EPA's Proposal for MOBILE
     I/M Credits: Separation of Start and Running Emissions
                        October 7, 1997
                                
Abstract

     Inspection and Maintenance credits are based on two
criteria.  The first criterion is the identification rate, which
is the percent of emissions identified by a given test.  The
second criterion is repair effects, which is the difference
between the emissions before and after repair, based on a
mathematical curve determined for each test and cut-point
combination.  Credits in MOBILE5 are based on Emission Factor
Program repairs.  Currently, the same percent emission reduction
rate is used for both start and running emissions.  

Proposal

     For MOBILE6, EPA proposes using the same identification
rates that are currently  in MOBILE. EPA proposes to  examine
emissions reduction from repairs of start and running emissions
separately, to assess what effects, if any, this separation it
would have on credits.  

Preliminary Analysis

      EPA used it's Emissions Factor database to take a first
look at the separation of start and running emissions.  The
vehicles in this database are tested both in house and by
contractor.  The Emissions Factor Program is one of the only data
sources which contains effects on starts. Starts are available
since all the vehicles in this database were tested using the
Federal Test Procedure (FTP).  

     For this analysis, EPA tried to be as representative of an
I/M lane as possible.  Vehicles in an I/M lane must fail their
first test, in order to be sent for repairs.  After repairs, the
vehicle returns to the lane to be retested.  Vehicles for this
analysis were chosen in a similar fashion.  A vehicle had to have
failed its "as received" I/M test and passed its "after repairs"
I/M test to be chosen for this analysis.  

     A preliminary analysis was performed using vehicles that had
failed and then passed the I/M240 test.  The sample size was 22
cars and 5 trucks with model years ranging from 1981 to 1995. 
All the vehicles were classified as Tier 0 and were defined as
either normal or high emitters.  High emitters were determined
using these cut points from the LA4, HC > .8 and CO > 15.  This
preliminary analysis showed a reduction in emissions due to
repairs.  Though there was an effect on both cold start and
running emissions, the effect on running emissions for high
emitters was more notable. 



The following table is a summary of the result of this analysis.  

                        Normal Emitters           High Emitters
As Received Cold Start
CO Emissions (grams)         25.9                      69.6
 
After Repair Cold Start
CO Emissions (grams)         20.5                      34.9

As Received Running
CO Emissions (grams)          6.77                     83.6

After Repair Running 
CO Emissions (grams/mile)     2.77                     10.09

Sample Size                    9                        18

As Received Cold Start
HC Emissions (grams)          3.32                      3.5

After Repair Cold Start
HC Emissions (grams)          2.79                      3.01

As Received Running
HC Emissions (grams)          0.39                      6.60

After Repair Running 
HC Emissions (grams/mile)     0.29                      0.64 

Sample Size                     8                         19


Issues

     This first look has brought up several issues concerning
separating start and running emissions. While the analysis
examined cold starts ( after a 12 hour soak), it may be necessary
to determine the relative change to start emissions for all soak
times.  Also, a more thorough analysis that includes the effects
of repair on both older and newer-model vehicles is necessary.

     A large issue is the lack of "real world" data.  There are
relatively few studies that have the information that is needed
for this type of analysis.  To be considered for this analysis a
study should possess the following criteria.

o    Mechanics would not know they are being evaluated
o    The same vehicles are tested before and after repair
o    A transient test is performed (results in grams or grams per
     mile)
o    Results would allow evaluation of effects for start as well
     as running emissions
o    Vehicle selection is randomized
o    All (almost all) vehicles that fail initially, pass after
     repairs

Continuing Analysis

      EPA has only completed a preliminary analysis concerning
the concept of separating start and running emissions.  More in
depth analysis is necessary.  EPA will examine other test types,
such as the idle test, and other cut-point combinations.  Also,
EPA will look at the statistical significance of separating these
emissions.

Comments

Comments on this proposal should be sent to the attention of
Janet Kremer.  Comments may be submitted electronically to
mobile@epamail.epa.gov, or by fax to (313)741-7939, or by mail to
"MOBILE Review Comments", US EPA Assessment and Modeling
Division, 2565 Plymouth Road, Ann Arbor MI 48105.  Electronic
submission of comments is preferred.  In your comments, pleas
note clearly the document that you are commenting on including
the title and the code number listed, when applicable.  Please be
sure to include your name, address, affiliation, and any other
pertinent information.