71-29
EXHAUST EMISSIONS FROM TWO GENERAL MOTORS STEAM CARS
May 1971
H. Anthony Ashby
Bureau of Mobile Source Pollution Control
Office of Air Programs
ENVIRONMENTAL PROTECTION AGENCY
-------
71-29
EXHAUST EMISSIONS FROM TWO GENERAL MOTORS STEAM CARS
May 1971
H. Anthony Ashby
Bureau of Mobile Source Pollution Control
Office of Air Programs
ENVIRONMENTAL PROTECTION AGENCY
-------
Introduction
As part of a planned program to develop new techniques and
test procedures for measuring exhaust emissions from alternate
powerplants for automobiles, tests were conducted on two
Rankine cycle powerplants installed in chassis and supplied
by the General Motors Corporation.
The development of alternate powerplants to the conventional
internal combustion engine is the goal of the Advanced Automotive
Power Systems Program of the Environmental Protection Agency.
Those alternates which can meet or lower the Federal exhaust
emission standards of 1976 will be developed through to demon-
stration before 1975. The exceedingly low levels of HC, CO, and
NOx to be permitted by that date coupled with the unique mass flow
characteristics of most alternates under study warrant that
considerable research into new measurement methods be undertaken;
thus, the testing reported herein.
The two steam cars were made available, at no charge to the
Federal Government, by the General Motors Corporation. Delivery
of the cars was made to Building 2042 on March 29.
Vehicle Description
The cars had been designated previously as SE 101 and the SE 124
by General Motors. Both cars burn kerosene, use water as the
working fluid, use reciprocating expanders, and drive the rear
wheels through more or less conventional trnasmissions.
-------
-2-
The SE 124 was built by Besler Developments, Inc. of Emeryville,
California. This engine uses a double acting, double expansion
V-2 expander with a displacement of 124 cubic-inches and is mounted
in a 1969 Chevelle sedan. This powerplant is considered to repre-
sent a 1930 technology level.
The SE 101 uses a 1969 Pontiac Grand Prix body, with a longer than
normal engine compartment to accommodate the powerplant. The
expander is an in-line four cylinder, single stage, single acting
engine of 101 cubic-inches displacement. The design philosophy
employed in the SE 101 was that this powerplant should provide the
conveniences of the contemporary automobile including power steering,
power brakes, and air conditioning. A detailed description of the
vehicle can be found in SAE Paper No. 700163.
Test Procedures
The 1972 Federal Test Procedure was followed as closely as
possible, however, certain deviations were made in the test
procedure (intended for conventional internal combustion engine
powerplants) because of the nature of these unconventional engines.
The high exhaust flow rates of the two test vehicles dictated
use of a Constant Volume Sampler (CVS) with a larger capacity
than that specified in the 1972 Federal Test Procedure. The
0
CVS capacity was 400 scfm for the SE 124 and approximately 1000 scfm
for the SE 101.
During the testing, the combustor exhaust duct was not positively
connected to the CVS inlet duct. Instead, a gap of about one
inch was left. This was done to avoid a large pressure drop across
-------
-3-
the combustor, which might have adversely affected combustion.
Thus the combustor exhaust and dilution air were pulled through
the same duct to the CVS pump. Sample bags of the diluted exhaust
were read on two different sets of instruments, which were cali-
brated to low concentration ranges.
In console #1, NDIR is used for measuring CO, CC>2, and NO,
with a FID used for hydrocarbons. The other console (#2) uses
a NDIR for CO and C02, with a FID for hydrocarbons. The NOx is
measured using a chemiluminescent analyzer with the Saltzman
technique as a backup.
In addition, undiluted exhaust hydrocarbons were monitored using
a Beckman heated FID. Fuel consumption was determined by GM
personnel using a GM digital fuel meter.
Because of the relatively long warm-up time required by each of
these steam cars to get underway from a cold start, the LA 4
driving schedule was not started until the cars were fully capable
of following the schedule. This warm-up procedure, which required
between one and two minutes, consisted of getting up a head of
steam in the boiler (achieving full pressure) , followed by a short
expander idle period until full steam pressure again was attained.
Only then was the driving schedule started.
Sample bag filling began with initial lighting of the combustor,
and continued through the warm-up period as well as the entire
driving schedule. On one test of each car a sample bag for the
-------
-4-
start and warm-up was filled and measured, with a separate bag
then being used during the LA 4 driving schedule.
Results and Discussions
The mass emissions data for both cars are presented in grams per
vehicle mile in Table I. These are based on bag concentrations
as measured on console #2. The NOx data are corrected to 75 grains
humidity and are presented as NC>2 • The SE 101 data have been
corrected according to our latest CVS calibration at the high flow
rate. The masses shown are for the total test including the start
and warm-up period.
The hydrocarbon data presented in this report are based on readings
of a cold bag of a diluted sample on an unheated FID. It is be-
lieved that these figures may be significantly less than the actual
HC emissions, because of condensation of the heavier HC molecules
1) on the walls of the CVS inlet duct, 2) in the CVS heat exchanger,
sample pump and plumbing, and sample bag, and 3) in the console #2
plumbing.
Data from the hot FID were observed on each engine to study combustor
transients during the run. It is interesting to note that .these
traces clearly show large concentration spikes when the combustors
go on and off.
On all SE 124 tests the inertia weight was 3500 pounds, while on
the SE 101 5500 pounds was used. The final test of the SE 101,
test no. 12-1453, was run at 3500 pounds in an attempt to evaluate
the two systems on an equal weight basis. This resulted in a ten
-------
-5-
percent decrease (approximately) in fuel consumption and little
else. In that test a smaller-than-normal sample bag was inadvert-
ently used and sampling was halted at 1200 seconds of the LA 4
driving schedule. As a result, the concentrations of pollutants
in the bag, and the resultant mass emissions, ignore the final 171
seconds of the driving schedule.
Inspection of Table I shows that neither car meets the 1975
exhaust emission standards. The SE 124 consistently displayed
lower emissions than the SE 101 by a factor of two or more for
each pollutant.
The hot start emissions from each car are presented in Table II.
As expected, emissions of HC and CO are lower, NOx emissions are
higher, and fuel consumption is lower than when warm-up is included.
Again, 1975 emission standards were not met by either vehicle.
The total mass emissions and fuel consumed during the warm-up
period of each vehicle for the tests reported in Table II are
presented in Table III.
Fuel consumption data are presented in Table IV. The miles per
gallon figures were calculated from the CQ2 emissions data as
well as the GM fuel data. It is likely that the fuel consumption
of both cars would be considered unacceptable by contemporary
standards. The SE 124, a 3500 pound car, delivered an average
of about 6.5 miles per gallon of kerosene during the test procedure,
including warm-up and LA 4 driving schedule, based on the GM fuel
meter data. The SE 101, weighing 5500 pounds, gave 3 miles per
-------
-6-
gallon over the same conditions. The hot start data show that
the fully warmed-up SE 124 delivered 8.6 mpg over the LA 4
driving schedule, and the SE 101 delivered 3.3 gpm.
For comparison, 5000 pound ICE-powered prototype vehicles with
advanced emission control systems tested at Bldg. 2042 gave about
8 mpg on the cold start 1972 FTP. These tests were reported in
Test § Evaluation Branch Report No. 71-19. Conventional 5000
pound ICE-powered cars give 9 to 11 mpg on the cold start test,
and 3500 pound cars give 12 to 16 mpg.
To summarize a comparison of the two cars, the SE 101 emitted
more than two times as much HC, CO, and NOx as the SE 124, and
consumed over twice as much fuel. When tested at the same inertia
weight as the SE 124, the SE 101 still consumed almost twice as
much fuel.
-------
Table I
GM Steam Car
Mass Emissions
Modified Cold Start 1972 Federal Test Procedure
CVS
HC CO NOX NOx CO2 Flow Fuel
Test No. FID C.L. Saltzman Standard Flow
gm/mile gm/mile gm/mile gm/mile gin/mile cu. ft. cc
12-1440
12-1442
12-1446
12-1450
12-1444
12-1443
12-1451
12-1453*
.92
1.19
.64
.70
3.73
3.47
2.36
3.36
3.
4.
7.
4.
14.
17.
14.
14.
80
46
88
49
08
26
10
48
SE
.87
1.89
1.03
1.21
SE
2.00
2.10
2.44
2.45
124
.87
1.44
1.48
1.32
101
3.48
2.56
3.44
1.02
1148.
1313.
1322.
1370.
2056.
3059.
2732.
2700.
51
60
74
51
18
57
72
93
8691
8230
8560
8666
26085
23802
25331
25566
No
Data
4023
4360
4743
9418
9301
9451
8399
* Inertia weight = 3500 pounds. Sample bag pulled at 20 minutes of
the LA4 driving schedule.
-------
Table II
GM Steam Cars
Mass Emissions
Hot Start 1972 Federal Test Procedures
Test No.
CVS
HC CO NOX NOx C02 Flow, Fuel
FID C.L. Saltzman standard Flow,
gm/mile gm/mile gm/mile gm/mile gm/mile cu. ft. cc
12-1445 3.29
SE 124
14.77 2.09
2.55
12-1442 1.15 2.33 1.76 1.28 1109.3 7641 3298
• SE 101 '
2853.98 22607 8584
Test No.
12-1442
HC
FID
grams
Table III
GM Steam Cars
Mass Emissions During Warm-up
CO
grams
NOX
C.L.
grams
NOX
Saltzman
grams
CO 2
grams
CVS
Flow, Fuel
standard Flow,
cu. ft. cc
SE 124
.28
15.61
93 1.21 1532.24
SE 101
589
725
12-1445 1.30 18.69 No data No data 1545.67 1199
717
-------
Table IV
GM Steam Cars
Fuel Consumption
Test No.
12-1440
12-1442
12-1442
12-1446
12-1450
12-1444
12-1445
12-1445
12-1451
12-1453*
co2
gm/mile
SE
1148.51
1313.60
Hot 1109.3
1322.74
1370.51
SE
3056.18
3059.57
Hot 2853.98
2732.72
2700.93
Fuel
cc
124
No
Data
4023
3298
4360
4743
101
9418
9301
8584
9451
8399
Miles/gal
(C02)
8.3
7.3
8.6
7.2
7.0
3.1
3.1
3.3
3.5
3.5
Miles/gal
(Fuel)
No Data
7.1
8.6
6.5
6.0
3.0
3.1
3.3
3.0
3.4
* Inertia weight = 3500 pounds. Sample bag pulled at 20
minutes of LA4 driving schedule.
-------
Appendix
Bag Concentrations
GM Steam Cars
Test No.
12-1440
Console
Console
12-1442
Console
Console
12-1442
Console
Console
12-1444
Console
Console
12-1445
Console
Console
12-1445
Console
Console
12-1446
Console
Console
12-1450
Console
Console
12-1451
Console
Console
12-1453
Console
Console
2
1
HC
ppm
48.
56
C
6
CO
ppm
100
100
NOX
ppm
16
104
CO 2
%
1.912
1.75
- Start & warm-up
2
1
_
2
1
2
1
29.
31
Hot LA 4
69
78
65.
93
1
7
800
740
70
80
120
160
33
128
36
96
12
82
5.02
3.70
2.10
1.82
1.695
1.49
- Start & warm-up
2
1
_
2
1
2
1
2
1
2
1
2
1
66.
67
Hot LA 4
66.
83
34.
51
37.
74
41.
70
60.
80
3
9
2
2
7
3
470
510
150
170
210
240
120
150
130
150
130
150
no data
60
14
80
18
84
22.40
no data
15.0
no data
15.4
no data
2.487
2.24
1.826
1.57
2.235
1.82
2.287
2.15
1.56
1.44
1.528
1.47
------- |