
XGMXR0124912
VNS1110AYMEA
YGMXE0111911
TFM1045AYPAB
YGMXE0095905
TCR1073AYPOB
Dataset
DENVER
CARYL
LIP AN
CARYL
CARYL
DENVER
DENVER
CARYL
Model Year
1997
1998
1999
1997
2000
1996
2000
1996
Make
FORD
HOND
OLDS
NISS
CHEV
FORD
OLDS
DODG
Model
ASPIRE
ACCORD
ALERO
PATHFINDER
ASTRO VAN
CONTOUR
BRAVADA
RAM 1500
Should
Malfunction
be
detectable
by OBD?
no evap
codes yet*
?
?
no evap
codes yet*
yes
no evap
codes yet*
no
no evap
codes yet*
Vapor/
Liquid/
No
Leak
Found
Vapor
Vapor
No
Leak
Found
Vapor
Vapor
Vapor
Liquid
Liquid
Comments
No visible liquid leaks, but occasional
HC from gas analyzer. Filter mount is
broken. Fuel filter was replaced, and
the mount was repaired. Fuel spilled
down the driver's side strut mount
during replacement. This was picked
up during after repair 1 test.
Sniffer leak detected at vapor canister.
No non-OEM installations. Did not
find fuel filter. Canister leak = vapor
not liquid.
Nothing Found
There is some sort of contraption that
sets sniffer off back between fuel tank
and canister (see photo).
Suspect that fuel pump replacement
damaged gasket. Areas on top of tank
and near tank to fill pipe connect set
sniffer off. Also evap canister set
sniffer off. No access to engine and
that part of fuel system.
Significant HC from top of tank.
Visible fuel stainage bottom of tank.
Tank looks stained possibly from
gasoline. I can smell gasoline when
lying under vehicle by filler neck.
The vent line clamp is rusted and
looks like it has been wet but sniffer is
inconsistent about sensing fumes in
that area.
A-3
-------
Appendix A: Details for All Study Vehicles with SHED Results > 0.3 g/Qhr
PacketID
HE-6725
HE-6647
HE-6759
LIP- 193
CRL-321
HE-6620
HE-6519
HE-6702
HE-6763
CRL-316
Evap Family
WTYXE0095AE1
YSZXT0095MEO
VGM1070AYMAA
WFMXE0045AAA
VGM1095AYMEA
TNS1057BYMBB
YTY1047DYMAO
TGM1089AYMEA
VFM1057BYMA
3HYXR0105PEA
Dataset
DENVER
DENVER
DENVER
LIP AN
CARYL
DENVER
DENVER
DENVER
DENVER
CARYL
Model Year
1998
2000
1997
1998
1997
1996
1997
1996
1997
2003
Make
TOYOTA
ISUZU
OLDS
FORD
CHEV
NISS
TOYT
CHEV
MERCURY
HYUN
Model
CAMRY
RODEO
DELTA 88
MUSTANG
S10
QUEST
COROLLA
CAVALIER
VILLAGER
ACCENT
Should
Malfunction
be
detectable
by OBD?
yes
no
no evap
codes yet*
?
no evap
codes yet*
no evap
codes yet*
no evap
codes yet*
no evap
codes yet*
yes - purge
solenoid
small leak -
.02", fuel
level sensor
Vapor/
Liquid/
No
Leak
Found
Vapor
Vapor
No
Leak
Found
No
Leak
Found
Liquid
Vapor
No
Leak
Found
Liquid
Vapor
Vapor
Comments
Gas cap was cause of leak.
There is a fuel odor under hood.
Sniffer in V engine valley under
intake manifold sets it off.
No comments
Nothing Found
Major gasoline leak with gas puddling
on engine. Not sure where leak is
exactly. Some part of supply or
injector to TBI system?
Major HC at mid tank area, driver's
side.
No comments on MCM test.
Comment on repair note said that the
trouble code indicated P0401 EGR
flow. Investigate to see why this was
not in IM.
#3 fuel injector leaks, gas cap is
missing, #1 fuel injector started
leaking.
Vapor/purge hose cracked.
Something seems to be damaged or
leaking at top of fuel tank. Sniffer
active many places around top pipe to
tank joint. Looks dry - tank appears to
have evaporated fluid of some kind
but that did not activate sniffer.
A-4
-------
Appendix A: Details for All Study Vehicles with SHED Results > 0.3 g/Qhr
PacketID
HE-6011
HE-6799
HE-6018
Evap Family
2FJXR01251CC
TFM2045AYPBA
WFMXE0115BAE
Dataset
DENVER
DENVER
DENVER
Model Year
2002
1996
1998
Make
SUBA
FORD
FORD
Model
LEGACY
RANGER
TAURUS
Should
Malfunction
be
detectable
by OBD?
purge
solenoid
purge
solenoid
Vapor/
Liquid/
No
Leak
Found
Vapor
Vapor
Vapor
Comments
Vapor lines at the top of filler neck not
connected.
Found wires corroded/broken that
activated the purge valve. Supplying
power and ground showed the purge
valve did not operate. Removed
canister and evacuated over night.
Extreme HC odor at right rear.
Exhaust collector is right behind this
area.
* 1996 and 1997 vehicles and 1998 federal-only OBD vehicles did not have OBD evaporative system leak checks but did have purge solenoid checks.
A-5
-------
Appendix B
-------
\ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
NATIONAL VEHICLE AND FUEL EMISSIONS LABORATORY
2565 PLYMOUTH ROAD
PB0^ ANN ARBOR, MICHIGAN 48105-2498
OFFICE OF
AIR AND RADIATION
February 22, 2013
MEMORANDUM
SUBJECT: Preliminary Analysis of On-Board Diagnostic (OBD) Evaporative System
Information from State Inspection and Maintenance (I/M) Stations - California
and Texas
FROM: Carl Fulper, Chemical Engineer, Assessment and Standards Division
TO: Docket EPA-HQ-OAR-2011-0135
Introduction:
EPA is reviewing data gathered from selected state I/M test programs on vehicle
evaporative emission system OBD codes and their evaporative monitor "ready" information to
help inform our understanding of the incidence rate of evaporative system monitor "not ready"
events and the incidence of evaporative system related DTCs. While five states have been
selected for analysis, this memo summarizes completed analyses for Texas and California. Since
the data was gathered by the states under different protocols and time periods, the contents of the
data sets are not identical. To provide some degree of uniformity in our analysis, we examined
the data for five model years (MY) covering (2000 - 2004) when all systems (absent specific
exemptions called "deficiencies") were required to be OBDII compliant (which includes
evaporative system monitoring) and all passenger cars and light trucks would meet the enhanced
evaporative emissions standards promulgated in 1993. We looked at only calendar years of data
beginning after the initial state I/M new vehicle exemption period. A state's new vehicle
exemption period can vary between 2 to 4 calendar years and require a vehicle to be test with
annual, biennial or a hybrid of these two frequencies (see Table 1). This memorandum shows
the preliminary analysis of evaporative monitor readiness and their associated evaporative
diagnostic trouble codes (DTCs).
Internet Address (&RL) • http://www.epa.gov
Recycled/Recyclable • Printed with Vegetable Oil Based Inks on 100% Postconsumer, Process Chlorine Free Recycled Paper
-------
Table 1. List of I/M OBD Data
State
TX
CA
MY
Exemption
2
4
Calendar
years
1/Mdata
Evaluated
2004-2010
2004-2010
Network Type
Decentralized
Hybrid
Program
Frequency
Annual
Biennial
OBD
MIL+DTC
Yes
Yes
Evap
Testing
Gas Cap
Gas Cap
The OBDII system monitors the performance of various engine and emission control
systems, including the vehicle's evaporative emissions control system. The vehicle software
monitors these systems and when monitoring enable conditions are met, conducts self-
diagnostics tests to see if they are performing properly. Whether or not a vehicle's evaporative
control system monitoring has been conducted and completed (self diagnostic program on the
vehicle), it will indicate by a "ready" or "not ready" code that is stored in the vehicle's computer.
If the monitoring has been completed and a problem has been found, using established protocols
a DTC will be set. The presence of an evaporative DTC indicates a system malfunction has been
detected which could result in elevated evaporative emissions from that vehicle. These OBD
DTCs are stored on the vehicle and can be retrieved when the vehicle goes for its routine I/M
inspection. It is these recorded monitor codes and DTCs that were used by EPA to conduct its
analysis.
A 'not ready" monitor on the vehicle's evaporative system indicates that the system has
not completed monitoring. Normally, if the system is ready, the OBD evaporative system will
run and update results at each key on event. There are some circumstances, normally called
"global disenables", when the system is not expected to run. Key global disenables are based on
factors such as tank fill level, outside air temperature, and elevation. In addition, there are some
relatively rare instances when a manufacturer is granted an individual "deficiency" for a given
vehicle system. This occurs when the OBD system cannot give a reliable monitoring result in
certain operating circumstances even though it would be expected to do so under the regulations.
Furthermore, this can occur if the readiness code was cleared as a result of external actions such
as a battery disconnect or system maintenance. When the codes are cleared, the vehicle's
evaporative emissions system monitor will indicate a "not ready" code until the vehicle
completes a self diagnostic test and will change the vehicle's status from "not ready" to "ready'
and record any DTCs on the vehicle's evaporative system.
Numerous filters were applied to each state's data to remove any inspection records that
should not be used for this analysis. This included for example; heavy-duty vehicles, vehicles
with a fuel type other than gasoline, special inspections performed at government-operated
stations and inspections performed outside the enhanced I/M areas. All records with invalid
vehicle identification numbers (VINs) and passenger cars and light trucks with model year 1995
and older were removed since they are not required to have enhanced evaporative emission
controls or OBDII.
-------
Analysis:
Our analysis of two out of the five states (Texas and California) has been completed. The
findings are discussed below. We will first describe the I/M OBD data sets, then the evaporative
emissions monitor readiness rates, next the rates for evaporative system DTCs and finally
evaporative system leak DTC rates. For internal consistency, this analysis looked at MYs 2000-
2004 vehicle evaluated at I/M during calendar years 2004-2010. However, additional
information is included in the tables.
The State of Texas has an annual I/M vehicle inspection program with in most cases a
two year exemption for new vehicles. The dataset that was analyzed for Texas includes seven
calendar years of data (2004-2010) containing over 41 million inspection cycles and covers over
11 million individual vehicles. The State of California has a hybrid I/M vehicle inspection
program with in most cases a four year exemption for new vehicles.1 The dataset that was
analyzed for California includes seven calendar years of data (2004-2010) containing over 47
million inspection cycles and covers over 17 million individual vehicles.
Tables 2 and 3 shows the percent of the vehicle's evaporative monitors that are "ready"
in a given I/M cycle by calendar year and vehicle model year. Data from these tables indicate
that about 3-16 percent of the vehicles arrived at the I/M station with their evaporative emission
monitors in a "not ready" mode. Furthermore, data from both states shows that as the vehicle
population ages, they are more likely to have a "not ready" evaporative monitor in a given
calendar year. Evaporative emission monitor readiness at the I/M inspection improved by 1 -2
percentage points over time relative to the first year for two model years of the ten model years
studied (2000-2004 for CA and TX) but generally decreased by 2-10 percentage points for
model year 2000-2004 vehicles evaluated at I/M during calendar years 2004-2010.
For the vehicles in these two states that had their evaporative monitor "ready" for at least
one inspection of their I/M cycle in each calendar year, EPA conducted additional analysis to
determine if one or more evaporative DTCs were set at any time during the I/M cycle. A list of
the evaporative DTCs that were analyzed for is listed in Table 4. Tables 5 and 6 show that for
the first inspections for a model year of vehicles evaporative system DTCs were set on 0.2 to 1.1
percent of vehicles. However, as the fleets aged, these percentages increased. For model year
2000-2004 vehicles evaluated at I/M during calendar years 2004-2010, evaporative related DTC
rates increased by a factor in the range of 1.7 to 5. The median value was about 3. This is an
important finding, since it can at least be hypothesized that a DTC found in I/M in Texas or
California is usually fixed. Thus, this indicates not only that similar problems occur in
subsequent years within a model year's vehicles but usually at a greater rate. Absent repair, the
total number of evaporative DTC occurrences may be more cumulative in nature.
EPA also analyzed the two states for vehicles that had their evaporative monitors convert
from "not ready" in their previous I/M cycle to "ready" when they came back to the I/M station.
The results show a higher percentage of evaporative DTCs set than the whole dataset by 0.1 to
1.5 percent depending on vehicle model year to calendar year. This could suggest that some of
1 California began a six year exemption after the 2004 model year.
2 In some cases these vehicles are required to return at a later date.
3
-------
that vehicle's previous I/M cycle evaporative monitor "not ready" could be masking a set of
evaporative DTCs.
Leaks from fuel and evaporative systems are potentially significant source of VOCs.
Since OBD systems are required to monitor for vapor leaks when the system is ready, EPA
examined the evaporative DTC code data for the prevalence of vapor leak DTCs (P0455, 0456,
0457, 0442). For both Texas and California, a review of the data shows that leaks represent 60-
80 percent of all evaporative system DTCs. As is shown in Tables 7-8 this is generally the same
for all model years evaluated (2000-2004) in calendar years (2004-2010). As before for
evaporative DTCs, this suggests that, even assuming repair for problems found at I/M, the
absolute number of leaks in the fleet increases with time since the overall number of vehicles
with evaporative-related DTCs increases with time.
Preliminary Conclusions.
A review of the information leads to the following conclusions:
• Depending on age, 0.3-2.5 percent of vehicles with evaporative monitors ready came into the
I/M stations with evaporative related DTC
• The percent of vehicles with evaporative emission related MILs set increased by a factor of 2
- 4 over about 5 years
* Evaporative emission monitors were not ready for 3-16 percent of vehicles when arriving at
the I/M station.
• The percent of vehicles with monitors ready at the I/M station generally decreased by 3 to 7
percentage points over about 5-6 years; decrease was less for model year vehicles less than
five years old.
• While it varies by age, 60-80 percent of evaporative systems DTCs are leak related.
-------
Table 2. State of Texas: Percent of Evap Monitors Ready for All Test Cycles by Calendar Year and Model Year
Calendar
Year
2003
2004
2005
2006
2007
2008
2009
2010
2011
Ready
Total
% R.lv
Ready
Total
% Rdy
Ready
Total
% Rdy
Readv
Total
% Rdv
Ready
Total
% Rdy
Ready
Total
% Rdy
Ready
Total
% Rdv
Readv
Total
% Rdy
Ready
Total
% Rdy
Vehicle Model Year
1996
212,417
220,933
96.1%
313,693
326,483
96.1%
291,195
304,452
95.6%
250.340
263,481
95.0%
216,940
228.994
94.7%
190,786
201,734
94.6%
174,496
185,249
94.2%
153,703
164,677
93.3%
135,428
146,713
92.3%
1997
251,581
269,143
93.5%
367,795
394,701
93.2%
345,213
372,957
92.6%
311,992
341.632
91.3%
273,782
301.642
90.8°.,,
242.440
268,180
90.4%
223,925
250, 148
89.5%
200.074
225,624
88.7%
179,677
204,628
87.8%
1998
272,443
290,293
93.9%
399,109
426,793
93.5%
376.605
407,451
92.4%
347,948
383,548
90.7%
317,575
354.138
89.7%
283,472
318,843
88.9%
260.65 1
297,498
87.6%
234.004
271,529
86.2%
211,545
248,507
85.1%
1999
309,056
332.586
92.9%
449,390
485,114
92.6%
423,980
464.217
91.3%
395,327
441.628
89.5%
368,364
416.017
88.5%
341,183
391,220
87.2%
317,243
369,098
86.0%
285,068
337,927
84.4%
260.075
312.484
83.2%
2000
355,644
376.551
94.4%
515,902
548,579
94.0%
480,111
516.785
92.9%
461,071
504,553
91.4%
437,128
483.946
90.3%
412,567
463,453
89.0%
394,155
451,067
87.4%
357,546
416.928
85.8%
327.158
388,235
84.3%
2001
379,122
396,569
95.6%
547,999
579,579
94.6%
504,195
542,086
93.0%
482,983
528,661
91.4%
458.359
507.627
90.3%
435.470
487,595
89.3%
426,793
474,273
90.0%
396,415
447,224
88.6%
368.489
421,481
87.4%
2002
154.252
158,674
97.2%
572,621
591,211
96.9%
548,960
577,221
95.1%
529,371
565,441
93.6%
505,696
545.934
92.6%
484,262
528,414
91.6%
475,295
518,062
91.7%
448,922
495,984
90.5%
423,436
473,674
89.4%
2003
13,419
14,610
91.8%
157,985
163.647
96.5%
520.428
541,005
96.2%
520,779
552,213
94.3%
498,690
536.160
93.0%
477,145
520,419
91.7%
470,690
515,973
91.2%
447,643
498,683
89.8%
427.437
482,296
88.6%
2004
15,197
16,370
92.8%
135,366
139,968
96/7%
497,530
516,906
96.3%
513.471
542.118
94.7%
499,715
533,560
93.7%
498,950
533,256
93.6%
480,563
519,797
92.5%
464.433
508,086
91.4%
2005
14,489
15,490
93.5%
131,305
135,905
96.6%
508,401
526,145
96.6%
533,651
562.201
94.9%
535.342
564.203
94.9%
516,386
548,346
94.2%
501,214
538,304
93.1%
2006
13,181
14,329
92.0%
133,867
139.238
96.1%
523,024
545,437
95.9%
565,315
595,716
94.9%
545,403
579,195
94.2%
529,345
567,942
93.2%
2007
2008
.
12,569
13,679
91.9%
1 72,294
178,659
96.4%
610,822
632,552
96.6%
635,317
668.581
95.0%
611,814
648,549
94.3%
12,679
13.651
92.9%
163,513
167,504
97.6%
575,293
592,703
97.1%
611.047
638,320
95.7%
2009
7,972
8.462
94.2%
131,103
133,948
97.9%
372,079
382.941
97.2%
2010
9.493
10,041
94.5%
126,440
129.255
97.8%
2011
f
.
t
11,775
12,544
93.9%
-------
Table 3. State of California: Percent of Evap Monitors Ready for All Test Cycles in an I/M Cycle by Calendar Year
and Model Year
Calendar
Year
2004
2005
2006
2007
2008
2009
2010
Ready
Total
% Rdy
Ready
Total
% Rdy
Ready
Total
% Rdy
Ready
Total
% Rdy
Ready
Total
% Rdy
Ready
Total
% Rdy
Ready
Total
% Rdy
Vehicle Model Year
19%
810,561
843.246
96.1%
432,051
456,604
94.6%
667,343
699,816
95.4%
387,300
411,795
94.1%
540,399
571,533
94.6%
348,025
374,438
92.9%
463,884
497,470
93.2%_
1997
516,697
569,106
90.8%
842,942
917,775
91.8%
499,630
557,066
89.7%
704,889
775,171
90.9%
439,432
495,384
88.7%
600,445
672,594
89.3%
409,166
470,704
86.9%
1998
971,755
1,065,028
91.2%
501.595
572,812
87.6%
836,110
937,095
89.2%
478,407
557,093
85.9%
695.175
794,806
87.5%
432,646
519,581
83.3%
611,897
723,119
84.6%
1999
511,922
583,806
87.7%
1.040.290
1.150,154
90.4%
5 1 8,456
602,608
86.0%
893,653
1,012,319
88.3%
484,571
576,994
84.0%
769,712
896,492
85.9%
469,722
573,883
SI. 8%
2000
1,245,836
1,343,798
92.7%
296,247
349,579
84.7%
1,350,221
1,481,334
91.1%
442,636
518,372
85.4%
1,140,170
1,280,294
89.1%
452,706
546,853
82.8%
1,012,825
1,168,494
60.7%
2001
308,108
359,696
85.7%
243.609
290,644
83.8%
259,989
306.798
84.7%
1,314,166
1,439,185
91.3%
409,179
478,032
85.6%
1,147,469
1,287,831
89.1%
462.347
540.824
85.5%
2002
256,568
284,912
90.1%
118,992
138,695
85,8%
229,412
268,900
85.3%
239,320
277,118
86.4%
1,290,781
1,395,503
92.5%
380,440
440.307
86.4%
1,193,482
1,305,366
91.4%
2003
247,453
268,590
92.1%
79,784
91,474
87.2%
99.549
117,918
84.4%
215,297
254,939
84.5%
208,776
245,367
85.1%
1,308,578
1.430,133
91.5%
388,749
447,463
86.9%
2004
143.357
154,769
92.6%
77,762
85.050
91.4%
68.995
77,522
89.0%
87,833
100,926
87.0%
182,380
212,833
85.7%
188,646
216.017
87.3%
1,355,367
1,437,297
94.3%
2005
7,887
8,863
89.0%
51.387
55,153
93.2%
75.619
80,950
93.4%
64,136
69,741
92.0%
77,783
87,004
89.4%
167,453
189.553
88.3%
218,014
236,415
92.2%
2006
2,688
3,071
87.5%
45,474
48,352
94.0%
68,348
71,905
95.1%
58.321
62.259
93.7%
78,192
84,929
92.1%
196,057
209,837
93.4%
2007
3,122
3,424
91.2%
40,544
41,804
97J3%
58,816
60.203
97.7%
55,180
57,323
96.3%
95.001
98,327
96.6%
2008
2,115
2,137
99.0%
28.066
28,240
99.4%
56,542
56,966
99.3%
68,529
69,384
98.8%
2009
»
1,758
1,779
98.8%
19,145
19.266
99.4%
48,734
49,147
99.2%
2010
1,639
1,657
98.9%
23.220
23,496
98.8%
2011
2,013
2,034
99.0%
-------
Table 4. For List of Evap OBD DTCs
DTC Code
P0440
P0441
P0442
P0443
P0444
P0445
P0446
P0447
P0448
P0449
P0450
P0451
P0452
P0453
P0454
P0455
P0456
P0457
P0458
P0459
P0465
P0466
P0467
P0468
P0469
DTC Description
Evaporative Emission Control System Malfunction
Evaporative Emission Control System Incorrect Purge Flow
Evaporative Emission Control System Leak Detected (small leak)
Evaporative Emission Control System Purge Control Valve Circuit
Evaporative Emission Control System Purge Control Valve Circuit Open
Evaporative Emission Control System Purge Control Valve Circuit Shorted
Evaporative Emission Control System Vent Control Circuit Malfunction
Evaporative Emission Control System Vent Control Circuit Open
Evaporative Emission Control System Vent Control Circuit Shorted
Evaporative Emission Control System Vent Valve/Solenoid Circuit Malfunction
Evaporative Emission Control System Pressure Sensor Malfunction
Evaporative Emission Control System Pressure Sensor Range/Performance
Evaporative Emission Control System Pressure Sensor Low Input
Evaporative Emission Control System Pressure Sensor High Input
Evaporative Emission Control System Pressure Sensor Intermittent
Evaporative Emission Control System Leak Detected (gross leak)
Evaporative Emissions System Small Leak Detected
Evaporative Emission Control System Leak Detected
Evaporative Emission System Purge Control Valve Circuit Low
Evaporative Emission System Purge Control Valve Circuit High
Purge Flow Sensor Circuit Malfunction
Purge Flow Sensor Circuit Range/Performance
Purge Flow Sensor Circuit Low Input
Purge Flow Sensor Circuit High Input
Purge Flow Sensor Circuit Intermittent
7
-------
Table 5. State of Texas: For All Test Cycles with Evap Monitor Ready: Percent with Evap DTCs Set,
by Calendar Year and Model Year
Calendar Year
2003
1
2004
2005
2006
2007
2008
2009
2010
2011
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Vehicle Model Year
1996
1411
212417
0.66%
2220
313693
0.71%
2240
291195
0.77%
2056
250340
0.82%
1727
216940
0.80%
1571
190786
0.82%
1542
174496
0.88%
1239
153703
0.81%
1060
135428
0.78%
1997
2050
251581
0.81%
3314
367795
0.90%
3265
345213
0.95%
3335
311992
1.07%
2968
273782
1.08%
2757
242440
1.14%
2737
223925
1.22%
2303
200074
1.15%
2028
179677
1.13%
1998
2083
272443
0.76%
3663
399109
0.92%
3536
376605
0.94%
3914
347948
1.12%
3769
3 1 7575
1.19%
3756
283472
1.32%
3769
260651
1 .45%
3206
234004
1.37%
2753
211545
1.30%
1999
2053
309056
0.66%
3435
449390
0.76%
3533
423980
0.83%
3901
395327
0.99%
3942
368364
1.07%
3771
341183
1.11%
3982
317243
1.26%
3500
285068
1.23%
3027
260075
1.16%
2000
1927
355644
0,54%
3419
5 1 5902
0.66%
3959
480111
0.82%
4731
461071
1.03%
5165
437128
1.18%
5184
412567
1 .26%
5563
394155
1.41%
4958
357546
1.39%
4246
327158
1.30%
2001
2429
379122
0.64%
4496
547999
0.82%
5052
504195
1.00%
6069
482983
1.26%
6289
458359
1.37%
6297
435470
1.45%
6971
426793
1.63%
6542
396415
1.65%
6158
368489
1 .67%
2002
528
154252
0.34%
2688
572621
0.47%
3634
548960
0.66%
5283
529371
1.00%
5863
505696
1.16%
6074
484262
1.25%
7019
475295
1.48%
6943
448922
1.55%
6259
423436
1,18%
2003
30
13419
0.22%
533
157985
0.34%
3029
520428
0.58%
4916
520779
0.94%
6093
498690
1.22%
6693
477145
1.40%
7860 j
470690
1.67%
7766
447643
1.73%
7213
427437
1.69%
2004
48
15197
0.32%
728
135366
0.54%
3444
497530
0.69%
4674
513471
0.91%
5340
499715
I.(l7"o
6491
498950
1.30%
6757
480563
1.41%
6698
464433
1.44%
2005
44
14489
0.30%
675
131305
0.51%
3285
508401
0.65%
•-1020
533651
0.75%
5068
535342
0.95%
5589
516386
1.08%
5927
501214
1.18%
2006
39
13181
0.30%
565
133867
0.42%
2S9S
523024
0.55%
4176
565315
0.74%
4672
545403
0.86%
5193
529345
0.98%
2007
41
12569
0.33%
787
172294
0.46%
3284
610822
0.54%
4199
635317
0.66%
4549
611814
0.74%
2008
4S
12679
0.38%
666
163513
0.41%
2640
575293
0.46%
3322
611047
0.54%
2009
35
7972
0.44%
381
131103
0.29%
1303
372079
0.35%
2010
18
9493
0.19%
264
126440
0.21%
2011
.
18
11775
0.15%
-------
Table 6. State of California: For All Test Cycles with Evap Monitor Ready: Percent with Evap DTCs Set,
by Calendar Year and Model Year
Calendar
Year
2004
2005
2006
2007
2008
2009
2010
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Yes
Total
%w/DTC
Vehicle Model Year
1996
10.665
810,561
1.3%
7.163
432,05 1
1.7%
9,483
667,343
1.4%
6,632
387,300
1.7%
8.033
540.399
1.5%
6,167
348,025
1.8%
6.938
463,884
1.5%
1997
7,627
516,697
1.5%
12.622
842,942
1.5%
9,439
499,630
1.9%
12,149
704,889
1.7%
9.262
439,432
2.1%
11.432
600,445
1.9%
8.694
409,166
2.1%
1998
11.746
971,755
1.2%
8,082
501,595
1.6%
12,739
836.110
1.5%
9.407
478,407
2.0%
12,569
695,175
1.8%
10,015
432,646
2.3%
1999
6,137
511,922
1.2%
12,960
1,040,290
1.2%
7,916
518,456
1.5%
12,771
893,653
1.4%
8,683
484,571
1.8%
13,019
769,712
1.7%
12,665 9.486
611,897
2.1%
469,722
2.0%
2000
13,539
1,245,836
1.1%
3,330
296,247
1.1%
i 9,307
1.350,221
1.4%
7,976
442,636
1.8%
19,051
1,140,170
1.7%
9,709
452,706
2.1%
18,930
1,012,825
1.9%
2001
2.969
308.108
1.0%
2,739
243,609
1.1%
3,359
259.989
1.3%
23.772
1,314,166
1.8%
9.354
409.179
2.3%
23,801
1.147.469
2.1%
11,408
462,347
2.5%
2002
2,058
256,568
0.8%
1,209
118,992
1.0%
2,590
229,412
1.1%
3,022
239,320
1.3%
24,675
1,290,781
1.9%
9.136
380,440
2,4%
26,055
1,193,482
2.2%
2003
1,398
247,453
0.6%
671
79,784
0.8%
902
99,549
0.9%
1.966
215,297
0.9%
2.524
208,776
1.2%
23,952
i. 308.578
1,8%
8,010
388,749
2.1%
2004
515
143,357
0.4%
408
77.762
0.5%
482
68,995
0.7%
664
87,833
0.8%
1,566
182.380
0.9%
1,957
188,646
1.0%
20.256
1,355,367
1.5%
2005
8
7,887
0.1%
109
51,387
0.2%
288
75,619
0.4%
308
64,136
0.5%
391
77,783
0.5%
886
167,453
0.5%
1.269
218.014
0.6%
2006
1
2,688
0.0%
44
45,474
0.1%
81
68,348
0.1%
130
58,321
0.2%
188
78,192
0.2%
'159
196,057
0.2%
2007
5
3,122
0.2%
32
40.544
0.1%
64
58,816
0.1%
81
55,180
0.1%
125
95,001
0.1%
2008
0
2,115
0.0%
5
28,066
0.0%
17
56,542
0.0%
27
68,529
0.0%
2009
0
1,758
0.0%
4
19,145
0.0%
15
48.734
0.0%
2010
0
1,639
0.0%
7
23,220
0.0%
2011
0
2,013
0.0%
-------
Table?. State of Texas: Sum of Evaporative DTC Leaks/All Evap DTCs
By Calendar Year and Model year
Calendar
Year
2003
2004
2005
2006
2007
2008
2009
2010
1998
0.58
0.54
0.52
0.54
0.52
0.49
0.52
0.50
Model Year
1999
0.42
0.41
0.41
0.37
0.37
0.37
0.38
0.37
2000
0.70
0.68
0.67
0.66
0.67
0.64
0.62
0.62
2001
0.77
0.76
0.76
0.75
0.75
0.72
0.72
0.68
2002
0.72
0.73
0.75
0.74
0.72
0.72
0.71
2003
0.66
0.72
0.68
0.67
0.68
0.66
2004
0.74
0.73
0.69
0.71
0.69
2005
0.86
0.87
0.85
0.83
2006
0.87
0.84
0.84
2007
0.87
0.85
2008
0.89
DTC Ratio = Evaporative DTC Leak (P0442+P0455+P0456 +P0457)//Total Evap DTCs
10
-------
Table 8. State of California: Sum of Evaporative DTC Leaks/All Evap DTCs
By Calendar Year and Model year
Calendar
Year
2003
2004
2005
2006
2007
2008
2009
2010
Model Year
1996
0.11
0.10
0.11
0.11
0.11
0.12
0.13
1997
0.36
0.34
0.33
0.32
0.32
0.31
0.33
1998
0.72
0.68
0.67
0.63
0.65
0.62
0.65
1999
0.66
0.66
0.63
0.62
0.58
0.59
0.57
2000
0.80
0,76
0.76
0.72
0.73
0.69
0.71
2001
0.83
0.80
0.78
0.77
0.75
0.74
0.74
2002
0.86
0.83
0.84
0.80
0.80
0.76
0.78
2003
0.84
0.75
0.76
0.78
0.77
0.78
0.74
2004
0.87
0.78
0.77
0.81
0.78
0.80
0.79
2005
1.00
0.83
0.72
0.83
0.89
0.82
0.83
2006
0.77
0.80
0.80
0.85
0.82
2007
0.84
0.89
0.86
0.90
DTC Ratio = Evap (P0442+P0455+P0456+P0457)/Total Evap DTCs
Note: Highlight (yellow) are high vehicle recruitment years (biannual)
11
------- |