EPA-AA-IMS/81-1
Recommendations Regarding the
Selection of Idle Emission Inspection Cutpoints
for Inspection and Maintenance Programs
by
Douglas Noddings
Inspection/Maintenance Staff
January, 1981
Notice
Technical Reports do not necessarily represent final EPA decisions or posi-
tions. They are intended to present technical analysis of issues using data
which are currently available. The purpose in the release of such reports is
to facilitate the exchange of technical information and to inform the public
of technical developments which may form the basis for a final EPA decision,
position or regulatory action.
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air, Noise, and Radiation
U.S. Environmental Protection Agency
-------
Preface
This report, which was developed as part of EPA's Model Program Guidance,
addresses idle HC and CO cutpoints and their resulting failure rates in an
Inspection/Maintenance (I/M) program. Recommended cutpoints have been
developed for various desired failure rates both at the beginning of I/M and
after a year of the program. The analysis applies to I/M programs beginning
either 1/1/82 or 1/1/83.
The author acknowledges David Hughes of the EPA Inspection and Maintenance
Staff, who contributed the section on 1981 and later vehicles.
-------
I. Background
Careful attention should be given when selecting idle emission standards
(outpoints) as this choice will effect several factors of an I/M program. The
cutpoints used will determine how many vehicles fail the emission test which
in turn will determine the cost or inconvenience to consumers for maintenance
as well as the additional capacity needed to reinspect those failed vehicles.
With stricter cutpoints, more vehicles will be maintained and the repairs made
will result in higher emission reduction benefits. There is also some
potential for cutpoints to influence the balance between hydrocarbon and
carbon monoxide reductions. Not all I/M programs will be in areas requiring
reductions of both Ozone and CO. Some programs may consider selecting
cutpoints which would achieve acceptable emission reduction benefits for only
one of these pollutants by failing most or all vehicles for that pollutant
(e.g. HC for ozone) while keeping the total number of failed vehicles con-
stant. For all of these reasons, simply adopting another area's cutpoints may
not be in the best interest of the I/M program if more effective cutpoints can
be found to fulfill the program's emission reduction needs.
-------
II. Discussion
A. Policy Considerations and Recommendations In Selecting Outpoints
Certain policy decisions must first be made by each I/M program before the
technical problem of cutpoint selection can be dealt with. It must be deter-
mined which groups of vehicles will be classified into the same cutpoint
categories subject to identical inspection standards. The desired fraction of
the vehicles to be failed both overall and within each cutpoint category is of
principal importance. Finally, it must be decided what proportion of these
failed vehicles are failed for HC emissions and/or CO emissions. An I/M
program could fail all vehicles only for HC or only for CO or distribute the
failures in several ways among both emissions with the same overall failure
rate.
These policy decisions will affect the emission reduction benefits derived
from I/M. Clearly there is a direct relationship between the overall failure
rate and the emissions reduction benefits which will be achieved. For a given
failure rate these benefits will vary among different emission control tech-
nologies represented by different cutpoint categories. Whether vehicles are
failed for HC or CO matters because different emission problems will be
identified and emission repairs will be performed so that vehicles will be
able to pass different cutpoints.
Cutpoint categories may be defined in terms of a single vehicle characteristic
or in terms of a unique combination of several characteristics. Characteris-
tics considered in different I/M programs now in operation, include model
year, model year groups based on similarities in technology, number of
cylinders, presence of an air pump or catalyst, and the make and model of each
vehicle.
The cutpoint categories which this report suggests are defined in terms of
model year groups based on similarities in the federal emission standards they
were designed to meet and/or emission control technologies. These suggested
cutpoint categories are pre-1968, 1968-1971, 1972-1974, 1975-1979, 1980, and
post-1980. It is believed that a more complex structure of cutpoint cate-
gories does not warrant the necessary additional administrative burden. Also,
model year failure rates within these model year cutpoint categories (as
illustrated in Table 3 and 4, below) do not vary much indicating that finer
model year divisions are not necessary .JL/ With these model year group
cutpoint categories, an I/M program can be assured that virtually all failed
vehicles will exceed federal emission standards and will benefit from repair.
I/ MOBILE2 assumes equal failure rates for each model year. Equal cutpoints
within technology groups with equal failure rates among groups is close enough
to the standard assumption that MOBILE2 may be used. Widely varying failure
rates among model year groups require special analysis.
-------
Among pre-1981 model year groups, it is suggested that outpoints be selected
which will result in a failure rate of 35% within each group. This is
considered the most effective failure rate for I/M within reasonable limits.
This report presents cutpoints for a 35% failure rate and also for other
reasonable 1/M failure rates.
In the selection of the ratio of HC to CO I/M failures, careful attention must
be paid to several factors. As mentioned above, there are many pairs of HC/CO
cutpoints which will result in the same total failure rate within a cutpoint
category.
There are different approaches to selecting the HC/CO ratio of I/M failures.
If a specific pair of HC/CO cutpoints are selected directly, this will
determine the HC/CO ratio of failures for a given vehicle population and also
the total failure rate. If these results are judged to be acceptable, the
cutpoint selection process is completed. If not, further trial-and-error is
needed. There are also two other approaches which are more systematic. One
is to decide first what the total failure rate, the HC-only failure rate, and
the CO failure rate will be (the total failure rate is the sum of the HC-only
and CO failure rates); once this decision is made, actual cutpoints that will
result in this total failure rate and ratio of HC/CO failures can systematic-
ally be selected for any given sample of vehicles' idle scores. A computer
program such as the one contained in the Appendix of this report can be used
for this final step. The other systematic approach to selecting the HC/CO
ratio of failures is to specify first an HC/CO cutpoint relationship, e.g.,
"HC cutpoint = CO cutpoint x (100 ppm/1%)", and then specify the desired total
failure rate. Once these decisions are. made, the unique pair of HC/CO
cutpoints which satisfy the relationship and result in the desired total
failure rate can be found systematically, again using a computer program. EPA
prefers this last approach to selecting the HC/CO distribution; the remainder
of this report uses it.
In selecting a specific recommended cutpoint locus (HC/CO pairs) for each of
the model year groups listed earlier from the great number of alternatives,
several factors were considered. Changes in the percentage of excess FTP HC
and CO emissions identified in failed vehicles resulting from moderate shifts
in HC/CO locus emphasis are negligible. Even shifts to the extreme cases of
failing vehicles only for HC or only for CO have little impact as illustrated
in Figure 1. Repair effectiveness, however, also determines emission
reduction, so excess FTP emissions identified is only a partial indication of
the relative emission reductions which would be observed when using such
extreme HC/CO cutpoint emphasis. EPA has found that cutpoint sets which
emphasize HC are accompanied by high rates of errors of commission (failed
vehicles which meet federal emission standards). Due to the resulting high
rate of errors of commission and the uncertain though apparently small benefit
of extreme HC cutpoint emphasis, it is suggested that I/M programs avoid using
extreme HC emphasis cutpoint loci. Also because of uncertain benefits I/M
programs which need HC reduction benefits should avoid extreme CO emphasis
loci.
-------
Figure 1
FTP
EXCESS
EMISSIONS
EXCESS FTP EMISSIONS IDENTIFIED
flT CONSTflNT FfllLURE RflTE UITH
DIFFERENT HC/CO OUTPOINT EMPHPSIS
C035
C020
I-
Fail Vehicles
Only for CO
IDLE HC/CO OUTPOINT EMPHflSIS
Fail Vehicles
Only for HC
-------
EPA recommends two specific HC/CO cutpoint loci for pre-1981 vehicles, one for
pre-1968 vehicles and the other for 1968-1980 vehicles. Suggested loci are
listed in Table 1 and illustrated in Figure 2. These loci are believed to be
reasonable, moderate, and administratively convenient. They are suitable for
I/M programs needing HC only or both HC and CO reductions. Advice for I/M
programs requiring only CO reductions will follow this report. I/M benefits
predicted by MOBILE2 are based upon these loci. Special help is available for
states who wish to depart significantly from them. The 207(b) standards of
220 ppm HC and 1.2% CO are recommended for use in all I/M programs for 1981
and later model year vehicles.
\J Excess FTP emissions, are defined to be emissions above the FTP standard.
The percent identified in the fraction of all excess FTP emissions in the
vehicle fleet which are accounted for by vehicles failing the short test. The
percentage of excess emissions identified is a rough indicator of likely
emission reductions possible. If only a small proportion of the fleet excess
emissions are identified there is probably not much potential for emission
reductions.
-------
Table 1
Suggested HC/CO Cutpoint Loci (Pairs) For Pre-1981
Model Year Vehicles
Model Year
Pre-1968
HC (ppm)
200
225
250
275
300
350
400
450
500
550
600
650
700
750
800
1400
1467
1533
1600
CO (%)
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
1968-1980
HC (ppm)
200
225
250
275
300
350
400
450
500
550
600
650
700
750
800
850
900
950
1000
CO (%)
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
-------
Figure 2
EPfl RECOMMENDED HC/CO CUTPOINT LOCUS
11.0
10.0
9.0
8.0
7.0
5.0
4.0
3.0
2.0
1.0
0.0
I
(300. .3.)
(200. , 1.1
_L
(800..8.)
(1000. .10.)
(1600.. 10. J
..8.51
MYB 1968 -
1980
«YR PRE 68
J !
J I
0 200 400 600 800 1000 1200 1400 1600 1800 2000
100 300 500 700 900 1100 1300 1500 1700 1900
HC (PPM)
-------
10
B. Technical Problems in Selecting Initial Outpoints (Pre-1981 Vehicles)
Once a locus of possible HC/CO cutpoints is selected, selecting specific
cutpoints for initial use in an I/M program is simply a matter of estimating
the failure rates that will result from different possible cutpoints and
choosing the cutpoints that produce the desired failure rate. Estimating the
failure rate resulting from a set of cutpoints requires attention to an I/M
program's unique attributes. Consideration of the geographic location and
starting date of the program are both important.
The same cutpoints may yield substantially different failure rates depending
on the geographic location of the I/M program. EPA samples indicate varia-
tions in failure rate of as much as 5 percent when identical cutpoints are
applied in different states. Altitude, fleet composition and local main-
tenance habits could be contributing factors to this variation. The use of
local data therefore is probably better for failure rate estimation than data
collected in other areas, provided enough local data is collected to keep
sampling error at an acceptable level. Table 2 gives sample sizes appropriate
for sampling one model year group with varying tolerances, confidence levels
and estimated model year group failure rates. For exampl.e, an I/M program
which desires an accuracy of +2% with 95% confidence for each model year group
failure rate estimate and is interested in failure rates of 35 percent should
sample 2185 vehicles per model year group in its cutpoint strategy (10925
total if, as is recommended, five model year groups are used for pre-1981
vehicles). The data from this sample should be preserved in the form of one
record for each vehicle, the record should consist of the HC and CO scores and
the model year or model year group.
Table 2
Appropriate Sample Sizes
for Estimating Failure Rates at Different
Levels of Confidence and Tolerance
Confidence
Level—
95%
99%
Failure
Rate
20%
25%
30%
35% '
20%
25%
30%
' 35%
+_ 1%
6147
7203
8068
8740
10618
12443
13936
15097
Tolerance
+ 2%
1537
1801
2017
2185
2655
3111
3484
3775
± 5%
246
289
323
350
425
498
558
604
-------
11
The best data to use when estimating the failure rate in an I/M program is
data from the operating program itself. However, to select initial outpoints
it may be desired to obtain these estimates before the actual program is in
place. Two possible ways of collecting useful data are a pilot program (e.g.,
shopping center testing) and a mandatory inspection/voluntary maintenance
phase.jY A strong possibility of recruiting or sampling bias in the former
makes the latter preferable. For either option, though, test accuracy must be
assured by good quality control. Otherwise the sample estimates may be
meaningless.
Failure rates observed now for a given set of cutpoints and vehicles are
probably lower than they will be in the future. As vehicles age and are
driven more miles they get dirtier. For this reason, a given pair of HC/CO
standards for a given model year will result in higher failure rates with
time. Failure rate estimates derived from a current sample will be less than
what will actually be observed in a program beginning a couple years from
now. The most recent data is therefore the best sample to use when estimating
failure rates.
When the actual I/M program does begin, states should plan to use the actual
I/M data to measure the actual failure rate in the initial months. A month or
two should be sufficient time to collect enough data to make these measure-
ments.
Once the necessary data has been gathered, estimates for the failure rates
resulting from various cutpoints can be achieved using computer software
similar to the FORTRAN program attached. This program can test any cutpoint
pair input by the user and return the resulting failure rate, or accept as
input a desired failure rate and HC/CO cutpoint relationship and return the
appropriate cutpoints. With data sets of reasonable sizes, such as mentioned
in Table 2, this analysis can be managed on most computer systems once the
data has been transcribed into computer usable form.
JY Either of these approaches may over predict the failure rate which will be
observed at the beginning of the mandatory maintenance I/M program. During
the mandatory maintenance phase, some vehicle owners will probably have their
vehicles maintained anticipating the emissions inspection. No data is avail-
able to investigate this further.
-------
12
For the benefit of states which do not have local data available, EPA has
developed a model to predict failure rates for various cutpoints in the
future.I/ Data used in creating this model came from 10,450 light duty
vehicles of all model years sampled nationwide by EPA's I/M Demonstration Vans
during the past year, 2,513 1975-1980 model year vehicles tested in Houston,
Phoenix, St. Louis, Chicago and Washington D.C. by EPA's Emission Factor
Program between 1975 and 1979, and 10,656 vehicles of all model years sampled
in New Jersey's I/M program during the past year. The EPA model accounts for
the fact that failure rates will be higher when I/M programs start than when
these vehicles were tested.
Tables 3 and 4 present prediced minimum, maximum, and mean failure rates for
all pre-1981 model year groups. Table 3 is for programs beginning 1/1/82;
Table 4 for those starting a year later. The minimum is the failure rate for
the most recent and therefore the youngest and cleanest model year in the
group; the maximum is the failure rate for the oldest and therefore the
dirtiest model year; and the mean is the average for all the model years in
the group, taking the national average registration distribution of model
years within model year groups at the projected time into account. As the
tables show, there is little variation in failure rates among model years
within a model year group, as is usually desired. This verifies the appropri-
ateness of the model year groupings chosen.
JY Details concerning this model are to be released in a technical paper
titled: "Failure Rates in Inspection and Maintenance Programs". Copies will
be distributed to state I/M officials and will also be available from the
EPA's I/M Staff in Ann Arbor, Michigan (312)668-4367.
-------
Table 3
Initial Model-Year Group Failure Rates
Predicted by EPA Outpoint Model
At I/M Program Start Date 1/1/82
Outpoints
Idle
HC CO
(ppm) (X)
200 1.0
220 1 .2
225 1.5
250 2.0
275 2 .5
300 3.0
350 3.5
400 4 .0
450 4.5
500 5.0
550 5.5
600 6.0
650 6.5
700 7.0
750 7.5
800 8.0
1400 8.5
1600 10.0
Registration
Fraction
Pre-1968
Mini/
90.0
86.6
81.1
75.4
69.8
65.0
62.2
56.7
51.0
46.6
4.1.9
26.3
16.0
Max2_/
90.3
87.2
81.1
77.6
71.0
65.7
62.8
57.8
52.2
48.3
43.9
29.0
16.7
.038
Mean3_/
90.1
86.9
81.1
76.5
70.3
65.5
62.5
57.4
51.7
47.5
42.9
27.8
11.4
1969-71
Min
76.8
71.3
66.3
59.9
53.5
48.6
42.6
37.0
34.3
29.1
24.8
Max
76.8
72.0
67.5
61.4
54.2
49.9
44.1
39.4
35.1
30.2
27.4
.110
Mean
76.8
71.8
66.5
60.9
54.2
48.9
43.2
38.2
34.4
29.6
25.8
Min
86.0
81.2
75.2
68.9
65.3
58.7
.53.3
47.9
42.5
36.3
31.4
27.3
23.6
19.4
16.6
Model
Year Group
1972-74
Max
88.7
83.4
77.5
71.4
68.3
62.1
56.6
51.2
46.5
39.7
34.8
30.1
25.8
22.7
18.9
.184
Mean
87.2
82.4
76.3
70.1
66.3
60.4
54.7
49.6
43.9
38.0
33.1
28.6
24.5
20.8
17.6
Min
58.8
52.0
47.3
43.2
40.0
35.1
30.8
26.3
22.4
19.8
16.7
14.4
12.0
10.2
8.0
1975-79
Max
77.9
60.0
54.8
50.3
48.7
44.1
39.3
35.1
31.2
27.6
23.6
20.7
18.4
16.0
13.9
.449
Mean
66.1
56.5
51.1
46.9
44.7
39.9
35.4
31.2
27.2
23.7
20.3
17.8
15.4
13.3
11.2
1980 1981+4/
Mean
48.5
5.0-10.0
43.0
36.6
31.0
24.2
19.4
16.6
13.9
11.5
8.9
7.2
5.3
4.0
3.0
1.9
.095 .124
(1) Minimum model year failure rate within model year groups.
(2) Maximum model year failure rate within model year groups.
(3) Average failure rate in model year groups weighted by projected model year registration fractions.
(4) EPA recommends use of 207(b) short test standards of 1.2% CO and 220 ppm HC.
EPA expects that the failure rate from these outpoints will never exceed 5-lOfc.
-------
Table 4
Initial Model Year Group Failure Rates
Predicted by EPA Outpoint Model
At I/M Program Start Date 1/1/83
Outpoints
Idle
HC CO
(ppm) (%)
200 1.0
220 1 .2
225 1.5
250 2 .0
275 2.5
300 3 .0
350 3.5
400 ' 4 .0
450 4.5
500 5.0
550 5.5
600 6 .0
650 6.5
700 7.0
750 7.5
800 8 .0
1400 8.5
1600 10.0
Registration
Fraction
Model
Pre-1968
Mini/
90.0
86.6
81.1
75.4
69.8
65.5
62.2
57.4
51.4
47.2
42.3
27.0
16.3
Max2/
90.3
87.2
81.1
77.6
71.0
65.7
62.8
57.8
52.2
48.3
43.9
29.0
16.7
.027
Mean3/
90.2
87.0
81.1
77.0
70.6
65.6
62.6
57.7
52.0
47.9
43.3
28.4
16.6
1969-71
Min
76.8
72.0
66.3
61.4
54.2
48.6
43.2
38.1
34.3
29.6
25.8
Max
76.8
72.0
67.5
62.2
55.2
49.9
44.1
39.6
35.1
31.4
27.4
.084
Mean
76.8
72.0
66.7
61.5
54.6
49.1
43.6
38.9
34.6
30.0
26.4
Min
87.2
83.0
76.7
70.3
65.7
60.9
54.6
50.1
43.4
38.5
33.7
28.8
24.4
20.9
17.6
Year Group
1972-74
Max
90.3
84.6
79.2
73.0
69.8
63.5
57.6
52.5
47.8
40.7
35.6
30.7
27.3
23.1
19.4
.152
Mean
88.5
83.5
77.6
71.4
67.6
62.0
56.0
51.1
45.5
39.5
34.5
29.7
25.6
22.1
18.5
Min
60.6
54.8
49.6
45.7
42.6
37.7
33.0
28.7
25.1
21.6
18.8
16.3
13.9
11.4
9.9
1975-79
Max
79.8
63.1
56.6
51.0
50.1
46.2
41.1
36.9
32.3
28.7
25.1
22.0
19.8
16.8
15.0
.447
Mean
69.5
58.3
52.6
48.1
46.3
41.7
37.1
32.9
28.8
25.1
21.7
19.1
16.7
14.4
12.3
1980 1981+4/
Mean
51.4
5.0-10.0
45.9
39.2
35.2
27.0
23.6
18.9
16.5
13.9
11.6
9.6
7.8
5.9
4.4
3.4
.073 .217
(1) Minimum model year failure rate within model year groups.
(2) Maximum model year failure rate within model year groups.
(3) Averge failure rate in model year groups weighted by projected model year registration fractions,
(4) EPA recommends use of 207(b) short test standards of 1.285 CO and 220 ppm HC.
EPA expects that the failure rate from these outpoints will never exceed 5-10%.
-------
15
C. Outpoints For Initial and Subsequent Years (Pre-1981 Vehicles)
Because of emission deterioration, failure rates based on data available today
are expected to increase up to the time at which the I/M program begins. I/M
will thereafter reduce failure rates relative to what they would have been
without I/M, since maintenance received to pass the I/M test will still have
some effect when vehicles come due for their next annual inspection. How much
effect varies with the standards the vehicles were initially required to pass.
During the first twelve months of an annual I/M program, failure rates should
remain approximately constant if vehicle inspections fall on vehicle birthdays
since the inspected vehicles will be the same age each month. This may not be
the case in states which have recently switched to a staggered registration
system or still have another registration system in place.
At the second annual inspection, failure rates will probably be different, for
a given set of outpoints, than they were at the first inspection. Two prin-
ciple factors influence how failure rates will change. Those vehicles which
were maintained as a result of the first inspection will still be cleaner at
the second inspection than they would have been without emissions main-
tenance. They may even be. cleaner than they were one year earlier at the time
of their first inspection, depending on how much cleaner maintenance made them
and how fast they have deteriorated. Passing vehicles will have deteriorated
during the interim period between inspections thus tending to result in an
increase in the failure rate. The change in failure rate will depend upon the
balance among these influences. Since sufficient maintenance would have been
performed on failed vehicles to enable them to pass the original outpoints,
the failure rate change will largely depend upon the initial failure rate and
outpoints.
EPA's cutpoint model predicts that without I/M, model year group failure rates
will increase 1~4 percentage points in the course of a year while with I/M
these failure rates will decrease 4-14 percentage points for a given set of
outpoints designed to achieve an initial failure rate of 35% in each model
year group.
States have two general options to choose from in selecting their I/M cutpoint
strategy. They may leave outpoints fixed from year to year and let failure
rates change, or they may adjust outpoints from year to year to achieve a
constant failure rate or some other desired pattern of failure rate._l/ The
second option puts the state in control of failure rate, and thereby also in
control of the demand for I/M- repair and for reinspection.
JL/ Fixed outpoints with changing failure rates are modeled in MOBILE2.
Special help is available if a state wishes to adjust outpoints during an I/M
program.
-------
16
The effectiveness of an I/M program in reducing emissions from pre-1981
vehicles increases as cutpoints are made stricter, up to the point at which
clean cars are being failed frequently. In practice, this only means that
failure rates should not exceed 50 percent. If a state allows cutpoints to
stay fixed and as a result the failure rate declines (as it generally will
between the first and second year), the state is losing an opportunity for
greater emission reductions. These greater reductions could be achieved
without increasing the failure rate from the rate to which the public, the
repair industry, and the I/M employees are already accustomed. All that would
be needed to achieve the greater emission reduction would be to tighten the
cutpoints just enough to keep failure rates constant. EPA therefore recom-
mends that states plan to adjust cutpoints in this way.
EPA has estimated the initial cutpoints and the revised cutpoints that would
be needed at the start of the second year, for a range of failure rates.
These estimates are shown in Table 5. By the start of the third year, each
I/M program should be able to revise its cutpoints nore accurately by using
local data than by relying on EPA's general estimates, so third and subsequent
year cutpoints are not included. Table 5 gives appropriate cutpoints for I/M
programs beginning 1/1/1982 and 1/1/1983. The second year cutpoints shown in
these tables should be planned for implementation at the beginning of the
second inspection cycle. To insure a program's capability of adjusting
cutpoints to maintain a constant failure rate, it is further suggested that
the stricter subsequent year cutpoints be on the books and ready to take
effect from the start of the program.
-------
Table 5
EPA Recommended I/M Outpoints
Outpoints predicted to give constant failure rates among pre-1981 vehicles for I/M beginning
1/1/82 and 1/1/83 in initial and second year inspections by EPA cutpoint model.
I/M Nominal
Beginning Failure
Date Rate (%)
1/1/82
20
25
30
35 21
1/1/83
20
25
30
35
Program
Sequence
initial
2nd year
initial
2nd year
initial
2nd year
initial
2nd year
initial
2nd year
initial
2nd year
initial
2nd year
initial
2nd year
Model Year Groups
Pre-1968
HC
(ppm)
1550
1400
1450
1400
1400
800
1400
800
1550
1400
1450
1400
1400
800
1400
800
CO
(%)
9.5
8.5
9.0
8.5
8.5
8.0
8.5
8.0
9.5
8.5
9.0
8.5 '
8.5
8.0
8.5
8.0
1968-71
HC
(ppm)
850
750
800
700
750
600
700
600
900
750
800
700
700
650
700
600
CO
(%)
8.5
7.5
8.0
7.0
7.5
6.0
7.0
6.0
9.0
7.5
8.0
7.0
7.5
6.5
7.0
6.0
1972-74 1975-79
HC
(ppm)
750
600
700
500
650
400
600
400
800
600
700
550
650
450
600
400
CO HC
(%) (ppm)
7.5 600
6.0 500
7.0 550
5.0 400
6.5 450
4.0 350
6.0 400
4.0 300
8.0 650
6.0 500
7.0 550
5.5 400
6.5 500
4.5 350
6.0 400
4.0 300
CO
(%)
6.0
5.0
5.5
4.0
4.5
3.5
4.0
3.0
6.5
5.0
5.5
4.0
5.0
3.5
4.0
3.0
1980
HC
(ppm)
350
450
300
400
275
300
250
275
400
450
350
350
300
300
275
275
198H-I/
CO
(%)
3.5
4.5
3.0
4.0
2.5
3.0
2.0
2.5
4.0
4.5
3.5
3.5
3.0
3.0
2.5
2.5
HC
220
220
220
220
220
220
220
220
220
220
220
220
220
220
220
220
CO
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
I/ All I/M programs should use the 207(b) cutpoints for 1981 and later vehicles
for all program years. EPA expects that the failure rate from these cutpoints
will never exceed 5-10%.
2/ EPA recommends this failure rate for best effectiveness.
-------
18
D. Subsequent Outpoint Revision (Pre-1981 Vehicles)
EPA outpoint predictions for pre-1981 vehicles are not tailored to specific
local conditions such as fleet composition, altitude, and maintenance habits
indicating that both initial and subsequent failure rate projections will
probably not be completely accurate. For 1980 vehicles in particular, EPA's
predictions are based on limited data and consequently may be inaccurate.
Thus it is likely that states will find it necessary to revise their cutpoints
after I/M has begun. States should be prepared to revise cutpoints within the
first one or two months to correct for overall or model year group failure
rates which differ from those expected and desired. If a state wishes to
achieve a constant or specified failure rate in the second and later years,
subsequent revisions become imperative.
In designing an I/M program it may be desirable to specify in advance (by law
or regulation) how cutpoints will be revised. One alternative is to give
administrators a free hand in cutpoint revision. If this is not preferred,
there are other alternatives retaining administrative flexibility while
keeping program administrators from having a completely free hand. By law or
regulation a state could establish a range (such as +_ 2.0% CO, _+ 200 ppm HC)
in which cutpoints may be modified without new laws or time-consuming ruleraak-
ing. A variation of this alternative would be an enforcement tolerance or one
sided range allowing administrators to tighten but not loosen cutpoints.
Conversely, a state could establish by law or regulation a range within which
the program administrator would be directed to confine the failure rates,
requiring him/her to revise the cutpoints if the failure rates should be found
outside of these boundaries for some appropriate length of time.
States should be able to make cutpoint revisions quickly. This will be found
to be especially desirable during the first few months at the start of the I/M
program when local influences on failure rates become apparent. Similar
corrections may be necessary at the beginning of the second inspection cycle.
To accomodate this, procedures for data collection, handling, and analysis
should be developed. What is necessary is a continuous sample of idle scores
and failure frequencies at first inspections identified by model year and test
date, and the capability of analyzing these data in a rapid manner.
E. Cutpoints for 1981 and Later Vehicles
The I/M process for 1981 and later vehicles will be different in several ways
as compared to earlier model year vehicles. This results from the signifi-
cantly different technology which will be employed on the 1981 and later
fleet. This technology utilizes an on-board computer control system to
monitor the operation of the engine and continuously adjust engine parameters
such as air/fuel ratio, spark timing, EGR flow rate, etc. This is in contrast
to earlier model year vehicles which rely on a combination of fixed internal
parameters (passage sizes, spring stiffnesses, etc.) and adjustable external
parameters (adjusting screws, variable distributor alignment, etc.) For these
earlier model year vehicles, one of the main causes of vehicles emitting
excessive emissions in-use is improper adjustment of the external engine
parameters (e.g. idle mixture). This occurs in a fairly large percentage of
-------
19
the in-use fleet. (The other main cause of excessive in-use emissions is
normal deterioration of replaceable components such as spark plugs, the air
filter, distributor cap and rotor, spark plug wires, diaphragms and rubber
hoses.) For the 1981 and later fleet, however, the few remaining external
adjustable parameters are of minor importance in terms of emissions, and the
more critical internal parameters are not adjustable and fall under control of
the on-board computer. The main cause of vehicles emitting significantly
above the standard is failure of the computer control system. (The other main
cause of high emissions for earlier model year vehicles as described above is
also still present but plays a secondary role.) Failure of the computer
control system can occur in a variety of different ways, but it is not
expected to occur very often. When it does occur, HC and CO emissions rise
dramatically: hundreds of percent over the design standard. The I/M process
for 1981 and later vehicles will focus in on this relatively small group of
grossly emitting vehicles^.
As indicated above, 1981 and later vehicles are also expected to experience a
phenomena common to all model years of vehicles: ignition and misfire problems
due to deterioration and failure of replaceable ignition parts. These
problems are, in a certain way, similar in nature to the computer control
system failures described above. That is, they do not occur very often, but
when they do they can cause a dramatic increase in HC emissions.
Given the preceding background on the differences in 1981 and later technology
and the types of failures which will be encountered, issues related to
cutpoint selection and failure rates can now be discussed. There are two
issues: the first is that the selection of the short test to be used for 1981
and later vehicles is of more importance than which outpoints are used. The
second is the impact of the 207(b) warranty on cutpoint selection.
The first area listed above stems from the fact that due to the sophistication
of 1981 and later vehicles, some I/M short tests will be more capable of
identifying the gross emitters than others. For example, a test which simply
measures a vehicle's emissions at idle will be slightly less effective than a
test which also measures and evaluates (i.e. passes or fails) a vehicle's
emissions at either an elevated speed (2500 rpm) or under a loaded condition
(30 mph at 9.0 AHP load). Evaluating a vehicle's emissions at modes other
than the basic idle mode puts the computer control system more to the test by
giving it more than one operating condition to respond to. This conclusion is
based upon analyses of data from actual in-use vehicles equipped with tech-
nology representative of the 1981 and later model year fleet. While cutpoints
still play an important role, their role is secondary in comparison to the
choice of which short test will be used. Since the computer control system is
assumed to either be operating properly with very low emissions or to have
failed with resultingly very high emissions, the in-use emissions picture will
1 A more complete discussion of this subject can be found in a technical
report entitled: Derivation of I/M Benefits for Post-1980 Light Duty Vehicles
for Low Altitude, Non-California Areas. [EPA-AA-IMS/81-2]
-------
20
be much more quantized. This is as opposed to the continuous range of
emission levels seen among fleets of earlier model year vehicles with adjust-
able parameters. Thus, varying the cutpoints up and down within reasonable
limits will not yield significant changes in the failure rate. Grossly
emitting vehicles will for the most part be identified by any reasonable
cutpoint.
The discussion so far has focused on vehicles with a failure of the computer
control system. The other major source of excessive in-use emissions is
vehicles with ignition and misfire problems. These vehicles will have an
in-use emissions performance similar to current model year cars. Therefore it
is reasonable to conclude that currently available cutpoints (such as the
207(b) cutpoints discussed below) will work equally well in identifying those
vehicles requiring maintenance.
The failure rates resulting from the identification of vehicles with computer
system failures and of vehicles with severe ignition/misfire problems were
combined to arrive at the estimated failure rate of 5-10% shown in Tables 3
and 4. In comparison to earlier model year vehicles, there are not large data
bases from 1981 and later vehicles participating in representative I/M
programs. Thus, failure rates cannot be more closely quantified as they have
been for earlier model year fleets.
The second area of discussion has to do with the impact of the 207(b) warranty
upon the selection of cutpoints. The 207(b) regulations established three
basic tests as acceptable for use in the warranty process: the basic
Idle-in-Neutral test with or without a 2500 rpm preconditioning, the Two Speed
Idle test (ldle-in-Neutral/2500 rpm/Idle-in-Neutral) and the Loaded Two Mode
test (30 mph @ 9.0 AHP/Idle-in-N7eutral) . The regulations also provided, for a
number of variations on test format, that is, which test modes would be used
in making the pass/fail decision. Cutpoints were established for each test
and are EPA's recommendation for use for 1981 and later vehicles in operating
I/M programs. The basic cutpoints are 1.2% CO and 220 ppm HC. The one
exception to this is for the Two Speed Idle test where either the 2500 rpn
emissions or the lower of the two Idle-in-Neutral emissions are the criteria
for failure. For these cases the cutpoints are 1.0% CO and 200 ppm HC. Table
6 lists the basic tes-fe-s, the variety of formats and the cutpoints for each.
-------
21
Table 6
207(b) Tests and Associated Cutpoints
(1981 and Later Models)
Basic Test
Idle-in-Neutral
Mode(s) to be Evaluted
Idle-in-Neutral
Cutpoints
1.2% CO, 220 ppm HC
Two Speed Idle
(Idle-in-Neutral/
2500 rpm/Idle-in
Ne.utral)
Lower of two Idles-in-
Neutral
2500 rpm
2500 rpm and Lower of two
Idles-in-Neutral
1.0% CO, 200 ppm HC
1.0% CO, 200 ppm HC
1.0% CO, 200 ppm HC
2500/Idle-in-Neutral!/
2500 rpm
Idle-in-Neutral
1.0% CO, 200 ppm HC
1.2% CO, 220 ppm HC
Loaded Two Mode
Idle-in-Neutral 1.2% CO, 220 ppm HC
30 mph 1.2% CO, 220 ppm HC
Idle-in-Neutral and 30 mph 1.2% CO, 220 ppm HC
.!/ This test combines the 2500 rpm test and standards from the two speed
idle test with the basic idle test, and is 'EPA's recommended test for 1981 and
later models. It differs from the two speed idle in that only one
idle-in—neutral is performed. Because the standards are relatively
insensitive to failure rates for post-1980 models, 1.2% CO and 220 ppm HC are
recommended for both modes for the purpose of simplicity in administration.
-------
23
LDVI/
Pre-1968
1968-1971
1972-1974
1975-1979
1980
1981+
Table 7
Vehicle Model Year Groups
Corresponding To Similar Emission
Control Technologies
LDTll/ LDT21/
Pre-1968 Pre-1970
1968-1971 1970-1972
1972-1974 1973-1978
1975-1983 1979-1983
1984+ 1984+
J./ LDV = Light Duty Vehicle.
2] LDT1 = Light Duty Truck with GVW less than 6,000 pounds.
_3/ LDT2 = Light Duty Truck curb weight greater than 6,000 pounds or Light Duty
Truck with GVW between 6,000 pounds and 8,500 pounds.
-------
22
There are several points which lie behind EPA's recommendation that the 207(b)
cutpoints be adopted. First, and perhaps most obviously, a tighter cutpoint
should not be used. This would result in problems with warranty coverage.
For example, if an I/M cutpoint was set lower than the established 207(b)
cutpoint, an auto manufacturer would not need to honor a warranty claim from a
vehicle which failed the lower cutpoint since the regulation specifically
defines the higher cutpoint as the lowest acceptable criterion. On the other
hand, while a state or locality could select cutpoints higher than the 207(b)
cutpoints without loss of warranty coverage, there is no good reason to do so
since the failure rate using the 207(b) cutpoints is already very low
(5-10%). This failure rate is well below the limit necessary for good public
acceptance.
F. Cutpoints for Light Duty Trucks
The same considerations apply when selecting cutpoints for light duty trucks
(LDTs) as in selecting cutpoints for light duty vehicles (LDVs). LDTs are
classified into 2 weight categories for purposes of federal emission stan-
dards. Each category is subject to a different set of certification schedules
for meeting federal emission standards. LDT model year groups recommended in
Table 7 are chosen so that each group has similar emission controls to the .
corresponding LDV model year groups.
The same cutpoints should be applied to both LDT categories as well as LDVs
within similar emission control groups. Further refinements are possible once
the I/M program begins.
III. Summary
Cutpoints should be selected and adjusted to maintain a constant failure
rate. Appropriate initial cutpoints and revised cutpoints for the second
inspection cycle are presented in Table 4. If an I/M program results in a
failure rate lower than anticipated, the failure rate should be allowed to
climb, or cutpoints should be revised. Whatever the desired failure rate,
HC/CO cutpoint pairs should be selected from HC/CO loci in Figure 2 and Table
1.
-------
Attachment
FORTRAN program for automating
outpoint selection
-------
is
I CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC
2 CC CC
3 cc THC PROGRAM, 806<*ti-cuT. is 10 STUDY THE nc f» CO CUT POINTS ANO cc
<» CC OtTtKMlNt THE STANDARDS FROM A DATA bASE FILE. CC
5 CC. CC
6 CC HUN SAUP:6004H-CUT 1 =SGWH: EFLINE . T2 6=»MSOURCE* 6=<*MSINK* CC
7 CC m. KS.O.F5.2 CC
a cc cc
<» CC WRITTEN rtY J.P.CHENG ON 07-G7-riO. CC
10 CC REVISED rtY J.P.CrttNG ON 07-14-80. , CC
11 CC CC
12 CC CC
13 CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC
14 CC
15 CC
16 DIMENSION HC(bOOO), VA 11 (5000 ) • VA22 (SOOO ) , COMl(20) ' I
17 DlMcNSlON C0«5000>t Vfl21 (SOOO) . VA12 (5000) , FMT(IO) f.
ia cc
19 DATA fMT(l)f FMf(10)i 6UNK.ONO /•( ••• )•.' '»»N •/ 1
no cc
21 100 FOKMATI//' JfiPUT IHC AC.'D JCO HtAO FORMAT «»»»«««««>»•, <,
Z2 »/• XAXAXXXXXXXXXXXXXXXXXXXXXXXAAXXX • )
23 10? FyH,iA7 ( 2'JA*. ) T
2«» 10<» FOMHAU IX, 20A«. ) ft
*s CC
26 106 FOHMAT ( IriOt • IN THE SAQPs^ub^H-CUT PKOOHAM. THE FUNCTIONS OF ' i 7
27 w 'KuUhi (4) OPCODES /'.Kt. LISTED bELOv/ : '/1HO,
2d •*/• 1 = GIVEN COTS TO PROJECT FREO. PERCENTAGES. • ,
2V »/' 2 = GIVEN KwEO. PLrtCE.NTftGES TO PROJECT CUTS.',
30 •/' 3 = GIVLN
33 CC
3«. CC
35 lOfl FORMAT CO INPUT COMMENTS AS THE HEADER OF 0/P ( SUCH AS', M
36 M ' UATA StT NA.«Ei FORMAT ETC. )«/lH • 203 CC
55 CC
56 WKIT£(6,10U) I'A
57 CC
56 REAU (5,102) COMl 1*9
-------
t* C *l
C>E d>
59
60
hi
62
63
&^t
GO
66
67
64
69
70
71
Id
73
7<»
Vb
76
77
7«J
79
ttO
81
H2
y3
<)<«
as
COMl
UO 10 N=l,bOOO
rfEAO <1,F>1T,END*1S) HC(N), CO(N)
CONTINUE
" N " IS THE SAMPLE SlZt
N = N - 1
SORTING
CALL SOKTPRC HC, co, VAiitVAiZt N, IRTN )
CALL SOHTPR( CO, HC. VA«?1.VA22. M, I«TN )
CALCULATE PERCENTILES AND LIST THEM OUT.
wki rtfti.iio)
00 3U 1P=2,99,2
PtKT = ( IP • N ) / 100
INT = ( IP u N ) / 100
DLL = PEKl - INT
HCTIL = VAH(INT) » DEL « VAll(INT*l) « ( 1. - OEL )
COTIL = VA21UNT) » OEL * VA2KINT+1) « ( 1 . - DEL )
WKlIt(6,112) IP, HCTIL. COTIL
CUNiIliNUE
IP = 100
WKITE(0,112) IP, VAll(N), VA2KN)
MAJOK OPF.HATIONS by USINli " CUTOUT " SUBROUTINE
whUTt (b, 1 1<»)
RC.AO (5,116) ANS
IK{ ANS .tU. ONO ) GO TO 999
CALL CUTOUT( VA11.VA22, VA21,VA12, COMl, N )
•
GO TO 777
STOP
ENO
IttblS^i: SAOP tlILE;!«06'.G-CUT 1'
20
2 1
2?
23
2^»
—
23
2f>
2V
?_=>
?°
30
31'
3^
33
3<»
3?
36
3f
3*
3°
u(3
u\
U'.
ul
4°
-------
IUU W
110
111
112
113
1 1*
us
lib-
117
118
119
120
121
122
123.
U*
US
12b
127
12«
129
130
131
132
133
1 3*
135
1 36
137
13d
139
1*0
1*1
1*2
1*3
!<**
1*5
!<AGt 3>
1'3
i'-f
2?
23
-------
*,„._.-—.-•—.———..—— ———. — — ———»'
I t)ftIE:iO-16-80»OB!22 QjiN|F.a:SAQP F.1LF.I t»0*><»r,-CUT I'
cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
169 CC C
170 CC THL iUiJMOUTINE, CUTOUT, IS TO CALCULATE THE HC & CO CUTOUT POINTS C
171 CC fwOM OIVtN FKtOUtNCIES OR VICEVEKSA. C
172 CC C
1 73 CC C
I7a CC C
175 CC C
\fh CC C
177 CC C
I7d C.C WKITTEN eiY J.P.CHENG ON 07-15-BO. . C
.179 CC C
180 CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC __
1B1 CC
Id2 SUcJWUUTlNE CUTOUTl VA11.VA22, VA21,VA12, COMl, N ) l'
1*3 • CC
CC
DIMENSION VAl 1(SOOO),VA 22 (5000), VAiM (5000), VA12 (5000) ,COMl (20) ?.
DIMENSION S^CISOOO), SCO(5000), KS<5000)» PCT4(4),COK2(20) J
DIMENSION KTIM <•
CC
DATA MLNK/' '/ ^
5*0 FUHrlATCO INPUT COMMENTS WHICH WILL rfE USED AS 0/P HEAOEN', 6
191 w /JA»' ( OEFLT = PHEVIOUS COMMENT INPUT )', 21X, 6('COMM») )
192 • SVS FOKMATI ««««»« INPUT OPCODE (1/2/3/4): • ) V
196 bOl Fu«M«T('u INPUT : », l'n
197 ' •> '/»0 HC. CUT CO CUT" /' XXX.XXX XXX.XXX' )
198 00? FOrfKAH «0- INPUT ' F2 (HC-F-Co-P) f. F3*F'»(CO-F) IN •» IT
199 *'HtrtCfiMTAl»E.S. '/'O ^ ?. '.'> F3«F4 A' /' XXX.XXX XXX.XXX' )
200 603 FUrtMATCO INPUT !'/'0 MEHCENT HCCUT/COCUT ', lV?
«' '/' F2 + f-3«F'» HA Til) • /' XXX.XXX XXX.XXX' )
60<< fUKMA'ft'O INHUT :'/'U PEKCt'NT '• 13
*' '/' F2»F3*F<» • /i XXX.XXX • )
CC
2*05 202 KUHMATI II ) l<»
206 212 Furi.«iAT( 2FV.3 ) • Ti
2oe cc
^09 -220 Fux.MATCO ««««« THE NEGATIVE Fl = ', F10.3) l'7
^'0 22^ FUKHAFllHU,' «« yo« uo«ou*> uttuo 11 uu tto«ttOiiouuo«ooo» i f ^
?11 •/' H C H C H C H C SIZE «I
212 »/' CO CO CO CO',
213 »/' P P F P P F F F '»
?»•» «/ IX, M7» 3X»I7, i FREO. ' *
21^ »/ 2X, <»F7.3, ' F«F.O. IN PERCENTAGES ' ,
216 »//• »•» HC CUT =', F9.3,' / CO CUT =',Fft,3 // IX, 40<»<>«) )
217 244 I-U^MATCO »*««•«»«» SO«HY CHARLIE ««»••««•«««•«/. fy
2M • ' UNA6LL TO CALCULATE WITH Fl = ', 14
'219 «/' HC CUT a •, F7.3,' N CO CUT «= ', F7,3 )
?20 CC
221 cc INITIALIZATIONS AND HEAOEH FIXING
222 CC
^223 UU 110 K=l»<» 20'
?><» K T <<<} a n iff
-------
231
232
233
23A
237
2<»6
?<*7
2*0
2bl
2:>2
2S3
200
263
.«
i
CC
CC
CC
CC
115
HtAO (S.bVb) COM2
KNT = U
DO UU K = l»20
IF( COM2(M .E'O.BLNK > KNT = KNT » 1
CONTINUE
IK ( M-Jf.tu.20 ) GO TO 116
00 115 K=1.20
COM1(K) = COM2(K) '
CONllNOE
CC
CC
]
CC
-,?cc
116
127
CC
CC
CC
CC
CC
COMl
MAJOK UPEHATIONS START HERE *»**««»«««nnnnnn»i»»»
w«lTt(o.60U)
KtAO .KODE
OHCOOE = 1 < GIVEN HC CUT AND CO CUT TO GENERATE RESULTS. >
CC
«c.Alj (^,212) HCCUT, COCUT
HCCUT = HCCUT + .b
COCUT = COCUT » .005
HCCUT. CHCUT
no
[=I.N
26A
265
266
0
2*1
212
•» 11 ^
150
CC
CC
CC ,
CC !
2
•
CC
CC
CC
CC
Kb 1) =
IK Vil
IF VAl
U VAl
KT ^S<
1
1
1
1
(I).
(1).
(1).
GT. HCCUT
Lt.HCCuT
GT.MCCUT
.ANO.
. ANU.
.A NO.
VA12
vAii;
VA12
(1)
(I)
(I)
.LE
.GT
.GT
.COCUT ) KS(I) = 2
.COCUT ) KSU) = 3
.COCUT ) KS(I) = <»
I) > = KT< KSU) ) « I
CONTINUE
Gu TO
OHCOOE
WK! TE(b
KtAO (-3
WK1 TE (o
voo
=
f
«
,
^
602)
212)
21A)
< GIVEN
F2^CT.
F2PCTi
OtTEKMlNE NFCO AND
NFCO =
N
« (
100. -
F2<*>
F3AHCT
F3APCT
AND
NFHC CUT
F34PCT.
» * k >c r+i\ .
)
1
F3»F*
t-fi) TO GENERATE RESULTS. >
OFF POINTS.
/
t
100
. »
0.999
V/N O^. *
. S>
IS'?'
2°
31'
33
3S
36
-JV
34
AJ
A«
A'-;
fcS
/./
A><
<.;
50
SI
S-
b'S
-------
I QAJ£; 10-16-BOtOHJ22 Ciitlf.E.SSAQP FJJ.£:ti06<»G-CUT I'
..__i . -.—..r ..i... . »•• is-;
2b'» 260 IF(VA21 (NKCO) .EO.VA21 (MFCO'l) ) NFCO =« NFCO • 1 S7
?MS 1F.(VA2J (NFCO).EU.VA21 (NFCO-1) ) GO TO 260 S3
?16 CC
2«7
NKHC = NFCO « < F1PCT / ( F1PCT » F2PCT ) ) * 0.999
IF < SHC(NKHC) .N£. SHC(NKHC»1) ) GO TO 2til
IFf iiMC(NFMC) .EO. SHCINFHC*!) ) NFHC = NFHC » 1
IF { briC(NFHC) .tO. SHC(fiFHC-l) ) GO TO 280;
HCCUV = SKC(NFHC) » O.b
TO COUNT F(l) i F(2) WHEN CO PASSES.
KT(1) = NF'MC
Kl (2> = NFCO - NFHC
IM = NFCO » 1
GO 10 bSO
O^COOE = J < GIVEN F2*F3»F^(9S) AND HCCUT/COCUT RATIO
TO GENERATE HESULTS. >
wKIft (0.603)
HtAU (b,212) F23a, HATIO
WKlILlb.21^) F23<», HATIO
CONST o 0.
GO TO J10
OPCyUE » f < WITH BUILT-IN RELATIONS (I.E. HCCUT=KATIO»COCUT»C)
GIVEN K2*FJ»F*(*) TO GENEKATE RESULTS. >
WHITE (0.60
-------
.» *HAOfc ft
I
.«• is:
CC
310 F1PCT = 100. - F23<» P*?
CC
NrHC = N « F1HCT / 100. » 0.999 H»
CC
CC
COLO = VA211NFMC) » .OOb- *<•
COHI = VA21 (N) * .005 «S"
CC
CC
00 3BO iNTKY=l»20 H'*
CC
CoCUT = ( COLO * COHI ) / ?.. H'7.
CC .
CC
IF ( KOuE.fc.0.3 ) GO TO 3<»0 ti*
CC
CC IF( NTHY.tU,! .AND. J.OOb.itT.CUhl ) COCUT = 3.005
rUTIO = 100. 6=*
If ( CUCUT .LE. 3.00 > KATIO = SO. 9'i
CONST = 0. 91'
IFC COCUT .LE. 3.00 ) CONST = 150.25 9»
CC .
3*»o HCCUT = CUCUT « HAT'IO » CONST 91
CC
UO JSi> I=NFHC%N Vfc
IF( VA2KI) .GT4 COCUT ) GO TO 3btt W-:
355 COiNTlNUL ' 9'»
CC
• WHlTt (ftt?**^) NFhC,HCCUT«COCUT 9/
00 Tu (V99.V99, 3. <» J , KOOC <»>
CC
3!>8 CoCUT = VA21(I-1) « .OOb <*••»
HCCUT = COCUT • RATIO * CONST I'OO
CC
00 3t>0 K = 1.4 101'
KTU) = 0 I'u^
360 CONTlNUt 103
CC
CC
IM = I - 1 1'0<.
CC
DO 370 1=1.IM JOS
^S^I) = 1 I'ejf,
IK I VA22II) .GT. HCCUT ) KSd) = 2 I'D f
CC
KT(KSm) = KT(KS(D) » 1 lO'l
CC
'370 CONTlNUt ' 10'>»
CC
IF( KT(1) - NFHC ) 373, 800. 376 1'l'n
CC
373 COLO = COCUT J'lY
GO TO 360 llV
CC
376 CO«I 5, COCUT 11'3
CC
3ttO CONTlNUt 11'4
-------
*
I UAJ£:iO-16-a0.08s22 QttlWSAOP
« .. , ..__...._... .»•' is ;
400 CC
401 CC
402 CC
403 HOO IM = IM • 1 11^"
40<» IKdAHbt KTtll- NFHC ) .UT. S ) GO TO 850 11'"
W«Ht(o»2<»'«) NFMC, hCCUTt COCUT 11'
GO TO (9Wt999» 3t 4 ) , KOOE I'l''1.
i.07 CC
CC
CC STANT TO COUNT F(3) S F(<») WHEN CO FfllLS.
410 CC
Ml CC
412 bSO 00 0*5 I = IM,N It-i
413 KS«I) = J I'?'1
4U IM vA^atn .GT. HCCUT > KSID = 4 121'
4)S KT( ^S^I) > = KT( KSII) ) » 1 Mi*
416 ttbS CUNT1NUE • l?i
417 CC
4lB CC WKlIt UOT HtSULTS
419 CC
4 • 100. / N 12'i
4
-------
T A H L E OF CONTENTS
(ALPHABETICAL)
MOOUI £ FILE PAGE
CUTOUT t SUttrtOUUNfc.' ............................. B06<»G-CUr (
MAIN HHOGKAM .................................... B06<«G-CUT ( Ibt 109) .............. .. ............ ; ......... ' 1
SORTPR i SUUttUUTiNt •• ..... • ..................... f»06<»G-CUT ( 12
------- |