76-13 GS
The Mark II Vapor Injector:
/.n Air-Vapor Bleed Device Evaluated
January 1976
Technology Assessment and Evaluation Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Environmental Protection Agency
-------
Background
The APO Mark II Vapor Injector marketed in the United States by APO
of America, Inc., Dallas, Texas is essentially an induction system air-
vapor bleed device. It is the fifth device of this basic type to be
tested by TAEB in the past five years. ' ' * The general conclusions
of the four previous air-vapor bleed device tests were that fuel economy
improvements if any, were small and were attributed to enleanment of the
air fuel mixture as opposed to the effects of the added vapors. Similarly
exhaust emissions changes were minor and were typical of the results of
enleanment of air-fuel ratios near stoichiometry. The Mark II contains
several variations on the basic vapor bleed device theme which could
alter its performance relative to other devices of this type. In light
of these variations and an interest in the Mark II exhibited by the
public and some sectors of the government, EPA evaluated the device.
The Environmental Protection Agency receives information about many
devices for which emission reduction or fuel economy improvement claims
are made. In some cases, both claims are made for a single device. In
most cases, these devices are being recommended or promoted for retrofit
to existing vehicles although some represent advanced systems for meeting
future standards.
The EPA is interested in evaluating the validity of the claims for
all such devices, because of the obvious benefits to the Nation of
identifying devices that live up to their claims. For that reason the
EPA invites proponents of such devices to provide to the EPA complete
technical data on the device's principle of operation, together with
test data on the device made by independent laboratories. In those
cases in which review by EPA technical staff suggests that the data
submitted hold promise of confirming the claims made for the device,
confirmatory tests of the device are scheduled at the EPA Emissions
Laboratory at Ann Arbor, Michigan. The results of all such confirmatory
test projects are set forth in a series of Technology Assessment and
Evaluation Reports, of which this report is one.
The conclusions drawn from the EPA confirmatory tests are neces-
sarily of limited applicability. A complete evaluation of the effectiveness
of an emission control system in achieving its claimed performance
improvements on the many different types of vehicles that are in actual
use requires a much larger sample of test vehicles than is economically
feasible in the confirmatory test projects conducted by EPA. For
promising devices it is necessary that more extensive test programs be
carried out.
The conclusions from the EPA confirmatory tests can be considered
to be quantitatively valid only for the specific type of vehicle used in
the EPA confirmatory test program. Although it is reasonable to extra-
polate the results from the EPA confirmatory test to other types of
-------
vehicles in a directional or qualitative manner, i.e., to suggest that
similar results are likely to be achieved on other types of vehicles,
tests of the device on such other vehicles would be required to reliably
quantify results on other types of vehicles.
In summary, a device that lives up to its claims in the EPA confirma-
tory test must be further tested according to protocols described in
footnote 5, to quantify its beneficial effects on a broad range of
vehicles. A device which when tested by EPA does not meet the claimed
results would not appear to be a worthwhile candidate for such further
testing from the standpoint of the likelihood of ultimately validating
the claims made. However, a definitive quantitative evaluation of its
effectiveness on a broad range of vehicle types would equally require
further tests in accordance with footnote 5.
System Description
The Mark II Vapor Injector is a device for inducting an air-vapor
mixture into the intake system of the conventional spark ignited gasoline
engine. The point of induction can be the Positive Crankcase Ventilation
(PCV) line, a spacer plate installed between the carburetor and the
intake manifold, or an idle adjustment screw with a hole through the
center. The latter method was chosen for this evaluation because the
previous evaluations used the PCV induction point. The spacer plate is
essentially the same as the PCV induction point.
The Mark II consists of a large glass jar approximately two-thirds
full of a fluid comprised of one part Mark II Econo Mix fluid and two
parts water. Air is drawn in through a brass needle valve mounted on
the cast aluminum jar cap, and passes through a plastic tube to a plastic
bubbler located near the bottom of the jar. After bubbling through the
fluid, the resulting air-vapor mixture is drawn out of the bottle through
a vacuum hose connected to the jar cap. A .022 inch diameter metering
orifice mounted in the vacuum hose restricts the flow of air-vapor
mixture to the Vapor Jet, an idle adjustment screw with a hole drilled
through the center. The air-vapor mixture is drawn through the Vapor
Jet by manifold vacuum. A check valve in the Vapor Jet is intended to
prevent reversal of flow.
Figure 1 is a photograph of the Mark II as installed in the vehicle
used in this evaluation, a 1971 Chevrolet Vega. As the Mark II is
"driven" by manifold vacuum the volume of vapor delivered to the intake
manifold is virtually independent of engine displacement. To maximize
the effect of the device, a vehicle with a small displacement engine was
selected. Table 1 Ls the vehicle description of the Vega. Figure 2 is
a schematic of the Vega carburetor with the Vapor Jet installed.
-------
hose to Vapor Jet
mounting bracket
carburetor
N Mark II reservoir
Figure 1 Mark IT install; tion on 1971 Vega (air cleaner removed).
choke valve
hot idle compensator
lower idle air bleed
throttle valve
top air bleed
idle channel restriction
idle tube
float bowl
main metering jet
off-idle port
to Mark II reservoir
idle discharge hole
Vapor Jet
Figure 2 Schematic of Vega carburetor showing idle circuitry
with Mark II Vapor jet installed,
-------
4
Table 1
TEST VEHICLE DESCRIPTION
Chassis model year/make - 1971 Chevrolet Vega Kammback
Emission control system - PCV
Engine
type 4 cyl. OHC
bore x stroke ....... . 3-50 x 3.625 in./88.9 x 92.1 mm
displacement .'.!!!".!!!'.! 1*0 CID/2300 cc
compression ratio 8.0:1
maximum power @ rpm ....... 90 hp/67 kW @ 4400 rpm
fuel metering l barrel
fuel requirement 91 RON
Drive Train
transmission type 3 speed manual
final drive ratio 2.53
Chassis
type front engine, rear drive, unitized body
tire size A .78 x 13
curb weight 2340
inertia weight 275°
passenger capacity *
Emission Control System
basic type PCV
mileage at beginning of test ~
program 11,000 miles
-------
According to Mr. Allen Best, the technical advisor of APO, the
composition of the Econo Mix by volume is 65% methanol, 34% acetone, and
1% propylene glycol. The benefits of the vapors of Econo Mix-water
mixture claimed in the Mark II owner's manual are a decrease in required
octane number of the gasoline used, increased fuel economy, increased
power, elimination of carbon deposits, extension of engine life, and
reduction of exhaust emissions. EPA evaluates devices in terms of their
effects on vehicle emissions, fuel economy, and occasionally performance.
Additions of methanol and water to gasoline are known to increase the
octane number of fuel and additions of methanol under certain conditions
to increase power. Therefore, it was felt desirable to conduct a pre-
liminary evaluation of the possible benefits of the APO device based on
information available in the technical literature concerning the various
constituents of the APO fluid and measurements of the operating variables
of the device. This evaluation is presented in Appendix I.
Test Procedure
Exhaust emissions tests were conducted according to the 1975 Federal
Test Procedure ('75 FTP), described in the Federal Register of November 15,
1972, and the EPA Highway Fuel Economy Test (HFET), described in the
Federal Register, Volume 39, Number 200, October 15, 1974. Both of
these tests are conducted on a chassis dynamometer and employ the Constant
Volume Sampling (CVS) procedure, which gives exhaust emissions of HC,
CO, NO and CO in grams per mile. Fuel economy is calculated by the
carbon balance method. The fuel used was Indolene unleaded 96 RON
gasoline.
The vehicle was tested in three different configurations: baseline,
with the Mark II installed but without any fluid in the jar, and with
the Mark II functioning with fluid. The second configuration was tested
in order to separate the effect of the fluid vapors from the effect of
the air bleed.
Before the baseline testing, the vehicle was tuned to manufacturer's
specifications. The carburetor idle mixture adjustment was adjusted to
lean best idle, which for this vehicle resulted in a 0.2% idle CO. The
idle mixture was adjusted to lean best idle after the installation of
the Mark II in each configuration tested. In both cases the idle CO was
again 0.2%.
The test schedule plan was two '75 FTP's and two HFET's for each of
the following test conditions:
1. Baseline
2. Mark II installed but without fluid
3. Mark II with fluid
4. Mark II with fluid after 1500 miles of operation on the
durability driving schedule described in the Federal Register
Vol. 37, No. 221, November 15, 1972
5. Mark II without fluid after mileage accumulation
6. Baseline after mileage accumulation
-------
Due to difficulties encountered in the test program additional
tests were conducted as discussed in the following section.
Test Results
The test schedule was initially conducted according to plan with
some additional tests being conducted to increase confidence in the
data. After the 1500 miles, the idle specifications were again set to
manufacturer's specification. The two tests following this idle tune
had '75 FTP and HFET fuel economy decreases of 8 and 16 percent respectively
from the baseline data. Inspection of the vehicle revealed the distributor
vacuum line disconnected at the carburetor. The exact time when the
vacuum hose was disconnected is not known but it is our belief that the
hose was not reconnected during the idle timing check after the mileage
accumulation. The results of the above two tests were discarded, the
hose was reconnected, and the tests were repeated. Subsequent tests
displayed a steady increase in hydrocarbon emissions apparently independent
of the test configuration. Concurrent with the hydrocarbon increases
was a smaller but still discernible decrease in fuel economy.
Thorough diagnostics revealed low compression in number two cylinder
(120 psi vs. 180 psi for the other three cylinders). The head was
removed and the exhaust valve of the number two cylinder was observed to
be mildly burned due to valve seat warpage. In retrospect it is possible
that operating the vehicle with the vacuum leak and no vacuum advance
contributed to the valve seat warpage. This valve was replaced and the
valve seat ground; great care was taken not to disturb the deposits on
the head. After the head was replaced, the car was driven approximately
150 miles with the Mark II functioning. A second series of "after
mileage accumulation" tests were run. These are identified alternatively
as "after valve replacement" or "after 2,000 miles," as opposed to the
original "after 1,500 miles."
The composite results of the '75 FTPs are presented in Table 2;
detailed individual bag results are given in Appendix II. The HFET
results are given in Table 3. Figures 3 and 4 display the fuel economy
values of the '75 FTPs and HFET's respectively. Figure 5 displays the
hydrocarbon, carbon monoxide and nitrogen oxides emissions of the '75
FTPs. Three tests (16-1280, 76-1310 and 76-2455) were not included in
these Figures because they were determingd to be statistically invalid
using the Dixon method on the C02 data.
The representatives of the Mark II claim that the device cleans the
engine and that it is only after mileage accumulation that the true
effects of the device can be measured. Therefore the effect of the
exhaust valve-to-seat seal deterioration on the long term effects of the
Mark II must be addressed. First, as evidenced in Figure 5 the change
in HC emissions, which are indicative of the leak, did not increase
until after the mileage accumulation was completed. Thus the mileage
accumulation can be assumed to have occurred normally. Second, when the
HC emission increased from approximately 1.7 to 2.7 grams per mile (gpm)
the fuel economy decreased only 4 to 5 percent and the other gaseous
emissions remained virtually unchanged. This indicates that the loss in
compression in the one cylinder was relatively minor and that the cylinder
continued to fire normally. Again the operation of the Mark II was not
*
Only the tests before mileage accumulation and after the v;ilve replace-
ment were considered because the remaining, tests were taken on a mal-
functioning engine, as evidenced by the steadily increasing 1 ydrocarbon
emissions, which was remedied by the valve replacement.
-------
Figure 3
'75 FTP Composite Fuel Economies
by individual test
c
o
oc
CO
a)
1-1
•rl
o
o
K
0)
30.0
25.0
20.0
15.0
10.0
5.0
• Before mileage
accumulation
A
A-r
o
—TV
u^GJO
After 1500 miles
Q - no device
Q]- Mark II "dry1
- Mark II
A
30
25
20
After valve replacement
mean value
-------
Figure 4
Highway Cycle Fuel Economies
by individual test
(6
00
14
0)
40
0-0-0
30
Before mileage
accumulation
A
A
A
O
After 1500 miles
40
30
After valve
replacement
oo
o
c
o
o
w
0)
20
10
G - no device
El - Mark II "dry"
A - Mark II
mean value
0
-------
Figure 5
'75 FTP Composite Emission
3.5-
3.0-
2.5 -
2.0-
1.5.
i.o_
uy xuuxvxuuax uc=,u Symbol Key
HC CO NOx
A
A
A A
A AA
O
O ^ *k *
3 •
-Ui — r> n — — — "-
•
-< — before mileage *>•
accumulation
A A A A
* A
•
O
n
(9 (9 ' ' O
^1 H r-™ m !"• n n -^™ — ^"~
• • • -B-a— * -Q— u— Q-J-J— Q- • "
After 1 500 miles — - ~^
0 A
0 A
3 A
A A A
i
A •
• •
• • •
-, Af te
repl
rt] - no device
| - Mark II
(J - Mark II "dry"
— mean value
A
A A
A
O 3 . c
QJ
acement - 1 oo
K
•rl
S5
0
-------
10
Table 2
"75 FTP Composite Results
Mass Emissions, grams per mile (grams per kilometer)
Fuel Economy, miles per gallon (liters per 100 kiloim.-ter)
TVf^L dull li'.urjiMon
Be 1 '_<•> r>' m I 1 K a BC a c c i imu 1 a t 1 oil
no cIcvu-L-
ti
Mark IJ "dry"
it
Mark II
ti
"
After 1500 miles
Mark II
it
"
»
Mark II "dry"
ii
M
no device
"
11
tl
II
Mark II
"
After valve replacement
Mark II
"
»
11
tl
no device
M
Hark II "dry"
ii
HC
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
2.
2.
2.
2.
2.
2.
2.
2.
2.
1.
1.
1.
1.
1.
1.
1.
1.
77
65
66
48
43
56
65
70
64
78
67
88
15
13
17
16
10
25
37
68
53
47
58
84
43
61
59
73
61
46
(1.10)
(1.03)
(1.03)
(0.92)
(0.89)
(0.97)
(1.03)
(1.06)
(1.02)
(1-11)
(1.04)
(1.17)
(1.34)
(1.32)
(1.35)
(1.34)
(1.31)
(1.40)
(1.47)
(1.67)
(1.57)
(0.91)
(0.98)
(0.52)
(0.89)
(1.00)
(0.99)
(1.08)
(1.00)
(0.91)
CO
33. 4
25.8
29.4
28.6
25.2
25.7
31.3
29.9
28.5
29.6
29.1
29.7
32.3
30.6
29.2
30.4
27.8
31.3
31.3
31.8
30.3
27.6
26.0
23.1
26.0
26.2
26.1
28.0
26.4
24.9
(20.8)
(16.0)
(18.3)
(17.8)
(15.7)
(16.0)
(19.5)
(18.6)
(17.7)
(18.4)
(18.1)
(18.5)
(20.1)
(19.0)
(18.1)
(18.9)
(17.3)
(19.5)
(19.5)
(19.8)
(18.8)
(17.2)
(16.2)
(14.4)
(16.2)
(lfi.3)
(16.2)
(17 4)
(16.4)
(15.5)
£0j
295
302
292
296
314
305
305
297
309
311
306
306
311
311
292
311
306
310
310
310
320
331
301
268
290
306
298
293
. 296
299
(183)
(188)
(181)
(184)
(195)
(190)
(190)
(185)
(192)
(193)
(190)
(190)
(193)
(193)
(181)
(193)
(190)
(193)
(193)
(193)
(199)
(206)
(187)
(167)
(180)
(190)
(185)
(182)
(184)
(186)
NUx
2.12
2.04
2.08
2.35
1.87
2.21
2.15
1.93
2.17
2.05
2.03
1.99
2.02
2.03
1.93
2.02
1.97
2.05
1.95
2.09
2.11
1.96
1.76
1.43
1.68
1.76
2.21
1.75
2.10
2.35
rue i
IZcononiy
(1.32)
(1.27).
(1.29)
(1.46)
(1.16)
(1.37)
(1.34)
(1.20)
(1.35)
(1.27)
(1.26)
(1.24)
(1.26)
(1.26)
(1.20)
(1.26)
(1.22)
(1.27)
(1.21)
(1.30)
(1.31)
(1.22)
(1.09)
(0.89)
(1.04)
(1.09)
(1.37)
(1.09)
(1.31)
(1.46)
X3.1
25.5
25.8
25.7
24.8
25.3
24.7
25.4
24.7
24.4
24.8
24.7
24.0
24.2
25.7
24.3
24.9
24.2
24.2
24.1
23.6
23.',
25.6
28.9
26.5
25.2
25.8
25.9
25.9
25.9
C9.37)
(9.23)
(9.12)
(9.16)
(9.49)
(9.30)
(9.53)
(9.26)
(9.53)
(9.64)
(9.49)
(9.53)
(9.80)
(9.72)
(9.16)
(9.68)
(9.45)
(9.72)
(9.72)
(9.76)
(9.97)
(10.1)
(9.19)
(8.14)
(8.88)
(9.34)
(9.12)
(9.08)
(9.08)
(9.08)
-------
Test
configuration
Before milenge accumulation
Mark II
Mark 11 "dry"
no device
Mark II
After 1500 miles
Mark II
Mark II "drv"
no device
Mark II
Af ter__valve replacement
Mark II
nu devir.e
M-irk 11' 'Mr/'
11
Table 3
Highway Cycle Results
Kass emissions ( grams per mile (j;ram per kilometer)
Fuel economy, miles per gallon (liters per 100 kilometer)
HC
0.91
0.88
0.88
0.90
0.91
0.90
0.88
0.88
0.88
0.75
0.86
0.89
0.87
0.92
0.91
0.89
0.96
1.00
0.93
1.00
1.02
1.03
1.13
1.01
0.81
0.88
0.36
0.75
0.83
0.90
0.89
0.93
0.88
0.88
(0.57)
(0.55)
(0.55)
(0.56)
(0.57)
(0.56)
(0.55)
(0.55)
(0.55)
(0.47)
(0.53)
(0.55)
(0.54)
(0.57)
(0.57)
(0.55)
(0.60)
(0.62)
(0.58)
(0.62)
(0.63)
(0.64)
(0.70)
(0.63)
(0.50)
(0.55)
(0.22)
(0.1,7)
(0.52)
(0.56)
(0. Vi)
(0. 5B)
(0.55)
(n'.55)
O)
8.97
8.40
9.16
8.10
8.12
8.46
8.30
8.31
8.02
5.84
7.55
9.66
7.88
8.60
8.75
7.78
9.62
10.10
7.15
8.30
8.66
10.18
8.55
6.92
9.10
8.01
7.31
7.70
7.92
8.89
7.97
9.57
8.31
7.90
(5.57)
(5.22)
(5.69)
(5.03)
(5.0b)
(5.26)
(5.16)
(5.16)
(4.93)
(3.63)
(4.69)
(6.00)
(4.90)
(5.34)
(5.44)
(4.84)
(5.98)
(6.28)
(4.44)
(5.16)
(5.38)
(6.33)
(5.31)
(4.30)
(5.66)
(4.98)
(4.54)
(4.79)
(4.92)
(5.51)
(4.rr>)
(5.9r>)
(5.16)
(4.91)
CO
207
205
210
211
210
210
210
206
206
222
221
213
223
227
217
222
222
222
204
224
216
225
222
220
236
211
185
226
205
208
207
207
216
215
(129)
(127)
(131)
(131)
(131)
(131)
(131)
(128)
(128)
(138)
(137)
(132)
(139)
(141)
(135)
(138)
(138)
(138)
(127)
(139)
(134)
(140)
(138)
(137)
(147)
(131)
(115)
(140)
(127)
(129)
(139)
(129)
(134)
(134)
NOx
3.55
3.32
3.37
3.33
3.32
3.40
3.41
3.36
3.38
3.86)
3:77
3.32
3.66
3.64
3.32
3.44
3.46
3.25
3.06
3.38
3.04
3.55
3.55
3.55
3.07
3.04
2.45
3.40
2.94
2.9V.
3.W
2.Hb
3.51
3.56
Fuel economy
(2.21)
(2.06)
(2.09)
(2.07)
(2.06)
(2.11)
(2.12)
(2.09)
(2.10)
(2.40)
(2.34)
(2.06)
(2.'27)
(2.26)
(2.06)
(2.14)
(2.15)
(2.02)
(1.90)
(2.10)
(1.89)
(2.21)
(2.21)
(2.21)
(1.91)
(1.89)
(1.52)
(2.11)
(1.83)
(1.82)
(2.21)
(1.77)
(2.18)
(2.21)
39.6
40.1
39.1
39.2
39.3
39.2
39.3
39.9
40.0
37.9
37.7
38.4
37.2
36.5
38.0
37.4
37.0
36.8
40.6
36.9
38.2
36.4
37.2
37.9
35.1
39.2
44.8
•Jfi.9
40.3
39.4
19.9
' 39.4
38.2
38.5
(5.94)
(5.87)
(6.02)
(6.00)
(5.99)
(6 . 00) '
(5.99)
(5.90)
(5.88)
(6.21)
(6.24)
(6.13)
(6.33)
(6.45)
(6.19)
(6.29)
(6.36)
(6.39)
(5.80)
(6.38)
(6.16)
(6.46)
(6.33)
(6.21)
(6.70)
(6.00)
(5.25)
(6.38)
(5.84)
(5.97)
(5.90)
Cj.1)?)
(6.16)
(6.11)
Temp.
°F
68.0
63.0
68.0
65.5
66.0
66.0
68.0
65.0
66.0
65.5
69.0
70.5
72.0
66.5
65.0
70.5
68.0
71.0
69.0
71.0
73.5
70.0
70.0
73.0
77.0
71.0
71.0
70 . 0
69.0
67.5
70.0
67. r>
74.0
74.0
Rel.
hum.
. Z
76
61
61
64
52
58
55
52
58
45
55
42
44
51
52
47
53
53
47
46
43 .
46
49
44
56
49
. 46
46
51
3 J
76
48
33
31
Bare, P
JlHfi_
2H.57
2;:. 50
2,'i.4S
21.46
23.47
23.46
23.48
23.48
23.44
2:1.34
2i.83
J i.91
29.09
29.01
28.84
29.16
28.98
28.99
29.18
28.98
28.70
28.75
28.98
29.25
29.12
29.07
29.04
29.21
29.13
29.03
79.08
28.87
29.05
29.05
-------
12
changed and the cylinder deposits if affected at all would only be
slightly altered in the one cylinder. Third, when the valve was replaced
considerable care was taken not to disturb the combustion chamber deposits,
It is our conclusion therefore that the tests taken after the valve was
replaced are valid representations of the effect of the Mark II after
approximately 2,000 miles.
Table 4 contains the average fuel economies and standard deviations
of the three configurations tested before and after mileage accumulation
for tjje '75 FTP. Also shown are the percent improvement and t test
value using the no-device configuration before mileage accumulation as
the base sample. The final column in this table is the resolution to
the t test null hypothesis, i.e., that there is no significant difference
between the configuration and the base sample with a confidence level of
90%. A "yes" in this column indicates that there is a significant
difference at that confidence level. By this test none of the configura-
tions showed a significant difference from the base sample. While some
of the configurations showed mean percent improvements of 2.0 percent,
these changes were not sufficiently different from the test-to-test
variability to be considered significant.
Table 5 is the same format as Table 6 for the Highway cycle fuel
economies. While the configurations showed decreases in fuel economy up
to 3.3 percent, none were found to be significantly different from the
observed variability.
Table 6 is again the same format for the hydrocarbon, carbon
monoxide, and nitrogen oxides emissions of the '75 FTPs. Many of the
configurations showed mean percent reductions in the order of 10 percent,
and the Mark II after 2000 miles showed a 16.7 percent reduction in
nitrogen oxides. Despite the apparently large percentages of reduction,
none of these were found to be significantly different from the test-to-
test variability.
The t tests were calculated using an overall standard deviation calculated
by averaging the variances of the six test sets (3 configurations before
and 3 after 2000 miles) using the equation:
2 6 2 6
o = Z n a. /£ n.
av. i=l ili=l1
where n. = sample size of test set i
2
a. = the variance (standard deviation squared) of test i
This involves the reasonable assumption that while the mean value of
the test sets may vary, the test-to-test variability within each set is
the same for all six sets.
-------
13
Table 4
'75 FTP Fuel Economy Statistics
Significantly
Test
configuration
Before mileage
accumulation
no device
Mark II
Mark II "dry"
After 2000 miles
no device
Mark II
Mark II "dry"
Sample
size
2
3
2
2
3
2
Mean
mpg
25.3
24.9
25.8
25.8
25.8
25.9
Standard
2ES.
+0.3
+0.3
+0.1
+0.1
+0.7
+0.0
dev.
%
+1.1
+1.3
+0.3
+0.3
+2.6
+0.0
Percent ^
improvement t
-1.6 +1.21
+2.0 -1.38
+2.0 -1.38
+2.0 -1.52
+2.4 -1.66
different at
90% confidence
No
No
1
No
No
No
'All percent improvement and t tests conducted with no device before mileage
accumulation used as the base sample. The t tests were calculated using
an overall standard deviation of +0.36 mpg for the *75 FTP fuel economy.
Table 5
Highway Cycle Fuel Economy Statistics
Significantly
Test
configuration
Before mileage
accumulation
no device
Mark II
Mark II "dry"
After 2000 miles
no device
Mark II
Mark II "dry"
Sample
size
3
5
3
2
3
2
Mean
mpg
39.7
38.9
39.2
39.6
39.6
38.4
Standard dev.
mpg
+0.4
+1.0
+0.1
+0.4
+0.6
+0.2
%
+1.0
+2.7
+0.1
+0.9
+1.5
+0.6
Percent
Improvement
-2.1
-1.3
-0.3
-0.3
-3.3
^ different at
t 90% confidence
+1.72 No
+0.96 No
+0.17 No
+0.19 No
+2.23 No
All percent improvement and t tests conducted with no device before mileage
accumulation used as the base sample. The t tests were calculated using an
overall standard deviation of +0.64 mpg for the Highway Cycle fuel economy.
-------
Table 6
175 FTP Composite Emissions Statistics
Hydrocarbons
Carbon Monoxide
Test Sample
Cf»n figuration : Size
Before mileage
accumulation
no device 2
Mark II 3
Mark II "dry" 2
After 2000 miles
no device 2
Mark II 3
Mark II "dry" 2
Mean
g/mi.
1.71
1.55
1.57
1.66
1.54
1.54
Standard dev.
g/mi.
+0.08
+0.11
+0.13
+0.10
+0.10
+0.11
%
+5.0
+7.2
+8.1
+6.0
+6.3
+6.9
*
Percent
reduction
—
+9.6
+8.2
+2.9
+9.9
+10.2
Significantly
a different at
t 90% confidence
—
+1.69 No
+1 . 35 No
+0.43 No
+1.79 ' No
+1.63 No
1
ft Significantly
Mean Standard dev. Percent A different at
g/mi. g/mi.
29.6 +5.4
27.4 +3.4
29.0 +0.6
27.1 +1.3
26.1 +0.1
25.6 +1.1
% reduction t 90% confidence
+18.2
+12.4 +7 ..4
+ 2.0 +2.0
+5.0 +8.6
+0.4 +11.9
+4.1 +13.3
—
+0.91 No
+0.23 No
+0.94 NO
41 .45 NO
+1.51 No
Nitrogen Oxides
Test Sample
ronfieuration size
Before mileage
accumulation
no device 2
Mark II 3
Mark II "dry" ' 2.
After 2000 miles
no device 2
Mark II 3
Mark II "dry" 2
Mean
g/mi.
2.08
2.07
2.22
1.98
1.73
2.22
Standard dev.
g/mi.
+0.06
+0.18
+0.19
+0.33
+0.05
+0.18
%
+2.7
+8 ..7
+8.6
+16.4
+2.7
+7.9
A
Percent
reduction
—
+0.5
-6.5
+4.8
+16.7
-7.0
Significantly
a different at
t 90% confidence
'
—
+0.06 No
-0.77 No
,
+0. 53 No
+2.12 No
-0.77 No
All percent reduction and t tests conducted with no
device before mileage accumulation used as a base
sample.
standard
+0.
+2.
+0.
The t tests were
deviations of:
10 g/mi. for HC
65 g/mi. for CO
18 g/mi. for NOx
calculated using overall
-------
15
Table 7 gives the levels of change that were necessary to be con-
sidered significantly different at 90% confidence with the observed test
variability and different sample sizes. This shows that a sample size
of 7 is required to be able to detect with 90% confidence a difference
equal to the standard deviation. Sample sizes of this order can be
obtained in this analysis if all the before mileage accumulation tests
are grouped and compared to all the after 2000 miles tests. This grouping
should reveal any overall shifts in emissions or fuel economy with
mileage accumulation and thus reveal any long term benefit of the Mark
II. Table 8 shows the results of this grouping for the emissions over
the '75 FTP and for the fuel economies over the '75 FTP and HFET. Under
the column titled "Comparative Statistics" are given the percent change
in the group means, the t test score and the resolution of the same t
test null hypothesis as used in Tables 4, 5, and 6.
The CO and NOx emissions showed reductions of 7% and 8% respec-
tively but just missed being significantly different from test-to-test
variability. The '75 FTP fuel economy improvement of 2% was found
significantly different. From this there appeared to be a slight but
real improvement: in the 75 FTP fuel economy after mileage accumulation
with the Mark II. There was no corresponding improvement in the HFET
fuel economy.
Any fuel economy benefits that would result from the alteration of
the combustion chamber deposit, would be expected to be reflected in
both the 75 FTP and HFET. Thus it was difficult to envision a long term
effect of the Mark II that would tend to improve only the low speed
stop-and-go type driving fuel economy.
Table 9 shows the combined city/highway fuel economy of the test
vehicle in the different configurations tested. Also shown is the fuel
consumed and its cost over a period of one year of average driving,
assuming the annual mileage of 10,000 miles and gasoline cost of $.60/
gallon. The Mark II after mileage accumulation showed a savings of
$2.58 over the no-device configuration before mileage accumulation.
The price of the Mark II with Vapor Jet is listed at $47.90. The
owner's manual recommends refilling the Mark II with Econo Mix ($1.95
for a 15 oz. can) every 90 days, yielding an annual operating expense of
$7.80. Thus, at least for the test vehicle the Mark II does not appear
economically justifiable.
-------
16
Table 7
Percent difference between sample means detectable
at 90% confidence as a function of sample size
Percent difference, detectable with
Overall
Std. dev.%
' 75 FTP Emissions
HC +5
CO +9
NOx +8
Fuel Economy
75 FTP +1
HFET +1
.8
.0
.7
.4
.6
no.
2 3
+17.1 +10.2
+26.1 +15.6
+25.3 +15.1
+4.2 +2.5
+4.7 +2.8
of tests
4.
+8.0
+12.3
+11.9
+2.0
+2.2
equal to
5 7 10
+6.9
+10.5
+10.2
+1.7
+1.9
+5.6 +4.f.
+8.5 +6.9
+8.2 +6.7
+1.4 +1.1
+1.5 +1.2
*
No. of tests is for both samples, each having the same indicated number of
individual tests, i.e. the number 5 indicates two samples of 5 tests each for
a total of 10 tests.
Before mi.
'75 FTP Sample
emissions size
HC 7
CO 7
NOx 7
Fuel Economy
'75 FTP 7
HFET 11
accum.
Mean
g/mi
1.60
28.5
2.11
mpg
25.3
39.2
Table 8
After 2000 mi.
Sample Mean
size g/mi
7 1.57
7 26.2
7 1.94
mpg
7 25.8
7 39.3
Comparative.
*
Percent
change
-1.9
-7.1
-8.1
+1.9
+0.3
*
t
0.56
1.61
1.78
2.92
0.32
Statistics
Significantly
different at
90% confidence
No
No
No
Yes
No
Before mileage accumulation mean values were used as base sample for
percent change and t test. Overall standard deviations used are those
given in Tables 4, 5. and 6.
-------
17
Table 9
Configuration
Before mileage
accumulation
no device
Mark II
Mark II "dry"
After 2000 miles
no device
Mark II
Mark II "dry"
Composite
fuel economy
mpg
30.2
29.7
30.4
30.6
30.6
30.3
Gasoline used
per year
gallons
**
331.1
336.7
328.9
326.8
326.8
330.0
Gasoline
cost per
year
***
$198.66
$202.02
$194.34
$196.08
$196.08
$198.00
Performance oi: the vehicle was not specifically examined, but the
vehicle was unable to maintain the hard acceleration occurring from 180
to 200 seconds into the transient cycle of the '75 FTP. Figure 6 shows
this section of the driving cycle, with the cross hatched area represen-
ting the difference between the prescribed speed time trace (upper
curve) and the vehicle's actual speed time trace. The "WOT" on the
trace was written by the driver indicating that the throttle was wide
open. Had any power improvements occurred, the vehicle would have been
better able to follow the prescribed trace and the cross hatched area
would have been smaller. Since no noticeable changes in this area were
produced by any of the configurations tested, it was concluded that no
noticeable changes in vehicle performance occurred.
Composite fuel economy =
.55
.45
'75 FTP F.E.
Highway F.E.
**
***
Annual mileage 10,000 miles
Gasoline cost of $.60/gallon
-------
18
Figure 6
i i
Typical speed-time trace for the Vega in the region of 180 to 240
seconds into the '75 FTP. Cross-hatched area added for clarity.
»
Conclusions
The calculations of the preliminary analysis show that the quantities
of water, methanol, and acetone added by the Mark II are considerably
smaller than additions reported in the literature that produced measure-
able changes in octane requirement, power, or fuel economy. The Mark II
can be considered as a small auxiliary carburetor, which under favorable
conditions delivers a mixture with an air fuel ratio near stoichiometric.
If this addition to the total carbureted mixture were ignored, the
apparent increase in fuel economy was calculated to be around 0.3 percent
on the '75 FTP. If the Mark II were operated "dry" it could increase
the overall air fuel ratio by 0.28 at idle. Depending on the original
air fuel ratio, this could produce small but measurable changes in
emission levels and low speed fuel economy. Possible long term effects
of the Mark II were not considered in the preliminary analysis.
-------
19
The test results show that all configurations tested yielded the
same emissions and fuel economy within test-to-test variability. By
combining all tests before mileage accumulation and comparing them .to
all the after-2000 mile tests, a significant 2% increase in '75 FTP fuel
economy was observed with mileage accumulation. There was not a cor-
responding increase in HFET fuel economy. Throughout the testing
sequence no improvements in vehicle performance were observed.
Based on the results from the test car, the operating expenses of
the Mark II exceeded the savings in fuel by a factor of three. It is
the conclusion of the analysis that the purchase price and operating
expenses of the Mark II do not appear to be justified by the insignifi-
cant changes in emission levels and minor fuel economy improvement
produced by the Mark II.
-------
References
"Emission Results from an Automobile Using the Frantz Vapor Injector,"
TAEB Report #72-5, 9/71.
o
"Evaluation of the Turbo Vapor Injector," TAEB Report #73-22, 3/73.
3 "Evaluation of the SCATPAC Device," TAEB Report #74-6, 7/73.
4 "An Evaluation of the Econo-Mist Device," TAEB Report #75-19, 3/75.
See Federal Register 38 FR 11334, 3/27/74, for a description of the
test protocols proposed for definitive evaluations of the effectiveness
of retrofit devices.
Natrella, M.G. Experimental Statistics, Nat. Bur. Stand. Hand.
91, Aug. 1, 1963.
7 Potter el al,. "Weather or Knock," Trans. SAE, 62, 1954, p. 346.
g
Ingamells, Stone, Gerber, Unzelman, Effects of Atmospheric Variables
on Passenger Car Octane Number Requirements," SAE Paper #660544.
a
Nicholls, El-Messiri, and Newhall, "Inlet Manifold Water Injection
for Control of Nitrogen Oxides - Theory and Experiment," SAE Paper
#690018.
Obert, "Detonation and Internal Coolants"
Trans, SAE, 2, Jan. 1948, p. 52.
Ingamells and Lindquist, "Methanol as a Motor Fuel or a Gasoline
Blending Component," SAE Paper #750123 Feb. 1975.
12
Wigg and Lunt, "Methanol as a Gasoline Extender - Fuel Economy,
Emissions, and High Temperature Driveability SAE," Paper #741008,
Oct. 1974.
13
Powell, "Racing Experience with Methanol and Ethanol-Based Motor-Fuel
Blends," SAE Paper #750124, Feb. 1975.
14
Private communication with R. Campion, Exxon Research and Engineering.
"Passenger Car Fuel Economy - Dynamometer vs. Track vs. Road"
EPA, ECTD, TAEB Report #76-1.
Private communication with H. Toulmin, Sun Oil Company.
-------
Appendix I
Preliminary analysis of Mark II Vapor Injector
The purpose of this preliminary analysis is to determine the approxi-
mate concentration of the various vapors in the carbureted air fuel
mixture, and to compare them to concentrations, reported in the literature,
known to produce measurable effects.
Mark II Econo Mix fluid is 65% mei.hanol, 34% acetone and 1% propylene
glycol by volume. This is mixed one pc-rt to two parts water by volume.
As only the vapors of this mixture are used and the vapor pressure of
propylene glycol is low (less than 1 mm of Hg at 100 F), we will not
include it in our analysis.
Water vapor is normally added to air-fuel mixtures because it is
present in the air. The effects of increasing humidity are fairly well
known. It lowers the octane requiremert of the engine, i.e. it acts as
a knock suppressor. Potter et al found that at 70 F a change in
relative humidity from 30 to 60 percent decreased the required motor
octane number (MON) of the fuel for an automobile engine from 88 to 86.
Ingamello, Stone, Gerber and Ungelman found studying eight automobiles
that the effects of humidity changes on required octane number was
linear with the equation:
A O.N. = - K AH (grains/lb absolute humidity)
K was observed to vary from 0.04 to 0.09 for the cars tested with an
average of 0.045. This is in good agreement with Potter, yielding a 1.4
O.N. decrease versus the 2 from Potter for the 30% change in relative
humidity at 70°F.
As a diluent, water vapor also decreases the charge density and
indicated thermal efficiency. Slight power improvement is possible with
increasing humidity if the engine was previously spark limited and can
take advantage of the increased octane number by increasing spark advance
and/or increasing charge density (opening the throttle more for a
normally aspirated engine).
9
Nichols, El-Messiri and Newhall investigated the effects of inlet
manifold water injection on oxides of nitrogen emissions. They found
that at air-fuel ratios near stoichiometry, 30 to 50 percent reduction
of nitric oxide emissions were observed with a water to fuel weight
ratio (W/F) of 0.50. The effects of water injection on percent reduction
of NOx appeared linear in the range of W/F =0-0.5. The effectiveness
of water in this regard was attributed primarily to its high latent heat
of vaporization resulting in lower peak combustion temperatures.
Obert investigated injections of liquid water and water-alcohol
mixtures into the intake manifold as a means of knock suppression.
While effective, it required large amounts of water, around 50% of the
fuel volume. This technique has been used for airplane engines during
take-off. Much of the effectiveness of this method has been attributed
by Obert to the high latent heat of vaporization of the liquids.
-------
Methanol has been widely investigated as a possible fuel blending
component and alternate fuel. Ingamells and Lindquist found, using
different unleaded gasolines, that the addition of 5% by volume methanol
increased the MON by 0.1 to 1.5 octane number (ON) while the addition of
10% methanol increased the MON by 1.6 to 2.5 ON. They also reported
that on a miles per gallon basis the addition of 10% methanol produced
an average 3.2% fuel economy loss for a six car fleet in commuter type
driving. Wigg and Lunt reported the effects of the addition of 15%
methanol by volume to gasoline on the exhaust emissions of cars operated
over the '75 FTP. For a 1973 car the methanol addition resulted in a
36% increase in HC, a 50% decrease in CO, and a 24% decrease in NOx.
These effects however were attributed to 1) the enleanment effect of
using the alcohol blends in carburetors designed for gasoline (The
addition of 15% methanol increased the equivalence ratio by about 0.1
unit.), and 2) the relatively higher latent heat of vaporization of
methanol resulting in..a cooler inlet charge and lower peak flame temper-
atures. In Powell's review of racing motor-fuel blends he estimates a
20% gain in power for high compression ratio engines using methanol
instead of regular gasoline. This was attributed to methanol's 1)
ability to burn richer of stoichiometric than gasoline, 2) higher
octane number and 3) higher heat of vaporization producing cooler,
denser inlet charges. This power increase also entailed a specific fuel
consumption of methanol being about three times that of gasoline.
The use of acetone in fuels has not been considered very much
because it and other ketones promote the formation of gum. For this
reason, production gasolines generally limit naturally occurring ketones
to .005 weight percent. Powell reported that acetone was sometimes
used in racing fuels in concentrations of 5% or less. It was used with
methanol blends as a blend stabilizer and water tolerance booster because
of its high solvent powers. Gumming and other fuel system problems
associated with acetone and methanol are prevented by draining and
flushing of the system after each race.
Turning now to the Mark II we can reasonably approximate the con-
centrations of water, methanol, and acetone contributed by the device to
the induction mixture. First the operating variables of the device were
measured. With the aerator valve on the Mark II reservoir adjusted
according to the manufacturer's instructions, the air flow into the
valve was measured with a wet test meter. The air flow was 1.4 cubic
feet per hour with the engine idling and at steady state 50 mph cruise.
At wide open thottle the flow was less than .02 cubic feet per hour.
The maximum flow rate measured, 2.6 cubic feet per hour, was with the
valve wide open (an improper setting according to the instructions).
The temperature of the reservoir fluid was measured throughout a '75 FTP
and Highway test with the reservoir installed in the car but not connected
to the vacuum source. As there was no air bubbling through the fluid
the evaporative heat losses associated with it were not present. The
initial temperature was 80 F and the final temperature after 60 minutes
was 98 F and still rising slowly. The vacuum in the bottle with the air
bubbling rate properly adjusted was 10 inches of mercury. For convenit-.nce
-------
we will use an air flow of 1.4 cubic feet per hour, reservoir temperature
of 100 F and an absolute pressure in the bottle of 20 in. Hg in our
calculations below.
Calculation of weight of air, water, methanol and acetone delivered
per hour by the Mark II:
Assumptions
1. Air entering bottle is at 80 F and at 50% relative humidity.
2. Fluid mixture obeys Raoult's Law, i.e. each component's equili-
brium vapor pressure above the liquid is equal to the vapor pressure of
the pure component's equilibrium vapor pressure at that temperature
times the mole fraction of that component in the liquid mixture.
3. The vapor-liquid concentrations are in equilibrium after
bubbling.
Given the Econo Mix composition of 65% methanol, 34% acetone and 1%
propylene glycol, mixed with 2 parts water; the resulting mixture per
litre is:
water
methanol
acetone
propylene glycol
667 ml
217 ml
113 ml
3 ml
1000 ml
Component
Density
g/litre
water 1.00
raethanol 0.79
acetone 0.79
propylene glycol 1.04
Molec. Wt.
18.0
32.0
58.
76.
.1
.1
Moles/litre
pure component
55.6
24.7
13.6
13.7
Mole fraction
in mixture
.842
.122
.035
.001
Absolute pressure in bottle is 20 in. Hg or 510 mm Hg.
Component
water
methanol
acetone
propylene glycol
air
Mole fraction
in liquid
Vapor pressure of
pure component
at 100°F in mm Hg
Partial pressure
above mixture
.842
.122
.035
.001
49
230
380
1
43.3
28.1
13.5
0.0
425.
Mole fraction
in air vapor
mixture
.085
.055
.026
.000
.834
Total
510
1.000
-------
Below the composition of one mole of air-vapor mixture (22.4 Si at
standard conditions of 0 C and 1 atm.) and the weight and volumes of
these components delivered per hour by the Mark II are shown.
Component
Weight per 1
mole of mixture
in grams
water 1.53
methanol 1.76
acetone 1.51
propylene glycol 0.00
air 24.0
Weight delivered
in one hour in grams
3.28
3.77
3.23
.00
51.4
Liquid volume
delivered in one
hour in ml.
3.28
4.77
4.09
.00
**
Molecular weight x mole fraction in air-vapor mixture
Total volume of vapor mixture delivered to the engine by the Mark II
at STP is:
(1.4 cu. ft./hr.) (
510 mm Hg
425 mm Hg
) = 1.7 cu. ft./hr. or 48 «./hr.
Calculating the fuel consumed by the test vehicle and the vapor
components delivered by the Mark II during the '75 FTP, Highway test
and idle:
'75 FTP
Highway
Idle
Driving Time
hours
.521
.21
1.0
Distance
miles
11.1
10.2
0.0
Gasoline consumed
Vega F.E.
25.7
39.3
Gallons
.432
.260
.4+
ml.
1635
982
1500
Estimated from the following idle fuel consumption data from reference
13. '75 Ford Pinto - .426 gal./hr. '75 VW Rabbit - .388 gal./hr.
Component
water
methanol
acetone
gasoline
'75
Volume
ml.
1.71
2.49
2.13
1635.
FTP
Vol.%
0.10
0.15
0.13
99.63
Highway
Volume
ml.
0.69
1.00
0.86
982.
Vol.%
0.07
0.10
0.09
99.70
Idle
Volume Vol.:
ml.
3.28
4.77
4.09
1500.
0.22
0.32
0.27
99.20
Total
1641.
985.
1512.
-------
Thus we see that considering the vapor components contributed by
the Mark II as part of the fuel, they represent a very small fraction:
only 0.30% on the Highway test, 0.37% on the '75 FTP and 0.80% at idle.
This checks well with the observed consumption of 550 ml of reservoir
fluid during the accumulation of 1600 miles. With a composite fuel
economy of 30.4 mpg, 52 gallons or 200 litres of gasoline were used
yielding a 0.28% by volume addition of the reservoir fluid.
Water addition due to the Mark II amounted to at most 0.22% and of
that, slightly over 50% was the original humidity of the air entering
the Mark II. Assuming an overall stoichiometric air fuel ratio, this
amounts to a 1.4 grains of water addition per pound of incoming air.
This is equivalent to a humidity change of a little less than one relative
humidity point at 80 F. Using Ingamells et al equation for the effect
of humidity on octane requirement we can expect a decrease of 0.06 O.N.
due to the water contributed by the Mark II. This small change in O.N.
is not measurable. Using the linear relationshipqof water addition to
percent reduction of NOx observed by Nichols et al , we would expect a
0.2 to 0.3 percent reduction in NOx emissions due to the water additions
of the Mark II if this water were in a liquid state when it entered the
intake manifold. Since the Mark II adds only water vapor the benefits
of the high latent heat of vaporization are lost. Thus the actual
effect would be smaller than the 0.2 to 0.3 percent reduction above,
which is already way below our test-to-test variability. Most important
of all however is the fact that normal day-to-day weather variations
produce humidity changes that dwarf those produced by the Mark II.
The maximum methanol addition of 0.32 volume percent is an order of
magnitude smaller than reported additions of 5% that produced a .1 to
1.5 octane number change. Assuming that the effects of methanol addition
to gasoline are linear with the percentage of volume addition, we can „
estimate the emissions changes over the '75 FTP from the Wigg and Lunt
data. That is a 15% by volume addition of methanol to gasoline resulted
in a 36% increase in HC, a 50% decrease in CO, and a 24% decrease in
NOx. Thus we might expect a 0.8% increase in HC, a 1.1% decrease in CO,
and a 0.5% decrease in NOx. Again with the Mark II the effect of the
high latent heat of vaporization of methanol is lost so the effect on
NOx would be less. These small changes are not measurable on the '75
FTP because of the test-to-test variability. Methanol additions of
0.25% are..routinely added to production winter gasolines by some oil
companies to prevent ice crystal formation in the fuel. This small
addition is not known to have any measurable effect on any engine
variable.
^ 1.00 density H~0 Ib. gasoline
.0022 vol. fraction H-0 x -oft , : 7 —— x ..,. .. :—
2 .739 density of gasoline 15 Ib. air
7000 grains/lb. =1.4 grains H20/lb. air.
-------
As mentioned earlier acetone is avoided in production gasolines.
It has been used in methanol blend racing fuels, in concentrations up to
5%, as a blend stabilizer, not for any known benefits as an octane or
power booster. The maximum concentration of acetone contributed by the
Mark II was 0.27%. It is unlikely that this small a concentration woulc
produce any measurable effects.
As methanol and acetone are combustible, the Mark II can be con-
sidered as an auxiliary carburetor. Below is a calculation of its
equivalence ratio. (Equivalence ratio is the observed air-to-fuel ratio
by weight divided by the stoichiometrically correct air-fuel ratio for
that fuel. A rich mixture will have an equivalence ratio less than 1.0
and a lean mixture will have one greater than 1.0)
Component
methanol
acetone
air
.21 x .834 =
Mole fraction
in air vapor
mixture
.055
.026
.834
.175
moles of 0- per
mole of component
for complete oxidation
1.5
4.0
Mole fraction of
QO require
.083
.104
Total
,187
So equivalence ratio is .175 _
.187
= .935 or slightly rich.
So when the Mark II reservoir fluid is fresh and at 100 F the air-vapor
mixture is rich. When the reservoir is cooler, the mixture will be
leaner. Also with mileage accumulation, as the conentrations of methanol
and acetone are depleted, the mixture will become leaner. Since the
methanol and acetone represent at most only 0.59% of the fuel their
effect on the overall air-fuel ratio is minimal. If the Mark II were
operated without fluid ("dry"), the air entering would lean the carbureted
mixture. At idle an original A/F of 15.0:1 would go up to:
.079 Ib/cu. ft air) =
6.0 Ib. air/hr. + (1.4 cu. ft air/hr. x
.4 Ib. fuel/hr.
or a 1.8% increase. The effects of air-fuel ratio changes on fuel
economy and emissions are well documented; and while this is the maximum
increase expected, it is sufficient at certain air fuel ratios to produce
small but measurable changes in emissions and low speed fuel economy.
This would be operating the Mark II strictly as air bleed, and similar
results could be obtained by leaning the normal idle mixture adjustment.
-------
Since the fuel economies of this report are calculated by the
carbon balance method the carbon added by the Mark II is counted. In
the calculation, a carbon-to-hydrogen ratio and a density typical of
gasoline are used. This creates an error if a gasoline is used that is
not typical.
For the '75 FTP we can determine the error in the calculated fuel
economy which resulted when 0.28% of the fuel was not typical gasoline
but was acetone and methanol.
Fuel
Gasoline
Methanol
Acetone
1
Density
'grams/gallon
2798
2990
2990
(Grams of Carbon)
(Gram molec. wt. of fuel)
.866
.375
.620
1x2
2423
1120
1850
Total
3
Volume
fraction
.9972
.0015
.0013
1x2x3
2416
1.7
2.4
2420
Percent error = (1 -
2420
2423
) x 100 = 0.12%
That is the calculated fuel economy is 0.12% higher than actuality. At
25.0 mpg this error amounts to .030 mpg, or beyond the significance to
which we report fuel economy.
If the methanol and acetone were not considered as fuel during the
'75 FTP, the calculated fuel economy would be
higher than actuality.
(1 -
2416
2423
) x 100 = 0.29%
For 25.0 mpg this amounts to .072 mpg. This then is the magnitude
of change we would expect by ignoring the fuel content of the Mark II
vapors, measuring only the volume of gasoline consumed and dividing it
into the miles traveled, as the typical motorist might do. Even though
this is technically incorrect it still represents a very minor change in
fuel economy.
Not discussed above is the possible long term effects of the device
such as altered combustion chamber deposit quality or quantity. In
evaluating aftermarket devices, TAEB is not particularly concerned with
these changes unless they affect the emissions, fuel economy, or per-
formance of the vehicle.
-------
Test Number HC
19-8205
19-8224
19-8277
19-8324
16-8721
15-3740
15-8810
15-9009
15-9019
15-9017
15-9064
16-9036
15-9162
16-9179
16-9192
16-9219
16-9237
15-9427
15-647
15-719
15-720
16-1280
76-1988
76-1310
76-2454
76-2552
76-2629
76-2641
76-3-V.5
76-14',8
2.41
2.69
2.42
2.36
2.34
2.90
2.41
2.63
2./i2
2.98
2.50
3.31
4.06
3.77
4.04
3.91
3.18
3.15
3.41
3.73
3.48
2.59
2.76
2.13
2.60
2.39
2.54
3.15
2.41
?.OS
Bag 1
CO
53.8
41.9
52.9
57.3
47.3
48.2
50.0
45.9
44.1
47.4
46.5
48.4
56.0
51.1
53.2
55.2
48.4
53.9
51.9
52.9
54.1
50.8
47.2
41.3
55.7
51.4
48.6
54.8
50.1
44.7
Cold Transient
co2
295
316
293
307
303
299
292
298
302
307
301
298
302
303
291
304
?99
308
300
"311
311
318
290
258
281
291
284
295
289
294
NOy.
2.26
2.35
2.18
2.49
2.44
2.25
2.30
2.22
2.31
2.29
2.24
2.17
2.13
2.24
2.14
2.18
2.13
2.14
2.16
2.37
2.37
2.00
2.03
1.52
1.71
1.74
2.36
1.85
2.19
2.40
Fuel
Economy
22.9
22.8
23.1
21.9
23.0
23.1
23.4
23.4
23.4
22.7
23.2
23.1
22.0
22.5
22.9
22.0
23.1
22.0
22.6
21.9
21.8
21.9
23.8
26.9
23.6
23.3
24.1
22.7
23.6
23.9
HC
1.45
1.27
1.30
1.16
1.02
1.10
1.27
1.31
1.35
1.39
1.43
1.47
1.59
1.65
1.67
1.69
1.79
2.06
2.07
2.46
2.30
0.99
1.11
0.46
0.95
1.12
1.20
1.21
1.22
1.12
8
Appendix II
'75 FTP Individual B8.'Ki
»».<*
29.01
2«.B4
29'. 17
28. 9S
28.99
29.13
28.98
28.70
28.73
28.96
29.13
29.25
29.12
29.06
29.01
?','.] 3
29.03
20.08
28.87
fi.n
^.09
* US. GOVERNMENT PRINTING OFFICE: 1979-651-112/0121
------- |