EPA-AA-EOD/TPB-85-2
Technical Report
Assessment of the Hot Start Fuel Economy Effects of a New
CVS Exhaust Connector Pipe Design
August 1985
Marty Reineman
Douglas DeVries
Testing Programs Branch
Engineering Operations Division
Office of Mobile Sources
Environmental Protection Agency
2565 Plymouth Road
Ann Arbor, Michigan 48105
NOTICE
Technical reports do not necessarily represent final EPA decisions or'
positions. They are 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 improvements
in emissions measurement. Their publication or distribution does not
constitute any endorsement of equipment or instrumentation that may have been
evaluated.
-------
Background
In June of 1985, the EPA-MVEL initiated a program to convert the • EOD,
fixed length CVS exhaust connector pipe to a variable length design, which
is., described in Equipment/Procedure Change Notice No. 64. Prior to the
usage of these new style exhaust connectors, the Facility Support Branch
performed a series of checks on each new .connector. These tests included
a bench static pressure leak check, a propane injection test with each new
connector, and emissions and fuel economy verification tests using EOD's
Volvo REPCA. By September, all EOD CVS's had been converted to the new
design.
Although the REPCA tests permitted a comparison of CVS results obtained
with the new system to results obtained with the old system, the REPCA
data were slow to accumulate due to the heavy test schedule of
Certification and In-use vehicles. Consequently, as the testing load
slackened, it was determined that it would be meaningful to run a series
of A-B type tests comparing the new and old exhaust collection systems.
It was tho_ught that the change to the new style connector pipes might
reduce an apparent carbon balance fuel economy offset between EPA and the
manufacturers, particularily General Motors. These offsets were first
observed early in 1985.
Program Design
The test program consisted of A-B type fuel economy tests with a
repeatable vehicle using CVS's 21C, 22C, 29C, and 25C. Each CVS was used
with two different connector pipes - one set of tests was run with the
CVS' new pipe and one set was run with the old connector pipe from CVS
29C. The other elements of the test design included:
Vehicle - The test vehicle was a GM repeatable vehicle, a 1.8 liter, four
cylinder, TBI Pontiac J-2000 equipped with a fuel flow meter, torque
wheels, and fifth wheel distance pickup.
Driver - The same EPA driver drove a particular test sequence, but three
different drivers were used during the course of the program.
Prep - The same tank of test fuel was' used for all tests except the last
set when it was necessary to add additional fuel. Each test sequence was
started by warming up the vehicle and dynamometer for 15 minutes at 50 mph
before the first test, with additional warm-up for 3 minutes at 50 mph
preceding each LA-4.
Sequence - A daily test sequence consisted of a series of six hot start
LA-4 (two bags) dynamometer tests, while changing the exhaust connector
pipe after each LA-4. Two test sequences started with the old system in
place, and two test sequences started with the new system in place.
Volumetric fuel consumption, wheel torque/horsepower, and distance,
measured in roll-feet, were recorded for each phase of the LA-4 cycles.
-------
-2-
Results
This study; conducted between August 16 and August 21, showed a reduction
in composite hot start 'LA-4 fuel economy in the range of 0.6 to 0.8
percent. An analysis of the hot start fuel economy effects is summarized
in Table 1. Tables A-l through A-4 and Figures A-l through A-6 in the
Appendix present detailed summaries of the bag and composite fuel economy
results.
Table 1 is a summary of the overall average percent differences for phase
1, phase 2, and composite LA-4 results. The average percent difference
between carbon balance and volumetric fuel economy is given for the new
and old CVS exhaust connectors in columns one and two, respectively. In
Table 1, the percentages -0.9, -1.0, and -0.8 are the overall changes in
carbon balance versus volumetric fuel economy for phase 1, phase 2, and
the composite LA-4 results, respectively. The last column summarizes the
•fuel economy effect based on the change in carbon balance fuel economy for
the two types of exhaust connectors.
Because carbon balance and volumetric fuel economy are determined
simultaneously, the overall carbon balance vs. metered differences for
phase 1, phase 2, and composite results from the new and old connector
pipes are determined by averaging the 12 individual test pair differences
in each configuration.
By nature of the experimental design, old versus new comparisons could not
be run simultaneously, and thus they are not paired results. Therefore,
the overall percent differences in carbon balance fuel economy between the
old and new connectors, shown in column four of Table 1, are calculated by
taking the percent difference between the two grand means of carbon
balance fuel economy.
The fuel economy effect is larger using the change in carbon balance
versus metered fuel economy comparisons., although either method provides a
valid means of estimating the fuel economy impact of the change in exhaust
connectors. Thus, a reasonable estimate of the LA-4 fuel economy impact
is in the range of -0.6 to -0.8 percent.
Tables A-l through A-4 in the Appendix present the raw carbon balance and
metered fuel economy data for the individual tests. In these figures, the
numbers 1-6 to left of the phase 1 carbon balance data indicate the order
in which the fuel economy measurements were obtained. Inspection of
Tables A-l through A-4 shows that fuel economy tests using CVS's at D001
and D006 were started with the new connector pipe in place, while tests on
D002 and D005 were begun with the old connector in place.
In Figures A-l through A-3, the carbon balance versus volumetric percent
differences obtained with the old and new sampling systems are presented
-------
-3-
as a function of the actual test sequence. This information is summarized
for composite, phase 1, and phase 2 results, respectively. Figures A-4
through A-6 of the Appendix are plots of the individual carbon balance and
volumetric .fuel economy for the composite, phase 1, and phase 2 results,
respectively, versus the test sequence order.
Discussion
All of the test data were closely examined to verify their validity and to
check for effects other than those which may be attributed to the change
in the sampling system. The following observations are made with respect
to wheel torque and horsepower, exhaust emissions, and fuel economy.
Wheel torque and horsepower:
The same driver drove all six tests once a particular test sequence was
started, but three different drivers were used during the four days of
testing. Nevertheless, the wheel torque and horsepower measurements were
very precise for a given test day, and very repeatable among the four
dynamometers. Positive and negative torque repeated within 2 percent
during phase 1 and 2 dynamometer operation. Integrated horsepower
repeated within 2 percent for all test phases except phase 1 where
negative horsepower repeated within 7 percent. These levels of torque and
horsepower measurement precision are well within the range of good
repeatability for EPA's Clayton dynamometers.
Emissions:
No change in HC, CO, or NOx emissions were observed as the sampling system
was switched between the new and the old configurations. HC emissions
varied about _+ 3 percent around a composite mean of 0.063 g/mi. Composite
CO emissions averaged 1.56 g/mi with a range of +6 percent. Composite NOx
emissions averaged 0.53 g/mi and varied ^+4 percent around this value.
These emission results demonstrate very good test precision.
Fuel Economy:
Figures A-4 and A-5 clearly demonstrate that the fuel economy was lower on
the first test of each six test sequence, despite a 15 minute warm-up at
50 mph. Figure A-5 indicates that this was a phenomenon of the phase 1
test results, but this effect also caused a decrease in the composite fuel
economy (Figure A-4). The increasing fuel economy as a function of time
is likely due to tire and lubricant' warm-up effects, and possibly
influenced by driver familiarity with the vehicle after the first phase of
the first test. Although this warm-up effect confounds the interpretation
of the old versus new sample system results, the experimental design
mitigated its impact. This is true because two of the four dynamometer
test sequences began with the old connector pipe in place, and two test
-------
-4-
sequences were started with the new- connector. Even when the first test
was deleted from each series of dynamometer tests, the effect of the
change to 'the new system was the same - composite fuel economy decreased
0.6 percent. " '
Figures A-4 through A-6 show that metered, and to a lesser degree, carbon
balance fuel economy, increased over time. This effect was small and very
gradual, therefore, it was not expected to affect the findings summarized
in Table 1.
Summary
The overall composite difference of -0.6 to -0.8 percent provides an
estimate of the change in fuel economy attributable to the new exhaust
sampling system. The effect of the sampling system with respect to
regulated emissions was not observable.
The limitations of this program must be understood when extrapolating
these results to FTP and HFET fuel economy results. The limitations are:
1. A single, old type exhaust connector was used for all tests.
2. The real fuel economy effect will be a function of the particular
vehicle (this vehicle may have over- or understated the effect).
3. The hot start effects which were estimated in this program are not
directly relatable to cold start fuel economy.
4. The fuel economy effect of the new connectors will likely change as
the new connectors wear.
Recommendations for Future Actions
1. Continue to monitor the carbon balance and volumetric fuel economy data
from the Volvo REPCA to assess the long term effect of the change in the
exhaust sampling system.
2. Concentrate on examining the CVS sampling system design and calibration
practices as possible explanations for the difference between EPA and GM
carbon balance fuel economy measurements.
0703c
-------
Table 1
Fuel Economy Effects [1]
New vs. Old Exhaust-Connectors, % Difference
New CB-New M [2]- Old CB-Old M Fuel Economy, New CB-Old CB
i Old CB
-0.5
-0.8
-0.6
Phase 1
Phase 2
Composite
New M
(2.1
(3.3
(2.5
Old M
3.0)
4.3)
3.3)
A CB vs
-0.9
-1.0
-0.8
[1] Based on 24 tests on 4. dynamometers; 12 with old connector, 12 with new
connectors.-
[2] CB = Carbon balance fuel economy, mpg.
M = Metered fuel economy, mpg.
-------
APPENDIX
-------
Table A-l
Dyno 1 (CVS 21C) Fuel Economy Results
New vs. Old Exhaust Connectors, mpg
'Old
New
Sequence
Phase 1 2
4
6
Phase 2
Composite
CB
29.4
29.0
29.3
26.7
27.0
26.7
27.9
27.9
27.9
M
28.5
28.4
28.6
. 25.9
26.2
25.9
27.2
27.3
27.3
uence
1
3
5
CB.
28.6
28.9
28.9
26.8
26.7
26.5
27.7
27.8
27.7
M
. 28.3
28.5
28.7
26.3
26.1
26.0
27.3
27.3
27.3
-------
Table A-2
Dyno 2 (CVS 22C) Fuel Economy Results
New vs. Old Exhaust Connectors, mpg
Old
New
Sequence
Phase 1 1
3
5
Phase 2
Composite
CB
28.6
29.4
29.5
26.7
26.6
26.8
27.6
27.9
28.0
M
28.0
28.5
28.4
25.9
25.7
25.8
26.9
27.1
27.1
;uence
2
4
6
CB
29.1
29.1
29.3
26.4
26.4
27.0
27.7
27.6
28.0
M
28.3
28.5
28.6
25.8
25.7
26.1
27.0
27.1
-27.4
-------
Table A-3
Dyno 5 (CVS 29C) Fuel Economy Results
New vs. Old Exhaust Connectors, mpg
•Old
New
Sequence
Phase 1 1
3
5
Phase 2
Composite
CB
29.2
29.3
29.5
27.0
27.3
27.2
28.0
28.2
28.3
M
28.1
28.3
28.3
25.7
25.7
25.7
26.9
27.0
27.0
uence
2
4
6
CB
29.2
29.3
29.4
27.0
27.2
27.0
28.0
28.2
28.1
M
. 28.4
28.4
28.4
25.6
25.8
26.1
27.0
27.1
27.0
-------
Table A-4
Dyno 6 (CVS .25C) Fuel Economy Results
New vs. Old Exhaust Connectors, mpg
'Old
New
Sequence
Phase 1 2
4
6
Phase 2
Composite
CB
28.8
28.5
29.1
26.7
26.8
27.2
27.7
27.6
28.0
M
27.9
28.0
28.3
25.6
25.5
25.8
26.7
26.8
27.1
Sequence
1
3
5
CB
28.7
28.8
28.6
26.2
26.4
26.6
27.3
27.5
27.5
M
27.8
28.3
28.4
25.3
25.6
25.7
26.6
26.9
-27.1
0703c
-------
2
sf
8
W
O,
6
0
0
FIGURE ' A" I
EPA SITE HARDWARE EVALUATION - 1985
11 OLD"vs"NEW - DYNOS 1,2,5 & 6
I T
2 4
DYNO 16
A
"OLD" HARDWARE
"NEW" HARDWARE
"A A
8 10
DYNO #5
I
12
14 16
DYNO 12
I
18
20 22
DYNO #1
24 26
TEST SEQUENCE NO.
-------
Figure 'A~2.
s
sf
w
1 -
•0
EPA SITE HARDWARE EVALUATION - 1985
"OLDllvs"NEW" - DYNOS 1,2,5 & 6
I I
2 4
DYNO «6
"OLD" HARDWARE
"NEW" HARDWARE
8 10
DYNO #5
12
I 1
14 16
DYNO #2
I
A-'
7
18 20 22 24 26
DYNO
TEST SEQUENCE NO.
-------
Figure :A-2>
w
CM
.3-
EPA SITE HARDWARE EVALUATION - 1985
"OLD"vs"NEWn - DYNOS 1,2,5 & 6
i r
2 4
DYNO 16
"OLD" HARDWARE
"NEW" HARDWARE
A
6
8 10
DYNO #5
12
I I
14 16
DYNO t2
18
i I
20 22
DYNO #1
24 26
TEST SEQUENCE NO.
-------
Figure ?A-
a
u
29
28.5
28-
27.5
27-
26.5
26
EPA SITE HARDWARE EVALUATION - 1985
11 OLD"vs"NEW" - DYNOS 1,2,5 & 6
:.--<$>'
= "OLD" CAR.BAL. MPG
= "NEW" CAR.BAL. MPG
A--A-
"OLD" METER MPG
"NEW" METER MPG
1 I
2 4
DYNO *6
I I
8 10
DYNO #5
12
14 16
DYNO n
18
• I I
20 22
DYNO #1
24 26
TEST SEQUENCE NO.
-------
&.
30
29.5
29-
28.5-
28-
27.5
27
0
Figure /A-
EPA SITE HARDWARE EVALUATION - 1985
11 OLD"vs"NEW" - DYNOS 1,2,5 & 6
A
A"
2 4
DYNO #6
I I
8 10
DYNO #5
12
I I
14 16
DYNO #2
= "OLD" CAR.BAL. MPG
= "NEW" CAR.BAL. MPG
"OLD" METER MPG
"NEW" METER MPG
18 20 22 24 26
DYNO #1
TEST SEQUENCE NO.
-------
Figure
O CN
H W
28
27.5
27
26.5~
26~
25.5
25
0
EPA SITE HARDWARE EVALUATION - 1985
11 OLD"vs"NEW" - DYNOS 1,2,5 & 6
I
4
:A
-O
"OLD" CAR.BAL. MPG
"NEW" CAR.BAL. MPG
"OLD" METER MPG
"NEW" METER MPG
T
6
I
12
DYNO #6
8 10 12 14 16
DYNO #5 DYNO #2
TEST SEQUENCE NO.
I Tl I I
18 20 22 24 26
DYNO (HI
------- |