RENTAL VEHICLE
    SURVEILLANCE PROGRAM
March  1968 to  January  1970
                •/      /

-------
       RENTAL VEHICLE

  SURVEILLANCE PROGRAM

  March 1968 to January 1970
             Daniel B. Pinkert,
           James M. Marzen, and
             Charles J. Domke
     Bureau of Mobile Source Pollution Control
   ENVIRONMENTAL PROTECTION AGENCY
           Office of Air Programs
     Research Triangle Park, North Carolina
              November 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office
          Washington, D.C., 20402 - Price 25 cents

-------
The AP series  of  reports is issued by  the Environmental Protection Agency to
report the results of scientific and engineering studies, and information of general
interest in the field of air pollution.  Information presented in this series includes
coverage of intramural activities involving air pollution research  and control tech-
nology and of cooperative programs and studies conducted in  conjunction with
state and local agencies,  research institutes,  and industrial  organizations.
Copies  of AP reports are available free of charge - as supplies  permit - from the
Office  of  Technical Information and Publications, Office of Air  Programs,  Envi-
ronmental Protection  Agency, Research Triangle Park, North Carolina  27711-
            Office of Air Programs Publication No. AP-97
                                      11

-------
                        CONTENTS

                                                  Page
LIST OF FIGURES	 iv
LIST OF TABLES	 iv
ABSTRACT 	  v
INTRODUCTION	  1
TEST VEHICLES AND PROCEDURES  	  5
    Test Vehicles	  5
    Exhaust Emission Study Procedures 	  8
RESULTS  	  9
SUMMARY AND CONCLUSIONS  	19
                          in

-------
                        LIST OF FIGURES

Figure                                                        Page
1      Unbumed Hydrocarbon Emissions versus Cumulative Percen-
       tage  of Domestic Cars Represented by Test Sample.  (Test
       Sample Represents 65  Percent of  Combined 1968 Plus 1969
       Production, with Percentages Weighted  According to Produc-
       tion.) 	  17

2      Carbon Monoxide Emissions versus Cumulative  Percentage
       of Domestic Cars Represented by Test Sample. (Test Sample
       Represents 65 Percent of Combined 1968 Plus 1969 Produc-
       tion, with Percentages  Weighted According to Production.) ... 17
                         LIST OF TABLES

Table                                                       Page
1      Test Vehicles and Their Estimated Distribution in Manufac-
       turer and National Vehicle Populations   	   6
2      Mean Emission Concentrations for All Displacement Groups
       with 15 or More Tests	10
3      Mean Emission Concentrations for All Displacement Groups
       with 15  or  More Tests  (Tests on Vehicles  with  Less
       than 3,500 Miles Eliminated)   	  11
4      Percentage of Cars with Emissions Exceeding Standards (For
       All Displacement Groups with 15 or More Tests)	  12
5      Mean Emission Concentrations for All Displacement Groups
       with Less than 15 Tests _	  14
6      Mean Emission Concentrations  for All Displacement Groups
       with Less than 15 Tests (Tests on Vehicles with Less than
       3,500 Miles Eliminated)  	  15
7      Percentage of Cars with Emissions Exceeding Standards (For
       All Displacement Groups with Less than 15 Tests)	  16
8      Data Summary  	  18
                                IV

-------
                              ABSTRACT
    Exhaust-emission tests were performed on rental vehicles representing 26
vehicle types to ascertain  the  effectiveness of the Federal exhaust emission
standards. Most of the rental vehicles tested had odometer readings of less
than 15,000 miles, and had received no engine maintenance other than minor
carburetor  adjustments. Approximately half  the cars with odometer readings
in excess  of 15,000 miles had received one engine tune-up during  their fleet
operation.

    All vehicles were tested by the standard Federal seven-mode, seven-cycle
cold-start  test  procedure, and  by the New Jersey ACID cycle test and the
Clayton Key Mode test.

    Although exhaust emissions varied  considerably among automobiles  of
different manufacturers, among  engines of  different types, and among dif-
ferent  engines of the same type, most types of engines tested produced emis-
sions  at  levels below those  permitted by Federal  standards.  More than 80
percent,  however,  of  the  Chevrolet  307-CID  engines  operated  for  at least
3,500  miles exceeded the Federally established levels. Ford 289- and 302-CID
engines also had high emission levels.

    Approximately  75 percent  of the  test vehicles,  of  types  representing
about  two-thirds  of the  1968 and  1969 domestic automobile population,
emitted hydrocarbons at  levels lower than  those permitted by the Federal
certification standards, and about 60 percent emitted carbon monoxide  at
levels lower than those permitted. Projections indicate, however, that a smaller
percentage of the vehicles represented in this test program will meet certifi-
cation  standards when they have been operated 50,000 miles.

-------
        RENTAL VEHICLE SURVEILLANCE PROGRAM,

                MARCH 1968 TO JANUARY 1970



                          INTRODUCTION
    The nuisance problem and health hazards of automotive air pollution have
received increasing recognition  and study  over  the past decade. The United
States Congress passed the Clean Air Act of 1963 to improve, strengthen, and
accelerate  programs for the prevention  and abatement of air pollution. This
act, as amended in 1965, authorized the United States Department of Health,
Education, and Welfare—under  which the  national air pollution control pro-
gram  was  administered  until late  1970—to  set national standards  for the
control of air  pollution  from new motor  vehicles. Standards  were issued in
early  1966 that established maximum permissible concentrations for hydro-
carbon (HC) and  carbon monoxide (CO) emissions from  automobiles and
light-duty  trucks sold  in the United States,  beginning  with the 1968 model
year. The  Federally established maximum allowable concentrations of HC and
CO emissions  from 1968  and 1969  vehicles, which  vary  according to the
cubic-inch displacement (CID) of the engine,* are listed below:

                   CID           HC, ppm        CO, %
                50 to 100          410             2.3
               100 to 140          350             2.0
                 =»140             275              1.5

    Procedures  for testing and certifying  prototype vehicles for  compliance
with Federal exhaust emission  standards were drafted  and  published in the
Federal Register  31(61), March 30, 1966. These procedures  required  each
manufacturer to build  and test  a specified number of prototype vehicles that
would meet Federal emission standards before  Certificates of Conformity were
issued to  the manufacturer. The procedures did not  require  that each vehicle

*For a 4,000-pound car with an automatic transmission, 275 parts per million
 (ppm) HC is  equivalent to 3.5 grams  HC per mile,  and 1.5 percent CO is
 equivalent to  36  grams  CO per mile. The 1970 standards are 2.2 grams HC
 per mile  and  23 grams CO per mile. The vehicles involved in this test pro-
 gram were not required to meet  the 1970 standards, however, inasmuch as
 they were 1968 and 1969 models.

-------
produced meet the applicable  standards. They did require  that  the  average
emission values from a sample  of two or four emission-data vehicles of each
displacement, depending  on projected sales volume,  be less than the values
stated in the standards, when the values had been modified by a deterioration
factor obtained from a fleet of four to ten durability-data vehicles, represen-
tative  of at  least  70 percent of  the  manufacturer's engine displacement and
transmission  options.

    The Federal  procedures  and standards  applicable to  1968- and 1969-
model vehicles  were similar to  those of California,  which became effective
with  1966 models built  for sale in that state. After the 1966 models were
introduced for  sale, the  California  authorities,  with Federal  financial  assis-
tance, initiated a program by which to survey the effectiveness of the certified
emission-control systems  on  production, as  opposed to prototype, vehicles
operated routinely by the ultimate consumer. Periodic  reports resulting  from
this  continuing program  indicated that emissions from the  1966 models de-
viated substantially from  the standards, especially with increasing mileage. The
performance  for  1967  and  subsequent  model-year vehicles  was  generally
better. The results reported by California were obtained from "hot start" tests
corrected to  "cold start" values  by means of average adjustment factors that
had been determined experimentally for  carbon monoxide and hydrocarbons.

    The purpose of the Rental Vehicle Survey was to ascertain the effective-
ness of the  Federal air  pollution  regulations by  measuring emissions  from
vehicles driven by the motoring  public. Between March 1, 1968, and January
13, 1970,  705  exhaust-emission tests  were performed by the  Environmental
Protection  Agency (EPA) at the Los Angeles, California, and Ypsilanti, Mich-
igan,  test facilities of EPA's  Bureau  of  Mobile Source Pollution  Control on
rental  vehicles  representing 26  different vehicle types. Besides  being used to
give  an indication  of the  exhaust-emission levels of vehicles  in  general use,
these tests were designed  to provide the background information necessary for
efficiently planning future surveillance programs for privately owned vehicles.

    Vehicles for  this  program  were  obtained from  the  Hertz and Airways
rental  companies  in Los  Angeles and from the Hertz and Avis companies in
Detroit. Although  such rental vehicles may differ  from privately  owned ve-
hicles  with respect to the care with which they are  driven,  the maintenance
they  receive, and  the  type and rate of mileage  accumulated,  no data are
available that define such differences. Lacking data to the contrary, there  is
no reason to believe that  the two groups have different exhaust-emission  char-
acteristics.

    The original plan for this program was to start with a basic  fleet  of 138
vehicles, 1968-model year, that  would be retested at 3,000- to 4,000-mile
intervals. These vehicles were chosen to represent many of the high-production
vehicles sold  in the United States. Within a short period of time,  however,  it
became apparent that the rental  companies could not deliver the vehicles for
repetitive testing at the required intervals. In  addition, policy changes  by the
                              RENTAL VEHICLE SURVEILLANCE PROGRAM

-------
rental companies resulted in the retention of most rental vehicles for less than
1  year  of operation, thereby  eliminating any possibility of obtaining large
samples of vehicles with  more than  20,000 accumulated miles.  Hence the
surveillance program had to be reorganized to reflect non-repetitive testing  of
vehicles and the  inclusion of  low-mileage,  1969-model  vehicles in  the test
fleet.
Introduction

-------
                  TEST VEHICLES AND PROCEDURES
TEST VEHICLES

    The vehicles  tested in this  program are listed in Table 1.  All vehicles,
except Volkswagens, were equipped with automatic transmissions, and all ve-
hicles except Cadillacs had engine-modification emission-control systems. The
Cadillacs used an air-injection emission-control system. Most engines of a given
make and  displacement constituted a homogeneous population  inasmuch as
only  one  version  was  tested. Thus, even  though  Ford Motor Company's
390-CID engine was produced in both  two-barrel- and four-barrel-carburetor
versions, data  were  obtained with the two-barrel model only. The 1968 Chev-
rolet 327-CID engines differed from the  same 1969 engines in carburetion and
in compression ratio. The  difference between 1968  and 1969 versions of all
other engines was considered small enough to allow data from  the two model
years to be combined. Because the 350-CID engines  used by General Motors'
Buick, Chevrolet,  Oldsmobile, and  Pontiac  Divisions differed  in design,  a
separate analysis was needed for each division's engines.

    All test results  for a given engine type were combined without regard for
body style. Thus, all test results on Ford Motor Company's 302-CID engines
were combined even though the test fleet for this engine included a mix of
Mercury Cougars,  Ford  Galaxies, and  Ford  Mustangs. Similarly, data on
Chrysler Corporation's 318-CID  engine  were  obtained by testing both Ply-
mouths and Dodges.

    The number of 1968- and 1969-model-year, and 1968- plus  1969-model-
year cars tested in  each displacement group is  given  in Table 1 along with
estimates of the proportionate use of a  particular engine in  the total popula-
tion of domestic cars. These population estimates are  described  below:

      Column heading                       Meaning
    % of manufacturer's      The percentage of the manufacturer's out-
    output                  put that uses the particular engine type.

    % of total vehicle        The percentage of the  total car population
    population              under consideration that  uses the particular
                            engine type.

    All three  rental-car companies received  their vehicles from local new-car
dealers and not directly from the manufacturers. They did not  usually rely on
one dealer for all vehicles of a certain model, but received vehicles from many

-------
                       Table 1.  TEST VEHICLES AND THEIR ESTIMATED DISTRIBUTION IN MANUFACTURER

                                        AND NATIONAL VEHICLE POPULATIONS
Manufacturer
Domestic
American Motors
Corporation
Total
Chrysler
Corporation


Total
Ford Motor
Company






Total
General Motors
Corporation
Chevrolet

CID

290
343

225
318
383
440

200
289
302
351
390
429
460
462



230
307
Carbu-
retor
barrels

2
2

1
2
2
4

1
2
2
2
2
4
4
4



1
2
1968
Number
tested3

19
8
27
0
40
2
0
42
0
45
78
0
56
8
1
5
193


0
24
% of
mfr. output

28.75
7.19
35.94
15.74
33.89
23.32
9.58
57.21
7.40
13.87
32.34
0.00
19.96
3.40
1.23
0.89
71.69


2.39
17.99
% of total
U.S. vehicles

0.92
0.23

2.99
6.44
4.43
1.82

1.74
3.26
7.60
0.00
4.69
0.80
0.29
0.21



1.30
9.77
1969
Number
tested3

13
4
17
21
9
5
3
38
11
0
32
13
62
0
4
0
122


5
24
% of
mfr. output

29.70
12.73
42.43
10.40
34.09
21.20
9.96
75.65
4.16
0.00
32.36
13.54
24.54
4.71
2.88
0.00
77.48


2.65
6.10
% of total
U. S. vehicles

0.98
0.42

1.88
6.16
3.83
1.80

1.07
0.00
8.32
3.48
6.31
1.21
0.74
0.00



1.40
3.23
1968 plus 1969
Number
tested3

32
12
44
21
49
7
3
80
11
45
110
13
118
8
5
5
315


5
48
% of
mfr. output

29.23
9.85
39.08
13.17
34.00
22.29
9.77
79.23
5.69
6.62
32.34
7.07
22.35
4.06
2.11
0.41
80.65


2.52
12.12
% of tota I
U. S. vehicles

0.95
0.32

2.44
6.30
4.13
1.81

1.40
1.63
7.96
1.74
5.50
1.00
0.52
0.10



1.35
6.50
m


O
GO



I
O
m

-a
x
O
O
•30

-------
                    Table 1 (continued). TEST VEHICLES AND THEIR ESTIMATED DISTRIBUTION IN MANUFACTURER
                                            AND NATIONAL VEHICLE POPULATIONS
CD
CD
(A

fi)

Q_
O
CD
Q.

^
CD
V)



Manufacturer
Domestic
General Motors
Corporation
Chevrolet
(continued)

Pontiac


Oldsmobile

Cadillac
Total
Total domestic
Foreign
Volkswagen
Grand Total



CID



327
327
350
350
400
428
350
455
472



92


Carbu-
retor
barrels



4
2
4
2
4
4
2
4
4



1

1968

Number
tested3



48
0
0
14
6
0
0
8
8
108
370

0
370
%of
mfr. output



6.17
0.00
0.00
5.75
6.17
0.00
5.06
2.87
4.97
43.92


-

% of total
U.S. vehicles



3.35
0.00
0.00
3.12
3.35
0.00
2.75
1.56
2.70

63.32

-

1969

Number
tested8



0
53
5
11
17
4
4
0
0
123
300

35
335
%of
mfr. output



0.00
13.83
6.90
5.16
4.04
2.80
5.57
7.14
5.06
47.05


-

% of total
U.S. vehicles



0.00
7.32
3.65
2.73
2.14
1.48
2.95
3.78
2.68

67.56

-

1968 plus 1969

Number
tested8



48
53
5
25
23
4
4
8
8
231
670

35
705
%of
mfr. output



3.13
6.83
3.39
5.45
5.11
1.38
5.32
4.98
5.02
55.25


-

% of tola I
U.S. vehicles



1.68
3.66
1.82
2.92
2.74
0.74
2.85
2.67
2.69

65.44

-

          3lf "Number tested" was zero, production percentage was excluded from totals.

-------
different dealers. All vehicles reportedly received normal dealer make-ready
servicing before delivery to the rental-car companies.

    Although  there were  significant differences in the maintenance practices
of the  three rental-car  companies used as vehicle sources for this program,
some  generalizations can be made about maintenance of the test vehicles. The
great  majority of vehicles tested in this program with odometer readings of
15,000  miles or  less had no engine maintenance other than minor carburetor
adjustments performed  on them between new-car dealer preparation and the
time  of test.  Approximately half of the cars  with  greater than 15,000  ac-
cumulated  miles had their  engines  tuned once during their  fleet  operation
prior to testing.

EXHAUST EMISSION STUDY PROCEDURES

    All vehicles  were tested by  the standard Federal  seven-mode, seven-cycle
cold-start test  procedure as described in the Federal Register 31^(61), II, Para-
graphs 85.70-85.83. Following each cold-start test, a New Jersey ACID Cycle
test and a Clayton Key Mode test were performed to determine the feasibility
of using short exhaust-emission tests as  predictors of the Federal seven-mode,
cold-start emission test. Data from these short  tests are being  used in a study
of inspection procedures and are not presented in this report.
                             RENTAL VEHICLE SURVEILLANCE PROGRAM

-------
                               RESULTS

    Table 2 presents the mean  exhaust-emission levels for those 12 displace-
ment  groups  for  which data from 15 or more vehicles were  obtained. The
table  includes the  average odometer reading and the averages of the three
lowest and of the three highest  odometer readings of the vehicles tested; the
last two values give the range of mileage for the vehicles tested. The average
exhaust emissions are given as arithmetic mean values.

    Analysis  of the exhaust-emission data for the individual engine-displace-
ment  groups  revealed  that the  hydrocarbon and carbon  monoxide  concen-
trations were  lognormally rather than  normally  distributed;  that  is, the
logarithms  of the  values  closely approximated  a  normal  distribution. The
geometric  mean is  the value corresponding to the average of the logarithms;
the HC and CO geometric mean concentrations are  listed in Table 2. (For a
group of values whose logarithms  are normally distributed, half of the values
will be above  and half below  the geometric mean.  The  probability is  95
percent that  the  true value of the  geometric mean  is  between  the  lower
confidence limit (LCL) and the upper confidence limit (UCL).)

    Table 3 presents the mean exhaust-emission levels for  those vehicles that
had accumulated  at least 3,500 miles. Although tests were conducted on a
number of vehicles with lower mileage to develop some background-emissions
information  on engines not  fully  broken  in, data from these tests were
eliminated from this analysis  for  two reasons. First, emissions produced  by
vehicles with  low  mileage—before piston  rings have seated and equilibrium
combustion chamber deposits  have accumulated—were suspected to be erratic
and to  change rapidly as  mileage was accumulated; thus  they could not  be
used  to develop  real-life data  that could be compared  with  the 4,000-  to
50,000-mile  values, modified by deterioration factors, that  were determined
from  certification prototypes. Second, some displacement groups contained
vehicles with  low  mileage and  others  did not, and the elimination of the
low-mileage tests  tended to equalize the  average odometer readings for each
group, making comparisons of emission data among  groups more meaningful.

    In general, the  average HC emissions were somewhat higher  for the groups
of cars with  more  than 3,500  miles than for the corresponding groups con-
taining  all  cars tested, regardless  of odometer readings.  The  differences  in
average CO values were slight.

    Table 4 presents the percentage of cars in each displacement group, from
which 15 or more cars were tested, that produced HC and/or CO emissions in
excess of  the  limits prescribed  by Federal standards.  Results  obtained after
eliminating data from cars with fewer than 3,500 miles are also presented.

-------
                 Table 2.  MEAN EMISSION CONCENTRATIONS FOR ALL DISPLACEMENT GROUPS WITH 15 OR MORE TESTS
Manufacturer
American Motors
Corporation
Chrysler
Corporation
Ford Motor
Company
General Motors
Corporation
Chevrolet
1968
1969
Pontiac
Volkswagen
CID
290
225
318
289
302
390

307
327
327
350
400
92
Mumber
tested
32
21
49
45
110
118

48
48
53
25
23
35
Odometer readings
Average
5,133
13,568
4,991
10,994
8,953
3,044

9,548
16,960
5,611
9,631
6,492
9,914
Average of
minimum 3
263
4,272
34
4,050
940
34

1,088
6,525
44
486
1,550
53
Average of
maximum 3
10,709
24,086
14,265
21,397
21,821
24,199

20,311
25,991
22,391
17,187
12,042
18,364
Hydrocarbon emissions, ppm
Arithmetic
mean
213
215
230
320
305
228

325
289
231
219
176
372
Geometric
mean
211
212
226
304
294
213

299
282
226
215
173
359
LCLa
199
195
214
279
279
201

268
264
213
198
159
328
UCLb
223
229
239
332
308
226

334
300
239
233
188
393
Carbon monoxide emissions, %
Arithmetic
mean
1.13
1.70
1.09
1.39
1.19
1.11

1.52
1.76
1.56
1.55
1.46
2.14
Geometric
mean
1.06
1.56
1.01
1.22
1.04
0.89

1.40
1.66
1.46
1.42
1.40
1.97
LCLa
0.91
1.29
0.91
1.06
0.95
0.80

1.23
1.49
1.32
1.17
1.23
1.74
UCLb
1.23
1.89
1.13
1.41
1.14
0.99

1.60
1.84
1.62
1.72
1.59
2.23
m
I

O
C/5


30
>


O
-o
30
O
O
30
aLCL = Lower confidence limit of 95-percent confidence

 UCL : Upper confidence limit of 95-percent confidence
interval on geometric mean.

interval on geometric mean.

-------
30
CD
(A
C
Table 3.  MEAN EMISSION CONCENTRATIONS FOR ALL DISPLACEMENT GROUPS WITH 15 OR MORE TESTS

                   (Tests on vehicles with less than 3,500 miles eliminated)
Manufacturer
American Motors
Corporation
Chrysler
Corporation
Ford Motor
Company

General Motors
Corporation
Chevrolet
1968
1969
Pontiac

Volkswagen
CID
290

225
318-
289
302
390


307
327
327
350
400
92
Number
tested
24

20
28
44
95
72


36
48
32
21
21
30
Odometer readings
Average
6,157

14,178
8,125
11,191
10,038
12,461


12,040
16,960
8,354
11,236
7,061
1 1 ,537
Average of
minimumS
3,690

6,739
3,661
4,902
3,827
4,384


4,773
6,525
3,817
5,536
3,703
5,531
Average of
maximum 3
10,709

24,086
14,265
21 ,397
21 ,821
24,199


20,311
25,991
22,391
17,187
12,042
18,364
Hydrocarbon emissions, ppm
Arithmetic
mean
218

215
240
323
314
252


355
289
251
215
177
391
Geometric
mean
216

212
234
308
303
232


328
282
248
211
173
380
Lcia
206

195
214
283
288
212


288
264
233
193
158
349
UCLb
228

230
256
336
318
253


372
300
263
230
190
414
Carbon monoxide emissions, %
Arithmetic
mean
1.12

1.71
1.18
1.40
1.23
1.36


1.56
1.76
1.59
1.61
1.50
2.17
Geometric
mean
1.05

1.56
1.06
1.23
1.08
1.04


1.43
1.66
1.50
1.50
1.44
1.97
LCLa
0.88

1.28
0.89
1.06
0.98
0.89


1.22
1.49
1.32
1.25
1.26
1.71
UCLb
1.26

1.91
1.26
1.42
1.19
1.22


1.68
1.84
1.70
1.80
1.64
2.28
           aLCL • lower confidence limit of 95 percent confidence interval on geometric mean.

            UCL :Upper confidence limit of 95 percent confidence interval on geometric mean.

-------
                             Table 4.  PERCENTAGE OF CARS WITH EMISSIONS EXCEEDING STANDARDS

                                       (For all displacement groups with 15 or more tests)
m


O
CO


30
O
m

-o
33
O
Manufacturer
American Motors
Corporation
Chrysler
Corporation
Ford Motor
Company
General Motors
Corporation
Chevrolet
1968
1969
Pontiac
Volkswagen
CID
290

225
318
289
302
390


307
327
327
350
400
92
Tests on all cars
Number
tested
32

21
49
45
110
118


48
48
53
25
23
35
Average
odometer
reading
5,133

13,568
4,991
10,994
8,953
3,044


9,548
16,960
5,611
9,631
6,492
9,914
Cars with emissions
over standard, %
HC
0

14
12
62
58
16


60
42
10
16
0
31
CO
13

48
10
22
19
15


54
58
47
56
48
23
Both
0

14
4
20
19
8


42
29
4
8
0
9
Either
13

48
18
64
58
23


72
71
53
64
48
45
Tests on cars with more than 3,500 miles
Number
tested
24

20
28
44
95
72


36
48
32
21
21
30
Average
odometer
reading
6,157

14,178
8,125
11,191
10,038
12,461


12,040
16,960
8,354
1 1 ,236
7,061
1 1 ,537
Cars with emissions
over standard, %
HC
0

15
18
62
63
25


75
42
16
14
0
37
CO
8

45
14
23
20
24


56
58
44
62
52
20
Both
0

15
7
20
20
13


47
29
6
10
0
10
Either
8

45
25
65
63
36


84
71
54
66
52
47

-------
    The average  Ford 289- and 302-CID engines  and the average  Chevrolet
307-CID engines had  HC emissions considerably  in excess of Federal stan-
dards. The average  1968 Chevrolet 327-CID engine had HC emissions slightly
higher than the  standards. Of the remaining displacement  groups (with 15 or
more test vehicles), all had average HC values less  than the limits imposed by
the Federal standards.

    Although several of the displacement groups  had average  CO  emissions
slightly  higher than the  standards,  none of the  averages was significantly
greater than the  standard.

    Results for the 14 engine-displacement groups represented by fewer than
15 test vehicles are  presented in Tables 5 through 7. A sample smaller than 15
was considered too  small  to allow meaningful statements about average emis-
sions for the displacement group because the effect of random fluctuations on
data from such a small sample could be relatively large.

    Estimates of the exhaust emissions of the population of domestic vehicles
were made by combining  exhaust-emission data  and the estimated production
percentages shown  in Table  1.  Foreign vehicles were excluded because only
Volkswagen data were  obtained and the Volkswagen, because of its small
engine size, was  subject to emission standards that were different from those
applicable to domestic vehicles. The distribution of HC and CO emissions is
shown  respectively  in  Figures  1  and 2. Similar plots were made after data
obtained from vehicles  with fewer than 3,500 miles were eliminated,  but these
are not  presented because they were almost identical in appearance to Figures
1 and 2. The data may be summarized as shown  in Table 8.

    As these data show, more  than half of the vehicle population represented
by the test sample  had exhaust emissions, at the mileage levels of the vehicles
sampled, below the  limits prescribed in the Federal standards.

    An  attempt  was  made to  estimate  the  change in  emission levels with
mileage  for each displacement  group. The sample,  however, was  not well
suited to  this type of analysis because  most  of  the cars had  accumulated
relatively few miles, ranging from a low of 29  to  a high of 27,493, with an
average  of only  7,708 miles. Past experience indicates, however, that vehicle
emissions  increase with mileage accumulation.  It  is  reasonable to  conclude,
therefore, that although this study  provides no basis for predicting emission
levels at 50,000 miles, many  of the vehicles will  have higher emission levels
when they  have been driven 50,000 miles.
Results                                                                13

-------
                Table 5.  MEAN EMISSION CONCENTRATIONS FOR ALL DISPLACEMENT GROUPS WITH LESS THAN 15 TESTS
Manufacturer
American Motors
Corporation
Chrysler
Corporation
Ford Motor
Company



General Motors
Corporation
Chevrolet

Pontiac
Oldsmobile

Cadillac
CID
343

383
440
200
351
429
460
462


230
350
428
350
455
472
Number
tested
12

7
3
11
13
8
5
5


5
5
4
4
8
8
Odometer readings
Average
3,846

3,050
3,883
741
4,397
11,013
4,221
11,910


5,060
12,704
5,089
4,155
13,255
6,300
Average of
minimum 2
1,593

478
2,873
32
1,275
6,129
3,735
6,936


3,368
3,827
2,328
2,377
10,208
652
Average of
maximum 2
6,130

5,096
4,704
3,156
6,904
15,806
4,813
15,757


7,303
21 ,628
7,852
5,934
16,234
11,494
Hydrocarbon emissions, ppm
Arithmetic
mean
203

212
217
212
210
215
186
203


185
214
324
187
138
146
Geometric
mean
201

207
214
211
209
196
178
197


183
212
319
186
137
145
LCLa
184

167
135
195
195
139
116
140


149
182
228
168
126
130
UCLb
220

257
341
228
224
276
273
278


225
248
448
206
149
163
Carbon monoxide emissions, %
Arithmetic
mean
0.94

1.17
0.89
1.07
0.83
1.76
1.42
1.51


1.37
1.95
2.72
2.34
2.09
1.08
Geometric
mean
0.90

1.08
0.88
1.05
0.74
1.60
1.28
1.50


1.32
1.93
2.68
2.31
2.07
1.08
LCLa
0.74

0.73
0.44
0.90
0.54
1.08
0.69
1.23


0.88
1.51
1.93
1.66
1.76
0.98
UCLb
1.10

1.61
1.76
1.35
1.00
2.36
2.36
1.83


1.98
2.46
3.73
3.23
3.10
1.19
m
I
CO


33
O
m

-o
30
O
O
•33
al_CL ; Lower confidence limit


bUCL : Upper confidence limit
of 95-percent confidence interval


of 95-percent confidence interval
on geometric mean.


on geometric mean.

-------
i?
en
Table 6.  MEAN EMISSION CONCENTRATIONS FOR ALL DISPLACEMENT GROUPS WITH LESS THAN 15 TESTS
                        (Tests on vehicles with less than 3,500 miles eliminated)
Manufacturer
American Motors
Corporation
Chrysler
Corporation

Ford Motor
Company



General Motors
Corporation
Chevrolet

Pontiac
Oldsmobile

Cadillac
CID
343

383
440

200
351
429
460
462


230
350
428
350
455
472
Number
tested
8

4
2

1
9
8
5
4


4
4
3
3
8
6
Odometer readings
Average
4,827

4,360
4,704

4,903
5,627
11,013
4,221
14,169


5,628
15,124
6,498
5,323
13,255
8,183
Average of
minimum 2
4,039

3,625
4,704

4,903
4,085
6,129
3,735
12,582


3,954
8,622
5,700
4,135
10,208
4,974
Average of
maximum 2
6,130

.5,096
4,704

4,903
6,904
15,806
4,813
15,757


7,303
21 ,628
7,852
5,934
16,234
11,494
Hydrocarbon emissions, ppm
Arithmetic
mean
215

224
210

267
213
215
186
201


187
221
355
182
138
155
Geometric
mean
214

222
206

267
212
196
178
194


184
220
355
182
137
154
LCia
192

171
22

—
194
139
116
117


136
183
327
161
126
140
UCLb
238

287
1,928

—
231
276
273
322


249
263
385
205
149
170
Carbon monoxide emissions, %
Arithmetic
mean
1.03

1.15
0.94

1.10
0.88
1.76
1.42
1.52


1.42
1.97
2.96
2.20
2.09
1.09
Geometric
mean
0.99

1.07
0.92

1.10
0.83
1.60
1.28
1.50


1.35
1.93
2.96
2.17
2.07
1.09
LCLa
0.76

0.56
0.03

—
0.64
1.08
0.69
1.12


0.75
1.35
2.40
1.31
1.76
0.95
UCLb
1.29

2.07
27.90

_
1.09
2.36
2.36
2.02


2.44
2.78
3.65
3.60
2.43
1.25
            LCL - Lower confidence limit of 95-percent confidence interval on geometric mean.
           bUCL r Upper confidence limit of 95-percent confidence interval on geometric mean.

-------
                            Table 7. PERCENTAGE OF CARS WITH EMISSIONS EXCEEDING STANDARDS
                                     (For all displacement groups with less than 15 tests)
m
I
o
oo

30
m
>
o
30
O
Manufacturer
American Motors
Corporation
Chrysler
Corporation
Ford Motor
Company



General Motors
Corporation
Chevrolet

Pontiac
Oldsmobile

Cadillac
CID
343

383
440
200
351
429
460
462


230
350
428
350
455
472
All tests
Number
tested
12

7
3
11
13
8
5
5


5
5
4
4
8
8
Average
odometer
reading
3,846

3,050
3,883
741
4,397
11,013
4,221
11,910


5,060
12,704
5,089
4,155
13,255
6,300
Cars with emissions
over standard, %
HC
0

14
0
0
0
13
0
0


0
0
76
0
0
0
CO
0

28
0
0
8
63
40
40


20
100
100
100
100
0
Both
0

0
0
0
0
13
0
0


0
0
76
0
0
0
Either
0

42
0
0
8
63
40
40


20
100
100
100
100
0
Tests on cars with more than 3,500 miles
Number
tested
8

4
2
1
9
8
5
4


4
4
3
3
8
6
Average
odometer
reading
4,827

4,360
4,704
4,903
5,627
11,013
4,221
14,169


5,628
15,124
6,498
5,323
13,255
8,183
Cars with emissions
over standard, %
HC
0

25
0
0
0
13
0
0


0
0
100
0
0
0
CO
0

25
0
0
11
63
40
50


25
100
100
100
100
0
Both
0

0
0
0
0
13
0
0


0
0
100
0
0
0
Either
0

50
0
0
11
63
40
50


25
100
100
100
100
0

-------
30

-------
                     Tables. DATA SUMMARY




Description of data
Cumulative % of
vehicle population
50.0
78.0
62.4
73.4
61.2
Production-weighted
means
Arithmetic mean
Antilog of mean log

Emissions from all
cars
HC,
ppm


216
275a
—
-
-


236
223
CO,
°/0


1.30
—
1.50a
-
-


1.41
1.25
Emissions from cars
with odometer
readings »3, 500 miles
HC,
ppm


226
—
-
275a
-


247
232
CO,
%


1.20
—
-
-
1.50a


1.45
1.29
     aFederal standard.
18
RENTAL VEHICLE SURVEILLANCE PROGRAM

-------
                  SUMMARY AND CONCLUSIONS
    Results  of 705  exhaust-emission tests  conducted on 1968- and 1969-
model-year rental vehicles by the Environmental Protection Agency's Bureau
of Mobile Source Pollution Control revealed the following:

    1.  Exhaust emission levels varied considerably among automobiles of
        different  manufacturers,  among  automobiles with different  engine
        types, and among different engines  of the same  type. Most types of
        engines had  emission levels below the limits imposed by Federal cer-
        tification requirements.

    2.  The  types  of vehicles sampled in this survey represent about two-
        thirds of the  population of 1968  and 1969  domestic  automobiles.
        Approximately 75  percent  of this population was projected to have
        hydrocarbon emissions that were less  than  the  levels prescribed by
        the Federal certification standards, and approximately 60 percent had
        carbon monoxide emissions that were less than the prescribed levels,
        as determined by weighting test-vehicle percentages in proportion to
        production  of  that type vehicle. No  statement can be made con-
        cerning compliance with standards by that portion of the population
        not represented in the test sample.

    3.  Some engines produced by Ford and Chevrolet had particularly high
        emission levels. More than  80 percent  of the vehicles equipped with
        Chevrolet 307-CID engines  that had been driven at least 3,500 miles
        produced either hydrocarbon or carbon monoxide emissions in excess
        of Federal  certification levels.  The  Ford 289- and 302-CID engines
        also had high  emission levels. These engines  were used in a relatively
        high percentage of the vehicles produced (estimated at  over 16  per-
        cent of  total  1968  and  1969 domestic production,  and over  24
        percent  of  the population represented  by vehicles tested in  this
        study).

    4.  Vehicles in  the test  sample had accumulated relatively few miles,
        ranging from a low of 29 to a high of 27,493 miles, with an average
        of 7,708 miles. Past  experience with the real-life vehicle population
        indicates  that   emissions increase with mileage  accumulation. It  is
        reasonable,  therefore, to project that a smaller percentage of the
        vehicles  represented in this test program will have exhaust emission
        levels below the Federal limits when they have been operated 50,000
        miles.
                                   19

-------