United States Air and Radiation EPA420-R-01-029
Environmental Protection April 2001
Agency M6.STE.004
vvEPA Exhaust Emission
Temperature Correction
Factors for MOBILE6:
Adjustments for Engine
Start and Running
LA4 Emissions for
Gasoline Vehicles
> Printed on Recycled Paper
-------
EPA420-R-01-029
April 2001
for
for
for
M6.STE.004
Edward L. Glover
David J. Brzezinski
Assessment and Standards Division
Office of Transportation and Air Quality
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.
-------
1.0 OVERVIEW
In the MOBILE model series vehicle exhaust emissions have always been adjusted for the
effects of ambient temperature. In previous versions of MOBILE these correction factors were
determined separately for each segment (bag) of the Federal Test Procedure (FTP), and were
applied in the model using the user supplied operating mode temperature correction factor inputs.
The new MOBILE6 model will estimate exhaust emissions separately for engine running
emissions and the effects of engine start. The operating mode temperature correction factor
inputs will be removed, and replaced by a count of engine starts per day.
For MOBILE6. the existing MOBILES temperature correction factors (one for each FTP
bag) will still be used. However, they will be applied separately to cold start, hot start, and
running LA4 emissions rather than weighted together into an overall FTP correction using the
operating mode temperature correction factor inputs like in MOBILES. For MOBILE6, the Bagl
temperature correction factor (Equation 1) will be applied to the cold start emissions (a 12 hour
soak prior the start). The Bag3 temperature correction factor (Equation 2) will be applied to the
hot start emissions (a 10 minute soak prior to the start). A VMT weighted combination of Bag2
and Bag3 temperature correction factor (Equation 3) will be applied to the running LA4
emissions. Linear interpolation between the hot start (10 minute soak) and the cold start (12 hour
soak) will be done to obtain temperature correction factors (TCF(x)) for starts with soaks in
between these two values (Soak value of 'x' in Equation 4). The one exception to this general
methodology are the CO start temperature corrections below 75 F. They use an additive
correction factor rather than a multiplicative correction factor.
TCFMOBILE6(COLD engine start) = TCFMOBILE5(Bag 1) Eqn 1
TCFMOBILE6(HOT engine start) = TCFMOBILE5(Bag 3) Eqn 2
TCFMOBILE6(running) = TCFMOBILE5(Bag 2) * 0.521 + TCFMOBILE5(Bag 3) * 0.479 Eqn 3
TCF(x) = TCF(h) + [(TCF(c)-TCF(h)) / (Soak(c)-Soak(h))] * (Soak(x)-Soak(h)) Eqn 4
where:
subscript 'h' means hot engine start and 'c' means cold engine start, and 'x' is the soak length of
interest.
TCF is the temperature correction factor;
Soak is the soak length in minutes. Soak(c) is defined to be 720 minutes (the FTP cold start soak
duration), and Soak(h) is defined to be 10 minutes (the FTP hot start soak duration). The values
of 0.521 and 0.479 are the VMT (vehicle miles traveled) weighting factors for the running LA4.
M6.STE.004 April 2001
-------
The two FTP bags (bags 2 and 3) are weighted together by these factors to produce the running
LA4.
The temperature correction factors in Equations 1 through 4 are not perfectly associated
with the basic emission rates to which they will be applied. This is because the individual Bagl
and Bag3 correction factors used in MOBILES contain both running emissions and start
emissions, and these components cannot be separated. The inconsistency arises because in
MOBILE6, both the cold and hot start emissions have been separated from the running
emissions, and the start estimates contain only the effect of engine starts on exhaust emissions
(See EPA document M6.STE.002 - "The Determination of Hot Running Emissions from FTP
Bag Emissions"). Similarly, the running LA4 estimates in MOBILE6 contain only running
emissions and do not contain start emissions. Thus, utilizing the MOBILES temperature
correction factors directly in MOBILE6 will add a small component of running emissions to the
start estimates and a small component of start emission to the running LA4 estimate.
Because of a lack of data, we cannot develop new correction factors based directly on
start and running LA4 emissions, nor estimate the magnitude of the error introduced by using the
individual bag correction factors from MOBILES. However, the error in applying the MOBILES
Bag 1 temperature correction factors to engine start emissions in MOBILE6 should be small.
This is because the temperature corrections for starts are based on bags. These include some
stabilized running emissions, but these stabilized running emissions are less sensitive to ambient
temperature, and should not strongly bias the results.
Similarly, VMT weighting the Bag 3 with Bag 2 temperature correction factors from
MOBILES to determine the hot running emission temperature correction factors for MOBILE6
assumes that the affect on the temperature correction factor from emissions from the engine start
following a 10-minute soak are negligible. The reason for combining Bag 3 and Bag 2 is to
make the underlying driving cycle identical to the cycle used for the basic hot running emission
estimates in MOBILE6 (the running LA4). Both Bag 1 and Bag 3 use the same driving cycle and
contain engine starts, so the difference in their temperature correction factors should be only a
function of their soak time.
Another possible source of error in applying the MOBILES temperature correction factors
to MOBILE6 start and running emissions is a possible dependence of vehicle model year on the
correction factors. The MOBILES correction factors are based mostly on testing from 1980's
model year vehicles ,whereas the model now predicts emission effects of 1990's and later model
year vehicles. Some error could occur in modeling 1990's vehicles in MOBILES if the
temperature dependence on emissions has changed with newer model year vehicles and
technology. For example, the emission control system on newer model year vehicles may now
reach full operating temperature sooner than vehicles of the past. Unfortunately, due to a lack of
data, the correction factors cannot be updated to fully reflect 1990's technology, nor any
subsequent changes due to advancing technology.
M6.STE.004 April 2001
-------
Since the existing MOBILES temperature correction estimates are being used in
MOBILE6, they will not be reproduced in their entirety in this document. The interested reader
is referred to Pages H-23 and H-24 in Appendix H of "Compilation of Air Pollution Emission
Factors, Volume II: Mobile Sources" (AP-42, June 1995) for a complete listing of the correction
factor coefficients.
2.0 TEMPERATURE AND RVP RANGE
In MOBILES, exhaust emissions for vehicles of model year later than 1979 are adjusted
for the combined effects of fuel volatility (as measured by the Reid Vapor Pressure (RVP)), and
temperature for temperatures between 45 ° and 95 °F. Within this temperature range, there are
two regions of interest— above and below the reference temperature of 75° F. All Temperatures
using the symbol "T" are in degrees Fahrenheit.
2.1 Corrections Above 75 °F
The form of the temperature correction factor for temperatures above 75 °F is shown in
Equation 5a for pre-1980 model years and in Equation 5b for 1980 and later model years. It is
used in MOBILE6 as a multiplicative factor. An example of the coefficients used in Equation 5a
and Equation 5b for light-duty gasoline vehicles are shown in Appendix A-2. The coefficients
labeled Test Segment 1 are used for the cold start, the coefficients labeled Test Segment 3 are
used for the hot start, and a weighted percentage of Test Segment 2 and Test Segment 3
coefficients are used for the hot running LA4 corrections. The weighting for Test Segment 2 and
Test Segment 3 is shown in Equation 3.
TCF(b) = EXP[ TC(b)*(T-75)] Eqn 5a
TRCF(b) = EXP[ RC(b)*(RVP-9.0) + TC(b)*(T-75) + TRC(b)*(RVP-9.0)*(T-75)] Eqn 5b
Where b = 1: Cold Start (Bag 1)
2: Hot Stabilized (Bag2)
3: Hot Start (Bag3)
2.1.1 Adjustment Between 75°F and 95°F
At temperatures above 75 °F, the temperature correction factor is a combined temperature
and fuel volatility factor. This factor is a function of both temperature in degrees Fahrenheit and
fuel RVP in units of psi. Graphically, it can be thought of as a family of curves, each curve
representing the impact of emissions versus temperature for a given fuel volatility. The emission
rate corresponding to 9.0 psi fuel RVP and 75 °F temperature is the base rate. It is the lowest
M6.STE.004 April 2001
-------
emission rate allowed for temperatures above 75 °F regardless of RVP. This means that the
combined affect of temperature and RVP for fuels which have RVP values less than 9.0 are the
same as the affect for a 9.0 psi RVP fuel at 75 °F. The effect only increases when the combined
affect of temperature and RVP for that fuel exceed the effect at 9.0 psi RVP and 75 °F. In other
words, the correction factor has a minimum value equal to the 9.0 psi and 75 °F case. This is the
same methodology as used in MOBILES.
2.1.2 Adjustments Above 95°F
Emissions at temperatures between 95 °F and 110°F are adjusted only as a function of
temperature. The terms in the equation related to fuel volatility (RVP) are set to their value at
95 °F and only the temperature effects are allowed to increase beyond 95°F. In some model years
the correction factors are further differentiated by fuel delivery system (carbureted versus fuel
injection). Equation 5b is used, and the RVP terms are removed.
2.2 Corrections Below 75 °F
2.2.1 HC and NOx Correction Factors
Below 75 °F there is only a temperature correction factor. There is no RVP component to
the overall temperature/fuel volatility correction factor. The form of the temperature correction
factor below 75 °F is shown in Equation 6. This equation applies to HC and NOx for cold start,
hot start and running LA4 emissions. It also applies for running LA4 CO emissions (the
coefficients for Equation 6 are shown in Appendix A-2), and for CO start emissions of pre-1980
model year vintage. The analogous weighting between cold and hot start like in Equation 4 is
also applied to these correction factors to make them a function of soak time.
TCF(b) = EXP[ TC(b) * (T-75) ] Eqn 6
Where b = 1: Cold Start (Bag 1)
2: Hot Stabilized (Bag2)
3: Hot Start (Bag3)
2.2.2 CO Start Correction Factors
The methodology for Start CO on 1980 and later model year vehicles is slightly different
than for the other pollutants. For Start CO the correction factor is additive, and there is a
temperature correction factor coefficient only for cold start CO (soak time = 720 minutes). The
correction factor will increase the CO emissions as temperature (T in Equation 7) is lowered. It
has the mathematical form shown in Equation 7.
TCF(l) = Coeff * (T - 75.0) Eqn 7
M6.STE.004 April 2001
-------
The values of the coefficients (Coeff) are shown in Table 1 below by model year for the
1980 - 1982 model years, and by technology type for the 1983 and later model years. They are
the values of the Start CO temperature correction factor coefficients at a soak length of 720
minutes in units of g/mi-°F. MOBILE will also correct Start CO emission for soak times other
than 720 minutes. The methodology for this is discussed in Section 3.4 because of the
relationship between the additive Start CO temperature correction factor and additive Cold CO
emission standard effects.
Table 1
Temperature Correction Factor Equation Coefficients for Start CO
Model Year Group
1980
1981
1982
1983 +
ALL or CARB
(g/mi-°F)
-1.24480
-1.30945
-1.28402
-1.22620*
TBI**
(g/mi-°F)
-1.22620
PFI**
(g/mi-°F)
-0.65727
* For Carbureted Vehicles Only
** TBI is throttle body fuel injection type
** PFI is ported fuel injection type
3.0 Effects of the Cold Temperature CO Certification Standard(s)
As part of the Clean Air Act requirements, EPA developed a new cold temperature CO
certification standard and a 20 °F test. The new requirements have two phases. Under the first
phase, passenger cars were built to pass a 10 g/mi cold CO standard, and light-duty trucks were
built to pass a 12 g/mi cold temperature CO standard as well as the traditional FTP CO standard
at 75 °F. The requirements for Phase 1 of the cold temperature CO standard were phased into
the fleet starting in 1994 and ending in 1996. For passenger cars and light trucks the phase-in
requirement was 40 percent of the fleet, 80 percent of the fleet and 100 percent of the fleet were
required to pass the cold temperature standard in 1994, 1995 and 1996 respectively.
Because of the unknown nature of the Phase 2 standard, the parameters will not be 'hard-
coded' into MOBILE6. Instead, the default case will be no implementation. Optional user inputs
will be coded in case a new cold temperature CO standard is implemented. These inputs will
include the implementation year, the phase-in schedule (three percentages), and the car and truck
standards in grams per mile.
M6.STE.004
April 2001
-------
3.1 Data
An additive temperature correction factor for cold start CO will be used in MOBILE6 to
model the effects of the cold CO certification standard. This assumes that only engine start
emissions will be affected by the cold temperature CO standards. Table 1 shows a summary of
the 1997 cold temperature CO certification data used to develop cold CO offsets for vehicles
certified for Phase 1 cold CO standards (10 g/mi certification standard at 20 °F ). The
summarized certification data were provided by the American Automobile Manufacturers
Association (AAMA).
Table 2
1997 Certification Data Summary
Manufacturer
All
Big Six1
Vehicle Type
Car
LOT
Car
LOT
FTP CO @ 75 °F (g/mi)
4K Miles
0.95
1.32
0.98
1.36
50K Miles
1.15
1.54
1.23
1.55
FTP CO @ 20 °F (g/mi)
4K Miles
4.68
4.81
4.80
4.60
50K Miles
5.51
5.76
5.86
5.48
1 Sales-weighted average of GM, Ford, Chrysler, Honda, Nissan and Toyota.
CO Offsets
CO Offsets for cars @ 4K
CO Offsets for trucks @ 4K
4.8-0.98 =
4.6-1.36 =
3.82 g/mi
3.24 g/mi
The results from the Big Six Dataset were used. The 4.8 g/mi number for cars is the 20 °F
FTP emission result and the 0.98 g/mi number is the 75 °F FTP emission result.
3.2 Use in MOBILE6
The Phase 1 CO offset (PI CO Offset) value in MOBILE6 is the CO offset for Bagl or
cold start emissions. It is used for all temperatures below 75°F, and is in units of g/mi-°F. It is
a negative number because it reflects the lower CO emissions due to the new CO Cold Start
certification standards. It is used in MOBILE6 to reduce the cold start CO emissions calculated
in Equation 7. Equation 8 shows how the PI CO Offset is calculated from the FTP difference
determined in the previous section. The value of 0.206 is the standard Bag 1 cold start FTP
M6.STE.004
April 2001
-------
vehicle miles traveled (VMT) weighting. The value of-3.82 g/mi is based on the entire FTP.
The Bagl VMT weighting of 0.206 is used to transform the full FTP emissions into equivalent
Bagl emissions. The temperature Delta is 55 °F. This is the difference between the standard
FTP temperature condition of 75 °F and the cold CO certification temperature of 20°F.
Estimate of PI CO Offset values:
P1 CO Offset = -CO Offset FTP / (Bagl FTP Weighting * Temp Delta) Eqn 8
For example, the offset for cars certified to the Phase 1 Cold CO Standard would be:
PI CO Offset = -3.82 g/mi / (0.206 * 55°F) = -0.33709 g/mi -°F
and
PI CO Offset for LDTs certified to Phase 1 cold CO = -0.28600 g/mi -°F
Since cold start emissions in MOBILE6 are in grams per start rather than grams per mile,
the value of-0.33709 g/mi -°F can be converted to g/°F by multiplying the value by 3.59 miles.
This is the distance of Bagl (cold start bag) of the FTP. This produces values of PI CO Offset in
grams/ start-°F of:
PICO Offset for cars certified to Phase 1 cold CO = -1.210 g/start - °F
P1 CO Offset for LDTs certified to Phase 1 cold CO = -1.027 g/start - °F
The cold start CO offsets are shown on a per degree basis with the baseline temperature
of 75 °F, and soak time length of 720 minutes (full cold start). An evaluation showing an
example calculation of the CO cold start correction factor at a soak time of 720 minutes and
application of the phase 1 offset at 60 °F is as follows:
TCF(T) = Coeff * (T - 75 °F) Eqn 9
Then, from Table 1, the TC(1) coefficient for 1992+ model years is -0.7739. Thus:
TCF(60) = -0.65727*(60-75) = 9.859 g/mi CO
Converting to grams per start by multiplying by 3.59 miles produces 35.39 g/start CO.
The CO offset as the result of the cold start CO rulemaking is: -1.210 g/start -°F
Multiplying this by the change in temperature from 75°F to 60°F (delta is 15°F) gives -18.15
g/start.
M6.STE.004 April 2001
-------
The final TCP at 60 °F for a soak time of 720 minutes (full cold start) is the sum of the cold start
temperature correction factor and the CO offset from the rulemaking. It is shown by:
Final TCF(T) = TCF(T)*3.59 - PI CO Offset*(T - 75) Eqn 10
or
Final TCF(Temp=60, soak=720) = 35.39 - 18.15 = 17.24 g/start
This value is added to the base CO emission factor at 75 °F to correct the CO emissions for the
lower temperature of 60°F.
3.3 The Effects of Potential Phase 2 Cold CO Standards
In the current MOBILESb, Phase 2 cold CO standards are assumed to affect only the CO
offset for Bag 1. For vehicles certified to Phase 2 standards, the CO offset for Phase 1 vehicles is
reduced by the difference between pre-Phase 1 cold CO emissions at 20° F and an estimated
Phase 2 CO level assuming a certification margin of about 20%. The MOBILESb approach has
two problems: (1) the CO offset for a low proposed Phase 2 standard (i.e. 3.4 g/mi) could be less
than zero, and (2) the CO offset for vehicles certified to the Phase 1 standard was higher than the
certification data now shows.
The cold CO offset for vehicles certified to the Phase 1 cold CO standard was discussed
in the previous section. The proposed method for Phase 2 also assumes that only Bag 1 CO is
affected by a potential Phase 2 standard. However, in this new method, the Phase 1 CO offset is
reduced in proportion to the standard, so that it is equivalent to the Phase 1 CO offset at a Phase
2 standard of 10 g/mi (equal to the Phase 1 standard for cars) and is zero at a Phase 2 standard of
3.4 grams/mi (the standard for cars at 75°F) by using Equation 11.
Phase 2 CO Offset = Phase 1 CO Offset * [l-((10-x)/(10-3.4))] Eqn 11
Where :
10 = Phase 1 CO standard (g/mi)
x = User Input Phase 2 CO Standard (g/mi)
3.4 = Minimum Phase 2 CO standard (g/mi)
In MOBILE6 the Phase 2 CO Offset will be applied as a multiplicative factor to the Phase
1 CO Offset.
The Bag 1 CO offset is linearly interpolated at all points in between 10 and 3.4 g/mi. In
this manner the CO offset cannot be negative, unless the Phase 2 CO standard is less than 3.4
g/mi (an unlikely scenario). Proposed Phase 2 standards of less than 3.4 g/mi will not be allowed
M6.STE.004 April 2001
-------
in MOBILE6. The temperature sensitivities of Bags 2 and 3 are assumed to be unaffected by the
Phase 2 standard (same assumption as for Phase 1).
3.4 Start CO Effects Versus Soak Time Length
The Start CO temperature correction factor is also a function of soak time which can
range from 0 minutes (an immediate restart after a fully warmed up engine is shut down) to 720
minutes. The relationship between soak time and the Start CO temperature correction factor was
developed from the value at 720 minutes, and the assumption that the start CO temperature
correction factor at a soak time of zero minutes is 0.00. This means that there is no temperature
correction for warmed up vehicles which were immediately restarted after being shut off. Also,
there is no PI CO Offset or P2 CO Offset at soak times of zero minutes. These are also assumed
to be 0.00, since if there is no effect, there could not be an offset to that effect. These are
reasonable assumptions also since ambient temperature is not likely to be a major factor on a
warmed up vehicle which has not had any time to soak and cool down.
For example for the 1983+ PFI vehicles the Start CO emissions at 500 minutes are:
Final TCF(Temp=60, soak=720) = 35.39-18.15 = 17.24 g/start
Final TCF(Temp=60, soak=500) = (17.24 / 720 minutes)*500 = 11.972 g/start
4.0 The Temperature Correction Factor in MOBILE6
The application of the temperature correction factor in MOBILE6 is slightly different
depending on whether it is a multiplicative correction factor or an additive correction factor.
Also, the running emission temperature correction factors are handled slightly different than the
start temperature correction factors.
The multiplicative correction factors are of the form in the MOBILE6 function BEF:
BEF = BEF * TEMPCOR * FUEL_CF Eqn 12
Where BEF is the basic emission factor, TEMPCOR is the multiplicative temperature
correction factor and FUEL_CF is the fuel type correction factor.
Subsequent to this subroutine multiplicative correction factors are applied to BEF. For
running emissions these correct for speed, A/C, Load, Humidity and RVP. For HC and NOx
start emissions, these correct only for humidity and RVP.
The additive correction factors are of the form in the MOBILE6 function BEF:
M6.STE.004 April 2001
-------
BEF = (BEF + TEMPCOR) * FUEL_CF Eqn 13
Where TEMPCOR is the additive temperature correction factor. Since the additive
correction factor is only for start CO emission, these are corrected subsequently only for
humidity and RVP.
M6.STE.004 April 2001
-------
APPENDIX A-l
LOW (< 75F) TEMPERATURE CORRECTION FACTOR COEFFICIENTS FOR
LIGHT DUTY GASOLINE POWERED VEHICLES
* TCF(l) = TC(1)*(T - 75.0), 1980+ CO,
TCF(b) = EXP [ TC(b)*(T - 75.0) ], all others
Pol Model Years Test Segment 1 Test Segment 2 Test Segment 3
HC
Pre-1968
1968-1969
1970-1971
1972-1974
1975-1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992 +
-0
20623E-01
0.24462E-01
0.21255E-01
0.21427E-01
0.23517E-01
0.26820E-01
0.32775E-01
0.32082E-01
0.36491E-01
0.35513E-01
0.32437E-01
0.30471E-01
0.30325E-01
0.27959E-01
0.26867E-01
0.24273E-01
0.23768E-01
-0.23768E-01
-0.24032E-02
-0.32017E-02
-0.52755E-03
-0.39442E-03
-0.88057E-02
-0.75815E-02
-0.83176E-02
-0.85130E-02
-0.74210E-02
-0.81506E-02
-0.78173E-02
-0.84450E-02
-0.90327E-02
-0.94236E-02
-0.85843E-02
-0.83468E-02
-0.82591E-02
-0.82591E-02
-0.10081E-02
-0.86884E-03
0.93659E-03
0.49731E-02
-0.16222E-02
-0.51660E-02
-0.90264E-02
-0.90264E-02
-0.59700E-02
-0.65977E-02
-0.63349E-02
-0.68826E-02
-0.73839E-02
-0.77326E-02
-0.70257E-02
-0.68413E-02
-0.67700E-02
-0.67700E-02
CO
Pre-1968
1968-1969
1970-1971
1972-1974
1975-1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992 +
-0.13487E-01
-0.21126E-01
-0.20843E-01
-0.19091E-01
-0.24835E-01
-0.12448E+01
-0.13095E+01
-0.12840E+01
-0.11761E+01
-0.11636E+01
-0.10515E+01
-0.10032E+01
-0.10146E+01
-0.94629E+00
-0.88655E+00
-0.79324E+00
-0.77390E+00
-0.77390E+00
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
.15784E-
.15289E-
.59951E-
.42373E-
.88336E-
.12478E-
.14584E-
.14584E-
.13550E-
.14658E-
.14282E-
.15277E-
.16146E-
.16807E-
.15614E-
.15360E-
.15250E-
.15250E-
02
02
02
03
02
01
01
01
01
01
01
01
01
01
01
01
01
01
0.
0.
0.
0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
.11097E-
.15749E-
.18253E-
.57982E-
.11553E-
.74106E-
.11371E-
.11371E-
.90777E-
.90777E-
.90777E-
.90777E-
.90777E-
.90777E-
.90777E-
.90777E-
.90777E-
.90777E-
02
02
02
02
02
02
01
01
02
02
02
02
02
02
02
02
02
02
NOx
Pre-1968
1968-1972
1973-1974
1975-1976
1977-1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992 +
-0.16897E-03
-0.25074E-03
0.38855E-02
-0.45504E-04
-0.76044E-02
-0.19000E-02
-0.45479E-02
-0.47657E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.43258E-02
-0.89245E-02
-0.59791E-02
-0.24156E-02
-0.12575E-02
-0.68045E-02
-0.61656E-02
-0.74823E-02
-0.69890E-02
-0.97539E-02
-0.93986E-02
-0.85213E-02
-0.78839E-02
-0.77871E-02
-0.70534E-02
-0.68079E-02
-0.60641E-02
-0.59229E-02
-0.59229E-02
-0.72580E-02
-0.62690E-02
-0.21188E-02
-0.53153E-03
-0.54198E-02
-0.49643E-02
-0.90882E-02
-0.90882E-02
-0.10132E-01
-0.10036E-01
-0.91794E-02
-0.88096E-02
-0.88966E-02
-0.83745E-02
-0.79177E-02
-0.72042E-02
-0.70563E-02
-0.70563E-02
* WHERE :
TCF(b) = Low temperature correction factor for appropriate pollutant,
ambient temperature (< 75F), and model year, for test segment b,
T = Ambient temperature (Fahrenheit),
TC(b) = Low temperature correction factor coefficient for appropriate
pollutant, reference temperature, and model year, for test segment b.
NOTE : The low temperature correction factor is used in conjunction with
the correction factor given in Table 1.7C.
DATE : JUNE 30, 1995
M6.STE.004
April 2001
-------
APPENDIX A-2
HIGH (> 75F) TEMPERATURE CORRECTION FACTOR COEFFICIENTS
AND FUEL RVP CORRECTION FACTORS FOR
LIGHT DUTY GASOLINE POWERED VEHICLES
* TCF(b) = EXP [ TC(b)*(T - 75.0) ], Pre-1980
TRCF(b) = EXP [ RC(b)*(RVP - 9.0) + TC (b) * (T - 75.0)
+ TRC(b)*(RVP - 9.0)*(T - 75.0) ], 1980 +
Pol Model Years Parameter Test Segment 1 Test Segment 2 Test Segment 3
HC
CO
Pre
1968
1970
1972
1975
1980
-1968
-1969
-1971
-1974
-1979
-1982
1983 +
Pre
1968
1970
1972
1975
1980
-1968
-1969
-1971
-1974
-1979
-1982
1983 +
TC
RC
TC
TRC
RC
TC
TRC
TC
RC
TC
TRC
RC
TC
TRC
-0.
-0.
-0.
-0.
-0.
0.
0.
0.
0.
0.
0.
-0.
-0.
-0.
-0.
-0.
0.
0.
0.
0.
0.
0.
.14381E-
.12552E-
.10888E-
.66107E-
.14095E-
.91402E-
.44270E-
.29466E-
.23202E-
.OOOOOE+
.OOOOOE+
.14691E-
.38767E-
.21165E-
.13146E-
.19612E-
.91345E-
.62182E-
.OOOOOE+
.40748E-
.35170E-
.OOOOOE+
01
01
01
02
01
01
02
02
01
00
00
01
01
01
01
01
01
02
00
01
02
00
0.
0.
-0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
.13219E-
.42667E-
.47925E-
.26288E-
.26179E-
.42060E-
.48358E-
.OOOOOE+
02
02
03
02
01
01
02
00
.15373E+00
.86550E-
.OOOOOE+
.37462E-
.84685E-
.23603E-
.24717E-
.48537E-
.13968E+
.14943E-
.OOOOOE+
02
00
02
02
01
01
01
00
01
00
.26214E+00
.14966E-
.56416E-
01
02
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
.34799E-
.75843E-
.76666E-
.12320E-
.24297E-
.93179E-
.74688E-
.47276E-
02
02
02
01
01
01
02
02
.13263E+00
.83730E-
.56009E-
.11014E-
.25179E-
.28483E-
.25848E-
.31439E-
02
02
01
01
01
01
01
.16322E+00
.14923E-
.OOOOOE+
01
00
.23218E+00
.20695E-
.82344E-
01
02
NOx
Pre-1968
1968-1972
1973-1974
1975-1976
1977-1979
1980-1982
1983 +
TC
RC
TC
TRC
RC
TC
TRC
-0
0.38841E-02
0.10389E-02
0.18301E-01
0.71420E-02
26153E-01
O.OOOOOE+00
O.OOOOOE+00
O.OOOOOE+00
0.14219E-01
O.OOOOOE+00
O.OOOOOE+00
-0.87325E-02
-0.92466E-02
-0.10925E-01
-0.87910E-02
-0.18603E-01
-0.40024E-01
O.OOOOOE+00
O.OOOOOE+00
0.27491E-01
0.37789E-02
O.OOOOOE+00
-0.10839E-01
-0.10108E-01
-0.18042E-01
-0.75470E-02
-0.20878E-01
O.OOOOOE+00
O.OOOOOE+00
O.OOOOOE+00
O.OOOOOE+00
O.OOOOOE+00
O.OOOOOE+00
WHERE :
TCF(b)
= High temperature correction factor for appropriate pollutant,
ambient temperature, and model year, for test segment b,
T = Ambient temperature (Fahrenheit),
TC(b) = High temperature correction factor coefficient for appropriate
pollutant, temperature, and model year, for test segment b,
TRCF(b)= High temperature and fuel RVP correction factor
for appropriate pollutant, ambient temperature, fuel RVP,
and model year, for test segment b,
RC(b) = Fuel RVP correction factor coefficient for appropriate
pollutant, fuel RVP, and model year, for test segment b,
RVP = Fuel volatility in psi,
TRC(b) = Combined temperature and fuel RVP correction factor coefficient
for appropriate pollutant, fuel RVP, ambient temperature,
and model year, for test segment b.
NOTE : The temperature correction factor is used in conjunction with
the correction factor given in Table 1.7C in APP42.
M6.STE.004
April 2001
-------
Appendix
Response to Stakeholder Comments
AAMA Comments:
The American Automobile Manufacturers Association had the following comment
regarding temperature corrections:
1. Regarding the application of various correction factors to the start emissions, AAMA is
unsure whether the use of Bag 1 factors is appropriate. The Bag 1 factors reflect a
significant amount of warm running operation. It is unlikely, therefore, that the Bag 1
temperature correction factors at 20 degrees Fahrenheit would be appropriate for use on
100 seconds of cold start emissions. However, AAMA also recognizes that there is little
other data on which to base these factors. Therefore, AAMA recommends a longer term
testing program in which these correction factors (for temperature and fuels) for start
emissions can be developed. Such a testing program should probably focus only on Tier I
andLEV-type vehicles (to assist in making accurate future projections), and the ten
model years prior to 1996.
Certainly data specific to the operating condition we are modeling (i.e., engine start
emissions) will be preferable to Bag 1 correction factors. However, as AAMA noted, this will
require new testing, which will not be available in time for use in MOBILE6.
2. We think more attention should be given to updating the model in the area of cold
temperature CO emissions. In our meeting with you on this subject on November 14th, we
shared a number of specific concerns, and you indicated that there was possibility that a
separate CO model could be issued prior to MOB1LE6 if it is determined that this was
necessary. We fully support that concept and will continue to provide you the analyses
you requested.
A special version of the MOBILES model was produced to address the AAMA concerns.
These changes are included into the MOBILE6 model as well.
3. EPA is proposing to use the MOBILES Bag 1 temperature correction factors to adjust
start emissions for soaks 12 hours or longer; to use the Bag 3 temperature correction
factors for start emissions for soak periods 10 minutes and shorter; and to linearly
interpolate between these temperature correction factors for the intermediate soak
periods. EPA is further proposing to use the MOBILES Bag 2 temperature correction
factors for running emissions.
EPA 's temperature correction factors are based on FTP testing using a minimum 12-
hour soak. Examination of the temperature correction factor database shows that the
latest model year vehicles tested are 1987, which are now ten years old. AAMA realizes
that EPA may have to use these factors for MOBILE6, because of the lack of temperature
test data. However, the Bag 1 temperature correction factors may not be appropriate for
cold start emissions, because Bag 1 includes a certain amount of "stabilized" operation.
EPA 's use of Bag 3 temperature correction factors is probably appropriate for very short
M6.STE.004 April 2001
-------
soak periods (i.e., less than 15 minutes long). Finally, EPA 's use of the Bag 2
temperature correction factors for running emissions is probably appropriate.
No new data addressing the effect of temperature on exhaust emissions is available for
the development of MOBILE6. EPA believes that the existence of some stabilized vehicle
operation in Bag 1 emissions will have only a small effect on the overall estimate of the effect of
temperature on engine start emissions.
4. AAMA 's and EPA 's analysis of emissions data at normal summer temperatures (i.e., 75
degrees Fahrenheit) shows that later model Tier 0 vehicle (i.e., 1990-1993) generally
have much lower emissions than cars and light duty trucks built in the 1980s. These
vehicles probably also have less emissions sensitivity to variations in temperature. AAMA
believes that EPA 's approach of separating cold start and running emissions creates a
significant need for further temperature testing, both on later model year cars and light
duty trucks, and with a different test procedure. New testing should use a test procedure
that incorporates a cold start and no start, similar to EPA 's test program which it
conducted to determine emissions from a hot running 505 (HR505). In addition, there is a
need for data at different temperatures at intermediate soak periods, for example, 30
minutes, 1-hour, 3 hours, etc.
EPA is now planning a new generation of emission models which will address these
concerns, including the collection of data on later model year vehicles.
5. On November 14, 1997, AAMA members met with EPA OMS and AIR, Inc., to discuss
changes in the MOBILESb model that would more accurately reflect vehicle CO
emissions at cold temperatures. AAMA 's specific comments are contained in the
November 14, 1997 AIR briefing entitled "Impacts of MOBILE6Development on CO
Emissions at Cold Temperatures", which was provided to EPA at that meeting. AAMA
was pleased with the outcome of the meeting in that it resulted in an agreement to work
together to determine model modifications that should be incorporated to best reflect
vehicles designed to meet Phase 1 cold temperature CO standards. AAMA is confident
that our work efforts can continue and that they will result in model modifications that
are technically correct and agreeable to all involved. AAMA would also like to continue
work efforts related to Phase II cold temperature standards, should they be required.
AAMA is hopeful that agreed upon changes related to the Phase I standards can be
quickly incorporated and that states will be allowed to utilize a model having these
revisions in their SIP planning for CO attainment/maintenance. In this regard, we again
request that, once the revisions are completed, EPA make the revised model available for
the states/municipalities to use prior to release of the official MOBILE6 model. This is
critical because the changes result in major reductions in forecasted CO emissions, and
will thus have significant implications on possible CO control strategies and costs.
A special version of the MOBILES model was produced to address the AAMA concerns.
These changes are included into the MOBILE6 model as well.
M6.STE.004 April 2001
-------
AIR Comments:
Air Improvement Resource, Inc., has five areas of comment on the update to the hot soak
emission rates:
1. EPA should design and implement a temperature correction factor test program for Tier
1 and LEV vehicles and implement the results inMOBILE6.
EPA proposes to rely on the MOBILES temperature correction factors for MOBILE6,
when the MOBILES Temperature Correction Factors (TCFs) were based on early Tier 0
vehicles. These TCFs probably are inappropriate for late model Tier 0 vehicles (1998-
1993), Tier 1 vehicles (1994-2000), LEVs, and Tier 2 vehicles.
EPA has been in the process of updating MOBILES for about 3 years. Over the last few
years, EPA has committed resources to test programs to develop basic emissions rates at
FTP temperatures, and speed correction factors, however, little or no resources have
been devoted to updating the temperature correction factors. This is a very serious
problem for allMOBILE6 users, because allMOBILE6 users runMOBILE6 at
nonstandard conditions. For example, all states that prepare inventories, and those that
must submit ozone SIPs, run MOBILES at nonstandard temperature conditions. EPA,
when it produces inventories for the nation on a county-by-county basis for its Emission
Trends Report, also runs the model at nonstandard conditions.
AIR believes that EPA should commence a test program immediately to determine TCFs
for Tier 1 and LEV vehicles, at low and high temperatures.
No new data addressing the effect of temperature on exhaust emissions is yet available for
the development of MOBILE6.
2. EPA should examine the cold CO data on LEVs to determine a separate Cold CO offset
for LEVs.
EPA adopted AIR's suggestions for the cold CO portion of the model. AIR estimated cold
CO offsets for Tier 1 vehicles, which EPA is proposing to use for all Tier 1 and later
vehicles. However, the CO offsets for LEVs and Tier 2 vehicles may be very different. AIR
accessed the 1999 certification data for CO emissions at 20 degrees Fahrenheit and 75
degrees Fahrenheit for Tier Is, TLEVs, LEVs and ULEVs. The results for cars andLDTs
are shown in Table 1 below.
EPA is using a 3.8 g/mi increase for Tier 1 passenger cars for 20 degrees Fahrenheit
relative to 75 degrees Fahrenheit. This was based on the difference in 4,000 mile
certification results. The above table shows a 4.3 g/mi difference based on 50K results for
Tier 1 vehicles in 1997, 3.9 g/mi for 1998 vehicles, and 3.9 for 1999 Tier 1 vehicles.
Thus, the 3.8 g/mi for Tier 1 vehicles which EPA proposes to use still appears to be a
good value. For LEVs, the CO difference between 75 degrees Fahrenheit and 20 degrees
Fahrenheit is 3.0 g/mi in 1998, and 3.2 in 1999. For ULEVs, the increase is about 2.3
g/mi. These data show that the CO offsets for LEVs, ULEVs, and Tier 2 vehicles should
be substantially less than the Tier 1 offsets. AIR recommends using 3.1 g/mi for National
LEV vehicles (NLEVs).
M6.STE.004 April 2001
-------
Table 1
EPA 50,000 mile Certification CO Emission Rates
Vehicle
Class
LDV
LOT
MY
1997
1998
1999
1997
1998
1999
Tier
T1
T1
TLEV
LEV
ULEV
T1
TLEV
LEV
ULEV
T1
LEV
T1
TLEV
LEV
T1
TLEV
ULEV
CO (75F)
N
124
887
279
60
2
894
341
248
16
54
40
212
120
342
730
126
12
g/mi
1.156
1.121
0.916
0.727
0.314
1.285
0.964
0.798
0.311
1.598
1.362
1.533
1.382
1.491
1.897
1.361
0.757
Cold CO (20F)
N
328
209
33
7
None
213
49
48
2
203
7
57
24
65
153
22
1
g/mi
5.467
5.011
5.324
3.779
None
5.211
4.305
4.083
2.6
5.742
5.373
5.366
6.916
5.129
5.593
5.738
1.669
Diff.
g/mi
4.311
3.89
4.407
3.052
None
3.926
3.341
3.285
2.289
4.144
4.011
3.833
5.535
3.638
3.696
4.377
0.912
EPA had not proposed updating the cold temperature CO offset for MOBILE6. The
changes proposed by AIR are very reasonable and will be included in our plans for updates to the
MOBILE6 model after it's initial release.
3. EPA should examine the interaction between the air conditioning correction factor, the
cycle or speed correction factors, and the temperature correction factors, to ensure that
no double-counting of the emission effects at higher temperatures takes place in the
model.
AIR is unsure, and the report is unclear about how EPA intends to handle the interaction
of the high temperature correction factors, the air conditioning correction factor, and the
cycle correction factors. Previous reports from EPA have developed the air conditioning
correction factors at 95 degrees Fahrenheit, and air conditioning activity (M6.ACE. 001
and M6. ACE.002). If these factors have been developed at 95 degrees Fahrenheit, then
perhaps the use of a separate temperature correction factor is not necessary. Also,
depending on the cycle that was used to develop the air conditioning correction factors, a
M6.STE.004
April 2001
-------
cycle correction factor may not be necessary also. EPA should clearly explain how all of
these correction factors are applied at high and low temperatures, and what data sources
they are being estimated from, so that it is clear that EPA is not double-counting the
effects of any given parameter (cycle, temperature, or activity).
Air conditioning effects are only applied to running emissions (not including engine start
emissions). The effect is calculated as an additive emission offset from a basic exhaust emission
rate which has not yet been adjusted (multiplicative) by the temperature correction factor. The
basic emission rate is then adjusted by the temperature correction factor before the air
conditioning effect is added to the basic exhaust emission rate. In this way the temperature
correction is never applied to the emissions represented by the air conditioning offset.
The air conditioning offset itself is calculated by comparing emissions with and without
air conditioning at 95 degrees Fahrenheit. Therefore, the air conditioning offset is primarily the
effect of the air conditioning load on emissions (at 95 degrees), rather than a combination of
temperature and air conditioning effects. These air conditioning effects are, in turn, adjusted to
account for differences in air conditioning load at different temperatures. Air conditioning
effects are discussed in the reports, "Air Conditioning Activity and Intermediate Conditions,"
(M6.ACE.001) and "Air Conditioning: Full Usage Correction Factors," (M6.ACE.002) found on
the EPA web site (http://www.epa.gov/otaq/m6.htm).
M6.STE.004 April 2001
------- |