A Study of
Mandatory Engine Maintenance
for Reducing Vehicle Exhaust Emissions
Volume VIII. Experimental Characterization of
Vehicle Emissions and Maintenance States
FINAL REPORT
July 1973
In Support of:
APRAC Project Number CAPE-13-68
for
Coordinating Research Council, Inc.
Thirty Rockefeller Plaza
New York, New York 10020
and
Environmental Protection Agency
Air Pollution Control Office
5600 Fishers Lane
Rockville, Maryland 20852
TRWk
TRANSPORTATION*
'ENVIRONMENTAL
OPERATIONS
SCOTT RESEARCH LABORATORIES, INC
P. O. BOX «4I«
AN BERNARDINO. CALIFORNIA
ONE SPACE PARK PEDONOO BEACH CALIFORNIA 90?/8
-------
A Study of
Mandatory Engine Maintenance
for Reducing Vehicle Exhaust Emissions
Volume VIII. Experimental Characterization of
Vehicle Emissions and Maintenance States
FINAL REPORT
July 1973
In Support of:
APRAC Project Number CAPE-13-68
for
Coordinating Research Council, Inc.
Thirty Rockefeller Plaza
New York, New York 10020
and
Environmental Protection Agency
Air Pollution Control Office
5600 Fishers Lane
Rockville, Maryland 20852
TRW/
TRANSPORTATION t
ENVIRONMENTAL
'oreRATIONS
ONf SPiCC PABK KCOOHOO BCtCH CMIfORNIA 90J/8
SCOTT RESEARCH LABORATORIES, INC
'. O. BOX >4I«
AN BERNARDINO. CALIFORNIA M4O*
-------
PREFACE
This report, "A Study of Mandatory Engine Maintenance for Reducing
Vehicle Exhaust Emissions," consists of eight volumes. The following
are the subtitles given for each volume:
o Executive Summary, Volume I, Final Report, July 1973
o Mandatory Inspection/Maintenance Systems Study,
Volume II, Final Report, July 1973
o A Documentation Handbook for the Economic Effectiveness
Model, Volume III, Final Report, July 1972
o Experimental Characterization of Vehicle Emissions and
Maintenance Studies, Volume IV, Year End Report, July 1972
o Experimental Characterization of Service Organization
Maintenance Performance, Volume V, Final Report, July 1972
o A Comparison of Oxides of Nitrogen Measurements Made with
Chemiluminescent and Non-Dispersive Radiation Analyzers,
Volume VI, Final Report, July 1972
o A User's Manual and Guide to the Economic Effectiveness
Computer Program, Volume VII, Final Report, July 1973
o Experimental Characterization of Vehicle Emissions and
Maintenance States, Volume VIII, Final Report, July 1973
The first volume summarizes the general objectives, approach and
results of the study. The second volume presents the results of a manda-
tory inspection/maintenance system study conducted with a computerized
system model which is described in Volume III. The experimental programs
conducted to develop input data for the model are described in Volume IV
(Interim Report of 1971-72 Test Effort), V, VI, and VIII. Volume VII
is a user's manual for the computer code and Volume VIII reports the
experimental program and data obtained in the final test phase of the
investigation.
The work presented herein is the product of a joint effort by TRW
Systems Group and its subcontractor, Scott Research Laboratories. TRW,
as the prime contractor, was responsible for overall program management,
experimental design, data management and analysis, and the economic
effectiveness study. Scott acquired and tested all of the study vehicles
Scott also provided technical assistance in selecting emission test pro-
cedures and in evaluating the test results.
-------
TABLE OF CONTENTS
Page
1.0 INTRODUCTION 1-1
2.0 EXPERIMENTAL CHARACTERIZATION OF VEHICLE EMISSIONS AND TUNE UP
PARAMETER DETERIORATION 2-1
2.1 INTRODUCTION 2-1
2.2 TEST PROCEDURE 2-2
2.2.1 Vehicle Sample Identification and Acquisition Procedures 2-3
2.2.2 Vehicle Inspection and Tune-Up Procedures 2-13
2.2.3 Exhaust Emission Tests 2-29
2.2.4 Instrumentation and Test Equipment 2-31
2.3 DATA REDUCTION AND ANALYSIS SYSTEM 2-38
2.4 ANALYSIS OF TEST DATA 2-42
2.4.1 Discussion of Results 2-42
2.4.2 Analysis Method 2-86
2.4.3 Variability of Coefficients 2-101
2.4.4 Summary of Data 2-105
3.0 EXPERIMENT TO DETERMINE THE EFFECT OF REPEATED TESTING 3-1
3.1 INTRODUCTION 3-1
3.2 OBJECTIVE 3-2
3.3 TEST PROCEDURE 3-2
3.3.1 Test Vehicle and Preparation 3-2
3,3.2 Test Sequence and Measurements 3-2
3.4 ANALYSIS OF TEST DATA 3-3
3.4.1 Summary of Results 3-3
3.4.2 Discussion of Analysis 3-4
4.0 EFFECT OF COLD SOAK TEMPERATURE ON EXHAUST EMISSIONS 4-1
4.1 INTRODUCTION 4-1
4.2 OBJECTIVE 4-1
4.3 TEST PROCEDURE 4-1
4.3.1 Test Vehicles 4-1
4.3.2 Vehicle Preparation 4-2
4.3.3 Vehicle Soak Temperature and Sequence 4-2
4.3.4 Exhaust Emission Test 4-3
4.4 ANALYSIS OF TEST DATA 4-4
4.4.1 Summary of Results 4-4
4.4.2 Discussion of Analyses 4-4
-------
LIST OF FIGURES
Figure Title Page
2.1 Test Vehicle Processing Sequence 2-7
2.2 Vehicle Control Log 2-8
2.3 Test 4 Vehicle Scheduling Control 2-11
2.4 Test 5 Vehicle Scheduling Control 2-12
2.5 Engine Parameter Inspection 2-14
2.6 Engine Analyzer Ignition Pattern 2-16
2.7 Engine Parameter Adjustment/Repair Record 2-26
2.8 Federal Short Cycle Driving Schedule 2-32
2.9 Short Diagnostic Cycles 2-33
2.10 Data Reduction and Analysis System 2-39
2.11 - Computer Program Interfaces 2-41
2.12 Cold 1972 Federal HC Emissions - Pre-Emission Controlled
Vehicles 2-96
2.13 Cold 1972 Federal CO Emissions - Pre-Emission Controlled
Vehicles 2-97
2.14 Cold 1972 Federal NO Emissions- Pre-Emission Controlled
Vehicles x 2-98
2.15 Cold 1972 Federal HC Emissions - Emission Controlled
Vehicles 2-110
2.16 Cold 1972 Federal CO Emissions - Emission Controlled
Vehicles 2-111
2.17 Cold 1972 Federal NO Emissions- Emission Controlled
Vehicles x 2-112
2.18 Cold 1972 Federal HC Emissions - NOV Controlled Vehicles 2-113
/\
2.19 Cold 1972 Federal CO Emissions - NO Controlled Vehicles 2-114
J\
2.20 Cold 1972 Federal NOV Emissions- NOV Controlled Vehicles 2-115
X X
2.21 Basic Timing - Pre-Emission Controlled Vehicles 2-116
2.22 Basic Timing - Emission Controlled Vehicles 2-117
2.23 Basic Timing - NOV Controlled Vehicles 2-118
/\
2.24 Idle Speed - Pre-Emission Controlled Vehicles 2-119
2.25 Idle Speed - Emission Controlled Vehicles 2-120
2.26 Idle Speed - NOY Controlled Vehicles 2-121
/\
-------
LIST OF FIGURES (CONT'D.)
Figure Title Page
2.27 Air Cleaner Restriction - Pre-Emission Controlled
Vehicles 2-122
2.28 Air Cleaner Restriction - Emission Controlled Vehicles. 2-123
2.29 Air Cleaner Restriction - NOV Controlled Vehicles 2-124
/\
2.30 PCV Valve Restriction (33/30 MPH Cruise) - Pre-Emission
Controlled Vehicles 2-125
2.31 PCV Valve Restriction (33/30 MPH Cruise) - Emission
Controlled Vehicles 2-126
2.32 PCV Valve Restriction (33/30 MPH Cruise) - NOV Controlled
Vehicles x 2-127
2.33 Idle Mode HC Emissions - Pre-Emission Controlled Vehicles 2-128
2.34 Idle Mode HC Emissions - Emission Controlled Vehicles 2-129
2.35 Idle Mode HC Emissions - NOV Controlled Vehicles 2-130
A
2.36 Idle Mode CO Emissions - Pre-Emission Controlled Vehicles 2-131
2.37 Idle Mode CO Emissions - Emission Controlled Vehicles 2-132
2.38 Idle Mode CO Emissions - NOV Controlled Vehicles 2-133
A
2.39 Idle Mode NO Emissions - Pre-Emission Controlled Vehicles 2-134
A
2.40 Idle Mode NOX Emissions - Emission Controlled Vehicles 2-135
2.41 Idle Mode N0₯ Emissions - NOY Controlled Vehicles 2-136
A A
2.42 49/45 MPH Cruise Mode HC Emissions - Pre-Emission
Controlled Vehicles 2-137
2.43 49/45 MPH Cruise Mode HC Emissions - Emission Controlled
Vehicles 2-138
2.44 49/45 MPH Cruise Mode HC Emissions - NO Controlled
Vehicles 2-139
2.45 49/45 MPH Cruise Mode CO Emissions - Pre-Emission
Controlled Vehicles 2-140
2.46 49/45 MPH Cruise Mode CO Emissions - Emission Controlled
Vehicles 2-141
2 47 49/45 MPH Cruise Mode CO Emissions - NO Controlled
Vehicles 2-142
2.48 49/45 MPH Cruise Mode NO Emissions - Pre-Emission
Controlled Vehicles 2-143
2.49 49/45 MPH Cruise Mode NO Emissions - Emission Controlled
Vehicles X 2-144
-------
LIST OF FIGURES (CONT'D.)
Figure Title Page
2.50 49/45 MPH Cruise Mode NO Emissions - NO Controlled
Vehicles x x 2-145
3.1 Effect of Test Sequence on Hydrocarbon Emissions -
(1972 Cold Cycles) 3-9
3.2 Effect of Test Sequence on Carbon Monoxide Emissions -
(1972 Cold Cycles) 3-10
3.3 Effect of Test Sequence on NO Emissions -
(1972 Cold Cycles) x 3-11
4.1 1972 Federal Cold HC VS Water Temperature - Pre-Emission
Controlled Vehicles 4-6
4.2 1972 Federal Cold CO VS Water Temperature - Pre-Emission
Controlled Vehicles 4-7
4.3 1972 Federal Cold NO VS Water Temperature - Pre-Emission
Controlled Vehicles x 4-8
4.4 1972 Federal Cold HC VS Water Temperature - Emission
Controlled Vehicles 4-9
4.5 1972 Federal Cold CO VS Water Temperature - Emission
Controlled Vehicles 4-10
4.6 1972 Federal Cold NO VS Water Temperature - Emission
Controlled Vehicles x 4-11
4.7 1972 Federal Cold HC VS Water Temperature - NO Controlled
Vehicles 4-12
4.8 1972 Federal Cold CO VS Water Temperature - NO Controlled
Vehicles x 4-13
4.9 1972 Federal Cold NO VS Water Temperature - NO Controlled
Vehicles x 4-14
-------
LIST OF TABLES
Title Page
In-Use Percent of Vehicle Population by Make and
Number Desired for Each Fleet Classification 2-5
2.2 Crankcase Ventilation System Type Codes 2-9
2.3 Idle CO % Specifications for Vehicles Equipped with
Exhaust Emission Controls 2-24
2.4 Key Mode Cycles 2-34
2.5 Fleet Distribution over 16 Month Test Program 2-44
2.6 Summary of Deterioration Rates (Change Per Mile) -
Pre-Emission Controlled Vehicles 2-46
2.7 Summary of Deterioration Rates (Change Per Mile) -
Emission Controlled Vehicles 2-47
2.8 Summary of Deterioration Rates (Change Per Mile) -
NOV Controlled Vehicles 2-48
A
2.9 Failure Rates of Engine Parameters - Pre-Emission
Controlled Vehicles 2-49
2.10 Failure Rates of Engine Parameters - Emission
Controlled Vehicles 2-50
2.11 Failure Rates of Engine Parameters - NO Controlled
Vehicles x 2-51
2.12 Failure Rates of Engine Parameters - Vehicles with
Minor Adjustments - Pre-Emission Controlled Vehicles 2-53
2.13 Failure Rates of Engine Parameters - Vehicles with
Minor Adjustments - Emission Controlled Vehicles 2-54
2.14 Failure Rates of Engine Parameters - Vehicles with
Minor Adjustments - NOV Controlled Vehicles 2-55
A
2.15 Comparison of Measured and Predicted Emission
Variation Rates 2-58
2.16 Fraction of Predicted Emission Deterioration Rate
(Slope) - Basis Mean Rate 2-61
2.17 Fraction of Predicted Emission Deterioration Rate -
Basis Maximum Rate (Upper Limit) 2-62
2.18 Emission Response Coefficients Hot 1972 Federal CVS
Procedure 2-65
2.19 Calculation of Average Change in Emissions -
Pre-Emission Controlled Vehicles 2-67
-------
LIST OF TABLES (CONT'D.)
Title Page
Calculation of Average Change in Emissions -
Emission Controlled Vehicles 2-68
2.21 Calculation of Average Change in Emissions -
NOV Controlled Vehicles 2-69
J\
2.22 Comparison of Predicted and Measured Cold 1972
Federal Emissions - Final Initialization Period
(Test 5A, 5B) 2-71
2.23 Comparison of Predicted and Measured Hot 1972
Federal Emissions - Final Initialization Period
(Test 5A, 5B) 2-72
2.24 Comparison of Predicted and Measured Cold 1972
Federal Emissions - Pre-Deterioration Experiment
Initialization Period (Test 1A, IB) 2-73
2.25 Estimate of Standard Deviation of Influence
Coefficients 2-75
2.26 Fraction of Total Emission Change Due to Change in
Tune Parameter 2-77
2.27 Summary of Estimate of Uncertainty (Standard Devia-
tion) of Average Predicted Emissions 2-78
2.28 Effect of Re-Initialization Tune Parameters -
Pre-Emission Controlled Vehicles 2-80
2.29 Effect of Re-Initialization Tune Parameters -
Emission Controlled Vehicles 2-81
2.30 Effect of Re-Initialization Tune Parameters -
N0₯ Controlled Vehicles 2-82
A
2.31 Effect of Re-Initialization Emission Parameters -
Pre-Emission Controlled Vehicles 2-83
2.32 Effect of Re-Initialization Emission Parameters -
Emission Controlled Vehicles 2-84
2.33 Effect of Re-Initialization Emission Parameters -
NOV Controlled Vehicles 2-85
X
2.34 Comparison of Slopes - Cold 1972 Federal Emissions -
Pre-Emission Controlled Vehicles 2-89
2.35 Comparison of Slopes - Cold 1972 Federal Emissions -
Emission Controlled Vehicles 2-90
2.36 Comparison of Slopes - Cold 1972 Federal Emissions -
N0 Controlled Vehicles 2-91
-------
LIST OF TABLES (CONT'D.)
Title Page
Comparison of Slopes - Engine Tune and Key Mode
Parameters - Pre-Emission Controlled Vehicles 2-93
2.38 Comparison of Slopes - Engine Tune and Key Mode
Parameters - Emission Controlled Vehicles 2-94
2.39 Comparison of Slopes - Engine Tune and Key Mode
Parameters - NOX Controlled Vehicles 2-95
2.40 Parameter Variation Rate - Change in Cold 1972
Federal Emission with Mileage 2-102
2.41 Parameter Variation Rate - Fractional Change in
Cold 1972 Federal Emission with Mileage 2-103
2.42 Parameter Variation Rate - Change in Federal Short
Emission with Mileage 2-104
2.43 Parameter Variation Rate - Change in Key Mode
Emission with Mileage 2-106
2.44 Parameter Variation Rate - Change in Engine Tune
Parameters with Mileage 2-107
2.45 Summary of Mean and 95$ Confidence Limits of
Influence Factors 2-108
2.46 Average Cold 1972 Federal Emission 2-146
2.47 Average Change in Parameter from Test IB - Cold 1972
Federal Emissions and Federal Short Emissions 2-147
2.48 Average Change in Parameter from Test IB - Cold 1972
Federal Emissions and Federal Short Emissions 2-148
2.49 Average Change in Parameter from Test IB - 49/45 MPH
Cruise Mode and Idle Mode 2-149
2.50 Average Change in Parameter from Test IB - 49/45 MPH
Cruise Mode and Idle Mode 2-150
2.51 Average Change in Parameter from Test IB - Tune
Parameters, Total Set 2-151
2.52 Average Change in Parameter From Test IB - Tune
Parameters 2-152
2.53 Average Slope of Fractional Change of Cold 1972
Federal Emissions with Mileage 2-153
2.54 Average Slope of Cold 1972 Federal Emissions with
Mileage and Average Slope of Federal Short Emissions
with Mileage 2-154
vm
-------
LIST OF TABLES (CONT'D.)
Table Title Page
2.55 Average Slope of 49/45 MPH Cruise Emissions with
Mileage and Average Slope of Idle Emissions with
Mileage 2-155
2.56 Average Slope of Tune Parameters with Mileage 2-156
2.57 Linear Regression of Cold 1972 Federal Emissions with
Mileage 2-157
2.58 Linear Regression of Federal Short Emissions with
Mileage 2-158
2.59 Linear Regression of 49/45 MPH Cruise Emissions
with Mileage 2-159
2.60 Linear Regression of Idle Key Mode Emissions with
Mileage 2-160
2.61 Linear Regression of Tune Parameters with Mileage 2-161
3.1 Summary of Repeatability Experiment Emission Response 3-5
3.2 Repeatability Test Analysis of Variance Summary 3-8
4.1 Test Vehicle Description 4-2
4.2 Statistical Summary of the Effect of Soak Temperature
on 1972 Federal Emissions 4-16
4.3 Comparison of Response Coefficients 4-17
IX
-------
1.0 INTRODUCTION
A twenty month vehicle emission and related engine tune-up
parameter deterioration investigation was performed to provide empiri-
cal data for the Economic Effectiveness Model which was developed in the
Study of Mandatory Engine Maintenance for Reducing Vehicle Emissions.
This was the most recently completed in a series of experimental studies
that were conducted as part of the Vehicle Inspection and Maintenance
Study.
The "Experimental Characterization of Vehicle Emission and Tune-
Up Parameter Deterioration" involved a large scale fleet evaluation to
determine vehicle tune-up setting and component deterioration character-
istics with time and mileage. The results of the first year of this
program were reported in Volume IV, "Experimental Characterization of
Vehicle Emissions and Maintenance States," Year End Report, July 1972
(Reference 1). This report presents the procedures, data, and results
for the entire twenty month program.
Interim analyses of data suggested that two sub-experiments were
necessary to properly interpret the data taken during this experiment.
One consisted of the development of the run-to-run repeatability of
emission measurements and the other was the characterization of emissions
measured using 1972 Federal Procedure as a function of the cold soak
temperature.
The Deterioration Experiment is presented in Section 2.0, and
the two sub-experiments investigating measurement repeatability and
the effect of cold soak temperature on emissions are respectively
presented in Sections 3.0 and 4.0.
1-1
-------
2.0 EXPERIMENTAL CHARACTERIZATION OF VEHICLE
EMISSIONS AND TUNE-UP PARAMETER DETERIORATION
2.1 INTRODUCTION
The objective of this program was to determine the deterioration
rates of engine tune-up settings and components and the corresponding
emissions of privately owned and operated vehicles. The deterioration
of emissions and adjustments with both time and mileage were studied
for a period covering sixteen months of typical driving. These de-
terioration rates were required to provide a more accurate data base
in the Economic Effectiveness model for the model's predictive output.
The tune-up settings and components that were to be evaluated
consisted of the following:
1. Basic ignition timing
2. Idle speed
3. Idle air/fuel ratio (% CO)
4. PCV valve and system performance
5. Air filter restriction
6. Primary ignition system condition
7. Secondary ignition system condition
8. NO control system operation
A
9. Choke diaphragm or piston setting
10. Heat riser valve operation
11. Air injection system performance
The exhaust emission data that were required included measurement of
unburned hydrocarbons, carbon monoxide, and oxides of nitrogen. Exhaust
gas mass emissions were measured using both the current Federal test
procedure and a short cycle. Emission concentration measurements were
also made in selected engine operating modes.
2-1
-------
2.2 TEST PROCEDURE
Four hundred fifty (450) vehicles were initially obtained from
private owners for testing. The vehicles were subsequently recalled
for testing at scheduled four-month time intervals. Four recalls were
scheduled, thus providing five test periods covering sixteen months of
owner operation for each vehicle. Of the original 450 vehicles, 413,
392, 367, and 330 were obtained for each subsequent test. Testing was
initiated in May of 1971 and completed in January of 1973.
Engine tune-up adjustments and component replacements as well as
vehicle exhaust emissions were quantitatively measured at each scheduled
test. The following measurements were made on each vehicle at the
four-month intervals:
o Quantitative measurement of engine components,
parameters, and settings that were shown in
Phase One of the CAPE-13-68 Program to be
significant in an emission control program.
o Exhaust emissions using the 1972 Federal test
procedure and the Federal short cycle. During
the eighth month of the program the 1972 Federal
test procedure was replaced with the 1975 pro-
cedure.
o Closed, hot seven-mode cycle CVS mass emissions,
at the time of the first test.
o Concentration of exhaust emissions at idle and
under loaded engine operating conditions.
The total fleet was divided equally into three groups of 150
vehicles based upon model years which reflected the extent of their
emission control equipment. The basic description of the fleets were
Pre-Emission Controlled Vehicles (pre 1966), Vehicles with HC and CO
exhaust emission controls (1966-1970), and NOX Emission Controlled
Vehicles sold in California during the 1971 model year.
2-2
-------
To characterize deterioration of engine parameters with both time
and mileage accumulation, it was necessary to establish a baseline from
which deterioration could be monitored. This was accomplished by adjust-
ing or replacing components in accordance to the manufacturer's specifi-
cations at the beginning of the deterioration test period. The emission
and engine diagnostic tests were made both before and after the initial
tune-up. After the vehicles received their last test for deterioration,
they were again returned to specifications and retested.
2.2.1 Vehicle Sample Identification and Acquisition Procedures
Fleet Acquisition
Three fleets of 150 vehicles were each selected and acquired
by Scott. Candidate vehicles were first identified from a file of Scott
questionnaires. Only the questionnaires returned by owners who indicated
an interest in participating in an experimental program were considered.
During the first four months of initial vehicle acquisition, the owners
of needed vehicles were contacted, given an explanation of the program
and incentive provided, and if they were willing to participate, their
vehicle was scheduled for testing.
The owners of vehicles used in the program were always
supplied with a 1971 loan vehicle while their car was under test. In
addition, maintenance of their vehicle was performed at the beginning and
end of the program at no cost. A cash and gasoline incentive was also
given to owners at each test interval when no tune-up was performed. The
cash incentive was $15 for 1971 model year vehicles and $10 for all earlier
model year vehicles. Each owner's vehicle received five gallons of premium
grade gasoline upon completion of tests.
The vehicles that were supplied as loan vehicles were new,
major manufacturer, 1971 intermediate size vehicles, equipped with stand-
ard V-8 engines and automatic transmissions. There were twenty of these
loan vehicles and they were included in the California NO Controlled
/\
vehicle fleet.
2-3
-------
Fleet Identification and Description
Vehicles included in the experimental program were limited
to passenger vehicles and light duty trucks which were classified in one
of three categories. Each category consisted of 150 vehicles at the in-
itial implementation of the program (test phase 1). The specific classi-
fications of each category were:
Fleet I - Vehicles with no exhaust emission controls (pre-emission
controlled vehicles), which included all 1960 through
1965 model year vehicles and 1966 and 1967 model year
vehicles that were originally purchased outside California.
(The majority of vehicles were originally purchased in
California and were equipped with Positive Crankcase
Ventilation Systems.)
Fleet II - Vehicles equipped with exhaust emission controls (emission
controlled vehicles), which included California vehicles
from model years 1966'through 1970 and out-of-state pur-
chased vehicles from model years 1968 through 1970.
Fleet III - Vehicles originally sold in California for the 1971
model year (NO controlled vehicles), which were
required to meet NO emission standards.
/\
Vehicles of various manufacture were obtained in proportion
to the national in-use population, within the fleet groupings, as deter-
mined by vehicle registration data provided in the most recent Automotive
News Almanac, (Reference 2). The proportioning was based on vehicle make
and model year and within attainable goals; vehicle size, transmission
type and number of engine cylinders. A description of the initial pro-
gram goals for the distribution of makes within each fleet is presented
in Table 2.1.
The distribution of vehicle makes was controlled for only
U.S. makes and Volkswagens. All other foreign vehicles were combined
into the miscellaneous category. The percentage of foreign vehicles
(Volkswagens included) was to be limited to 10 percent of Fleets I and II
and to 20 percent of Fleet III. Deviations from the above criteria were
required due to the low number of available 1960 through 1962 vehicles
as well as the low availability of American Motors Corporation and
Oldsmobile vehicles for most years.
2-4
-------
Table 2.1
In-Use Percent of Vehicle Population by Make and Number Desired
for Each Fleet Classification
Fleet
#1
Uncontrolled
Vehicle Make
Buick
Cadillac
Chevrolet
Chrysler
Dodge
Ford
Mercury
Oldsmobile
Plymouth
Pontiac
Rambler
Vol kswagen
Miscellaneous
TOTAL
%
5.6
2.0
24.8
1.8
4.8
19.8
3.8
6.0
5.6
7.7
4.2
4.0
9.9
100
Number
9
4
41
3
8
33
6
10
9
13
7
7
0
150
Fleet
#2
Controlled
%
6.1
2.1
20.1
2.0
5.6
18.3
3.4
5.9
7.0
7.8
2.6
6.3
12.2
100
Number
11
4
35
3
10
31
6
10
12
13
4
11
0
150
Fl
1971
%
5.8
1.9
20.2
2.5
6.3
22.8
4.2
6.1
8.5
6.0
3.3
4.2
8.2
100
eet #3
Cal i form" a
Number
10
3
33
4
10
37
7
10
14
10
5
7
0
150
2-5
-------
The ratio of automatic to manual transmissions was matched
to production statistics for each fleet. The percentage of six cylinder
and V-8 engines was maintained only across the entire 450 vehicles sample.
The number of vehicles in Fleet II that were equipped with Air Injection
emission control systems was limited to 20 percent of that fleet.
Initial Vehicle Scheduling and Control
Vehicles were scheduled for testing on a continuous basis
during the first four months of testing. Once these vehicles were ob-
tained they were processed according to the Test Vehicle Processing
Sequence illustrated in Figure 2.1. The specific details of the sequence
elements are described as follows:
Sequence Element 1 - Owners were contacted as vehicles were re-
quired and appointments were made for delivery of the
vehicles to the Scott facility. Each vehicle was in-
spected by Scott personnel to determine the condition
of the body sheet metal, interior trim and mechanical
items. The exhaust system of the vehicle was leak
checked by blocking the tail pipe outlet. Repairs were
made if leaks were detected in the exhaust system.
Sequence Element 2 - Vehicle numbers were assigned on the following
basis: Fleet I - 1 through 150, Fleet II - 151 through
300, and Fleet III - 301 through 450. A data package
folder was placed in and remained with the vehicle during
test and inspection. This folder was kept on file until
the next time the vehicle was recalled for testing. The
Vehicle Control Log, Figure 2.2, was included in the
folder. The vehicle identification items 1.1 through 1.8
and scheduling information items 2.3 and 2.4 were filled
out. The Engine Parameter Inspection Forms
Section 2.2.2 were also included. All forms
fied with the vehicle and test numbers. The
then parked in a designated cold soak area.
Identification Sheet was placed in the vehicle to identify
the vehicle and test number along with the date and start-
ing time of the soak period.
Sequence Element 3 - While the vehicle was being cold soaked, the
power train was identified by use of engine codes and
carburetor numbers. Items 1.9 through 1.14 of the
described in
were identi-
vehicle was
A Vehicle
2-6
-------
Obtain Vehicle From Owner
Inspect Vehicle
a. General Condition
b. Exhaust System
Repair Exhaust System if
Necessary
Assign Vehicle Number
Fill in Vehicle Identification and
Scheduling Information on Form 1.2
Put Data Package Folder in Vehicle
Park in Soak Area
Attach Vehicle Identification Sheet
Record Shut Down Time and Date
Identify Power Train
(Items 1.9-1 .14 on
Form 1 .2)
Determine Dyno Settings _.
Identify and Order Tune-Up
Parts
Enter Tune-up Specs on
Form 2.5
Complete "As-Received"
Emission Test and Dyna-
mometer Inspection
Complete "As-Received" Floor
Inspection (Form 1.3)
no
--j
Tune-Up Vehicle
Perform Post Tune-Up
Floor Inspection (Form 1.3)
Park in Soak Area
Identify Shut Down Time
and Data
Complete "As-Tuned"
Emission Test and
Dynamometer Inspection
Clean Vehicle
Remove Data Package
Folder
Return Vehicle to Owner
- Elements Required for Recall Tests Except Test 5
FIGURE 2.1 TEST VEHICLE PROCESSING SEQUENCE
-------
Figure 2.2
Vehicle Control Log
FLEET DETERIORATION PROGRAM
FORM 1.2
Vehicle Control Loq yp<;t Nn 1
205
Car No ?
Curb Weight 3750 + 300 Ibs = 3 4050
Inertia Weiaht 4000 InHJrated HP 8.0
1. Veh
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
Icle Identification
Y~,r ,, 68 1.9 EnqineSize, CID 6 383
Make 5 Plymouth 1.10 Horsepower 7 330
Model RoadRunner !-n Carb Make and Bbls ft Holley_, 4
Color Blue 1.12 Transmission Type 9 Auto
License No. ZZT 088 ]-13 Exn- Em'ss- SXS- 10 C.A.P.
Owner's Name J. Welstand ] l4 Crankcase Emiss. Sys. n 4
Telephone No. 862-7757
Address 26021 18th Street, San Bernardino. California
2. SCHEDULING INFORMATION
2.1 Next Due Date 10/28/71 2/28/72 6/28/72
2.2
2.3
2.4
2.5
2.6
2.7
2.8
Test No. IA B 2 3 4 5A 5B
Odometer Reading 12 35414 38185 || 40327
Date Obtained 13 6-22-71 1 1 -1 -71 1 1 2-24-721
Vehicle Checked ^./ &8. M-t
Tune-Up Completed iX/
Inspection Completed W / w s^C 1\K I
Test Competed ~W 3
-------
Table 2.2
Crankcase Ventilation System Type Codes
1. PCV Valve Only - Controlled by Manifold Vacuum (Open System)
2. Valve Only - Controlled by Crankcase Vacuum
3. Hose from Crankcase to Air Cleaner Only (Like VW)
4. Combination System - PCV Valve and Hose to Air Cleaner
(Closed System)
Vehicle Control Log were completed. The crankcase
emission system type was identified and coded as
described in Table 2.2. The codes, 1 through 4.
are consistent with the nomenclature used for
California vehicle emission installation and in-
spection stations, (Reference 3). The vehicle
curb weight was determined from specifications
listed in the Automotive News Almanac. The
appropriate dynamometer inertia and indicated horse-
power setting was then determined in accordance with
the 1972 Federal test procedure. These settings
were entered at the top of the Vehicle Control Log.
The vehicle drive train description was used to
identify and order the basic tune-up parts. The tune-up
specifications were also determined and entered on the
Engine Parameter Inspection Forms (Section 2.2.2).
Sequence Elements 4,
5, 6, 7, and 8 - The required emission tests, parameter
inspections and tune-up were then performed in the
sequence depicted in Figure 2.1. The details of these
elements are fully described in Sections 2.2.2 and 2.2.3.
Sequence Element 9 - Upon completion of all required testing,
the vehicle was washed, the data package folder was
removed, and the owner was contacted to arrange for
the return of the vehicle.
Recall Scheduling and Control
The first test sequence was designated as Test 1A and Test IB for
the "as-received" and "post tune-up" tests, respectively. The four month recall
tests were sequentially numbered Tests 2 through 5. The recall tests 2 through
4 required only the elements 1, 2, 4, 5, and 9 of the Test Vehicle Processing
Sequence, Figure 2.1. The final test sequence designated as Test 5A and 5B was
performed in a manner similar to the first test sequence.
2-9
-------
Vehicle recalls were scheduled at approximate four-month intervals
from the initial test date. Minor deviations from the four-month interval
were reqjirea due to the specific availability of vehicles from their owners
and for efficient usage of facilities and time. At each recall the vehicle
information such as mileage and owner maintenance was recorded on the appro-
priate forms. Additional procedures were implemented during the program
because an increasing number of vehicles received owner provided maintenance
as the program progressed. The procedures for identifying unscheduled main-
tenance and repairs are presented in Section 2.2.2.
The additional procedures which were implemented during the test program
for vehicles that received unscheduled maintenance were designed to maximize
the useful data obtained for the basic program. During Tests 2 and 3 those
vehicles receiving minor tune-up repairs and adjustments were processed
according to the basic test plan. The repairs and adjustments were documented
for future data analysis. Those vehicles that received major or extensive
engine maintenance were reinitialized. Reinitialization consisted of assign-
ing a new number to the test vehicle, performing an initial tune-up and/or
inspection, and starting the test series over, beginning with Test 1.
/
At Test 4 it was found that a substantial number of vehicles were re-
quiring tune-ups after twelve months of operation. It was deemed advisable
on these vehicles to perform the deterioration test, the tune-up, and "post
tune-up" test at this time. The vehicles were assigned new numbers as in the
previous cases. In addition, these vehicles were recalled at the fifth test
period for one additional test which would be similar to Test 2 of the original
sequence. This was possible since the tune-up and emission test at Test 4
served as an initialization. Also, vehicles that had received unauthorized
maintenance since the previous test were processed in the same manner - "as-
received" test, tune-up, "post tune-up" test, assignment of a new vehicle
number, and scheduled recall for one additional Test 2. A Fleet Deterioration
Vehicle Scheduling Control, Form 1.5 (Figure 2.3} was used at Test 4 to facili-
tate decision making by test personnel, assure proper handling as outlined
above, and document the work performed. At test period 5, a similar form was
used as a follow-up and for the same purposes as at Test 4. This Form is
shown in Figure 2.4.
2-10
-------
FLEET DETERIORATION VEHICLE SCHEDULING CONTROL
FORM 1.5
SITUATION
GO AHEAD
AUTHORIZATION
PERFORM
REGULAR
TEST AS
RECEIVED
WORK REQUIRFD
PROJECT 2037-01
CAR NO.
LICENSE NO.
DATE
PERFORM
SPECIFIED
TUNE-UP
PERFORM
POST TUNE
TEST
REPAIR
NON-SPECIFIED
COMPONENTS
PERFORM
INITIALIZING
TEST
FUTURE
SCHEDULING
COMMENTS
1. NO MAINTENANCE,
COMPLAINTS,
OR PROBLEMS
SINCE LAST TEST
2. CAR IS NO LONGER
ACCEPTABLE - POOR
CONDITION
3. CAR WILL NO LONGER
BE AVAILABLE
4. OWNtl O* SRL HAS
PERFORMED MAIN-
TENANCE SINCE
LAST TEST
5. OWNER HAS COMPLAINT
AT TIME OF TEST
DOES IT REQUIRE TUNE-UP OR ADJ?
NO
NO TUNE-UP
NEW CAR NUMBER
TEST IB, INSPECTION,
AND NEW CAR NUMBER
TEST 2, 3, OR 4, AND
INSPECTION. DATA EQUIVALENT TO TEST 5A
INDICATE ACTION TO BE TAKEN
Figure 2.3 Test 4 Vehicle Scheduling Control
-------
FLEET DETERIORATION VEHICLE SCHEDULING CONTROL
FORM 1.5
WORK REQUIRED
PROJECT 2037-01
CAR NO.
LICENSE NO.
DATE
SITUATION
GO AGEAD
AUTHORIZATION
PERFORM
REGULAR
TEST AS
RECEIVED
PERFORM
SPECIFIED
TUNE-UP
PERFORM
POST TUNE
TEST
REPAIR
NON-SPECIFIED
COMPONENTS
PERFORM
INITIALIZING
TEST
FUTURE
SCHEDULING
COMMENTS
ro
2.
NO MAINTENANCE,
COMPLAINTS, OR
PROBLEMS SINCE
LAST TEST
CAR IS NO LONGER
ACCEPTABLE - POOR
CONDITION
i. CAR WHICH HAS
RECEIVED 5A, 5B
TEST
4 OWNER OR SRL HAS
PERFORMAD MAIN-
TENANCE SINCE
LAST TEST
TEST 3, 4, O
AND INSPEC
TEST 2
i
1
1
[
1
*^"\
-v >
S 5A 1 TEST 5B AND /'*' "T YES
INDICATF ACTION
ORDER PARTS IF YES
AND INSPECTION
~\
1
1
MO J^ N^NE /
\ /
( END OF TEST
Jv. MONF /
( END OF TEST
i J
INSPECTION ONLY
(END OF TEST )
NOTE: OWNER COMPLAINTS WILL BE CORRECTED AFTER TEST 5B OR 2.
Figure 2.4 Test 5 Vehicle Scheduling Control
-------
2.2.2 Vehicle Inspection and Tune-Up Procedures
Inspection Procedures
A comprehensive engine parameter adjustment and component
inspection was performed in conjunction with each scheduled emission test.
The engine inspection was designed to quantitatively evaluate those engine
settings and components that affect vehicle emissions. The inspection
covered the ignition system, induction system, and emission control devices.
Measurements o.f ignition misfire, air cleaner blockage, PCV system per-
formance and air injection pump system performance were made with the en-
gine under load on a chassis dynamometer. Figure 2.5, Engine Parameter
Inspection, was used to record all engine inspection information.
The procedures used for obtaining the measurements recorded
on the Engine Parameter Inspection form are outlined below and relate to the
item numbers on Figure 2.5. Figure 2.6 shows an acceptable ignition pattern
and its nomenclature.
(2.) Ignition System Inspection
(2.1) Required Voltage
The parade display of the ignition scope is
selected with the engine operating at 1500 rpm, no load. The minimum
and maximum firing line voltage readings are recorded.
(2.2) Coil Available Voltage
The same operating conditions as described in
item 2.1 are employed, except one secondary ignition wire is disconnected
for an open ignition circuit condition. The maximum voltage output that
is observed for the open circuit cylinder is recorded.
(2.3) Spark Line
With the engine still operating at 1500 rpm and
the transmission set in neutral, the scope is set for stacked or raster
display. The spark lines are rated OK if they are generally clean and
level. Excessive slope or noise for any cylinder is rated NO (no good).
2-13
-------
FLEET DETERIORATION PROGRAM
FORM 1.3
Engine Parameter Inspection
Performed By
SCOTT RESEARCH LABORATORIES, INC.
Vehicle Identification
1.1 Test No.
1.2 Car No. 2
1 . 3 License N
1.4 Inspected
1.5 Date
il 1A
205
o. ZZT °88
By JH
6-24-71
Ignition System Inspection
2.1 Required Voltage, kv at 1500 rpm
2.2 Coil Available Voltage, kv at 1500 rpm
2.3 Spark Line (OK, NG)
2.4 Coil Oscillations (OK, NG)
2.5 Point Opening Variation, degrees
2.6 Coil Polarity (OK, NG)
2.7 Ignition Point Dwell, degrees
2.8 Condenser Oscillations (OK, NG)
2.9 Basic Ignition Timing, degrees
2. 10 Total Advance at 2500 rpm, degrees
2.11 Mechanical Advance, at 2500 rpm, degrees
2. 12 Vacuum Advance at 2500 rpm, degrees
Induction System
3.1 Idle Speed, rpm(Chrys. in Neutral) N DrX_
3.2 Manifold Vacuum, in. Hg.
3.3 Air Cleaner Angle, degrees
3.4 Float Level, inches*
3.5 Choke, Vacuum Kick, inches
3.6 Choke Vacuum Diaphragm (OK, NG, None) ^
3.7 Heat Riser Valve (None, Free, Frozen)
On parking lot survey only
3
5
7
8
9
10
11
13
14
16
18
20
Measurement
or Analysis
5-8
36
NG
NG
2
OK
35
OK
BTDC
50
28
22
kv
kv
o
o
o
o
0
o
12
15
17
19
21
Manufacturer's
Specification
kv
kv
27-32
BTDC
41-50 1/2
23-26 1/2
18-24
1
3
5
7
9
1
2
650
15
0
X
.081
OK
FREE
rpm
6
8
10
600
.081
rpm
it
Figure 2.5 Engine Parameter Inspection
2-14
-------
FLEET DETERIORATION PROGRAM
FORM 1.3 (Continued)
Emission Control
4. 1 PCV Perf. at Idle, inches H2O**
4.2 Vacuum Leaks (Yes or No)
4.3 Idle rpm change (Leaks Eliminated)
4.4 NO Control Device (Ok, NG, None)
s\.
4.5 Timing Retard Mechanism (OK, NG, None)
13
15
16
17
Measurement
or Analysis
-.1
NO
NONE
NONE
Keymode Diagnostic Inspection
Dyno Load Set to 30 HP at 50 MPH
5.1 49/45 MPH Cruise
o Plug Req'd Volt, kv
o Misfire Rate, %
o Air Cleaner Restriction, in H^O
o PCV Flow, inches
o Air Pump Disconnected, Emissions
Completed By
rpm
Manufacturer's
Specification
2
3
7-9
0
.2
.40
kv
5.2 33.5/30 MPH Cruise
o Plug Req's Volt, kv
o Misfire Rate, %
o Air Cleaner Restriction, in
o PCV Flow, inches
o Air Pump /Disconnected, Emissions
5
6
7
8
Co
8-10
0
.1
.40
npleted By
kv
%
II
II
1
5.3 Idle (in Drive)
o Plug Req'd Volt, kv
o Misfire Rate, %
° Air Pump Disconnected, Emissions
9
10
lor
12-15
0
npleted By
kv I ]
% (fei) |
REMARKS: High point resistance
**
Vacuum is minus (-), and Pressure is plus (+)
Figure 2.5
2-15
Engine Parameter Inspection
(Continued)
-------
PO
03
CD
tO
Firing
Line
(Required
Voltage)
Zero Line
Firing
Section
Spark
Line
Intermediate
Section
Coil Oscillations
Duration
Dwell
Section
Point
Closing
Point
Opening
Figure 2.6 Engine Analyzer Ignition Pattern
-------
(2.4) Coil Oscillations
Four or more oscillations are required for an OK
rating. The ignition scope and engine rpm are set as described for
item 2.3.
(2.5) Point Opening Variation
The engine is operated as described above and
the scope is set for a superimposed display of cylinder ignition patterns.
The difference in timing (point opening) between the cylinder with the
earliest timing and the cylinder with the latest timing is recorded.
(2.6) Coil Polarity
If the ignition pattern is reversed on the
vertical scale (upside-down), the coil polarity is reversed and rated NG.
(2.7) Ignition Point Dwell
The conventional (average point) dwell is recorded
with the engine at idle.
(2.8) Condenser Oscillations
Oscillations must be observed and the first oscilla-
tion must have the greatest magnitude for an (OK) rating.
(2.9) Basic Ignition Timing
The distributor vacuum line is disconnected and the
timing is recorded in accordance with the manufacturer's specified pro-
cedure.
(2.10) Total Advance at 2500 rpm
The engine is run at 2500 rpm in neutral and the
total advance in ignition timing from the basic timing is recorded.
(2.11) Mechanical Advance at 2500 rpm
The same operating conditions as 2.10 are employed,
except that the distributor vacuum line is disconnected. The advance in
ignition timing from the basic timing is recorded.
2-17
-------
(2.12) Vacuum Advance at 2500 rpm
The reading for 2.11 is subtracted from the
reading for 2.10 to obtain the value for the vacuum advance.
(3.) Induction System
(3.1) Idle Speed
The idle rpm is measured with a tachometer
according to the manufacturer's specified procedure (includes proper
operation of transmission, air conditioning, evaporative control
system, headlights, etc.).
(3.2) Manifold Vacuum
The manifold vacuum is obtained with the
engine idling and the transmission set in neutral.
(3.3) Air Cleaner Angle
The vehicle's air cleaner element is placed
on a commercial air cleaner tester and the relative restriction is
indicated in angular degrees.
(3.4) Float Level
Not measured
(3.5) Choke Vacuum Kick
The choke vacuum diaphragm or piston setting
is measured according to the manufacturer's procedure with a gauge pin,
Various nomenclatures for this setting are Vacuum Kick, Vacuum Break,
Choke Plate or Valve Pulldown, Choke Opening, Choke Vacuum Piston,
Intermediate Choke Rod and Initial Choke Opening.
(3.6) Choke Vacuum Diaphragm
On those vehicles equipped with a choke vacuum
diaphragm, vacuum is applied to the diaphragm. If the diaphragm
does not retract, it is failed and NG is recorded. OK is recorded
if it is operational.
2-18
-------
(3.7) Heat Riser Valve
On vehicles equipped with an exhaust heat riser
valve the valve is inspected manually for freeness of operation. If
the operation is not free, a FROZEN rating is entered. Acceptable
operation is rated OK.
(4.) Emission Control
(4.1) PCV Performance at Idle
With all the vehicle's PCV system components
properly installed, the crankcase vacuum is measured at the oil dipstick
tube. The vacuum or pressure is recorded in inches of water. The fresh
air inlet to the PCV system is not blocked off with this procedure.
(4.2) Vacuum Leaks
Accessory vacuum hose leaks are diagnosed at idle
by pinching off each hose individually near its source of vacuum. A
leak is detected if the idle quality changes noticeably (or the idle rpm
changes).
(4.3) Idle rpm Change
If an idle rpm change is detected in item 4.2,
the difference in rpm is recorded.
(4.4) NO Control Device
/^
Vehicles that have a distributor vacuum advance
that is controlled by either speed or transmission gear engagement
are evaluated for proper operation of the vacuum advance control.
The vehicle is gradually accelerated from idle to about 40 mph on the
chassis dynamometer. If the vacuum advance control does not operate
according to the manufacturer's specifications within reasonable
limits, the device is rated NG.
2-19
-------
(4.5) Timing Retard Mechanism
Dual diaphragm distributors and solenoid
operated retard mechanisms are evaluated for proper operation.
Either the retard side vacuum hose or electrical connection is dis-
connected and the change in ignition timing is observed. If there
is a shift in ignition timing in the correct direction, the unit is
rated OK.
(5.) Keymode Diagnostic Inspection
An evaluation of firing voltage, ignition misfire,
air cleaner restriction, PCV system flow, and air injection pump per-
formance is made using the Clayton Keymode Cycle diagnostic modes,
(Reference 4). The Clayton Keymode Cycle consists of steady state cycles
at high engine load, moderate engine load and idle. The Keymode cycle
is described in Table 2.4, Section 2.2.3. All of the performance evalu-
ations are made during the high and low cruise. In addition, the ignition
system, firing voltage and air injection performance measurements are
performed at idle. The measurement techniques that are employed are
described as follows:
a) Spark Plug Required Voltage
The ignition analyzer scope is connected and set to
the parade display. The minimum and maximum observed firing voltage
readings are recorded.
b) Misfire Rate
With the same ignition scope set-up that is described
in a), misfire is considered to be present if an open circuit is observed
in the firing line and no spark line is present. The percentage is based
on the total number of spark plug firings that are available, i.e., if
one spark plug in an eight cylinder engine is misfiring all of the time
the misfire rate is 12.5%.
2-20
-------
c) Air Cleaner Restriction
The complete air cleaner assembly and attachments are
installed in their normal arrangement. The air cleaner hold-down wing
nut is removed and the end of a hose is butt connected at the air cleaner
housing opening around the attachment stud. The other end of the hose
is connected to a "U" tube water manometer. The air cleaner housing
vacuum that is observed is recorded for the two loaded engine conditions.
d) PCV Flow
A laminar element flow gauge is connected to the up-
stream side of the Positive Crankcase Ventilation Valve and hose.
The downstream side of the PCV Valve is connected in its normal operating
configuration. The flow rate is recorded for the two loaded engine
conditions.
e) Air Pump Disconnected
The air injection pump is disconnected from the air
distribution manifold and the manifold inlets are plugged. The direct
concentration exhaust emissions of carbon monoxide, carbon dioxide, and
hydrocarbons for the three operating modes are monitored and recorded
on standard strip charts. The exhaust emissions at this condition are
compared to the exhaust emissions obtained during the standard emission
test sequence when the air injection pump is normally connected.
Any deficiencies and pertinent items that are observed
beyond those recorded on the standard inspections sheet are recorded
under "Comments."
The vehicle parameter inspection was expanded during the seventh
month of the experimental program (midway through the first recall). The
measurement of idle carbon monoxide emissions during the engine parameter
inspection (floor inspection) was added. This measurement was performed
on Vehicle Fleets II and III to verify the fuel/air mixture adjustments
made during initialization of the vehicles to manufacturers' specifications,
2-21
-------
This additional measurement was initiated to establish the relationship
between the Keymode idle CO and the idle CO obtained during a garage
floor measurement. The original initialization of idle CO was made under
the garage conditions.
The other parameter diagnostic procedure that was added was the
performance of a cylinder balance check. The cylinder balance check is
performed with the ignition analyzer scope by sequentially shorting
ignition of individual cylinders, i.e., there is no ignition spark for a
selected cylinder. The ignition scope provides a reading of percent loss
of engine speed for each cylinder shorting. This test was performed at
the low speed cruise of the Keymode Cycle and at 1500 rpm in neutral
during the floor inspection. This procedure was undertaken due to the
indication that ignition misfire was not being accurately detected by
the previously discussed procedure.
Vehicle Initialization Procedures
Each vehicle in the fleet received a comprehensive engine tune-up
after the "as-received" emission test and parameter inspection had been
performed. This tune-up was performed in order to initialize the
vehicle so that the deterioration of new components and adjustments
with mileage and time could be monitored. The baseline or reference
emission level prior to deterioration was measured in test IB which was
performed immediately after vehicle initialization.
Regardless of their condition, each vehicle was equipped with new
spark plugs, ignition condenser, PCV valve and air cleaner element.
In every case, idle rpm, idle fuel/air ratio, basic timing, ignition
point dwell and the choke vacuum kick were set to manufacturers' specifi-
cation. In many cases, failed distributor and choke vacuum diaphragms
were detected and therefore replaced. All stuck heat riser valves were
freed. The secondary ignition components, i.e., high tension cables,
rotors, and distributor caps were replaced whenever they were found to be
in poor condition. Occasional major carburetor and distributor repairs
2-22
-------
were performed in order to obtain acceptable vehicle operation. The
tune-up parts used were major Original Equipment Manufacturer (OEM)
brands.
All of the mechanical settings were accomplished using conventional
mechanical and electrical equipment and standard garage procedures.
The distributor advance specifications were obtained from the Sun
Specification Service Manual, (Reference 5), and all other tune-up
specifications were obtained from National Service Data manuals,
(Reference 6). The settings of idle rpm and basic timing were made
in strict compliance with the manufacturers' published procedures.
The idle mixture settings were set for "best lean idle" on the vehicles
with no emission controls (Pre-Emission Controlled Vehicles, Fleet I).
"Best lean idle" was accomplished by leaning the mixture until a lean
engine roll or rpm drop was observed and then just richening the mixture
to return the idle rpm back to the value prior to the drop. Emission
Controlled Vehicles (Fleet II and Fleet III) idle settings were made
using the most definitive procedures known. These procedures resulted
in either a % CO setting or best idle rpm drop setting. The specifica-
tions that were employed are listed in Table 2.3.
2-23
-------
Table 2.3
Idle CO % Specifications for
Vehicles Equipped with Exhaust Emission Controls
General Motors - Follow engine compartment or tune-up manual procedures.
These procedures are based on rpm drop with the mixture
screw. 1971 Chevrolet, Pontiac and Oldsmobile have % CO
specifications listed in the tune-up manual.
Chrysler Corporation - Set 1.0% CO at the specified rpm with transmission
set in neutral.
Ford Motor Company - Set the idle % CO at the specified rpm with the
transmission in neutral.
The % CO specifications are listed in the Ford (Reference )
Emissions Analyzer Manual. The specifications listed for
1968 model vehicles are applicable to 1967 and 1966 Cali-
fornia Vehicles. The idle % CO setting for 1971 vehicles
is listed in the tune-up manual.
American Motors - 1968 through 1971 vehicles are set from 1.0 to 1.5 % CO.
1966 and 1967 California vehicles are set for best lean idle.
All Others - Set to whatever specifications are available in the tune-up
manual.
2-24
-------
Unscheduled Maintenance and Repairs
The program goal was that a minimum number of vehicles
would receive additional tune-up maintenance during the course of the
vehicle deterioration test program. All participating vehicle owners
were informed at the beginning of the program that their vehicles
would undergo an additional major tune-up at the end of the test series,
sixteen months from the time of their vehicle's first test. It was
anticipated that the vehicle owner would therefore tend to delay his
routine, periodic tune-up maintenance until the end of the test series.
During the period of the first recall testing (Test 2), it was discovered
that some owners had obtained various and sometimes extensive tune-up
related repairs. Therefore, at the. beginning of the second recall test-
ing (Test 3) a letter was sent to the vehicle owners reminding them of
the scheduled tune-up at Test 5. However, extensive maintenance and
even complete tune-ups continued to be performed on many of the vehicles.
Vehicles were also returned to Scott at either the scheduled recall or
between tests with various performance complaints. These complaints,
however, were corrected only when judged to be necessary. A coding
system was established to maintain a complete record of all unscheduled
maintenance and to facilitate automatic data processing. The Engine
Parameter Adjustment/Repair Record (Figure 2.7) was used to document
parameter adjustments and repairs that were performed at the Scott
Research Laboratories facilities.
Due to this greater than expected degree of unscheduled
maintenance, additional procedures were instituted in order to maximize
the usefulness of data obtained from vehicles which received unscheduled
maintenance. As described in Section 2.2.1, during the first and second
recalls (Tests 2 and 3) the vehicles that had received excessive mainten-
ance were reinitialized and retested. In cases where only minor adjustments
were required, the adjustments were documented along with the corresponding
before and after diagnostic readings. Reinitialization involved retuning
the vehicle as described in Section 2.2.1 , assigning a new vehicle number
2-25
-------
Fleet Deterioration Program
Form 1.4
Engine Parameter Adjustment/Repair Record
1. Vehicle Identification
1.1 Last Test No.
1.2 Car No.
1.3 License No.
1.4 Repaired by
1.5 Date
1.6 Odometer
2. Complaint or Problem
"*-*" -
n^e&AjnrL stsAaAstA
3. Repair(s) or Adjustment(s) Made
Ar,
4. Diagnostic Readings
Item Adjusted
2
3
4
5
Measurement Before
Adjustment
Measurement After
Adjustment
Figure 2.7 Engine Parameter Adjustment/Repair Record
2-26
-------
to that vehicle, and reinstituting the vehicle in the program at Test IB.
At Test 4, vehicles that had received any tune-up related maintenance
since Test 3, and those vehicles that required maintenance due to owner
complaints were given their final "as-received" test, the specified
tune-up, and their "post tune-up" test. Since this procedure was equiva-
lent to reinitialization, these vehicles were assigned a new number and
recalled at the fifth test period for an additional Test 2.
Those vehicles that required testing under the alternate
procedures due to excessive maintenance were identified by owner
volunteered information or from records of Scott performed interim
maintenance. An inquiry was made of each owner at the scheduled recalls
to determine whether he had any tune-up related work performed on his
vehicle since the last test. All positive responses were recorded on
appropriate forms. It was apparent that complete identification of
interim maintenance was not being obtained by owner inquiries. Addi-
tional techniques were instituted throughout the test program to
identify additional maintenance as often as possible.
At the beginning of the third test period the carburetor
mixture and speed screws were painted with nail polish to identify their
position. At the subsequent recalls the mechanics inspected the screws
during the floor diagnosis. The results of this inspection were recorded.
At the beginning of the fourth test period the replaceable tune-up
parts were color coded with paint. These parts were retained after
their removal at the final tune-up for evaluation. The evaluation of
the tune-up parts was made by inspecting them for paint coding, their
condition, and their part number and manufacturer's brand. Most of the
original parts were listed by manufacturer and part number on individual
requisitions by car number. All probable discrepancies were noted.
In addition to these parts inspections the tune-up parameter
inspection sheets for each individual vehicle were screened. Pertinent
parameters such as timing, dwell, idle rpm, idle CO, air cleaner measure-
ments, and spark plug required voltage were compared for the sequential
2-27
-------
tests. Suspicious discrepancies were noted and occasionally a follow-up
inquiry was made of the owner. When unscheduled maintenance was detected,
the results were recorded as additional tune codes for data processing.
Only those determinations of unreported maintenance that were considered
to be positive were recorded. There were approximately 100 vehicles
where additional maintenance was suspected, but where the probability
was not considered to be high. Tune codes were not entered for these
vehicles.
Final Tune-Up Procedures
At the time of the final tune-up the same procedures used
for the original vehicle initialization were employed except for the
postponement of some maintenance. Only those tune-up adjustments and
component replacements that are characterized in the Economic Effective-
ness Model were made. These characterized tune-up elements are de-
scribed as follows:
1. Basic ignition timing
2. Idle speed
3. Idle air/fuel ratio (% CO)
4. PCV valve and system
5. Air filter element
6. Primary ignition system
7. Secondary ignition system
8. NO control system
9. Choke piston setting
10. Heat riser valve
11. Air injection system
Only those NO system components were maintained that controlled vacuum
/\
spark advance and only when replacement parts were immediately available.
The air injection system performance was not diagnosed at Scott and was
not repaired, if defective.
The failed components that did not fall into the above
categories were repaired after the performance of the "post tune-up"
emission test. These repairs included such items as carburetor overhaul
or replacement, distributor replacement, replacement of failed choke and
2-28
-------
distributor vacuum advance diaphragms, and other choke mechanism repairs
or adjustments. An additional "non-specified maintenance" emission test
and engine diagnosis was performed after repair of these maintenance
items for information. The application of these procedures would allow
for comparison of actual versus the Economic Effectiveness Model pre-
diction of emission response to tune-up.
2.2.3 Exhaust Emission Tests
Exhaust emission measurements were performed on each vehicle before
and after the initial tune-up and at each scheduled four-month recall.
The "as received" and "post tune-up" tests were designated Tests 1A and
IB, respectively. The first recall test, conducted four months after
Test 1, was identified as Test 2 and the second recall test, after an
additional four-month period, was identified as Test 3, etc.
Exhaust Emission Test Procedures
The exhaust emission tests were made in accordance with
applicable Federal Register test procedures and instrumentation specifi-
cation (References 7 and 8). The exhaust emission tests involved both
Constant Volume Sampling mass emission and direct^ tail pipe concentration
measurements. The mass emission, dilute exhaust bag samples were analyzed
with non-dispersive infrared instrumentation for carbon monoxide (CO)
carbon dioxide (C02) and nitric oxide (NO). Flame ionization detection
instrumentation was used for the analysis of total unburned hydrocarbon
(HC), and non-dispersive ultraviolet instrumentation was used for the
analysis of nitrogen dioxide (N02). The direct exhaust emission concentra-
tions were analyzed for carbon monoxide, carbon dioxide, n-hexane equivalent
unburned hydrocarbons, and nitric oxide with non-dispersive infrared
instrumentation. In addition, for both mass and concentration measure-
ments, a chemiluminescence analyzer, i.e., converter and chemiluminescent
NO analyzer system, was incorporated during the third month of the testing
program and was used for the analysis of total oxides of nitrogen. The
NDIR/NDUV NO measurements were deleted during the fourteenth month of
/\
testing.
2-29
-------
Since all of the emission tests involved either the 1972 or 1975
Federal Procedure, the vehicles were soaked for the minimum twelve hour
time period before conducting the cold start tests. The program was
initiated using the 1972 Federal Procedure but a change to the 1975
Federal Procedure was incorporated at the end of the first recall period
(Test 2). Indolene 30 test fuel, which meets the Federal test requirements,
was used whenever possible, during all of the emission test cycles. Vehicles
that were equipped with in-tank fuel pumps and foreign and U.S. vehicles
having fuel lines that were difficult to disconnect were consistently tested
using the vehicle's tank fuel. Approximately 10 to 13% of the vehicles
were tested on tank fuel. The dynamometer inertia weight and road load
horsepower were set according to the 1972 Federal Test Procedure.
Exhaust Emission Test Cycles
Each vehicle test incorporated the measurements of the exhaust
emission using the 1972 or 1975 Federal Test Procedure from a cold start,
the Federal Short Cycle and the short diagnostic cycles. The Federal Test
Procedure and the Federal Short Cycle tests require CVS mass emission measure-
ments. The short diagnostic cycle emissions were measured on a direct con-
centration basis. Only the Federal Test Procedure was run from a cold start,
the other two test cycles were run with a warmed-up engine. The emission
testing was performed using the following sequence:
a) Federal 1972 or 1975 Cold Start
b) Two hot 7-mode cycles (CVS test at 1A only)
c) Federal Short cycle
d) Clayton Keymode cycles while performing parameter inspection
e) Short Diagnostic cycles while measuring exhaust emissions
(which included the 49/45 mph, 33/30 mph aNd idle Clayton
Keymodes).
As indicated above, at the time of the "as-received" test (Test 1A)
an additional emission test was run. Two 7-mode cycles, the 1968-71 Federal
Test Procedure driving cycle, were run and the closed cycle, Constant
Volume Sample mass emissions were determined. In addition, the direct
concentration readings were recorded simultaneously over the 7-mode cycle.
2-30
-------
The 1975 Federal Test Procedure was substituted for the 1972 Federal Test
Procedure at the beginning of the eighth month of testing. The Federal
Short Cycle driving scheduled is presented in Figure 2,8. The short
diagnostic cycles which include the Clayton Keymode Cycles are illustrated
in Figure 2.9. The specifications used for the Keymode cycles are shown
in Table 2.4.
2.2.4 Instrumentation and Test Equipment
Whenever possible, conventional equipment was used to obtain
both the engine parameter and emission measurements. A brief description
of all equipment that was employed is given below.
Engine Parameter Inspection Equipment
The measurements described in Section 2.2.1 and the tune-up
adjustments described in Section 2.2.2 were performed with the following
equipment:
A) Engine Analyzer with Ignition Scope
An Autos can Model 4000 Series Diagnostic System was used.
This engine analyzer included the ignition oscilloscope,
tachometer, dwell meter, vacuum gauge, % speed power
change test meter, timing advance meter, and timing light.
B) Air Filter Tester
The air cleaner element tester was an AC Model 0 air filter
tester. A protractor was installed on the face of the
tester in order to obtain the readings in angular degrees
from 0 to 180.
C) Garage Carbon Monoxide Meter
An Horiba, Type MEXA-200 Motor Vehicle Exhaust Gas Analyzer
was used to set and measure the garage floor idle CO con-
centration. This is a non-dispersive infrared instrument,
with 0 to 5 Volume % range and claimed accuracy of +5%
of full scale.
2-31
-------
FIGURE 2.8
FEDERAL SHORT CYCLE DRIVING SCHEDULE
Driving Mode
Time In Mode,
Seconds
0-16 mph acceleration
16-29 mph acceleration
29 mph cruise
29-37 mph acceleration
37-42 mph acceleration
42-37 mph deceleration
37-20 mph deceleration
20-0 mph deceleration
Idle
6
23
10
18
4.5
2.5
32
7.5
21.5
125 sec.
.7536
21.7 mph
Elapsed Time, seconds
-------
Figure 2.9
SHORT DIAGNOSTIC CYCLES
Selected 7-Mode
Cycles
Speed,
MPH
50 -.
40 _
ro -*"
£ 20 -
10 -
0
Key Mode Cycles
SET DYNO LOAD
HIGH CRUISE
\
LOW CRUISE
IDLE
0
Time, min
Driving Schedules for Vehicles vs. Inertia Weight Settings
1. 4000 Ibs and heavier Inertia
2. 3000 - 4000 Ibs Inertia
3. 2500 Ibs and lighter Inertia
-------
Notes:
Table 2.4
KEY MODE CYCLES
Vehicle Inertia
Weight
4,000 Ibs. & up
3,000-3,500 Ibs.
2,000-2,500 Ibs.
Horsepower
Setting
30 HP @
50 MPH
30 HP @
50 MPH
15 HP @
38 MPH
High Cruise
49 MPH
45 MPH
37 MPH
Driving Cycles
Low Cruise
33 MPH
30 MPH
23 MPH
.
Idle
0
0
0
1. 2,000-2,500 Ib. vehicles with four speed transmissions are
driven in third gear.
2. Automatic transmissions are set in neutral at idle.
2-34
-------
D) PCV Flow Rate Meter
A Model 50 Vol-0-Flow meter, manufactured by National
Instrument Laboratories, Inc. was used to measure the
PCV system flow rate. This is a laminar flow meter
with a differential pressure gauge and has a nominal
flow range of 0 to 10 CFM.
E) Crankcase Pressure Gauge
A "U" Tube water manometer was employed to determine
the crankcase pressure in inches of water.
F) Choke Vacuum Kick Gauges
The choke vacuum kick settings were measured with a
Kent-Moore J-9789-01 plug gauge set.
Emission Test Equipment
The equipment used for the exhaust emission measurements was
constructed with standard commercial components in accordance with the
Federal Test Procedure requirements. The several instrument systems were
combined so that direct concentration samples could be taken simultaneously
with the CVS bag samples. The direct concentration samples were returned
to the dilution duct so that they would not bypass the bag collection.
Dual range instrumentation was also employed whenever possible and all
standard operating modes such as calibration, sampling, and analysis and
purging were semiautomatically controlled from a single, push button
operated control panel for simplicity and speed of operation. The specific
instrument units are described below.
a) Non-Dispersive Infrared System for Measurement of
CO, C02, and HC
A Scott Research Laboratories Model 103-11X instrument
system employing Beckman 315A analyzers was used for the analysis of
Carbon Monoxide, Carbon Dioxide and n-hexane equivalent Hydrocarbons.
Two ranges of HC analyzers were used for direct concentration readings:
1,500 ppm and 10,000 ppm full scale. The C02 analyzer had a full scale
range of 16% and was used for all analyses. A stacked cell CO analyzer
2-35
-------
was used. The direct concentration measurements required the full scale
range of 12% and the Dag sample analysis required the 1% of full scale
range. The low range cell was operated on two ranges, 0.4% and 2.0%
full scale when the 1975 Federal Test Procedure was implemented.
b) FID Analyzer for Measurement of Total Hydrocarbons
A Beckman Instrument Model 108A Flame lonization
Detection hydrocarbon analyzer was used to measure the dilute bag sample
total hydrocarbons. Multirange attenuation was used to set the appro-
priate full scale values of from 30 to 1000 ppm C^.
c) NDIR/NDUV Analyzer for Measurement of NO and N02
A Scott Research Laboratories Model 107-2 NO analyzer
J\
was used to measure both nitric oxide and nitrogen dioxide emissions.
A Beckman Instrument Model 315-A, non-dispersive infrared analyzer was
used to measure NO tailpipe concentration and dilute bag sample emissions.
The full scale ranges were 5000 ppm and 500 ppm, respectively. A Beckman
Instrument 315-A, non-dispersive ultraviolet analyzer was used to measure the
dilute bag sample emissions of NO- with a full scale range of 200 ppm.
d) Mass Sampling System
A Scott Research Laboratories Model 301 Constant Volume
Sampler was used to collect the mass emission samples. Five sample bags
were incorporated in the system, with the ability to analyze one bag
sample while collecting another.
e) Chemiluminescence NO Analyzer
/\
A government furnished Chemiluminescence analyzer (N0?
Converter/Chemiluminescent NO analyzer system) was incorporated for the
analysis of oxides of nitrogen during the third month of the testing
program. This unit was operated in conjunction with the NDIR/NDUV analyzer
until the fourteenth month of the program. This unit was a Thermo Electron
2-36
-------
Corporation Model 10A Analyzer. The Chemiluminescence analyzer and
thermal converter were assembled in a sample train which was operated
independently of the above described emission equipment. The NO exhaust
X
emission samples were always analyzed concurrently with the other instru-
ments. The CVS bag samples were analyzed in the NO mode (converter on)
X
with range attenuation set for 250 ppm full scale. Direct concentration
NO emissions were analyzed with the converter bypassed and the attenuator
set to 5500 ppm full scale. The results of data obtained with the two
different sets of NO instruments were compared and evaluated. A special
/\
detailed evaluation was performed. The results of this evaluation were
published in Volume VI, "A Comparison of Oxides of Nitrogen Measurements
Made with Chemiluminescent and Non-Dispersive Radiation Analyzers," Year
End Report, 1972.
f) Chassis Dynamometer
A Clayton Manufacturing CT-200 Chassis Dynamometer was
used for all emission tests. The dynamometer had a 2000 to 5000 pound
inertia system, adjustable in 500 pound increments. A 50 horsepower
torque bridge was used for the road load horsepower settings.
Instrument Maintenance and Calibration
The emission analysis equipment received preventive maintenance
on a bi-weekly basis. The emission instruments were calibrated monthly.
Most of the instruments were calibrated with at least a five point curve
plus zero. All of the calibration gases, including the span gases used to
set the instruments for each test were +2% tolerance. A system start-up
procedure which included an instrument curve verification and leak check
was followed every day. Engine tachometer calibrations were checked on
a weekly basis, at the beginning of the program and monthly in the second
year.
2-37
-------
2.3 DATA REDUCTION AND ANALYSIS SYSTEM
A detailed description of the system developed for processing and
analyzing all of the data taken in the various experiments conducted in
support of surveillance, inspection, and maintenance for minimization of
vehicle emissions was presented in the report presenting the first year's
activities (Reference 1). Only an overview of the system is therefore
presented. For further information, the reader is referred to Reference 1.
Review of the requirements of the program indicated that in order
to properly process data for publication and to obtain the maximum amount
of information from the experimental program, a system which would provide
the following capabilities would be required:
o reduce test data o develop graphical presentations
o store test data o develop regression equations
o retrieve data o perform analyses of variance
o compute statistics
A system was therefore developed which made maximum use of the capabilities
of Scott Research Laboratories (SRL) and TRW Systems. The partitioning of
the activities is illustrated in Figure 2.10. Since all of the testing was
to be undertaken by SRL, all of the recorded emission data and the reduced
CVS data were combined to form a total data package that was to be sub-
mitted to TRW. TRW, in turn, was charged with the responsibility for
systematically processing the data for further analysis and publication.
The program required that a large amount of data be recorded by
the technicians on the test floor. Because the information was to be
submitted to the computer, special forms were developed which constituted
a compromise between the specific formats required for computer processing
and the descriptive information that is customarily acquired by technicians,
A sample of the types of forms that were used is presented in Section 2.2.
The completed forms, as previously discussed, together with strip
chart records and the computed results of the mass emission tests, were
submitted to TRW as a data package. The inspection data was submitted to
2-38
-------
Figure 2.10
DATA REDUCTION AND ANALYSIS SYSTEM
ro
i
CO
ORTHOGONAL
ANALYSIS
TRW SYSTEMS
DATA
PACKAGE
DATA REDUCTION
OF DIAGNOSTIC
CYCLE DATA
KEY PUNCH
OF
INSPECTION DATA
CRC DATA
REDUCTION
PROGRAM
DATA FILE
SCOTT RESEARCH LABORATORIES
TEST
DATA REDUCTION
OF
MASS EMISSION
INSPECTION
SUMMARIES
DATA PACKAGE
ANALYSIS
OF
VARIANCE
RETRIEVAL
HISTOGRAM
STEPWISE
MULTIPLE
REGRESSION
-------
keypunch and the mass emissions data was merged with the diagnostic cycle
emission test data processed by TRW.
To reduce transportation and procedural errors, a system of data
formatting, editing and review was initiated. The data were initially
keypunched and submitted to the computer in a batch mode. Data listings
were reviewed and corrections were made by editing the data bank using the
remote terminal, TRW Time Share System. Review of the data and editing
of the data was therefore performed on a real time basis, on the computer,
with full control by the data processing technician. All of the process
computer programs are also controlled by teletype terminal control and
large data outputs are developed on a batch mode with the high speed
printer. The program therefore has the capability of submitting large
banks of data to the computer in batch mode with total control of storage,
retrieval and analyses on a demand basis by use of the remote terminal,
Time Share System.
*
A very extensive computer oriented analysis system was developed
to support all of the experiments of the Emissions Program. The total
system is depicted in Figure 2.11. The system was designed to facilitate
efficient processing and analysis of the large amount of data taken in
the Parameter Deterioration Experiment. Description of each of the
computer codes is presented in Reference 1.
2-40
-------
Figure 2.11
COMPUTER PROGRAM INTERFACES
INSPECTION AND
PARAMETER DETERIORATION
DATA CARDS PUNCHED
FROM DATA SHEETS
ro
-P.
PRINTED ANALYSIS
OF VARIANCE
OUTPUT (16 TEST)
/PRINTED ANALYSIS
/OF VARIANCE OUTPUT/
' (24 TEST - 700 SERIES)
/PRINTED ANALYSIS OF/
/VARIANCE OUTPUT
'(24 TEST -800 SERIES)
I PRINTED ANALYSIS
OF VARIANCE
OUTPUT (6 TEST)
ORTHO16
ANALYSIS OF VARIANCE
FOR 16 TEST SERIES
ORTH700
ANALYSIS OF VARIANCE
FOR 700 SERIES CARS
24 TEST SERIES
ORTH800
ANALYSIS OF VARIANCE
FOR 800 SERIES CARS
24 TEST SERIES
CHOKE
ANALYSIS OF VARIANCE
FOR COLD ENGINE
PARAMETER ORTH-
OGONAL TEST SERIES
ORTHOGONAL
TEST SERIES
DATA FILE
(EMISSIONS)
60 RANDOM ACCESS
RECORDS OF 63
WORDS
ONE FILE FOR EACH
ORTHOGONAL TEST
SERIES
MAIN DATA
STORAGE FILE
(EMISSION,
INSPECTION
AND PARAMETER
DETERIORATION)
4076 RANDOM
ACCESS RECORDS OF
127 WORDS
EDITD
EXAMINE AND CHANGE
THE VALUE OF ANY
PARAMETER ON ANY
RECORD OF THE FILE
FIX
SPECIAL PROGRAMS TO
ALTER LARGE AMOUNTS
OF DATA ON THE FILE
IN A SYSTEMATIC MANNER
CLEAND
PERIODIC FILE
CLEANING
PROGRAM
TABD
RETRIEVE SELECTED
DATA AND GENERATE
TABULAR LISTINGS
AND POLONOMIAL
REGRESSIONS WITH PLOTS
HISTO
RETRIEVE SELECTED
DATA AND GENERATE
PARAMETER HISTOGRAMS
WITH PLOTS
/TABULAR LISTINGS/
/AND REGRESSION/
/STATISTICS /
CIATA TYPED
IT TERMINAL
DATA TYPED \
AT TERMINAL/
TESTCAL
CHECKOUT
COMPUTATION OF
SEER'S LAW CALIBRATION
COEFFICIENTS
TIMESHARE
PROGRAM
BATCH
PROGRAM
TIMESHARE
TERMINAL
INPUT
LISTD
DUMP PRINTOUT
OF ENTIRE DATA
FILE CONTENTS
-------
2.4 ANALYSIS OF TEST DATA
2.4.1 Discussion of Results
The objective of the Parameter Deterioration Experiment was to:
1) determine the rate (change per mile of vehicle use) of variation of
emission and continuous engine tune parameters and 2) for those parameters
which were discontinuous, i.e., either operative or non-operative, de-
termine failure rates. Review of the data developed over a sixteen
month test period indicated that statistically significant deterioration
rates for Cold 1972 Federal Emissions were obtained. In contrast, for
many of the key mode emissions, as well as the engine parameters, very
few cases of statistically significant deterioration rates were obtained
from the experimental data.
The tangible results of the experimental program are therefore
the mean value of the deterioration rates (regardless of whether or
not statistically significant regressions were obtained) and upper and
lower limits at the 95 percent confidence level. The results therefore
establish the limits to which parameter deteriorations can be expected to
occur. It is speculated that the many cases of non-significant deteriora-
tion rates resulted because: 1) the sixteen month driving period was not
sufficiently long to result in parameter malfunctions or variations, and
2) sample size was too low at the completion of the program.
This section presents a summary of the results of analyses per-
formed with data taken in the Deterioration Experiment. The best
estimates of the parameter deterioration rates and the parameter failure
rates are summarized, and a comparison of predicted emission deterioration
rates using the empirically derived parameter variation rates (AP/AMiles)
and influence coefficients (AE/AP), with the empirically derived emission
deterioration rates (AE/AMiles) are presented. A comparison of predicted
and measured changes in emissions prior to and following vehicle initiali-
zation is also presented.
2-42
-------
Motor VehicSe Emission Lab
LIBRARY
Fleet Attrition
One of the reasons for the resulting low number of statistically
significant parameter deterioration rates was the small sample size of
vehicles which received no maintenance that remained at the end of the
experiment. An overview of the vehicle population at the end of the
sixteen month test program is illustrated in Table 2.5. The program
was initiated with 150 vehicles in each of three fleets, which represented
vehicles with different degrees of emission control equipment. These were:
Pre-Emission Controlled Vehicles (Fleet 1), Emission Controlled Vehicles
(Fleet 2), and NO Controlled Vehicles (Fleet 3); a very detailed pre-
/\
sentation of this distribution of the vehicle make and model year of the
vehicles which were included in each of. the vehicle fleets is presented
in Reference 1. At the end of the test program there were 24, 27, and 29
vehicles, respectively, in Fleet 1, Fleet 2, and Fleet 3, which had not
received any maintenance throughout the program. There were 43, 38, and
43 vehicles, respectively, which had received minor adjustments. Another
20 to 30 percent of the vehicles in each fleet required major adjustments
and finally, approximately 30 percent of the vehicles were lost due to
attrition.
A very salient conclusion with regard to a program of this type
which makes use of in-field vehicles driven by owners is that the vehicles
cannot be adequately controlled to obtain a data set of vehicles which
will not be maintained over a prolonged (greater than six month) test
program. Another observation is that owners are either (1) unaware that
their vehicles are being maintained, (2) not aware of the nature or extent
of maintenance, or (3) are unwilling to report repairs even though specific
questions regarding maintenance are asked of them. As is indicated in
Table 2.5, of the vehicles which had minor adjustments, respectively 49%,
32%, and 19% of these vehicles of the three fleets, which received main-
tenance were not reported. The number of unreported cases of vehicle
maintenance was obtained either by examining the parts at the completion
of the program (manufacture of component parts were examined for consistency)
2-43
-------
Table 2.5
FLEET DISTRIBUTION OVER 16 MONTH TEST PROGRAM
SUBSET DESCRIPTION
VEHICLES WITH NO
MAINTENANCE
VEHICLES WITH MINOR
ADJUSTMENTS
REPORTED
UNREPORTED
VEHICLES REQUIRING
MAJOR ADJUSTMENTS
VEHICLES LOST DUE
TO ATTRITION
TOTAL
PRE-EMISSION CONTROLLED
FLEET 1
NO. %
24 16
43 29
(22) (51)
(21) (49)
38 25
45 30
150
EMISSION CONTROLLED
FLEET 2
NO. %
27 18
38 25
(26) (68)
(12) (32)
43 29
42 ' 28
150
NOY CONTROLLED
TLEET 3
NO. %
29 19
43 29
35 (81)
8 (19)
46 31
32 21
150
-------
or examination of parts that were painted. The very important unanswered
question is the number of additional cases of maintenance that were un-
reported. Examination of emission and parameter measurements, in many
cases, very definitely suggested additional cases in which vehicles were
maintained.
Deterioration Rates
Several methods were employed to develop estimates of deterioration
rates. A summary of the deterioration rates that are considered to be the
best estimates is presented in Tables 2.6, 2.7, and 2.8, respectively,
for the Pre-Emission, Emission, and NO Controlled Vehicles. As was
A
previously described, the mean and the upper and lower 95 percent confidence
limits are summarized. The results are also graphically depicted in the
Figures 2.12 to 2.50. A cursory review of the data indicates that statisti-
cally significant deterioration rates were consistently obtained for only
;
the Cold 1972 Federal Emissions and Air Cleaner Restrictions. There are
only three additional cases in which statistically significant results were
obtained. These were HC emissions during 49/45 mph Cruise and PCV flow
rates with Emission Controlled Vehicles and NO emissions during 49/45 mph
P\
Cruise with NO Controlled Vehicles. It is concluded that the meaningful
/\
results obtained from this experimental investigation are the establishment
of the upper and lower limits of the rates at which emission and engine
tune parameters can deteriorate. In most cases it is concluded that the
results tend to be conservative, in that for a deterioration period of
approximately sixteen months, the deterioration rates would be expected to
be less in magnitude than is indicated by the limits.
Parameter Failure Rates
In addition to the deterioration rates of parameters, the data taken
in the Deterioration Experiment was analyzed to develop the failure rates
of discrete parameters. Summaries of the parameter failure rates of
interest for vehicles which were not^ maintained at any point during the
deterioration phase of the program are respectively presented in Tables
2.9,2.10, and 2.11, for the three test fleets.
2-45
-------
Table 2.6
SUMMARY OF DETERIORATION RATES (CHANGE PER MILE)
PRE-EMISSION CONTROLLED VEHICLES
FLEET 1
PARAMETER
COLD 1972 FEDERAL HC, gm/mi?
COLD 1972 FEDERAL CO, gm/nii ,
COLD 1972 FEDERAL NO , gin/mi^
/\
49/45 MPH CRUISE HC, ppm/mi
49/45 MPH CRUISE CO, % v/mi
49/45 MPH CRUISE NO , ppm/mi
A
IDLE HC, ppm/mi
IDLE CO, % v/mi
IDLE NO , ppm/mi
J\
TIMING, degrees/mi
IDLE RPM, rpm/mi
AIR CLEANER, degree/mi
PCV FLOW (33/30), cfm/mi
UPPER LIMIT
1+ kS
9.332x10"^
7.569x10';?
-1. 230x1 0~*
3.176x10';?
4. 725x1 Op
6.053x10'^
3.174x10"?
3.202x10':;
-1. 561x1 0~J
1.131x10"!
10.715x10"^
7.729x10"^.
3.513xlO"D
MEAN
I
5.740x10"^*
4.673x10"^*
-2.327x10"^*
-0.978x10"^
-1.610x10";
2.306x10"^
-0. 382x1 0"?
1.334x10"^
-3.140xlO"J
-1.481x10"!
3.766x10"^
5.307x10"^*
-1.424x10"°
LOWER LIMIT
X - kS
2.1 48x1 0~!J
1.477x10";:
-3.354x10"^
-5.132x10"^
-7.945x10"^
-1.441x10"^
-3.938x10"^
-0.534x10'.:
-7.841xlO"J
-4.093x10"^
-3.183x10"^
2.885x10"^
-6.361x10"°
r\s
i
cr>
*Statistically significant at the 90% confidence level.
-------
Table 2.7
SUMMARY OF DETERIORATION RATES (CHANGE PER MILE)
EMISSION CONTROLLED VEHICLES
FLEET 2
PARAMETER
COLD 1972 FEDERAL HC, gm/mi.
COLD 1972 FEDERAL CO, gm/mi ,
COLD 1972 FEDERAL NO , gm/mr
/\
49/45 MPH CRUISE HC, ppm/mi
49/45 MPH CRUISE CO, % v/mi
49/45 MPH CRUISE NO , ppm/mi
X
IDLE HC, ppm/mi
IDLE CO, % v/mi
IDLE NO. ppm/mi
A
TIMING, degrees/mi
IDLE RPM, rpm/mi
AIR CLEANER, degrees/mi
PCV FLOW (33/30), cfm/mi
UPPER LIMIT
X + kS
2.570x10"^
4.054xlO~;J
-2. 636x1 0"^
0.334x10';?
2.568xlO~X
2.137x10"^
0. 629x1 0~^
1.062x10,
2.267xlO"J
1. 240x1 0~J
4.011x10,
3.426x10';:
-1. 946x1 0"D
MEAN
X
1.333x10"^*
2.351x10"^*
-3.744x10"^*
-2.444x10"^*
-0.278x10",
-1.203x10"^
0.046x10"^
0.274x10 ,
-1. 088x1 0"J
-0.409x10"^
0.053x10,
2.555x10"^*
-4.758x10 *
LOWER LIMIT
X - kS
0.096x10"^
0.648xlO~;J
-4. 852x1 O'4
-5.222x10"^
-3.124x10",
-4.543x10'^
-0.537x10"^
-0.514x10 ,
-4.443xlO"J
-2.058x10"^
-3.905x10",
1.684x10";:
-7.570xlO"b
ro
i
*Statistically significant at the 90% confidence level.
-------
Table 2.8
SUMMARY OF DETERIORATION RATES (CHANGE PER MILE)
N0v CONTROLLED VEHICLES
X
FLEET 3
PARAMETER
COLD 1972 FEDERAL HC, gm/mi.
COLD 1972 FEDERAL CO, gm/mr 9
COLD 1972 FEDERAL NO , gm/mr
A
49/45 MPH CRUISE HC, ppm/mi
49/45 MPH CRUISE CO, % v/mi
49/45 MPH CRUISE NO , ppm/mi
A
IDLE HC, ppm/mi
IDLE CO, % v/mi
IDLE NO , ppm/mi
A
TIMING, degrees/mi
IDLE RPM, rpm/mi
AIR CLEANER, degrees/mi
PCV FLOW (33/30), cfm/mi
UPPER LIMIT
X + kS
1.191xlO"5
1.474xlO"c
0.653x10"°
1.598x10"?
0.320x10"°
0.271x10"^
_2
0.493x10'!!
2.762xlO"J
1.117x10"!
0.856x10'::
2.711x10";:
2.1 99x1 0
MEAN
X
0.703x10"!!*
O.SlOxlO'r*
-3.381x10"°*
-0.228x10";?
-0.716x10";
-3.002x10"^*
-0.013x10"^
-0.115x10"^
0. 200x1 0"15
0.345x10"^
-1.510x10"^
2.093x10"^*
0. 970x1 0"b
LOWER LIMIT
X - kS
0.215x10"^
0.147xlO'r
-7.415x10"°
-2.054xlO"|?
-1.752x10",
-6.275x10"^
-0.286x10"^
-0.723x10"!:
-2.362xlO"J
-0.427x10"^
-3.876x10"^
1. 475x1 0"c
-0.259x10"
ro
i
00
*Statistically significant at the 90% confidence level
-------
Table 2.9
FAILURE RATES OF ENGINE PARAMETERS
PRE-EMISSION CONTROLLED VEHICLES
FLEET 1
PARAMETER
Heat Riser,
% failed
Vacuum Diaphragm,
% failed
No Control Dev. ,
% failed
Air Pjmp,
% failed
Misfire, 49/45 Cruise
%
Misfire, 33/30 Cruise
%
Misfire, Idle
°/
h
Mileage, Miles
failed
N
failed
N
failed
N
failed
N
I
n
N
X"
n
N
I
n
N
1A
51.8
85
21.4
28
_
-
-
-
0.505
5
150
0.560
5
150
0.732
5
150
0
IB
7.0
86
0
28
_
-
-
-
0
0
150
0
0
150
0
0
150
0
2
33.3
66
0
19
_
-
-
-
0
0
100
0
0
100
o
0
100
3500
3
38.5
39
0
12
_
-
-
-
0
0
67
0
0
67
0
0
67
5900
4
40.9
22
0
5
_
-
-
0
0
43
0
0
43
0
0
43
8600
5A
40.0
15
20.0
5
-
-
1.472
2
24
1.312
2
24
1.292
2
24
10200
5B
40.0
15
20.0
5
-
-
0
0
24
0
0
24
0
0
24
10200
n = Number of cases in which misfire was detected
N = Samples in experimental test set
-------
Table 2.10
FAILURE RATES OF ENGINE PARAMETERS
EMISSION CONTROLLED VEHICLES
FLEET 2
PARAMETER
Heat Riser,
% failed
Vacuum Diaphragm,
% failed
NO Control Dev.,
x% failed
Air Pump,
% failed
Misfire, 49/45 Cruise
%
Misfire, 33/30 Cruise
%
Misfire, Idle
%
Mileage, Miles
failed
N
failed
N
failed
N
failed
N
X"
n
N
X"
n
N
jf
n
N
1A
43.8
73
20.6
92
0
12
7.7
39
.0750
2
148
.0818
2
148
.208
2
148
0
IB
4.1
73
0
92
0
14
2.6
39
0
0
148
0
0
148
0
0
148
0
2
27.1
59
1.4
70
0
13
3.6
28
0
0
105
0
0
105
0
0
105
4700
3
21.9
32
8.7
46
0
12
0
15
0
' 0
67
0
0
67
.188
1
67
7500
4
57.9
19
7.4
27
20.0
10
0
5
0
0
40
0
0
40
0
0
40
10900
5A
45.4
11
5.3
19
16.7
6
0
3
0
0
27
0
0
27
0
0
27
13400
5B
0
12
0
19
16.7
6
0
3
0
0
27
0
0
27
0
0
.27
13400
t\J
n = Number of cases in which misfire was detected
N = Samples in experimental test set
-------
Table 2.11
FAILURE RATES OF ENGINE PARAMETERS
NOX CONTROLLED VEHICLES
FLEET 3
PARAMETER
Heat Riser,
% failed
Vacuum Diaphragm»
% failed
No Control Dev.,
x% failed
Air Pump,
% failed
Misfire, 49/45 Cruise
%
Misfire, 33/30 Cruise
%
Misfire, Idle
V
h
Mileage, Miles
failed
N
failed
failed
N
failed
I
n
N
X"
n
N
X"
n
N
1A
3.6
28
0
122
4.8
84
-
0
0
150
0
0
150
0
0
150
0
IB
0
28
0
122
2.4
85
_
-
0
0
150
0
0
150
0
0
150
0
2
0
20
1.1
91
8.1
62
_
-
0
0
113
0
0
113
0
0
113
5600
3
0
11
1.5
65
6.8
44
_
-
0
0
75
0
0
75
0
0
75
9300
4
0
5
4.8
42
6.9
29
_
-
0
0
48
0
0
48
0
0
48
12900
5A
0
0
3.8
26
11.1
18
_
-
0
0
29
0
0
29
0
0
29
18000
5B
0
0
4.0
25
11.8
17
_
-
0
0
28
0
0
28
0
0
28
18000
ro
i
in
n = Number of cases in which misfire was detected
N = Samples in experimental test set
-------
As has been consistently, observed throughout the program, the heat
riser has a high frequency of failure. Vacuum diaphragms have comparatively
low rates of failure. In contrast, NO controlled devices have a sur-
^
prisingly large number of failures. The sample size of vehicles equipped
with an air pump was comparatively small. During the initialization, the
one vehicle with a failed air pump was not corrected. This vehicle is the
only one in which a measured air pump failure was observed.
Engine misfire is a very major cause for high HC emissions and it
was an objective of this experiment to develop the frequency of occurrence.
It was, however, concluded early in the program that no reliable method
for determining misfire existed. A method which utilized HC emission
measurements was therefore adopted to measure misfire (a discussion of the
method is presented in Reference 1). This approach was selected even though
confounding with other malfunction, i.e., valve failure, flooding carburetor,
etc., could occur and errors of commission could result. There were very
few cases of misfire detected with those vehicles which were not maintained
during the course of the Deterioration Experiment. In fact, during the
entire experimental program only two vehicles were observed to have misfire.
These two vehicles were Pre-Emission Controlled Vehicles and the misfire
rate occurred after approximately 10,000 miles of use. No measureable
amount of misfire was observed in the vehicles of the two other fleets.
The failure fractions presented in Tables 2.9, 2.10,and 2.11 were
developed from inspections of vehicles which did not receive maintenance
during the deterioration period. The failure fractions could therefore be
misleadingly low because only those vehicles which were inherently more
stable were left in the fleet. Secondly, because of the high number of
vehicles which received maintenance, the fleet sample sizes were greatly
reduced. In order to develop more representative estimates of failure
fractions, the failure fractions of discrete parameters of vehicles which
received minor maintenance were developed. These results are presented
in Tables 2.J2, 2.13, and 2.14. Particular emphasis should be placed on
the number of misfirings that occurred with the vehicles that had received
minor maintenance. It is speculated that: 1) many cases of misfire were
2-52
-------
Table 2.12
FAILURE RATES OF ENGINE PARAMETERS
VEHICLES WITH MINOR ADJUSTMENTS
PRE-EMISSION CONTROLLED VEHICLES
(FLEET 1)
PARAMETER
HEAT RISER,
% FAILED
VACUUM DIAPHRAGM,
% FAILED
NO CONTROL DEV.,
% XFAILED
AIR PUMP,
% FAILED
TEST
failed
N
failed
N
failed
N
fai 1 ed
N
MISFIRE, 49/45 Cruise jf
%
n
N
MISFIRE, 33/30 Cruise J
%
MISFIRE, Idle
%
MILEAGE, miles
n
N
X"
n
N
1A
51.8
85
21.4
28
_
-
_
-
.508
5
150
.564
5
150
.737
5
150
0
IB
7.0
86
0
28
_
-
_
-
0
0
150
0
0
150
0
0
150
0
2
38.4
73
0
22
_
-
_.
-
.638
4
126
.739 .
4
126
.574
4
126
3600.
3
49.2
65
0
19
_
-
_
-
.307*
2
no
.350*
3
no
.354*
3
no
6200
4
48.3
60
0
19
-
_
-
0.133*
1
105
0
0
105
0.217*
2
104
9100
5A
52.7
38
16.7
12
-
_
-
0.914*
3
81
0.899*
3
81
0.545*
3
80
11200
5B
5.4
55
10.5
19
-
_
-
0
0
97
0
0
97
0
0
97
11200
ro
en
CO
n = Number of cases in which misfire was detected
N = Samples in experimental test set
*Distributive values
-------
Table 2.13
FAILURE RATES OF ENGINE PARAMETERS
VEHICLES WITH MINOR ADJUSTMENTS
EMISSION CONTROLLED VEHICLES
(FLEET 2)
no
i
tn
p*^
t
PARAMETER
HEAT RISER,
% FAILED
VACUUM DIAPHRAGM,
% FAILED
NO CONTROL DEV.,
% TAILED
AIR PUMP,
% FAILED
MISFIRE, 49/45 Cruise
%
MISFIRE, 33/30 Cruise
%
MISFIRE, Idle
MILEAGE, miles
TEST [ 1A
failed
N
failed
N
failed
N
failed
N
I
n
N
I
n
N
I
n
N
43.8
73
20.6
92
0
12
7.7
39
.0750
2
148
.0818
2
148
.208
2
148
0
IB
4.1
73
0
92
0
13
2.6
39
0
0
148
0
0
148
0
0
148
0
2
25.0
72
2.4
85
0
15
2.8
35
0
0
130
0
0
130
0
0
130
5000
3
35.9
64
5.6
72
0
15
3.1
32
0
0
112
0
0
112
.112
1
112
8100
4
48.1
52
4.8
63
13.3
15
0
27
.300
2
98
.304
2
98
.247
2
98
11700
5A
48.7
35
9.3
43
11.1
9
0
25
0
0
85
0
0
85
0
0
85
14600
5B
0
53
6.6
61
8.3
12
0
25
0
0
95
0
0
95
0
0
95
14600
n = Number of cases in which misfire was detected
N = Samples in experimental test set
-------
Table 2.14
FAILURE RATES OF ENGINE PARAMETERS
VEHICLES WITH MINOR ADJUSTMENTS
NO CONTROLLED VEHICLES
X (FLEET 3)
PARAMETER
HEAT RISER,
% FAILED
VACUUM DIAPHRAGM,
% FAILED
NO CONTROL DEV.,
% XFAILED
AIR PUMP,
% FAILED
MISFIRE, 49/45 Crui
%
MISFIRE, 33/30 Crui
%
MISFIRE, Idle
%
MILEAGE, miles
TEST
failed
N
failed
N
failed
N
f ai 1 ed
N
se x"
n
N
se I
n
N
X"
n
1A
3.6
28
0
122
4.8
84
_
-
0
0
150
0
0
150
.0993
1
0
IB
0
28
0
122
2.4
85
_
-
0
0
150
0
0
150
0
0
0
2
0
25
1.8
113
6.8
74
_
.-
.0544
1
136
.0434
1 '
136
.0301
1
5700
3
4.8
21
4.1
98
4.4
68
_
-
0
0
114
0
0
114
0
0
9600
4
0
21
2.1
96
10.4
67
_
-
0
0
111
0
0
111
0
0
13900
5A
0
10
1.6
61
13.0
46
_
-
.165*
1
86
.173*
1
86
.153*
1
17400
5B
0
19
2.2
90
14.3
63
_
-
0
0
106
0
0
106
0
0
17400
en
en
n = Number of cases in which misfire was detected
N = Samples in experimental test set
*Distributive values.
-------
not detected due to the lack of sensitivity of misfire measurements, and
2) those vehicles which resulted in significant misfire may have been
repaired in the field with no record of maintenance. It is therefore
concluded that the failure rates measured with vehicles which received
minor maintenance are better representations of the expected frequencies
of occurrence of misfire of vehicles in the field. It should be noted
that the misfire rates, expressed in percentage of the total vehicle
fleet, are given as a distributive value. A cumulative frequency of mis-
fire would therefore be a summation of the values presented in the Tables.
Effort to quantify the deterioration characteristics of the continuous
parameters for vehicles which received minor adjustments was not pursued
because it was speculated that in addition to the minor adjustment per-
formed, other parameters would be maintained with no record of the degree
of maintenance. This consideration further supports the observation that
the frequency of misfire detected in this program is low.
Comparison of Predicted and Measured Emission Deterioration Rates
Two important phases of the Emissions Test Program were the develop-
ment of experimental data to: 1) determine the deterioration rate of
engine tune components and settings and exhaust emission parameters
(Cold 1972 Federal emissions or concentration measurements for the 49/45
mph loaded mode or idle mode) and, 2) the influence coefficients which
relate changes in parameters to changes in emissions. Since both the
emissions as well as the engine tune parameters were measured during the
Parameter Deterioration Experiment, a direct comparison of how well varia-
tions in emissions can be accounted for by changes in the selected parameters
monitored during the experiment could be made. The primary parameters
which were considered to influence emissions were basic timing, idle rpm,
air cleaner restriction, PCV flow restriction, and idle CO. Cold engine
parameters were not considered and misfire was treated separately.
Certainly there are other parameters that would influence emissions;
however, these other parameters were considered to be either too costly
to repair (carburetor, valves, rings, etc.) or on an individual basis
2-56
-------
were concluded to effect small changes in emissions. In addition, misfire
was not included in these analyses because within the scope of this ex-
perimental program the number of cases of misfire with unmaintained vehicles
was so small that the results were not considered statistically meaningful.
The comparison was made to determine the degree to which the deteriora-
tion rates on engine parameters would account for the deterioration rate of
cold emissions as measured by the 1972 Federal Procedure. The results are
summarized in Table 2.15. The measured and predicted values using the mean
and the upper 95 percent confidence limits of the deterioration rates and
influence coefficients (the change in emission with change in parameter
as developed in the analysis of variance of data taken in the Orthogonal
Test Program, Reference 1) are given. Further discussion, which presents
the development of the confidence limits of both the deterioration rates
and the influence coefficients, is presented in Section 2.4.3.
The data presented in Table 2.15 clearly indicate that as is the
case for comparisons of predicted and measured changes that occurred during
the initialization phase (initial and final tune-up) the best agreement
between predicted and measured values is obtained with CO emissions. Clearly,
an intercept of the predicted value with the measured value within the
range of variation of the experimentally measured rates resulted; e.g., with
the Emission Controlled Vehicles the maximum predicted value or the maximum
-3 2
predicted deterioration rate is 1.762x10 gm/mi , and the measured values
resulted in a lower limit of 0.648x10 gm/mi, and an upper limit of
4.054xlO"3 gm/mi. The results with the Pre-Emission Controlled Vehicles, as
well as with the NO Controlled Vehicles, are comparable. Comparison of
Ai
measured and predicted values for Cold 1972 Federal HC Emissions resulted
in an intercept of limits with Pre-Emission Controlled and Emission Con-
trolled Vehicles. In contrast, with the NO Controlled Vehicles, the
4 2
predicted upper limit of 0.190x10 gm/mi was less than the lower limit
of the measured HC emission deterioration rate. For these 1971 NO Con-
/\
trolled Vehicles both the emission rates and the engine tune parameter
2-57
-------
Table 2.15
COMPARISON OF MEASURED AND PREDICTED EMISSION VARIATION RATES
ro
i
un
oo
STATISTIC .
MEAN
UPPER LIMIT
LOWER LIMIT
MEAN
UPPER LIMIT
LOWER LIMIT
MEAN
UPPER LIMIT
LOWER LIMIT
PRE-EMISSION
CONTROLLED
MEASURED PREDICTED
EMISSION
CONTROLLED
MEASURED PREDICTED
NOX
CONTROLLED
MEASURED PREDICTED
COLD 1972 FEDERAL HC EMISSIONS
2 2
gm/mi gm/mi
5.740x10~4 0.676xlO"4
9. 332x1 0"4 2. 533x1 O"4
2.148xlO~4
2 2
gm/mi gm/mi
1.333xlO~4 0.048xlO~4
2.570xlO~4 0.498xlO"4
0. 096x1 O"4
2 2
gm/mi gm/mi
0.703xlO~4 0.027xlO"4
1.191xlO"4 0.190xlO"4
0.215xlO"4
COLD 1972 FEDERAL CO EMISSIONS
2 2
gm/mi gm/mi
4.673xlO"3 1.408xlO"3
7.869xlO"3 3.787xlO"3
1.477xlO"3
2 2
gm/mi gm/mi
2.351xlO~3 0.543xlO~3
4.054xlO"3 1. 762xlO~ 3
0.648xlO"3
2 2
gm/mi gm/mi
O.SlOxlO"3 -O.OSlxlO"3
1.474xlO"3 0.734xlO'3
0.147xlO"3
COLD 1972 FEDERAL N0x EMISSIONS
2 2
gm/mi gm/mi
-2.327xlO"4 -0.079xlO"4
-1.230X10"4 -1.081xlO'4
-3.354xlO'4
2 2
gm/mi gm/mi
-3.744xlO"4 -0.126xlO"4
-2.636xlO~4 -0.424xlO~4
-4. 852x1 O"4
2 2
gm/mi gm/mi
-0.3381xlO"4 -0.139xlO"4
0.0653xlO"4 -0.628xlO"4
-0.741 5x1 0"4
-------
deterioration rates were extremely small, and therefore resulted in the
lack of agreement between measured and predicted results. Furthermore,
the HC deterioration rates with the NO Controlled Vehicles were more
A
repeatable than with the other fleets, and therefore resulted in a much
tighter band of variation of measured emission deterioration rates.
The inclusion of Misfire should greatly improve the correlation.
The predicted values of NO emission deterioration rates on the
A
basis of five engine tune parameters clearly did not account for the
experimentally measured emission deterioration rate. In all cases the
upper 95 percent confidence limit of the predicted emission rate was
smaller in magnitude than the upper limit of measured deterioration
rates. (It should be noted that NO deterioration rates are negative).
A
Additional experimental data are required to determine the reasons for the
discrepancy between measured and predicted NO emission deterioration rates.
A
The method selected to develop the deterioration rate weights high values
heavily, and therefore may have resulted in an overestimation of the rate.
Although the approach gives reasonable results with HC and CO emissions,
the NO emissions appear large. The use of alternate approaches for
A
development of deterioration rates of different parameters could not be
justified. The same approach was therefore applied to all parameters.
Clearly, further review of the existing data or development of additional
test data is required.
As a part of the effort to verify the experimentally developed
deterioration rates and influence coefficients, a measure of the degree to
which the combined effects of the magnitude of the deterioration rate and
the magnitude of the influence coefficient contributes to the total emission,
was developed. The fraction of total emissions was developed using the
predicted values, for both the cases in which predictions were made using
1) the mean values and 2) the upper 95 percent confidence limits (values
selected to obtain the largest magnitude of emission change). The fraction
2-59
-------
of total emissions for the two cases studied are presented in Tables
2.16 and 2.17. Although there are some cases of differences between
the two cases examined, there are some fairly consistent indications.
In all cases, as would be expected, idle CO greatly influences the Cold
1972 Federal CO emission deterioration rate. Air cleaner restriction
is a strong influence on the CO deterioration rate. Both idle CO and
air cleaner restriction greatly affect the deterioration rate of HC
emissions for Pre-Emission Controlled and NO Controlled Vehicles. For
/\
Cold 1972 NO emission deterioration rates, timing appears to be the
A
strongest influence, with air cleaner also being a very strong influence.
These data appear consistent with opinions generally accepted in the
industry and therefore tend to further establish credibility of the data
developed in this program. One inconsistency of a prior concept which
resulted from the detailed experimentation conducted during the deteriora-
tion experiment is the low deterioration of PCV flow rate, together with
the very minor influence of PCV flow rate variations on emissions. The
other is that, surprisingly, idle rpm appears to deteriorate at a lower
rate than anticipated prior to the experimental program. It should be
emphasized, however, that as summarized in Tables 2.6, 2.7, and 2.8, timing,
idle rpm, PCV, and flowrate resulted in low deterioration rates which were
not statistically significant. It is speculated that these parameters
operate in a non-linear manner and the length of the program was not
sufficient to result in real physical changes in these parameters.
Comparison of Predicted and Measured Changes in Emissions due to
Engine Tune
The influence coefficients which will be used in the Economic
Effectiveness Model to predict a change in emissions, on the basis of a
known or predicted change in parameter, were developed experimentally
(Orthogonal Experiments). The results of the Analysis of Variance (AOV)
conducted with the test data are presented in Reference 1. Coefficients
were developed for General Motors, Ford, Chrysler, American Motors, and
other categories of vehicle make. All the data clearly indicated that
2-60
-------
Table 2.16
FRACTION OF PREDICTED EMISSION DETERIORATION RATE (SLOPE)
BASIS MEAN RATE
PARAMETER
TIMING
IDLE RPM
AIR CLEANER
PCV FLOW
IDLE CO
PRE-EMISSION CONTROLLED
(FLEET 1)
HC CO NO
X
-0.1472 0.0888 2.5988
-0.1632 0.0773 -0.3463
0.4449 0.4226 2.0369
-0.0048 0.0247 0.3546
0.8703 0.3866 -3.6440
EMISSION CONTROLLED
(FLEET 2)
HC CO NO
X
-0.4280 0.0258 0.3088
-0.0612 0.0020 -0.0011
0.4832 0.3595 0.4294
0.8413 0.2506 0.3265
0.1647 0.3621 -0.0636
NOX CONTROLLED
(FLEET 3)
HC CO N0v
J\
0.9655 0.4355 -5.5375
0.2372 0.4846 1.7971
0.6932 -1.7759 6.4033
-0.5560 0.8936 -1.2440
-0.3399 0.9622 -0.4189
ro
en
-------
Table 2.17
FRACTION OF PREDICTED EMISSION DETERIORATION RATE
BASIS MAXIMUM RATE (UPPER LIMIT)
PARAMETER
TIMING
IDLE RPM
AIR CLEANER
PCV FLOW
IDLE CO
PRE-EMISSION CONTROLLED
(FLEET 1)
HC CO N0x
0.0371 0.1180 0.6314
0.0509 0.1132 0.0323
0.2385 0.2933 0.2890
0.0043 0.0579 0.0473
0.6692 0.4176 0
EMISSION CONTROLLED
(FLEET 2)
HC CO N0x
0.1526 0.0536 0.5386
0.5462 0.0618 0.0320
0.0743 0.1737 0.2217
0.1486 0.1419 0.2077
0.0783 0.5690 0
NO CONTROLLED
(FLEET 3)
HC CO N0x
0.4841 0.0669 0.2842
0.0604 0.0395 0.1919
0.1631 0.3214 0.4121
0.0268 0.0318 0.0168
0.2656 0.5404 0.0949
ro
i
CTl
r\)
-------
the coefficients were both repeatable for like power trains, and in most
parameters, similar for all power trains. It was considered necessary to
apply these coefficients to the measured changes in parameter to determine
how well the change in emissions could be predicted. This section presents
the methods used to compute the predictions and to determine whether or
not the differences were statistically different.
Influence Coefficients
The influence coefficients developed from the data obtained for
each vehicle tested in the Orthogonal Experiments were used to develop a
weighted value for each class of major vehicle manufacturer. Also, the
estimates for each vehicle make were weighted by using the National Dis-
tribution of Vehicles to develop a weighted average coefficient representing
the in-field population of vehicles. Results of the weighted values for
the parameters timing, idle rpm, air cleaner restrictions, PCV flow restric-
tion (measured in the 33/30 keymode) and idle CO were developed. These
values were reported in Reference 1.
It was noted that in the testing of the Emission Controlled
Vehicles (Fleet 2, 1966-1970 Vehicles) that there were very few cases of
two-factor interactions. In contrast, in the testing of the Pre-Emission
Controlled Vehicles (Fleet 1, Pre-1966 Vehicles), and the NO Controlled
A
Vehicles (Fleet 3, 1971 Vehicles), the analyses presented many cases of
two-factor interactions which, although small in comparison to the main
effects, were considered significant. For use in predictions, it was
therefore necessary to make use of the two-factor interactions.
Although two-factor interactions can be included in a computer
model, for use in quick or hand calculations in predicting emission changes,
it was decided that main effects would be adjusted for two-factor inter-
actions. For those parameters which varied both in a positive and negative
direction from mean values, it was decided that the two-factor interaction
could be ignored. This decision is considered valid because predictions
are made on the basis of an average set of data and the effect of a two-
factor interaction would be negligible if the parameter varies both in a
2-63
-------
positive and negative direction. In contrast, there are some parameters
which, in general, did not vary in a positive and negative direction.
For example, the Orthogonal Experiment was conducted measuring the influence
of a change in Air Cleaner from a fully restricted Air Cleaner to a new,
unused Air Cleaner (approximately 180° change using the Air Cleaner Tester).
Examination of the average value of Air Cleaner Restriction in the test
data, in contrast, suggested that the mean value of Air Cleaner Restriction
was approximately 20 degrees. Since there was an apparent bias in Air
Cleaner Restriction from the median value of the change imposed in the
Orthogonal Experiment, the effect of a two-factor interaction which involved
Air Cleaner and a second parameter was used to adjust the main effect of
the second parameter. This adjustment was used to nullify the bias that
existed in the Air Cleaner Restriction in comparison to the restriction
used in the Orthogonal Experiment. A similar correction was used for
two-factor interactions which involved PCV flow rates and NO control de-
/\
vices. The investigation of the effects in variations in these two parameters
in the Orthogonal Experiment consisted of examining a fully operative or
non-operative system (failed or operative). In a deterioration experiment
very few cases of PCV valve failure or NO control system failure resulted.
/\
The effect of two-factor interactions was therefore eliminated by imposing
the effect of fully operative PCV valve and NO control devices.
A
The coefficients that resulted following the adjustments de-
scribed in the previous paragraph are summarized in Table 2.18. It should
be emphasized that no two-factor interactions are included in the Table.
The influence coefficients for changes in the parameters indicated are
representative of the change that would be expected to result for a given
change in the main parameter, given that the air cleaner restriction is
approximately 20 degrees and the PCV valve and NO control valve are
^
operative. Since these coefficients are used primarily to predict the
change in emissions that would result in an average change in a given
parameter for a fleet of vehicles, they are considered the best estimates
and most easily usable influence coefficients. To obtain a more exact
2-64
-------
i
CD
en
Table 2.18
EMISSION RESPONSE COEFFICIENTS HOT 1972 FEDERAL CVS PROCEDURE
TIMING,
IDLE CO (-TOO), g/mi/%v
MANUFACTURER
GM
FORD
CHRYSLER
AMC
OTHER
COMPOSITE
GM
FORD
CHRYSLER
AMC
OTHER
COMPOSITE
HC EMISSIONS
FLEET 1 FLEET 2 FLEET 3
.06788 .0742 .1087
.06173 .0604 .06898
.07457 .0180 .04618
.01581
0 .04000
.06710 .0498 .07558
CO EMISSIONS
FLEET 1 FLEET 2 FLEET 3
-1.234 -.5128 -.5659
-.3661 -.1575 -1.513
-.2972 -.2011 -1.880
-1.388
-.3299 -.1327
-.8444 -.3423 -1.022
IDLE RPM, gram/mile/rpm
-.005205 -.00584 -.001747
0 -.00711 .001227
0 -.00354 0
-.0002768
-.00346 -.0003846
-.002930 -.00550 -.0004243
.01191 .00541 .01216
.05264 .03256 .07750
.04704 .03618 0
0
.02979 .001572
.02888 .02120 .02598
NOX EMISSIONS
FLEET 1 FLEET 2 FLEET 3
.09036 .1041 .1631
.2257 .1126 .09282
.1579 .1030 .1403
.1080
.0480 .09501
.1394 .09524 .1299
.001298 .00062 .003639
0 0 -.001820
000
-.00002991
0 0
.0007306 .000255 .0009632
PCV, granv/mile/cfm
GM
FORD
CHRYSLER
AMC
OTHER
COMPOSITE
0 -.08332 -.1126
.07922 -.1114 -.2047
0 -.1785 -.1439
-.1025
0 -.2116
.02283 -.08415 -.1548
GM
FORD
CHRYSLER
AMC
OTHER
COMPOSITE
.007773 .00297 .0006598
.003819 0 .001076
.001293 0 .001733
.001152
-.00161 .00001800
.005668 .000900 .0008945
-3.355 -5.476 -6.561
-1.167 -.7393 -8.164
-1.454 -3.979 -9.591
-1.182
0 -7.508
-2.441 -2.861 -7.458
AIR CLEANER, gram/mile/degree
.1482 .1091 .05896
.07568 .0525 .07879
.04606 .0501 .1105
.05317
.0570 .02390
.1121 .07642 .06869
.1029 .06041 .01232'
0 .1035 .2348
0 .3217 .5361
.1692
0 -.5069
.05792 .08654 .1038
-.003286 -.00258 -.001500
-.003069 -.00157 -.003482
-.002113 -.00298 -.004896
-.002694
-.00157 0
-.003049 -.00212 -.002476
HC EMISSIONS
FLEET 1 FLEET 2 FLEET 3
.2018
-.008467 -
0 - -
-
_
.1112
CO EMISSIONS
FLEET 1 FLEET 2 FLEET 3
7.877
3.355
2.381
_
-
5.756
NOX EMISSIONS
FLEET 1 FLEET 2 FLEET 3
-.4582
-1 . 200
.01786
-
_
-.6011
IDLE CO (0 TO +), grom/mile/% y
.4171 .0696
.4122 0
.5880 0
.
0
.4411 .02860
3.295 9.66
4.365 6.01
6.506 9.31
-
-.03805 -.0489
.8272 .1702
00-
-
Z.71 \ 0
4.081 7.177 - | .2170 .02926
IDLE CO (- TO +), gnmv/mile/% v
. 09530
.04900
.1450
-.03225
.03600
.07983
9.927
6.744
"*.351
3.380
2.337
6.773
-.1564
0
.1912
0
0
-.02948
NOX CONTROL DEVICE, gram/mile/unit
.8491
0
0
.07562
.6560
.4235
-6.876
0
-3.423
-12.62
3.104
-3.374
.3311
0
.8375
1.209
2.081
.5753
-------
functional relationship of emissions with parameter changes, the full
change of main effects and two-factor interactions can be obtained from
the Tables presented in Reference 1.
An initial attempt to develop an indication of how well
predictions of Cold 1972 Federal Emissions could be made on the basis of
measured changes in parameters was made using the test data developed in
the final vehicle initialization process performed at the end of the
Parameter Deterioration Experiment (Tests 5A and 5B). Calculations were
made on the basis of average changes that were calculated for each fleet
of vehicles. The weighted influence coefficients representative of all
in-field vehicles were therefore used. The method used to apply the.
coefficients and to indicate the general agreement of the results is
illustrated in the calculation sheets respectively representing the
results with each of the three test fleets (Tables 2.19, 2.20, and
2.21). It should be emphasized that the estimates presented in these
tables are the results developed following elimination of emission values
greater than two times the estimate of standard deviation obtained in an
initial retrieval of all of the data. This smoothing of the data was con-
sidered necessary because of the existence of the resulting large changes
in some of the vehicles. For example, these were in the order of magnitude
of 20, 120, and 3 gm/mi, respectively, for HC, CO, and NO emissions as
s\
measured using the Cold 1972 Federal Procedure. Changes of this magnitude
were considered beyond expected changes that would result from minor
variations in parameters.
A cursory review of the results obtained by applying the
influence coefficients using average changes per fleet indicates the best
agreement between predictions and measured changes in CO. In general,
the coefficients resulted in larger predictions than measured. In contrast,
the predictions for HC changes were markedly smaller than measured. For
predictions of NO changes, there appeared to be no correlation between
J\
measured and predicted results. A cursory conclusion was that the pre-
dictions of CO would be expected within the experimental uncertainty of
2-66
-------
Table 2.19
CALCULATION OF AVERAGE CHANGE IN EMISSIONS
PRE-EMISSION CONTROLLED VEHICLES
(FLEET 1)
PARAMETER
TIMING, degrees
IDLE RPM, rpm
PCV, cfm
A/C, degree
ICO, % v
PREDICTED VALUES X"
MEASURED VALUES I
s
d.f.
t
CHANGE
-.2037
61.71
-.1306
18.24
0.5700
COLD 1972 FEDERAL
AE/AP HC AE
gm/mi
.06710 -.013
-.002930 0.181
.02283 -0.003
.005668 0.103
.4411 0.251
0.519
1 .212
3.6
81
3.07
COLD 1972 FEDERAL
CO
AE/AP AE
gm/mi
-.8444 0.17
.02888 1.78
-2.441 0.32
.1121 2.04
4.081 2.32
6.63
5.34
35
81
1.40
COLD 1972 FEDERAL
NOx
AE/AP AE
gm/mi
.1394 -0.028
.0007306 0.045
.05792 -0.008
-.003049 -0.056
.2170 0.124
0.077
0.026
0.91
81
0.26
no
cr>
-------
Table 2.20
CALCULATION OF AVERAGE CHANGE IN EMISSIONS
EMISSION CONTROLLED VEHICLES
(FLEET 2)
PARAMETER
TIMING, degrees
IDLE RPM, rpm
PCV, cfm
A/C, degrees
ICO, % v
PREDICTED VALUES I
MEASURED VALUES X~
s
d.f.
t
CHANGE
-0.5488
- 11.12
-0.1193
17.07
1.174
COLD 1972 FEDERAL
HC
AE/AP AE
gm/mi
0.0498 -0.027
-0.00550 0.061
-0.08415 0.010
0.00090 0.015
0.02860 0.034
0.093
0.72
1.8
81
3.54
COLD 1972 FEDERAL
CO
AE/AP AE
gm/mi
-0.3423 0.19
0.02120 -0.24
-2.861 0.34
0.07642 1.30
7.177 8.42
10.01
9.00
28
81
2.88
COLD 1972 FEDERAL
NOX
AE/AP AE
gm/mi
0.09524 -0.052
0.000255 -0.003
0.08654 -0.010
-0.00212 -0.036
0.02926 0.034
-0.067
-0.222
0.90
81
-2.23
CO
-------
Table 2.21
CALCULATION OF AVERAGE CHANGE IN EMISSIONS
NOX CONTROLLED VEHICLES
(FLEET 3)
PARAMETER
TIMING, degrees
IDLE RPM, rpm
PCV, cfm
A/C, degrees
ICO, % v
PREDICTED VALUES I
MEASURED VALUES I
s
d.f.
t
CHANGE
-0.1686
1.977
-0.1598
26.12
0.8568
COLD 1972 FEDERAL
HC
AE/AP AE
gm/mi
0.07558 -0.013
-0.0004243 -0.001
-0.1548 0.025
0.0008945 0.023
0.07983 0.068
0.102.
0.345
1.6
87
2.06
COLD 1972 FEDERAL
CO
AE/AP AE
gm/mi
-1.022 0.17
0.02598 0.05
-7.458 1.19
0.06869 1.79
6.773 5.80
9.00
6.70
17
87
3.64
COLD 1972 FEDERAL
NO
AE/AP AE
gm/mi
0.1299 -0.022
0.0009632 0.019
0.1038 -0.016
-0.002476 -0.065
-0.02948 -0.025
-0.109
-0.07
0.67
87
-1.03
ro
cr>
vo
-------
the data, but that the HC and NO emissions predictions would result
x\
in poor predictions of emissions. In order to further investigate
the validity of the influence coefficients, computer runs were made
applying the coefficients developed for each major manufacturer to
the corresponding change measured in the test program. A correspond-
ing comparison was made using the major change in Hot 1972 Federal
emissions. In addition, predictions were made using results obtained
in the first vehicle initialization period (Test 1A and IB). The
results of the computer runs are summarized in Tables 2.22, 2.23, and
2.24. These results agree with the comparisons developed using average
changes in parameters. The best agreement was obtained for CO emission
measurements; HC emission measurements correlated, but the predicted
values were markedly smaller than the measured values. The statistics,
however, did not show significant differences in many cases. For the
data obtained for NO emission changes, there appeared to be a great
J\
deal of scatter between predicted and measured changes in emissions. In
some cases there was a lack of agreement in direction between predicted
and measured quantities. The predictions were also both larger and smaller
than the measured quantities for the different cases considered. There
appears to be a general lack of agreement between predicted and measured
values of NO emissions.
/\
Development of Uncertainty of the Influence Coefficients
In order to facilitate the statistical comparison of major changes
in emissions and predicted changes, a measure of uncertainty of the
influence coefficients was developed. The primary source of uncertainty
was obtained from the residual variance estimates developed in the analysis
of variance of test data. The uncertainty of the coefficient was developed
by applying the following relationship:
2-70
-------
Table 2.22
COMPARISON OF PREDICTED AND MEASURED COLD 1972 FEDERAL EMISSIONS
FINAL INITIALIZATION PERIOD (TEST 5A, 5B)
VEHICLE FLEETS
X"
PRE-EMISSION CONTROLLED ST
VEHICLES n
(FLEET 1) t
d.f.
Conf.
I
EMISSION CONTROLLED Sy
VEHICLES n
(FLEET 2) t
d.f.
Conf.
T
NO CONTROLLED ST
x VEHICLES n
(FLEET 3) t
d.f.
Conf.
COLD 1972 FEDERAL
HC
PREDICTED* MEASURED
gm/mi gm/mi
0.563 1.062
0.048 0.47
62
-1.05
61
70
0.125 0.738
0.018 0.20
67
-3.00
66
>99
0.095 0.242
0.010 0.57
75
-0.84
74
60
COLD 1972 FEDERAL
CO
PREDICTED* MEASURED
gm/mi gm/mi
7.658 5.412
0.55 4.7
62
0.47
61
<50
10.19 9.33
0.93 3.5
67
0.23
66
<40
8.61 6.68
0.58 1.9
75
0.95
74
70
COLD 1972 FEDERAL
N0x
PREDICTED* MEASURED
gm/mi gm/mi
0.221 -0.020
0.060 0.12
62
1.98
61
95
-0.095 -0.260
0.013 0.11
67
1.44
66
85
-0.111 -0.051
0.012 0.077
75
-0.74
74
60
PREDICTIONS MADE ON THE BASIS OF INFLUENCE COEFFICIENTS DEVELOPED BY MAKE
-------
Table 2.23
COMPARISON OF PREDICTED AND MEASURED HOT 1972 FEDERAL EMISSIONS
FINAL INITIALIZATION PERIOD (TEST 5A, 5B)
VEHICLE FLEETS
J
PRE-EMISSION CONTROLLED ST
VEHICLES n
(FLEET 1) t
d.f.
Conf.
I
EMISSION CONTROLLED %
VEHICLES n
(FLEET 2) t
d.f.
Conf.
X
NO CONTROLLED Sj
VEHICLES n
(FLEET 3) t
d.f.
Conf.
HOT 1972 FEDERAL
HC
PREDICTED* MEASURED
gm/mi gm/mi
0.591 1.533
0.051 0.314
59
-2.93
58
>99
0.115 0.502
0.016 0.170
67
-2.26
66
97
0.074 0.264
0.008 0.095
74
-1.98
73
95
HOT 1972 FEDERAL
CO
PREDICTED* MEASURED
gm/mi gm/mi
8.14 10.24-
1.59 3.24
59
-0.60
58
50
9.61 10.08
0.87 2.78
67
-0.15
66
<40
8.192 8.129
0.557 T.63
74
0.04
73
<40
HOT 1972 FEDERAL
PREDICTED* MEASURED
gm/mi gm/mi
0.243 -0.089
0.066 0.104
59
3.10
58
>99
-0.096 -0.298
0.013 0.117
67
1.71
66
91
-0.144 -0.055
-0.016 0.079
74
-1.10
73
96
ro
INJ
* PREDICTIONS MADE ON THE BASIS OF INFLUENCE COEFFICIENTS DEVELOPED BY MAKE
-------
Table 2.24
COMPARISON OF PREDICTED AND MEASURED COLD 1972 FEDERAL EMISSIONS
PRE-DETERIORATION EXPERIMENT INITIALIZATION PERIOD (TEST 1A, IB)
VEHICLE FLEETS
I
PRE-EMISSION CONTROLLED ST
VEHICLES n
(FLEET 1) t
d.f.
Conf.
I
EMISSION CONTROLLED ST
VEHICLES n*
(FLEET 2) t
d.f.
Conf.
X"
NO CONTROLLED ST
x VEHICLES n
(FLEET 3) t
d.f.
Conf.
COLD 1972 FEDERAL
HC
PREDICTED* MEASURED
gm/mi gm/mi
0.215 0.608
0.020 0.279
74
-1.40
73
85
0.250 0.395
0.035 0.176
TOO
-0.82
99
59
0.082 0.395
0.009 0.122
104
-2.53
]S2
99
COLD 1972 FEDERAL
CO
PREDICTED* MEASURED
gm/mi qm/mi
5.48 6.691
0.39 3.32
74
-0.35
73
<40
12.02 7.58
1.09 2.47
100
1.49
99
86
5.50 8.07
0.37 2.02
104
-1.24
103
80
COLD 1972 FEDERAL
N0x
PREDICTED* MEASURED
gm/mi gm/mi
0.088 -0.240
0.024 0.126
74
2.59
73
99
-0.016 -0.230
-0.002 0.167
100
1.27
99
80
0.009 0.056
0.001 0.102
104
-0.46
103
<40
* PREDICTIONS MADE ON THE BASIS OF INFLUENCE COEFFICIENTS DEVELOPED BY MAKE
-------
SB = SR
where;
SB = Estimate of standard deviation of the
influence coefficients
SR = Square root of the residual variance
DF = Degrees of Freedom associated with the
residual variance
S,, = Estimate of standard deviation of the
variations imposed on the parameters
Estimates were obtained for each parameter and each vehicle tested.
Since the emission levels of the vehicles tested were different, even
though they were representative of vehicles within the same test fleet,
i.e., Fleet 1, 2, or 3, the estimate of uncertainty of the coefficient
was divided by the measured effect and adjusted to give an estimate
presented as a percent of uncertainty. Estimates were developed only
when the effects were statistically significant. These estimates
were subsequently pooled to develop an overall estimate of uncertainty
for each main effect. The resulting estimates of uncertainty are
presented in Table 2.25.
In order to develop the functional relationship of the uncertainty
in the coefficient and the uncertainty of the final predicted result, a
measure of the fraction of total emission change that results from the
change in a given tune parameter was developed, using the data presented
in Tables 2.19, 2.20, and 2.21. Although it is realized that the distribution
of changes in parameters will vary, depending on the particular set of
vehicles tested, it was considered that the sample sizes were sufficiently
large to give valid estimates of the changes in parameters. It is
assumed the mean change in parameters would result regardless of the
group of vehicles tested. This assumption is certainly acceptable for
2-74
-------
Table 2.25
ESTIMATE OF STANDARD DEVIATION OF INFLUENCE COEFFICIENTS
PARAMETER
TIMING, deg
IDLE RPM, rpm
PCV FLOW, Cfm
AIR CLEANER, deg
IDLE CO, % v
NOX CONTROL, units
DEGREES OF FREEDOM
PRE-EMISSION CONTROLLED
FLEET 1
HC CO NOX
11.8 17.0 8.33
13.4 14.0 18.3
8.86 7.76 17.3
11.5 8.48 14.8
14.3 15.8 14.0
_
60
EMISSION CONTROLLED
FLEET 2
HC CO NOY
%o/
-------
use in prediction of overall uncertainties. These results also give
an indication of the combined influence of the magnitude of the
parameter change and the influence coefficients. The fraction of
the changes that are expected as a result in change of parameter
is summarized in Table 2.26. These results clearly suggest the
importance of idle CO and air cleaner performance for control of
emissions. Idle RPM appears significant in some cases, and PCV and
basic timing appear least important. This latter conclusion is
consistent with the observation that basic timing and PCV performance
are somewhat invariant in vehicle use.
The overall uncertainty of the predictions (Sy), as was
summarized in Tables 2.22, 2.23, and 2.24, was developed by statisti-
cally summing the uncertainties of each of the coefficients and the
fraction of total emission change as summarized in Table 2.26.
These uncertainties, expressed in percent of measured values, are
given in Table 2.27.
The values presented in Table 2.27 represent the uncertainty
of predicted changes in emissions associated with changes only in the
five parameters under consideration, i.e., timing, idle rpm, idle CO,
air cleaner restriction, and PCV valve restriction. Variabilities in
emissions of vehicle fleets in the field will be greater because other
parameters will malfunction. The magnitude of the uncertainties do,
however, suggest that within the limitations of the test program, the
uncertainties are consistent with test measurement uncertainties.
Repeatability tests conducted with NO Controlled Vehicles (Section 3.0)
s\
resulted in test-to-test estimates of standard deviation of 8.6%, 12%,
and 4%, respectively, for HC, CO, and NO emissions measured using the
X
Cold 1972 Federal Procedure. These can be compared with 10.6%, 6.8%,
and n.3/=, respectively, for uncertainties of predictions. The uncer-
tainty of predictions of NO emissions is the only case that appears
/\
excessively large. The uncertainty is great for this case because the
tune-up parameters considered did not correlate well with the NO
}\
emissions.
2-76
-------
Table 2.26
FRACTION OF TOTAL EMISSION CHANGE DUE TO CHANGE IN TUNE PARAMETER
PARAMETER
TIMING, deg
IDLE RPM, rpm
PCV, cfm
AIR CLEANER, deg
ICO, % v
PREDICTED, gm/mi
COLD 1972 FEDERAL HC
FLEET 1 FLEET 2 FLEET 3
-0.0250 -0.2903 -0.1274
0.3487 0.6559 -0.0098
-0.0058 0.1075 0.2451
0.1984 0.1613 0.2255
0.4836 0.3656 0.6667
0.519 0.093 0.102
COLD 1972 FEDERAL CO
FLEET 1 FLEET 2 FLEET 3
0.0256 0.0190 0.0189
0.2685 -0.0240 0.0056
0.0483 0.0340 0.1322
0.3077 0.1299 0.1989
0.3499 0.8412 0.6444
6.63 10.01 9.00
COLD 1972 FEDERAL NOY
^
FLEET 1 FLEET 2 FLEET 3
-0.3636 -0.7761 -0.2018
0.5844 -0.0448 0.1743
-0.1039 -0.1492 -0.1468
-0.7273 -0.5373 -0.5963
1.6104 0.5075 -0.2294
0.077 -0.067 -0.109
ro
i
-------
Table 2.27
SUMMARY OF ESTIMATE OF UNCERTAINTY (STANDARD DEVIATION)
OF AVERAGE PREDICTED EMISSIONS
EMISSION
COLD 1972 FEDERAL HC
COLD 1972 FEDERAL CO
COLD 1972 FEDERAL NOX
PRE-EMISSION
CONTROLLED
(FLEET 1)
%
8.7
7.2
27.4
EMISSION
CONTROLLED
(FLEET 2)
% '
14.2
9.1
13.4
NOX
CONTROLLED
(FLEET 3)
%
10.6
6.8
11.3
2-78
-------
Maintenance Effects on Emissions and Engine Parameters
The vehicles selected for the Parameter Deterioration Experiment
S
were restored to manufacturers' specifications prior to release in the
field. In this initial engine tune process all engine tune related
components, including carburetor replacement on some vehicles, were
replaced or repaired. At the completion of the deterioration phase of
the experiment, vehicles were again restored to manufacturers' specifi-
cations. In this final engine tune, misfire related parameters, i.e.,
spark plugs, wires, etc., and parameters for which response coefficients
were developed in the Orthogonal Experiments (Reference 1) were repaired.
In both engine tune processes, emission and parameter inspections were
made prior to and following the engine tune.
The average state of parameters prior to and following the tune
process at the beginning and end of the Deterioration Experiment are pre-
sented in Tables 2.28 through 2.33. The tune parameters for the three test
fleets are respectively presented in Tables 2.28 through 2.30 and the
average emission levels as measured with the 1972 Federal Procedure are
presented in Tables 2.31 through 2.33.
The data given as Test 1 are indicative of the tests prior to
(Test 1A) and following (Test IB) engine tune-up performed at the beginning
of the test program. For most vehicles Test 5 was performed prior to re-
lease of the vehicles following the deterioration phase of the experiment.
A few vehicles which required major repair during the deterioration phase
of the program were also tested prior to and following repair. These
tests were also designated as Test 5.
Although the data were indicative of the state following different
time periods of deterioration, a large number of parameters were at a
state close to the "as received" condition (Test 1A) when Test 5A was
performed. Timing varied non-systematically. However, the agreement
of the state of idle rpm, idle CO, and PCV flow rate, and air cleaner
restriction for Pre-Emission Controlled Vehicles (Fleet 1) and Emission
2-79
-------
oo
o
Table 2.28
EFFECT OF RE-INITIALIZATION
TUNE PARAMETERS
PRE-EMISSION CONTROLLED VEHICLES
FLEET 1
PARAMETER
TIMING, I
degrees s
d.f.
t
Conf.
IRPM, rpm X"
s
d.f.
t
Conf.
ICO, % v I
s
d.f.
t
Conf.
AIR CLN, X"
degrees s
d.f.
t
Conf.
PCV FLO Dl , I
cfm s
(49/45 Cruise)d-J-
Conf.
MILEAGE
TEST DATE
1A
0.44
5.8
14R
-
-
81 .2
141
148
-
-
6.08
3.0
147
_
-
60.0
53
137
-
-
2.87
1.3
126
-
83248
202
IB
0.014
0.28
146
-
-
12.4
45
142
-
-
5.86
2.6
148
-
-
31.7
44
136
-
-
2.93
1.1
126
-
83438
204
1A-1B
0.47
5.8
147
0.97
67
67.1
138
145
5.' 86
99
0.27
3.2
147
1.04
69
28.4
51
133
6.43
99
-0.09
1.4
124
-0.72
52
0
0
5A
-0.39
6.2
96
-
-
63.5
150
94
-
6.50
2.8
96
_
-
61 .1
41
91
-
-
3.09
1.5
83
-
93038
634
5B
-0.24
2.2
96
-
-
4.3
18
96
-
6.12
2.7
96
-
-
40.5
35
87
-
-
3.32
1.3
83
-
93233
635
5A-5B
-0.16
4.7
95
-0.34
< 4Q
56.5
144
93
3.79
99
0.39
3.0
95
1.27
78
19.3
31
87
5.79
99
-0.22
1.3
82
-1.52
86
n
0
-------
Table 2.29
EFFECT OF RE-INITIALIZATION
TUNE PARAMETERS
EMISSION CONTROLLED VEHICLES
FLEET 2
PARAMETER
TIMING, X"
degrees s
d.f.
t
Conf.
IRPM, rpm X"
s
d.f.
t
Conf.
ICO, % v I
s
d.f.
t
Conf.
AIR CLN, I
degrees s
d.f.
t
Conf.
PCV FLO Dl, X"
cfm s
(49/45 Cruise) d'[-
Conf.
MILEAGE
TEST DATE
1A
0.53
4.7
147
_
-
-17.3
108
147
-
-
3.98
2.5
147
-
45.9
53
136
-
-
2.619
0.97
136
-
43357
187
IB
0.063
0.88
141
_
-
-1.5
12
145
-
3.15
2.2
147
-
-
21.7
37
137
-
-
2.619
0.80
136
-
43357
189
1A-1B
0.33
4.5
141
0.87
61
-17.4
109
145
-1.92
94
0.83
2.7
147
3.77
99
22.5
49
133
5.36
99
0
1.0
136
0
<40
0
0
5A
-0.32
3.4
94
_
-
-15.2
89
94
-
3.67
2.8
94
-
-
43.1
40
88
-
-
2.872
0.88
89
-
53809
622
5B
0.19
1.2
94
-
-
-0.54
6.4
92
-
-
2.50
1.7
94
-
-
23.2
31
84
-
-
3.093
0.89
88
-
53826
622
5A.-5B
-0.51
3.5
94
-1.43
84
-12.8
91
92
-1.36
82
1.17
2.6
94
4.48
99
20.0
34
84
5.43
99
-0.227
0.80
88
-2.66
99
0
0
ro
Oo
-------
Table 2.30
EFFECT OF RE-INITIALIZATION
TUNE PARAMETERS
NOX CONTROLLED VEHICLES
FLEET 3
PARAMETER
TIMING, X~
degrees s
d.f.
t
Conf .
IRPM, rpm X"
s
d.f.
t
Conf.
ICO, % v I
s
d.f.
t
Conf.
AIR CLN, X"
degrees s
d.f.
t
Conf.
PCV FLO Dl , I
cfm s
(49/45 Cruise) d.f.
I*
Conf.
fllLEAGE
TEST DATE
1A
0.19
3.5
147
-
-
-5.7
138
149
-
-
3.34
2.7
149
_
-
28.8
43
130
-
-
2.574
0.90
135
8059
215
IB
0.04
1 .2
145
-
-
1 .8
35
148
-
-
2.81
2.5
148
_
-
16.0
32
133
-
-
2.63
1.1
134
-
8059
217
1A-1B
0.24
3.2
145
0.91
68
-14.2
94
148
-1.84
93
0.52
2.5
148
2.52
98
13.2
34
128
4.39
99
-0.057
0.82
133
-0.81
58
0
0
5A
0.14
4.6
104
-
-
10.6
136
105
-
-
3.00
2.6
104
_
-
42.3
38
98
-
-
2.82
1.0
97
-
24221
642
5B
0
0.55
104
-
-
1 .00
8.2
103
-
-
1.76
1 .8
105
_
-
17.8
22
98
-
-
3.04
1.2
97
-
24345
643
5A-5B
0.16
4.5
103
0.36
<40
11.1
138
103
0.82
58
1.24
2.4
104
5.32
99
24.5
31
97
7.86
99
-0.219
0.95
97
-2.27
97
0
0
00
ro
-------
Table 2.31
EFFECT OF RE-INITIALIZATION
EMISSION PARAMETERS
PRE-EMISSION CONTROLLED VEHICLES
FLEET 1
PARAMETER
72 COLD HC, X"
gm/mi s
d.f.
t
Conf.
72 COLD CO, X"
gm/mi s
d.f.
t
Conf.
72 COLD NOXP, I
gm/mi s
d.f.
t
Conf.
MILEAGE
TEST DATE
1A
12.48
6.9
147
-
-
135.4
63
147
-
3.68
1.7
146
-
83248
202
IB
11.17
4.6
148
-
-
125.3
53
148
-
3.88
1.9
148
-
-
83438
204
1A-1B
1.41
5.5
147
3.11
99
10.6
49
147
2.61
99
-0.24
1.4
146
-2.07
96
0
0
5A
16.0
12
96
-
-
146.0
84
96
-
~
3.40
1.6
96
-
-
93038
634
58
12.62
9.3
96
_
-
137.9
85
96
-
3.62
1.8
96
_
-
93233
635
5A-5B
3.34
9.5
95
3.46
99
7.7
59
95
1 .29
80
-0.22
1.5
95
-1.39
83
0
0
ro
oo
GO
-------
Table 2.32
EFFECT OF RE-INITIALIZATION
EMISSION PARAMETERS
EMISSION CONTROLLED VEHICLES
FLEET 2
PARAMETER
72 COLD HC, I
gm/mi s
d.f.
t
Conf.
72 COLD CO, I
gm/mi s
d.f.
t
Conf.
72 COLD NOXP, I
gm/mi s
d.f.
t
Conf.
MILEAGE
TEST DATE
1A
7.39
3.8
147
-
93.5
44
147
-
-
5.80
2.2
145
-
43357
i 187
I
IB
6.55
2.2
147
-
84.6
35
147
-
5.91
2.0
144
-
43357
189
1A-1B
0.84
3.4
147
3.01
99
9.0
34
147
3.18
99
-0.11
2.0
143
-0.69
51
0
0
5A
8.G5
10
94
-
-
96.6
60
94
-
-
4.72
1.5
94
-
-
53809
622
5B
6.40
3.1
94
-
-
82.4
34
94
_
-
5.. 02
1 .6
94
-
-
53826
622
5A-5R
2.25
9.1
94
2.42
98
14.2
51
94
2.70
99
-0.30
1.1
94
-2.54
98
0
0
I
00
-------
Table 2.33
EFFECT OF RE-INITIALIZATION
EMISSION PARAMETERS
NOX CONTROLLED VEHICLES
FLEET 3
PARAMETER
72 COLD HC, X"
gm/mi s
d.f.
t
Conf .
72 COLD CO, I
gm/mi s
d.f.
t
Conf.
72 COLD NOXP, X~
gm/mi s
d.f.
t
Conf.
MILEAGE
TEST DATE
1A
4.98
1.8
149
-
-
71.0
37
149
-
5.40
1.7
149
-
8059
215
IB
4.60
2.3
148
-
-
61.0
32
147
-
-
5.27
1.7
148
-
8059
217
1A-1B
0.37
2.0
148
2.29
97
10.0
28
147
4.32
99
0.12
1.4
148
1.07
70
0
0
5A
5.80
4.7
104
-
-
71 .1
44
104
-
-
4.45
1.1
104
-
24221
642
5B
4.63
1.5
105
-
-
57.7
29
105
_
-
4.65
1.3
105
_
-
24345
643
5A-5B
1.16
4.3
104
2.74
99
13.1
26
104
5.15
99
-0.193
0.88
104
-2.24
97
0
0
00
en
-------
Controlled Vehicles (Fleet 2) was extremely close. With the NO
J\
Controlled Vehicles (Fleet 3) the state of idle CO and PCV flow rate
at Test 1A and 5A showed fair agreement. As expected, since the
Fleet 3 vehicles were fairly new (8,000 miles) at the time of initial
tune process, air cleaner restriction measured at Test 5 was greater
than measured at Test 1A.
The average HC and CO emissions tended to be larger, and the
NO emissions smaller, at TEst 5A than at Test 1A. Normal wear of
A
all components other than those repaired would be expected to cause
these differences. In all cases, however, emission levels following
the two tune-up processes showed extremely good agreement. The magnitude
of the changes in emissions that were measured at the completion of the
program (after 10,000 to 15,000 miles of use) were consequently larger
in magnitude.
The data taken prior to and following the tune-up process appear
consistent. Attempts made to correlate predicted changes in emissions
as developed on the basis of measured changes in parameters and response
coefficients did not fully account for the changes observed in the
emission levels (this effort is described in Section 3.1.5). The mag-
nitude of the changes should, however, be considered as best estimates
for the changes in emissions that could be expected by restoring
vehicles to manufacturers' specifications.
2.4.2 Analysis Method
Comparison of Approach
Several approaches for developing the parameter deterioration rates
of continuous functions were investigated prior to the selection of the
method ultimately used to develop the rates. The final method used to
compute the parameter deterioration rates was one in which the slope, i.e.,
change in parameter with change in mileage, was calculated using the data
taken in two consecutive tests with a given vehicle. Averages for each
pair of test periods were computed, e.g., Test 2 - Test 1, Test 3 - Test 2,
2-86
-------
etc. The average values for each pair of test periods were ultimately
statistically pooled to develop the finalized deterioration rates. A
detailed description of the procedure is further described later in this
Section.
The initial approach used to develop the deterioration rates was
development of a linear regression of the change in emissions with
mileage accumulation from the values that existed following initializa-
tion of the vehicles (Test IB). The results of this initial effort
indicate, with the exception of the NO Controlled Vehicles, a very low
X
number of statistically significant relationships, particularly with the
Cold 1972 Federal Emissions. The results of the repeatability tests,
which were described in Section 3.0, suggested that particularly with HC
emissions a systematically lower value will be obtained if two consecutive
tests are conducted using the Cold 1972 Federal Procedure. Therefore,
on the basis of the assumption that deterioration rates are linear, linear
regressions were developed using Test 2 measured values as the reference
condition. These results gave more cases of statistically significant
results.
It was speculated that if the changes in emissions were expressed
as a fraction of the value obtained following the initialization of the
vehicles, a more meaningful result would be obtained. This speculation
had been previously suggested by members of the CRC CAPE-13 Committee.
Therefore, in an attempt to further improve the results, linear regressions
were developed using the changes expressed as a fraction of the Test IB
measured values, or the Test 2 measured values (two cases were investigated)
This approach further improved the correlations in that the Index of
Determinations improved for most cases considered.
A careful review of the data suggested that systematic biases in
the difference from the reference test values (either Test IB or Test 2)
resulted in the data. This would be caused by a high or low reference
value that could result from random variations of measured emissions. In
2-87
-------
order to eliminate these systematic biases the slope, i.e., change in
emissions with change in parameter, was calculated on the basis of two
consecutive tests. As would be expected, in many cases close examina-
tion of the results for a given vehicle indicated that slopes varied
both in a positive and negative manner. This random variability could,
however, be reduced by averaging the data. The arithmetic average of
the data for each of the vehicles between two adjacent tests was calcu-
lated to develop an estimate of the deterioration rate between two
discrete measurement points. These results were further examined for
consistency and those values which were statistically different from the
remaining set at the 95 percent confidence level were rejected. The
remaining estimates were finally pooled to develop the best estimate of
the deterioration rate.
The results of all of the investigations conducted to determine
the deterioration rates of Cold 1972 Federal Emissions are presented in
Tables 2.34, 2.35, and 2.36. In addition to the results of regressions
described in the previous discussion, the statistics for linear regressions
of the measured emissions, i.e., no reference value subtracted, are in-
cluded in the tables. The results, also summarized in Tables 2.34
through 2.36 indicate fairly consistent results between vehicle fleets.
It is apparent that the data obtained with the Pre-Emission Controlled
Vehicles exhibited the greatest amount of scatter, and the results with the
1971 NO Controlled Vehicles resulted in the least amount of scatter in
/\
the data. With the exception of this trend, a comparison of the different
approaches shows'comparable results. The significance levels, using Test
2 as a reference, tend to be higher than those obtained using the Test IB
values as a reference. The regressions were better, as indicated by the
larger student(t)values. The correlations were further improved when the
regressions were developed using the fractional (ratio) change in emissions.
In contrast, due to the large variability in emission levels of vehicles
in a given fleet, linear regressions using the "as measured" values (in
contrast to using the change in emissions) resulted in statistically
significant correlations (greater than 90 percent confidence) with only
2-88
-------
Table 2.34
COMPARISON OF SLOPES - COLD 1972 FEDERAL EMISSIONS
PRE-EMISSION CONTROLLED VEHICLES
(FLEET 1)
PARAMETER
Slope
COLD SEE
1972 FEDERAL t
HC d.f.
Conf.
Slope
COLD SEE
1972 FEDERAL t
CO d.f.
Conf.
Slope
COLD SEE
1972 FEDERAL t
NO d.f.
x Conf.
REGRESSION RESULTS
MEASURED CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
2 2
gm/mi . gm/mi
2. 695x1 O"5 1. 759x1 O"4
2.83 3.185
0.83 2.60
384 235
59 99
4.221xlO"4 1. 643x1 O"3
28.5 28.9
1.28 2.68
384 235
80 99
-2. 608x1 O"5 -1.571X10"5
0.891 0.766
2.54 0.97
382 229
99 76
FRACTIONAL CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
gm/mi gm/mi
5.103xlO"6 2.080xlO"5
0.284 0.256
1.56 3.83
384 235
90 99
5.992xlO"6 1.857xlO"5
0.297 0.261
1.75 3.36
384 235
93 99
-4.345xlO"5 6.559xlO"7
0.232 0.537
1.63 0.06
382 229
90 <40
MEASURED
p
gm/mi
5. 537x1 O"5
4.93
1.01
389
70
3.922xlO"5
52.6
0.65
389
47
-4. 035x1 O"5
1.66
2.18
387
97
RETRIEVAL
RESULTS
gm/mi
5.740x10"?.
22. 7x1 O"4
3.15
154
99
4.673x10":*
23.2X10"-3
2.85
154
99
2.327x10"!
6.78xlO~^
4.51
168
99
ro
oo
-------
Table 2.35
COMPARISON OF SLOPES - COLD 1972 FEDERAL EMISSIONS
EMISSION CONTROLLED VEHICLES
(FLEET 2)
PARAMETER
Slope
COLD SEE
1972 FEDERAL t
HC d.f.
Conf.
Slope
COLD SEE
1972 FEDERAL t
CO d.f.
Conf.
Slope
COLD SEE
1972 FEDERAL t
NOV d.f.
x Conf.
REGRESSION RESULTS
MEASURED CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
2 2
gm/mi gm/mi
-9.240X10"6 3.784xlO"5
1.88 1.33
0.52 1.77
393 244
40 92
2. 448x1 O"4 5. 255x1 O"4
22.6 18.1
1.17 1.80
393 244
75 92
-1.1 14x1 O"4 -3. 756x1 O"5
1.49 0.961
8.07 2.25
383 233
99 97
FRACTIONAL CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
gm/mi gm/mi
1. 987x1 O"6 9. 660x1 O"6
0.278 0.206
0.77 2.91
393 244
60 99
1. 429x1 O"5 1.1 84x1 O"5
0.354 0.226
4.37 3.25
393 244
99 99
-1.439xlO"5 -4.600xlO"6
0.203 0.179
7.62 1.48
383 233
99 85
MEASURED
2
gm/mi
-2.781xlO"5
2.54
1.19
395
76
1. 643x1 O"4
34.0
0.52
395
40
-8. 763x1 O"5
1.74
5.50
387
>99
RETRIEVAL
RESULTS
2
gm/mi
1. 333x1 0~J
"8.28x10"^
2.12
173
95
2.351xlO~o
11.4xlO"J
2.72
173
99
-3.744x10"!
7.59x10"^
181
6.65
99
ro
i
10
O
-------
Table 2.36
COMPARISON OF SLOPES - COLD 1972 FEDERAL EMISSIONS
NOV CONTROLLED VEHICLES
x (FLEET 3)
PARAMETER
Slope
COLD SEE
1972 FEDERAL t
HC d.f.
Conf.
Slope
COLD SEE
1972 FEDERAL t
CO d.f.
Conf.
Slope
COLD SEE
1972 FEDERAL t
NOY d.f.
x Conf.
REGRESSION RESULTS
MEASURED CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
gm/mi gm/mi
2.990xlO"5 8.392xlO"5
1.11 1.05
3.53 6.04
412 265
99 99
4.065xlO"4 6.101xlO"4
15.7 15.8
3.39 2.92
407 265
99 99
-6. 689x1 O"5 -1. 823x1 O"5
0.998 0.685
8.80 2.00
410 260
99 96
FRACTIONAL CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
I/mi I/mi
9.619xlO"6 2.132xlO"5
0.316 0.226
4.00 7.13
412 265
99 99
1. 339x1 O"5 1. 540x1 O"5
0.344 0.296
5.09 3.95
407 265
99 99
-1.030xlO"5 -2.651xlO"6
0.171 0.169
7.90 1.18
410 260
99 78
MEASURED
gm/mi
2.146xlO"5
1.84
1.56
426
87
3. 788x1 O"4
37.2
1.57
425
87
-3.940xlO"5
1.39
3.80
424
>99
RETRIEVAL
RESULTS
2
gm/mi
0.703xlO~4
3.73x10"^
2.83
186
99
0.8105xlO~o
4.57xlO~d
2.40
183
98
-1. 803x1 0"4
3.40xlO~4
7.45
207
99
ro
i
-------
NO emissions. The approach using the overall pooled value of the slope
/\
of the emission between tests (indicated as Retrieval Results in Tables
2.34 through 2.36) resulted in statistically significant values in all
cases. It should be emphasized that in many cases the approach in which
the slope between adjacent tests is used resulted in a higher magnitude
for the overall deterioration rate. As will be explained later in detail,
some of the results are due to the rejection of data between given tests.
The corresponding results obtained for the other parameters of interest
are presented in Tables 2.37, 2.38, and 2.39, for the three test fleets.
Selection of Method
The different deterioration rates that would result, depending
upon the method used to analyze the data, are very vividly illustrated
in Figures 2.12, 2.13, and 2.14. The three figures respectively represent
the plots of the change in Cold 1972 Federal HC, CO, and NO emissions
A
for the Pre-Emission Controlled Vehicle Fleet after vehicle initialization.
The various plots presented in each figure illustrate the different deterior-
ation rates of emissions developed by alternative methods for analyzing
the test data. The data points presented are those that represent the
change from the measurements taken following vehicle initialization. The
mean values obtained following each test during the deterioration phase
of the program are respectively represented with triangles and circles for
the total data set (A) and the set that would result with only vehicles
that completed the entire program with no maintenance (0). Review of
these discrete average values suggests and "s" shaped non-linear deteriora-
tion with a net effect of marginal change in emissions at the end of the
deterioration program, with approximately 10 to 17,000 miles of average
mileage accumulated on the vehicles. It should be observed that the agree-
ment of both the results with the total data set and the constant n data
set illustrate the validity of including all of the data in the analysis
regardless of whether or not the test vehicle remained non-maintained
throughout the deterioration experiment.
-------
TABLE 2.37
COMPARISON OF SLOPES - ENGINE TUNE AND KEY MODE PARAMETERS
PRE-EMISSION CONTROLLED VEHICLES
FLEET 1
PARAMETER
TIMING, degrees/mi
IDLE RPM, rpm/mi
AIR CLEANER, deg./mi
PCV FLOW, cfm/mi
CHOKE KICK, in/mi
49/45 MPH HC, ppm/mi
49/45 MPH CO, % v/mi
49/45 MPH NO v, ppm/mi
X
IDLE HC, ppm/mi
IDLE CO, % v/mi
IDLE NO , ppm/mi
A
REGRESSION RESULTS
MEASURED CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
3.107x10'!? -6.606x10"^
4.540x10":?* 1.838x10";:
1. 064x1 0"?* 9.509xlO~7*
-1.619xlO~7* -3. 241x10":?*
-4.828x10"'* -2.303x10"'
-1.295x10";?* 2.306x10"^*
-2.820x10"^* 4.062x10"^*
2.685x10* -2.473x10"^*
6.621xlO"J 8.871x10"^
2.414x10 "J 3.647x10"^
7.715xlO"4 3.887x10"^
FRACTIONAL CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
-
-2.503xlO"|j 1. 753x1 0"5*
-8.169xlO"r* 2.652x10"^*
2.060x10"° -6.638xlO"b
1.149x10";?* 2.138x10"^*
2.256x10"^ 3.448x10"^*
6.343X10"13* 2.837xlO"b*
RETRIEVAL
RESULTS
-1.481x10"^
3.766x10"^
5.307xTO~;:
-1. 424x1 0"^
-1. 889x1 0"X
-0.978x10"^
-3.366x10"^
5.644x10 ^
-0.382x10"^
1.334x10",
-3.140xlO~J
ro
i
10
co
*Regression significant at 90% confidence level
-------
Table 2.38
COMPARISON OF SLOPES - ENGINE TUNE AND KEY MODE PARAMETERS
EMISSION CONTROLLED VEHICLES
FLEET 2
PARAMETER
TIMING, degrees/mi
IDLE RPM, rpm/mi
AIR CLEANER, deg./mi
PCV FLOW, cfm/mi
CHOKE KICK, in/mi
49/45 MPH HC, ppm/mi
49/45 MPH CO, % v/mi
49/45 MPH N0v, ppm/mi
/\
IDLE HC, pptn/mi
IDLE CO, % v/mi
IDLE NOV> ppm/mi
X
REGRESSION RESULTS
MEASURED CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
-4.114x10";?* -1.214x10':?
-2.207x10".! 1.689x10",
1. 820x1 0~7* 2.016x10'^*
5.746x10"' -1.342x10"?
-4.929x10 '* -1.198x10 D
-1.210x10";?* 1.032x10';?*
5.769x10'; 1.210x10'^*
-1.198x10" * -2.877x10*
2.267x10'^ 5.252x10"^*
5.358x10 7 4.161x10 ,*
-5.845x10'^ -2.279X10'"*
FRACTIONAL CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
-
-
-6.064x10"^* 8.790x10'^*
1.676x10"?* 1. 860x1 0"c*
3.052xlO"4 3.125x10"^
1.039x10"^* 3.954x10'^*
6.879x10 ^* 5.904x10 ^
1.786x10" -7.928x10
RETRIEVAL
RESULTS
-0.409x10"^
3.766x10",
5.307x10"^
-1.424xlO~7
-1.889x10 '
-2.444x10'^
-0.278x10?
-1.203x10'*
0.046x10"^
0.274x10 ,
-1. 088x1 Q~
*Regression significant at 90% confidence level
-------
Table 2.39
COMPARISON OF SLOPES - ENGINE TUNE AND KEY MODE PARAMETERS
NOY CONTROLLED VEHICLES
A FLEET 3
PARAMETER
TIMING, degrees/mi
IDLE RPM, rpm/mi
AIR CLEANER, deg./mi
PCV FLOW, cfm/mi
CHOKE KICK, in/mi
49/45 MPH HC, ppm/mi
49/45 MPH CO, % v/mi
49/45 MPH NOX, ppm/mi
IDLE HC, ppm/mi
IDLE CO, % v/mi
IDLE NOV, ppm/mi
A
REGRESSION RESULTS
MEASURED CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
-3.187x10";? -7.051x10";?*
-4.645x10", 2.345x10",
1. 895x1 0~2* 2.205x10"^*
8.467x10"°* 6.650x10"°
-4.959x10"° -5.283x10"°
4.681xlO~7 -4.302xlO~J
-6.270x10", 6.1 93x1 0"°,*
8. 168x1 0~6* -1.572x10"^*
2.413x10";?* 3.303x10"^*
3.437x10"^* 2.870x10",*
-2.529x10"^ -1.888xlO"J*
FRACTIONAL CHANGE FROM
REFERENCE TEST
TEST IB TEST 2
-
6.551x10"!? 2.916xlO~J?
9.456x10"° 1.458x10"^*
3.705x10"°* -6.631x10"°*
1.681xlO"r* 2.177x10"^*
6.362x10"°* 1.400xlO">
1.141x10" -3.502x10"°
RETRIEVAL
RESULTS
0.345x10"^
-1.510x10",
2. 093x1 0"-%
0. 9700x1 0;°
-3.310X10"7
-0.228x10"^
-0.716x10";
-3.002x10"^
-0.013x10"?
-0.115x10",
0. 200x1 0"1*
i-O
on
*Regression significant at 90% confidence level.
-------
Figure 2.12
COLD 1972 FEDERAL HC EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
UPPER AND LOWER LIMITS OF
AVERAGE Of SLOPES BETWEEN TESTS
2 - TEST 3 DATA NOT INCLUDED
l;lillll:!lil:llil ill lillllll
TEST 2 SUBTRACTED
SLOPE = 1.759 x 10 gm/mi/m1
ffl
fEST IB SUBTRACTED
SLOPE = 2.695 x 10 " gm/mi/mi ffl
ATotal Set (Varying N)
OConstant N Set
CHANGE FROM TUNE CONDITION
-------
Figure 2.13
COLD 1972 FEDERAL CO EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
ro
i
UPPER AND LOWER LIMITS OF
HVERAGE OF SLOPES BETWEEN TESTS
TEST 2 - TEST 3 DATA NOT INCLUDED
TEST 2 SUBTRACTED
SLOPE = 1.643 x 10 " gm/mi/mi
FEST IB SUBTRACTED
SLOPE = 4.221 x 10 ' gm/mi/mi
A Total Set (Varying N)
OConstant N Set
(CHANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.14
COLD 1972 FEDERAL NOX EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
ro
i
-------
The deterioration rate of Cold 1972 Federal HC (gm/mi2) as is
illustrated in Figure 2.12, is representative of all of the figures.
For this case the slope of the linear regression developed using all of
the points plotted in the graph resulted in a slope of 2.695xlO"5 gm/mi2.
As depicted in Figure 2.12, this slope represents a negligible deteriora-
tion rate and is indicated to be negligible by the low value of index of
determination given in the regression results (summary of the results of
regressions are presented in Section 2.4.4). As is discussed in the
analysis of repeatability test data because there was some question as
to the validity of the test following vehicle initialization, a regression
was developed using the data taken at Test 2 as a reference. The resulting
linear curve is presented in Figure 2.12 with a slope of 1.759x10
2
gm/mi . Finally, by taking the difference between adjacent tests and
developing the arithmetic mean of all slopes calculated using each pair
of tests resulted in an average value of 5.74x10" gm/mi (not plotted).
The curves presented in Figure 2.12 give the upper and lower 95 percent
confidence limits of this mean value. The upper and lower limits are,
respectively, 9.32x10"4 gm/mi and 2.148x10" gm/mi2.
In the development of the slopes by using the calculated slope
between adjacent tests, clear indications of differences in average values
for given pairs of tests which were statistically different were observed
(Test 2 - Test 2). This difference is apparent in the plot of the average
values for Tests 2 and 3, presented in Figure 2.12. The upper and lower
limits indicated in the figure represent the results obtained following
rejection of the statistically significant outlier set. The rejection of
the outlier set is the primary reason why the slopes obtained by the
averaging method are markedly different than those obtained by the re-
gressions.
Clearly, the different approaches give different results. The
best method for developing the deterioration rate is not apparent. There
is a large amount of data scatter and the linear regressions obtained are
not statistically significant when the total data set is used. In contrast,
2-99
-------
if data subsequent to Test 2 are used, a statistically significant
(99 percent) regression is obtained. However, the data used in
regressions were in some cases obviously biased by the fact that sub-
tracting a given reference value from all subsequent values, a systematic
bias in all of the data points would result because of the random varia-
tion that resulted in the data measured at the reference point. Review
of computer listings clearly indicated positive and negative deviations
for certain vehicles. For example, with some vehicles negative devia-
tions resulted at every test and therefore suggested an erroneously high
value for the reference test. The calculation of individual slopes
between adjacent tests and averaging of the data tends to eliminate
systematic biases. Random variations will, however, exist in all of the
data and the development of the arithmetic average will tend to eliminate
the effects of random variation. The averaging technique was therefore
selected as the best method for developing the deterioration rates.
It should be emphasized that the averaging method tends to result
in high values if a situation in which large changes resulted due to
random variations of measurements and low mileage was accumulated between
tests. Obvious outliers have been rejected during the computational process;
however, this approach will weight the data in favor of the high deteriora-
tion rate values. Consistency of the standard deviations of all the
individual pairs of tests, however, suggests that the overall data set
was consistent. Further improvement may possibly be made by further
rejecting extreme values.
The results of the analyses conducted using the data taken with
Emission Controlled Vehicles and the NO Controlled Vehicles, respectively
A
presented in Figures 2 and 3, illustrate comparable results. The format
described in Figures 2.12, 2.13, and 2.14 is maintained throughout all of
the figures and is presented in the data summary in Section 2.4.4. Plots
representing the Federal Emissions of all the vehicle fleets, together
with the five continuous parameters considered most important to emissions
2-100
-------
control, e.g., timing, idle rpm, air cleaner, PCV flow (33/30 mph cruise),
and choke kick, and the emissions measured in the 49/45 mph cruise and idle
modes of the Clayton keymode cycle, are included in Section 2.4.4.
2.4.3 Variability of Coefficients
In addition to the development of the influence coefficients and
the deterioration rates in order to meaningfully make use of the Economic
Effectiveness model, it was necessary to develop a measure of uncertainty.
of the coefficients in the model. The final approach selected to develop
the deterioration rates, i.e., values obtained as the weighted overall
pooled value of individual slopes between adjacent tests readily allowed
development of the confidence limits of the values. The statistics
associated with each of the deterioration rates, together with the upper
and lower 95 percent confidence limits of all coefficients, are presented
in Tables 2.40 through 2,44.
The statistics associated with the emissions, as measured using
the Cold 1972 Federal Procedure, are presented in Table 2.40. The results,
in general, appear consistent between the three test fleets. As a whole,
the deterioration rates of HC and CO emissions decreased from the older
Pre-Emission Controlled Vehicles (Fleet 1) to the newer 1971 NO Con-
X
trolled Vehicles (Fleet 3). In contrast, the variation rate of NO
A
emissions was largest with the Emission Controlled Vehicles (Fleet 2)
as would be expected and substantially smaller with the NO Controlled
/\
Vehicles (Fleet 3).
From an engineering standpoint rather than differential changes in
emissions with mileage, it could be more practical to explain the change
as a fraction of the value that would be obtained with an initialized
vehicle. The deterioration rates, expressed in terms of a fractional
change from a reference value, are therefore presented in Table 2.41 .
To illustrate the magnitude of the changes that would result by using
a hot cycle method for determining emissions, the deterioration rates
obtained using the Federal Short Emission Cycle are presented in Table
2.42. The deterioration rates of the emissions using the High Speed
2-101
-------
Table 2.40
PARAMETER VARIATION RATE
CHANGE IN COLD 1972 FEDERAL EMISSION WITH MILEAGE
PARAMETER
X"
COLD SY
1972 FEDERAL d.f.
HC, 10"4 gm/mi2 ^1
X+kSTT
I-kSf
Conf.
I
COLD SY
1972 FEDERAL d.f.
CO, 10"3 gm/mi2 ^X~
X+kSY
X-kST
Conf.
X"
COLD SY
1972 FEDERAL d.f.
NO JO"4 gm/mi2 :T
X L
X+kSy-
I-kSf
Conf.
PRE-EMISSION
CONTROLLED
(FLEET 1)
5.740
22.7
154
1.82
3.15
9.332
2.148
99
4.673
20.2
154
1.622
2.88
7.869
1.477
99
-2.586
7.49
171
0.571
-4.53
-1.461
-3.711
99
EMISSION
CONTROLLED
(FLEET 2)
1.333
8.28
173
0.628
2.12
2.570
0.096
95
2.351
11.4
173
0.864
2.72
4.054
0.648
99
-3.744
7.59
181
0.563
-6.65
-2.636
-4.852
99
NOX
CONTROLLED
(FLEET 3)
0.703
3.39
186
0.248
2.84
1.191
0.215
99
0.8105
4.57
183
0.337
2.40
1.474
0.147
98
-0.3381
3.27
254
0.205
-1.65
0.0653
-0.7415
90
2-102
-------
Table 2.41
PARAMETER VARIATION RATE
FRACTIONAL CHANGE IN COLD 1972 FEDERAL EMISSION WITH MILEAGE
PARAMETER
X"
COLD Stf
x
1972 FEDERAL d.f.
HC, 10"5 mi"1 ^1
X+kSy-
x-ksi
Conf.
I
COLD Sy
1972 FEDERAL d.f.
_K _1 S
CO, 10 mi tX
X+kS^-
Conf*
X"
COLD Sx
1972 FEDERAL d.f.
NO , 10"5 mi"1 ^X"
X I*
X+ksx-
Conf.
PRE-EMISSION
CONTROLLED
(FLEET 1)
6.617
20.1
155
1.61
4.11
9.787
3.447 -
99
5.56
18.8
155
1.51
3.69
8.525
2.595
99
-5.092
16.9
172
1.28
- 3.96
-2.561
-7.623
99
EMISSION
CONTROLLED
(FLEET 2)
2.553
12.8
173
0.970
2.63
4.465
0.641
99
3.739
14.3
173
1.08
3.45
5.875
1.603
99
-5.073
9.12
181
0.676
- 7.50
-3.741
-6.405
99
NOX
CONTROLLED
(FLEET 3)
2.825
9.87
186
0.722
3.91
4.247
1.403
99
2.270
9.42
183
0.694
3.27
3.638
0,902
99
-0.884
5.58
254
0.349
-2.529
-0.196
-1.572
99
2-103
-------
Table 2.42
PARAMETER VARIATION RATE
CHANGE IN FEDERAL SHORT EMISSION WITH MILEAGE
PARAMETER
X"
FEDERAL Sv
SHORT d.f.
HC, 10"4 gm/mi2 ^
X+kST
T-kS
Conf.X
J
FEDERAL Sy
SHORT d.f.
CO, 10"3 gm/mi2 Jl
T+kSY
X-kSy
Conf .
I
FEDERAL Sy
SHORT d.f.
NO. 10"4gm/mi2 +X"
X U
y 4.L. c
A 1 N,O"y"
X"-ks4
Conf.X
PRE-EMISSION
CONTROLLED
(FLEET 1)
1.503
13.5
214
0.921
1 .63
3.317
-0.311
85
4.068
13.8
149
1.13
3.61
6.29
1.85
99
-0.3118
4.44
195
0.317
-0.983
0.313
-0.937
70
EMISSION
CONTROLLED
(FLEET 2)
0.2546
3.60
239
0.232
1.10
0.712
-0.203
70
1.218
6.48
238
0.419
2.90
2.044
0.392
99
-2.287
5.19
180
0.386
-5.929
-1.527
-3.047
99
NOX
CONTROLLED
(FLEET 3)
0.3427
2.14
250
0.135
2.54
0.609
0.077
98
0.2951
4.77
248
0.302
0.98
0.890
-0.300
70
-0.3076
2.54
240
0.164
-1.880
0.015
-0.630
90
2-104
-------
Cruise Mode (49/45 mph cruise) of the Clayton keymode cycle and the
idle mode, are presented in Table 2.43. Deterioration rates of engine
tune parameters, which are primarily used in controlling emissions,
are presented in Table 2.44.
In addition to the upper and lower 95 percent confidence levels
of the deterioration rates the corresponding limits of the influence
coefficients, i.e., change in emission with change in parameter, were
computed and are presented in Table 2.45. The estimate of standard
deviation used to develop the confidence limits was obtained as pre-
viously described in Section 2.4.1.
The approach in which individual slopes were calculated and
ultimately pooled to develop the overall deterioration rate is considered
to be the most appropriate because it eliminates the systematic biases
that might occur as a result of a random variation in the measurement of
a parameter during the reference test. The additional advantage of
utilizing this approach is that the statistical limits of the deteriora-
tion rate can easily be developed. In particular, in most of the engine
tune parameters, statistically significant values for deterioration rates
were not obtained (Air Cleaner Restriction was the only parameter that
resulted in significant deterioration rates). The results, however,
clearly establish the upper and lower bounds at which parameter deteriora-
tions can be expected. Investigations with the Economic Effectiveness
Model can therefore be made to determine the maximum and minimum expected
effect of mandatory inspection and maintenance of specific parameters.
2.4.4 Summary of Data
As was described in this section, a large amount of effort has been
directed in the analysis of data to develop the deterioration rates. A
method was ultimately chosen which is considered to be the most consistent
and meaningful. However, in order to present all of the information
developed in the course of the analysis, data summaries as well as graphical
2-105
-------
Table 2.43
PARAMETER VARIATION RATE
CHANGE IN KEY MODE EMISSION WITH MILEAGE
PARAMETERS
49/45 MPH
CRUISE ,
HC, 10~J ppm/mi
49/45 MPH
CRUISE ,-
CO, 10"D % v/mi
49/45 MPH
CRUISE ?
NO , 10 ppm/mi
X
IDLE 2
HC, 10" ppm/mi
IDLE .
CO, 10 % v/mi
IDLE 3
NO , 10 ppm/mi
X
X
SX
d.f.
SY
t
X+ks
X-ks
Conf.
X"
Sv
dXf.
t^
X+ks
X-ks
Conf.
JT
sx
dXf.
S)T
tX
X+ks
X-ks
Conf.
X"
SX
dxf.
SX
, A
X+ks
X-ks
Conf.
X"
SX
dxf.
So-
tX
X+ks
X-ks
Conf.
I
Sy
dXf.
s_
tX
X+ks
X-ks
Conf.
PRE-EMISSION
CONTROLLED
(FLEET 1)
-0.978
30.7
211
2.11
-0.46
3.176
-5.132
<50
-1.610
46.6
209
3.22
-0.50
4.725
-7.945
<50
2.306
26.9
199
1.90
1.21
6.053
-1.441
80
-0.382
26.1
208
1.805
-0.21
3.174
-3.938
90
1.334
14
217
0.948
1.41
3.202
-0.534
70
-3.14
24.1
101
2.39
-1.32
1.561
-7.841
80
EMISSION
CONTROLLED
(FLEET 2)
-2.444
21.8
238
1.410
-1.73
0.334
-5.222
90
-0.278
22.1
233
1.44
-0.19
2.568
-3.124
<50
-1.203
24.8
213
1.70
-0.71
2.137
-4.543
50
0.046
4.55
235
0.296
0.16
0.629
-0.537
<50
0.274
6.18
238
0.400
0.68
1.062
-0.514
50
-1.088
18.5
117
1.70
0.64
2.267
-4.443
<50
NOX
CONTROLLED
(FLEET 3)
-0.228
14.8
254
0.927
-0.25
1.598
-2.054
<50
-0.716
8.43
256
0.526
-1.36
0.320
-1.752
80
-3.002
19.8
141
1.66
-1.81
0.271
-6.275
90
-0.013
2.22
256
0.134
-0.09
0.260
-0.286
<50
-0.115
4.93
254
0.309
0.37
0.493
-0.723
<50
0.200
20.4
245
1.30
0.15
2.762
-2.362
<50
2-106
-------
Table 2.44
PARAMETER VARIATION RATE
CHANGE IN ENGINE TUNE PARAMETERS WITH MILEAGE
PARAMETER
J
TIMING, SY
-4 H f
10 degrees/mi ''
Y
tA
X+kST
Y-ksf
Conf.
X"
SY
IDLE RPM, d.f.
10"3 rpm/mi ^X~
X+kS
X-kS
Conf.
I
SY
AIR CLEANER, d.f.
10 degrees/mi ^X"
X+kS
X-kS
Conf.
I
PCV FLOWRATE Sy
(33/30 MPH d.f.
CRUISE), Sy
10-5cfm/nii I+kL
X-ks4
Conf.
I
CHOKE KICK, Sx
10"7 in/mi d'Jl
O \/
tX
X+kSy
X-ksf
Conf.
PRE-EMISSION
CONTROLLED
(FLEET 1)
-1.481
19.8
222
1.32
- 1.12
1.131
-4.093
70
3.766
52.2
218
3.53
1.07
1 0.. 71 5
-3.183
70
5.307
14.7
142
1.23
4.32
7.729
2.885
99
-1.424
31.8
160
2.51
- 0.57
3.513
-6.361
<50
-1.889
67.2
215
4.57
- 0.41
7.118
-10.896
<50
EMISSION
CONTROLLED
(FLEET 2)
-0.409
12.3
215
0.837
-0.489
1.240
-2.058
<50
0.053
29.8
219
2.01
0.03 .
4.011
-3.905
<50
2.555
6.14
192
0.442
5.78
3.426
1.684
99
-4.758
15.7
120
1.43
- 3.33
-1.946
-7.570
99
- 5.87
132
155
10.57
- 0.56
14.950
-26.690
<50
NOX
CONTROLLED
(FLEET 3)
0.345
6.20
249
0.392
0.880
1.117
-0.427
60
-1.510
19.1
253
1.20
- 1.26
0.856
-3.876
75
2.093
4.60
214
0.314
6.67
2.711
1.475
99
0.970
9.38
225
0.624
1.55
2.199
-0.259
85
- 3.31
70.4
193
5.05
- 0.65
6.647
-13.267
<50
2-107
-------
Table 2.45
SUMMARY OF MEAN AND 95% CONFIDENCE LIMITS OF INFLUENCE FACTORS
PARAMETERS
TIMING, X+ks
degrees X
X-ks
IDLE RPM, X+ks
rpm X
X-ks
AIR CLEANER, T+ks
degrees X
X-ks
PCV(33/30), X+ks
cfm X
X-ks
IDLE CO, I+ks
% v X
X-ks
PRE-EMISSION CONTROLLED
(FLEET 1)
HC CO NO
A
gm/mi/AP gm/mi/AP gm/mi/AP
0.08321 -0.5964 0.1667
0.06710 -0.8444 0.1394
0.05099 -1.0923 0.1121
-0.001802 0.04000 0.001097
-0.002930 0.02888 0.0007306
-0.004058 0.01776 0.0003639
0.007816 0.1437 -0.002057
0.005668 0.1121 -0.003049
0.003520 0.0805 -0.004041
0.03212 -1.434 0.08034
0.02283 -2.441 0.05920
0.01354 -3.448 0.03806
0.5295 4.939 0.2669
0.4411 4.081 0.2170
0.3527 3.223 0.1670
EMISSION CONTROLLED
(FLEET 2)
HC CO NO
X
gm/mi/AP gm/mi/AP gm/mi/AP
0.06155 -0.2259 0.1111
0.04980 -0.3423 0.09524
0.03805 -0.4587 0.07937
-0.004026 0.02714 0.0003483
-0.005500 0.02120 0.0002550
-0.006974 0.01526 0.0001617
0.001107 0.08938 -0.001492
0.0009000 0.07642 -0.002120
0.0006930 0.06346 -0.002748
-0.06924 2.417 0.1165
-0.08415 -2.861 0.08654
-0.09906 -3.305 0.0566
0.03678 9.445 0.03745
0.02860 7.177 0.02926
0.02042 4.909 0.02107
NO CONTROLLED
X(FLEET 3)
HC CO NOX
gm/mi/AP rp/mi/AP gm/mi/AP
0.08248 -0.893 0.1435
0.07558 -1.022 0.1299
0.06869 -T.151 0.1163
-0.0002964 0.03392 0.0010670
-0.0004243 0.02598 0.0009632
-0.0005522 0.01804 0.0008592
0.001145 0.08709 -0.001675
0.0008945 0.06869 -0.002476
0.0006443 0.05029 -0.003277
-0.1127 -5.904 0.1396
-0.1548 -7.458 0.1038
-0.1969 -9.012 0.0680
0.1025 8.052 -0.01749
0.07983 6.773 -0.02948
0.05718 5.494 -0.04147
o
CD
-------
plots are presented in this section to illustrate the various methods
used to develop indicators of the deterioration rates.
Graphical presentations of all the data points given as a change
from the Reference IB Test, are presented for each of the parameters
considered in the investigation. Graphs of the emissions as measured
using the Cold 1972 Federal Procedure and the emissions measured using
the 49/45 mph cruise and the idle mode, are presented for each of the
vehicle test fleets. In addition, data of changes in timing, idle rpm,
air cleaner restriction, PCV flow rate (33/30 mph cruise) and the choke
kick are presented. The plots of all parameters for the three test
fleets are presented in Figures 2.15 to 2.50.
On each plot the linear regressions obtained using Test IB for a
reference, or using Test 2 as a reference, are indicated. In order to
further illustrate the consistency of the data when averages are ob-
tained for each test period, discrete data points are plotted to indicate
the change that was observed in each test phase. Plots representing the
total data set (data set in which vehicle number varied throughout the
program) and the constant n set (n equal to the number of vehicles re-
maining at the end of the program) are included in the graphs. Also,
the upper and lower limits as obtained by calculating the slope between
tests and ultimately pooling the results are included in the graphs.
The summary of the data of all of the curves and points included
in the graphs is included in Tables 2.46 to 2.61. These tables represent:
1) Summaries of all of the statistical regressions
conducted;
2) The average changes that occurred during each
test phase, and
3) The average slope and pooled value obtained by
calculating the slope between adjacent tests.
2-109
-------
COLD
Figure 2.15
1972 FEDERAL HC EMISSIONS - EMISSION CONTROLLED VEHICLES
{iiffjflM^
UPPER AND LOWER LIMITS OF
£AVERAGE OF SLOPES .BETWEEN TESTS
;:; TEST 2 - TEST 3 DATA NOT INCLUDED
TEST 2 SUBTRACTED
SLOPE = 3.784 x 10~ gm/m1/m1
SLOPE = -9.240 x 10 " gm/m1/mi
tu
g ATotal Set (Varying N)
OConstant N Set
HANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.16
COLD 1972 FEDERAL CO EMISSIONS - EMISSION CONTROLLED VEHICLES
ro
i
UP ER AND LOWER LI ITS OF
AVERAGE OF SLOPES BETWEEN TESTS
tTEST 2 - TEST 3 DATA NOT INCLUDED
I
TEST 2 SUBTRACTED
SLOPE = 5.255 x 10 gm/mi/mi
FEST IB SUBTRACTED
SLOPE = 2.448 x 10 gm/mi/mi
A Total Set (Varying N)
O Constant N Set
i (CHANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.17
COLD 1972 FEDERAL NOX EMISSIONS - EMISSION CONTROLLED VEHICLES
ro
ro
TEST 2 SUBTRACTED
SLOPE = -3.756 x 10 gm/ml/ml
TEST IB SUBTRACTED
SLOPE = -1.114 x 10
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN' TESTS
TEST 3 - TEST 4 DATA NOT INCLUDED
A Total Set (Varying N)
O Constant N Set
CHANGE FROM TUNE CONUIUUNj, luuu roues
-------
Figure 2.18
COLD 1972 FEDERAL HC EMISSIONS - NOX CONTROLLED VEHICLES
ro
TEST 2 SUBTRACTED
SLOPE = 8.392 x 10 " gm/mi/mi
FEST IB SUBTRACTED
SLOPE = 2.990 x 10 gm/mi/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
FEST 2 - TEST 3 DATA NOT-INCLUDED
A Total Set (Varying N) i
OConstant N Set
-------
ro
t
Figure 2.19
COLD 1972 FEDERAL CO EMISSIONS - NOX CONTROLLED VEHICLES
TEST 2 SUBTRACTED
SLOPE = 6.101 x ICf gm/m1/m1
SLOPE = 4.065 x 10 ^ gpi/mi/mi
ER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS )
i:iii!|!ii!ii;;|! TEST 2 - TEST 3
DATA NOT INCLUDED
A Total Set (Varying N)
O Constant N Set
(CHANGE FROM TUNE CONDITION), 1000 miles
-------
ro
i
Figure 2.20
COLD 1972 FEDERAL NOX EMISSIONS - NOX CONTROLLED VEHICLES
zmmmm
urrQi\ nnu i_uncn. uirn i j ur
AVERAGE OF SLOPES BETWEEN TESTS
TEST 2 SUBTRACTED
SLOPE = -2.910 X 10
TEST IB SUBTRACTED
SLOPE = -6.689 x 10"b qm/mi/nri
A Total Set (Varying N)
O Constant N Set
CHANGE FROM TUNE CONDI
-------
Figure 2.21
BASIC TIMING - PRE-EMISSION CONTROLLED VEHICLES
TEST IB SUBTRACTED
SLOPE = 3.107 x 10 degrees/ml
TEST 2 SUBTRACTED
SLOPE = -6.606 x 10 " degrees/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
(CHANGE FROM TUNE CONDITION), 1000 miles
-------
BASIC TIMING -
Figure 2.22
EMISSION CONTROLLED VEHICLES
TEST 2 SUBTRACTED
SLOPE = -1.214 x 10 " degrees/mi
TEST IB SUBTRACTED
SLOPE = -4.114 x 10 degrees/mi
UrrtK t\nU l_UHC.r\ U It'll 10
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
(CHANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.23
BASIC TIMING - N0y CONTROLLED VEHICLES
TEST IB SUBTRACTED
SLOPE = -3.187 x 10 - degrees/ml
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
TEST 2 SUBTRACTED
SLOPE = -7.051 X
10 degrees/ml
A Total Set (Varying N)
O Constant N Set
liiliiii II 1 ;
I (CHANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.24
IDLE SPEED - PRE-EMISSION CONTROLLED VEHICLES
TEST IB SUBTRACTED
SLOPE = 4.540 x 10 rpm/rrn
TEST 2 SUBTRACTED ,
SLOPE = 1.838 x 10 rpm/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
ATotal Set (Varying N)
O Constant ft Set
CHANGE FROM TUNE CONDITION)
-------
Figure 2.25
IDLE SPEED - EMISSION CONTROLLED VEHICLES
TEST 2 SUBTRACTED
SLOPE = 1.689 x 10 rpm/mi
TEST IB SUBTRACTED
SLOPE = -2.207 x 10 rpm/mi
UPPER AND LOWEK L1MIIS Uh
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying Njftffl
O Constant N Set
CHANGE FROM TUNE CONDITION). 1000 miles
-------
Figure 2.26
IDLE SPEED - NOX CONTROLLED VEHICLES
i
ro
TEST 2 SUBTRACTED
SLOPE = 2.345 x 10"* rpm/mi
TEST IB SUBTRACTED
SLOPE = -4.645 x 10 " rpm/mi
U PER AN LOWER LIMITF
AVERAGE OF SLOPES BETWEEN TESTS
ATotal Set (Varying N)
OConstant N Set
-------
Figure 2.27
AIR CLEANER RESTRICTION - PRE-EMISSION CONTROLLED VEHICLES
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
TEST IB SUBTRACTED
SLOPE = 1.064 x 10 degrees/ml
TEST 2 SUBTRACTED
SLOPE = 9.509 x 10 degrees/ml
A Total Set (Varying N)
O Constant N Set
-------
Figure 2.28
AIR CLEANER RESTRICTION - EMISSION CONTROLLED VEHICLES
ro
CO
TEST IB SUBTRACTED
SLOPE = 1.820 x 10"" degrees/mi
TEST 2 SUBTRACTED
SLOPE = 2.016 x 10 " degrees/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS 3
A Total Set (Varying N)
O Constant N Set
-------
Figure 2.29
AIR CLEANER RESTRICTION - NOY CONTROLLED VEHICLES
A
TEST IB SUBTRACTED
SLOPE = 1.895 x 10 " degrees/mi
TEST 2 SUBTRACTED
SLOPE = 2.205 x 10 " degrees/ml
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
OConstant N Set
-------
Figure 2.30
PCV VALVE RESTRICTION (33/30 MPH CRUISE) - PRE-EMISSION CONTROLLED VEHICLES
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
TEST 5 - TEST 4 DATA EXCLUDED
TEST IB SUBTRACTED
SLOPE = -1.619 x 10" cfm/mi
TEST 2 SUBTRACTED
SLOPE = -3.241 x 10 - cfm/mi
ATotal Set (Varying N)
OConstant N Set
(CHANGE FROM TUNE CONDITION), 1000 miles
I
-------
Figure 2.31
PCV VALVE RESTRICTION (33/30 MPH CRUISE) - EMISSION CONTROLLED VEHICLES
TEST IB SUBTRACTED
TEST 2 SUBTRACTED
ATotal Set (Varying N)
OConstant N Set
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
TEST 3 - TEST 2 DATA EXCLUDED
I:;|H|:; s- :;;:::;; a;:;;
FROM TUNE CONDITION), 1000 miles
-------
ro
Figure 2.32
PCV VALVE RESTRICTION (33/30 MPH CRUISE) - NOX CONTROLLED VEHICLES
TEST IB SUBTRACTED
SLOPE = 8.467 x 10 " cfm/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
OConstant N Set
CHANGE FROM TUNE CONDITION , 1000 miles
-------
Figure 2.33
IDLE MODE HC EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
00
TEST 2 SUBTRACTED
SLOPE = 8.871 x 10 ppm/mi
TEST IB SUBTRACTED
SLOPE = 6.621 x 10 ppm/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
CHANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.34
IDLE HODE HO nssn* -
LEO VEHICLES
ro
i
ro
to
A Total Set (Varying
O Constant N Set
-------
Figure 2.35
MODE HC EMISSIONS - NOV CONTROLLED VEHICLES
ro
i
CO
o
-; TEST 2 SUBTRACTED
:; SLOPE = 3.303 X 10 " ppm/m1
TEST IB SUBTRACTED
SLOPE = 2.413 x 10 " ppm/ml
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
CHANGE ROM TUNE CONDITION
-------
Figure 2.36
IDLE MODE CO EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
SMS
ro
CO
TEST IB SUBTRACTED
SLOPE = 2.414 x 10 % v/mi
TEST 2 SUBTRACTED
SLOPE = 3.647 X 10 % v/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
OConstant N Set
(CHANGE FROM TUNE CONDITION), 1000 mile:
-------
Figure 2.37
IDLE MODE CO EMISSIONS - EMISSION CONTROLLED VEHICLES
oo
ro
TEST IB SUBTRACTED
SLOPE = 5.358 x 10 " * v/nri
TEST 2 SUBTRACTED
UP ER AN OWER LI ITS OF
AVERAGE OF SLOPES BETWEEN TESTS
ATotal Set (Varying N)
OConstant N Set
CHANGE FROM TUNE CONDITION
-------
Figure 2.38
IDLE MODE CO EMISSIONS - NOX CONTROLLED VEHICLES
ro
CO
CO
FEST IB SUBTRACTED
SLOPE = 3.437 x 10 I v/mi
ffiffi
TEST 2 SUBTRACTED
SLOPE = 2.870 x 10 % v/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
-------
IDLE MODE NOX EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
to
-p*
TEST 2 SUBTRACTED
JiHt+fflHIIIIIIIIUJIIIIIlllMII-Him
TEST IB SUBTRACTED
UPPER AND LOWER LIMITS OF
WERA6E OF SLOPES BETWEEN
A Total Set (Varying N)
OConstant N Set
CHANGE FROM TUNE CONDITION), 1000 miles
-------
Figure 2.40
IDLE MODE NOX EMISSIONS - EMISSION CONTROLLED VEHICLES
co
in
FEST IB'SUBTRACTED
SLOPE = -5.845 x 10 ' ppm/mi
TEST 2 SUBTRACTED
SLOPE = -2.279-x 10 J ppm/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
-------
IDLE MODE NOX EMISSIONS - NOX CONTROLLED VEHICLES
TEST IB SUBTRACTED
SLOPE = -2.529 x 10 ' ppm/m1
TEST 2 SUBTRACTED
UPPER AND LOWER LIMITS OF
SLOPE - -1.888 x 10 " ppm/nri
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
CHANGE FROM TUNE CONDITION, 1000 miles
-------
Figure 2.42
49/45 MPH CRUISE MODE HC EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
ro
i
CO
Itbl f. iUBIKRLItU
SLOPE -- 2.306 x 10 ppm/mi
TEST IB SUBTRACTED
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
-------
Figure 2.43
49/45 MPH CRUISE MODE HC EMISSIONS - EMISSION CONTROLLED VEHICLES
CO
I
CO
CO
TEST 2 SUBTRACTED
SLOPE = 1.032 x 10 ppm/ml
TEST IB SUBTRACTED
UPPER AND LOWER LIMITS
AVERAGEOFSLOPESBETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
-------
Figure 2.44
49/45 MPH CRUISE MODE HC EMISSIONS - NOX CONTROLLED VEHICLES
ro
i
co
ID
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
OConstant N Set
-------
Figure 2.45
49/45 MPH CRUISE MODE CO EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
ro
.£»
O
PEST 2 SUBTRACTED
SLOPE = 4.062 x 10 % v/ml
TEST IB SUBTRACTED
SLOPE = -2.820 x 10 " % v/m1
UPPER AND LOWER
AVERAGE OF SLOPES
A Total Set (Varying
OConstant N Set
-------
Figure 2.46
49/45 MPH CRUISE MODE CO EMISSIONS - EMISSION CONTROLLED VEHICLES
rv>
-P.
TEST 2 SUBTRACTED
SLOPE = 1.210 x 10 " % v/mi
TEST IB SUBTRACTED
SLOPE = 5.769 x 10 % v/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
OConstant N Set
:HANGE FROM TUNE CONDITION, 1000 miles
-------
Figure 2.47
49/45 MPH CRUISE MODE CO EMISSIONS - NOY CONTROLLED VEHICLES
ro
ro
TEST 2 SUBTRACTED
6.198 x 10 " % v/m1
TEST IB SUBTRACTED
SLOPE - -6.270 x 10 " V v/ra1
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
A Total Set (Varying N)
O Constant N Set
-------
Figure 2.48
49/45 MPH CRUISE MODE NOX EMISSIONS - PRE-EMISSION CONTROLLED VEHICLES
CO
TEST IB SUBTRACTED
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTS
m
SLOPE = -2.473 x
PP
A Total Set (Varying N)
O Constant N Set
CHANGE FROM TUNE CONDITION
-------
ro
i
49/45 MPH CRUISE MODE NO,
Figure 2.49
EMISSIONS - EMISSION CONTROLLED VEHICLES
rEST IB SUBTRACTED
SLOPE = -1.198 x 10 - ppm/mi
TEST 2 SUBTRACTED
SLOPE = -2.877 x 10 ppm/mi:
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES BETWEEN TESTSffiffl
A Total Set (Varying N)
O Constant N Set
-------
cn
Figure 2.50
49/45 MPH CRUISE MODE NOX EMISSIONS - NOX CONTROLLED VEHICLES
I Li I IB bUBIRACIED
SLOPE = 8.168 x 10 " ppm/mi
TEST 2 SUBTRACTED
SLOPE = -1.572 x 10 ppm/mi
UPPER AND LOWER LIMITS OF
AVERAGE OF SLOPES 'BETWEEN TESTS
A Total Set (Varying N)
D Constant N Set
-------
Table 2.46
AVERAGE COLD 1972 FEDERAL EMISSION
I
^
O>
Cot-D J9Ti
&&£&)<-
"C,l
ff°Xj f^/>^-
TomL SET
COLD /9?i
FeoeeaL
Ht,^"-
corfsrAtSr
// SET
COLO >y)2-
F£I>£GAL-
C0,f~/i~
ctHiSTrtfT
// S£T
&X.D /97Z
F£D£&)L-
(V^f/*"
CofSSrtff/T
y ser
TEST
f/0
X
s
d.f
Conf.
MILEAGE
X
5
d.f
t
Conf.
MILEAGE
X
S
d.r
\
Conf.
MILEAGE
X
S
d.f.
f
Conf.
MILEAGE
X
S
d.f.
t
Conf.
MILEAGE
X
S
d.f.
t
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
Ifl
I/.4L
4.5
14°
83G6
J&.1
59
I4o
03438
\33±
1.1
I32_
83438
/0
/o-95
4.6
I4o
83439
IZ2.9
53
I4o
'3438
4.01
/)
13?
83*38
/o.9t
4-4
/9
83438
ni.o
4o
/?
.
83138
4.o8
'7
/ff
83tel
Z
ti^4(L
5:2.
95
8-7/05
25.4
Si
95
57/oS
3Ji
1-5
92
311 OS
//.42
6.L
/?
8t&3
125,8
55
11
ab?j
3.55
1.8
18
8*93
3
/o.Z/
4,£
t*
89542-
U4-T
42-
lt
8J54L
2.Z3
1.4
&
69^2-
/o.tn
f.o
/9
8903
/27-?
44-
U
884Z?
z.?f
/2-
J3
39^2
4
tt.fo
f.4-
46
.
?Z2o?
/3?.t
15
4C
"ftZoJ
3.5-Z
I.S
44
322??
/o.a?
/Z
/?
9°7zo
w-t
4t
/?
Jfftzo
3.48
/.(,
/a
.
Joflo
5rt
/2.&
U._
/?
-^r
?3$4/
141.8
47
/9
")384I
3.39
/.s
/8
93841
I2£?
4.?
/9
-
?&4(
14/.B
41
If
1&4(
3.38
/.*>
/8
?&f
ffl
//.Z7
3 5
IJ
13841
132.4
41-
11
JSdil
3.43
{ +
/8
'BQ+I
11,11
3.5
JJ
139(1
1324
^2_
/?
?38t(
3.4,3
1,4
18
13W
EMISSION CONTROLED
FLEET 2
JA
7 2C
3.7
US
«35?
92.0
4o
14*
~ .
43331
5.86
2.2.
141
43357
'
/#
i.Zo
y.z
Io4
. ......
«?57
83.£
33
I4f
43357
5.96
2.0
///
.
43357
t.tf
Z.I
Z?
43&1
85. f
24
2?
43357
4.87
2.4
Z£
46357
z
t.ll
34-
102.
48/0+
90.9
3t
107.
.
4sio4
4.&1
/.4
«"
48/04
5.74
_A?
ZJ
4M8
8/.?
32-
2?
4M8
f.3?
S.L
26
.
.
+X48
3
5.81
Z.Z
48
-zrr
5J3M.
81-4
23
(.*
5/9?Z
4-08
/t
62.
5?5>Z
5.3Z
/7
Z3
50310.
16.1
31
-11
5o3/0
44f
/5
7J.
5o3/o
4
i.oz
/. ff
43
.
sstst
8S.(.
-L-)
43
525"^
4-81
/«*
4o
jaz:s£
5:79
/3
29
53105
63.8
28
29
53/05
f.OO
/3
2.6
55*5
5>?
UZ
z.t
Z9
57^7
?/.8
40
Z?
57/n
4-t?
//
26
5?//7
<^.^
Z.t
29
57/n
?/!
40
^
*r»l
4.67
/.4
2(>
5"7//7
5ft
S.3t*
l.i
Z9
.
57/n
75.7
23
Z8
sun
y,f?
/.5
25
siin
s.at,
i.t,
28
57111
13.1
Z.3
Z8
57111
5.4l
1-5
75
57/77
NOX CONTROLLED
FLEETS
IA
4.18
1.8
/48
,
80S")
71.0
37
147
&5?
5.4o
1,8
14Q
8oS?
/#
4.to
2,3
H9
.
SoS")
(,1.0
3Z
141
805?
5:27
/7
/48
SoS?
4.30
A3
3o
eo5?
63. /
1<-
z?
80S?
5. 58
u
2.9
3oS1
2.
4J.lt,
1.2-
//o
13730
te.s
3/ _
IOJ
ffl3o
4.3?
/.2.
/O8
/3730
4.21
/./
3o
.
13331
58.5
Zt,
29
/3331
4.T2,
//
7.1
/33$7
3
4.11
/.?
7?
IKZ.9
58.5
33
7t
17621
t//
/2.
211/Z
tfZ
//
52
.
2J?/Z
4.32.
1-4
3o
Zo7/7
60, 7.
10
z?
_
237/7
4.44
Ao
29
2C7/7
^
5.5e
2.0
30
ZS^J
t4>'*
31
29
ZS4C3
/^
/.o
Z9
25443
S.50
Z.O
Jo
2544:
641
31
Z9
2544*
4,61
/.O
z9
*3W
5?
w
/.3
to
2$4S<
so. 9
ZO
Z9
ZS#3
4.14
/.2-
ZJ
1
Z5445
4.14
J.3
3o
75^3
56.9
ZO
21
_
25441
434
/Z
ZJ
.
ZS44.
-------
Table 2.47
AVERAGE CHANGE IN PARAMETER FROM TEST IB - COLD 1972 FEDERAL EMISSIONS
ro
i
COLD /97Z
TOTAL s£7
COLO /972.
F£D£ftyl
COLD i?TL
T£ST
f/0.
X
S
d.f
t
Conf.
MILEAGE
X
S
d.f
t
Conf.
MILEAGE
X
S
d.f
l
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
Ifl
o.5l
3.4
I.TI
42
0
10.4
4B
140
1°\
o
-O.Zt
1.4
134
-2.20
1?
0
*V
//./
4.C.
m
ni.&
53
- -
3.1)
1-1
1 41
2
3.5
45
2.47
10
3S3Z.
5.1
37
I-3C
86
353£
-0.51)
0.11
13
- 5
5V?
/Z.
4 i
o.&°l
62-
KI8*
\
-o>4Z.
J.I
-?./!
Ife
szgz.
5/^
IT
i. 36
gl
10 &m
//.3
21
il
1. 65
88
/.jj>44
-0-14
l.t
17
-7.57
/<>04<
^?
T75
|7
o.u^S
^4o
lootfq
/.(,
37.
i7
.)..;(
Z40
/DO 44
I.I
-2.t6
iix>44
EMISSION CONTROLED
FLEET 2
I/)
.0.%
3.3
I45
2.17
>11
0
*.4
145
2.^1
>71
c
2.0
HI
55
J,
"?&
6,55
|4T
84. C,
35
|4T
.
5.11
2.0
144
2.
°2\
\ot
0-50
^40
fil'Z?
4 5
26
1 0 £
/,t3
89
-/25
1.4
95
^-8.55
>11
4535
3
2. 1
fc?
-A2-J
f1/
75S3
-/4
/It
67
-0.43
"-40
7SM
1-7
6 /
-11:13
>17
746^
/
1. 8"
43
-i.se
81
IC/lfo
1-5
i7
43
o.3<:
i-yo
lo'tog
-l.U
I.I,
40
-7.52.
'11
/O<
S/)
o.l?
2.5
28
o,4\
^40
/Jg^S
7.6
37
/.;/
73
-2 /?
2.4
25
-4.6")
~>11
I32U,
^?
-0.^,
1. 8
z7
-'95
95
/3f/3
-/o.g
-v
,7
9o
/J9/3
-/.5/
2. I
24
-3. S3
7-f 9
/J^f^
NOX CONTROLLED
FLEETS
/tf
(7.J7
^.o
I4P>
17
o
/o.o
2g
141
4..V-
..
o./e
1-4
148
l.c>7
78
0
^
4.60
2.3
I48
1,1.0
3t
\41
5.27
I'l
146
.
2
-0.04
/3
1 16
~
ItbSf)
4,11
I7iol
S^
o.ga
/,5
28
Z.S4
?7f
I7£o(,
n.-4
Zo
27
0.01
'Mo
/7^J6
-0,70
\-6
Z7
-3,^i
T-^?
/7fo>
AVERAGE CHANGE IN PARAMETER FROM TEST IB - FEDERAL SHORT EMISSIONS
TOTAL. s£7
snoer
roTXL S£7
J8**!:
TEST
//O
X
S
d.f
t
Conf.
MILEAGE
X
S
d.f
t
Conf.
MILEAGE
X
S
d.f
t
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
1*
0.3Z
Z.I,
I4o
|.48
0
7.1
21
131
>41
0
-<>,o&
I-3
\3g
-0.11
S£
o
*^fr
v.i"!
3.Z
1 4^
.
7t>t
Hft
-
z.%
1.5
i47
',i
o.ooz
2. 1
14
*Ho
3S3Z
5.6
24
14
?.?6
17
35 13
-C.26I
0.7-?
#?
-3.6o
»>)i
3fcoC
3
-0.33
7.Z
-Lit
78
594?
3-5
z?
kZ
I. I3
T5
5830
-0,218
a?4
5"6
-2-67
7 71
(,04k
4
n "j 5
'?&
41
l.lt.
fio
/5.4
31
>?44
1-&
21
I./I7
gz
11ZI
- ft./j^
«.7<]
\i
-O.K.
$1
loS°l9,
EMISSION CONTROLED
FLEET 2
IA
o 30
2.2
145
I.Z8
80
t,.Z
21
z.st
99
o
j,<>a
/,5
HO
0,18
5/
0
I'C
3.86
1.8
l°*
I4T
.
.
4.Z4
1.7
143
C
-o.o(,
1. 1
IOZ
-0.57
«/s
4U6
2.7
IOZ
00
4^20
-o.i?
LZ
95
-5. Jo
?99
45?^
3
-0.07
I.I
tg
-0.5/
7^52
S.O
20
67
2.07
15
7S25
-I.Zi
I.Z
4/
-7-?7
711
?i/4
4
a. 34
I.I
4^
/.ee
11
!OlS4
'zi
42
3-10
?"»9
totio
-o.fZ
1.4
39
>14
/om
5/1
o.41
I-2-
?.o&
9k
13116
8-3
25
27
1.17
12-
/3?o4
-o.g/
l.fc
24
11
/3?#0
55
0.08
1-11
?.(,
0.43
^0
/i/0Z3
-3. 1
Z^
-tf.t?
SB
I3i?1
-0.02,
1.8
23
-o. 05
^VO
13311
NOX CONTROLLED
FLEET 3
//I
0,230
o.fZ,
M6
3.05
7 11
,^
4-4
15
147
3-49
???
o
0,28
1.7.
M7
2,3u
71??
O
^
2.472.
o.g?
I4B
e9.3
20
147
3.04
1,3
147
-
Z
0,1*54
o.T/
lo7
o.os
^0
55 7f
/.5
>1
I Ok
i.ia
7S
5S/?g
-0.5/
I.I
1 05
-4.76
79"?
5512.
3
0.077
O.l>1
70
0-^4
65
^2:
i3
7o
8L-
no*)
-o.U
\,1
67
711
<<25Z.
H
o.T,\l
o.gl
4t
2.4/
1&
12151
5.1
15
A(,
Z.35
18
12125
-0-25
M
43
-I.5o
8S
uin
5A
i.i1!
h 1
z7
5.43
/?5/^
//.f
2Z
2,85
7l1
I7&4&
-o.-^t
\.T-
25
-I.S5
88.
'7555
5-5
/.OS
i. 7
Z7
3,ZZ
7>ff
I7S/7
1.2.
II
26
0.54
11
/764B
-0,11
\.\
i5
-o.g4
(1
/7SS5
-------
AVERAGE CHANGE IN PARAMETER FROM TEST IB - COLD 1972 EMISSIONS
COLD 1972.
fns
t/C
O3U>
y SET
Cot.0
CD
esr
Conf.
d.f.
Conf.
MILEAGE
d.f.
Conf.
MILEAGE
PRE -EMISSION CONTROLLED
FLEET 1
I/I
3.5
'7
0.37
-0,70
55
-O.JS
I.Z.
17
-0.54
0.003
17
Q.ooS
-10,6
n
11
275?
^1^0
^757
-0.17
z.g
n
a^k
65
'1.0
-05/5
45
57 73
713
3.7
U2.S6
25
-o.H
-2."3
5/9
M;
_U3_
i7
se
.n
_I2-
-z.ife]
98
loo'H
EMISSION CONTROIED
FLEET 2
I.Oi
25
i i >
"
Tf
js
-/.<»
88
r>
0.1.1
2.6
20
-1.81
1 2.
^L
i'fl
on
25
l.ol
1.7
-3.4ft
^Li
95
~l'48_
i.f
zs
-i.57
1.0
20
-o.L
-fl.lt
25
2.8
27
LJJ12.
1±.
90
-/.S/
-3.53
7" 17
NOX CONTROLLED
FLEET 3
JM
2o
4-tJ
/.O
56
-^a.z3S ALli(-
Jfl
L11
l.o
-4.57
5373
"9.17
26
2-C.
15
0.11
ts
27
7./Z.
M
A z.
.-Y70
711
I'Kof
iA0_
_12_
Ilidl
50
I.S
Z.O
o.eff
-o,70_
7.o
27
ittol
AVERAGE CHANGE IN PARAMETER FROM TEST IB - FEDERAL SHORT EMISSIONS
PRE-EMISSION CONTROLLED
FLEET 1
EMISSION CONTROLED
FLEET 2
NOX CONTROLLED
FLEET 3
5*
50
10
ss
S/)
SB
St/oeT
0./3
-a. so
-0.1J
-0.13,
-0,137
n.oK
0.661
o./Zt
1.11
i.os
34
2.U
3.2
z.f
1.4
0.7?
O.IT,
l.i
0.12-
ff^Z.
d.f.
IT
-12.
n
17
27
in
77
27
27
Z7
27
27
2?
o.l.
-/.SO
-I.5I
l^L
0.30
-Ul
o.oZ._
5.43
3.22
Conf.
ja_
MILEAGE
zvs?
5/73
5349
(75/7
/7S/7
0-31
7.0
IS-*
-K
'ho
Z.2.
H.1
2Z
Ifl
Z?-
2Z.
25
17
14
2Z.
d.f.
SET
Conf.
-"to
It.
Z7
27
zc.
ill
il3_
-Q-/1
MZ.
_as_
fz-
₯0
70
"7Z-
<*
7ft
MILEAGE
J331
UU,
54 5i
/Wffj
-o./te
-t>,;33
-o,//
-l.ol
-0,02-
-o.8to
-6.W
-a. HI
^Jo.
oTL
1.1-
1.3
1.8
l.t
d.f.
15
o.ZZ
15
15
15
23
25
Lll
-0.3?
Z.S3
-.Ml
-2.34
-f.oS
_^S_
^.Zl
2.5
23
-Z.It
-/.55
25
Conf.
(.1
1.1
11
It
1V
It,
11180
ItSS
(7S2
-------
ro
i
TOT/H-
TOTAL
Table 2.49
AVERAGE CHANGE IN PARAMETER FROM TEST IB - 49/45 MPH CRUISE MODE
PRE-EMISSION CONTROLLED
FLEET 1
Conf.
MILEAGE
Conf.
MILEAGE
50.1
10
Ho
0.23
1.9
l.sg
81
Conf.
137.
\0l
141
3.01
1475.5
H7
z?. 4
354S
-0.21
3L
35 TJ
711
3U4
70_
u
-1.40
85
0.84
_lo_
57
2.0
51
_38.
0i3«L
-O.It
41
1030
556
A 20
5/I
n
0,5-}
Ji
0.30
loo 4 4
-O.Zl
. .11
-o.ol
"JO
10044
SB
-18.0
35
n
-0.306
0.51
-2.53
10044
1/0.4
356
IV
EMISSION CONTROLED
FLEET ?
/A
35.8
334
|43_
I.Z4
80
113
sg
-o.to
50
Iflo.s
13?
l.o
145
1.43
o.Sg
t?l
11
4415
Jl.
7^52-
0-03
5
13^45
27
0.7S
-11-0
<&
23
-0.85
60
lc>S
2?
rli^
13113
-^1Z
2.3
ZC
16
75
^z
1334^
NOX CONTROLLED
FLEET 3
(48
2.55
18
o
148
2.52
147
-0.50
2.81
2.S
148
1±_
^3,25
1-8
0-86
5-7
lot
7.13
97
S7-/
-0.00}
JJL
71
I.?
72-
J.SO
f/13
90
?773
47
. 87
o./B
1-5
47
sz
JH
I28I&
5A
f/.e
^,S_
48
6.31
HZ
I71S3
AL
2S
-^.77
-l.lo
75.5
It
-------
Table 2.50
AVERAGE CHANGE IN PARAMETER FROM TEST IB - 49/45 MPH CRUISE
MODE
ro
en
o
vfaMPitatti
4}/ff/*eHc#i*
«v.~-
riser
r)/jf/tficwr*.
ft/o^i/if^
ro/vSr/^A/T"
//o.
X
s
d.f.
I
Conf.
MILEAGE
f
X
S
d.f.
t
Conf.
MILEAGE
X
S
d.f.
,
Conf.
Miu AGI
IA
12.1
$1
i1)
3 -08
>1"l
a
-0.181,
0,11
n
-o.Rf,
to
o
75-8
471
0.40
53
o
PRE-EMISSION CONTROLLED
FLEET ]
Ifi
I
-zz.z.
3o
iT
-3./C
f1<\
?15l
0.73
n
2 75?
3 '8, 4
17
m?
3
-4-V
44
n
o.45
fVo
5/73
-0.3K.
1.1
-Z.lfc
_33_
Z8Z.6
371
n
't?
±113
4
-Z-f5
1 7
-0.24
«to
74o4
-o 138
o.Slj
J.T
-/.il
76t>6
2.V
17
a,P3
1lo(,
5*
-7.Z
sz
n
-o.5f
50
/oo4i
t.cfl
l-o
17
o.3o
^Vt)
/uoH
-o.g
55?
17
-0.01
IOOW
5£
-18-0
35
n
?£
10011
-0.304
o.s/
-/"
/%
i7
/-3*
-30
EMISSION CONTROUD
FLEET 2
)A
I3Z--1
450
Zl
1-06
Q
Z6
-1.08
I'l
O
11 0
;^i
iB
:.
---
^
-12.4
32-
-Z-01
li« '?S
"z7^
o 83
4337.
252.4
971
24
1.44.
4o7o
3
/t.o
35
27
17
4738
"" '??
0.7S
. 18-
^i f\
-^
1.13
01
444 (
/I
- /$, 7
38
-(.53
1154
- o. '3o
0.47
2fc
16
1173
-112.1,
44?
-0,-»Z
61
5A
-?4_.3
41_
Zl
711
i3U>j
-o. (2?_
0.78
26
-1.15
_J£__
)37//
- /W-4
not
21
50
-2? 4
Z1
26
-5.?7
/37//
-a/H
*.?S
2fc
-/..H
so
/^.6
/OSO
z4
0.0?
'32%
NOX CONTROLLED
FLEET J
Ifl
i? 3
41
t'fe
0.1?
75
o
o.os;
0-2/
/.Zfl
&o
o
-M-3
ZG
"-Tj|4
^8
Zfe
1.41
35
mil.
SA
iff. 3
45
,Z4
I. Oil
17^
j^.OJj
0.3?
Z8
0.47
^"/O
\ltoi
Zfc
0,5^
Llz,
/tist
523
zo.6
41
26
I, to
f?
l<7&5
0.010
' 0.33
26
O.It,
<»/0
/9M
Z53-4
<7St
Zt
2. SO
71f
/fKZ.
AVERAGE CHANGE IN PARAMETER FROM TEST IB - IDLE MODE
A/ ser
HX-C Utt
}
fit ser
fjp t^ng
ri se-r
res-r
f/0
X
s
d.f.
1
Conf.
MILEAGE
X
S
d.f.
t
Conf.
MILEAGE
X
S
d.f.
i
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET t
M
-n.t
244
i7
-0.20
<-fO
b
0.34
3.9
i7
o.4o
j-i\0
o
n.t
18
M
|.|0
71
0
/S
z
-54-4
I")}
n
-/34
SO
275?
o 30
| <^
17
0.47
57
24.0
feS
il
1.47
^
2627
3
-112.0
1oS
n
-M?
11
5/73
0.05
3.3
i7
a.ol,
"HO
S/73
31.5
8fe
11
I -73
lo
41Z5
?
-ff.O
«7
-a 23
/3865
-------
Table 2.51
AVERAGE CHANGE IN PARAMETER FROM TEST IB - TUNE PARAMETERS
PRE-EMISSION CONTROLLED
FLEET I
EMISSION CONTROLED
FLEET 2
NOX CONTROLLED
FLEET 3
-resr
//O.
5/1
0.47
0.01
-Jo
-17
.06
5.B
0.21
J-5
-.32
2.8
o.oj
-/2_3
2.1,
-l°<>
0.005
O-2o
32
-l.oo
3.57
3.2
3.T-
2.1
3.9
3.o
^L
57
97
23
115
/o?
45
27
o.W
o.o2
-.24
-.08
Conf.
70_
-/. go
1o
-l.lt,
.3.20
'V
-2.30
IK
70
Tft
ff
0.55
-/77
45
40
35/9
5792
$359
414-1
75-75
HOOD
5639
1301$
/7136
IKPM,
HJL**^
TPTAI- $er
&7J_
JSLl
So.O
^1
-7.3
-o.i
-23.1,
6.3
-3.2-
138
33
87
in
154
/o
ID'/-
7?"
?f
SS
67
d.f
14^_
20
20.
/oo
22-
11
45
SM-
3.
-.33
-i.n-
-/.OS
-2.71
0.4?
77
Conf.
IK
15
7?
56
MILEAGE
5314-
9/U
4749
)347
J2?74
23/30
-.3?
-.0/5"
-.13-7
-.034
-.123
-.-23
o.l4l
0,021
-.Z-C2-
-.021-
-14?
0.218
o.ni
l.-i
0.67
o.(,5
O.Zo
ZlL
oM.
oJ4_
0.52
037
"37
JZ2,
13-
50
31
12_
95
2^-
\33
64
1*
2-1
-J.S5
-/f
-4.11-
32.
11A.
35-
20
22
12-7
Sii
Jo
z/
21
36
23
22.
5.39
3.53
2.JO
Conf.
712.
-rtf
-77
4,01-
4,84
5.83
223-
MILEAG
77tf
2^1
13520
-.ooolB
.o^^
1am- 5er
52.
52.
35
/,sy
-.031
Z/
-/./o
.00.1
9
OOOiO
"n
2.
.0024
.047
/.oo
Ilk
.041-
-,o/ai
.014
-,oo3 1
-oo3o
OPS?!
.0028
-54
72
O.S2.
M-
Conf.
7°
45
70
-_iU
&S_
.o4o
.022.
021
.0084-
.010
-tf
_L2J.
53
35
2.0
-Z.ff?
-I.*}
-12-
IL27
85
33=^
A2/_
_Z4_
1JM.
4926
S7/7
9775
13/53
/7220
-------
Table 2.52
AVERAGE CHANGE IN PARAMETER FROM TEST IB - TUNE PARAMETERS
//o
PRE-EMISSION CONTROLLED
FLEET 1
I/I
EMISSION CONTROLED
FLEET 7
3
NOX CONTROLLED
FLEET 3
iff
54
5,5"
°j_5A
LJ=_
21
o. //
A/5"
-o<3
-A 23
.11.-
0.11
zz
Coof.
4°
ST
ML
MILEAGE
2553
22-
-/00
_70
57
z3
.-.a
so
^735
1.5
23
-A 7?
2,6
Z3
-Z.V
°I7
23
S-SL.
1.2
21-
SAoL
17 _
.2-5.
-A 77
szr
121,0
25- 0
36.0
/o,o
S
d.r.
Jf*
3 ,'sj'
1°L
Zo
(25-
^i
1-7
-/a
-22. o
1o
Conf.
15
^u£/_
.z^
A 7o
10
ZT-
IX
^1Z
18
r^i
_-^L
Mo
2/
A2£
^3_
_Zi_
..5^_
£7
Z/
^£
.z*.
zJZ
2Q_
o
53-19
\IJ1I
-.038
-.17.5
-.z^z
-ft?Z
-.So
0.5T-
0.3 S
0.32
^14
o^
-/24.
0,33
0,3%
..J2 -J.3-
1T,
-2.1}
22.
2-1-
11.
11-
Z.T-
a7/
-.11
H-
l-ii-
2/
(-04-
Conf.
JSS-
85
Qo-^^2.
^99
5fi_
2<2_
MILEAGE
9257.
5733
I
tn
MK
.ZJ_
//.*
17:5
12.0
lo.T,
32-
£Z_
Zo
It*
18
d.f.
/.Oil
^/
/73
Conf.
zy
J_§
-LI
^/_..JV._.ll
23
Z3
23
Z3
Z.I-
z3
2.3;
25^
MILEAGE
1//1
3.
-.oooi
-.oaoZi -.pan
.ooola.
,o/7/
.OZI
.0/6
.oool.'i
,03J_
.080
.0355
Cadsmmr
fj
ser
d.f,
-02-
_2
-.//
Ai?o
SS
It
__
-A 79
/3
,002$
13
.045
.012
',00 II
.0047
oil
.0028
.010
Zo
Zo
Zo
ZO
n
-/.37
1.17
ffo
75
MILEAGE
/TZi"
-------
Table 2.53
AVERAGE SLOPE OF FRACTIONAL CHANGE OF COLD 1972 FEDERAL EMISSIONS WITH MILEAGE
ro
in
CO
cfteeAL
CO, 16
-5 I .1
COLO
MILEAGE
d.f.
MILEAGE
d.f
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
TO
353 1
1.111.
I6.o
U
-S.&7
/S.fl
13
-3.5J
Z/.7
-2,35
aw 2.
-/.V/
n./
10
2 712.
z-f.S
JlL
43
J2_
v/
17
I2.7
\1
ui
/r.5
17
'7
TJ
i.t/7
/55
/S5
/t.t
IfffAc-
i.tiv
20.1,
tl.l
-l.Ztl
n-i
713
EMISSION CONTROLED
FLEET 2
~resr
2.
lol
1. 51
10. t
loi-
3.31
1 7
-5.36
V535
r£sr
3
I0.1.
3o/a
Uu-
3375
-33?*
^LZL
3olO
14
LI
resr
TJ>_
4}
3375
l.lfi
3311
5f\
1-K5
3WJ
zg
U
37V/
Z5
-t.ei
PootfD
3 731
3.45
-5.073
I0I
9?
in1)1/
2ZI
NOX CONTROLLED
FLEET 3
Uz.
sai
J.50Z.
5.2.
-5.30
797
-2.67
TS08
7Z.
-^?/
3871
JT^
-i.tt
10
5.271
11
-5-2L
??r
J.JVOJg)
37^
jiL
3732.
-13-
17 If
/i.'i
"Izl
1.10
I. ill
z?
f3et£DlffTA<-
HIM
i.z7o
3.27
l
-------
AVERAGE SLOPE OF COLD 1972 FEDERAL EMISSIONS WITH MILEAGE
INS
in
J:, c ' '
~! 'jt '
J
>nj> ii-il
hiuHi
'O^f./^
Com /9r-
fr »>*<
AQ/O/vi,..1
X
s
d.f
Conf.
MILEAGE
X
S
d.f.
t
Conf.
MILEAGE
X
S
d.f
t
Conf.
MILEAGE
RUE-EMISSION CONTROLLED
FLEET 1
-
ifcr
5V:V
11.1
If
rr.z7
j'sfr
7
-?'W/y
T&
tr
-t.u
w
"10
. -' "f-S*
)/}
i-1
/,-/
.0
' V«
J'*.T
,-%
il
-All
;;
*tl«"
/_"
HkO
-tTf
45
/1C
94
*3?r
;.4??
4.":
1 43
/«9
«
??7r
.-«*"
7^7
m
!76
W
*Z~l
^
l.zti
9.Z/
ze
1.91
03
3741
4tW
" ^
".&
/.if
re
U4I
-:>4C
£Cr
f~
-/.tc
7rt
7., -
£sv-i£
/.}{}
&1t
'73
r 'Z
Qf
:.?ri
14
/7*
'.n
VI
\M4
/«>
ISI
-ti,1:
<) "9
/Of)
'-..g
n
ntjr
-IK3
?.-y
{
-(-.3
V
r&4
*'t>
-u4*
4.S-
fl
-,"'-
er
HCj 1C ff~
srtoer
~3 / 1
CO^ioy*/*
F£V£#AL
srt&T
tot,r~/~~
X
s
d.f
1
Conf.
MILEAGE
X
S
d.f.
1
Conf.
MILEAGE
X
S
d.f
t
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
TTfirr
2
&-5J(
12. S
)4
0.3-)
«/o
3532.
3.IS2
/3.7
34
2.21
9?
3573
-/-Of*
3.^2
58
-3. IK
>9?
J636
-resr
3
-/Itz
I3.&
t2
-1.03
70
2(n4
O.ZSJ
(4.9
tz
o (4
<4o
2(.8t
0.353
T.40
Si
0.4?
*4o
24)9
z*tz
">.t?4
/4.4
39
?./?
38
Z3kS
1.133
4.-lo
3t
2.11
96
1633
TfST
5*
2455
If. 3
/7
7.07
?S
2.431
S.V3
iZ.f,
li
/.£)
9?
2491
-37?
3.ol
-LS_
-O.So
'<4o
/a 1?
fbKfE
/.6o3
13. S
1/4
1.13
35
<<*/
13. S
t4?
3.t/
??
-.312.
iff
135
-0.98
10
EMISSIONCONTROLED
FLEET 2
Tesr
2.
-.033
J.TO
132.
-o.oi
<-4o
4(,U
t 335
t 6f
1C 2.
/.Zfi
3f
ms
-2.Z/3
5".J?
95
-i.zi
'93
4S?t
-resr
3
-.671
4 el,
63
-113
Bl
3o31
/ OOo
*.8(
i7
I.4Z
85
z?$?
-Z.H4
(,.oi
t/
-278
^9
3ol3
fesr
4
I.5JZ
l.ol,
42.
34i
*?)
3257
3.Z04
8./1
12-
2.58
?2
33ol
l-TIS
1-32.
39
z.-si
??
3o?y
TEST
5/«
1-1/t
z.ts
21
2.8Z
-??
3803
o-ffl
1/0
?1
o/4
«jo
3535"
-2.919
z.53
Z4
-0^8
SO
Z5£7
/%*«£
o. 255
3 tO
23?
I./O
7o
1.218
1.48
Z3&
2. JO
9?
-2.ZK1
5. /I
180
-5.J2.
??
NOX CONTROLLED
FLEETS
TTFSr
z.
O.tOo
/t?
101
o./8
-to
5579
o.Zii
3.83
Jot
OJJ^.
So
5S9S
'.%0
?.37
106
-4-n
Zl2_
5St2
resr
3
-oil
2.K
10
-O.li
-1o
38)0
-241
t.3t
7a
-o.3Z
<-4°
ZBSo
-SS-?
3.ol
(1
-I-S2-
88
3L70
TZ-S-T
4
O.SI5
1.59
46
3.Q
79?
37^5
O.tfo
3^
46
/.Z3
id
3757
I.1o5
2.31
43
3.tS
'9?
55/7
-resr
SA
IW)
/??
27
4.5')
'99
4t(,o
I-Z21
5.03
2&
1.11
79
46Q±_
o.?75
2. IS
25
o.ts
So
4T5L
Ftxx.£t>
0.343
2,/f
250
2-54
78
OttS
4.17
Z1R
0.9 ^
70
-.3o8
251
240
-1.88
?o
Dtp fO~-
ni ttaLfD
-------
en
en
Table 2.55
AVERAGE SLOPE OF 49/45 MPH CRUISE EMISSIONS WITH MILEAGE
nfe * M*
W, /arfT^-
vv%~^,*
s / *
C"0 10 7.t/r~-
11/45 nff Cf«sf
-2 , 3
X
S
d.f
t
Conf.
MILEAGE
X
5
d.f.
1
Corf.
MILEAGE
X
S
d.f
1
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
rfS{
31-5
11
1?
35 Ti
a. OH
21.1
88
<) g^
»11
3tt4
resr
3
5.I7I
30.3
(.2.
1.3C.
as
Ct'fj'l
-fl Co-)
45-0
fco
Ml)
Zlot
-i.! SC-
SI 4
s?
i? 11
il
2718
rrsr
'll'i
16_
o.8fi
£444
^O.J
11
'.45
16
/(. M;
-'2.3
-.iT
isio"
77F37-
sA
-5/H5
_17
-1 01,
75
lO.tJ
I'/
0,
?< 7
-V
/ z/
80
1 MISSION CONIROLED
FUET 2
rysr
-faJII
25-2.
lol
«n
"n.i
ICO
'^
V67V
oP£.
.1 , -L
CO^O?,'/*"
TU£/*>0*
^^rr/^
X
s
d.f
t
Conf.
MILEAGE
X
S
d.f.
t
Corf.
MILEAGE
X
S
d.f
1
Conf.
MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
r£sr
2
-o.i 17
_1SA
M
-o.oC
^6
3SJ1
i.oJo
IZ.I
15
l.fcl
81
3532.
u.m*
^7.1
00
l.5t
af
IStt
rrsr
^
-5.813
40. t
fco
-Lit-
7t
?tsg
-0.111
it.o
bl
-o.16
^10
Z667
-o.lo1?
zt.t
«
-o. zfe
-Ho
/ft»
r«r
4
H 515
14. f
31
111
15
j'55J
-0.216
-J>.|
42
-0 13
^HO
25I3
-1.151
_Z3.0
^2
-l.lt
a.o^i
4.4
235
a, /£
^5o
o.T-W
(,.T,
23?
o.i?
So
-1.083
11. S
in
-0.6,£
^5o
NOX CONTROLLED
FLEET 3
rzfs-r
-2.
'0-252
t. 1
10 1
-l.il.
SO
5534
0-ol8
<.?
(07
O.o4.
^^o
5551
^.0*
;4. o
(Qt>
Z.24
«,7
_S£2i
-resr
3
j iii
^.3
Tl
0.42
^40
3«B7
-O.lii.
Lit
12,
-0,/(>
^0
?M1
i.ool
ZS.8
10
CL3L
*-4o
3110
resr
t
a.llo
t-z-
47
I.Zo
11
37Z0
-0. Z&.
2.5
^^
-a.SS
50
}1tt
~gos£
II.SL.
41_
-MS
?£
J?53
TEST
54
-0.072.
^.s
^8
-
-------
Table 2.56
AVERAGE SLOPE OF TUNE PARAMETERS WITH MILEAGE
ro
i
to
.3
/o
-3
lo
-S
PRE-EMISSION CONTROLLED
FLEET 1
d.f
1
Corf.
MILEAGE
d.f.
Conf.
MILEAGE
d.f
Conf.
MILEAGE
d.f
Conf.
MILEAGE
d.f.
Conf.
MILEAGE
-/Si?
r) :
97
- -So
rfST
2
511
OS
*?-*
_11_.
$-150
^L/_
34??
->U77
3336
I.&72
34.4
52
3142
-2.11&,
0 02
2&5J
-i. 3S2
9.?
4$
L-J3-
23 a
2538
-/3.33S
35
-.8k
1.5
- $4
,--5-.<9
r- o
ir^/±
2431
1£^£.
3/
^?^-
i-HO
21,11-
-.23
7^
J2.__
JZi^
-.OS
/731
43 2
.U3
2.1
50
2Z5/
z/5
^.3Z
Ili..
-1-931
7. Z
2/5"
-.-.4/
EMISSION CONTROLEO
FLEET 1
TETT" rezr re^r
O.I
mo
'. -21
/I.
""" J*.P
5o
JL4
73
-4I.I33
-A55"
61
58
15
-TJlt
?
-Li' -
r-?^
25:
-lZi'5
-?.37
'32
/55
-.56
NOX CONTROLLEP
FLEET J
-/C77
o.io
03
-£.7
10
'IS
-V. I
J
1?
/7/i
J.-C-
6.'
-A^/-
!). I
75
-J1
-3.11
/// POOLED
-------
Table 2.57
LINEAR REGRESSION OF COLD 1972 FEDERAL EMISSIONS WITH MILEAGE
PRE-EMiSSION CONTROLLED
FLEET I
18
[MISSION CONTROLLED
FLEET 2
.e
NOX CONTROLLED
FLEET 3
AlB
£OLZ> /97Z
'.08m 10 i
A(0)
/79/v/o
SEE
&254_ r_i
J/35
d.f.
.0-234
^Q06±_
334
3./55
/. S3
Conf.
225"
O.Q3
>99
1-33
_A77_
.227
4.00
7.13
3.5
87
A(0)
0215
to.
125.1,
-0215
2CI
23. 5
CO
00^
, 0460
52.'.
, QOll
d.F.
84
235
235
A 75
A 23
IX)
en
Conf.
_^3_
30
'99
^^i/a
, V9.Z
2^.i
^Z5_
.2%
15-3
24
-------
Table 2.58
LINEAR REGRESSION OF FEDERAL SHORT EMISSIONS WITH MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
/e
42.
EMISSION CONTROLLED
FLEET 2
18
NO\ CONTROLLED
FLEET 3
'&
^3 Z
ro
i
tn
00
Ad)
A(0)
0, Qlil*
_SEE
d.f.
t
'/.fi
Conf.
O.C34J \-./Oi,8
i-tl \l.9}
,oo$4 \.oZZt,
_3lo \ 225 _
1.43 | 4.5(~
83
I/I1/
ill.
. (-So
. l$$c
Z44
A (01
SEE
3.0k? -.
O.'/Ol
.n-2.
d.r.
z.y.
0,0151
?35
'M6.
Conf.
95
15.1.
2. MIL
' W
i-Cl
22 i
4.11
A(0)
-o.ofcZU
O.Oftl'J
5££
0./37
\-oli3
1-k.V^'"
OjOSLP
0 ZiS
l.ot
d.f.
?6o_
c,S5
Ml
Conf.
.txoZ. .o/oS
370
.?-Z?-l/.5/_
_581
7S
j!LS5_
'?7
S. 2. 1
.??
J?.<₯ it
» '-'fr
n.iill'f
SZ4
1841
C5(,
10. Oi
ZS8
5./8
5J3of.ro
-.000 5
.(.It,
ooi3
--3-5/
-.-±9.
A.
leaf" 2. .
t- .0
,ri
o-t
'
-------
Table 2.59
LINEAR REGRESSION OF 49/45 MPH CRUISE EMISSIONS WITH MILEAGE
PRE-EMISSION CONTROLLED
FLEET 1
AI&
EMISSION CONTROLLED
FLEET 2
NOX CONTROLLED
FLEET 3
A/ 3'
\\C,
Co, 7,
tn
MPH C*
alit
.7S3M0
(V110KIO
-1.2101(10
1.0% A fo
A(0)
_SEE
d.f.
u-aoW
o.zsi
0.111
3.772.
41.8
.0/3?
3-1.7
30.
.0/80
0,317
0.0018
-3.ZL
_ 3g_. 0_
381_
235
t
Conf.
fcZ
77
.3/
95
IfO
3,2?
244
4ZZ.
J..13.
IB
JJk
I.It
75
53
/.&/
_8.2_
64
_A(0)_
SEE
R/4
o.oi^l
0.374
d.f.
o- oo9
381 "
MS
.035?
o.o&3^ \.o2H
,-o. oi jj
o-ZII
.650
0.351
Conf.
-.0502 J-.0/47 |
.455 '-297 j
..-^2. :.ozs^ I
3S3 j Zf / |
80 | 99
o.oZfcS
O.OZSS
.301
Oooi.
-,o2oi,
.Z(,l
4Z3
3.0SZX/01
A(0)
CZ.Z3
SEE
Mfe
;.fc7
0.00
-------
Table 2.60
LINEAR REGRESSION OF IDLE KEY MODE EMISSIONS WITH MILEAGE
PRE- Mi'iiotj cONiRoui r
FLEET 1
EMISSION COMT ROLLED
FLEET ^
^3 /f
NOX CONTROLLED
FLEET 3
.42
ro
i
CT(
o
7 IJLI ,1 lol/£.
Ad'
A (Oi
SEt
R7
d.f.
~3
,.0fc"f
.10
J.Oi
7 11
" mil
0.7ft
o.oooS
r.l
(.0
60
7.50-7
4Z.o
iS?
_A*6 J
_55
2.Z1
-/. 35*
217
-II.iz
.0126
I 4II
1.K,
Z12.
fs?
z.tf
.i?«2
A 5^
42Z.
.»?£
2to
?8
5.0C-/Z
g.oo
e.iool
-i.no
St. 8
,oo/o_
4o-L-\255
-------
Table 2.61
LINEAR REGRESSION OF TUNE PARAMETERS WITH MILEAGE
ro
i
~TMItJ6- )
AflMU.
_£*/>/» ,
/ye-""^
XI /A CtftolfK. ,
byit*-*.
Us f^J
Oj/-'" «Wl)j
y~
(TrfoAf A1** ,
,c~o>-/ U*.
A(l)
A(0)
SEE
~R2 ~
d.f.
t
Conf.
Afl)
A(0)
SEE
R'
d.f.
I
ConF.
A(l)
AfO)
SEE
R'
d.r.
t
Conf.
Ad)
AfO)
SEE
R2
d.f.
t
Conf.
A(l)
AfO)
SEE
#
d.f.
t
Conf.
PRE-EMISSION CONTROLLED
FLEET 1
A/e1
3.io7*io
0. (Ill)
Alfe^
D.OOI6
315
o.ez
__5f
'I.S'lOJIO
f.foty
it,. i
/>.0lt4
3fe?
5.l>
>'»1
l.oMxio1
7.1313
zt.z
0.03^
213
3-lt
>11
i.inwi
-D. to7&
0.551
o.0l?0
^15
Z.53
S8
-18Z8XI01
a. ooo 7
0.014
_o.pza
-J^3_
-.Z-.^T
11
A2
tUfeMO"
o.ollZ
z.feZ
0-OOoZ
zzi
I.M
^8
BWxio"3
3.5^o4_
(,7.t
o.ooni
ZH
1.26
8\
t-SoKlb1*
0.7^3
14.0
0.04IS
Ht
2.11
?11
-3 2^1X10
-O.oiil
0.433
0.0523
'It
j.ai
>11
-?319»|6"
-O.OOI6
0:018
O.OOll
Mo
o.^i
'HD
EMISSION CONTROLLED
FLEET 2
A/5'
1.IMW05
-O.OJ6L
?.11
o.oogt,
3bi
I.T7
13
-^zflxiu11
-O.tl01
^5. 4
O.OOP2
_3TL
o.z?
^10
\.tlof.x>
Z.OJtft
23.o
o.'5f5
332-
V.71
'11
-mzxib5
-o.«?3
O.Z.7I
0.0353
z>ft
Z.gL
>11
1.118X10
-o.ooU
o.o^o
o - oiot*
I0Z-
I.4D
ffs
NOX CONTROLLED
FLEET 3
/
*1/S
-J.lffl««>'
o. /353
2-fcl
o.oo«
100
1.51
«1
-ftlSmo
1.2381
53.8
0. ooji
HO
1.13
u
l.fffijio3
o.SSil
H.fe.
0.2«l
3 St.
I?-ZI
>11
a^tyxio1
0.033£-
O.SIt
O.o3i4
3t8
J.4S
>11
-l.rejxio"
-0
AZ1-
-1.05l»M?
«.1«74
1.57
fl,\
231
|.5t
86
-J.ZWri?
aoc»6
(J.ofg.
O.OOOZ
no
o.^o
-------
3.0 EXPERIMENT TO DETERMINE THE
EFFECT OF REPEATED TESTING
3.1 INTRODUCTION
the variation of HC and CO emissions as periodically measured in the
Emission Test Program using the 1972 Federal Procedure following operation
1n the field did not follow a smooth predictable function. The data clearly
indicated variabilities that were both random within each test period and
systematic over certain periods within the test program. Increases in
emissions after the first four month operating interval resulted in average
emission increases that were approximately as large as the decrease in
emissions that were measured when the vehicles were tested after initializa-
tion to manufacturers' specifications. The subsequent deterioration periods
over the remaining eight months of the year resulted in deteriorations that
were on the average negative during the second period and positive during
the third period. Although the general deterioration rate over the year
of operation appeared reasonable, the individual results over each operational
period appeared anomalous. The variabilities of measured emissions precluded
precise interpretations of experimental results. An experimental investiga-
tion was therefore performed to determine the test-to-test variability of
emission measurements.
Certainly, the variabilities were caused by a variety of influences.
The influences that are hypothesized to influence emissions are the
following:
o length of cold soak
0 temperature at which the cold soak is made
o test sequence (effect of repeated testing)
o preconditioning prior to cold soak
p previous driving history
o fuel type
o climate during previous driving
o post tune-up conditioning
3-1
-------
The study of influences was not within the scope of this investigation.
A cursory study was made however to determine the effects of test
sequence on emissions. Repeat emission tests were made using 1971 ve-
hicles used in the NO Controlled Vehicle Fleet.
/\
3.2 OBJECTIVE
The objective of the experiment was to characterize the test-to-test
variability of emission measurements using the 1972 Federal Procedure.
3.3 TEST PROCEDURE
3.3.1 Test Vehicle and Preparation
Nine vehicles were selected for the repeatability experi-
ments. All of the vehicles were the Scott leased loan cars which were
supplied to the owners of the vehicles used on the Deterioration Experiment.
Six of the vehicles were 1971 Ford Torinos equipped with 302 cubic inch
engines with two barrel carburetors and automatic transmissions. The other
three vehicles were 1971 Chevrolet Malibus equipped with 307 cubic inch
engines with two barrel carburetors and automatic transmissions. The ve-
hicles had been driven by many individuals and therefore represented usage
under all types of driving conditions.
The vehicles were not given any special maintenance or pre-
conditioning treatment prior to these tests. However, all of the vehicles
had received manufacturer's recommended, periodic maintenance and a complete
tune-up about four months previously. They were processed directly from
normal service to the test series. This procedure was equivalent to that
used during the Deterioration Experiment for the recall tests.
3.3.2 Test Sequence and Measurements
The vehicles were brought off normal service and initially
stored in the soak area for a minimum of twelve hours. The next day they
were installed on the chassis dynamometer, and Indolene fuel was connected
for a cold start emission test. The emission test sequence was similar to
3-2
-------
that used for the Deterioration Experiments. A 1975 Federal Test Procedure
exhaust emission test was conducted followed by the performance of the
Clayton Keyrnode Cycle emission tests. The vehicle was then returned to
the soak area and shut down. The vehicle would then be emission tested on
the following day, approximately twenty-two hours later, except over weekends,
The above sequence of daily testing was repeated until a total
of five to six tests had been made on each vehicle. The majority of the
vehicles were started on this test sequence so that the first weekend-long
soak occurred after the first three to five tests had been conducted using
the one day soak period.
3.4 ANALYSIS OF TEST DATA
3.4.1 Summary of Results
The analysis of the data indicated a systematic bias in the
cold or hot HC emission levels measured between the first and second test
when the 1972 Federal Procedure was used to measure emissions. Statistic-
ally significant differences of 0.68 gm/mi and 0.26 gm/mi, respectively,
for the cold and hot HC emissions at the 95 percent confidence level
resulted. Comparable results were not obtained with CO or NO emissions.
A
The test-to-test repeatability of emission measurement made
using the 1972 Federal Procedures and representative of data taken in the
Parameter Deterioration Experiment are as follows:
COLD 1972 PROCEDURE
HOT 1972 PROCEDURE
Estimate of
Standard
Emission
Specie
HC
CO
N(L
I
gm/mi
4.202
31.27
4.349
Deviation
gm/mi
0.362
3.79
0.174
%
8.6
12.1
4.0
d.f.
14
23
23
X
gm/mi
3.423
18.65
4.443
Estimate of
Standard
Deviation
gm/mi
0.217
1.49
0.150
%
6.3
8.0
3.4
d.f
14
23
23
3-3
-------
3.4.2 Discussion of Analysis
The set of data analyzed consisted of from five to seven 1972
Federal Hot and Cold Cycle emission response tests conducted on a fleet of
nine vehicles. The vehicles were selected to represent a homogeneous fleet
(i.e., same engine and drive train configuration for each vehicle). Each
emission test was conducted after a cold soak period of sufficient length
such that the vehicle was at or near 70°F. The data set, then, was assumed
to be representative of a single vehicle undergoing repeated tests at the
same starting conditions with all other factors (e.g., variations between
test cells and test crews) removed which could contribute to variations in
emissions. A summary of all the test data is presented in Table 3.1.
In order to verify this assumption, the mean and standard
deviation statistics for each vehicle and for each emission specie, i.e.,
hydrocarbon, carbon monoxide, and mixed oxides of nitrogen, were developed
for both the Cold and Hot Cycle 1972 Federal Emissien measurements. Review
of statistics presented in Table 3-1 indicated that for Vehicle 302, both the
mean level of hydrocarbon emissions and the scatter of emissions were grossly
larger in comparison to the rest of the fleet. On the basis of a Cochran's
test for homogeniety of variances, the precision of the hydrocarbon measure-
ments for hydrocarbon emissions were determined to be significantly different
(95 percent confidence level) and therefore the data from these tests were
removed from the overall data set and not used during any of the subsequent
analyses.
In order to verify that the remaining data set was homogeneous
(i.e., no significant differences between vehicles), an analysis of variance
was conducted for each emission specie on the data acquired with the Cold
1972 Federal Procedure. The results indicated that there were significant
differences (i.e., at the 95 percent confidence level) between vehicles for
all emission species. Since the scatter of the emission data from the tests
conducted on the eight remaining vehicles was approximately the same, the
between vehicle effect was removed by subtracting the mean value of each
test series from the raw values for that series. Further, it was observed
by plotting the emission data (mean values subtracted out) versus test
3-4
-------
Table 3.1
SUMMARY OF REPEATABILITY EXPERIMENT EMISSION RESPONSE
u>
1
Ul
Vehicle
No.
302
305
306
307
309
Test
Number
1
2
3
4
5
Mean
Std Dev
d.f.
1
2
3
4
5
6
7
Mean
Std Dev
d.f.
1
2
3
4
5
6
Mean
Std Dev
d.f.
1
2
3
4
5
Mean
Std Dev
d.f.
1
2
3
4
5
6
Mean
Std Dev
d.f.
1972
HC
gm/mile
9.07
4.86
10.73
9.75
5.16
7.91
2.7
4
4.66
5.26
4.95
4.55
4.44
4.39
4.89
4.73
0.31
6
4.29
4.56
4.22
4.26
4.48
5.11
4.50
0.32
5
5.30
4.19
3.90
4.61
4.10
4.42
0.56
4
4.49
4.12
3.82
3.64
4.20
4.11
4.06
0.30
5
Cold Cycle
CO
gm/mile
47.50
36.13
29.22
35.23
36.96
37.01
6.6
4
34.10
49.79
43.75
40.31
40.24
39.07
40.94
41.17
4.8
6
44.18
45.15
41.84
40.62
40.82
50.05
43.78
3.6
5
32.53
27.39
22.61
29.04
.24.22
27.16
3.9
4
28.30
23.32
23.31
23.47
27.90
30.42
26.12
3.1
5
NOx
gm/mile
4.83
4.96
4.69
5.26
4.91
4.93
0.21
4
4.84
5.01
4.90
4.71
4.53
4.81
4.88
4.81
0.15
6
4.16
4.57
3.98
4.28
4.13
4.28
4.23
0.20
5
3.96
4.07
4.14
4.69
4.14
4.20
0.20
4
4.22
4.04
4.09
3.72
4.14
4.37
4.10
0.22
5
1972
HC
gm/mi 1 e
8.01
6.19
11.63
11.16
5.39
8.59
3.3
4
4.31
4.19
4.34
4.02
4.46
4.09
3.70
4.16
0.25
6
4.35
4.35
4.21
3.88
4.47
4.01
4.21
0.23
5
3.69
3.42
3.15
3.33
3.38
3.39
0.20
4
3.69
3.34
3.35
3.27
4.54
3.21
3.57
0.50
5
Hot Cycle
CO
gin/mile
15.17
14.20
10.47
14.63
16.52
14.20
2.3
4
25.07
28.13
31.20
27.96
27.96
31.08
22.04
27.89
3.36
6
26.64
29.80
27.26
26.10
25.63
29.28
27.45
1.71
5
18.01
17.50
14.58
15.63
14.50
16.04
1.6
4
19.32
16.05
17.71
17.84
20.90
19.99
18.64
1 .8
5
NOx
gm/mile
5.02
5.02
5.10
5.13
5.16
5.09
.06
4
4.79
4.96
4.91
4.80
4.46
4.86
4.93
4.82
0.17
6
4.37
4.58
4.33
4.43
4.28
4.37
4.39
0.10
5
4.18
4.28
4.13
4.92
4.21
4.34
0.33
4
4.09
3.94
4.00
3.73
4.01
4.16
3.99
0.15
5
-------
Table 3.1 (Continued)
SUMMARY OF REPEATABILITY EXPERIMENT EMISSION RESPONSE
Vehicle
No.
310
313
317
320
Test
Number
1
2
3
4
5
6
Mean
Std Dev
d.f.
1
2
3
4
5
6
Mean
Std Dev
d.f.
1
2
3
4
5
6
Mean
Std Dev
d.f.
1
2
3
4
5
6
Mean
Std Dev
d.f.
1972
HC
gin/mi 1 e
3.47
3.13
3.33
3.27
3.40
3.51
3.35
0.14
5
4.26
3.44
3.46
3.40
3.23
3.60
3.56
0.36
5
5.54
4.08
4.13
3.98
4.09
4.41
4.37
0.59
5
4.76
3.70
3.37
4.27
3.78
3.62
3.92
0.51
5
Cold Cycle
CO
gm/mi le
27.09
24.57
27.19
25.42
28.42
31.99
27.45
2.6
5
22.51
21.75
19.79
23.51
20.81
23.34
21.95
1.5
5
35.22
27.09
33.42
32.25
28.86
30.76
31.27
3.0
5
44.84
33.34
26.85
30.58
31.23
27.24
32.35
6.6
5
NOx
gm/nri 1 e
4.14
3.93
4.21
3.69
4.55
4.40
4.15
0.31
5
4.62
4.40
4.42
4.39
4.39
4.85
4.51
0.19
5
4.98
4.75
4.24 .
4.49
4.57
4.85
4.65
0.27
5
4.52
4.08
4.41
4.76
4.30
4.75
4.47
0.26
5
1972
HC
gm/mi 1 e
3.21
2.92
3.61
3.21
3.74
3.75
3.41
0.34
5
3.80
2.94
3.10
3.04
3.97
3.26
3.35
0.43
5
3.61
3.33
3.31
2.77
3.31
3.40
3.29
0.28
5
2.83
2.71
2.67
3.40
2.62
2.62
2.81
0.30
5
Hot Cycle
CO
gm/mi 1 e
14.94
12.94
14.86
14.97
19.54
16.97
15.70
2.3
5
14.03
14.38
12.95
14.53
15.01
16.42
14.55
1.2
5
15.81
12.74
12.66
20.75
15.02
17.24
15.70
3.04
5
18.72
16.34
16.06
17.19
17.64
18.81
17.46
1.16
5
NOx
gm/mi 1 e
4.11
4.00
4.30
3.57
4.27
4.38
4.10
0.30
5.
4.59
4.44
4.55
4.43
4.25
4.72
4.50
0.16
5
5.18
4.91
4.40
4.67
4.69
4.89
4.79
0.26
5
4.78
4:39
4.57
4.96
4.78
4.82
4.71
0.20
5
3-6
-------
number that an increase in the scatter of the data occurred after the
third test. More importantly-, however, abrupt shifts in emission levels
for specific vehicles occurred following the third test in most of the
cases. A typical example for this phenomena is shown in Figure 3.1 for
HC emissions. Comparable figures presenting data typical of CO and NOV
y\
emissions are presented, respectively, in Figures 3.2 and 3.3. In
particular, the reader should note the change in emission level for
Vehicles 307 and 320 between the third and fourth test. A review of the
conduct of the experiment, i.e., date of test and crew used, did not
reveal any reasonable systematic cause for either the increased scatter
or the emission level shifts. Without attempting to further explore the
reason for this observed characteristic, it was decided to remove the data
after the third test from the data set and recalculate the differences from
the average, based on the first three tests.
This final data set, including both cold and hot cycle data,
was'then subjected to an analysis of variance. The results of this analysis,
shown in Table 3.4, indicated that for HC emissions a statistically signifi-
cant difference between tests was present. Further comparisons by means of
the Duncan Multiple Range test indicated that the first tests for both 1972
Federal Procedures (Cold and Hot) were significantly different (95 percent
confidence level) from the second and third tests. The second and third
tests, however, were not found to be significantly different. The net
effects (differences between the first test and the average of the second
and third test) are, respectively, 0.68 gm/mi and 0.26 gm/mi for the cold
and hot cycles. The significance levels for between test effect of cold
CO emissions was 75 percent and was less than 50 percent for hot emissions.
The significance levels of repeat test effects of hot and cold N0x emissions
were approximately 50 percent.
The natural product of repeat emission testing was the develop-
ment of run-to-run variability of vehicle emission measurements. The results
summarized in Table 3.1 are representative of the random uncertainty of
emission measurements caused by pure vehicle non-repeatability (excluding
fuel, climate, preconditioning, etc.) and measurement system uncertainty.
3-7
-------
Table 3.2
REPEATABILITY TEST ANALYSIS OF VARIANCE SUMMARY
EMISSION
TEST CYCLE SPECIE
1972 COLD HC
CO
NOX
1972 HOT HC
CO
NOX
SOURCE OF DEGREES OF
VARIATION FREEDOM
Between Vehicles
Between Tests
Between Vehicles
Between Tests
Between Vehicles
Between Tests
Between Vehicles
Between Tests
Between Vehicles
Between Tests
Between Vehicles
Between Tests
7
2
7
2
7
2
7
2
7
2
7
2
TEST
STATISTIC*
0
9.70
0
1.63
0
1.12
0
3.92
0
0.78
0
1.01
SIGNIFICANCE
LEVEL
%
0
99.5
0
75+
0
50+
0
95
0
<50
0
50+
CO
CO
*Ratio of Source of Variation Mean Square to Residual Mean Square. Significance determined by
comparison of Test Statistic to Standard "F" distribution value. Degrees of Freedom for
Residual Mean Square is equal to 14 for all cases.
-------
-------
i
c
ra
a.
z:
4-
11
4J
C
CO
z
c
IS
IS.
5
CJ
..-:|.-:
20
;1 8 "
. . . i : ' .
'1J4 '
; 1
12
i
" 1
i'o
i
8
5
4
i
2
"1
0
- 1
1
i
;j-:7
-10-
' '|--: C
, i- ,
: I'.
i' '
:::;
t " "
-rrt
- ',
' ' < t
:-; :
:-
-
: :!
]
j Tj
i !:
-
t.
I." -
'(,
y '
r .
-H
:: I1
-;4
.:<
;; s
j . . ,
-H
.
t "
. ;
..M..^.
: : : :
'
::.!;::
SEQUE
197
-
N
- -
I
^
M
--:H
:::f
r-- - ^
1 'i " -.
NCE ON
|V.T;
2 COLD
;'
- . .
L
t--
t
L:
{::
f
5-^
:::
!L
.:.
:::
N
H-
M
' i . .
si i
32T::!:4::H
4^^-lrti]
CARBO
:--;:
CYCLE
. . j
. , .
- . i
:::
^UMl
-,:^
-..,
^tf
1
rrrr1
h
L
5FD
5tK
N MONO
.::-,.:-
:iri
,::
'FT
....
r i
i n r
'.in
UU
.::
. . .-
....
XID
i-"
E E
Ttr
MIS
^VEHICLE
O302
A 305
D306
rrr rrr- TTTT
.K:
rrr
::::
rrr
_^)
-!::
;-a
r|lt
^
'*'!'
...
;:.: ...
, L. ,.
:;.:»
t-r
^J
T-f
) ' 1
SIO
,.:,
r-i
;u-i
NS;
:.ir
^
:;:-.
ri:
t r
IDENTIFICATION
§307 0313
309 O317
310 X320
[^ ri: . f: : i i , . j : i i-j
TTT ~T*T ' PTT t':TttTTJT
\-
"'] 'T
tt.n
M
in:
!::!
JU,
'if-
; --[j-
f-i
y
b
mr
* . .
IDj
-H-
' ' 1 ;
TT I r
"rrjl
-{^
Pi
J<
rtrt
4-
^if
:;;:
i;:
l '
H
n
- ^ -
....
-i,:;
i'r
1 1-
li::
I IV
4-4
~ ' *
:.::
Lf:
!f:r
,-:.
»
P-
ftr
rtif
trri-
i::r
rrr^
iiij:
....
:;r
rrr'j-
1 r
-Mil
mv
S'j
i
TTt]-
- - -
3-10
-------
g=
en
(0
H
£
4-
10
O
oo
o
t t
oo
oo
LU
X
O
... L -
" '7
.. r
' .6 -J
::;:1
R :
.4
I
- i
3
<± -
1
: i:-
0
1
.t
1
i
1-
2-
- .3-
- 4
-i
B -
'
--.6-
. 1
-.7-
i
:. : 1
TrTl
- 1
i . '
i , :.
:.L
1
_
' ; j;:-:
-4~
^
...
- i--
i
i
f -
-
-i:i-
- 1
"".:
'-' : '
P
;«
:!/
: ;
' 1 -
-
T
- .
:_;_
.
T
-1-
. _
- i
,. .
:,_ .:_
-'-1-7
_.
-4- 1
""I
".":
:ii; -:
:::_] ; ;
-~
. ,i :
: '
)^~
-
,
> : .'..
^ ::
D...
Jrf-
'
*
1_1^.
. . .
.
:;
1 " -
1 :
: :
-
. ' ' :
':''.'-
:~-
,..,
iEFFEC
i: '
--
--
1 . .
' : ^
:::<
:;3
! i
..I . .
......
~rr. "
--'}"
T C
FT
. ''I'.::
.-:::::.-:
-t
.:
-^
i : :
>.::
j::i
.
L
j
/7J- -
T
,
J
. t
": Figure
EST'SE
1972 (
-r:
:--
---
" '. 7-
- -
:::i
:';i
3/3
-) -::
QUE
:OLI
,:.:
i.-.
*-
-
.J. ..
:
-.;!..
r.
"1
..
:.,
-:i-
NCE ON NO
;;;!£].;
) CYCLES'
:::±
::i
'
' ' : : '
! ::
I
: -U
):...
v-
rr
:::-
. . i .
. . . ^
Hi:
* r
^
i
...j
::-:
-rr|
N
-«
: ;:i j.
. . . «
: 1: :S
:: :
; :
. ! .
:J"i
PTEST NUMBER
! ,
TTT
--
'"
i
x EMISSIONS!
;;>.
":f:
;i;:
i77"
V1-
5,:::
tTTTJ
Jrm
f;-
t'
-rr-
"-
':
;-
: :
:!:
i : '.-
':''<:
i ; i
.'":
^T^
fT
% :
i
TTT-I
rrrr
T"]
:hi
....
"r
j.:.
t .,
HT
f-i-
Lli
; H-
i il
'! :-
"r
--
-iL
-HI
-'a
>i
^
<
!-f
:Jr.
1 1
:l ;;
14::
_rj|'i
1 j_-
:'!',
^7
; ,:
i-rr
:VE
O3
A3
D3
^
L. - -
rrr
M
;-i
"i:
HIC
02
05
06
-
~
N
-:M
_: I:.-?1
-.:q:::^
Jrr
v-
IT:
r-
B;..
1 ' i
. ..,
;iji
i:
:j|
iri'
-."-
'
.,,-
L,
....
': iit
TTTt
T-Tt
4 i
; t
. ;
[ ;
4:.'
fHi
' r' I
"£-.
1 ' ' i
';-
I"::":
-ft ^>-
\l t-
ri.*:
^ 1-
:; r
r^
.E IDEC
<
'-r-
....
Ttir
rr:J
L
:i;i
"i
.'s;r
T
rnr
H t|t
j^f
HTT
.4 t i
m.
HH:
u.
; , ^l
03
0>3
73
:!<
- ,-4
;u
trrt1
ptij
:::;
'"" ;
~|T7
-i-ll
:hr
irf
uir
:?:.
TJTr
rhtt
4;:^
1 'i i
; 1 1 -
:r.:
t;
-
JTIFIC/
07
09
10
r;-. :
-;;;
f
1U
". I , '.-
:}
" :
^;:r
.:V:
1 "
....
H-
jjji-
! i .
:i -
::l
', i~i
i: "
1 !
^TIC
03
03
X?
. + , -
;:E
Ui
~^T
hf
; M
-^
i :i
r^l
;r,:-l
rrpr
ir";-
::f.
T-:-;
)Nj|
13i
17f
20[:
"Hi
.._.
.....
~v
i ' 1 "
.1
i,;'.:
: :r;
»::t
-Uii
rr:.
lilL
;.;t--j
:;lr
i;-i
3-11
-------
4.0 EFFECT OF COLD SOAK
TEMPERATURE ON EXHAUST EMISSIONS
4.1 INTRODUCTION
A series of experimental tests was performed in which eight vehicles
were emission tested after being soaked at different ambient temperatures.
The soak temperature is the ambient temperature at which a motor vehicle
is stored prior to conducting a Federal exhaust emission test on that
vehicle. The soak temperature was varied over a range of 60°F. Other test
variables were held as constant as possible. The 1972 and 1975 Federal
Test Procedures were employed to determine the CVS mass emission response.
4.2 OBJECTIVE
The objective of these experiments was to develop the response of
automobile exhaust emissions to variations in soak temperature. This
characterization was made for three classes of vehicles; 1) vehicles
with no exhaust emission controls, 2) vehicles with controls of HC and
CO, and 3) 1971 California vehicles having controls of HC, CO, and NO .
A
These three classes can be related directly to the three fleets that were
tested in the Deterioration Experiment. The emission responses developed
for this experiment could then be used to correct data from the Deteriora-
tion Experiment where soak temperatures were not controlled within specified
1 i mi ts.
4.3 TEST PROCEDURE
4.3.1 Test Vehicles
Eight vehicles were selected to approximately represent
the three classes of emission controls. Two and three vehicles per class
do not provide a firm base for representing populations. However, an
understanding of the possible lack of correlation to a population could be
assessed by observing the variabilities between vehicles in the same class.
Only popular makes of vehicles with the most commonly employed driveline
4-1
-------
options were used. This selection provided for the best possible corre-
lation to the populations being evaluated. Table 4.1 presents the
description of the eight vehicles.
TABLE 4.1
TEST VEHICLE DESCRIPTION
Repairs
Installed correct model
year carburetor
Replaced flooding car-
buretor
None
None
Replaced ignition wires
and vacuum advance
Connect and adjust choke
1inkage
None
Installed new distributor
Fleet
Represented
I
I
I
II
11
II
III
III
Year
1964
1964
1965
1967
1968
1969
1971
1971
Make/Model
Ford/Gal axie
Chevrolet/Impala
Dodge/Dart
Ford/Mustang
Pontiac/Tempest
Chevrolet/Impala
Ford/LTD
Chevrolet/Camaro
Engine/
Carburetor
289/2 bbl.
327/4 bbl.
273/2 bbl.
289/2 bbl.
350/2 bbl.
427/4 bbl.
351/2 bbl.
350/2 bbl.
4.3.2 Vehicle Preparation
Vehicle preparation was limited to making repairs on com-
ponents that would effect the stability of emissions or grossly effect
emission levels relative to their nominal level. No other tune-up repairs
or settings were made as it was desired that these vehicles approximate
the in-use condition of vehicles. Table 4.1 above describes those repairs
that were made.
4.3.3 Vehicle Soak Temperature and Sequence
Soak temperature experiments on two vehicles were first
completed covering a temperature range of 40 to 90°F in ten degree incre-
ments. Based on the data obtained from these two vehicles the soak tempera-
tures were set at 50, 65, 85, and 100°F for the remaining six vehicles.
Apparent outlying test results occasionally occurred at these temperatures.
In these cases the tests were repeated. If no repeat tests were required
4-2
-------
an additional test was run at 75°F. The soak temperatures for each
vehicle were randomized (not statistically) to preclude any bias due
to prior soak temperature from affecting the data. The vehicle soak
temperature was maintained by storing the test vehicle in Scott's All-
Weather Room facility. Heating and air conditioning equipment and
controls in this facility allowed for maintaining the desired soak
temperature within - 2° during the soak period.
In order to eliminate any influence due to the length of
time of the soak period, the time of each test was regulated. After the
first two vehicles mentioned above were tested, a twenty-four hour test
cycle was instituted. The vehicle was brought out of soak between 9:00
and 10:00 p.m. to receive its emission test. The emission test took
approximately one hour to perform and the vehicle was returned to the
soak facility. The vehicle was allowed to stand overnight at nominal
temperature as the air conditioning equipment could not'be operated
overnight. The temperature controls were set and the air conditioning
equipment was turned on at 8:00 a.m. the next morning. By no later than
10:00 a.m. the desired temperature was reached. Thus, the vehicles were
soaked for at least twenty-two hours, the last eleven being at the desired
soak temperature. This precise routine was violated only on weekends when
the vehicle was soaked for three days.
4.3.4 Exhaust Emission Test
After completing the soak described above, the vehicle was
installed on a chassis dynamometer at a test station that was immediately
adjacent to the All-Weather Room facility. A 1975 Federal Test Procedure
exhaust emission test was then conducted from a cold engine start. As
three sample bags are collected on this test, the equivalent 1972 Federal
test results could be obtained from the first two sample bags. The 1972
test was of primary concern in this experiment. The emission tests were
conducted within the nominal ambient temperatures of 68 - 86°F.
4-3
-------
The vehicle tail pipe pressures, mixing chamber pressure,
and CO analyzer were compliant with the 1972 Federal Test. Procedure but
were not up-to-date with respect to the 1975 Procedure. The instrumenta-
tion employed was also compliant with Federal test regulations. Both
Chemiluminescence and NDIR/NDLN NO instrumentation was used to determine
X
the NO emissions. The three bag data was converted to mass emission
A
results using the 1975 procedure. A dilution factor is used for correcting
the background level to obtain the net vehicle emissions.
4.4 ANALYSIS OF TEST DATA
4.4.1 Summary of Results
The analysis of data taken in the Cold Soak Temperature
Experiment indicated significant dependency (greater than 99 percent) of
CO emissions with soak temperature with vehicles representative of all
three vehicle fleets tested in the Parameter Deterioration Experiment. A
comparable highly significant dependency (98 percent) of HC emissions with
Emission Controlled Vehicles (Fleet 2) and NOV Controlled Vehicles (Fleet 3)
/\
was noted. Dependency of NO emissions was observed only with Emission
J\
Controlled Vehicles.
4.4.2 Discussion of Analyses
The set of data which was obtained from the Soak Temperature
Experiments consisted of from five to seven emission tests using the Cold
1972 Federal Procedure with eight vehicles which had been cold soaked at
temperatures ranging from 50 to 100°F. The overall fleet in this case was
selected to be representative of the general population and was partitioned
into three subsets (Pre-Emission Controlled, Emission Controlled, and NO
J\
Controlled Vehicles).
It was assumed for the purpose of analysis of these data
that all testing associated variables which could affect the emission
response other than pre-emission test soak temperature, were closely
controlled and could be ignored. It was also assumed that the three sub-
sets were significantly different and should be analyzed separately.
4-4
-------
Based on the analysis of the data obtained from the
Repeatability Experiment, two operations were performed on the Soak
Temperature Experiment data set. The first was to remove from the
data set the HC emissions from the first test conducted on each
vehicle. This operation was conducted on the basis that the first
test had been shown to be significantly different for HC emissions
during the Repeatability Experiment. The second operation was to
subtract the average emission response of the remaining tests for
all emission species and within each vehicle test series from the raw
test values. This was done to remove the between vehicle effect from
each subset. The resulting delta emission values for each emission
specie and for each subset of vehicles are shown in Figures 4.1
through 4.9.
A set of regressions was then conducted on the adjusted
data set to determine the effect of soak temperature on emission re-
sponse. A cursory review of the delta emission levels following the
regression analysis indicated that, for the pre-emission controlled
set, the significance of the HC response was largely dependent upon
two extreme values from one vehicle. This particular case was then
re-analyzed without these two data points.
The final set of regression results, shown in Table 4-2,
are summarized by the following.
1) A highly significant dependency (greater
than 99 percent) of CO emission response
with soak temperature across all three
fl eets.
2) A significant dependency (98 percent) of
HC emission response with soak temperature
for the emission controlled fleet and a
highly significant dependency for the NOX
controlled fleet.
4-5
-------
-------
±rr
U:r
#t
rm:-r!-r t7
Figure 4.2! ij
m
rtrr
^2-
if-.l- -
1972
FEDERAL COLD CO VS WATER TEMPERATURE
PRE-EMISSION CONTROLLED VEHICLES
rMU
IT:
-------
-f*
00
i....
i
... a
....
OJ
r-
E
" 1
x >
LJ_
_<
=»
LU
CD
"LU
>
i "^
g
p>
t
1
i
L
i
t
1 ' '
i
LU i
Q :
x:
_O 1 1
z
_LU
O
LU
U.
CVJ
Q
1
O
CJ
,
,-
1
.;.:;.
3.0-
- t-- -
i
: . : : r '
]" :
z'.Q--
\
1.0-
- 1
W
f
-i n
' '
; :
-..-.!.-_,
i
-2.0-
. ,, . ^ .
-slo-
.'l' 3
;;-i|. .
::-= U;.
.. , 1 : .
... j. .
::; r:
ii
' i
i
!'':;{: :i
....
... j. .
-
,y
:0
i
:-- JO
I . *
OL ,.:;:
.,] -
^ I *
:.'
j-,.*. ;-
44-
^
I
i
j
_
-*~».^.
::::lii';
: ] :
i
*
' ' :
;;
: :
,::::
1
tr
. . .
1 . '
." I i
r~-
: . .
: ; : ! ' '
64 FORD
65 DODGE
64 CHEV
i
'.\:
~
,.v
1
l
i
i
'
\
i
QJ
: , '
t , * - -
- 1 - * *
. '. . . i .
,.
^J'-
.': i;: -
-.: ;
r~
::
,
: :
::
-i--:
i:,.::;-
i : '
...
i. . "
-»-j"- ^
, l '. \
\ , ,
. . i
' ' i
;::.
. . .
r *
-, - t
!F
* 1
...
;; '
tr:14^
. . .
::;:
f_--_
^M !
ft'
t
; . .
.
.
.
,
* i
i
t
o:.
1
--
1972 F
;;
1 '
PRE
. . '
rH
r J
....
r
: : '
1 :l:,M
,F
igure
4.3\
EDERAL COLD NOX VS
;'" 1
-EMISS
" ' . i . .
,,,. ,,;,.
!
j .
i
r
J . .
\ -
1
. -\:
'"!- ..
- i .
i
'.'.
160
~ - T ~ ;
ION CO
. '. '. i ' ; .
: -!- .
-4,-
yji 1 i \f
' .! '
* ',
'. I.
NTROLL
i
I-
n
:: i;!
. i
WATER
.. .
ED VEH
. -----
ip
r : :'
?if- :.::
: :. :
. 1 . .
.'!.;
...
:-::
::. ; !
;;1:-:;:
TEMPERATURE !:.
.
ICL
~
ES
, -'
-
"
^?1"
' . '
EMISSION = -0.0041
, ..
i . -
. . 1: '
. 1 . . .
:'|"-
:'.: ;::.
'.'
....;.. .
; ' .-
: : . ' - :
: :<;i~70:
i
i
, . i
: -j.i
i -8
1
~
-
_
'
'
i '
! 1
;
. .~j...
t . . . .
r.
: . . i : : .
I ; .
- ' L'
ffrr
;
....
r
L ' .-
TTT"-
"=!
. -
t
r
(
- 1 t
. "
\
^--i
"Tfr
i ^
1 X TEMPERATURE -
, -]
i ' (
1
.
I
;::
'. '. '
Q u .
1
i COLD SOAK TEMPERATURE (WATER), °F ::
i .
I :
i r i : 1 -. '
,'i
.'
9
: !'
-
' . , '.
.
, .~|, .
F
i
;
o.:
*
. 1
l
i-
-
: ' : :
'.' ] :"
i
)
0.3118
! .
: i . : . .
::;-| -.:
'"i':-
\
t "
.
'.:.'
::: . ..
* ' ' . J
:..'!:::
jy;:
;;';ii.!
100
;
i:' '
-
i
.:
v:
til"
. :
ii i
....
. . r -
r '. '.
.~,^~.
. ;:i...V
F
.-
.........
;
:
"* t
- 1
';;-
. 1 1 ,
-
i
]
. i * . r ,
:;;: :.::
. i .
L
\\^:'
'::.::. ':":.
'^':
Q' :'.':
:;;i
::
i
. *. .
;!;
!'
ill
H|
I,
';
liil
' ".
)';
-
'.r::
;':':
iiji
III!
< , . ..
:,!:
\ . .
::' ':
: . . i ;
' '...
rrf
.
::..
'.:; :;.:
. ' : : '
" i 1 r "
....
!<::
....
-------
j|Figure 4.4
1972 FEDERAL COLD HC VS WATER TEMPERATURE-H?'
^EMISSION CONTROLLED VEHICLES:;
r||j EMISSION = -0.06300 X TEMPERATURE +4.429
COLD SOAK TEMPERATURE (WATER)
-------
m
Ol
tr1 E
iri °f
-li'sol
TTTttTT
-J40f
r. rr;
«30-
-20
; 'i::
-T^SffHlOr
II . - | . .
I
__J
O
Q.-
0:
uur|.;;-r
i^4i-2C
CVJ .
r-.-
O :
-O
m
il
m
a;ii
TTtt
IiU
ffi
is?
:".]:A
r:;i:;;ii:Ht
mt
TTE
«
.
mi
I'.i
trt.tt
tt
-
tt;m
tit:
Ut:
it:
m
f
HM
69 CHEV
68 PONT
67 MUSTANG
t-U
i - --»
:t:r
1 T
ttt
ttt
TT
ii 1972
Ui.
till
ThT
hll:
H-
iiir
Figure 4.5 :j:::':
h:|::;:i -I r.t
rrt-^--;
S
FEDERAL COLD CO VS WATER TEMPERATURE i
;:.;.i..!L!!iri
i EMISSION CONTROLLED VEHICLES I ii;
11:!
.
.t:::
EMISSION
-l
ij;
t
1
III i
t
rl
F
t*F
MM'
= -1.074 X TEMPERATURE +74.16 i
ttjj
;m
!!it
rU.
TF
f ' !1
W
ii
tt~-i
+.1-1 * *
rrr
1
-Ut
r 1
*l
lit
stl
JHr
^E
«
Hi
tr.t
^Lp.
ITTT
t;
H
^
m
m
#3
1
m
m
^4-
«»!
r-ir
ill
I
-50
itr;
^T
i-50,:
ij:: .
70 :';
4f
nt
*£
oo
rrn
ii
1rHt
IJfl
^LL±L
«tt
OltE
COLD SOAK TEMPERATURE
iiU'lIUtLL
1 '
til
T'l
iLL-lLuaL:l-Luu.j_iiri: I'
iilili^i^ij-^-TH7
liMMtrr- i- ip. i
-------
FFigure 4.6
T3;jj.JiiJi..: 11. :r.
uittt'.l::::.; :
+J.I!,, .,...,,,,
WATER TEMPERATURE
MISSION CONTRO
VEHICLES }i
E'-'tt.y&g
TEMPERATURE
69 CHEV
68 PONT
67 MUSTANG i
(WATER), °F 1
COLD SOAK TEMPERATURE
-------
-------
gure 4.8^;;
1972 FEDERAL COLD tb VS "WATER TEMPERATURE
NOV CONTROLLED .VEHICLES
/\
14-r4 1 | - 4 i " *
44- .Ml -r-M
71 CHEV
71 FORD
EMISSION
TEMPERATURE
SOAK TEMPERA
-------
I
*.
.:_..:.
4 OJ
1
r
; : ***-.
i :-.f !:.k.
AVERAGE VALUE
rf "i - ; ?
O
o:
1 KM
LU '
. :o :,
; x
::H
?
-2
~°1
*
n^
; : ' '
-i
io j
t- -
.....
'CM _| 1
;Q
1
:o
. .' ~ ....
r .
.... :::,
-c
. '
r
0-
-
r
-3
, . , .
*«,.
n
L.3
-..,.
\i-i\- :
.;
::
*.«. » *.
.
;
r " *
: : . :
: : . : T:
: : < O
o.j.
:-:)::::|
';
- it i
1^ r ,
^ . ',
i \ '. :
1 ,. '
rrft
:: ::
...
|:;;:
i
- 5
;..
.1
i
<
:,.
. i .
-
:. . . i :
71 CHEV"!
71 FOPn
i
i
,' "
,. .
1 - ;
^ j
;:i:t;::i
._._
i
. .
I . '.
'':
:::'
...
..,._.
\ '. .
" :;
l:.::.'
_
-
.
.'
. .
:-
; .T '
-'
; ;
:
197
-- -
:
:: ~
y
1 . _
t
!
' . !
f-T\:
. ;;.-
:' 50- i- ;
.r~|
*'. ' i ^ ' .
- - - - T
! _
2 FEDE
' ' , . i , ; ;
: !:-:'
. 1 1 j . , .
i"
f *-
j i
i
.1
l ' ' '
I:;-.
i - .
- r -
RAL
NO
CO
KC
... ,
.,_.
1
. .
j: ::(;:
Fiq
LD
f «_
DNT
^
....
.--*.
ure
X
P~*«-,
ROL
-
Kjp
_>-
4.
vs
!
LED
9
WA'
VE
L
'.. ~ »
EMISSION =
i ; ' .
~r~
::j.;.
i so ;
COLD S
-~~
. . ,
. : i .
.^.
;;
....
: "
,_, ...
FER TD
.... .
HICLES
- i
t
i
. . i
1
;
;-.:
i . -
1PERAT!
.....
-
~
.. ..i
Ud
0.0007905 X
-
;- ;. ..
; 70; ;
OAK TEMPERATURE (V
; "i
H-rt: -H
'''''i
'.'. " ^'
'..: ' "
JRE '
- ; : '.
; - i
-~f:
i .
--H-~7;
(
f
1
L -1
".
> i
L.::;
*^ !
:i::
t
i
: .:
1
'
N
..t,-.. .» i..
; ;
F
.
;
.
_,
-x
TEMPERATURE - 0 . 05903 i ': j : ; ! i
;;; !:;i
* i
r '-,
--' -
: 80; i
JATER), °F
^::{;?|;-^ I
. " !
..;.. .:_
1__
'<
-
-
r;..>
--r1
. .}
1
./ 90 ;
i .
!.:| i::
'
.'-
._^
-\ .:.
N
L^.
:rr-
. ' . '.
. : : . i
; 100:; .
i
i
!
: . - 4 - -
* *
:;
i!
-
_»,_>«-
1
:i::
.. .
' ' '.'.4 .'. ,
.:..
r
i
r
,;..:
i . -
* , ', .
,, . ,
i::
ill
h:
:!:
; : 1:1
; i i I
';
_.
17! ! '.
Inoil!!'.
:\l\\:':\l
'. ': \".
;.:;:
; ^;
. ...
:i:
;::;
rrr-
; : '1
L . . .
1 . .
' ' 1
:::'
i ;
^ ! I
-.:::
~. t .' .
.;:::
;::':
-fill
. i .
i-i ::
'T.'-
__.
, 1 . . «
: :l .:
."
....
.::.
-------
3) A significant dependency (95 percent)
of NO emission response for the emission
X
controlled fleet.
The response coefficients of emissions with cold soak temperature of the
three vehicle fleets are presented in Table 4-3.
4-15
-------
Table 4.2
STATISTICAL SUMMARY OF THE EFFECT OF SOAK TEMPERATURE ON 1972 FEDERAL EMISSIONS
VEHICLE
FLEET
PRE-EMISSION
CONTROLLED
VEHICLES
EMISSION
CONTROLLED
VEHICLES
NOV
X
CONTROLLED
VEHICLES
PARAMETER
HC
CO
NOV
X
HC
CO
NO
X
HC
CO
NOX
STANDARD
ERROR
gm/mi 1 e
0.56
8.50
0.48
1.10
11.10
0.38
0.34
9.60
0.52
DEGREES OF
FREEDOM
14
20
20
9
18
16
8
10
10
INDEX OF
DETERMI-
NATION
0.029
0.380
0.019
0.475
0.725
0.188
0.783
0.688
0.001
t
0.619
3.410
0.607
2.690
6.690
1.860
5.020
4.450
0.095
SIGNIFICANCE
70.5
99.5
70.0
98.0
99.9
95.0
99.9
99.9
<50.0
-------
Table 4.3
COMPARISON OF RESPONSE COEFFICIENTS
VEHICLE
FLEET
PRE-EMISSION
CONTROLLED
VEHICLES
EMISSION
CONTROLLED
VEHICLES
NOX
CONTROLLED
VEHICLES
PARAMETER
HC
CO
NOX
HC
CO
NOX
HC
CO
NO,
RESPONSE COEFFICIENT
g/mile/degree
-.00767
-.4140 *
0.0041
-.06300*
- 1.074*
0.01072
-.03590*
-.7160
0.0008
*Statist1cally significant with confidence
graater than 95 percent.
4-17
-------
REFERENCES
1. Volume IV, "Experimental Characterization of Vehicle Emissions
and Maintenance States," Year End Report, July 1972.
2. 1970 Almanac, Automotive News, Slocum Publishing Company,
Detroit, Michigan, April 27, 1970.
3. Handbook for Installation and Inspection Stations, California
Highway Patrol, August 1969.
4. A Realistic Vehicle Emission Inspection System, E. L. Cline
and Lee Tinkhan, Clayton Mfg.(a, El Monte, California, June
1968, (APCA - Paper No. 68-152).
5. Sun Service Control System, Sun Electric Corporation, Chicago,
Illinois.
6. 1961-1971 Rational Service Data, National Automotive Service,
Inc., Division of Glenn Mitchell Manuals, Inc., San Diego,
Cal iform'a.
7. Operating Manual, Rotunda Model JJRE-21-1 Exhaust Emissions
Analyzer.
8. 1972 Procedure Federal Register, Volume 35, No. 219 - Tuesday,
November 10, 1970 - Part II and Volume 36 - No. 128, Friday,
July 2, 1971.
------- |