EPA-650/4-75-024-a
                        Environmental Monitoring Series
                                  GUIDELINES
     FOR  QUALITY ASSURANCE  PROGRAMS
           FOR MOBILE SOURCE EMISSIONS
                   MEASUREMENT  SYSTEMS:
          PHASE I, LIGHT-DUTY GASOLINE-POWERED VEHICLES -
                        QUALITY ASSURANCE GUIDELINES
                                         \
                                          LU
                               U.S. Environmental Protection Agency
                               Office of Research and Development
                                    Washington, D.C. 20460

-------
                                     EPA-650/4-75-024-a
                 GUIDELINES


FOR QUALITY  ASSURANCE  PROGRAMS


   FOR  MOBILE  SOURCE EMISSIONS


         MEASUREMENT SYSTEMS:


  PHASE  I,  LIGHT-DUTY GASOLINE-POWERED VEHICLES -

           QUALITY  ASSURANCE GUIDELINES

                        by

         Harold Wiraette, Rod Pilkington, and Tom Kelly

                 Olson Laboratories. Inc.
                 421 East Corritos Avenue
                Anaheim, California 92805

                 Contract No. 68-02-1740
                   ROAP No. 26BGC
                Program Element No. 1HA327


                  EPA Project Officers:

                    R. C. Rhodes
     Quality Assurance and Environmental Monitoring Laboratory
          Research Triangle Park, North Carolina 27711
                        and
                    C. Don Paulsell
               Office of Program Management
                Ann Arbor, Michigan  48105

                     Prepared for

          U.S. ENVIRONMENTAL PROTECTION AGENCY
           Office of Mobile Source Air Pollution Control
                        and
              Office of Research and Development
                Washington, D. C. 20460

                      June 1975

-------
                        EPA REVIEW NOTICE

This volume has been prepared by Olson Laboratories,  Incorporated
consistent with the Environmental Protection Agency Quality Assurance
principles and concepts and with the Environmental Protection Agency Mobile
Source Testing Practices at Ann Arbor, Michigan.

The guidelines and procedures are generally applicable to mobile
source testing operations and are intended for use by those engaged in such
measurement programs

It is requested that recipients and users of this document submit any
comments and suggestions  to the Project Officers.

Mention of trade names or commercial products does not constitute
Environmental Protection Agency endorsement or recommendation for  use.
                    RESEARCH REPORTING SERIES

 Research reports of the Office of Research and Development, U.S. Environ-
 mental Protection Agency, have been grouped into series.  These broad
 categories were established to facilitate further development and applica-
 tion of environmental technology.  Elimination of traditional grouping was
 consciously planned to foster technology transfer and maximum interface
 in related fields.  These series are:

           1.  ENVIRONMENTAL HEALTH EFFECTS RESEARCH

           2.  ENVIRONMENTAL PROTECTION TECHNOLOGY

           3.  ECOLOGICAL RESEARCH

           4.  ENVIRONMENTAL MONITORING
           5.  SOCIOECONOMIC ENVIRONMENTAL STUDIES

           6.  SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS
           9.  MISCELLANEOUS

 This report has been assigned  to the ENVIRONMENTAL MONITORING
 series. This series describes  research conducted to develop new or
 improved methods and instrumentation for the identification and quanti-
 fication of environmental pollutants at the lowest conceivably significant
 concentrations. It also includes studies to determine the ambient concen-
 trations of pollutants in the environment and/or the variance of pollutants
 as a (unction of time or meteorological factors.
  This document is available to the public for sale through the National
  Technical Information Service, Springfield, Virginia 22161.

                 Publication No. EPA-650/4-75-024-a
                                  11

-------
                             FOREWORD
          All mobile source testing facilities have some elements  (acti-
vities)  of a quality assurance system built into their routine testing
operations.  These activities may not have been identified and/or  inte-
grated into a formal quality assurance program.  It is the objective of
these guidelines to provide guidance to both  (1) facilities which  desire
to organize an integrated quality assurance program, and (2) facilities
which may have already organized towards an integrated quality assurance
program, but may desire to review their program as a result of the
recommendations and suggestions included in these guidelines.  The
extent of implementation of these guidelines will depend upon the  re-
quirements of each individual test facility.
                                111

-------
                          EXECUTIVE SUMMARY
          Quality Assurance guidelines for light duty mobile source
emission measurements are presented in this document.  The guidelines
are modeled after the concept of "total quality assurance" developed in
recent years to meet the quality requirements of industrial programs.
Many of the quality concepts presented in this document are presently
utilized as part of the overall management program of numerous organiza-
tions.
          In order to evaluate this concept in terms of mobile source
emissions, the existing testing facilities at the EPA, Ann Arbor facil-
ity and Olson Laboratories were studied for the purpose of identifying
those elements requiring quality consideration.
          Basic concepts of quality assurance as they apply to the
measurement of mobile source emissions involve such areas as procurement
control, test quality control, data validation, corrective action, stan-
dards and calibration.  The guidelines offer guidance in the application
of quality assurance techniques in these areas.
          The measurement system used for light duty mobile sources is
described in detail in Volume I, and Test Procedures to meet the appli-
cable requirements of the Federal Register for the 1975 model year, used
by the EPA, Ann Arbor facilities appear in Volume II.
          Methods of performance checks and preventive maintenance are
discussed.  Quality management procedures and responsibilities of the
quality functions are included as Appendix C to Volume I.  Suggested
formats for documentation of test data, inspection reports, failure
reports, and other form requirements of a quality assurance program are
specified.
          Statistical methods are a valuable tool in the quality assu-
rance program.  Pertinent statistical methods are described with spe-
cific applications in emission testing.  Test variability is discussed
and test variables have been identified.  Methods for controlling or
reducing test variability are described.
          The report is divided into three parts, (1) a general quide-
line (Volume I) containing quality functions and provisions which are
generally applicable to organizations performing emission measurements
(2)  quality management procedures (Volume I, Appendix C) which define
the organizational procedures to be used and assign responsibilities
for the quality functions of a model quality assurance program  (3) test
procedures (Volume II) written for the EPA laboratory in Ann Arbor.  A
document control system is incorporated to facilitate updating of these
procedures as required by changes in the measurement system.

-------
                 TABLE   OF   CONTENTS
FOREWORD	      iii

EXECUTIVE SUMMARY 	      v

Section

1         INTRODUCTION	      1-1
1.1       Objective and Scope of Guidelines	      1-1
1.2       Formation of Quality Assurance Guidelines 	      1-2
1.2.1       Section 1 Introduction	      1-2
1.2.2       Section 2 Organizing for Quality	      1-2
1.2.3       Section 3 Measurement System Analysis 	      1-2
1.2.4       Section 4 Guidelines for Performance,
              Audits and Maintenance Procedures 	      1-2
1.2.5       Section 5 Quality Assurance Guidelines for
              Documentation of the Measurement System 	      1-3
1.2.6       Section 6 Application of Statistical Quality
              Assurance Methods to the Emission
              Test System	      1-3
1.2.7       Section 7 Analysis of Variability in the
              Measurement of Emissions from Light Duty
              Vehicle	      1-3
1.2.8       Section 8 Quality Assurance System
              (On Site) Survey	      1-3
1.2.9       Appendices	      1-3

2         ORGANIZING FOR QUALITY	      2-1
2.1       Operations Management 	      2-2
2.1.1       Quality Assurance Management	      2-2
2.1.2       Emission Test Facility Management 	      2-13

3         MEASUREMENT SYSTEM ANALYSIS 	      3-1
3.1       Applicable Federal Register Procedures	      3-1
3.2       Elements of a Measurement System for Light Duty
            Vehicle Emission Measurement  	      3-2
3.2.1       Evaporative Emission Measurement	      3-2
3.2.2       Exhaust Emission Measurement	      3-10
                                VII

-------
4         GUIDELINES FOR PERFORMANCEAUDITS AND MAINTENANCE
           PROCEDURES	     4-1
4.1       Performance Audits	     4-1
4.1.1       System Performance Characteristics, and Acceptable
             Limits	     4-3
4.1.2       Independent Performance Check Procedures	     4-7
4.1.3       Reporting and Corrective Action Procedures	     4-9
4.2       Preventive Maintenance	     4-10
4.2.1       Preventive Maintenance Procedures 	     4-11
4.2.2       Preventive Maintenance Action 	     4-14
4.2.3       Maintenance Log Procedures	     4-23

5         QUALITY ASSURANCE GUIDELINES FOR DOCUMENTATION OF THE
           MEASUREMENT SYSTEM	     5-1
5.1       Development of an Operations Manual	     5-1
5.1.1       Document/Manual Control	     5-2
5.1.2       Quality Management Procedures 	     5-2
5.1.3       Testing Procedures	     5-3
5.1.4       Related Information	     5-5
5.2       Documentation Requirements of a Quality Assurance
           System	     5-5
5.2.1       Recording Inspection Results	     5-7
5.2.2       Recording Calibration Results 	     5-7
5.2.3       Recording Maintenance Actions 	     5-15
5.2.4       Reporting Unacceptable Results	     5-22
5.2.5       Failure Reporting and Analysis.	     5-22
5.2.6       Initiating and Assuring Closed-Loop Corrective
             Action	     5-25
5.2.7       Recording Audit Results 	     5-28
5.2.8       Initiating Procedural or Equipment Change Notices .     5-28

6         APPLICATION OF STATISTICAL QUALITY ASSURANCE METHODS
           TO THE EMISSION TEST SYSTEM	     6-1
6.1       Statistical Methods 	     6-1
6.1.1       Special Applications of Statistical Methods ....     6-2
6.1.2       Statistical Techniques and Nomenclature 	     6-2
6.2       Control Charts	     6-2
6.2.1       Definition and Purpose of Control Charts	     6-2
6.2.2       Format	     6-3
6.2.3       Types of Control Charts	     6-3
6.2.4       Applications of Control Charts in Mobile Source
             Emission Testing	     6-5
6.2.5       Precision Control Charts	     6-5
6.2.5.1     Construction of Range Precision Control Charts
             (R-Charts)	     6-10
6.2.5.2     Construction of Relative Range Control Charts . . .     6-13
6.2.5.3     Construction of Coefficient of Variation Control
             Charts	     6-16
                                Vlli

-------
6.2.6       Accuracy Control  Charts 	     6-19
6.2.6.1     Construction of a Difference Control Chart	     6-19
6.3       Statistical Inference and some Applications of
           Acceptance Sampling	     6-20
6.3.1       General Context	     6-20
6.3.2       Definition of Statistical  Inference 	     6-20
6.3.3       Application of Sampling Theory in Mobile Source
             Emission Testing	     6-23
6.4       Analysis of Variance	     6-23
6.4.1       Basic Theory	     6-23
6.4.2       Analysis of Variance Implementation in Mobile
             Source Emission  Testing	     6-24
6.5       Data Validation	     6-27
6.5.1       Data Validation for Manual Techniques 	     6-28
6.5.2       Data Validation for Computerized Techniques  ....     6-28
6.5.3       Statistical Validation in  Maintaining Data Quality.     6-29
6.5.3.1     Outlier Analysis	     6-29
6.6       Methods of Calibration Curve Construction 	     6-31
6.6.1       General Content of Calibration Curve Construction .     6-31
6.6.2       Curveall	     6-32
6.6.3       Summary of Curve  Fitting Techniques 	     6-32
6.6.4       General Considerations	     6-32
6.7       The Use of Probability Paper	     6-34

7         ANALYSIS OF VARIABILITY IN THE MEASUREMENT OF
           EMISSIONS FROM LIGHT DUTY VEHICLES 	     7-1
7.1       Variables Associated with the Measurement of
           Evaporative Emissions	     7-2
7.2       Variables Associated with the Measurement of
           Exhaust Emissions	     7-3
7.2.1       Analysis of Variables Associated with Measurement
             and Reduction of Data	      7-3
7.2.1.1     Selection of the  Mathematical Model 	      7-6
7.2.1.2     Effect of Variables in the Measurement of Ambient
             Conditions and  the Calibration of Total Exhaust
             Volume and Mass  Emission  Values	      7-7
7.2.1.3     Variation in the  Determination of Exhaust
             Emission Concentrations	      7-10
7.2.2       Variation Associated with  the Equipment and
             Test Procedures  used in the Measurement System  .      7-14
7.2.3       Emission Measurement Variability Contributed by
             the Operator and Driver	     7-17
7.3       Measurement of Variability in Emission Measurement
           Systems	     7-18
7.4       Fuel Economy Measurements	     7-26
7.5       Quality Assurance and Test Variability	     7-27
                                 IX

-------
8         QUALITY ASSURANCE (CM-SITE)  SURVEY 	   8-1
8.1       General Requirements	    8-1
8.2       Administration Guideline, Quality Assurance System
            Survey Report 	    8-2
8.3       Quality Assurance System Survey Report	    8-10

9         REFERENCES. 	    9-1

Appendices

A-l       Statistical Techniques and Nomenclature 	    A-1-1
A-2       Control Chart Multiplication Factors	    A-2-1
B-l       Glossary of Terms	    B-l-1
B-2       List of Abbreviations	    B-2-1
C-l       Quality Management Procedures 	    C-l

-------
                          LIST OF FIGURES
Figure No.                                                         Page

2-1       Function/Responsibility Chart 	     2-3
2-2       Data Validation Control Network 	     2-11

3-1       Elements of a Measurement System for Light Duty
            Vehicle Emissions 	     3-5

4-1       Daily Start-up Checksheet 	     4-13
4-2       Preventive Maintenance Checklist CVS, Weekly	     4-15
4-3       Preventive Maintenance Checklist CVS, Monthly ....     4-16
4-4       Preventive Maintenance Checklist Analysis System,
            Weekly	     4-18
4-5       Preventive Maintenance Checklist Analysis System,
            Monthly	     4-19
4-6       Preventive Maintenance Checklist Dynamometer,
            Weekly	     4-20
4-7       Preventive Maintenance Checklist Dynamometer,
            Monthly	     4-21
4-8       Preventive Maintenance Checklist Individual
            Instruments, Monthly	     4-22
4-9       Maintenance Log Form	     4-24

5-1       Typical Receiving Inspection Form 	     5-10
5-2       Instruction for Receiving Inspection Report 	     5-11
5-3       Procurement Control List	     5-12
5-4       Calibration Tags	     5-13
5-5       Calibration Control Punch Card	     5-14
5-6       Calibration History Evaluation	     5-16
5-7       CVS Calibration Sheet	     5-17
5-8       Analyzer Curve Generation Data	     5-18
5-9       Monthly Dyno Calibration Log	     5-19
5-10      Gas Analysis Report	     5-20
5-11      Equipment Repair Authorization	     5-21
5-12      Rejection Report	     5-23
5-13      Failure Analysis Report 	     5-24
5-14      Corrective Action Request 	     5-26
5-15      Corrective Action Request - Flow Chart	     5-27
5-16      Performance Audit Summary Sheet 	     5-29
5-17      Procedure/Equipment Configuration Change Notice . . .     5-31

6-1       Control Chart Configuration - Propane Injection
            Test - % Error	     6-4
                                XI

-------
6-2       Scatter Diagrams for Determining the Type of
            Control Charts to use	    6-8
6-3       Range Control Chart 	    6-11
6-4       Relative Range Control Chart	    6-14
6-5       Coefficient of Variation Chart	    6-17
6*-6       Signed Differences Control Chart	    6-21
6-7       Normal Probability Paper	    6-35
6-8       Normal Probability Paper (CO Emission Level)	    6-38

7-1       Estimated Sources of Variability and Probable
            Relative Contribution for Mass Emission Errors on
            the CVS Cold Start Test at 1975 California Levels  .    7-4
7-2       Variability Associated with each Component Measured
            During each Phase of 1975 FTP	    7-22
7-3       Effect of Ambient Temperature on Exhaust Emissions
            During the CVS Cold Start Test	    7-24
                                xii

-------
                          LIST OF TABLES
Table No.

3-1       Subpart A - Emission Regulations  for New Gasoline
            Fueled Light Duty Vehicles	     3-3
3-2       Federal Emission Test Procedure - Preparation and
            Preconditioning 	     3-6
3-3       Federal Emission Test Procedure - Evaporative
            Emission Collection and Measurement 	     3-8
3-4       Federal Emission Test Procedure - Exhaust Emission
            Test	     3-11

4-1       Federal Register Specifications 	     4-5
4-2       Methods of Monitoring Variables 	     4-8
4-3       Major System Components  of Mobile Source Emissions.  .     4-14

5-1       Summary of Forms Referenced in Volume I	     5-8

6-1       Applications of Statistical Control  Charts in 1975
            FTP Testing	     6-6
6-2       Measured Data Used In Constructing Scatter Diagrams  .     6-9
6-3       Data Values and Computations  for  Constructing Range
            Control Chart Limits.  ...  	     6-12
6-4       Concentration Measurements -  Relative Range
            Calculation	     6-15
6-5       Test Measurements - Coefficient of Variation
            Calculation	     6-18
6-6       HC Concentration Measurements  - Correlation Vehicle
            vs Current Test Vehicle	     6-22
6-7       Analysis of Variance - One Way Classification ....     6-25
6-8       Analysis of Variance - Two Way Classification ....     6-26
6-9       Merits and Disadvantages of Two Curve Fitting
            Techniques	     6-33
6-10      Tabular Description of CO Emission Levels in PPM. .  .     6-37

7-1       Effect of Variables on the Determination of Mass
            Emissions Ambient Conditions and the Calculation
            °f Vmix	     7-8
7-2       Effect of Variables on the Determination of Mass
            Emissions - Measurements of  Diluted Exhaust and
            Ambient Air Concentrations	     7-12
7-3       Selected Vehicle Emission Test Variability from
            Several Sources 	     7-20
7-4       Effect of Barometric Pressure and Humidity on
            Exhaust Emissions 	     7-25
7-5       Summary of Test Variables and Methods used for their
            Control	     7-28
                               xiii

-------

-------
                                                            Section:   1(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page  1 of 3
                             Section 1

                           INTRODUCTION
          The Quality Assurance Staff of the EPA Quality Assurance and
Environment Monitoring Laboratory ,Research Triangle Park, North
Carolina is responsible for providing guidance for Quality Assurance Pro-
grams in the measurement of mobile source emissions.  Standards for the
emissions from light and heavy duty mobile sources have been promulgated
and procedures published for the measurement of their emissions and
certification.  Quality assurance guidelines, however, have not been
previously specified for the testing procedures.  Such quality assurance
programs are necessary to assure the integrity of the data resulting
from these tests.  This report presents guidelines for quality assurance
programs for measurement systems used in mobile source testing according
to the applicable requirements of the Federal Register for the 1975
model year.
          The guidelines for the Quality Assurance Program for mobile
source measurement systems are prepared in four phases.

          o    Phase I - For light duty gasoline powered
               vehicles (cars and trucks)

          o    Phase II - For heavy duty diesel engines

          o    Phase III - For light duty diesel powered vehicles (cars
               and trucks)

          o    Phase IV - For heavy duty gasoline engines

          This document presents the guidelines for implementing a
Quality Assurance Program for the measurement of emission from light
duty gasoline powered vehicles (Phase I).  Guidelines for the other
phases are reported in separate documents.
1.1       OBJECTIVE AND SCOPE OF GUIDELINES

          The measurement system for light duty vehicles consists of the
testing, calibration and analytical requirements, the operational and
measurement procedures used, and the operational and measurement data
obtained.  The primary objective of this program was to analyze this
system and apply the principles and techniques of modern quality assur-
ance systems to the total testing process to assure the validity and
reliability of the tests and the resulting test data.

-------
Section:   1(LG)
Revision:  0
Date:      June 1975
Page  2 of 3
          These guidelines provide information on general quality methods
which may be used in emission testing.  They were primarily designed for
use by management and supervisory personnel involved in the development
or operation of quality programs.  Upper management may use the guide-
lines to evaluate the quality programs which presently exist within
their own laboratory or organization.

1.2       FORMATION OF QUALITY ASSURANCE GUIDELINES

          These guidelines have been written in two volumes.  Volume I
contains the general guidelines which may be applied to almost any
mobile source testing facility.  Appendix C of Volume I contains general
Quality Management Procedures (QMP) which define those functions identi-
fied as being necessary in a quality program.  Volume II contains the
detailed testing procedures which are used by the EPA Ann Arbor facility
for 1975 certification testing.
          The quality assurance guidelines for light duty vehicle emis-
sion measurement systems are contained in Sections 1 through 8, with all
references appearing in Section 9.  A summary of the contents of each
section is as follows

1.2.1     Section 1  Introduction

          A description of the make-up and organization of the guidelines.

1.2.2     Section 2  Organizing For Quality

          A typical Quality Assurance Organization is presented.  Qual-
ity functions are identified and the various key elements of a quality
program are described.

1.2.3     Section 3  Measurement System Analysis

          A description of the measurement system defining the equip-
ment, test procedure specifications and tolerances, quality provisions
and other requirements necessary for emission testing of light duty
vehicles.

1.2.4     Section 4  Guidelines for Performance Audits \nd
          Maintenance Procedures

          General guidelines are presented for performance inspection
and maintenance of instruments and equipment used in the measurement
systems.  Preventive maintenance programs are described for increasing
the reliability and efficiency of the test equipment.

-------
                                                            Section:    1(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  3  of  3


1.2.5     Section 5  Quality Assurance Guidelines for_ Documentation
          of the Measurement System

          Guidelines for the development of a documentation system are
presented along with representative forms and description of the manuals,
data recording etc., required by a Quality Assurance program.

1.2.6     Section 6  Application of Statistical Quality Assurance Methods
          To The Emission Test System

          Basic statistical techniques such as control charts, analysis
of variance and data validation as applied to a quality system are
described.

1.2.7     Section 7  Analysis of Variability in the Measurement of
          Emissions from Light Duty Vehicles

          Sources of variability are identified and, where possible,
quantified to show their effect on the data.  A mathematical model
selected to give emissions similar to the 1975 Federal emission stan-
dards and to show the effect of the variability in data inputs on mass
emissions is discussed.

1.2.8     Section 8  Quality Assurance System (On Site) Survey

          A procedure and survey form for conducting a Q.A. survey of a
laboratory conducting light duty vehicle emission testing is presented.

1.2.9     Appendices

          Statistical techniques and nomenclatures appear in Appendix A-
1. Appendix A-2 contains control chart multiplication factors.  Appen-
dices B-l and 2 include a glossary of terms and a list of abbreviations
commonly used in the measurement system.

-------

-------
                                                           Section:   2(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 1  of 15
                             Section 2

                      ORGANIZING FOR QUALITY
          There are several ways in which a quality assurance program
may be incorporated into an organizational structure.  The management
level at which this function is introduced can greatly determine the
effectiveness of any quality assurance program.  Therefore, it is
necessary in the early stages of quality program planning to study
existing functions and responsibilities of each group or department
within an organization.  The scope of an organization studied should be
determined by defining all departments involved in the quality assurance
program and the management level at which the quality responsibility is
introduced into an organization.
          Basically a facility can be divided into four major management
functions, Quality Assurance, Data Analysis, Administrative Services and
Laboratory Operations.  A typical function/responsibility chart will
show the four primary functions and the various subfunctions considered
to be necessary in a quality organization.  The subfunctions should
maintain a high degree of flexibility, with assignments made on the
basis of manpower proficiency and availability within the major de-
partments.  Management should conduct frequent reviews of the effect-
iveness of the delegated authorities and assigned responsibilities in
order to make decisions on possible reassignments or establishing new
subgroups as necessary.
          A summary of the four primary functions follows.

          o    Quality Assurance - Has the overall responsibility for
               insuring adherence to the quality requirements recommended
               by EPA to comply with Federal regulations through all
               phases of testing emissions from light duty vehicles.

          o    Data Analysis - Develops computer programs and processes
               and monitors test-related data to insure the accuracy and
               reliability of the emission measurement.  Maintains data
               files of test information and provides statistical
               programs to assist quality assurance  in the improvement
               of test data.

          o    Administrative Services - Performs all the necessary
               peripheral functions required by the  laboratory such as
               purchasing, facility engineering, contracts administra-
               tion, and the training and certification of personnel.

-------
Section:     2(LG)
Revision:    0
Date:        June 1975
Page 2  of   15
          o    Laboratory Operations - Performs  emission tests  on light
               duty vehicles in accordance with  documented procedures.
               Is responsible for the calibration, maintenance  and
               control of the equipment used in  the  facility.


2.1       OPERATIONS MANAGEMENT

          Upper management should actively participate in establishing
quality policies, quality objectives and plans for meeting these  objec-
tives.  However, instead of providing active leadership  of the  quality
function, upper management may choose to delegate authority for this
leadership to some subfunction with a direct line of authority  from
upper management.  A positive management attitude towards quality should
stimulate an aggressive quality consciousness among all  employees.  In
establishing a quality assurance program, it is  important that  the
organization be structured to produce a high degree of quality  and
communication among functional groups with a minimum of  personal  friction
and overlap of authority.
          A separate mechanism should be established to  assist  in inte-
grating these responsibilities,  measuring their  success, and performing
functional responsibilities not assigned to other groups.  This mechan-
ism is Quality Assurance Management.
          A typical functional organization chart for an emission measure-
ment system is presented in Figure 2-1.  The actual organizational chart
at a given facility will depend largely on the size of the operation and
the assignment of the quality assurance responsibilities.  Assignment of
the functions should be on the basis of "best able" to accomplish the
job rather than trying to set up an "ideal"  organization.  These
functions will be discussed under two major topics; the  Quality Assurance
Management, and the Emission Test Facility Management.

2.1.1     Quality Assurance Management

          It is the primary responsibility of quality assurance to
assure the accuracy, precision and completeness  of the data from  the
test system by assisting and integrating the quality development,  quality
maintenance and quality improvement efforts of the various groups in the
organization.  The quality assurance program should stress prevention
rather than after-the-fact correction of errors  on involved tests.
Quality assurance also has the overall responsibility for assuring (1)
adherence to the procedures required by the applicable Federal  regula-
tions and specific contract requirements and (2) adherence to adequate
quality control practices.

-------
                 DIRECTOR

QUALITY
ASSURANCE

PROCUREMENT
CONTROL


TEST
QUALITY
CONTROL

CORRECT
ACT 10
VE
N

AUDIT










STANDARDS
AND
CALIBRATION

DATA
VALIDATION

DEFICIENCY
REVIEW

          DATA
        ANALYSIS
       STATISTICAL!
        ANALYSIS
         SYSTEMS
       DEVELOPMENT
         COMPUTER
        OPERATIONS
ADMINISTRATT
   SERVICES
  CONTRACTS
  PURCHASING
   FACILITY
   SERVICES
                           RECORDS
                         MANAGEMENT
                         TRAINING 5-
                       CERTIFICATION


LABORATORY
OPERATIONS


TEST
OPERATIONS



VEHICLE
TEST

FUEL Er GAS
STORAGE



— I





SUPPORT
OPERATIONS
1
i
CHEMICAL
ANALYSIS

EQUIPMENT
SERVICES

TEST
SCHEDULING
                                                               •fl O 50 W
                                                               P> P> (D fl>
                                                               ifl ft < O
                                                               (B (D H- ft

                                                                   H- O
                                                               U   03
Figure 2-1.  FUNCTION/RESPONSIBILITY CHART
                                        U1 C
                                          3
                                          fD
                                                                   O NJ

-------
Section:   2(LG)
Revision:  0
Date:      June 1975
Page  4 of 15
           The functional responsibilities assigned to Quality Assurance
 Management as shown on the functional chart (Figure 2-1) are procurement
 control, test quality control, corrective action, standards and calibra-
 tions, data validation, deficiency review and audit.  To accomplish these
 functions, the Quality Assurance Management may require assistance from
 the other groups in the organization or from outside sources.  For example,
 the Data Validation group might utilize the Data Analysis section to
 perform the statistical or other analyses of data they require.  The
 Standards and Calibration group might purchase certified gas standards
 from outside suppliers.  It must be remembered, however, that assignment
 of the responsibility for total quality assurance to a particular section
 does not relieve the other functional groups from performing their assigned
 quality responsibilities.

 Procurement Control

           A test facility purchases equipment, supplies and services from
 outside sources.  The function of Procurement Control is to assist Pur-
 chasing in determining qualified suppliers and to assure quality require-
 ments are met by monitoring an order from its inception to completion.
 This is accomplished in three basic steps.

           o    Procurement Document Review.  The purchase request and the
                related program are reviewed by Quality Assurance to
                determine if it includes the correct and adequate descrip-
                tion, specifications and requests for analysis and certifi-
                cation when required.  In addition, standard purchase
                order paragraphs are incorporated covering such items as
                warranty, materials of construction, packaging and shipping
                information, disposition of rejected material and failure
                to meet specified requirements or delivery time.

           o    Supplier Review.  A request for quote on the purchased
                material should be sent out to at least three suppliers
                unless for some reason the material or service is available
                from only a single source.  An actual on-site supplier
                review is usually unnecessary; however, responding suppliers
                should be reviewed by Procurement Control on the basis of
                past performance and ability to meet specifications of the
                purchase request.  Many of the problems encountered in
                purchasing equipment and instruments can be avoided by
                careful procurement document review.

           o    Receiving Inspection.  After the supplier is selected.
                Purchasing issues a purchase order including all the
                                  8

-------
                                                          Section:    2(LG)
                                                          Revision:   0
                                                          Date:      June 1975
                                                          Page  5 of 15
               requirements contained in the original request.  When the
               material is received,  it is subjected to the appropriate
               receiving inspection to insure that all the requirements
               of the purchase order are met.  Receiving inspection
               issues a receiving report noting any discrepancies.  This
               is sent to Purchasing and maintained in a supplier file
               by Procurement Control.  Should corrective action be
               required. Procurement Control will initiate a request for
               corrective action which is sent to the supplier through
               purchasing.  Procurement Control will then follow up this
               request to assure supplier compliance.

          Procurement Control is concerned with those items, materials
and services that can affect the quality of the test data.  A list of
these items, materials and services should be generated by and jointly
agreed upon between Quality Assurance and Purchasing.
          In an emission test facility a minimum of the following items
should be subjected to a procurement document review on an initial order
basis, with an analysis of any discrepancies/failures that may occur.

          1.   Pure gases; fuels and chemicals other than solvents and
               cleaning agents

          2.   Calibration gases

          3.   Filtering or gas absorbing material  (i.e., dryrite,
               ascarite, charcoal, etc.)

          4.   Carbon canisters

          5.   Replacement parts for calibration, analytical and/or test
               equipment

          6.   Analytical instruments or systems

          7.   Dynamometers

          8.   Constant volume samplers

          9.   Any sampling equipment used in the analytical process,
               such as tubing, flow controllers, meters,  pumps, valves,
               flowmeters and sample bags

          10.   Computer systems

-------
Section:   2(LG)
Revision:  0
Date:      June 1975
Page  6 of 15
          11.   Temperature and pressure measuring/controlling  items

          12.   Chart recorders,  chart paper, driver's  traces

          13.   Standards and  calibration  test  instruments

          14.   Vehicle diagnostic equipment

          15.   Fuel conditioning system

          16.   Special instrumentation/equipment  designed  in-house and
               purchased from outside suppliers  (i.e.,  sulfate tunnel,
               filter holders, etc.)

          17.   Standard reference materials  (i.e.,  weights (NBS),  etc.)

           In  addition to the  purchasing-procurement control relationship,
 it is  important  that the person  or  group  requesting an item be involved
 in the initial review.  The concept of cost usage should be considered in
 the initial procurement review.   (Ref. 2-1.)   A careful analysis of  the
 utility of a  particular item  should be performed  by reviewing  the  specified
 requirements  with  the user.   In  many cases, items are  purchased because
 they are considered to be the "best" on the market.  However,  often  a
 more cost-effective item can  be  utilized  without  affecting the quality
 desired.  For example, if calibration gases are subjected  to analysis
 during a receiving inspection and measured against primary standards, it
 would  be of no value to order gases with  a certified ±1.0  percent  analysis.
 This would be especially true if the gases are to be utilized  as routine
 span gases.   Gases other than primary standards can be ordered without
 analysis or with a specified  "make" tolerance  (a  guarantee that the
 actual value  will  fall between certain limits), or batch analysis  at a
 lower  cost.   The effect on quality  would  be negligible. On the other
 hand,  if these same gases were to be used as a primary standard, the
 ±1 percent tolerance might not be good enough  if  the desired end result
 of the instrument  being calibrated  was ±2 percent.   Usually a  factor of
 four is considered acceptable from  the primary to the  secondary standard,
 which  would dictate a required accuracy of ±0.5 percent for the primary
 standard.

 Test Quality  Control

           Since  Quality Assurance has the final responsibility for
 assuring the  quality of the emission data, it  must define  and  implement
 the necessary quality controls.  To improve the quality of the data  and
 decrease the  number of voided tests, prevention rather than correction
                                 10

-------
                                                          Section:   2(LG)
                                                          Revision:  0
                                                          Date:      June 1975
                                                          Page  7 of 15
should be stressed.  Many of the quality provisions are presently required
by the Federal Register, or by good engineering practices developed from
experience with the analytical process.  These provisions are presented
in the following section which describes the test procedures.  Test
quality control is a continuing process since the system is constantly
generating new data and subject to change.  The data generated can be
used to identify those areas which have need for better control and those
areas which may be over-controlled.  As new instruments are introduced
into the system, their characteristics and operating parameters must be
carefully studied for revision of quality control procedures.  The methods
for evaluating the system can be functional or statistical.  Functional
evaluations would uncover particular operation needs and requirements
such as calibration and maintenance.  Statistical evaluation involves a
system study utilizing data generated by the system.  This data history
might be used to construct control charts, define acceptable limits or
predict a need for calibration or preventive maintenance.  These techniques
are discussed to a greater degree in Section 6.  Quality analysis should
also be applied to evaluation, development or research programs where the
test data will be used for some special reason such as determining the
effectiveness of tune-ups on emissions.  In this case, new variables
which require controlled provisions will be added to the system.

Corrective Action

          The feedback of error information to the originator of the
error, with a request for corrective action to prevent recurrence of such
an error, is a vital part of an effective quality assurance system.  The
corrective action system must be provided with a "closed loop" mechanism,
namely, persistent follow-up until satisfactory corrective action has
been accomplished and documented.  Failure to follow up on a corrective
action request nullifies the power of this important quality tool.
Further discussion of corrective action is presented in Sections 4 and 5.

Standards and Calibration

          A primary function in any system is the maintenance of standards
and calibration for measuring devices.  Vehicle emissions are determined
on a mass basis from gas concentration, volume, flow rates and density
measurement.  In the early days of testing, all measurements and emission
standards were based on a volumetric measurement.  Measurements were made
by non-dispersive infrared instruments which were calibrated from standards
usually prepared by partial pressure and  analyzed by gas chromatography.
Primary standards were generated by partial pressure using mixtures
prepared  in all-glass manifolds at only slightly above atmospheric pressure
 to avoid  compressibility problems.  Hydrocarbon standards were based on a
                               11

-------
Section:   2(LG)
Revision:  0
Date:      June 1975
Page  8 of 15
hexane standard which was liquid at room temperature and small amounts
could be weighed and introduced into a container which was then.pressurized
with nitrogen to a convenient pressure to give the desired concentration.
These procedures received some measure of success in the research labora-
tory but did not lend themselves to the procedure for measuring exhaust
emissions.  Several problems were encountered primarily because of the
volume of calibration gases required by the instruments.  Because of
economic and instrument requirements, calibration gases were prepared at
high pressures in steel cylinders.  It soon was discovered that what went
into the cylinder, did not necessarily come out.  Also because of the
uncertainty of actual temperatures and pressure within the cylinder,
blending by partial pressure was at best a rough estimate of the actual
concentration.  Measurement of the raw exhaust on a volumetric basis was
also unsatisfactory since it was felt that volumetric emission standards
penalized the smaller engines.
          Mass emission testing began with the introduction of the
constant volume sampler (CVS), which allowed the calculation of the
individual exhaust components by determining the concentration in a known
dilute volume of exhaust at specified constant temperature and pressure.
By knowing the density of the component and the volume at these specific
conditions the mass of each could be calculated.  By careful control of
the vehicle cycle and knowing the miles driven by the vehicle it could
then be converted to a mass rate in grams per mile.
          The "state of art" for preparing gas mixtures has improved
greatly since the early days of vehicle testing.  Stable blends are pre-
pared routinely by gas suppliers.  NBS and EPA have been cooperating in
a program to prepare Standard Reference Material (SRM) i.e., NBS certified
gas mixtures, for vehicle emission testing.  Gravimetric blends have been
prepared by the EPA for carbon monoxide, carbon dioxide, nitric oxide,
and propane mixtures.  Gravimetric blends of NO in nitrogen are presently
being prepared by NBS.
          Any vehicle emission testing facility should maintain a complete
set of instrument calibration standards which are traceable to the EPA
primary standards.  Working standards, used on a daily basis are analyzed
using this calibration set.  Annual correlation of the calibration set
with EPA is recommended.  Another possible method of checking these
standards is through a gas cross-referencing program whereby cylinders of
unknown blends are sent to each test site on a quarterly basis and analyzed
against the standards.  The reported analyses are treated statistically
and those analyses which are suspect may indicate a need for auditing the
particular calibration gases involved.
          In addition to the calibration of the analytical instruments
there are several other tests or calibrations required within the mea-
surement system.  Of primary concern is the calibration of the pump in
the CVS which determines the total amount of exhaust-air mixture.  This
                                12

-------
                                                             Section:    2(LG)
                                                             Revision:   0
                                                             Date:       June  1975
                                                             Page  9 of  15
is usually performed using a laminar flow element with traceability  to
NBS with an accuracy of ±1 percent.
          Other instruments which require calibration are:

          1.   Vehicle diagnostic equipment
          2.   Temperature recorders
          3.   Barometers
          4.   Hygrometers
          5.   Dynamometers
          6.   Recorders or output measuring devices.

          Quality Assurance should be responsible for maintaining the
calibration procedures and records in a file by specific item and up-
dating and reviewing the calibration results as they are performed.  The
Laboratory Operations in turn should report all calibrations results to
Quality Assurance.

Deficiency Review

          In any measurement system, deficiencies, repetitive errors and
an  inordinate void test rate may be encountered.  When any of these
situations are discovered, a review should be initiated by Quality
Assurance Management to determine the cause and recommend a plan for
correcting the situation.  This review is sometimes conducted by a
Deficiency Review Committee which is composed of representatives from the
departments involved. Management, and Quality Assurance.  Their recommenda-
tions are presented to the Laboratory Management.  Quality Assurance has
the responsibility for measuring the effect of implementing these correc-
tive actions.  This review process should be repeated until the desired
results are achieved.

Data Validation

          A control network must be established to assure a smooth flow
of  all data collected during an emission test.  Data Validation should
perform the control function and also check the data forms to confirm the
validity of the results and assure the data is within specified limits.
This function should be independent of the test technicians and should
report directly to Quality Assurance Management or the Laboratory Director.
Validation is accomplished usually by personnel with extensive experience
in  vehicle emission testing.  The actual checks should be done against
specified documented control limits.  Data transfers should be verified
and all forms checked for completeness.  Invalid or incomplete data  should
 be reported to the testing supervisor  and Quality Assurance.   The  rate
 of error  should be monitored to  determine trends  and  the need for  corrective
                                 13

-------
Section:   2(LG)
Revision:  0
Date:      June 1975
Page 10 of 15
 action.  Data validation procedures will vary with the size and structure
 of the laboratory.  Their merit will be evaluated in terms of user
 acceptability of the validity of the data.  Figure 2-2 shows an example
 of a Data Validation network and illustrates the direction of the collected
 data flow.  A detailed procedure for Data Validation is contained in
 EPA, Ann Arbor, Test Procedure #TP-801,  (Volume II).

 Audits

           Independent and random audits should be made to further assure
 the reliability of the measurement system.  Two types of audits are
 generally considered.

           o    Performance Audit - a planned independent random check of
                the data output and personnel in order to evaluate the
                quality of the output from the total system.  Section 4.0
                discusses performance audits in greater detail.

           o    System Survey - a systematic on-site qualitative review of
                facilities, equipment, training, procedures, record-
                keeping, validation and reporting aspects of a total
                (quality assurance) system, to arrive at a measure of the
                capability and ability of the system.  Even though each
                element of the system survey is qualitative in nature, the
                evaluation of each element and the total may be quantified
                on some subjective basis.  A typical quality system survey
                is discussed in Section 8.

           There are several performance audits which are done in the
 emission measurement system.  Routine performance checks are not con-
 sidered as part of the audit system since they are performed on a sche-
 duled rather than a random basis and are usually performed by the test
 operators as an integral part of their activities.  However, an auditor
 may use any of the performance checks such as NO  converter efficiency,
 propane injections, etc., as an audit check.  These checks are discussed
 in Section 4.
           Audits are important to the Quality Assurance Management as the
 only objective method available to determine the data quality and to
 assure that the emission test is being conducted according to the pre-
 scribed procedures.
           The audit report is the most important part of the audit pro-
 cedure, but to be effective, it must reach the management level having
 the authority to initiate corrective action; Quality Assurance Management
 should have the authority to shut down any part of the testing system
 producing invalid test data until the non-conforming condition is corrected.

-------
                1975 FTP
                REQUESTOR
   DATA
VALIDATION
                PRODUCTION
                 CONTROL
                    I
                 VEHICLE
                  PREP/
                 DIURNAL
                  EVAP.
                    I
HOT SOAK
EVA P.


—~

VEHICLE
TEST
I
XDATA\.
&i& \ \ RAT iriMV
                           NO
                     YES
                   DATA
                PROCESSING
 FTP - FEDERAL TEST PROCEDURE

HWFET - HIGHWAY FUEL ECONOMY TEST
                                                         Section:   2(LG)
                                                         Revision:  0
                                                         Date:     June 1975
                                                         Page 11 of 15
                                         1975 HWFET
                                          REQUESTOR
                                                        PRODUCTION
                                                         CONTROL
                                                            I
                                                         VEHICLE
                                                           TEST
                                                             J
                                                           DATA
                                                        VALIDATION
                                                           DATA
                                                        PROCESSING
                                                             1
                                                         REQUESTOR
                                                             t
             Figure  2-2.   DATA  VALIDATION CONTROL NETWORK
                                 15

-------
Section:   2(LG)
Revision:  0
Date:      June 1975
Page 12 of 15
Other Quality Assurance Elements

          There are other elements which should be considered in a total
quality assurance program, such as:
          Reliability is becoming an increasingly important considera-
tion in emissions measurement due to the complex systems involved.  The
probability of failure tends to increase as equipment becomes more
complex.  A comprehensive reliable testing program must rely on many
tools:

          1.   Accurate and complete record-keeping, with a data
               feedback loop built into the program.

          2.   Specific preventive maintenance schedules including
               replacement schedules to remove and replace low re-
               liability parts before they reach wear-out stage.

          3.   Complete descriptions of the products that are required to
               undergo reliability testing.  These descriptions will
               include specifications for both quality and reliability.

          4.   Concise specifications for the performance of tests,
               including meticulous attention to the ambient conditions -
               such as number of operating cycles and times, temperatures,
               shock, pressures and vibrations - that are to prevail
               during testing.

          5.   Definite sampling procedures, sample sizes, criteria for
               judging the success or failure of a test and acceptance
               and rejection values for action on a measurement.

          6.   Knowledge of the calculated sampling risks, such as those
               embodied in operating characteristic curves or tabulated
               data on the probabilities of sampling errors.

          A failure analysis report should be prepared for each occurrence
of  equipment failure.  There is a further discussion of failure reporting
and analysis in Section 5.
          Configuration and Documentation Control in a testing laboratory
is  primarily concerned with assuring that all similar equipments have
the same  configuration and that all hardware and document changes,
including computer program revisions have been recorded.  There is a
further discussion of these controls in Section 5.
          Quality costs should be readily identifiable in any effective
quality assurance system.  These costs are usually categorized into:

          1.   Prevention Costs
          2.   Appraisal Costs
          3.   Internal Failure Costs
          4.   External Failure Costs

                                 16

-------
                                                            Section:    2(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page 13 of  15

          Reference 2-3 discusses these costs in greater detail.
          The benefits from implementing a quality cost program include:

          1.    Overall quality performance can be measured in terms
               easily understood by management.

          2.    Problem areas can be defined.

          3.    Input for budgeting purposes can be easily obtained.

          4.    Cost savings can be readily identified.

          A system should be designed to collect and report these  costs
accurately, completely and in a meaningful manner, and the data should be
properly organized and available when needed.

2.1.2     Emission Test Facility Management

          Basically, the management of an emission test facility can be
divided into four major activities; Quality Assurance Management,  as
discussed in the previous section, Administration, Data Analysis and
Laboratory Operations which includes Test Operations and Support Opera-
tions.  In a total quality assurance program, the organizational struc-
ture may appear as shown in Figure 2-1, with Quality Assurance on  the
same management level as Administration and Laboratory Operations.  There
is general agreement among the experts in the field of quality assurance
that the introduction of a quality function below this level will  not
provide the necessary line of authority to succeed.
          A brief description of the primary functions performed by the
departments are:

Administrative Services

          Administrative Services performs all the necessary peripheral
functions required by the emissions facility such as contracts, pur-
chasing, facility services, personnel, records management, training and
certification.

          1.   Purchases from the Quality Assurance Approved Supplier
               List all materials, equipment, instruments, expendable
               items, office equipment, etc., which affect test data
               quality.

          2.   Routes purchase requests for those items which affect
               test data quality to Quality Assurance for approval of
               specifications and drawings and inclusion of standard
               quality clauses where applicable.
                                 17

-------
Section:   2(LG)
Revision:  0
Date:      June 1975
Page 14 of 15

          3.   Requests approval and review of suppliers' product from
               Quality Assurance as required.

          4.   Provides for all facilities engineering requirements
               which may have an effect on data quality such as plumbing,
               heating, cooling, electrical wiring, ventilation and fuel
               storage.

          5.   Initiates, recommends, and implements safety programs and
               procedures for the facility which meet personnel, equip-
               ment and building requirements in accordance with OSHA,
               EPA and State regulations.

          6.   Formulates, recommends and implements administrative
               policies in accordance with the Quality Management
               Procedures.  (See Appendix C.)

          7.   Controls and maintains inventory of all parts, supplies,
               and equipment used in the normal operation of the emission
               test facility.

          8.   Responsible for training personnel involved in any phase
               of vehicle emission measurement and should assist in
               training quality assurance personnel.  Administrative
               Services should assist Test Operations and Quality Assur-
               ance in the development of adequate training programs and
               the evaluation of programs using written or practical
               "hands-on" examinations.  A report describing examination
               development methodology is available from the Environ-
               mental Protection Agency.  (Ref. 2-2.)

Data Analysis

          Data Analysis develops computer programs, processes and monitors
test-related data to assure the accuracy and reliability of the emission
measurements.  Maintains data files of test results and provides statisti-
cal programs to assist Quality Assurance in the evaluation and improve-
ment of  test data quality.

          1.   Develops and processes computer programs for the reduction
               of test data to provide emission results on a grams per
               mile basis for carbon monoxide, hydrocarbons, carbon
               dioxide and nitric oxide.  Provides results for fuel
               economy on a mile per gallon basis utilizing the carbon
               balance method.
                                 18

-------
                                                           Section:   2(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 15 of 15
         2.   Maintains all test data  in  a  data file.

         3.   Provides statistical analysis for Quality  Assurance require-
              ments such as determination of acceptable  test  parameter
              limits, preparation of control charts, analyses of variance
              and cost effectiveness analyses.

         4.   Develops computer programs  for calibration data,  maintains
              calibration data file, and  computes instrument  calibration
              curves.  Informs Quality Assurance and Test Operations
              when calibration and maintenance  has not been performed
              according to the intervals  prescribed by Support Operations.

         5.   Assists Quality Assurance in  monitoring  all data to
              verify the accuracy and  reliability of emission measure-
              ments .

         6.   Maintains the paperwork  inventory for calibration gas
              cylinders.
          7.    Assists Quality Assurance  and Laboratory Operations in the
               construction  of mathematically correct formulas for the
               reduction of  data  for  non-routine test programs.

          8.    Assists Quality Assurance  in developing and implementing
               correlation and audit  programs to assure the reliability
               of the data on a "cell-to-cell" basis,  and for comparison
               with other laboratories performing emission testing.

          9.    Documents all program  changes, forms,  etc.

Laboratory Operations

          Laboratory Operations has the responsibility for the daily
operation of the vehicle test section. This includes the performance of
emission testing, calibration, maintenance, sample analysis and the
supervision of personnel, equipment and vehicles utilized in the perfor-
mance of emission testing.
          Quality Management Procedure Number 2.3 in Appendix C details
the primary functions of the Test Operations and Support Operations
departments.
                                19

-------

-------
                                                            Section:   3(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page  1 of 13
                            Section 3

                   MEASUREMENT SYSTEM ANALYSIS
          A total Measurement System can be defined as an orderly arrange-
ment consisting of the analytical method, the test sampling procedure, the
instruments or analyzers, the supporting functions, the management organi-
zation and the technicians or personnel involved in performing specific
functions within the system.  Applying this definition to the measure-
ment system for light duty gasoline vehicle emissions, the process is
composed of:

          o    The test procedure defined by the Federal regulations

          o    The preparation of the vehicle for the emissions test

          o    The exhaust emission sampler and analytical bench  (train)
               consisting of instruments for the measurement of carbon
               dioxide (CO ) carbon monoxide (CO) hydrocarbons (HC) and
               nitrogen oxides (NO )
                                  X
          o    The Laboratory Operations management including Test
               Operations and Support Operations

          This measurement system was subjected to a functional analysis
to determine and define the basic elements which require attention in a
total g_uality assurance program.
3.1       APPLICABLE FEDERAL REGISTER PROCEDURES

          Measurement Systems for which Quality Assurance  guidelines  and
procedures have been developed are defined  in  the  Federal  Register.
Those portions of the Federal Register which define  the measurement
systems covered by this document are:

             Date              Vol.     No.             Page

     1. November 15, 1972      37       221         24265-25277
     2. June 28, 1973          38       124         17149-17162
     3. October 31, 1973       38       209         30080-30081
     4. January 21, 1974       39         14            2364
     5. February 27, 1974      39         40         7548-7551
     6. May 23, 1974           39       101         18077-18080

                                21

-------
Section:   3(LG)
Revision:  0
Date:      June 1975
Page  2 of 13
           The cut-off date for Federal Regulations considered in this
 report was July 31, 1974.   Subsequent revisions should be incorporated
 into this document by the user.
           The paragraphs of the Federal Register subparts defining the
 scope of the measurement system for light duty vehicles appear in Table 3-1.
 3.2       ELEMENTS OF A MEASUREMENT SYSTEM FOR LIGHT DUTY
           VEHICLE EMISSION MEASUREMENT

           A requirement of a total Quality Assurance Program is to have
 control at all important stages of a process.  In this
 measurement system, an analytical process, it is necessary to first
 identify its functional elements.  In order to categorize these
 elements the measurement system has been divided into 3 basic operations:

           o    Vehicle Preparation
           o    Evaporative Emissions Measurement
           o    Exhaust Emission Measurement

           These three job categories are further separated into the tasks
 or elements requiring quality consideration in Figure 3-1.
           Vehicle preparation and preconditioning are generally
 accepted as part of evaporative emissions measurement  and the general
 procedure is shown in Table 3-2.

 3.2.1     Evaporative Emission Measurement

           A summary of evaporative emission collection and measurement
 procedures is shown in Table 3-3.  The purpose of this matrix is to show
 a general overview and does not attempt to include every detail required
 for the collection and measurement process.  The information discussed
 in the table consists of:

           o    A brief description of the tasks

           o    Applicable Federal Register paragraphs

           o    Applicable EPA, Ann Arbor, Test Procedure numbers

           o    Specifications and tolerances included in the Federal
                Register and from Engineering practices

           o    Quality provisions

           o    Invalid tests  (determination)

           o    Corrective actions required

           o    Training and skill level required

                                 22

-------
                                                            Section:    3(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  3  of  13
           Table 3-1.  SUBPART A - EMISSION REGULATIONS
           FOR NEW GASOLINE-FUELED LIGHT DUTY VEHICLES

APPLICABLE
REVISIONS*        SECTION                           TITLE

   2, 5           85.002            Definitions

   2              85.003            Abbreviations

   2, 5           85.006            Maintenance of records, submittal of
                                    information; right of entry

   2              85.075-6          Maintenance

   2              85.075-7          Mileage accumulation and emission
                                    measurements

   2              85.075-9          Test procedures

   2, 4           85.075-10         Gasoline specifications

   2, 3           85.075-11         Vehicle and engine preparation  (fuel
                                    evaporative emissions)

   2, 3           85.075-12         Vehicle preconditioning (fuel
                                    evaporative emissions)

   2, 6           85.075-13         Evaporative emission collection
                                    procedure

   2, 6           85.075-14         Dynamometer driving schedule

   2, 6           85.075-15         Dynamometer procedure

   2              85.075-16         Three-speed manual transmissions

   2              85.075-17         Four-speed and five-speed manual
                                    transmissions

   2              85.075-18         Automatic transmissions

   2, 3           85.075-19         Engine  starting and restarting

   2, 6           85.075-20         Sampling and analytical system
                                     (exhaust emissions)

   2, 6           85.075-21         Sampling and analystical  system
                                     (fuel evaporative emissions)

   2, 3,  6       85.075-22         Information to be recorded
                                 23

-------
Section:   3(LG)
Revision:  0
Date:      June 1975
Page  4 of 13


       Table 3-1.  SUBPART A - EMISSION REGULATIONS FOR NEW
            GASOLINE-FUELED LIGHT DUTY VEHICLES (Cont.)

APPLICABLE
REVISIONS*        SECTION                           TITLE

   2, 3, 6        85.075-23         Analytical system calibration and
                                    sample handling

   2, 3           85.075-24         Dynamometer test runs

   2              85.075-25         Chart reading

   2, 3, 6        85.075-26         Calculations (exhaust emissions)

   2              85.075-27         Calculations (fuel evaporative
                                    emissions)

   2, 3           85.075-28         Compliance with emission standards

   2, 5           85.075-29         Testing by the Administrator
                  Appendix   I - EPA urban Dynamometer Driving
                                 Schedule

                  Appendix  II - Procedure for Dynamometer Road
                                 Horsepower Calibration

                  Appendix III - Constant Volume Sampler Flow
                                 Calibration

                  Appendix  IV - Durability Driving Schedule
 *  1) 15 Nov  72  2)  28 June 73  3) 31 Oct 73
   4) 21 Jan  74  5)  27 Feb  74  6) 23 May 74

-------
                                                  Section:   3(LG)
                                                  Revision:  0
                                                  Date:      June 1975
                                                  Page  5 of 13
           FEDERAL TEST PROCEDURE

                     FOR

            LIGHT DUTY  VEHICLES
             VEHICLE PREPARATION
                     AND
               PRECONDITIONING

             o  Receive vehicle
             o  Vehicle inspection
             o  Preparation
             o  AMA route
             o  Driveability
             o  Preconditioning
EVAPORATIVE EMISSION

o

o
o
o
o
MEASUREMENTS
Install collection
system
Diurnal/heat build
Running loss
Hot Soak
Weigh canister
              EXHAUST EMISSION
                 MEASUREMENT

             o  Dyno warm-up
             o  Start-up procedure
             o  CVS sampling
             o  Analytical
                measurements
             o  Data reduction
Figure 3-1.  ELEMENTS OF A MEASUREMENT SYSTEM
       FOR LIGHT DUTY VEHICLE EMISSIONS

-------
Table 3-2.  FEDERAL EMISSION TEST PROCEDURE

  Vehicle Preparation and Preconditioning

PROCEDURE
ITEM
BRIEF
DESCRIPTION









FEDERAL
REGISTER
PARAGRAPHS
EPA PROCEDURE
NUMBER
SPECIFICATIONS

AND TOLERANCES
Federal
Register














Engineering
Practice




RECEIVE
VEHICLE
Visual inspec-
tion of test
vehicle to ver-
ify vehicle and
engine system
integrity





85.075-5,6,7


TP-701




None















Refer to manu-
facturers
specific, for
engine class

PRE-
CONDITIONING
AHA
All vehicles
driven over
sane route to
establish
similar his-
tories before
test




85.075-7,10,
12
Appendix IV
TP-702




Driving Time-
1 hour. Modified
routes must be
approved by the
Administrator .
Fuel-Tank fuel,
unleaded fuel
0.02 grams of
lead and 0.002
gin. phosphorus
per gallon
minimum.
Leaded Fuel -
1.4 gin. lead
per gallon,
minimum
Urban route ap-
proved by Ad-
ministrator



DRIVEABILITY

To determine
that veh.is
operating sat-
isfactorily.
safely and can
drive the
federal cycle







TP-702




None















Correct mal-
functions when
possible



VEHICLE
INSPECTION
To assure en-
gine parameters
are correctly
set. Chk IDLE,
CO, RPM, igni-
tion timing
dwell, centri-
fugal and
vacuum advance





TP-701


1

None















Manufacturers
range or speci-
fication


VEHICLE &
ENGINE
PREPARATION
Leak proof fit-
ting applied to
all fuel sys-
tems . External
vents to permit
collection of
emissions. Fuel
system leak-
checked . Install
thermocouple c
drain tank
85.075-11


TP-702




Fittings and
tubing for can-
isters 5/16
I.D.












Fuel system
should lose
not more than
2" HO at 14"
H20 in 5 min.

TEST FUEL
ADDED
Indolene 30
Indolene HO









85.075-10,11


TP-702




See above re-
ferenced para-
graph for de-
tailed specifi-
cations
















• PRE-
CONDITIONING
FEDERAL CYCLE
The vehicle is
driven on a dy-
namometer under
controlled con-
ditions






85.075-12


TP-703




Temp 77±9°F
Speed Tolerance
t4 MPH, 11 sec.
Hot start is
Acceptable
















11 HOUR
AMBIENT
SOAK
The vehicle is
stored in a con-
trolled environ-
ment







85.075-13


TP-703




1st hour 81°
±5°F. Followed by
10 hours 73
±13°F

















                                                                                                      flj Ql  (0 fp

                                                                                                      vo rt  < o
                                                                                                      fl> 0>  H- ft
                                                                                                         ..  to H-


                                                                                                            *8

                                                                                                      o
                                                                                                      Hi

                                                                                                      M C-l  O UJ
                                                                                                      CJ 0    ^-^

                                                                                                         CD    CD



                                                                                                         s
                                                                                                         Ul

-------
                                                              Table 3-2.  FEDERAL EMISSION TEST PROCEDURE

                                                                 Vehicle Preparation and Preconditioning

                                                                                (Continued)
PROCEDURE
ITEM
QUALITY
PROVISIONS
TEST INVALID
CORRECTIVE
ACTION
TRAINING OR
SKILL
REQUIRED
RESPONSIBLE
OPERATIONS
RECEIVE
VEHICLE
Inspection form
completed and
signed
Engine or vehi-
cle parts
missing or
disconnected
Vehicle re-
turned to
manufacturer .
Engine system
training
Receiving
inspection
Production
Control
PRE-
CONDITIONING
ANA

Failure to
complete route.
Accident.
Reschedule ve-
hicle. Repair
or replace
vehicle.
Normal driving
skills
Testing
Operations
DRIVEABILITY

Engine mal-
function,
brake failure,
vehicle un-
safe.
Return to
manufacturer
or supplier.
Driveability
characteristic
training
Testing
Operations
VEHICLE
INSPECTION
Calibration of
engine test
equipment
Incorrect en-
gine parameters
Adjust under
manufacturers
supervision
Mechanic
Testing
Operations
Support
Operations
VEHICLE S
ENGINE
PREPARATION

Failure to seal
system. Fuel
system leaks
Return to manu-
facturer
Installation
procedures
training
Testing
Operations
TEST FUEL
ADDED
Color coded
fuel pumps and
vehicle tags
and fuel
inlets. Fuel
analysis
Incorrect fuel
added. Fuel out
of specifica-
tion
Drain tank and
refuel with
correct fuel
None
Testing
Operations
Support
Operations
PRE-
CONDITIONING
FEDERAL CYCLE
Monitor temper-
ature and in-
spection of
drivers trace.
Dyno Calibra-
tion
Temperature
outside limits
Drivers trace
outside limits
Reschedule
test
Dynamometer
cycle drivers
training
Testing
Operations
Support
Operations
11 HOUR
AMBIENT
SOAK
Monitor tempera-
ture in soak
areas
Temperature
outside limits
Starting engine
gine during soak
Reschedule vehi-
cle. Correct
temperature
control .
None
Testing
Operations
Building
Maintenance
ro

                                                                                                                                                                        ?a
                                                                                                                                                                        n
                                                                                                                                                                           cn
                                                                                                                                                                     ft <  O
                                                                                                                                                                  (D (D H- ft
                                                                                                                                                                        w
                                                                                                                                                                        H-
                                                                                                                                                                        §  2
                                                                                                                                                                  h- C-| O
                                                                                                                                                                  we

-------
  TaMe  i-3    FEDERAL EMISSION TEST  PROCEDURE

Evaporative  Emission Collection  and  Measurement

PROCEDURE
ITEM
BRIEF
DESCRIPTION










FEDERAL
REGISTER
PARAGRAPHS
EPA TEST
PROCEDURE
SPECIFICATIONS
AND TOLERANCES
Federal
Register













Engineering
Practice






DRAIN
FUEL
Residual fuel
is drained from
tank after 11
hour soak.








85.075-13


TP-702



None














INSTALL CARBON
CANISTER
The carbon can-
ister traps the
emissions from
the fuel sys-
tem. Schematics
A75-3, A75-4,
A75-5, A75-6





85.075-13, 21


TP-702



Capacity- 3 00
12 ml. Length
to diameter
Ratio-1.4 iO.l
Inlet and out-
let tubes -
5/16 I.D.,
length 1 inch.
leak tight at
2 PSI 30 sec.,
150 ±10 gms.
of charcoal
conditioned
at 300 F for

ADD TEST
FUEL
A specified
test fuel with
known composi-
tion is added
to the tank,
Indolene 30 or
Indolene HO





85.002,
85.075-10, 13

TP-702



Charge 50-60 F
Start 60 ±2°F
End 84 ±2 r
Time 60 110
min. Charge to
40% of nominal
tank volume to
nearest gallon






DIURNAL
EVAP TEST
HEAT BUILD
Fuel vapors
emitted as a
result of a
specific in-
crease in fuel
tank tempera-
tures in a
specified time
are collected.
Record ambient
and fuel temp-
erature
85-075-13


TP-705



Temperature re-
corder, Range
50-100 ±1 F
Thermocouple -
Type J









(1) For more comple?erietail see Federal Register para. 85.075-2
Fuel pump
cart of nbt
more than 25
gallon capa-
city. Metts
OSHA require-
ments .







Heating rate
4 ±1.5 F Per
10 min.




Heating blanket
2000 watts to
cover min. 50%
of liquid fuel




DYNAMOMETER
PREPARATION
The vehicle is
placed on the
dynamometer
without start-
ing the engine
and the neces-
sary connec-
tions are made




85.075-13


TP-604



Soak vehicle at
76-86°F for a
min. of 1 hour
before running
loss test









..
Max. total soak
time from key
off to key on -
20 hours




RUNNING LOSS
TEST
Fuel vapors are
collected dur-
ing operation
of the vehicle
under the spec-
ified test
schedule





85.075-13


TP-706



See 1975 ex-
haust emission
test Table 3.4
Vapors are not
collected dur-
ing 10 min soak
or 505 second
"hot" start
test






See Table 3.4







1 HOUR HOT
SOAK LOSS
Fuel vapors are
collected for 1
hour beginning
immediately
after the en-
gine is turned
off.





85.075-13


TP-708



Ambient temp.
76-86°F





















CANISTER
HEIGHT
The collected
vapors are de-
termined by
weighing the
canister before
and after the
test.





85.075-27


TP-708



Weighing accur-
acy equip ±75 mg
weight deter-
mined to 20 mg.











Metier P1200 or
equivalent Reada-
bility 0.01 gram




t) O
(U pj
vQ rt
fl> fl>
                                                                                                           00


                                                                                                           O
                                                                                                           Hi
                                                                                                                fl> 
-------
                                                                Table  4-3.   FEDERAL EMISSION TEFT  PROCEDURE

                                                              Bvaporativ<  Emission Colloctior an*  «p,T3ur«"ii»-nt

                                                                                  (Cont mued:
PROCEDURE
ITEM
QUALITY
PROVISIONS
TEST
INVALID
CORRECTIVE
ACTION
TRAINING OR
SKILL REQUIRED
DRAIN
FUEL
Check-off sheet
signed by
witness
Failure to
drain tank.
Starting
engine.
Reschedule
Basic know-
ledge of fuel
system.
INSTALL CARBON
CANISTER
Installation
checked by team
leader. Canister
checked for
leaks by
comparing wt.
before test
with previous
tare weight.
Improper in-
stallation or
canister leaks
Correct in-
stallation. Re-
schedule if
heat build
had been
started.
Familiarity
with EPA ap-
proved in-
stallation
for engine
family
DIURNAL
ADD TEST EVAP TKST
FUEL HEAT BUILD
Ambient and fuel temp. Record
checked by data validation (DV)
step by step procedure check-
off form signed by witness
Incorrect temperature, heating
rate or time of heat applica-
tion.
Reschedule
Basic knowledge of heating and
temp measuring equipment.
DYNAMOMETER
PREPARATION
Ambient temp.
and soak time
by DV.
Failure to
preset dyno-
mometer load or
warm up dynano-
mometer Incor-
rect ambient
temperature
Reschedule
Knowledge of
dyno procedures
RUNNING LOSS
TEST
See
Table 3 . 4
Failure to
follow driving
cycle within
prescribed
tolerances. See
also Table 3.4
Reschedule
Trained driver
See Table 3,4
1 HOUR HOT
SOAK LOSS
Ambient temp.
record checked
by DV
Failure to
reconnect can-
isters after
"hot" start
test. Incorrect
soak temp.
Reschedule
evap. only.
Knowledge of
canister in-
stallation
CANISTER
WEIGHT
Data checked by
D.V.
Negative weight
gain is suspect.
Reschedule
using freshly
dried or new
canister
Knowledge of
balance
operation.
ro
                                                                                                                                                                               ^0  C3  JO  C/3
                                                                                                                                                                               Cu  Hi  fD  fD
                                                                                                                                                                              iQ  rt  <  o
                                                                                                                                                                               ro  it  H-  rt
                                                                                                                                                                                  ..  ui  p.
                                                                                                                                                                                     (-••  0
                                                                                                                                                                               VO     03
                                                                                                                                                                               0
                                                                                                                                                                               HI

                                                                                                                                                                                  C-i  O  OJ

-------
Section:   3(LG)
Revision:  0
Date:      June 1975
Page 10 of 13
  3.2.2     Exhaust Emission Measurement

            A summary of exhaust emission measurement procedures (FTP)  is
  shown in Table 3-4.  The overview represented by this matrix was designed
  to give a general understanding of the process involved in exhaust
  emission testing.  However, it is not our purpose to include every
  'detail required for this measurement.  The information discussed in the
  table consists of:

            o    A brief description of the tasks

            o    Applicable Federal Register paragraphs

            o    Applicable EPA, Ann Arbor, Test Procedure numbers

            o    Specifications and tolerances included in Federal
                 Register and from engineering practices

            o    Quality provisions

            o    Invalid tests  (determination)

            o    Corrective actions required

            o    Training and skill level required
                                 30

-------
                                                                   Table 3-4.   FEDERAL EMISSION TEST PROCEDURE

                                                                               Exhaust  Emission Test
UJ
H
PROCEDURE
ITEM
BRIEF
DESCRIPTION










FEDERAL
REGISTER
PARAGRAPH
EPA
PROCEDURE
NUMBER
SPECIFICATIONS
AND TOLERANCES
Federal
Register


























DYNO WARM-UP
AND HP SETTING
The vehicle is
placed on the
dynamometer
which has been
previously
wanned up and
the hp set





Appendix II
85.075-15


TP-604



Less than 2 hrs
between tests -
warm-up - 15
min & 30 MPH
within 1 hour
of test. Hp
setting - any
time prior to
test. For auto
1 hour prior
for manual. In-
flate tires to
45PSI. Use
vehicle re-
straint to
minimize rock-
ing.











CONSTANT VOLUME SAMPLER 

CALIBRATION:

The positive
displacement
pump is cali-
brated using a
laminar flow
element or
equivalent.



Appendix III



TP-201



See Appendix
III for equip-
ment toler-
ances. Measure
actual pump
cavity pres-
sure/tempera-
ture variation
during calibra-
tion ±2°F grad-
ual change.
Leak- free con-
nections.















OPERATION:

An integrated
portion of the
total exhaust-
air mix is
collected
during the
driving cycle
along with a
sample of dil-
ution air.
85. 075-20, -24



TP-706



CVS inlet
pressure less
than 1 in HO
Heat exchanger
±10 degrees of
set point temp
ace. ±2°F.
Flow rate 300-
350 cfffl. Dilu-
tion filters
consisting of
a charcoal
filter between
two particu-
late filters
Press, guage
±3 mm. Bag
sample flow
rate 10 cfh.
min. Specific
sampling pro-
cedure FR-24.






DRIVING
CYCLE
A driving cycle
typical of ur-
ban driving is
performed on
the dyno
according to
the FR driving
schedule.




Appendix I
85. 075-14, -15,
_19,-24

TP-706



Horsepower
setting - see
FR-15. Fan 18-
12 inches in
front or to
provide suffi-
cient cooling.
Driving trace
precision - 12
mph within 1
sec. Shift
points - see
FR-16-17. En-
gine shutdown
at 1369 sec-
conds . Time
between cold
and hot tests
10 ±1 minute
Engine starting
FR-19 Ambient
Temp 68-86°F






ANALYTICAL SYSTEM

CALIBRATION:

Primary gas
standards are
used to estab-
lish the in-
strument curve





85.075-23



TP-203



Calibration
performed every
30 days. Zero
gas impurity:
1 ppm HC
1 ppm CO
400 ppm CO
0.1 ppm NO
0 13-21 mole%
(AIR)
Calibration
Points :
HC & NO, 50 &
100%
CO & CO. -
10. 23, 40, 50,
60, 70, 80,
100% of full
scale. Analysis
of gas ±2% of
actual value.
Curve construc-
tion - best
judgement.
Analyzer warmup
- HC - 20 min.
CO, C0_, NO -
., L 2 x
2 hours
OPERATION:

The bag sam-
ples collected
by the CVS are
analyzed for
CO, CO,, HC and
NO . "
X



85. 075-23, -24



TP-707,711



Analysis per-
formed within
20 minutes from
end of sampling
Zero and span
instruments be-
fore and after
sample measure-
ment. Span gas
should have cone
of 80% of full
scale.
















DATA
COLLECTION
Ambient condi-
tions are re-
corded along
with instrument
outputs and
operating par-
ameters. Vehicle
and test cell
identification
and other per-
tinent informa-
tion.
85.075-22,25



TP-707



All information
is recorded ac-
cording to mea-
surement speci-
fications.























DATA
REDUCTION
The grams per
mile are calcu-
lated for each
component using
the formula in
the FR






85.07-26



TP-801



Reported to
three signi-
ficant figures
Density at 68
1 a tan.
HC 16.33
NO 54.16
COX32.97
C02 51.85



















                                                                                                                                                                                 V  O  50  to
                                                                                                                                                                                 pi  fa    fl>  H-  rt
                                                                                                                                                                                    ••  w  !•••
                                                                                                                                                                                 H     H-  0
                                                                                                                                                                                 *->     OS

                                                                                                                                                                                 o     H"
                                                                                                                                                                                 H)
                                                                                                                                                                                         *-•
                                                                                                                                                                                         IT"
                                                                                                                                                                                         O
                                                                                                                                                                                   <£>
                                                                                                                                                                                   -i
                                                                                                                                                                                   Ul

-------
Table  3-4.   FEDERAL EMISSION  TEST PROCEDURE

            Exhaust Emission Test

                 (Continued)
PROCEDURE
ITEM
SPECIFICATIONS
AND TOLERANCES
(Continued)
Engineering
Practice













QUALITY
CONTROL
PROVISIONS













DYNO WARM-UP
AND HP SETTING



Allowable
horsepower var-
iation less
than 10. 5 hp.











Calibration
performed
monthly with
weekly checks.
Correct setting
for vehicle
weight checked
by data valida-
tion (DV). Time
of previous
test run is
checked for
dyno warm-up
requirement.


CONSTANT VOLUME SAMPLER (CVS)



CALIBRATION:















Propane injec-
tion must agree
within ±2% of
calculated
value. Daily
propane injec-
tions plotted
on control
charts. Inter-
nal check of
calibration
data for uni-
formity.





OPERATION:
Tailpipe 15 in
HjO. Sample mix
temp, at pump
inlet 90-115 F
Heat exchanger
±5 F of set
point.
Dilution inlet
air 65 F min.
P. 70" HO max.
Bag construc-
tion 5 f ted-
lar film.


Weekly perfor-
mance checks
of equipment
Specifications
DV checks each
test for out-
of-control
operating
conditions.







DRIVING
CYCLE



Preprinted or
computer traced
driving sched-
ule. A minimum
of 12 hour and
maximum of 20
hour soak from
key off to key
on.






DV checks
speed, time,
trace, crank,
time, amb. temp
and all record-
ed information.
Daily span
check of driv-
ing aid. Driver
performance
audit.





ANALYTICAL SYSTEM




Calibration
points: CO, CO.
5 points & 0
across each
range . Curve
construction -
within 12% of
each point
value, smooth
curve passing
through zero
(origin) .
Weekly
calibration
check .
Calibration gas
analysis trace-
able to EPA
gravimetric
blends and/or
NBS-SRH's.
Inter-labora-
tory gas cross
check. Annual
restandardiza-
tion of gases.
Monthly instru-
ment perfor-
mance checks.
CVS gravimetric
injections.



Digital volt-
meter readings
of instrument
output record-
ed on chart.
Zero repeated
after each span
adjustment







Bags are leak
checked before
each test. NO
converter effi-
ciency check
performed daily
Analytical
system given
monthly per-
formance in-
spection and
preventative
maintenance
recorder
checked against
DVM each test.
DATA
COLLECTION


















DV inspects all
recorded infor-
mation for
spurious re-
sults and
facilitates
the smooth and
timely flow of
test documenta-
tion.






DATA
REDUCTION



NO and N02 re-
ported separate-
ly corrected and
uncorrected.











Data reduction
is usually per-
formed by com-
puter. Manual or
independent
check of the re-
duction program
should be per-
formed monthly &
whenever change
in program. Com-
puter output
checked by DV
for corrections.


                                                                                                         vQ
                                                                                                          CD
                                                                                                          N)

                                                                                                          O
                                                                                                          Ml
O
0)
                                                                                                                *
   w
   fl>
<  o
H- ft
CO  H-
H- O
                                                                                                             t4  O U>
                                                                                                             C    *-*

                                                                                                                  ^
                                                                                                             VO
                                                                                                             -J
                                                                                                             Ul

-------
                                                               Table  3-4.  FEDERAL  EMISSION TEST PROCEDURE

                                                                           Exhaust Emission Test

                                                                                 (Continued)
PROCEDURE
ITEM
TEST
INVALID









CORRECTIVE
ACTION






TRAINING
OR SKILL
REQUIRED





DYNO WARM-HP
AND HP SETTING
Failure to
warmup dyno.
Incorrect hp
setting for
weight. Vehicle
exhaust not
connected to
CVS.



Reschedule







Dyno operation







CONSTANT VOLUME SAMPLER (CVS)

Calibration in-
valid if pro-
pane injection
out of spec.







Repeat propane
injection. Re-
fer to Appendix
III for trouble
shooting.
Repeat calibra-
tion.

Special train-
ing in use of
calibration
equipment.
Experience in
emission
testing.

CVS flow rate
too low - in-
correct pump
speed used.
Equipment
failure or out
of spec.
Filters
plugged.


Reschedule







Special train-
ing in CVS op-
eration. Fami-
liar with
other test
equipment and
procedures.

DRIVING
CYCLE
Driver outside
specified lim-
its during
cycle . Improper
starting -
stalling proce-
dure . Out of
spec time
sequence. Soak
period to long
or short.
Reschedule
Driver perfor-
mance audit may
be necessary




Trained in
special driving
skills required





ANALYTICAL SYSTEM

Incorrect Stan-
dards or data
used to con-
struct curves.







Repeat calibra-
tion. Generate
new curve when
data points out
by more than
±0.5» deflec-
tion.

Special train-
ing in calibra-
tion procedure.
Previous exper-
ience as system
operator
desirable.

Leak in sample
bag detected
may invalidate
previous test.
Incorrect span
setting instru-
ment malfunc-
tion such as
span drift.


Repair or re-
place sample
bag. Reschedule
previous test.
Reschedule if
equipment fail-
occurs.

Training in
analytical sys-
tem operation.
Knowledge of
test procedures



DATA
COLLECTION
Incorrect data
or information









Correct infor-
mation when
possible. Re-
port all data
and information
errors. Resche-
dule if data is
not correctable
Data validation
should be
familiar with
test procedure,
basic statisti-
cal and techni-
cal knowledge
is desirable.
DATA
REDUCTION
Computer program
or data input
incorrect.








Correct program
or data input
and repeat cal-
culation.




Computer pro-
graming capa-
bility required
if done in-
house. Computer
operations
training.

UJ
                                                                                                                                                                          v  o jo w
                                                                                                                                                                          pJ  (U (D (D
                                                                                                                                                                         uQ  ft < O
                                                                                                                                                                          (D (D
                                                                                                                                                                               H- ft
                                                                                                                                                                               to H-
                                                                                                                                                                               P- 0
                                                                                                                                                                               OS
                                                                                                                                                                          O
                                                                                                                                                                          Hi
WC
   3
   n>
                                                                                                                                                                               O U>
                                                                                                                                                                                  —
                                                                                                                                                                                  f
                                                                                                                                                                                  o
                                                                                                                                                                            vo
                                                                                                                                                                            ~j
                                                                                                                                                                            Ul

-------

-------
                                                          Section:   4(LG)
                                                          Revision:  0
                                                          Date:      June 1975
                                                          Page  1 of 24
                             Section 4

               GUIDELINES FOR PERFORMANCE AUDITS AND
                      MAINTENANCE PROCEDURES
          Independent performance audits are conducted by a supervisor
or auditor to determine if the data collected is valid.  This is accom-
plished by defining system performance characteristics and acceptable
limits, and auditing to assure the instrumentation/equipment, and data
is acceptable. The occurrence of invalid data, non-acceptable instrumentation
and suspect values should be documented and corrective action initiated
to restore confidence in the system.
          Preventive maintenance routines can affect precision, accuracy
and reliability of a measurement system used in mobile source emission
monitoring.  Adequate routine preventive maintenance procedures will
minimize equipment failures.  Maintenance schedules should be related to
the purpose of the project, normal audit intervals, frequency of usage,
and frequency of failures.  Reporting procedures should include standard
checklists for ease of reporting and document control.
4.1       PERFORMANCE AUDITS

          Performance audits are those techniques implemented/used by
the Quality Assurance management to evaluate the total measurement
system.  These audits may involve the total system or specific portions
of the measurement system.  The techniques normally applied for perfor-
mance audits include the collection of normal operation data  (i.e., FTP
results, calibrations, etc.), replicate samples from the sampling or
analysis system and the subsequent plotting of the results on control
charts  (see Section 6.2 for methods of plotting control charts).  Fur-
ther audits may be made by the use of operational checks, visual checks
and standard reference samples.  The audit techniques should be applied
without the knowledge of the system operator/analyst, if possible, to
eliminate any bias, and to assure that the results are representative of
actual conditions.
          Performance audits generally are categorized as follows:

          o    Audit of the instrumentation
          o    Audit of operator's function
                               35

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page  2 of 24
          o    Audit of the sampling system
          o    Data processing audits

          These audits may be independent of or in conjunction with
normal quality checks.  Independence can be achieved by having an opera-
tor not normally assigned to the collection system in question perform
the audit.  An alternate method is to provide a standard reference
sample of unknown value (concentration)  to the operator and request that
he analyze and report the concentration values following normal operat-
ing procedures (Ref. 4-1.)
          In conducting the performance audit, various items should be
identified to assure all pertinent features of the system are identi-
fied.  The following criteria must be identified for each audited item
of the measurement system.

          o    Characteristics
          o    Acceptance criteria
          o    Frequency of checks
          o    Equipment used for checks
          o    Method (Procedure) used to perform check
          o    Corrective action requirements
          o    Recording of audit results

          Specific checks to be performed during a performance audit on
a random basis include the following:

          1.   Select data files at random  and check all forms and
               records.  Reduce all data manually to check its accuracy.
               Check data transfers and test vehicle information for
               correctness.

          2.   Observe the technicians performing the test.  Check that
               the forms are being properly filled out during the test.
               Check coriformance to test procedure methods.  Ask the
               technician questions concerning his job to determine that
               he possesses the required skill level.

          3.   Observe calibration procedures, check maintenance records,
               and all test log books.  Report all nonconforming conditions.

          4.   Check for proper data validation procedures and that
               proper authorization for the test sequence is being
               followed.
                                 36

-------
                                                           Section:   4(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page   3  of 24
          5.   Perform an exhaust emission test by operating the analy-
               tical console.  Check for system drift, noise, gain.
               Check the reproducibility of the span and sample.  Per-
               form a system leak check.

          6.   Independently monitor vehicle soak area for 48 hours,
               recording temperature and vehicle soak times.

          7.   Data collected from a correlation vehicle can provide an
               invaluable tool for audit purposes.  However, this method
               can be one of the most costly audit checks due to the
               measures taken to minimize the variables associated with
               the vehicle.
                    To minimize these variables, the selected vehicles
               should be transported to each participating laboratory.
               Statistical confidence levels should be established by
               testing each vehicle in each cell at least three or four
               times to establish a history of the normal expected
               variation.
                    Correlation vehicles vary in type, from a company
               vehicle selected for its low test-to-test variation to
               completely instrumented vehicles capable of measuring and
               recording torque, fuel flow, operating temperatures and
               operated by an automatic driver.  The latter example is
               the most costly and normally under controlled conditions,
               the most accurate audit method.

          When interlaboratory performance audits are being conducted
using a correlation vehicle, it is suggested that a cross-check be
performed by a laboratory independent of the one being audited.  This
will minimize errors due to a common inhouse problem.
          If these variables are minimized, a meaningful confirmation of
cell-to-cell compatibility may be accomplished.

4.1.1     System Performance Characteristics and
          Acceptable Limits

          Factors to be considered in the establishment of system quality
characteristics are quality of specifications, quality of conformance
and time-oriented factors such as availability, reliability and maintaina-
bility.  Specifications must have adequacy, attainability, and compatabi-
lity with the environment of the testing laboratory.  Quality of conformance
is the result of numerous variables:  test equipment, tools, supervision,
workmanship, etc.  Availability is measured by the extent to which a
user can obtain service when he wants it.  Reliability is classically
                                37

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 4   of 24
defined as "the probability of a system performing without failure, a
specified function, under given conditions, for a specified period of
time."  Maintainability is concerned with methods to improve the mainte-
nance of long-life systems, by planned preventive maintenance and unsche-
duled maintenance which consists of restoring service in the event of a
failure. (Ref. 4-2.)
          The total measurement system consists of the analytical method,
sampling method, operational conditions, the instrument or analyzer,
calibration, computation, data validation and the operator.  The criti-
cal characteristics of this complex system should be identified by
functional analysis and/or sensitivity analysis which are described in
Appendix K of Reference 6-1.  After identifying the critical character-
istics, the analysis should be extended to determine a means of control-
ling them and for detecting non-acceptable performance.
          Any feature (attribute, property, output, etc.) of the samp-
ling and/or analytical system which is required to achieve fitness for
use is normally classified as a quality characteristic.  For example,
any measurable or recordable output generated from the system or com-
ponent, such as flow rates, calibration data, concentration measurement
and gain settings exhibited by an analyzer are quality characteristics.
Each of these characteristics are subject to a performance audit
(Ref. 4-3).
          By classifying those numerous quality characteristics of the
system which may affect the precision and accuracy of system output, a
valuable tool for weighing the relative importance of system performance
is provided.  This  type of classification enables the quality effort to
concentrate on those characteristics which have a major effect on the
system output, thereby assuring quality and continued measurement at a
minimum quality cost.
          Each characteristic that is being subjected to a performance
audit has to be evaluated on the basis of the acceptance criteria for
that particular characteristic.  The development of these criteria
requires much discussion and common consent among parties of interest.
The acceptance criteria for a Light Duty Vehicle Emission Measurement
System are related to the requirements of the Federal Register (see
Table 4-1).  An expanded version of this table is shown as Table 3-2,
3-3, and 3-4.  These tables not only show the specifications and/or
tolerances required for the 1975 FTP, but they also outline the following;

          o    Engineering Practices
          o    Applicable EPA, Ann Arbor, Test Procedure numbers
          o    Quality Provisions
                                38

-------
   Table 4-1.   FEDERAL REGISTER SPECIFICATIONS

 FOR THE 1975  LIGHT-DOTY VEHICLE EMISSION TESTS
       Section:    4(LG)
       Revision:  0
       Date:       June 1975
       Page 5   of 24
REFERENCE
PARAGRAPH
85.002-a(20)
85.002-9(21)
85.075-7
85.075-7(0)
85.075-7
85.075-10(3)
(b)
85.075-ll(a)
(1)
85.075-12(c)
85. 075-12 (d)
85.075-13(3)
85.075-13(3)
85.075-13(a)
(1)
85.075-13
(a) t (d)
85. 075-13 (d)
(4)
85. 075-14 (b)
85. 075-15 (b)
85. 075-15 (d)
85.075-15(1)
85.075-16
85.075-17
85.075-18
85.075-19
PROCEDURE OR EQUIPMENT DESCRIPTION
Tank Fuel Volume
Zero Miles
Estimated Curb Height
Mileage Test Points
Emission Test Results
Gasoline 1. Evap. Testing
2. Exhaust Testing Only
3. Mileage Accumulation
Fitting and Tubing for Canisters
Ambient Temp 7.5 Mile Prep o Temp
o Speed Tolerance
Ambient Temp 1 Hour Soak
Ambient Temp 10 Hour Soak
Ambient Temp Fuel Tank Conditioning
Temperature Recorder
Fuel Temperatures o Charge
o Start
o End
Time between admittance to the 76 °F - 86°F soak
area and the placing of the vehicle on the dyna-
mometer for the running loss test.
Running loss and exhaust emission driving
schedule.
Cooling Fan o Engine in Front
o Engine in Rear
o Fan Capacity
Inertia Weights
Dyno prep manual controls o Idle
o Harm-up
o Hp setting
Dyno prep automatic controls o Hp setting
Three-speed manual transmissions. Shift points
when not specified by the manufacturer.
Four-speed and five-speed manual transmissions.
Shift points when not recommended by manufacturer.
Automatic transmission.
Engine starting and restarting:
SPECIFICATION
AND/OR TOLERANCE
40% of nominal capacity rounded to
the nearest gallon.
^.10 miles or 1 hour of operation.
+100 pounds measured curb weight.
+250 miles. May not exceed 3 valid
tests.
ASTM E29-67 (round off procedure) .
Reported as three significant
figures.
See reference paragraph (a) & (b) .
5/16 I.D.
77° +9°F (68-86)
+4 MP~H +1 Sec.
81° +5°F (76-86)
73° + 13°F (60-86)
81° + 5°F (76-86)
Chart Speed - 12 inches per hour.
50-60°F
60 +2°F
84 +2°F, 60 + 10 min.
Minimum of 1 hour.
Federal Driving Cycle 2 MPH within
1 Sec.
8-12 in. in front of the cooling
air inlets.
To provide sufficient cooling.
Not to exceed 5300 cfm.
See table of ref.
Less than 2 hours between tests
15 min. at 30 MPH
Within 1 hour of test
Any time prior to test.
First to second at 15 MPH.
Second to third at 25 MPH.
First to second at 15 MPH.
Second to third at 25 MPH.
Third to fourth at 40 MPH.
Decel in gear.

o 20 second idle period begins
o Place in gear            on
o Cranking time            Jy
o Stall
When engine starts.
After 15 seconds.
10 second mix.
Must restart within 1 minute.

-------
Section:     4(LG)
Revision:   0
Date:         June 1975
Page   6 of  24
          TahJe .4-1.   FEDERAL REGISTER SPECIFICATIONS

   FOR THE 1975 LIGHT-DUTY VEHICLE EMISSION TESTS (Continued)
       REFERENCE
       PARAGRAPH
     PROCEDURE OR EQUIPMENT DESCRIPTION
           SPECIFICATION
          AND/OR TOLERANCE
     85.075-20
                           Sampling and analytical system
                           Constant Volume Sampler
                              o CVS inlet filter pressure at the mixing
                                point
                              o Tail pipe pressure during test

                              o Heat exchanger
                              o Positive displacement pump capacity
                              o Pump inlet temperatures sensor accuracy
                              o Pump pressure gauge accuracy
                              o Sample collection flow rate
                           Exhaust gas analytical system
                           Wet CC>2 response for the CO-NDIR instrument
                                                        Schematic Drawing A7S

                                                        Less than 1 inch of water below
                                                        ambient.
                                                        +5 inches of water of the pressure
                                                        with no connection to the tail pipe

                                                        Preheat to within +10° of the oper-
                                                        ating temp.  Maintain +10° during
                                                        test.
                                                        300-350 cfm.

                                                        +2°F, continuous record
                                                        +3 mm Hg.
                                                        Minimum 10 cfh.
                                                        Schematic Drawing A-75-2.

                                                        It of full scale measured on the
                                                        most sensitive range or 3 ppm on
                                                        ranges below 300 ppm.
      85.075-21
                           Fuel evaporative sampling system

                              o Carbon canisters
                                   Capacity
                                   Length to diameter ratio
                                   Inlet and outlet tubes
                                   Pressure
                                   Charcoal content
                                   Conditioning temp
                              o Activated carbon meeting the following
                                specifications:
                                   Surface area minimum
                                   Absorption capacity minimum
                                      (carbon tetrachloride)
                                   Volatile material including adsorbed
                                      water vapor
                                   Screen analysis size
                               o Drying Tube
                               o Dryrite
                               o Sample collection  tubing
                               o Weighing equipment accuracy
                               o Weight of canister determined to
                                                        Schematics A-75-3. A-75-4 and
                                                        A-75-5.

                                                        Figure A-75-6.
                                                        300+ 2ml.
                                                        1.4*+ 0.1
                                                        5/16 I.D. x 1 inch
                                                        Leak tight at 2 PSI for 30 seconds.
                                                        150 + I" gm.
                                                        300°F for 3 hours.
                                                        N2 1000 sq.  meters per gram
                                                        60 percent by weight

                                                        None

                                                        Less than 1.4 mm         0%
                                                        1.7 - 2.4 mm         90% - 100%
                                                        More than 3.0 mm         0%
                                                        3/4 x 6 in.
                                                        8 mesh
                                                        Stainless steel or aluminum 5/16  I.D.
                                                        +75 Mg
                                                        Nearest 20 Mg

85.075-23
85.075-23 (a)
(2)
o Temperature measuring equipment
Analytical system calibration
Zero grade air
Dual Channel
Accuracy
Thermocouple
Every 30 days
HC less than 1 PPM
CO less than 1 PPM
50 -
Type


100°F
J


                                                                                    C02  less than 400 PPM
                                                                                    NO less than 0.1 PPM
                                                                                    O2 18-21 mole percent
      85.075-23  (a)
                (4)
Calibration points:
   o Hydrocarbon (FID)  and nitric oxide (NOX)
   o Carbon monoxide and carbon dioxide
   o Carbon dioxide
   o Calibration gas accuracy
                                                                                    50  and  100% of scale
                                                                                    10, 25,  40, 50. 60, 70% of scale
                                                                                    85  and  100.
                                                                                    +2%
      85.075-23  (a)
                (6)
                (7)
Sample conditioning efficiency check:
   o Frequency

   o Criteria for acceptable level of
     interference
Daily and consistent with observed
  column life
It of range or 3 PPM on ranges below
  300 PPM
      85.075-23  (b)
                           Analyzer warm-up
                                                                                   HC  -  20 min.
                                                                                   CO, CO2, NOX - 2 hours
      85.075-23  (b)
                (2)
Span gas concentration
Approximately 80% of full scale
      85.075-24  (b)
               (11)
               (12)

               (13)
               (16)
Analysis of sample gas

Engine abut-down after 'stabilized' cycle

Stop sampling
Time between cold and hot tests
As soon as possible and no longer
  than 20 minutes from end of test
2 seconds after the end of the last
  decelerations (1369 seconds)
5 seconds after engine shut-down
10 +1 minute
      Appendix  III
                           Equipment used for calibration of the CVS pump.
                           Register.                        1^0
                                                 See Table in Appendix III of Federal

-------
                                                           Section:    4(LG)
                                                           Revision:   0
                                                           Date:       June  1975
                                                           Page   7 of  24
          o    Corrective actions required
          o    Training and skill level required

          The economic impact of acceptance criteria should be taken
into consideration.  There have been situations in which an urge for
perfection has overshadowed the use of some equally applicable alterna-
tive criteria available at a considerably reduced cost.  For example, a
contractor specifies that a certain name brand instrument or analyzer be
used to fulfill a contract requirement.  In such cases, even though the
contracted laboratory had an instrument already in use, equivalent in
every respect, they would have to incur an extra cost despite the fact
that the data quality would not be increased.

4.1.2     Independent Performance Check Procedures

          The techniques employed in independent performance audit to
evaluate the quality of data produced by part of or the total measure-
ment system usually include the introduction of control samples into the
system.  The results are subsequently plotted on control charts and
evaluated.  A detailed discussion of types of control charts is contained
in Appendix H of Reference 6-1.  Various applications of these charts
are shown in Section 6.0.  These checks should be made independent of
the normal quality assurance checks.  A check could be made by a different
operator/analyst than the one normally involved in the measuring process.
A reference sample of a known concentration of pollutant could be sup-
plied to the operator/analyst with a request that he measure and report
its concentration  (preferably he would be unaware which is the reference
sample).
          There are a number of variables that can affect the expected
precision and accuracy of measurements made in the total system.  Some
of these are related to analysis uncertainties and others to instrument
characteristics.  Table 4-2 summarizes some of the more important vari-
ables and how they can be monitored.
          The recommendations from manufacturers of the various types of
instrumentation and equipment used in  the total measurement system
should provide an initial source of information on the methods and
frequencies of inspection.  These recommendations cannot always be
followed as specified due to the numerous sources of variability exhi-
bited by the system. Therefore, alternative methods of frequency deter-
mination must be considered.  A sensitivity  analysis  (discussed in   Appendix
K of Reference 6-1) can provide a basic insight into the frequency of
performance audits, and statistical sampling  techniques can be used  to
good advantage.

-------
                                                                                                                                         V O
                                                                                                                                         t» f»
                                                                                                                                         *Q ft
                                                                                                                                         o 0
                                                                                                                                         CO
                                                                                                                                         o
                                                                                                                                         M)
                                                                                                                                         to
                                                                                                            < o
                                                                                                            H- It
                                                                                                            M H-
                                                                                                            n
                                             Table 4-2.  EXAMPLE METHODS OF MONITORING VARIABLES
            VARIABLE
                                          MONITORING METHOD
                                           FREQUENCY OF CHECK
                                                                                                                  ACCEPTANCE CRITERIA
1. Calibration Gas Concentration
2. H20/CO2 Interference
3. Zero Drift
4. Span Drift
5. System Noise
6. Temperature Variation
Measurement of control samples as
a part of the independent audit-
ing program.
Checks performed on an audit
basis.
Zero check and adjustment prior
to each test period as part of
routine operating procedures.

Span check and adjustment prior
to and following each sampling
period.

Check the strip charts trace for
signs of noise during and after
each test period.
A thermometer or any other tem-
perature indicating device
placed near the analyzer or
sample system to monitor
unusual conditions.
Verify concentration when initially
purchased.  Audit concentrations
at monthly intervale and/or when
desired performance standards
cannot be met.

Perform check upon receipt of the
CO instrument to confirm that it
meets the specified acceptance
criteria.  Audit at periodic
intervals to assure performance
standards are met.

Perform as a routine operation and
periodically as a performance
audit.

Perform as a routine operation and
as a periodic performance audit
check.

Perform on a per test basis.
Check during performance audit.
Monitor trace as part of chart
recorder calibration.

Perform weekly/monthly as part of
a performance check.
                                                                                                                                           -j
                                                                                                                                           Ul
Zero gas within +0.1 deflection.
Span gas within +0.2 deflection.
Concentrations should be within
±2% of stated value.
See Federal Register, Volume 39,
No. 101, Thursday, May 23, 1974,
85.075-20(C)(11).
Drift should not exceed ±1% of
full scale.
Initial and post calibrations should
agree within .±1 deflection.
Noise should not exceed ±1% of  full
scale.
Should conform to  specifications as
indicated by manufacturer of each
instrument  checked.

-------
                                                            Section:    4(LG)
                                                            Revision:   0
                                                            Date:      June 1975
                                                            Page  9 of 24
4.1.3     Reporting and Corrective Action Procedures

          The value of any check is increased substantially if it can
help prevent repetition of some error which would not have met the
acceptance criteria.  It is essential that the laboratory quality assu-
rance program includes systematic procedures for recording and analyzing
check results, and the determination of the need and implementation of
corrective action.  Consideration should be given to the potential for
incorrect data entry each time a value is recorded, due to human error.
These kind of errors are sometimes difficult to detect, and the impor-
tance of accuracy in recording results must be continually stressed.
Control charts should be used, where appropriate, to monitor data quality.
Various statistical techniques, exemplified in Section 6, can be used to
analyze check results.
          When acceptance criteria are not met, the most effective means
of preventing further trouble is to implement corrective action to
eliminate the cause of nonconformance.  To maintain data quality at an
acceptable level, it is essential that the quality assurance system be
sensitive and timely in detecting out-of-control or unsatisfactory
conditions.  The basic steps in setting up a closed-loop corrective
action system are (Ref. 4-4):

          1.   Define the problem.

          2.   Assign responsibility for investigation of the problem.

          3.   Investigate and determine the cause of the problem.

          4.   Develop or determine a corrective action to eliminate the
               problem.

          5.   Assign responsibility for implementing the corrective
               action.

          6.   Establish effectivity and implement the correction.

          7.   Verify that the corrective action has eliminated the
               problem.

          The implementation of these steps often requires the coopera-
tion of many individuals and departments.  Corrective action requests
must be formally documented and reinforced with an effective follow-up
system to assure the closing of the loop.  Section 5.0 shows the typical
flow of the correction action loop with a specific example and forms
requirement.

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 10 of 24
4.2       PREVENTIVE MAINTENANCE

          Preventive maintenance, or the lack of it, can affect the
precision, accuracy and reliability (see Section 6.0)  of a measurement
system.  The concepts of preventive maintenance are not new; nor are they
confined to specific type of industry.   Preventive maintenance is as
practical to a laboratory situation as  it is to have a mechanic check
your car at regular intervals.
          The principle that underlies  preventive maintenance is that
every component or system has a basic engineered life.  They will be
clustered (predictably) if the product  is reliable and widely dispersed
for products of lesser quality which display erratic performance.  For
example, suppose the detector in an analyzer has a rated life of 2,000
hours.  If the design and quality of both the detector and the analyzer
in which it is used are superior, there may be very few failures before
2,000 hours and practically every detector will have to be replaced
before 2,500 hours of operation.  If the design is defective, however,
the failures will be spread over a longer period, starting within the
first hours of operation and continuing sporadically until the last
detector fails (Ref. 4-5.)
          Maintenance  factors will affect system reliability (see Sec-
tion 2.1.1) .  Some of  the factors are:

          a.   Training, experience and availability of instrument
               maintenance specialists.  Poor maintenance services will
               increase down-time and mean-time-between-fallures,
               increase costs and cause mistrust of data validity.

          b.   The physical conditions  under which maintenance tasks
               must be carried out can  affect reliability in a like
               manner, if work must be  done in extreme cold, heavy rain
               or snow or under inadequate lighting conditions.  Tied in
               with these is the presence of adverse factors such as
               lack of space, proper tools or supplies.

          c.   Responsibilities for various levels of repair and mainte-
               nance should be spelled  out so that no preventive mainte-
               nance task has been incorrectly assigned.

          d.   Control of spare parts should be exercised.

               1.   Inventory records should be kept to prevent stock-
                    outs and subsequent, increased down-time.

               2.   Policy should be established to control cannibali-
                    zation of parts when short-falls against requisi-
                    tions or purchase orders occur.

-------
                                                            Section:   4(LG)
                                                            Revision:   0
                                                            Date:      June 1975
                                                            Page 11 of 24
          e.    Scheduled coordination with calibration activities will
               save time and reduce maintenance down-time and equipment
               back-up requirements.

          f.    Repair reports such as operational logs (Figure 4-9)  and
               corrective action requests and recommendations should be
               sent through the quality assurance reporting loop.

4.2.1     Preventive Maintenance Procedures
          In order to minimize equipment degradation or failure,  scheduled
(i.e., periodic or routine)  preventive maintenance actions must be
performed.  The manufacturer's instrument/equipment manual is the logical
starting place for identifying systems or subsystems that require periodic
replacement or maintenance.   A more meaningful method of scheduling
maintenance is to plot past instrument/equipment performance on control
charts (Section 6.0)  and identify the optimum maintenance periods from
the frequency of malfunctions shown for each piece of equipment.
          Servicing and maintenance schedules should relate to the
purpose of testing, the environmental influences, the physical location
of the equipment/instrumentation, and the level of operator skills.  An
operational guideline showing time intervals for various types of service,
such as routine daily tasks and scheduled checks weekly, monthly, quarterly
and semi-anually must be developed from control charts and manufacturer's
recommendations to assure the quality of the total system.  Checklists
and station logs must be established and maintained routinely by the
system operator/ maintenance staff to record maintenance performed and to
insure that maintenance schedules have been met.  Service and maintenance
must be performed by personnel with the skill level required to assure
that efficient and effective repair/replacement is accomplished.  In
general, station operators should not attempt to perform more than
routine (daily) checks or diagnosis of a particular problem; they  (the
system operators) should definitely not attempt any repairs for which
they lack proper training or equipment, or for which the time required
would interfere with normal operations.
          An example of a routine, daily, preventive maintenance check
for mobile source emissions monitoring follows.  The items are arranged
in a systematic order for ease of checking by the auditor or operator.

Schedule for Daily Start-Up/Servicing  (General Guideline)

          o    Upon arrival start-up all instrumentation
          o    Check and record gain settings for FID, CO, CO  , NO^
               analyzers

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 12 of 24
          o    Check and record zero settings for FID,  CO,  CO ,  NO
               analyzers
          o    Check working gas cylinders for correct  pressures and
               record pressure
          o    Check recorders for zero,  gain and correlation
          o    Check all system pressures
          o    Check and record the NO analyzer for flow rate,  tempera-
               ture and reactor operating pressure
          o    Leak check by observing flowmeters and magnahelics and
               record results
          o    Check the CVS for the following and record:

               total counts for specified time period
               inlet pressure
               /\ pressure
               tail pipe pressure
               flexible exhaust tubing
               tailpipe adaptors
               sample collection bags (leak test)
               indicator lights
               sample pumps
               control switches
               temperature controller
               check filters for excessive carbon build-up

          o    Check driver's aid for paper and ink
          o    Record information on log sheet or log book

          Various items from this checklist can be usefully plotted on
control charts as shown in Section 6.0.
          Figure 4-1 shows a general form that may be used for routine,
daily, start-ups and preventive maintenance check sheets.
          A system of logs and check sheets  (such as those shown in
Section 5, and Section 4.2.2) to document that the required preventive
maintenance checks have been made and necessary work has been performed,
can be in the form of lab books or multiple copy forms.  Multiple copy
forms are an efficient means by which quality assurance can perform a
systematic review of maintenance accomplished during preventive  mainte-
nance periods.  A maintenance summary should be provided to outline
significant corrective maintenance.  It should include the replacement
of major components and required equipment changes.  This is normally an
inhouse change required to increase the system efficiency.  Analysis of
these reports will aid in developing a history of parts used, operations
performed and frequency of replacement, for use in determining optimum
parts replacement schedules, maintenance  schedules and optimum inventory
control.
                                 k6

-------
                                               DAILY START-UP  CHECKSHEET
DEPT NO

SHIFT

TRAIN

DATE

P. I.C.

                                                      CALIBRATION
HIGH
x7
CO
FIA
mx
co?
ZERO
RNG
750
0.3
100
300
IK
100
250
IK
4.0
X
GAIN









S
ZERO









^^
CYCL NO










RECORDER
CHART SP

ZERO

GAIN

DVM CORR

CONC










CEFL









S*
PRESS










^
TUNE
^^
^^
^^
^^
^^
^

^^

.s^
INTERMEDIATE
GAIN









^^
ZERO









^^
PRESSURE
^^
FIA
N0,

SAMPLE

^^^

FUEL

^^

AIR
^^


OZONE



CYL NO










BY PASS
FLOW RT


CONC










CONV
I NO TEMP
°C

DEFL










REACTOR
OPR PRESS
MM

PRESS










MFGRS MODEL NO










LEAK CHECK
^*" 	 " FIA CO C02 "°*
FL MTR OBS
MAG OBS

                                                            CVS
1 MIN COUNT   IN PRESS
T.P. PRESS
VOL/REV
FLEX
ADAP
BAGS
LIGHTS
PUMPS
SWITCHES
TEMP  CONTROL
COMMENTS
                                                                                                                               ^j o  po trt
                                                                                                                               JU pj  (0 r>
                                                Figure  4-1

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 14 of 24
4.2.2
Preventive Maintenance Action
          This section provides a system of check lists which outline
the preventive maintenance actions, frequency of performance, and indica-
tions of need for specific system components.
          The major system components of Mobile Source Emissions are
shown in Table 4-3.
          For each preventive maintenance period, the checks shown for
each type of equipment should be made to assure that the total system
retains the precision and accuracy required to produce acceptable data.
If a failure is discovered during any preventive maintenance period, a
failure report (Figure 5-13, Section 5.2.5) should be filed to document
the cause of failure, type of equipment, suggested corrective action and
final corrective actions taken.  An equipment repair authorization
(Figure 5-11, Section 5.2.3) should also be submitted to the particular
organization responsible for maintenance/repair, to document that
expedient repair or replacement was accomplished and that the costs
involved were recorded.
                            Table 4-3
MEASUREMENT SYSTEM
COMPONENT
CVS
CVS
Analysis System
(Includes HC, CO, CO ,
NO analyzers)
Dynamometer
(Includes speed and
torque meters)
Related Equipment
FREQUENCY
CHECK
Weekly
Monthly
Weekly
Monthly
Weekly
Monthly
Monthly
REFERENCE
CHECKLIST
Figure 4-2
Figure 4-3
Figure 4-4
Figure 4-5
Figure 4-6
Figure 4-7
Figure 4-8
(Ref. 4-6)
          Upon completion of the preventive maintenance checks, the
supervisor will perform an audit to assure the maintenance efficiency
and sign the checklists when satisfied with the results.

-------
                                                           Section:    4(LG)
                                                           Revision:   0
                                                           Date:      June 1975
                                                           Page 15 of 24
                PREVENTIVE MAINTENANCE CHECKLIST
 DEPT. NO.  	TRAIN	DATE	TECHNICIAN



EQUIPMENT:  Constant Volume Samplers

   PERIOD:  Weekly
               / /  Visual - Perform visual inspection on areas
                    of unit that are easily accessible.  (i.e.,
                    dirty lines, kinks,  electrical leads,  etc.)

               / /  Daily Log Book - Inspect daily log book  for
                    entries that might be pertinent in effecting
                    proper maintenance or repair.

               / /  Controller (100  F)  - Inspect  heat controller
                    and sensor indication to determine if proper
                    temperature of the heat exchanger  is  being
                    maintained.

               / /  Rustrak-Calib. - Inspect Rustrak temperature
                    recorder for correct operation.

               / /  Sample Lines (Leak)  - Inspect  all  sample
                    lines for leaks by the flow ball drawn down
                    method or by using a liquid test-detector on
                    positive pressure lines.
                         Supervisor Review_


                           Figure 4-2

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 16 of 24
                PREVENTIVE MAINTENANCE CHECKLIST
 DEPT. NO.	TRAIN	DATE	TECHNICIAN	



EQUIPMENT:  Constant Volume Sampler

   PERIOD:  Monthly


               / /  Visual - Perform visual inspection on areas
                    of unit that are easily accessible.

               / /  Daily Log Book - Inspect daily log book for
                    entries that might be pertinent in effecting
                    proper maintenance or repair.

               / /  Couplings - Inspect blower couplings for
                    security or excessive wear.

               / /  Water Leaks - Inspect all water lines and
                    fittings for leaks.

               / /  Water Pump Operation - Inspect water pump
                    for vibration, excessive noise, improper
                    water flow and leaks.

               / /  Sample Lines - Inspect all sample lines for
                    flow, leaks, excessive dirt, and wear.

               / /  Transmission Oil - Check level of transmission
                    oil.  Fill to oil level line if necessary.

               / /  Lubricate - Lubricate all the fittings for
                    which the CVS maintenance manual calls.

               / /  Flush Water System - Flush entire water system
                    with acidic solution to dissolve mineral
                    deposits on valves and heat exchanger tubes.

Page  1 of 2


                           Figure 4-3
                                 50

-------
                                                           Section:   4(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 17 of 24
                PREVENTIVE MAINTENANCE  CHECKLIST
 DEPT. NO.	TRAIN	DATE	TECHNICIAN



EQUIPMENT:   Constant Volume Sampler

   PERIOD:   Monthly (Continued)
               / /  Inlet Depressions Variation - If the inlet
                    depression varies from the normal by more
                    than 2 inches of oil,  heat exchanger should
                    be disassembled and cleaned.

               / /  Vertical Manometer Inspection - Inspection main-
                    tenance log for most recent comparison check
                    to master.  This check should be performed
                    every 90 days.

               / /  Dilution Box Filter and Pressure Check -
                    Measure negative pressure from atmosphere
                    at pressure tap on filter box.

               / /  Inspect appearance of filters.  If pressure
                    exceeds one  (1) inch of water or if filters
                    appear dirty, replace.
                         Supervisor Review_

 Page 2 of 2

                     Figure  4-3  (Continued)
                                  51

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 18 of 24
                PREVENTIVE MAINTENANCE CHECKLIST
  DEPT.  NO. 	TRAIN	DATE	TECHNICIAN



 EQUIPMENT:   Analysis System

    PERIOD:   Weekly
               / /  Visual  - Perform visual inspection of areas
                    of unit that are easily accessible.

               / /  Daily Log Book - Inspect daily log book
                    for  entries that might be pertinent in
                    effecting proper maintenance or repair.

               / /  Calibration Curve  (2 Points) - Pass a high
                    and  low standard gas through each analyzer
                    after making a set point using a standard
                    reference gas on the high end of the range.
                    If either point is off ±2% or more, inves-
                    tigate  further by  running complete curve.

               / /  NO   Converter - Perform converter efficiency
                    check.  See Test Procedure TP-303.

               / /  FID  Burner Peak -  See instrument manual of
                    specific manufacturer.

               / /  Oil  Level  (CL Pump)  - Check oil level in
                    vacuum  pump of the chemiluminescent
                    analyzer.  Refill  as necessary.
                          Supervisor Review


                            Figure  4-4



                                  52

-------
                                                           Section:   4(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 19 of 24
                PREVENTIVE MAINTENANCE CHECKLIST
 DEPT. NO.	TRAIN	DATE	TECHNICIAN



EQUIPMENT:   Analysis System

   PERIOD:   Monthly
               / /  Visual - Perform visual inspection on  areas
                    of unit that are easily accessible.

               / /  Daily Log Book - Inspect daily log book
                    for entries that might be pertinent in
                    effecting proper.maintenance or repair.

               / /  CL Pump (Change Oil)  - Change the  oil  in
                    the vacuum pump of the chemiluminescent
                    analyzer.

               / /  Perform complete monthly calibration on
                    all analyzers.  See Test Procedure TP-203.
                         Supervisor Review
                           Figure 4-5
                                53

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 20 of 24
                PREVENTIVE MAINTENANCE CHECKLIST
 DEPT. NO.	TRAIN	DATE	TECHNICIAN



EQUIPMENT:  Clayton Dynamometer

   PERIOD:  Weekly
               / /  Visual - perform visual inspection on areas
                    of unit that are easily accessible.

               / /  Daily Log Book - Inspect daily log book for
                    entries that might be pertinent in effecting
                    proper maintenance or repair.

               / /  Meter Operation - Visually inspect operation
                    for meter under normal testing conditions.
                    Observe for intermittent operation or no
                    movement at all.

               / /  Speed Calibration - See the Detailed
                    Test Procedures TP-202.

               / /  Coast Down (2 points) - See the Detailed
                    Test Procedures TP-302.

               / /  Noise - Listen for unusual noise from the
                    roll assembly or the inertial assembly
                    during normal testing conditions.
                         Supervisor Review


                           Figure 4-6

-------
                                                           Section:   4(LG)
                                                           Revision:  0
                                                           Date:      June  1975
                                                           Page 21 of 24
                PREVENTIVE MAINTENANCE CHECKLIST
 DEPT.  NO.	TRAIN	DATE	TECHNICIAN



EQUIPMENT:   Clayton Dynamometer

   PERIOD:   Monthly
               / /  Visual - Perform visual  inspection on areas
                    of unit that are easily  accessible.

               / /  Daily Log Book - Inspect daily log book for
                    entries that might be pertinent in effecting
                    proper maintenance and repair.

               / /  Adjust Roll Brake - Check and adjust,  if
                    necessary, the roll brake as described in
                    the manufacturer's maintenance manual.

               / /  Lubricate - Lubricate points on the dynamom-
                    eter as indicated in the manufacturer's
                    maintenance manual.

               / /  Magnesium Plug Check - Check Magnesium
                    plug for deterioration or leaks.  Replace
                    if 50% used.

               / /  Perform complete monthly calibration;
                    include speed and torque meters.  Test
                    Procedure TP-202.

               / /  Check hoses and connections for possible
                    leaks during operation.

               / /  Clean Water Strainer

                         o    Turn off water supply
                         o    Remove screen
                         o    Clean screen with compressed air
                         o    Replace screen and turn on water
                         Supervisor Review
                                 55
                           Figure 4-7

-------
Section:   4(LG)
Revision:  0
Date:      June 1975
Page 22 of 24
                 PREVENTIVE MAINTENANCE CHECKLIST
  DEPT.  NO.	TRAIN	DATE	TECHNICIAN



 EQUIPMENT:   Individual  Instruments

    PERIOD:   Monthly


               Check calibration  tag  on each of the following
               instruments  for calibration due date.  Submit
               a  job request  (Section 5.2.3) to the proper
               service  group  listing  those due for calibration.

               /  / Barometer (Test Procedure TP-206)

               /  / Digital Balance

               /  / Driver's  Aid  Recorder

               /  / Hygrometer

               /  / MV  Recorder

               /  / Manometers  (accuracy  check)
                         Supervisor Review


                           Figure 4-8


                                 56

-------
                                                            Section:    4(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  23 of  24
4.2.3     Maintenance Log Procedures

          Any maintenance (whether preventive,  routine or emergency)
performed on the equipment/instruments required in the measurement of
light duty vehicle emissions should be directly recorded into a mainte-
nance log.
          The maintenance log,  when properly used and maintained pro-
vides a valuable tool for documenting equipment breakdown histories,  a
guide for maintenance scheduling,  and a handy reference for trouble-
shooting problems.  Therefore,  it  is imperative that some sort of
maintenance recording procedure is developed and used regularly.
          Figure 4-9 is an example of a three-part maintenance log form.
          The following information should be reported on the form.

          1.   Responsible department

          2.   System number (for  multi-system facilities)

          3.   Date of equipment failure

          4 .   Person in charge-technician

          5.   Description of equipment and model
               (ex:  AIA-1, CO analyzer)

          6.   Equipment manufacturer (ex:  Horiba)

          7.   Serial number of equipment

          8.   Time of reported equipment failure

          9.   Summary of problem; any noticeable discrepancies from
               normal operating mode

         10.   Corrective action;  include steps taken to repair equip-
               ment, parts replaced, and equipment used for repair, or
               other action taken to preclude recurrence

         11.   Time and date of effective equipment repair

         12.   Signature of person performing repair

         13.   Signature of Supervisor responsible for equipment operation.

-------
                                        MAINTENANCE  REPORT
oo
                DEPT.  NO.
TRAIN
            EQUIPMENT DESCRIPTION; 5
            SERIAL NO.; 7
            SUMMARY OF PROBLEM: 9
            CORRECTIVE ACTION; 10
            TIME/DATE EQUIPMENT ON LINE;11
DATE
        WHITE/FILE   CANARY/QUALITY CONTROL   PINK/MANAGER
P.I.C,
                           MANUFACTURER:6
                           TIME OF  FAILURE: 8
                           CORRECTIVE ACTION PERFORMED BY; 12
                           SUPERVISOR   13
                                                                                                                 tn
13 O » W
PJ P) Q Q

13 rt <3 o
(P (P H- rt
  .. (o p.
                                           o
                                           H|
                                              0§
                                              3 ..
                                                                                                             04 O
                                                                                                             -J
                                                                                                             01
                                             Figure 4-9

-------

-------
                                                           Section:    5(LG)
                                                           Revision:   0
                                                           Date:       June  1975
                                                           Page  1  of  31
                             Section 5

                 QUALITY ASSURANCE GUIDELINES FOR
              DOCUMENTATION OF THE MEASUREMENT SYSTEM
          The responsibilities of Quality Management have been outlined
in Section 2; however, the implementation of the Quality Management
function depends upon the documentation of specific quality responsi-
bilities, departmental procedures and the interrelationship of each to
Quality Management.  The mechanism usually chosen to accomplish this is
the development of a Quality Management Procedures Manual.  Associated
with this manual are the separately documented test or operating pro-
cedures used by the various departments for performing their basic
functions such as specific emission test procedures, calibration, main-
tenance, and data analysis.
5.1       DEVELOPMENT OF AN OPERATIONS MANUAL

          Mobile source testing facilities generally incorporate some
elements of quality planning either formal or informal, into their
testing operations.  The end product of these facilities is test data,
and it is essential that authoritative information and control be imple-
mented to assure that the data produced is accurate and reliable.
Information and control can be obtained through the use of formalized
quality planning in the form of Quality Management procedures that
provide for documentation as objective evidence of information and
control.
          The Quality Management Procedures Manual specifies the neces-
sary paper work system for the documentation of the various quality
functions.  A smooth flow of data greatly enhances the auditing portion
of the Quality Assurance system.  A network of forms to be used in data
recording and reporting should be developed along with specific forms
instructions and processing procedures.  Establishment of a closed loop
corrective action process relies on the documentation and distribution
of the results of receiving inspections, audits, calibrations, etc.
          This report for Phase I of this program has been prepared in
two volumes.  Volume I contains the Quality Assurance Guidelines for
Light-Duty Mobile Emissions Measurement Systems.  The Quality Management
                               59

-------
Section:   5(LG)
Revision:  0
Date:      June 1975
Page 2 of 31
Procedures are included as Appendix C to this volume.  Volume II of the
report contains a Test Procedures Manual prepared for the EPA Laboratory
in Ann Arbor, Michigan.  This manual contains:

          o    Step by step testing procedures for direct use by tech-
               nicians performing the various portions of the test,
               inspection, calibrations and analysis.

          o    Standard data sheets and forms for use in recording and
               handling operational, inspection, calibration and analy-
               tical data and computational processes.

          Discussion of the guidelines for the preparation of a Procedures/
Operations Manual follows.

5.1.1     Document/Manual Control

          The responsibility and procedure for the implementation,
preparation, numbering and revision of Quality Management and Test
procedures and forms used in the measurement system must be clearly
defined.  Usually this is a function shared by Quality Management and
Administration Services.  Response to the changing requirements of the
measurement system is of utmost importance.  Timely reporting of change
notices, review of revisions, maintenance bulletins, etc., will prevent
the forms and test procedures manual from becoming obsolete.  The effec-
tiveness of document control may be directly judged by the universal use
of the forms and the consideration of the Quality Management and Test
Procedures manuals as worthwhile references.
          In addition, a master file of all procedures and subsequent
revisions showing effective dates and cross indexed for ease of refer-
ence should be maintained.  Responsibility for the actual revision of
the distributed manuals should be defined and manuals should be audited
on a random basis to determine compliance.

5.1.2     Quality Management Procedures

          The Quality Management Procedures Manual included as Appendix C
to Volume I of this report divides each department into various functional
units.  Specific operational functions, authorities, and responsibilities
are outlined.  In addition, the Quality Assurance provisions are assigned
and the interrelationship with other departments are defined.
          Specific management procedures are detailed for each function,
reflecting the organizational policy on the functional aspects of a
Quality Assurance program.  Other Quality Management procedures provide
the instructions required to implement a Quality Assurance program,
defining the purpose and procedure for implementing the policy, in-
cluding the assignment of functional responsibility.  The procedures are
                                60

-------
                                                            Section:    5(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  3  of  31
usually prepared and administered by the Quality Assurance Management
with the direct approval of the Laboratory Director/Manager.

5.1.3     Testing Procedures

          The only available published document outlining the testing
procedures to be used to measure the emissions from mobile sources is
the Federal Register (Reference Section 3.1).
          It is necessary that the Test Procedures be detailed and
developed in a logical sequence.  They should cover all phases of the
actual procedures performed in conducting an emission test and in cali-
brating and maintaining the test equipment.  The scope of the test
procedures manual(s) will be determined by the complexity of the equip-
ment used, the skill level of the people performing the procedures, the
number, size and location of the testing units and varied kinds of
testing performed in the facility.  As a general guide any procedure
performed as a matter of routine or on a periodic basis should be
documented.
          The Test Procedures contained in Volume II of this report have
been written in a standardized 13-point format.

          1.   Purpose - The reason or objective of performing the test
               is briefly described.

          2.   Test Article Description - This is a brief description of
               what is being tested, analyzed, calibrated, etc.

          3.   References - The Federal Register paragraphs, SAE Proce-
               dure, Manual or other documents that were the original
               source of the procedure are referenced along with litera-
               ture references which give additional background informa-
               tion on the procedure.

          4.   Required Equipment - Lists the necessary equipment includ-
               ing model number, manufacturer and other pertinent
               information.

          5.   Precautions - Lists safety precautions and points out
               certain procedures that are critical and require special
               attention.  Although specific safety precautions are
               documented in this section a general safety program is
               required by OSHA, especially in larger organizations, and
               is usually maintained as a separate manual.
                                 61

-------
Section:   5(LG)
Revision:  0
Date:      June 1975
Page 4 of 31
          6.   Visual Inspection - A check of the equipment, hook ups
               and general configuration of the equipment.  For example:
               a vacuum line disconnected on emission control equipment
               would have to be connected.

          7.   Test Article Preparation - Those steps performed immedi-
               ately prior to the actual test performance.  They may be
               referenced to a prior procedure or, if simple preparation,
               be outlined in this section.

          8.   Test Procedure - A numbered sequential step-by-step
               procedure used to accomplish the objective stated in
               Paragraph 1 above.  The points in the sequence where an
               entry or data output are required are described and noted
               in the right hand margin.

          9.   Data Input - A description of the information and data
               obtained during the test and the manner  in which it
               should be treated, stored or computed.

          10.  Data Analysis - A description of the data validation
               procedure used and any  subsequent statistical treatment
               to assure that it is within acceptable limits, complete,
               accurate and reliable.

          11.  Data Output - Descriptions of the data reporting and
               filing procedure, also  if applicable, examples of the
               computer output format.

          12.  Acceptance Criteria - A list of predetermined criteria
               which comprise a valid  test and are used in Paragraph 10
               for data analysis.

          13.  Quality Provisions - A  description of checks, calibra-
               tions, inspections, witnesses, specification, duplicate
               sampling, etc., specifically incorporated into the test
               procedure for controlling the quality of the data.

          A facsimile of the form used to document the  results of the
 test is included or referenced.  This  form should also  be  referenced  in
 Test Procedure, paragraph 8 above.
          This format is not the only  one which could be  selected.
 There are many acceptable methods for  writing  laboratory procedures;
 however, whatever format is selected should be used  consistently.  The
 format should be designed to facilitate change, clearly define objectives,
 and specify the quality acceptance provisions of the test procedure.
                                  62

-------
                                                          Section:   5(LG)
                                                          Revision:  0
                                                          Date:      June 1975
                                                          Page  5 of  31
5.1.4     Related Information

          Other sections as determined by individual needs may be added
to a Procedures Manual, such as a separate section on maintenance, cate-
gorized by equipment requiring the use of specific procedures not docu-
mented previously, particularly when a separate maintenance manual has
not been prepared.  These could be issued periodically as bulletins.
          Special test procedures may also be included to cover interim
modifications not requiring procedure change, or special contract
requirements for a single program.
          A glossary of terms and special sections on theory of opera-
tions of the equipment are sometimes included.
          In preparing a procedures manual for the first time it is best
to follow the rule of "keeping it simple".  Complexity and additions will
come with use, as the needs are identified through audit and review by
Quality Management.
5.2       DOCUMENTATION REQUIREMENTS OF A QUALITY ASSURANCE
          SYSTEM

          The most convenient and systematic way of developing a
Quality Assurance Plan or in summarizing and reviewing an existing
Quality Assurance program is to prepare a Plan Activity Summary
Matrix for each major activity or operation.  This matrix will include
the documentation requirements of the Quality Assurance system.
          For mobile source emission testing these major activities or
operations are:

          o    Procurement  (ordering)

          o    Procurement  (receiving)

          o    Gas Blending

          o    Calibration

          o    Verification and Correlation

          o    Test Operations

                    Vehicle Acceptance  and  Inspection
                    Vehicle Preconditioning
                    Evaporative Preparation
                    Diurnal Evaporative Test
                    Emissions Test
                    Highway Fuel  Economy Test

-------
Section:   5(LG)
Revision:  0
Date:      June 1975
Page 6 of 31
          o    Data Reduction and Validation

          o    Preventive Maintenance

          o    Auditing

          For each of the above activities the following items should be
considered:

          o    Characteristics checked
          o    Acceptance Limits
          o    Frequency of checking
          o    Equipment/Methods or Procedure used in checking
          o    Action if Acceptance Limits are not met
          o    Method of recording results of checks.

          A partial application of a Plan Activity Summary Matrix for
Procurement (receiving) is given in Section 5.2.1.  Similar tabular
summary matrixes should be prepared and kept up-to-date by the Quality
Assurance function for each test facility.  Only through the preparation
of such tabular summaries can the total "picture" of all quality checks be
seen.  These tabular summaries would consolidate all quality checks in
one place, including the quality  assurance provisions listed in all the
Test Procedures contained in Volume II of this report.
          As previously mentioned, the development of standard forms,
graphs, checksheets, etc., are necessary in a Quality Assurance System
for ensuring the completeness and traceability of data and information,
for facilitating validation and audit and for a systematic flow of
information throughout the system.
          In addition to simply recording and calculating data obtained
in the performance of a test, other items of documentation are required
for building reliability into the system, such as:

          o    Recording inspection and check results
          o    Recording calibration results
          o    Recording preventive maintenance actions
          o    Reporting unacceptable results
          o    Reporting failures
          o    Initiating and assuring closed-loop corrective actions
          o    Recording audit results
          o    Initiating procedural or equipment change notices
                                6k

-------
                                                            Section:    5(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page 7  of  31

          It is important that a form developed to accomplish a certain
quality function be carefully designed and be self-sufficient.  Manage-
ment should not allow a quality system to dead-end due to an incomplete
system of follow-up and distribution of the forms.  The following dis-
cussion and sample forms pertain to typical forms used in a quality
system.  These were designed as guidelines and may not satisfy all of an
individual laboratory's requirement.  A matrix of all forms exhibited in
Volume I and their location within the report is shown in Table 5-1.

5.2.1     Recording Inspection Results

          When recording inspection results it is important to record the
results, when the inspection was performed, how it was performed and
whether the test had an acceptable level of performance.  Many of the
Quality Control checks performed in conducting an emission test are
recorded as part of the test data and do not require separate forms, but
they need to be identified as quality checks.  However, other inspections
and checks not directly performed in conducting a test should be docu-
mented on a separate form.  As an example of an inspection form, Figure 5-1
shows a typical Receiving Inspection Form.  Figure 5-2 outlines the form
instructions which usually are printed on the back of the form.  In
addition to this form a reference matrix document is required, issued by
Procurement Control or Quality Engineering, outlining the inspection
procedures to be used for checking or ^-inspecting the material.  The
material purchased should have an identifying code number indicated on
the purchase order which corresponds with the item number on the plan
activity matrix. (Figure 5-3)
          The information contained in these forms should be logged in an
information file to establish a history which can be used for statistical
analysis such as the construction of control charts, for supplier ratings
and other purchase review requirements.

5.2.2     Recording Calibration Results

          Documentation of instrument or equipment calibration requires
the recording of the calibration data or set point in some chronological
form.  These calibrations should be performed on a periodic basis and the
equipment tagged to indicate the last calibration, status of the in-
strument, and calibration due date.  Different colored tags may be used
for example, white for calibration, yellow for instruments with limited
use, i.e., if only a single range has been calibrated, and red for inac-
tive instruments which require re-calibration before use.  Figure 5-4
shows  some examples of calibration tags and a rejection tag; Figure 5-5
shows  a calibration control card.  These cards can be processed by the
computer to show that periodic calibration has been done.  Each time the
calibration is performed a new card is issued showing the next due date
and remains with the instrument.  This eliminates the need for manual
audit.

-------
     Section:   5(LG)
     Revision:  0
     Date:      June 1975
     Page 8 of 31
            Table 5-1.   SUMMARY OF FORMS  REFERENCED IN VOLUME I
                                                      SECTION REFERENCE
         TITLE OF FORM
Daily Start-up Checksheet

Preventive Maintenance Checklist,  CVS, Weekly

Preventive Maintenance Checklist,  CVS, Monthly

Preventive Maintenance Checklist,  Analysis
  System, Weekly

Preventive Maintenance Checklist,  Analysis
  System, Monthly

Preventive Maintenance Checklist,
  Dynamometer, Weekly

Preventive Maintenance Checklist,
  Dynamometer, Monthly

Preventive Maintenance Checklist,
  Individual Instruments, Monthly

Maintenance Log Form

Receiving Inspection Report

Calibration Tags

Calibration Control Punch Card

Calibration History Evaluation

CVS Calibration Sheet

Analyzer Curve Generation Data

Monthly Dyno Calibration Log

Gas Analysis Report

Equipment Repair Authorization
Section 4, Figure 4-1

Section 4, Figure 4-2

Section 4, Figure 4-3


Section 4, Figure 4-4


Section 4, Figure 4-5


Section 4, Figure 4-6


Section 4, Figure 4-7
Section 4,

Section 4,

Section 5,

Section 5,

Section 5,

Section 5,

Section 5,

Section 5,

Section 5,

Section 5,

Section 5,
Figure 4-8

Figure 4-9

Figure 5-1

Figure 5-4

Figure 5-5

Figure 5-6

Figure 5-7

Figure 5-8

Figure 5-9

Figure 5-10

Figure 5-11
                                      66

-------
                                                                 Section:   5(LG)
                                                                 Revision:  0
                                                                 Date:      June 1975
                                                                 Page 9 of 31
       Table 5-1.   SUMMARY OF FORMS REFERENCED IN VOLUME I (Cont'd.)
         TITLE OF FORM
Rejection Report

Failure Analysis Report

Corrective Action Request

Performance Audit Summary Sheet

Procedure/Equipment Configuration Change
  Notice
     SECTION REFERENCE
Section 5, Figure 5-12

Section 5, Figure 5-13

Section 5, Figure 5-14

Section 5, Figure 5-16


Section 5, Figure 5-17

-------
Section:   5(LG)
Revision:  0
Date:      June  1975
page 10 of 31        RECEIVING  INSPECTION  REPORT
                    INVOICE  NO.
1.
2.
3.
k.
5.
6.
7.
9.
10.
12.
13.
Ut.
15.
16.
17.
18.
Rerpivpd From


Shipping Pamage

Received By
No. Pkg./Weiqht
PARTIAL Q] COMPLETE
Purchase Order No. 8. For Department
Shipped to At"tPnt"inn of
Packing Slip No. 11. UnparkpH By,













Invoice — Packing Slip— Purchase Order checked for correct count
and Material Part No.
Final inspprfion frt hp ropiplpt^d hy
Spnt for final in«;pprfion D^t**

MATERIAL INSPECTION REPORT
Inspected By Dept.




Q.C. Inspection Plan Procedure No.


Characteristics Acceptable Actual Measured
Checked Quality Level Conformance



Disposition of Material: ACCEPTED | | SEND TO
HOLD FOR ORDER COMPLETION | | SEND TO STORES



JSER Q
REJECTED | |
Reason Rei. Reoort No.



Distribution: 1. Purchasing 2. Requestor 3.
k. Procurement Control
Figure 5-1
68

Receiving File

-------
                                                  Section:   5(LG)
                                                  Revision:  0
                                                  Date:     June 197'
                                                  Page 11 of 31

       Instruction for Receiving Inspection Report

    (Printed on back of receiving inspection report)


1.   Print name of supplier and address of shipping point.

2.   Method of shipment and name of carrier.

3.   Record damage to shipping container or any other  visual
    damage observed.

4.   Signature of receiver.

5.   Record number of packages and total weight.

6.   Check one.

7.   Record purchase order number, if not available notify
    purchasing.

8.   Department originating order.

9.   Department or person requesting material.

10. File packing slip with receiving copy.

11. Person unpacking crate.

12. Compare documents for correct count and part numbers and
    other information on purchase order.  Record discrepancies
    and report to purchasing and procurement control.

13. Division or group responsible for receiving inspection.
    Determine from purchase order.

14. Date sent to inspector.

15.  Name and department of inspector.

16. File reference for quality planning procedure to  be used
    (see Figure 5-3).  Numbers should appear on purchase
    order.  Inspection procedure reference is contained in
    inspection procedure manual.

17. Characteristic and AQL recorded and results recorded.

18. Check appropriate boxes and give reason for rejected
    material, and rejection report number, if applicable.
                       Figure 5-2
                           69

-------
                                    PLAN ACTIVITY MATRIX - PROCUREMENT  (RECEIVING)
CHARACTERISTIC (1)
1. SPAN GAS-Concentration
of CO







2. GASOLINE-INDOLENE 30
Reid Vapor Pressure
(R.V.P.)

Lead Content

ACCEPTABLE
LIMITS (2)
Cone . Range
2350-2650
Analysis
Tolerance
±2%





8.7-9.2 R.V.P.


1 .4 Gram Min.

FREQUENCY
OP CHECK OR
MEASUREMENT (3)
Sample Each
Batch Re-
ceived







Sample Each
Batch Re-
ceived
it

METHOD OF
MEASUREMENT (4)
NDIR Gas
Comparator








Reid Bomb


Flame Photo-
meter
ACTION IF
REQUIREMENTS
NOT MET (5)
Out of Range.
Return to
Supplier
Analysis out
of Spec.
Label cylin-
der with
correct
analysis

Return to
Supplier

ii

RECORD OF
CHECKS (6)
Procurement •
Log Book








Procurement
Log Book

»

                                                                                                                v a
                                                                                                                0) fU
    50 W
..n>m
UJ ft < O
CD fl>
                                                                                                                N)
                                                                                                                U>
    H- ft
    0) H-
    H- 0

    § 5
                                                                                                                    O I/I
                                                                                                                  ID
                                                                                                                  -J
                                                                                                                  Ul
(1)  A list of the characteristics to be checked.

(2)  A list of the acceptable level of quality requirements  established by Quality Planning  (QA).

(3)  A description of the frequency of checking each  characteristic.

(4)  A brief description of the method, equipment or  reference standards to be used for

    checking each characteristic.
(5)  Directions for the inspector to follow if the  characteristic does not comply with

    acceptance limits.
(6)  A description of the type of record in which the accept/reject data is to reported.
                                                 FIGURE  5-3

-------
                                                   Section:   5(LG)
                                                   Revision:  0
                                                   Date:      June  1975
                                                   Page 13 of 31
                   Figure  5-4

                  CALIBRATION TAGS
CALIBRATION
Dot*
Ou.

       INACTIVE
Date
Must  be re-calibrated
prior to use
    LIMITED  USE

Date
Limitation
By order of
Expires
THERMOCOUPLE
CALIB/USE RECORD
WIRE: TYPE GAGE
INSUL LGTH
CERT CORR
TECH DATE
PURPOSE

REPLACE
USE NO.
1
2
3
4
5
6
7
8
9
10
DATE










BY










         GAGE AND INSTRUMENT REJECTION TAG
                                               No. 06738
        GAGE OR INSTRUMENT NOMENCLATURE
        MANUFACTURER
                                            IDENTIFICATION NO.
        REJECTED BY
                                                 ORGN. REJ. FROM
                            REASON FOR REJECTION
                 Figure 5-U.  CALIBRATION TAGS
                             71

-------
(D

Ul

Ul
       (D


,1  I  I  I  I   I
      CONTROL NO.
               I  I  I
        (4)
         MODEL
  CONTROL NO.

immm
191199)999
1 '  » 4 I I I I I II

                           (2)

                  I  I  I   I  I  I  I  I  I  I
           (5)
                       TYPE
(6)
                              MFR.
                                      NOMENCLATURE
                                   (7)
                                   -'CYCLE *'
(8)
                             « QR6N. NO. *>
(9)
                                           (3)



                                I  I  I   I  I  I  I  I   I  I  I  I  I   I
                                                                       MANUFACTURE
(10)
                                                              DUE
(11
                                                                         12;
13)
      NOMENCLATURE

  mmmitmt
    9 9 9 9 39 9 9 9 J 9 4 9 9
n i! ii u u ii ii ii ii M ii n n :< a ii ;i
         MANUFACTURE

    saiBtmssgitsi
    392999999999999
                                                       MODEL

                                                   ntnnnt
                                                   9999999S99
                                 n JIM II II11 II IS »» U MM (I 41 II « IS I! II « 
-------
                                                            Section:   5(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 15 of 31
          Figure 5-6 is an example of a calibration history evaluation.
The time in test, and out of test and number of failures are recorded for
evaluation purposes.  This type of form could also be used to record
daily checks in the instrument log books.  These records could be collected
and evaluated periodically for such things as indication of trends/plotting
of control charts or cost evaluation studies.
          Of primary concern in the measurement system is the calibration
of the constant volume sampler, the analytical instruments, dynamometer
and the gas mixtures.  Procedures for calibrating these items have been
detailed in Volume II of this report.  Examples of forms presently in
use are given in Figures 5-7, 5-8, 5-9 and 5-10.  These forms are used
to record raw data only.  The data output from this data is usually
audited manually or automatically, depending on the program used to
reduce the datTa, to locate points which may be out of tolerance, curve
slope changes and other types of errors.  The raw data should always be
maintained in the instrument log book.

5.2.3     Recording Maintenance Actions

          Preventive maintenance actions are performed on a periodic
routine schedule as outlined by the preventive maintenance guidelines,
discussed in Section 4, and chronologically recorded in the instrument
or test cell maintenance log book.  A typical Maintenance Report format
is illustrated in Figure 4-9.  Audit of this log book by quality assurance
usually will be sufficient to assure that the maintenance is done.
Entries in the log book should be signed by the person performing the
maintenance.
          Non-routine maintenance performed because of an equipment
failure can supply meaningful information to facility management.
Frequently, for the sake of expediency, the maintenance is performed but
never reported through the proper channels.  Reporting of all failures
should be mandatory as this information is invaluable in determining
equipment reliability and cost.  In addition, frequent failures of
certain equipment will indicate a need for corrective action.  One
method of recording these failures is through a work order or equipment
repair authorization form.  The work order should be issued by pro-
duction control and summarized in a weekly report.  Copies of the com-
pleted work order should be filed in the equipment records file.  A
typical work order request is presented in Figure 5-11.
          Equipment repair may be performed in-house or by outside
servicemen.  In either case the same job request form should be used and
completed indicating the service performed, man hours and parts replaced.
In addition the total charge for the service performed should be noted.
                                 73

-------
        Section:    5(LG)
        Revision:   0
        Date:        June 1975
        Page 16 of  31
CALIBRATION  HISTORY EVALUATION
     Figure 5-6
EQUIP TYPE 	
MFR. 	
MODEL 	
EVALUATED BY.
       TIMES FAIL
       DATE 	
                    TIMES CAL..
                    TIMES O.T. .
                    % O.T.	
                    AVG. CAL. TIME
PERCENT
HOURS
   CONTROL NO.
                    I.T.
                          O.T.
                                 FAIL
                                                                 REMARKS
Evaluation Summary:
Corrective Action:
                                              By
                                              Date
Follow-up Required:
Follow-up By:,
Dot* ,
.Remarks.

-------
                                                    Section:   5(LG)
                                                    Revision:  0
                                                    Date:     June  1975
                                                    Page 17 of 31
                CVS CALIBRATION  CALCULATION SHEET
DATA:
     Ambient
FORMULAE:
    AF =


   Pcf =
                              LFE
                                        CVS Pump
Tamb.
Pb
LFE Serial
Constants
°F AP
"Hg Tinl
Pinl
Mo.
A
R
"H?0
°F2
"H20

Tinlp
Toutp
Pinlp
Poutp
Counts
T ime
AP
°F
°F
H oU
"hUO

M i n .
n /^ \J
LFE A     LFE B

       - .0736 x
              Pb
                              Pinl
     Q = AF x  Pcf  x  Tcf
     Q = 	 * 	
    Vo =

    Vo =

    Vo =
                            x
                   528
                x
                        x
               760
               Pp
               760
                   528

                x        x
                                           Ap
                                    J/29.92 _
                                      TEST  SITE

                                            DATE

                                  CVS SERIAL  NO

                                          P.I.C.
                        Figure  5-7
                             75

-------
    Section:   5(LG)
    Revision:  0
    Date:     June 1975
    Page 18 of 31
                        Figure 5-8
                  TESTING SERVICES DIVISION

          ANALYZER CURVE GENERATION DATA
DEPT NO

TRAIN

DATE

P. I.C.

ANALYZER

CYLINDER NO







RANGE

DEFLECTION







CELL LENGTH

CONCENTRATION







ATTENUATION

FLOW RATE







NEXT RUN NUMBER

NO.







     ZERO
     GAS
AIR
NITROGEN
NEW
UPDATF
COMMENTS:
                            76

-------
                                  MONTHLY DYNO CALIBRATION LOG
DATE






















METER CALIBRATIONS
RHP
0




















ELECT
SPAN




















SPEED
MPH




















RPM*





















INERTIA




















RHP<5> 50
METER




















COAST DOWNS
TACH RPM
55 MPH




















45 MPH





















SEC
55-45 MPH




















ACT.
RHP




















CAL.
RHP




















A**
RHP




















INITIALS




















                                                                                                   ^^ C3 50 U5
                                                                                                   fU (U tD (I
                                                                                                   iQ ft < 0
                                                                                                   (0  H- ft
                                                                                                     .. 0) H-
                                                                                                   M   H- 0
                                                                                                   tf>   03
                                     U) C-l O
                                     M C

                                       n>
SITE LOCATION	

 *STROBATACH

**IF GREATER THAN
           METHOD:  5th WHEEL.
STROBE.
COMPLETE COAST DOWN MUST  BE  PERFORMED

                     Figure 5-9
                                                                                                     tn

-------
       Section:   5(LG)
       Revision:  0
       Date:      June 1975
       Page 20 of 31
     Figure  5-10
                           GAS ANALYSIS  REPORT
      CYLINDER »
MIXTURE REQUESTED
ANALYSIS
Supplier

Requestor
    Analyst.

      Date.
No. of  Cylinders  in order
          P.0.#__
         Project ».
         Invoice «.
Report  Data To

Comments
                                     78

-------
                                 Figure 5-11

                       Equipment Repair Authorization
                                                                   Section:    5 (LG)
                                                                   Revision:   0
                                                                   Date:       June 1975
                                                                   Page 21 of 31
                                                                 Job #
REQUEST

Name	 Date Submitted	
Branch__	Se c tion_	Extens ion
Equipment I.D. Number	
Job Description (Attach sketches needed)
Special Equipment Required
Proprietory Item:  | •  |  Yes   |   |  No
Craft Requested	
Date Item To Be Delivered for Test	
Latest Acceptable Completion Date 	
SCHEDULE - Equipment Maintenance            Equipment Repair Service Contacted
  Date Request Rec'd	   Date	Time	am/pm
  Craft(s)/Team Assigned	   Authorized by	
  Date	Time	am/pm
Equipment Back On Line                      Total Down Time	
Date	Time	am/pm   No. of Test Rescheduled
Equipment Repaired	             Test Supervisor	
          Replaced	
Equipment Maintenance Report                      Repair Service Report
Technician
Date Began	Time	am/pm   Date Began	Time	am/pm
Date Complete	Time	am/pm   Date Complete	Time	am/pm
Man-Hours	   Man-Hours	
Parts Replaced	   Parts Replaced	
Comments                                    Comments
                                            Service Charges
                                         WHITE: REQUESTOR'S COPY
                                         YELLOW: PRODUCTION CONTROLLER'S COPY
                                         PINK: LAB SECTION CHIEF'S COPY
                                         GOLD: REQUESTOR'S IN-PROCESS COPY
                                       79

-------
Section:   5(LG)
Revision:  0
Date:      June 1975
Page 22 of 31
5.2.4     Reporting Unacceptable Results

          Quality assurance has the responsibility for identifying areas
of the measurement system which need special consideration in order to
reduce the cost of the measurements, to increase production and improve
reliability.  One useful tool in determining these areas is an adequate
system for reporting unacceptable results.  These results should not be
limited to the tests rejected by data validation but should include any
determination made in the measurement system such as receiving inspec-
tions, equipment calibrations, test vehicle inspections, test cell
correlations and other auxiliary laboratory tests.  A typical Rejection
Report for use in reporting unacceptable results is shown in Figure 5-12.
This report should contain the type of result such as void test, the
unacceptable characteristic or data such as driver's trace error, the
reason for rejections and any immediate corrective action taken.  The
specific cause of the unacceptable results should be clearly identified.
          Analytical summaries of these rejection reports should be
prepared and reported to management by quality assurance.  Monthly and
yearly summaries by categories are quite helpful in identifying problem
areas and projecting realistic schedules.  Areas requiring corrective
action may be identified and reliability of equipment and personnel can
be objectively assessed from the information contained in these summary
reports.

5.2.5     Failure Reporting and Analysis

          A failure can be defined as the inability of a piece of equip-
ment or a vehicle to perform within previously specified limits.
          Failure rates can be reduced in magnitude with a resulting
reduction in testing costs if the following ground rule is applied.
Equipment/vehicles which have exhibited a trouble or failure, continuing
or intermittent, shall not be re-used or repaired until such time as the
trouble is isolated, the cause clearly established and corrective mea-
sures investigated, approved and taken to assure that the probability of
recurrence is minimized.
          The documentation of failures and the ensuing failure analysis
provides essential data for investigating the cause of failure and the
initiation of corrective action to preclude future recurrence.  A typical
Failure Analysis Report is shown in Figure 5-13.
                                 80

-------
REJECTION REPORT
Section:   5(LG)
Revision:  0
Date:      June 1975
Page 23 of 31

     NO.11101
PART NUMBER PART NAME SUPPLIER/MFR
REJECTED CONTRACT PURCHASE ORDER NO. REC. REPORT NO.
QUANTITY
ITEM NO.







DATE
DISCREPANC ES







REJECTED BY DATE SUPERVISOR APPROVAL DATE Q.A. APPROVAL DATE
DISPOSITION CHECK IF FAILURE ANALYSIS REQUIRED
USE AS IS
RETURN TO
FAIIIIRF ANAIYSK RFPORT NO
SUPPLIER
OTHER (SPECIFY) CORRECTIVE ACTION

Q.A. APPROVAL DATF
i
  Figure 5-12

    81

-------
      Section:    5{LG)
      Revision:   0
      Date:       June 1975
      Page 24 of 31
H
W
JM X
« p
<
COMPLETED
OPERATOR/AK
M
DATE

EQUIPMENT TYPE
NO. OF PAST FAILURE
REPORTS
ESTIMATED CAUSE
DETERMINATION

FAILURE ANALYSIS REPORT

MANUFACTURER
DATE-LAST FAILURE
EXPECTED, RANDOM
CAUSE DETERMINED OR
INSPECTOR
REPORT NUMBER

MOD. NO.
HOURS SINCE LAST
LOCATION FAILURE
ASSIGNMENT CAUSE
TYPE INVESTIGATION



SER. NO.

REQUIRES LAB
INVESTIGATION
NEEDED

      TESTS PERFORMED:
      FAILURE CAUSED BY:
      DISPOSITION:
                   f~f REPAIR
                   /"7 SCRAP
    REPLACE
/"7 OTHER
/"7 RETURN TO
    MANUFACTURER
      RECOMMENDED CORRECTIVE ACTION TO PREVENT RECURRENCE
      ACTION COPIES TO:
      RECOMMENDED BY
                                       REVIEWED BY
X O
CQ H
  PL,
0 E
w o
      CORRECTIVE ACTION TAKEN
RESULTS AND RECOMMENDATIONS
      APPROVED
      CLOSED OUT
                                            DATE_

                                            DATE
                   Figure 5-13  FAILURE ANALYSIS  REPORT
                                      82

-------
                                                           Section:   5(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 25 of 31
          The technique of Pareto analysis can be utilized very effec-
tively in analyzing failure types, as the bulk of failures, downtime,
etc., are traceable to a vital few failure modes.  Basically, the Pareto
analysis attempts to find the maldistribution for which the fewest poten-
tial failure modes provide the greatest potential for corrective action
applications.  This technique is discussed in Reference 5-1.

5.2.6     Initiating and Assuring Closed-Loop
          Corrective Action

          Corrective actions are of two kinds.  The more frequently
encountered type is immediate or on-the-spot corrective action to correct
non-conforming data or equipment.  It is important in this case to differ-
entiate between normal non-reportable procedural adjustments of equipment
that are performed as a matter of course during a test due to the charac-
teristic of the equipment, and those adjustments that are performed in
actual out-of-control situations, which should be reported as unacceptable
results (section 5.2.4).
          The second kind, long term corrective action, is invoked when it
becomes necessary to identify and eliminate the cause of non-conformance
and to prevent, if possible, the reoccurrence of the out-of-control
condition.  It is important that once a condition of unacceptable quality
is detected, a systematic and timely mechanism is established to assure
that the condition is reported to those assigned responsibility for
correction of the condition.  A positive closed loop mechanism must be
established to assure appropriate corrective action is taken.
          Documentation of closed loop corrective action usually takes
the form of the corrective action request.  A request for corrective
action can be initiated by anyone in the system, however, the formal
request is the responsibility of Quality Assurance management and is
usually assigned as a function of quality engineering.  A typical
corrective action request form is presented in Figure 5-14.
          To illustrate the use of a corrective action request form,
assume a test operator has observed that a CO Analyzer valve malfunc-
tioned.  The flow chart illustrated in Figure 5-15 traces the various
steps and interactions required to process a corrective action request.
          Generally, it is the responsibility of quality assurance to
utilize whatever means are available to see that the necessary actions
are completed.  Sometimes corrective action coordination responsibility
is assigned to an engineering function, with quality assurance monitoring
the effectiveness of the system.  Weekly status reports to management of
each of the assigned actions is usually adequate.  If the action is not
completed by the required date, quality assurance/engineering should
follow up, requesting an interim report of the progress and reasons for
the incompletion.  If the responsible organization is unable to meet the
deadline it should request an extension and any additional information or
assistance required for completion of the action.
                                83

-------
Section:   5(LG)

Revision:  0

Date:      June 1975

Page 26 of 31
           8
           w
           a

           a
           a
               Corrective Action Request
               1.  Request initiated by


                   Date
                                            Dept.
                           Authorization
               2.  Brief description of non-conformance
               3.  Recommended Action
                                                    Date
                                                File No
4.   Assigned to 	



5.   Quality Analysis (Attach complete report if necessary)
               6.  Action Required
               7.
Action to be initiated by



Expected Completion Date



Follow-Up Date 	
                                                     _Action Completed    Yes
               8.
Action Assigned To



Completion Date
                                                           Date
                                                   jSupervisor
                   Special Instructions
                9.  Action Completed -  Date
                                        Time
                    Quality Engineering Notified



                    Requestor Notified	


               10.   Comments
                                            Figure 5-1

-------
          CORRECTIVE  ACTION REQUEST -  FLOW CHART
                                                 Section:   5(LG)
                                                 Revision:  0
                                                 Date:      June 1975
                                                 Page 27 of 31
                 Gas Analyzer Malfunction
          Test Operator documents malfunction on
                  Rejection Report (R.R.)
         Forwards R.R.  to Supervisor for approval
   Test Operations Supervisor reviews and approves R.R.
    Determines need for Corrective Action.  Initiates a
  Corrective Action Request (C.A.R.).  Forwards R.R.  and
                C.A.R.  to Quality Assurance
   Quality Assurance reviews R.R.  and C.A.R.  Determines
  Action Addressee for C.A.R.,  Indicates required action
   and expected completion date.   Enters information in
C.A.R. Follow-Up Log.   Forwards C.A.R. To Action Addressee.
    Action Addressee reviews C.A.R. Determines need for
    equipment repair.  Initiates request for Equipment
                   Repair Authorization.       	
         Support Operations repairs Gas Analyzer.
         Completes Equipment Repair Authorization.
    Action Addressee completes C.A.R. indicating reason
      for malfunction and corrective action taken to
   preclude recurrence.  Obtains Supervisor's approval.
      Re-Routes completed C.A.R. to Quality Assurance
        Quality Assurance reviews completed C.A.R.
   If Corrective Action approved, closes out entries in
C.A.R. Log, indicates approval on C.A.R., forwards copy of
   approved C.A.R. to originator.  If Corrective Action
    disapproved, issues new C.A.R. to Action Addressee,
   notes status in C.A.R. Log.  Follows up on new C.A.R.
           until Action Addressee has completed
                approved Corrective Action.
                        FIGURE 5-15

                            85

-------
Section:   5(LG)
Revision:   0
Date:      June 1975
Page 28 of 31
                Upon completion of the action, management, the Quality Engi-
      neering Supervisor and the person or department initiating the request
      should be notified.
                In the interest of saving time and getting the job done the
      communication of the problems and progress should be done verbally by the
      Quality Engineering Supervisor; handwritten notes to the file would
      normally be sufficient for extensions and other analysis or agreements
      made.  Quality assurance should review all open files.

      5.2.7     Recording Audit Results

                There are two types of audit discussed in this report, inde-
      pendent performance audits and quality assurance system surveys.  The
      procedure and types of documentation required for a quality assurance
      system survey are outlined in Section 8 of this report.
                The documentation used in recording the results of independent
      performance audits would essentially be the same data collection forms as
      are normally used in the collection of that particular data.  In addition
      to these data records, control charts may be subsequently plotted using
      the audit results to determine if the element being audited is performing
      within established limits.  Control chart techniques are discussed in
      detail in Section 6.
                A performance audit summary sheet should be maintained by the
      auditor to provide a history of audits performed (Figure 5-16).  Periodic
      review of this summary will indicate whether the original audit schedule
      is effective or if a tightened or reduced schedule is required.  Separate
      summary sheets should be prepared for each of the major elements audited,
      i.e., Instrumentation, Operator, Sampling System and Data Processing.

      5.2.8     Initiating Procedural or Equipment Change Notices

                A clearly defined system is characteristic of a good quality
      system.  However, it must be responsive to changes resulting from ad-
      vances in the state-of-the-art in the measurement system.  Any change
      effective on a temporary basis or for a particular series of tests must
      be systematically documented to reflect evidence of such a change in
      subsequent analysis of the data.
                Changes in the design of the equipment used in the measurement
      system must also be carefully documented.  Configuration control of the
      total test system is important since not only do the basic sample hand-
      ling procedures change but actually instrumented analyses change with
      results that are not directly correlatable.  In many cases, changes
      should not be made until a comparative analysis has been completed in
      order to assure that the recommended changes do not affect accuracy and
      precision in a deleterious way.  For example, hydrocarbons may be mea-
      sured by non-dispersive infrared (NDIR) or by flame ionization  (FID).
                                       86

-------
                                                   PERFORMANCE AUDIT SUMMARY  SHEET
oo
AUDIT ELEMENT: Q INSTRUMENTATION Q OPERATOR Q SAMPLING SYSTEM p DATA PROCESSING
AUDIT
DATE

TYPE OF AUDIT

AUDIT RESULT
ACC.

UNACC.

CORRECTIVE ACTION TAKEN

AUDITED
BY

                                                                                                                                      hj O  50 W
                                                                                                                                      CU CU  CD CD
                                                                                                                                      iQ rt  < o
                                                                                                                                      CD CD  H- rt
                                                                                                                                        ..  w p.

                                                                                                                                      M    H- O



                                                                                                                                      :    !••
                                                                                                                                      l-h


                                                                                                                                      U) Oi  O Ul
                                                               FIGURE 5-16
                                                                                                                                        vc

-------
Section:   5(LG)
Revision:   0
Date:      June 1975
Page 30 of 31
      The numbers are only significant for those emission standards based on
      the particular method of analysis.  NDIR hydrocarbon measurements are of
      little use in the 1975 light duty procedure but are required by the
      1974 heavy duty gasoline engine procedure.
                The responsibility for procedural change and equipment config-
      uration control should be assigned by management policy.  Quality Assur-
      ance has responsibility for approval of all changes.  Distribution of the
      changes is usually performed by those with responsibility for manual
      control.  All affected individuals should be informed of the changes on a
      timely and formal basis.
                An example of a document used to effect configuration and
      procedural changes in the measurement system is given in Figure 5-17.  A
      similar document could be used to effectively control changes in computer
      programs in facilities which employ computer systems for testing and
      computational services.  Procedures for document control are given in
      Appendix C of this report.

-------
                                                 Section:   5(LG)
                                                 Revision:  0
                                                 Date:      June 1975
                                                 Page 31 of 31
                           Figure 5-17

         PROCEDURE/EQUIPMENT CONFIGURATION CHANGE NOTICE
1.  ORIGINATOR:
                                        2.  DATE:
 3.   TYPE OF CHANGE: EQUIPMENT /""/  PROCEDURE  /~7

 4.   REFERENCE DOCUMENT:
 5.   CHANGE REQUESTED BY:

 6.   PURPOSE OF CHANGE:
 7.   DESCRIPTION OF CHANGE:  (Attach  Details, Specifications or
     Drawings if Necessary).
 8.   EFFECTIVITY:
9.  DURATION OR EXTENT OF USE:
                                             TEMPORARY
                                             PERMANENT
10.   AREAS AFFECTED:  LOT £7  E&D £/  CHEM (~J  LAB LJ

                      HOT £7  I&E £7  C&M  £7 DATA

                      OTHER                      	

 11.  APPROVALS REQUIRED
                              YES
                                        NO
DATE
ECTD
0PM
CSD
     (If not approved please  discuss reasons on reverse  side)

 12. RETURN TO ORIGINATOR FOR DISTRIBUTION TO REVIEWERS  AND

     AREAS AFFECTED.           Qo
                               09

-------

-------
                                                           Section:   6(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 1 of 38
                            Section 6

          APPLICATION OF STATISTICAL QUALITY ASSURANCE
               METHODS TO THE EMISSION TEST SYSTEM
          An effective and efficient quality assurance system requires
the appropriate use of statistical methods.  The nature of the data
collected from the system requires the use of some specific statistical
methods, although practically all statistical tools can be applied to
quality assurance data at one time or another.
6.1       STATISTICAL METHODS

          Several of the most useful applications of statistical methods
as they apply to mobile source emission testing are as follows:

          1.   Use of statistical control charts for:

               a.   Successive zero/span checks
               b.   Constants of calibration curve solutions
               c.   Agreement between duplicate checks
               d.   Differences between original and independent audit
                    checks
               e.   Flow rate calibration checks

          2.   Regression analysis for:

               a.   Calculation of calibration curves
               b.   Determining relationships between variables in
                    measurements

          3.   Statistical sampling plans for:

               a.   Inspection of procured materials
               b.   Determining frequency of checks using standards, and
                    duplicate checks
               c.   Determining frequency of zero/span checks
               d.   Determining frequency of multipoint calibrations
                                91

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 2 of 38
           4.    Analysis of distributions of  data to measure  the  inherent
                variability in  the data, and  to  establish  limits  of
                agreement  for duplicate checks,  independent performance
                audit  checks, and other distributions  for  which control
                chart  limits need to be established

           5.    Analysis of failure rates to  determine optimum frequencies
                for preventive  maintenance  and scheduled replacement  of
                components

           6.    Use of probability paper to make predictions  based on a
                normal distribution.

 6.1.1     Special Applications of Statistical Methods

           There are specialized statistical  techniques which can be  used
 as effective tools in analyzing variables.  The analysis  of  variance can
 be used for performing  special comparisons of variables in the measure-
 ment system.   Statistical designs  for planning  special studies to deter-
 mine effects of suspected variables  can be developed, particularly useful
 in investigation of possible  causes  of quality  problems.

 6.1.2     Statistical Techniques  and Nomenclature

           Certain methods almost  always constitute,  in part, a good
 quality assurance system. Subsequently, an  understanding of certain
 fundamental statistical techniques and nomenclature  is necessary in
 establishing proper quality assurance procedures. Appendix  A-l  provides
 a glossary of such terms.
 6.2       CONTROL CHARTS

           This section describes the definition,  purpose,  format and
 application of control charts as they apply to mobile source emission
 testing.

 6.2.1     Definition and Purpose of Control Charts

           A control chart is a chronological graphic comparison of mobile
 source emission testing data to computed control limits that are drawn as
 limit lines on the chart.  The primary purpose of the control chart is to

-------
                                                           Section:   6(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 3 of 38
identify specific causes of variation.  Variation can be attributed to
two causes, assignable, i.e., as a result of "findable," and random,
i.e., small, nontraceable,  "chance" factors.  The role of the control
chart is concerned with assignable causes.
          The control chart presentation of quality is helpful for many
reasons; among them are:

          o    Detection of trends which could lead to "out-of-control"
               conditions,  or create problems if not corrected at time of
               detection

          o    Visual record assuring completion of routine checks

          o    Levels of quality can be more readily prescribed based on
               observed, obtainable, past levels.

          o    Management decision-making can be more readily based on
               easy access to past quality data

          o    A "picture" of quality as exemplified in quality charts is
               the single best description of performance

6.2.2     Format

          The format of the control chart usually follows the configura-
tion presented in Figure 6-1.  The upper and lower control limits define
the expected spread.  Plotting a central line  delineating the average
level of the values is helpful in evaluating biasing, and detecting
trends.

6.2.3     Types of Control Charts

           In Appendix  A-l the concepts  of precision and accuracy are
defined.  Generally, precision is the ability of a system to reproduce
its own levels of performance.  Accuracy  is the  difference between  a
measurement and a true value.  Precision  control charts delineate the
amount  of variability  among  replicate laboratory analyses results.
Variability can be expressed in the unit  of measurement of the variable
or in terms of percent.  When the extent  of variability is a  function of
the  level  of gas concentration, then  the  Coefficient  of Variation  (CV)
or Relative Range  (%R)  control charts are appropriate.  Control  charts
indicating levels of accuracy can also  be constructed.  The  standard
deviation  or range defined  in terms of  physical  units is a convenient
method  for measuring the variability  among  accuracy determination data.
A detailed discussion  of types of  control charts is contained in Appendix  H
of Reference 6-1.
                                 93

-------
+2.0.H
+ 1.0-
                i •»•• •••••• Li
                           r
A
/  \
                                                                  CVS FRAME 408-14
                                                       13 O 50 W
                                                       (U f  H- rt
                                                         •• tfl H-
                                                       .   H- O
                                                       •^   O 3

                                                       0
                                                       Ml
                                                                                                 oo
                                                                                                  ID
O
cc
a:
                                              t   •

                                              V
                                                                          CENTER  LINE
-i.o-
-2.0-11
               LOWER CONTROL LIMIT
                                          7     8     9   10    11    12    13    14   15


                                         SAMPLE NUMBER
                          Figure 6-1.  PROPANE  INJECTION TEST - % ERROR

-------
                                                           Section:   6(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 5 of  38
6.2.4     Applications of Control Charts in Mobile Source
          Emission Testing

          In mobile source emission testing, control chart techniques
are implemented to determine whether the errors associated with the
analytical data are within operational limits designated for the method.
For example, the precision of an exhaust emission meaurement system can
be evaluated from the use of replicate analysis results.  This can be
accomplished by performing replicate measurements using known HC, CO,
CO , and/or NO  concentrations and monitoring the degree of variability
among the replicates.  Other methods of replicate analysis include the
retesting of gas in a bag, and the use of a correlation vehicle.
          Some typical applications of control charts in 1975 FTP test-
ing situations are summarized in Table 6-1.  Construction of these
various types of charts is discussed in the following sections.

6.2.5     Precision Control Charts

          In using control charts, precision can be expressed in the
unit of measurement of the variable or in percent.  When expressing
precision in terms of units, variations can be expressed as a range,
using R-Charts, or as standard deviation using s-charts.  In air
pollution study applications, precision is often computed in terms of a
percent using the relative range  (%R) chart or the coefficient of varia-
tion (CV) chart.
          The following nomenclature should be noted:
          R = Maximum - Minimum
               ,     _ 2v
               (2.)(x-x)   )
               V  n-1    /
               D
         %R = 	 x 100%
         CV = -—  (100)%
                                95

-------
          Table 6-1.  APPLICATIONS OF STATISTICAL CONTROL CHARTS IN 1975 FTP TESTING
  TYPE OF CONTROL CHART
Coefficient of variation
control charts
Range chart
Signed difference chart
Relative Range, CV
charts

Relative Range, CV
charts

Percent error
Signed difference chart
Relative Range, CV
charts

Mean and Range charts
Percent defective charts
             APPLICATION
Monitoring precision of positive
displacement pump by maintaining
control charts on various para-
meters

Measurement of recorder chart
speed

Difference in coastdown time
Precision of speedometer, power
meter

Determining precision of gas
mixture

Propane injection test
Correlation Vehicle
Retest of Gas Bag
Measure variation in gain, zero,
  P, etc.

Monitoring rejection rate of test
data entries.
     AREAS OF APPLICATION
     WITHIN MOBILE SOURCE
   EMISSION TEST PROCEDURES
CVS calibration procedures for
positive displacement pump
Chart recorder calibration
Dynamometer calibration
Gas Mixture calibration
CVS accuracy checks
Daily Start-Up checks
Data Validation tests
tj O » in
pj PI (D (D
iQ (+ < n
n> a H- rt
  .. 01 H-
^   H- 0
o   Si?
Hi
to
oo d o en
                                                                                                     VO
                                                                                                     -J
                                                                                                     in

-------
                                                           Section:   6(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 7 of 38
Where:

          R = Range

          s = Standard Deviation

         %R = Relative Range

         CV = Coefficient of Variation

          x = Individual Value

          x = Mean Value

          n = Number of Replicates
          Replicate analyses should be made on known standards at differ-
ent levels and evaluated to determine the type of precision control
charts to use.  Standards should be used which represent the high and
low and at least one, however preferably two, intermediate concentrations
encountered during testing.  Between five and ten replicate analyses
should be made for each known concentration.
          The mean (x) and standard deviation (s)_for each_ concentration
should be found, i.e., calculate x , s , x  , s , x  , s  , x , s .  Plot
these values on a scatter diagram.  They wxll normally  coincide with one
of the two configurations:  (1) the standard deviation  is essentially
independent of the concentration mean, or (2) the standard deviation is
dependent upon changes in concentration.  Typical examples of these two
configurations are shown in Figure 6-2.  The plotted points were obtained
from the data compiled in Table 6-2.
          The standard deviation  (s) or range (R) control chart techniques
are applicable if Case 1 exists.  Note, however, that in the mobile
source emissions testing context R-Charts are normally  used, as the
range is an efficient estimator of the variation, and the number of
replicates do not usually exceed two.  CV-Charts or %R-Charts should be
implemented if Case 2 occurs.
          Relative Range or CV-Charts are derived from  measurements
obtained from replicate analysis of routine samples.  In mobile source
emission testing systems it is customary to use two replicates for pre-
cision determination and in such situations the use of  Relative Range
charts is recommended.  Where the number of replicates  exceed two, the
Coefficient of Variation chart is appropriate.
                                97

-------
Section:  6(LG)
Revision:  0
Date:     June 1975
Page 8 of  38
      CASE 1  - STANDARD DEVIATION ESSENTIALLY
            INDEPENDENT OF CONCENTRATION
o0.2

<

LU
O

So.i.
a
z
<
    0.0.
0    1
                                       ___	0

                           •• ** ^ "*
                  23456
                   CONCENTRATION MEAN
                                              8
       CASE 2- STANDARD  DEVIATION  INCREASES
         PROPORTIONATELY  WITH CONCENTRAT ION
    0.3n
   o

   £0.2H
UJ
o
o
OL

lo.i-
    0.0
           S
                   23456
                   CONCENTRATION  MEAN
                                               8
        Figure 6-2.   SCATTER  DIAGRAMS FOR DETERMINING
                    TYPE  OF CONTROL CHART
                            98

-------
Table 6-2.  MEASURED DATA USED IN SCATTER
            DIAGRAM CONSTRUCTION
                                               Section:   6(LG)
                                               Revision:  0
                                               Date:      June 1975
                                               Page 9  of  38
CASE 1
X
0.2
0.1
0.2
0.3
0.1
x = 0.2
1.0
1.1
1.2
0.9
1.2
JC = 1.1
3.0
2.9
3.1
3.0
2.9
x" = 3.0
7.4
7.5
7.3
7.6
7.5
x = 7.5
x - x
0
-.1
0
.1
-.1

-.1
0
.1
-.2
.1

0
-.1
.1
0
-.1

-.1
0
-.2
.1
0

(x - x)2
0
.01
0
.01
.01
s = .09
.01
0
.01
.04
.01
s = .13
0
.01
.01
0
.01
s = .09
.01
0
.04
.01
0
s = .12
CASE 2
x
0.2
0.1
0.2
0.3
0.1
x" = 0.2
1.0
1.1
1.2
0.9
1.2
x~ = 1.1
3.0
3.1
3.3
2.8
3.2
x = 3.1
7.4
7.0
7.5
6.9
7.1
x = 7.2
x - x
0
-.1
0
.1
-.1

-.1
0
.1
-.2
.1

-.1
0
.2
-.3
.1

.2
-.2
.3
-.3
-.1

(x - x)2
0
.01
0
.01
.01
s = .09
.01
0
.01
.04
.01
s = .13
.01
0
.04
.09
.01
s = .19
.04
.04
.09
.09
.01
s = .26
                    99

-------
Section:   6(1/3)
Revision:  0
Date:      June 1975
Page 10 of 38
 6.2.5.1   Construction  of Range Precision Control Charts  (R-Charts)

           The following procedure  should be used to construct a range
 control chart.   A typical example  is  shown in Figure 6-3.   The plotted
 points were obtained from Table 6-3.

           •    List the absolute values  of the range (R)  for each set of
                replicates (x ,  x )

           •    Compute  R, the average value of R for all  sets of replicates
                using the  formula
                 	    t>
                 R = 	 with N  = number  of sets of replicates

           •    Compute  the upper control limit, UCL, using the formula

                 UCL = D4R.

                The value  of D  is  obtained from Appendix  A-2.

           •    Compute  the lower control limit, LCL, using the formula

                 LCL = DR.

                The value  of D  is  obtained from Appendix  A-2.

           •    Draw the line for R on the control chart

           •    Plot the values  for ranges of each set of  replicates.

           For this control chart,  the computed control limits are 3
-------
      RANGE OF  PAIRS  OF REPLICATES
18-
16-
14-
12-
_10-
uj 8-
* 6-
4-
2-
n -
w

UPPER CONTROL LIMIT - 16.99





GL.
/ "•'^^
/ '° x-x ^'^- ,^*< AVERAC
/ ^ ^
i
LOWER CONTROL LIMIT - 0.0

I 1 1 1 1 1 1 	 1 | 	 1
1 23456789 10
          SAMPLE NUMBER
Figure 6-3.   RANGE CONTROL CHART
>O D » to
p) P> (D ID
^Q rt < n
(t> IB H- ft
  .. en H-
H   M- O
M   o a

o
                                                                      u>
                                                                      03

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 12 of 38
              Table 6-3.   DATA VALUES  AND COMPUTATIONS
             FOR CONSTRUCTING RANGE CONTROL CHART LIMITS
SAMPLE
1
2
3
4
5
6
7

8
9
10
Xl
10
15
21
11
30
45
50
s
42
10
21
X2
12
22
15
17
25
51
46

48
15
26
R
2
7
6
6
5
6
4

6
5
5
R TOTAL = 52
               R, * = 52
               R   N   10
             UCL = D,. R = 3.267 x 5.2 = 16.99
                    4
             LCL = D  R =  0  x5.2=0.0
             D_ and D. are multiplication factors when observations
             in each subgroup = 2
                                 102

-------
                                                            Section:   6(LG)
                                                            Revision:   0
                                                            Date:      June 1975
                                                            Page 13 of 38
6.2.5.2   Construction of Relative  Range  Control Charts

          •    Calculate the range, R,  established by each sampled
               duplicate set.

          •    Calculate the arithmetic mean,  x,  for  each  sampled
               duplicate set.

          •    For each sampled duplicate  set,  calculate the  relative
               range using the formula

               %R = -§- x 100%.
                     x

          •    Calculate the average relative  range using  the formula
                     N
                     V
                     s   * H •
                     f—i    j


               where N = Total number of sampled  duplicate  sets.


          •    Calculate the lower control limit  using the  formula

               LCL = D  %R.

               The value of D  is obtained from Appendix A-2.


          •    Calculate the upper control limit  using the  formula

               TT1T — T"»  ft-D
               UULi — U.  *K.

               The value of D  is obtained from Appendix A-2.


          •    Construct the Relative Range Chart delineating the values
               of %R, UCL and LCL.

          Figure 6-4 is an example utilizing the  above procedure.  The
hypothetical data used and the necessary calculations are given in
Table 6-4.
                                103

-------
  50
  30 i

  20 -
LU
or
  10 -
   0 -i
                       RELATIVE RANGE OF GAS MEASUREMENT  CONCENTRATIONS
                                                 UPPER CONTROL LIMIT = 43.03
O

A
/
                                        \
                                                                                a
   \
   \         	

   \AVERAGE %R=13.17 O
                                            \/
                                                  LOWER  CONTROL LIMIT = 0.0
                                 T
7    8    9    10     11
  DAY TESTED
                                          15
                                                      V D » W
                                                      p (U (D (D
                                                      U3 ft < O
                                                      (D fl> H- ft
                                                       .. 01 H-
                                                                                            o
                                                                                            H)
                                                                                              vo
                                                                                              -j
                                                                                              Cn
                          Figure  6-k.   RELATIVE RANGE CONTROL CHART

-------
                                                           Section:   6(LG)

                                                           Revision:  0

                                                           Date:      June 1975

                                                           Page 15 of 38
Table 6-4.  CONCENTRATION MEASUREMENTS - RELATIVE  RANGE CALCULATION
DAY
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
MEASUREMENTS, PPM
Xl
29.2
28.4
29.2
27.9
26.4
31.8
39.4
28.6
28.0
31.2
37.6
26.9
30.7
31.9
28.9
27.8
X2
22.7
25.2
26.4
30.2
31.8
31.5
29.1
29.2
26.2
35.2
31.8
29.0
28.0
26.8
36.2
31.4
R
6.5
3.2
2.8
2.3
5.4
0.3
10.3
0.6
1.8
4.0
5.8
2.1
2.7
5.1
7.3
3.6
X
25.95
26.80
27.80
29.05
29.10
31.65
34.25
28.90
27.10
33.20
34.70
27.95
29.35
29.35
32.55
29.60
TOTAL
%R
25.05
11.94
10.07
7.92
18.56
0.95
30.07
2.08
6.64
12.05
16.71
7.51
9.20
17.38
22.43
12.16
210.72
 —-   210.72       ._
 %R  =  —r-?	 = 13.17
         J.O
 UCL =  3.267  x 13.17 = 43.03
 LCL = 0 x 13.17 = 0
  D  = 0 and D. = 3.267 when observations in each subgroup = 2
                                105

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 16 of 38
6.2.5.3   Construction of Coefficient of Variation Control Charts

          •    Calculate the arithmetic mean x  for each sub-group of
               replicates.

          •    Calculate the standard deviation s for each sub-group of
               replicates using the formula
                         -2\°'5
                       (x-x) 2  I
                      ~^1    /    '
               For each  sub-group of replicates,  calculate  the coefficient
               of variation using the  formula

               CV = -^—  (100)%.
                Calculate the average coefficient of variation using  the
                formula

                      N


                CV =
                       N

               when N  = total  number  of sub-groups.


           •   Calculate the lower  control limit using  the  formula

               LCL = B  CV

               The value of B   is obtained from Appendix A-2.


           •   Calculate the upper  control limit using  the  formula

               UCL = B4 CV

               The value of B. is obtained from Appendix A-2.


           •   Construct the Coefficient of Variation (CV)  chart delineating
               the values of CV, UCL, and LCL.

           Figure 6-5 is an example  utilizing the above  procedure.   The
 hypothetical data used and the necessary calculation are given in Table  6-5.
                                106

-------
                   COEFFICIENT OF VARIATION  OF TEST MEASUREMENTS
  4.0-
  3.0-
   t£.

   >


O  lJ-
-4  O

   H-

   UJ


   O
  2.0-
  1.0

LJ
O
0
                                   UPPER  CONTROL LIMIT = 3.90
   AVERAGE  (CV)

                                                       o
                                                           \	t-

                                   LOWER  CONTROL LIMIT = 0.0
1     2    3
                                                                                 b
                             5    6    7    8     9     10   11   12    13    14    15
                                  TEST NUMBER  SEQUENCE
                                                                                              13 O 50 en
                                                                                              su BJ (D n>
                                                                                              IQ rt < n
                                                                                              01 n> P- rt
                                                                                               .. tn i-1-
                                                                                              M  P- 0
                                                                                              -J  03
                                                                                                 3 «
                                                                                              O
                                                                                              i-h
               Figure  6-5.   COEFFICIENT OF VARIATION  CONTROL CHART

-------
 Section:   6(LG)
 Revision:   0
 Date:      June 1975
 Page 18 of 38
Table 6-5.  TEST MEASUREMENTS - COEFFICIENT OF VARIATION CALCULATION
TEST
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Xl
.096
.189
.282
.378
.468
.556
.641
.720
.793
.856
.908
.922
.973
.988
.996
X2
.094
.191
.279
.375
.456
.548
.631
.708
.745
.828
.880
.947
.952
.966
.981
X3
.092
.187
.281
.369
.451
.541
.605
.670
.766
.841
.868
.934
.963
.951
.987
X4
.093
.184
.276
.361
.446
.551
.613
.678
.755
.805
.893
.890
.933
.974
.992
X5
.095
.185
.272
.370
.461
.553
.608
.687
.779
.816
.880
.900
.924
.981
.985
X
.0940
.1875
.2780
.3706
.4564
.5498
.6196
.6926
.7676
.8292
.8858
.9186
.9490
.9720
.9882
s
.0015
.0026
.0040
.0065
.0085
.0057
.0156
.0208
.0190
.0201
.0152
.0235
.0203
.0143
.0058
TOTAL
CV
1.60
1.39
1.44
1.75
1.86
1.04
2.52
3.00
2.48
2.42
1.72
2.56
2.14
1.47
0.59
27.98
UCL = 2.089 x 1.87 = 3.90
LCL = 0 x 1.87 = 0
      0 and B4 = 2.089 when observations in each subgroup = 5
                                 108

-------
                                                           Section:   6(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 19 of 38
6.2.6     Accuracy Control Charts

          Accuracy control charts are discussed in detail in Appendix H
of Reference 6-1.  There are occasions when variability in test results
has been significantly affected by testing conditions difficult to control.
For example, a sample may be selected and tested once a day in a situation
where weather conditions significantly affect the test.  Under such
circumstances, a x chart does not give a true indication of lack of control,
but only a lack of control of testing techniques, due to the confusing of
effects in the variations of weather with any real variations in the quality
of the test data.  To overcome such difficulties the difference control
chart has been devised (Reference 6-3).  This technique requires the use
of a standard unit or lot called the reference unit which is known to
have an output controlled at the desired level.  Such a unit or lot
could be taken as part of the output that had been produced under controlled
conditions, or it might have been made up as a result of artificial selection
and 100 percent inspection.  An application in mobile source emission
testing would be the use of a correlation vehicle in comparing test
measurements.

6.2.6.1   Construction of a Difference Control Chart

          o    Calculate the signed difference between the measurement
               from the current test unit (x ), and the reference unit
               (x ), i.e., x  - x .
                 r '        c    r
          o    Calculate the mean (x  ) and the standard deviation (s  )
               of the signed differences.

          o    The central line on the chart will be the mean of the
               signed differences.

          o    Calculate the upper and lower control limits for the
               chart using the formula

               x~    ±  3s         where x , = mean of signed differences
                sd       sd              sd
                                        s , = Std. deviation of signed
                                              differences

          o    Construct the Difference Control Chart delineating the
               central line, UCL, and LCL.

          o    Plot the signed differences  (x  - xr) on the chart.
                                109

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 20 of 38
 If points fall outside  the above  limits  and assignable causes  are found,
 the process is "out-of-control;"  if no points fall outside  the limits
 and there is no evidence  of non-random variation within the limits,  the
 process is said to be "under control with respect to its average," since
 variability in test results due to variations in testing conditions  from
 day to day have been eliminated by taking differences.
           Figure 6-6 is an example utilizing the above procedure  in  con-
 junction with the use of  a correlation vehicle.   Hypothetical  data were
 used in the computations  developed in Table 6-6.
 6.3       STATISTICAL INFERENCE AND SOME APPLICATIONS OF ACCEPTANCE
           SAMPLING

           This section discusses the meaning of statistical inference
 and how this concept can be used in developing an acceptance sampling
 procedure with respect to mobile source emission testing.

 6.3.1     General Context

           Statistical quality control is involved with quantitatively
 detecting and examining causes of variation in 1975 FTP testing and
 maintaining measurement quality at an optimum level.  Control charts
 which were previously described are typically statistical techniques
 applied to a continual process (e.g., charting measurement data for
 equipment performance).  However, what can be done to statistically
 analyze the properties of a group of data consisting of a finite number
 of measurements?  Statistical inference and sampling theory can provide
 a solution to such a problem.

 6.3.2     Definition of Statistical Inference

           Statistical inference is a method which allows one to infer
 what is true about a population from the results of a sample drawn from
 it.  This concept is very useful in that the quality of all elements
 within a group can be quantitatively determined without examining every
 element within the group.  Acceptance sampling is an application of
 this method.
           Why make use of statistical inference and sampling?  Why not
 inspect 100 percent of all the components or data which constitutes a
 group?  The answer, of course, is that it may be impractical (e.g.,
 testing may be destructive)  or too costly to inspect every element.
 Consequently, sampling can be a cost-saving statistical tool.
                                 110

-------
iu
iu
u.
5
+ 1.75



+ 1.50



+ 1.25



+ 1.00



+0.75



+0.50



+0.25



 0.00



-0.25



-0.50



-0.75



-1.00



-1.25
                             UPPER  CONTROL  LIMIT
                           LOWER CONTROL LIMIT
                                            .,	(	r	,	r-

                                            6     7     8     9    10
                          FIGURE 6-6.  SIGNED DIFFERENCES  CONTROL CHART
                                                                                                    S.D.
                                                                                                       t) o y} w
                                                                                                       Cu (U ft) ft>
                                                                                                       ua rt < O
                                                                                                       (D (D H- rt
                                                                                                         .. tn H-
                                                                                                       N>   H- 0
                                                                                                       H   03
                                                                                                           3 ••
                                                                                                       O
                                                                                                       i-h

                                                                                                       W C| O 0^
                                                                                                       CD C   •-»
                                                                                                         a   r
                                                                                                         (D   O

-------
Section:   6(LG)

Revision:  0

Date:      June 1975

Page 22 of 38




       Table 6-6.   HC CONCENTRATION MEASUREMENTS - CORRELATION

                   VEHICLE VS CURRENT TEST VEHICLE
TEST NO.
1
2
3
4
5
6
7
8
9
10
HC MEASUREMENTS, G/M *
X
c
2.06
2.46
2.21
2.37
2.34
2.85
2.91
3.29
3.36
2.40
X
2.43
2.26
2.46
2.20
2.60
2.28
2.79
2.32
2.24
2.27
SIGNED
DIFFERENCE
(Xc - V
- 0.37
+ 0.20
- 0.25
+ 0.17
- 0.26
+ 0.57
+ 0.12
+ 0.97
+ 1.12
+ 0.13
       UCL = x ,  + 3s .  = 0.24  + 3(0.51)  = 1.77
              sd      sd
       LCL - x .  - 3s .= 0.24 - 3(0.51)  = - 1.29
              sd      sd
                                                   x , = 0.24
                                                    sd
                                                   s   = 0.51
                                                    sd
                                 112

-------
                                                            Section-.   6(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 23 of 38
6.3.3     Application of Sampling Theory in Mobile Source
          Emission Testing

          Sampling theory can be used with respect to many aspects of
1975 FTP testing.  Some of these applications are as follows:

          o    Inspection of incoming lots of procured materials (e.g.,
               gas bottles)

          o    Determining frequency of checks in evaluating 1975 FTP
               measurement systems performance and 1975 FTP test results,
               sampling from past six month's data

          o    Determining frequency of zero/span checks in evaluating
               HC, CO, C02,  and NO  gas analyzer performance, sampling
               from past six month^s data.

          o    Determining frequency of multi-point calibration, sampling
               from past six month1s data

          o    Determining frequency of checks in validating data (e.g.,
               recorded gas analyzer strip charts), sampling from past
               six month's data.

          There is a detailed discussion of statistical sampling in
Appendix  I of Reference 6-1.
6.4       ANALYSIS OF VARIANCE

          This technique provides an objective method of dealing with the
total variation within a test.  By breaking down this variation into the
contributions of main effects, interaction and residual effects, valid
conclusions can be made regarding the test data through the use of statis-
tical methods.  The test must be designed to allow  extraneous influences
to operate in a truly random manner.  To obtain valid conclusions from
the test it is necessary to maintain proper control of other variables
in addition to those being investigated.  Uncontrollable or unknown
conditions occur in most tests.  Conditions such as temperature variation,
operator efficiency, equipment repeatability, and variation among related
items included in the test but not under control are only a few of the
possibilities to be considered.

6.4.1     Basic Theory

          The analysis of variance provides an indication as to whether   -
or not the observed differences among the means of  the samples are signi-
ficant, that is, greater than those variations which can be attributed
                                113

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 24 of 38
solely to sampling fluctuation.  To do this, the variance is computed
using two methods.  The F test is then used to quantitatively determine
the significance between the values obtained using each method.  A more
detailed description of analysis of variance theory and its applications
can be found in References 6-4 and 6-5.

6.4.2     Analysis of Variance Implementation in Mobile Source
          Emission Testing

          The analysis of variance objectively determines if significant
differences exist between groups of sampled data.  Such a technique is
useful in quantitatively examining the repeatability of a given measure-
ment system.  Accuracy of measurement systems can also be evaluated
using the analysis of variance.  In the measurement of exhaust emissions
from a given vehicle there are three levels of variability, i.e., vari-
abilities associated with a given test cell, cell-to-cell variability
within a given laboratory site, and laboratory-to-laboratory variability.
Factors affecting variability include the vehicle, driver, ambient
condition, dynamometer, CVS, analyzer, calibration gas, operator and the
computer.  The statistical significance of  any of these factors on the
test results can be evaluated by using the  analysis of variance technique.
This technique can be used to evaluate the  differences in performance of
various CVS systems, catalytic converters,  etc., and to determine the
significance of reduction in exhaust emissions as the result of scheduled
maintenance procedures.
          The following is an example comparing gas emissions from three
cars tested at five different times.  An analysis of variance test
(Table 6-7) is computed to determine if there are any significant
differences between cars.  The area of interest will be the effect of
one factor only on the gas emission measurements, in order to demon-
strate the computational set-up for a one-factor analysis of variance.
The factor, car type, is said to be in three categories as there are
three cars, and it is assumed that these are the only cars to be concerned
with.  It is not desired to generalize the  results to other car types of
which the three might be a random sample.   This is an important point.
As only these three car types are being considered, the factor is in a
fixed category.  If the engineer is interested in these three car types
as a random sample of a whole population of car types, car types would
be a random effect.  In a one way classification  (one factor) like this
one, the analysis used to obtain the results would be the same for
either a random or fixed effect, but the significance tests performed
would be interpreted differently.  This discussion will be confined to
designs with fixed factors only.
          Now, some engineer notes that five different sample gases were
used in these tests and realizes that further data analysis would determine
if there were possible differences due to the different gas samples.  The
problem now becomes an analysis of variance (Table 6-8) with a two-way
classification of the data, i.e., two factors: car type and gas sample, one
in three categories  (three car types) and the other in five categories
(5 gas samples).  Again it is assumed that  the five gas samples are the
                                114

-------
                                                       Section:   6(LG)
                                                       Revision:  0
                                                       Date:      June 1975
                                                       Page 25 of 38
Table 6-7.  ANALYSIS OF VARIANCE - ONE-WAY CLASSIFICATION
GAS CONCENTRATION MEASUREMENTS FROM
TEST
NUMBER
1
2
3
4
5
SUMS
SUMS OF
SQUARES
THREE CARS
EMISSION MEASUREMENTS
CAR TYPE 1
6.6
7.2
6.4
7.4
7.8
35.4
251.95
CAR TYPE 2
6.6
6.4
7.0
6.2
6.8
33.0
218.20
CAR TYPE 3
7.0
6.0
5.0
5.8
7.0
30.8
192.64
                           IQQ  9)
    TOTAL S.S. = 662.79 -    ^    =  6.75


               _ (35.4)2 +  (33.O)2 +  (30.8)2   _ (99.2)2
      CAR S * S • —        ••- - .                     —
                 658.16 -  656.04  = 2.12

SOURCE OF
VARIATION
TOTAL
AMONG CARS
ERROR
A.
S.S.
6.75
2.12
4.63
O.V. SUMMARY
d.f. M.S. F
14
2 1.06 2.74*
12 . 386

F.05

3.89

    *Not  significant at 5 percent level of significance.
                           115

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 26 of 38
     Table 6-8.  ANALYSIS OF VARIANCE - TWO-WAY  CLASSIFICATION
          GAS CONCENTRATION MEASUREMENTS FROM THREE CARS
                  USING FIVE DIFFERENT GAS SAMPLES
GAS
SAMPLE
1
2
3
4
5
SUMS
SUMS OF
SQUARES
EMISSION MEASUREMENTS
CAR TYPE 1
6.6
7.2
6.4
7.4
7.8
35.4
251.95
CAR TYPE 2
6.6
6.4
7.0
6.2
6.8
33.0
218.20
CAR TYPE 3
7.0
6.0
5.0
5.8
7.0
30.8
192.64
SUMS
20.2
19.6
18.4
19.4
21.6
99.2
662.79
    TOTAL S.S. = 6.75  (from Table 6-7)
      CAR S.S. = 2.12  (from Table 6-7)
    GAS SAMPLE S.S. =
                                 (19.6)2+  (18.4)2+  (19.4)2+  (21.6)2
                                 (99.2)
                                   15
= 657.89 - 656.04 = 1.85
A.O.V. SUMMARY
SOURCE OF
VARIATION
TOTAL
AMONG CARS
AMONG GAS
SAMPLES
ERROR
S.S.
6.75
2.12
1.85
2.78
d.f.
14
2
4
8
M.S.

1.06
0.46
• 0.35
F

3.03*
1.31*

F
.05

4.46
3.84

        *Not significant at 5 percent level of significance.
                                116

-------
                                                            Section:   6(LG)
                                                            Revision:   0
                                                            Date:      June 1975
                                                            Page 27 of 38
only gases of interest, i.e., gas samples are a fixed factor.  As each
gas has been used with each car type, the data can be analyzed for
differences in gas emissions among gas samples as well as among car
types.  The results show that neither the car types or the gas sample
types produce a significant difference in the gas emission measurements
even though the error term has been reduced by accounting for another
possible source of variation.  In the first example (one-way classifi-
cation) the gas sample effects were included in {i.e., "confounded")
with the error term.  In actual practice, this other source of variation
should have been foreseen in the original design and set up as a two-way
classification model.  Reference 6-4 contains applications involving
random effects.
6.5       DATA VALIDATION

          Documentation of measured emissions should precisely and
accurately indicate the concentration of the exhaust gases being sampled.
Accuracy in recording data, however, depends on the recording techniques
implemented.  Methods that have been extensively researched, evaluated,
and controlled should have minimal error.
          Error due to human factors is one source of inaccuracy in
measurement reporting.  Human errors include  (1) incorrect reading of
instrumentation, (2) mistakes in computing results, and  (3) mistakes in
transposing data from one form to another such as keypunching errors
when computers are used.  Human error cannot be totally eliminated; how-
ever, it can be considerably reduced.
          Instrumentation is another source of error in documenting
measurements, and cannot be totally eliminated as there continually
exists a random inaccuracy for any measurement system, which cannot be
completely removed, as was discussed earlier in this section.
          Data validation involves the processing of raw measurement
data generated from emission measurement systems.  This processing
includes a critical review of data in order to locate spurious, docu-
mented values.  It may consist of cursory scans to identify any extreme
values, or extensive examinations requiring sophisticated data processing
techniques.  In either case, when a spurious value is identified, it is
not immediately rejected.  Rather each questionable value must be checked
for validity.
          Data validation can occur at different steps within the total
measurement process.  Additionally, there exist numerous data validation
techniques.  Among the most commonly used are:

          o    Impossible value sorting  (i.e., identify and eliminate
               data with impossible values)

          o    Improbable value sorting  (i.e., identify and eliminate
               data with improbable values)
                                117

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 28 of 38
          o    Identification of abrupt shifts in data levels

          o    Identification of stuck values

          o    Analysis of calibration data acceptability

          o    Use of computer data checks

6.5.1     Data Validation For Manual Techniques

          Specified, experienced laboratory personnel should inspect
testing data.  At regular intervals, daily or weekly, results should be
scanned for questionable values.  This type of validation is most sensi-
tive to extreme values, i.e., either unusually high or low readings.
          The criteria for determining an extreme value are derived from
required, specified values expected, and from prior data.  The data used
to determine extremes may be the minimum and maximum concentrations from
prior data or may be derived from control chart limits established in
accordance with the techniques outlined earlier in this section.
          The time spent checking data that has been manually reduced by
technicians depends on the time available and on the demonstrated abili-
ties of the technicians to follow the detailed computation procedures.
At this time no agencies appear to be using a specific formula for
determining how much data should be checked for validity in a manual
data reduction system.  One air pollution control agency approached the
problem in the following manner:  (1) a senior technician or supervisor
was assigned to check approximately 10 percent of the data interpreted
by each of four or five technicians.  The 10 percent figure was arbitrary
based on time availability and experience in finding errors.  (2) Data
was checked for obvious trends or unusual values indicating possible
reader bias.  (3)  No statistical formula was applied to determine the
significance of differences between readings interpreted by the techni-
cian and readings interpreted by the senior technician or supervisor.
If the two values differed by more than two digits in the last signifi-
cant figure, the data was judged unacceptable.   (4) Each analyst's
technique of data interpretation was checked against written procedures
describing the use of graphic aids to determine if those graphic aids
had been properly used.  The most significant errors originated from the
technician deviating from the written procedures - not from random
error (Reference 6-10).

6.5.2     Data Validation For Computerized Techniques

          Computers are used not only to store and retrieve data but
also to validate data.  Data validation requires the development of a
specialized computer program.  The techniques for identifying and sorting
extreme values in manual techniques also apply here.
                               118

-------
                                                            Section:   6(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 29 of 38
          The extent of the decision elements to be used in data vali-
dation cannot be defined for the general case.  Rather, the validation
criteria should be tailored such that they coincide with time, man-power
required, accuracy, and cost constraints.

6.5.3     Statistical Validation in Maintaining Data Quality

          A statistical analysis of historical data can be used as a
diagnostic tool in data validation.  For example, the total data history
of homogenous groups can be compared for relationships in spatial pat-
terns of results.
          The output from the emission analyzer device is often an
analog trace on a strip chart.  Reading strip charts is a tedious job
subject to varying degrees of error.  A procedure for maintaining a
desirable quality for data manually reduced from strip charts is impor-
tant.  One procedure for checking the validity of the data reduced by a
technician is to have another technician or the supervisor check the
data.  Because the values have been taken from the strip chart by visual
inspection, some difference in the values derived by two individuals can
be expected.  When the difference exceeds a specified amount and the
initial reading has been determined to be incorrect, an error should be
noted.  If the number of errors exceeds a predetermined number, all data
for the strip chart are rejected and the charts are read again by a
technician other than the one who initially read the chart.  Acceptance
sampling techniques are appropriate for use in such situations.  These
techniques and the theory of statistical sampling are discussed in
Appendix I of Reference 6-1, and Part III of Reference 6-2.

6.5.3.1   Outlier Analysis

          The treatment of outliers has had to be considered by every
data analyst who at some time or another has obtained a set of observa-
tions, supposedly taken under the same conditions, in which one observa-
tion was widely different from the rest.  The problem is whether the
suspect observation should be kept in the computation or whether it
should be discarded as being a faulty measurement.  During mobile source
emission testing, frequently one value within a data set may appear to
be considerably different from the other values.
          Many criteria have been proposed as guidelines in the rejection
of observations.  An excellent summary and critical review of the classi-
cal rejection procedures and some of the more modern ones is provided in
Reference 6-6.  A famous classical rejection rule is "Chauvenet's
criterion," which is based on the normal distribution and advises rejection
of an extreme observation if the probability of occurrence of such a devia-
tion from the mean of the n measurements is less than l/2n.  For a small
n, such a criterion rejects too easily, and a more appropriate test in such
                                119

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 30 of 38
circumstances would be  the Dixon  Ratio  Test (Reference 6-7).   This test
makes use  of only  the data in  hand,  and implements the statistics:
           If  X1  is  suspect         If X   is  suspect   Most sensitive
                                                       criterion when

              X2  " Xl                     X - X  ,
       =      x   _ x       or            _n	n^l
              n     1                      X  - X,        3 <; n £  7
                                           n     1
              A^.  ^ A_                    A   ~~  A  _
               21                     n    n-1
                           or
             X  .  - Xn                    X   -  X_       8 <;  n s  10
              n—1     1                     n     2
                X^ Y                     Y   ^  Y
               _   A-                    A   *™  A  _
        _       3    1                     n     n-2
   r21  ~     X  , - X.      °r            X  - X.       11 s  n < 13
             n-1    1                     n    2
             X3 " Xl                    Xn  "  Xn-2
   r22      X  _ - X.      °r            X  - X.       14 s  n s 25
             n-2    1                     n    3
X. denotes either individual values or means  of data sets  arranged  in order
or magnitudes from X. to X  .  It is assumed that the distribution of X or
X is normal.  In using this method, the samples from which the means are
computed should all have the same size.  The  critical values  for rin/ r   ,
r  , and r5_ can be found in Table W of Reference 6-4.
          An example using this technique would be to suppose that  six
vehicles of the same type were tested for CO  exhaust emissions.  The CO
emissions in prrts per million were as follows:

                                           CO Emissions  in
               Vehicle                    Parts Per Million

                 A                               510
                 B                               521
                 C                               523
                 D                               501
                 E                               493
                 F                               605
                                120

-------
                                                            Section:   6(LG)
                                                            Revision:   0
                                                            Date:      June 1975
                                                            Page 31 of 38
The problem is to test whether vehicle F belongs with others of the
group.  To perform the test r   is computed where
    .   =   6 "  5   =  6Q5 - 523     _82_
    10    v            605 - 493     112  ~
          X6 " Xl
The critical value is 0.56 for  a = .05 per the referenced tables.
Therefore, since the computed valued of r   (0.732) is greater than
0.56, it can be concluded that F should be judged different from the
others.  Note that this technique bases its conclusion solely upon the
six values and not on an outside measure of error.
6.6       METHODS OF CALIBRATION CURVE CONSTRUCTION

          Least squares, and Curveall (modified least squares) are
numerical analysis techniques which can be used to construct calibration
curves.  Although other curve fitting techniques exist, the above are
among the most commonly used.  This section describes general considera-
tions in constructing calibration curves, the theory behind each of the
above techniques, and how each can be implemented.  Additionally, the
pros and cons of each method are discussed.

6.6.1     General Context of Calibration Curve Construction

          Instrumentation provides a means for describing the contents
of a sample in terms of specific, quantifiable measurement data.  By
translating the sample contents into meaningful data a functional rela-
tionship is constructed; in the case of calibrating gas analyzers, meter
deflection or digital display is expressed as a function of sample
content.  Construction of calibration curves is the process of attempting
to mathematically duplicate the aforementioned functional relationship
using numerical analysis techniques.  Several of these techniques,
including least squares, are discussed in Appendix J of Reference 6-1.
          Consideration should be given to the following when constructing
calibration curves.  Usually regardless of the technique, the error
between some or all of the data points and the corresponding estimated
dependent variable value should be computed.  Such a practice provides
an indication of the generated curve's accuracy.
          In general, it is recommended that the most accurately repre-
sentative curve fitting technique (i.e., in terms of realistic system
response and standard's accuracy) for a given procedure be determined
                                121

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 32 of 38
through experience.  For example, experience  dictates that the response
of a CO analyzer is not expected to be represented by a sixth order poly-
nomial.  This technique should then continue  to be used for that procedure
providing that the hardware or procedure remain unchanged.  It is recom-
mended that the curve fitting technique not be continually changed so that
the generated curve best fits a particular set of data for a given pro-
cedure.  In other words, the procedure and hardware  dictate the type of
technique to use and not the data set generated each time an item is
calibrated.

6.6.2     Curveall

          The Curveall curve fitting technique is a  modified version of
the least squares technique discussed previously.  Using Curveall, a
polynomial of the following form is assumed
where  c  is  the independent variable, d is the  dependent variable, and
A, are the  coefficients that will be estimated using the least  squares
technique.  The A. coefficients are determined by minimizing the sum
of the squares of1the errors.  A detailed discussion of the Curveall
techniques  is contained in Reference 6-8.

6.6.3     Summary of Curve Fitting Techniques

          The aforementioned curve fitting  techniques each have distinct
advantages  and disadvantages.  Table 6-9 is a  summary of the techniques
in this  regard.

6.6.4     General Considerations

          The number of data points which must be obtained to derive
1975 FTP calibration curves is specified in the Federal Register.  The
number of data points is roughly dependent  on  the order of the  poly-
nomial which realistically represents the system response being plotted.
However, it should be noted that specific curve fitting techniques are
better to use in particular situations.  For example,  in the case where
the system  response is not linear, the Curveall or other non-linear
methods  would generate a more accurate and  realistic curve.
                                122

-------
                           Table 6-9.   MERITS AND DISADVANTAGES OF TWO CURVE FITTING  TECHNIQUES
                    MERITS
lo
                    DIS-
                    ADVANTAGES
                                              LEAST SQUARES
Smooths data into a continuous
functional response
                                   Computer processing time is
                                   relatively short compared to
                                   other techniques
Care must be taken to deter-
mine which order polynomial is
most appropriate; e.g., second
order may not represent true
instrumentation response
                                              CURVEALL
                                      (MODIFIED LEAST SQUARES)
Smooths data into a  continuous
functional response
                                   Forces  curve through the origin
                                  Third  order fit determined to be
                                  an  appropriate response
Curve may fit data too closely;
inflection points introduced
which may not reflect true
instrumentation responses*
                                                                     Computer processing time rela-
                                                                     tively, large
                    *This situation occurs when standard gases have significant inaccuracies.  Hence,
                     the curve incorporates these inaccuracies since polynomials closely fit data points.
                                                                          ^d c? 50 c/3
                                                                          (V PI fD CD
                                                                          iQ rt- < n
                                                                          (5 (D p- rt
                                                                            .. w p-
                                                                          ui   P- O
                                                                          u>   O 3
                                                                              3 ••
                                                                                                             U) C| O O^
                                                                                                             oo c   —

                                                                                                               CD   O

-------
 Section:    6(LG)
 Revision:   0
 Date:       June 1975
 Page 34 of  38
6.7       THE USE OF PROBABILITY PAPER

          In the previous sections covering control charts and analysis
of variance, it was often assumed that the compiled data formed a normal
distribution.  Through the use of probability paper, one can determine
what the form of the distribution actually is, whether it be normal,
Pdisson, etc.  In addition probability paper graphically illustrates the
cumulative distributions as they relate to compiled data.
          Probability paper is ruled so that the plot of some particular
distribution function will appear as a straight line.  Normal probability
paper  (commonly called "probability paper") will straighten out the
normal distribution as shown in Figure 6-7.  This paper can usually be
obtained in various forms from any good source of drafting supplies.
          As an example of the use of this paper, the values of the data
versus the cumulative percent frequency (Table 6-10) are plotted in
Figure 6-8.  In this particular example, the variability of CO emission
levels from a fleet of catalytic converter vehicles during a cold stabil-
ized portion of the Federal driving cycle is being examined.
          The following are the steps taken to plot data on probability
paper:

          1.   Arrange the observations in ascending values.  The
               smallest value is given a rank of 1 and the largest value
               a rank of n.

          2.   For each value, calculate the cumulative frequency.

          _    „      1.1      i   -i ^  cumulative frequency
          3.   For each value, calculate 	=—a	*-  x 100.
                                                n + 1

               This provides the mean rank probability estimate, in
               percent, for plotting the data.

          4.   Plot the observed values against their mean rank proba-
               bility estimate.

When the observations are in a frequency distribution form, the procedure
is the same except that instead of using the observed values, the proba-
bility estimates are plotted against the cell boundaries, as illustrated
in Table 6-10.  The plot is shown in Figure 6-8.  Lower cell boundaries
are plotted against the last column of Table 6-10.  A straight line is
drawn in by eye and the fit appears to be reasonable.  If the sample is
supposedly representative of a universe, then that line characterizes the
distribution describing that universe.  From it, one can obtain the
probability corresponding to any of the values included in the population.
                                124

-------
                                          Section:   6(LG)
                                          Revision:  0
                                          Date:      June 1975


8 rn |— T:i:~~: T"T^1 ^
e

N


in
0






° 	 _.:::..:::.:.: _j 	 :::: :.:::_..;
• i![!!!!!!!!!l!!!!!!![i!!!l!!lii!!!!!!
i

syEEEiEEEEEEEEEEEEEMEEEEEEEEEEEEEEEEE!:E^E
s •--'-•-- 	 — 	 1 — — i 	 i-rJ^J-ip
S: :W^i
Q' - .- :£. _____ :: __ /, . +
»* s ' -i —
e| . "S^r^
*:8 .3^=-^-+ + 	 E:i:EE:E:~::±?E^gpE
o __^l..t .... _ m

i_0 ^
~UJ ^
5* --------------- /
<5 m _ _ _ /
S° -J- -
r* 7
i-x _______ _|_x _
§flO _ I i.._ ^
	 1 	 l_i_l 	 j — _^_- . .
"--- - ~~ - j~ \ :~





' 7\
eo .
01 / i
o> /
" SE:pEr:pE:::E:E:::Er: 	 "TjT^^^
•^"""^j 	 ' " -^ 	 ' -- - -J 	 i~| 	 1— | 	 "I-*--


• /

T*
it"

:|=f = = iEiEEE=E------ vrg=EEEEEE|g^ g
/ eo
7 S
*

t
^
	 /
7-- •-- £
7
f
: ::_::::
••-T'TS^~±~T5"ii:!± 	 r^'-'l s

^ 	 	 - - - - ; - «
— rT'T7^ 	 -:: 	 T 	 T"^ — T 9

	 -r 1 | ' ' "^TT ' [ N 	 > 	 ' 	 = = ::= = rr: S
. 	 1 	 j 	 I 	








	 __ 	 _ 	 1 — | — U^ 	 ! — -fl 	 h-

! r J 111
_ . .t . .

*n
°

[ . ~
j °
i ^
	 ::; = ;p; = = +^--ip;p±:EEr 	 ^j^7: r •"
- ^_^ — p 	 ^- -j — | 	 • 	 r—i — ' 1 M 	 '!' ! - ^
_j 	 j~
IIIIIII
	
j_ U_ ± lUJJJiL — L s

Figure  6-7.   NORMAL  PROBABILITY PAPER
                 125

-------
Section:   6(LG)
Revision:  0
Date:      June 1975
Page 36 of 38

Additionally, in Figure 6-8, an ordinate at 12 ppm has been erected to
show that this technique predicts that 9 percent of future tests will
result  in CO emission levels less than 12 ppm.  The standard deviation
can  be  estimated by the perpendicular distance between the intersections
of the  50 percent and 84 percent abscissas with the graph line.
          Since the data in this case tends to form a straight rather
than a  curved line on the probability paper, one could conclude that the
sample  did form a normal distribution.  If the data tends to form a
curved  line, other types of probability paper could be used to determine
the  type of distribution the data actually form.
          Probability graph paper is available for the normal, log-
normal, experimental, Weibull and other probability distributions.  It
can  be  used to detect outliers, to derive control charts limits and there
are  many other applications which are adequately discussed in Reference
6-9.
                               126

-------
                                              Section:   6(LG)
                                              Revision:  0
                                              Date:      June  1975
                                              Page 37 of 38
Table 6-10.  TABULAR DESCRIPTION OF CO
        EMISSION LEVELS IN PPM
EMISSION
LEVEL
9.5-10.4
10.5-11.4
11.5-12.4
12.5-13.4
13.5-14.4
14.5-15.4
15.5-16.4
16.5-17.4
17.5-18.4
18.5-19.4
19.5-20.4
20.5-21.4
FREQUENCY
2
2
6
18
26
32
42
30
24
12
4
2
CUMULATIVE
FREQUENCY
2
4
10
28
54
86
128
158
182
194
198
200
CUM. '
(CUM FREQ)
n+1
99
1.98
4.95
13.86
26.73
42.57
63.37
78.22
90.10
96.04
98.02
99.00
                  127

-------
 Section:   6(LG)
 Revision:   0
 Date:      June 1975
 Page 38 of 38
~
o T^- TT T


s ::::::::::::::::::::::::::::: 	 	
o -f
o
M
O


	 	 "" 	 	 	 	 	 "















g :EEE||||||||EEE|E||:EE|E:|EEE::EEE||E|E:|EE:








5 --T 	 ::::::::••••:•• 	 3j 	

\


	 1 	 1 	 T * • >-

s \\\ 	 — 1 — : j: 	 	 1 	 :: j( +-1—
• \jf*
— _ ii _-__ ti
0 „_________ _j_ 	 	 . . 	 -J 	 J 	 | i IJ-i—^ 	 ^
IB
Xi



g — t--4- 	
1 i


in / J
"2ii "
' i
S :::::::::: :::::::::) I::::::!:: ::::::::::::::
_i ^ |- _
S " — :::::: :±:a J^:± :- — : :3^i = 	



"• I
S i
2:nMnEEEEEEE=!iEEE=EEMEEMEE = E = EEEEEE!EEE


S j. j, j, f , J_f
9 10 11 12 13 14 15
uuun s


^ """""ii" -j- -)--, — . — i —
rs P
T «• |T
§: c
	 _ _1_ 	 	 L _ _ R


E
/ : -. - f '•
- :::: : 	 : 	 t 	 ::
,
-f -" u
" s/ N
? „ D
::::_" i-~ 	 * F

:::::: _. t _ _ __ __ 	 i _ R

..,.,. 	 i\' o
. — ::::: c —


? T1 T1
:::::] j?::::::::::=:::::|::::::::=: :::::::::: s

2
|j


........ j ................ . . 1 ............ . ..,,,.

P: = :: — ! 	 M-^ 	 L_^_^^_^ .^ _ _ _^_ n_^ . Q
~~_ " " ^t


	 j 	 	 j 	

• - -. d -- «H* --

	 ::::::::; 	 	 S

. 	 1 	 1 	 	 _. 	 j 	




=============== ================ =!===!========"
:EE^Ei§^pgK^H:::::::=::^:








i:::^:::::::::::::::::::::::::::::::::::::::
::::::::::::::::::::!:::: ;E::::::E::E:::i5|
i =>


j_ °
- - - • ••
- 	 	 -pi 	 -J 	 	 	 	 - -j_ J 	 _ 	 j__j 	 I-T , ^.; - o
TnT

16 17 18 9 20 21 (PPM)
Figure  6-8.   NORMAL  PROBABILITY  PAPER (CO EMISSION LEVEL)
                              128

-------

-------
                                                            Section:    7{LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page 1  of  29
                             Section 7

            ANALYSIS OF VARIABILITY IN THE MEASUREMENT
               OF EMISSIONS FROM LIGHT DUTY VEHICLES
          The precision and accuracy of emissions measurements from
mobile sources are dependent on the variability that exists in the
vehicle or engine being measured and the system used to measure the
emissions.  The measurements made on light duty vehicles for 1975
certification testing are evaporative emissions, the mass emissions
of exhaust hydrocarbons (HC) carbon monoxide (CO), carbon dioxide
(CO ), nitrogen oxides (NO )/ and fuel economy.
          A  program to define and quantify each variable in the
vehicle and measurement systems would be extremely costly and time
consuming.  The possibility of determining all variables would be
suspect and many of those defined would prove either of little signifi-
cance and/or difficult to control.  Consequently many laboratories have
conducted programs to determine total test variability usually by the
use of a vehicle which has been especially prepared to reduce its
test-to-test inconsistencies.  (See Section 4.1(7)).  This vehicle
can be used to determine test-to-test variation within a single measure-
ment system, cell-to-cell, and/or laboratory to laboratory.
          Although the total test vehicle and measurement system
variability is of prime importance, it is essentially a composite
of all variables and will only be useful for measurement systems
which were actually involved in the program.  Measurement systems
not involved in these programs may use this data as a guideline or
goal to improve test-to-test reliability within their own laboratory.
However, knowledge of specific variables significantly affecting
the data is a prerequisite for achieving a predetermined goal or
improving data reliability.  These variables are either determinate
or indeterminate.  Determinate variables may be objectively studied,
but the nature of indeterminate variables requires subjective
evaluation.  Indetenninates are usually estimated through experience
with the measurement system, engineering evaluation of the test pro-
cedure and statistical analyses of data.
                                 129

-------
Section:   7(LG)
Revision:  0
Date:      June 1975
Page 2 of 29
          In addition to vehicle inconsistencies certain measurement
system variables have been considered as prime sources of error.
Efforts to reduce these variables include the use of instruments and
calibration standards having increased precision and accuracy.
          The purpose of this section of the guideline is to identify
these major sources of variation, to quantify the effect of the deter-
minate variables and discuss the involvement of Quality Assurance in the
reduction of test variables.
7.1       VARIABLES ASSOCIATED WITH THE MEASUREMENT OF EVAPORATIVE
          EMISSIONS

          Evaporative emissions are presently being measured by trapping
the vapors with carbon canisters.  The emissions are reported as the
total gain in weight during the evaporative•test described in Section 3.
The major variables in this test are the vehicle and the means used to
install the canister.
          Primary variables associated with the measurement system,
exclusive of the vehicle, are the connections to the canister, the
conditioning of the carbon, the proper installation of the canister with
a drying tube and humidity.
          The balance, when properly calibrated with class S-2 weight or
better, is not considered a prime variable.  Its accuracy is specified
as ±0.075 grams, and results are reported to the nearest tenth of a
gram; therefore, a slight change in accuracy would have little or no
effect on the reported value.
          Other sources of variability are the fuel specifications and
handling procedures.  The use of "weathered" fuel or fuels with an
incorrect Reid vapor pressure could have a significant effect on the
emissions.  Other methods of charging and heating the fuel tanks will
also have an effect.  Therefore, in developing a new procedure, careful
consideration should be given to these sources of variation i.e., the
vehicle, the method of collection of the vapors, fuel specifications,
and fuel handling and control.
          The accuracy of the Carbon Canister Trap method is presently
under study due to the disparity in data collected when compared with
that collected in two recent EPA surveillance programs employing the
SHED test method.  (Reference 7-1).  New test procedures are presently
under investigation by EPA to determine the evaporative emissions more
accurately.
                                130

-------
                                                           Section:   7(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page  3 of 29
7.2       VARIABLES ASSOCIATED WITH THE MEASUREMENT OF
          EXHAUST EMISSIONS

          The measurement of exhaust emissions involves several sources
of variables.  An estimate of the sources and relative contribution of
various factors to the overall exhaust-emission variability of a 1975-76
California-type vehicle are shown in Figure 7-1.  (Reference 7-2).   This
figure indicates that the greatest source of variability is the vehicle
itself.  In addition, "ambient conditions" and "driver" are shown as the
next highest relative contributors.  This is primarily due to their
effect on the vehicle rather than on the measurement system.
          However, the estimate of vehicle variability can best be  made
by first considering the variables inherent in the measurement system
and by subtracting these variables from the total variability.
          The variability of the measurement system will be relatively
constant on a test-to-test basis with a coefficient of variation of less
than 5 percent whereas the vehicle may exhibit variation as high as
30 to 40 percent.  (Reference 7-3).
          The variables in the measurement system can be divided into:

          o    Variables associated with the measurement process and
               reduction of data as determined from the accuracy and
               precision of the instruments.

          o    Variables associated with the equipment and test pro-
               cedures which are usually more difficult to quantify

          o    Variables contributed by the operator and driver

          These variables will be discussed separately, but as with any
complex system, there is a definite interdependence on many of the
sources of variation.

7.2.1     Analysis of Variables Associated with Measurement
          and Reduction of Data

          The various parameters measured during the exhaust emission
test are each a source of variation.  In order to assess their impact on
the output of the system, i.e., the mass in grams per mile, it is first
necessary to understand the process of mass calculation and to assign a
numerical variation  that would be encountered in the system.
                                131

-------
 Section:   7(LG)
 Revision:  0
 Date:     June 1975
 Page 4 of 29
     ui
     ui
     >


z
o
H
3
00
GC
O
O
UJ

Jj
_l









































CO
z
2
t
o
z
o
o

5 i
> m _
i z u
^_ * h  H uJ Q
0 «0 5 N C
*^ *fc 1
• ^ M ^ ™
o S ^ «
rf * (L













;
J
J £
f K
0 < " ^
n° ^

O
1 	 1
Figure 7-1.  Estimated  Sources of Variability  and  Probable Relative
         Contribution for Mass Emissions Errors on the  CVS
             Cold Start Test  at  1975 California levels.  (Ref. 7-2)
                                  132

-------
                                                            Section:    7(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page 5 of  29
          Since the method of calculation has gone through several
revisions, it is necessary to identify the mathematical model used to
assess the variability.  For this study the calculations as prescribed
in the Federal Register dated June 28, 1973, 17161, were used.  A com-
plete description of the test procedure is given in Section 3 and in
Volume II, the Test Procedures Manual.  A brief theory of Constant
Volume Sampler operation follows:
          The Constant Volume Sampler  (CVS) is a device which provides a
flow of a mixture of vehicle exhaust gases and ambient air at constant
volumes and temperatures.  Revolution counters monitor each turn of the
constant volume pump.  The total revolutions registered at the end of a
test are used to calculate the exhaust volume for that test.   (Engine
cranktimes, stalls, and dieseling will alter test time and therefore
affect total revolutions.)  This provides a method for" calculating the
total exhaust gas and diluent air at a constant temperature.

          Thus V .   = V  x N  
-------
Section:   7(LG)
Revision:  0
Date:      June 1975
Page 6 of 29
          At the inlet of the positive displacement pump, small samples
of the gas-air mixture are continuously extracted at a constant flow
rate, and are collected in a tedlar bag.  At the completion of the test,
the collected sample gas is removed from the collection bag and analyzed
to determine the concentration levels of hydrocarbon, carbon monoxide,
carbon dioxide, oxides of nitrogen and oxygen.  The dilution air is also
collected and analyzed so that the background pollutants may be subtracted
from the concentrations in the exhaust sample bag.  If these dilution
air pollutants were directly subtracted, a negative value could result
on low emission vehicles.  Therefore, to correct this anachronism a
dilution factor is calculated which estimates the portion of pollutant
contributed by the background air in the exhaust-air mix.  The following
formula is used to determine the dilution factor (DF) .
          DF
                 C02 +  (HC + CO) X 10~4
          Where CO_ is the carbon dioxide concentration of the dilute
exhaust in mole percent, HC and CO are the concentration of HC and CO in
ppm in the dilute exhaust sample.
          The corrected concentration of each pollutant is equal to the
concentration in the exhaust air mix minus the concentration in the
dilution air times  (1 - 1/DF.)
          Humidity correction factors are applied to the NO  measurement
and to the CO measurement when the moisture and C0_ are removed from the
sampling stream with a desiccant and ascarite.  The correction factors
can be found in the referenced Federal Register; however, the CO correction
is seldom used with the new low interference NDIR instruments.

7.2.1.1   Selection of the Mathematical Model

          The total mass of pollutant emissions in the cold transient,
hot transient and the stabilized portions of the test are weighted by
0.43, 0.57, and 1.0 respectively.  Since a significant fraction of the
total mass of HC and CO is usually collected in the cold transient
phases of the test, the cold start weighting factor plays an important
role in determining the final weighted mass .   A mathematical model was
chosen to give emissions values similar to the 1975 Federal emission
standards with approximately 60 percent of the total weighted emissions
being contributed by the cold transient phase.  This is not only representa-
tive of a real test but also allows a look at the effect of making
instrument range changes during the test to measure the lower contra-
tion.

-------
                                                            Section:    7{LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page 7 of  29

7.2.1.2   Effect of Variables in the Measurement of Ambient Conditions
          and the Calibration of Total Exhaust Volume and Mass Emission
          Values

          The total exhaust volume is calculated using the calibrated
displacement volume of the pump.  The effect of these measurements on
mass emission values is presented in Table 7-1.  Relative humidity is
determined (usually by wet bulb-dry bulb measurements) and is used for
correction of the CO and NO  values.
          V  - Volume per revolution of the positive displacement
pump.  For this model a coefficient of variation of 1 percent was
chosen, based on the expected standard deviations from a series of
propane injection measurements.  This coefficient of variation, however,
takes into account the other variables (P , N, P, & Tp) that are used
in the propane injection test.  Therefore, this could actually be
considered the variation in the determination of V  .  .-  Of the five
measurements used to determine V .  , the determination of V  is of
prime concern, since the calibration itself has a coefficient of vari-
ation in the order of 0.5 percent and is affected by  ambient conditions
and subject to deterioration with time.  The accuracy and precision of
the ambient and pressure measurement is easily achieved in most labora-
tories.  However, the determination of V  is a complex calibration
process, utilizing a laminar flow element  (LFE).  The LFE is cali-
brated to ±0.5 percent of the manufacturer's NBS standards and is
therefore a second generation standard.  Appendix III of the Federal
Register dated June 28, 1973  (Vol 38 No. 124 pp 17167-17168) has  an
extensive discussion of the calibration procedure,  the acceptable
tolerances for the calibration  equipment, and  lists several sources
of error.  If the calibration is performed carefully, deviations  from
a least squares plot should be  less than 0.5 percent.  This is
considered an adequate calibration  for achieving the  ±2 percent accuracy
required by the propane injection check.
          N - Number of revolutions.  A deviation of  ±100 REV. was
chosen since greater values would void the test.  In  actual practice
deviations of less than ±50 RPM can be achieved for the entire test.
The theoretical N is determined from a plot of AP Vs  RPM which is
determined from the pump concentration data.   The RPM is multiplied
by the test time.  It is therefore  important  that the time be measured
from the time the counter  is  started until it is turned off, and  not
just the test time which may  not  include crank time,  stalls, etc.  The
counters used in the measurement  systems are  considered accurate  to
±1 revolution.  A malfunction of  the counter  results  in a void  test
since the test count will  exceed  the tolerances by  a  large  amount.
Establishment of a chronological  data  file of test  RPM's would
provide useful information on real  pump variation.  This data could be
used to predict  "out of tolerance"  situations or need for recalibrations.
                                 135

-------
                                                                                                                                   *0  O
                                                                                                                                   Ol  B)
                                                                                                                                     sr
                              Table 7-1.   Effect of Variables  on the Determination  of Mass Emissions  -
                                            Ambient Conditions  and the Calculation of V_.
     ..  a>
     <  n
     H- ft
  ..  M  p.
00    H- O

O    §  "
Hi
                                                                                                                                   VO C-l O -J

                                                                                                                                      §   B
I
DATA INPUT VARIABLES WHICH


V - Volume
per revolution
of the posi-
tive displace-
ment pump
N - Revolu-
tions of the
pump per test
interval


Wet Bulb - Dry
Bulb - Ra
Temperature
Relative
Humidity-R


Barometer P






Pump Inlet
Depression P.




Pump Inlet
Temperature
T
''


EFFECT DATA

CVS Calibration
Rev. Counter
AP Measurement
Ambient Condi-
tions Pump
Counter
accuracy
Test timer



Precision of
temperature
(Hygrometer)
measuring
device.
Ambient condi-
tions
Bar. Calibra-
tion
Bar. Precision





Accuracy of
pressure
measuring
device


Calibration of
of temperature
sensor.
Recorder pre-
cision. Aver-
jcjing method.
STANDARD
DEVIATIONS
OF INPUT

±.00287
Cubic ft/Rev


±100 Rev.





±1°F
% R




±0.03 inches
Hg





±0.2 inches
H20




±3°F





STD DEVIATION
DETERMINED
BY

Ref. 7-8
Propane
Injection
data

Test void if
counts exceed
±100 from theo-
retical. RPS=23
4.35 sec/100
rev
Calibration
against NBS
reference
with accu-
racy of
±0.2 F

Calibrated
against a
mercury
barometer
with a read-
ability of
±0.01 or
better
Readability
multiplied by
4



Worst case
Visual
Integration




INPUT RANGE


0.2780 -
0.2808


1. 11615-11715
revs .
2. 19941-20041
revs.
3. 11615-11715
revs.
Dry Bulb 78-79°F
Wet Bulb 64-63 F
R 46.5-41.0%




29.25-29.22 inches
Hg





40-39.8 inches
H20




110-113°F





PERCENT
VARIATION
OF MASS MEAN

1



0.1-0.7





HC - None
CO - 0.2
NOX - 2.87




0.1






Less than 0.1





0.6





S>
COMMENTS
r™
^P
-j
V is determined from least
square plot of V vs X .
Linear relationship
V . = Total Exhaust Volume

Effect varies depending on
the portion of the test in
which it occurs and the
contribution of that portion
to the weighted mass

Greatest effect is on NO
due to the humidity cor-x
rection factor. No
effect on CO if the drier
columns are not required.


Error associated with the
determination of V i .
Linear relationship.





Water manometers are used
for this measurement with
either 1.00 or 1.75 sp.gr.
fluid with a readibility
of better than 0.05 inches
of water.
Variability will depend on
method used for integration
of average inlet tempera-
ture.


OS

-------
                                                            Section:    7(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  9 of  29
          Relative Humidity - The determination of relative humidity
to correct apparent increases in CO concentration caused by water vapor
is rarely done.   Recent universal use of CO instruments with very low
interference from water and CO  has resulted in the deletion of this
requirement provided the instruments meet certain minimum interference
response requirements.  (Reference 7-4)
          Variations in the determination of relative humidity by as
much as 10 percent have only a slight (0.2 percent) effect on the mass
emissions.  However, the effect of relative humidity is significant
when used to calculate the correction factor for NO .  The correction
                                                   j£
factor is applied to correct for the change in actual NO  emissions as
ambient relative humidity changes.  This correction factor reduces the
variability of the NO  emission data by normalizing to a standard
relative humidity  or 75 grains of moisture per pound .of dry air.
          Relative humidity is almost universally determined in the
emission laboratory using the wet bulb-dry bulb hygrometer.  Other
methods of determining humidity are available but attempts to correlate
the various methods have usually met with some unsolved discongruity.
Therefore, it is mandatory that the equipment used for humidity deter-
mination should be specified.  Two basic types are presently used:  the
fan-type hygrometer with either thermocouples or thermometers and
electronic or visual read out.  The other is the sling-type psychro-
meter.  These two types are known to give equal readings.
          A comparison of readings, on an audit basis, of these two
types could be used as a check.  The sling psychrometer is the pre-
ferred audit tool because of its portability.
          Other recommended methods of reduction of variability include
a controlled test lab environment, and continuous recording of humidity
during a test.  Wicks and water supply should be inspected frequently
for contamination.  Thermocouples and thermometers should have a cali-
brated accuracy of ±0.5 F or better.
          P  - corrected Barometric Pressure - The temperature compensated
aneroid barometers, calibrated against a standard  laboratory mercury
barometer are frequently used in the measurement system.  In laboratories
with only a single test cell a mercury barometer is often used.  The
two primary sources of error for barometer readings are in calibrating
the aneroid barometer and errors in the reading of a mercury barometer.
Calibration errors are generally controlled through independent checks.
Errors in reading the barometer can be reduced by  recording the pressure
before and after the test.  Comparison of the range of the two readings
could then be done by data validation or computer  utilizing one of  the
control chart techniques described in Section 6.   In  addition, comparison
to the reading of the previous test on the same day would provide an
additional check.
                                 137

-------
Section:   7(LG)
Revision:  0
Date:      June 1975
Page 10 of 29

          P_., T^ - Pump Inlet Depression, Pump Inlet Temperature -
Measurement of pump pressure and temperature are highly accurate in
most laboratories and are not considered a significant contribution to
test variability.  Significant error would only occur through a failure
of the pressure or temperature measuring devices, which would be evident
through periodic propane checks.  TheAP of a pump will vary with time
and use and is an indicator of the condition of the CVS system.
          In summary, the error in measurement of ambient conditions is
most critical for the determination of NO , but has only a slight
effect on the determination of total exhaust volume.  Total exhaust
volume (V . )  is affected primarily by the determination of V  and
the variation in the positive displacement pump and other CVS com-
ponents.  This test variation is best controlled by careful calibration
of the pump, propane injection check and monitoring of the differential
pressure across the pump.

7.2.1.3   Variation in the Determination of Exhaust Emission
          Concentrations

          Exhaust emission concentrations are determined using an
analytical system calibrated with gas mixtures which have a specified
accuracy of ±2 percent.  Usually instrument curves are constructed
with gas mixtures having accuracies of ±1 percent or better.  Gravi-
metric standards prepared and used by the EPA have a reported accuracy
of ±0.5 percent or better.  In addition, reference standards are available
from the NBS (SRM's 1665-1669, 1673-1675, 1677-1681, and 1683-1687).
Instrument precision and reproducibility are specified by the Federal
test procedure and through experience have been found to conform to
these specifications when properly maintained.  Successive analyses of
the same sample give a precision of ±0.5 percent of the full scale
concentration.  (Reference 7-3)
          The primary sources of variability in the analytical system
are:

          o    Accuracy of the calibration gases
          o    instrument precision
          o    Accuracy of working or span gases
          o    Calibration curve construction
          o    Condition of the sampling system
          o    Full scale concentration
          o    Zero gas impurity
          o    Instrument drift (electronic)
          o    Operator
          o    NO  converter efficiency
          o    FlB Fuel
                               138

-------
                                                             Section:    7(LG)
                                                             Revision:   0
                                                             Date:       June  1975
                                                             Page 11 of  29
          The variables are controlled through a system of audits,
performance and receiving inspection checks, etc., previously described.
Detailed procedures for these appear in Volume II, the Test Procedure
Manual.  For determination of the effect of error in concentration
measurement, a coefficient of variation of 1 percent of the full scale
concentration was chosen.  However, variation between analytical systems
has been experienced as high as ±3 percent for the same sample.  Corre-
lation values in excess of this are, however, considered to be undes-
irable and suggest a need for corrective action.  Corrective action
usually involves a system leak check, reanalysis of the working gas
and construction of a new instrument curve followed by a systematic
check of the sources previously mentioned.
          An error in the measurement of an exhaust component would
obviously have a corresponding direct effect on the mass emissions.
However, because of the weighting factors, choice of instrument ranges
and measurement of emission components in the dilution air the effects
are not always readily apparent.
          Therefore, it was decided to vary not only the concentration
but also the ranges and in some cases only the dilution air readings
to assess their impact on the weighted mass measurement.  This data is
summarized in Table 7-2.  The concentration list in part A of Table 7-2
along with the data from Table 7-1 were used to calculate the "true"
mass values listed in the table.
          The concentrations in Bag 1, the cold transient phase were
varied by almost a factor of 10 from Bag 2  (cold stabilized) and
Bag 3  (hot transient).  This resulted in approximately 60 percent of
the total weighted mass being contributed by the cold transient phase.
The CO2 values were not varied by the factor of 10, however, since this
would not be realistic.  The C02 concentration was chosen to give a
dilution factor of approximately 8.
          In the first case  (Table 7-2, part B) the sample concentration
was varied by 1 percent of the maximum concentration of the range shown.
Ranges were selected to reflect, as near as possible, those ranges that
would normally be used in an actual emission test.  The resultant change
in concentration are as would be expected - less than 1 percent for
each value.  Table 7-2, part C, shows similar variation when varying only
the background air by the same  1 percent.  Since the concentrations as
measured are in the upper part  of the range curve, varying  the measured
concentrations for exhaust and  dilution air has a  small effect on
weighted mass.
          Now let us assume that the operator measures Bag  2 on the
same range as Bag 1 and  the concentrations  are  varied by  the same
value  of 1 percent of full scale.   The effect  is much more  pronounced
for both exhaust and dilution air  (Table  7-2,  parts  D & E)  giving a
deviation as high as 3.8 percent.   In part  F,  Table  7-2,  the ranges
are lowered  to give more accurate  readings  and the effect becomes more
like that experienced in Bag  1  (Table  7-2,  part B).
                                 139

-------
Section:    7{LG)
Revision:   0
Date:       June 1975
Page 12 of 29
                     Table 7-2.  EFFECT OF VARIABLES ON THE DETERMINATION OF MASS EMISSIONS -
                           MEASUREMENT OF DILUTED EXHAUST AND AMBIENT AIR CONCENTRATIONS

             A.   Exhaust Sample  Concentration Used to Determine Basic Mass Value


HC ppmc
CO ppm
co2%
NO ppm

Bag 1
373
1900
1.44
237.5

Bag 2
47.4
215
1.60
25.8

Bag 3
75.2
359.2
1.60
43.6
Background
All Bags
10
15
0.04
0.5
Mass Value
g/raile
1.49
14.96
758
3.14
             B.  Exhaust Samples  varied by  1% of Full Scale  for  the  Cold Transient Phase Only
                               Bag  1            Range             Mass Value            % Change
             HC ppmc              377            0-400                1.50                   0.7
             CO ppm              1930           0-3000               15.10                  0.94
             CO  %               1.48               0-4                761                   0.4
             NO  ppm              240            0-250                3.16                  0.63

             C.  Background Air Varied by 1% of Full Scale for the Cold Transient Phase Only
                             Background Air         Range           Mass Value          % Change
             HC ppmc              14                0-400             1.48                0.7
             CO ppm               45               0-3000            14.84                0.8
             C02 %                .08                   0-4              755                0.4
             NO  ppm              3.5                0-250             3.12               0.63

             D.  Exhaust Sample Varied by 1% of Full Scale for the Cold Stabilized Phase only
                 Using Same Range as the Cold Transient Analysis
                               Bag  2            Range             Mass Value            % Change
             HC ppmc             51.4            0-400                1.53                 2.7
             CO ppm               245           0-3000                15.53                 3.8
             C02 %               1.64               0-4                 771                 1.7
             NO  ppm             28.3            0-250                3.22                 2.5
               x

             E.  Background Air Varied by 1% of Full Range,  using  the Higher Range,
                 for the Stabilized  Phase Only

HC ppmc
CO ppm
co2%
N0x ppm
Background
14
45
.08
3.5
Range
0-400
0-3000
0-4
0-250
Mean Value
1.46
14.45
745
3.05
% Change
2.0
3.4
1.7
2.9
             F.  Exhaust Sample Varied by  1%  of Full Scale for the Stabilized Phase only, using
                 a Lower Range than used in the Cold Transient Analysis except for CO
                               Bag 1             Range           Mass Value             % Change
              HC ppmc           48.4             0-100               1.50                  0.7
             CO ppm              220             0-500             15.06                  0.7
             CO  *              1.62               0-2               764                  0.8
             NO  ppm            26.3              0-50               3.15                  0.3

-------
                                                            Section:    7(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page 13 of  29

          This shows the importance of selecting appropriate ranges for
the measurement of exhaust samples.  Variability can be reduced by
measurements made on the lowest possible range.
          Ambient-dilution air concentrations are usually measured on
the same range as the exhaust concentration since going to the lower
range would require an additional time consuming span and zero check.
Error reduction in measurement of dilution air concentrations becomes
more important as the exhaust concentrations decrease with improvements
in emission control devices.  Since the range change is not convenient
and further increase in instrument precision is unrealistic, other
sources must be considered for decreasing the variability.  Curve
construction should be considered a source of error for dilution air
measurements.  The method of curve generation chosen becomes more impor-
tant in respect to its definition as it approaches zero.  Some instru-
ments, for example, although they may be linear over several ranges will
exhibit nonlinearity as they approach zero.  In the past this was not
considered a problem as most measurements were made above 50 percent of
scale.  Now it becomes important that calibration gases of similar
concentrations to ambient air be used to define the lower end of the
range scale.
          Other sources which need further control are the instrument
zero drift, which should be checked periodically, and the contaminants
in the zero gases.  Nitrogen and air zero gases should be rigorously
analyzed by the receiving laboratory rather than the present practice of
accepting batch analysis from the supplier.
          Along with reducing contaminants in zero gas, the reduction of
contaminants in laboratory ambient air concentrations should also be
considered.  Humidity control and "make up" air units can help control
ambient conditions.  In addition, adequate removal of vehicle exhaust
from the diagnostic area, prohibiting the starting or driving of cars in
the test area and inspection of the heating system for leaks and proper
ventilation, will all help in achieving more desirable ambient conditions.
          Because of the variety of available certified accuracies for
calibration gases, a decision must be made based on cost versus relia-
bility desired when obtaining the laboratory standards and  "working
gases."  Naturally, as the certified accuracy of the blend  is improved,
the cost of the gas increases exponentially.  In all cases, however,
traceability to the EPA primary standards either through correlation
programs or by direct analysis by EPA is desirable.
          Taking into account the variables which are known to be
encountered in the measurement system  (Tables 7-1 and 7-2), it  is
apparent that for most measurement systems, excluding the vehicle
variables, an overall variability of 2-5 percent could be achieved.
This applies to variation only within a single measurement  system and
depends on the degree of control applied to the source of variability.
                                lUl

-------
Section:   7(LG)
Revision:  0
Date:      June 1975
Page 14 of 29
7.2.2     Variation Associated With The Equipment and Test Procedures
          Used in the Measurement System

          The equipment used in the measurement system, other than the
instruments and calibration standards previously mentioned, also contribute
to test variability.  These equipment variables are:

          o    Dynamometer
          o    Driver's Aid
          o    Constant Volume Sampler
          o    Computer
          o    Sample Handling System

Dynamometers

          A dynamometer attempts to simulate the road load at the
driving wheels of a vehicle.  The road load essentially consists of a
friction force independent of the speed, tire rolling resistance as
a linear function of the speed, and windage as a square of the speed.
In addition, the net force due to inertia.,  i.e. mass times acceleration
or deceleration, is also simulated.  Various types of dynamometers,
such as electric, direct drive, and belt driven, are used in light-
duty vehicle testing.
          Belt-driven dynamometers are subject to belt slippage, which
causes a greater deviation on a test-to-test basis than the direct-
drive type.  Consequently, many laboratories are converting to direct-
drive dynamometers to reduce this source of variability.
          Formerly calibrations were performed for a single inertia
weight with only one check at the other weights at the same horsepower
setting.  It has been determined that running the complete curve for
each weight reduces the variability.  Weekly quick checks of the
dynamometer can be used to detect changes in the calibration.
          Calibration of the speed and torque meters should be performed
each time a complete calibration is run.  The coefficient of variation
has been estimated as ±2.5 (Reference 7-3) at the EPA laboratories
and as high as 3 to 4 percent by a major automobile manufacturer.
(Reference 7-5)

          Emission test variability caused by the dynamometer can be
attributed to:

          o    Error in horsepower and inertial setting

          o    Differences in dynamometer absorber and friction
               characteristics (between dynamometers)
                                142

-------
                                                           Section:   7(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 15 of 29
          o    Differences in slip and resistance at the
               tire-roll interface

          o    Roll spacing and roll size differences

          o    Differences in extent and type of maintenance and the
               method of calibration (between laboratories).

          Very little can be done to reduce the variability associated
with the dynamometer except for proper calibration, maintenance and
checks.  Correlation of the different types could be achieved in an
extensive program involving a large sampling of different vehicle types.
Such a program would have to consider the fact that the variability of
the vehicle exceeds that of the dynamometer.  The use of correlation
type vehicles could possibly establish some significant data in less
extensive programs, such as comparing the belt driven to the direct-
drive dynamometer.

Driver's Aid

          Basically driver's aids are of two types - the preprinted
chart paper and the computer-printed driving cycle.  In both cases
one channel of the recorder is coupled to the speed output of the dyna-
mometer.  The driver's aid is not usually considered a significant
source of variability? however, if improperly used and/or maintained
it could introduce test error.  Variability could be caused by:

          o    Chart paper slippage
          o    Recorder speed calibration
          o    Computer malfunction
          o    Failure to start and stop at the correct intervals
          o    Failure to properly zero recorder
          o    Incorrect length of driving cycle
          o    Chart speed variation

          Most of the problems would be evident from incorrect test
times.  All can be controlled through routine audit, calibrations, and
maintenance of the computer and recorder.

Constant Volume Sampler

          There are other sources of test variability  in addition to
the previously mentioned V  associated with pump calibration and
                                3A3

-------
Section:   7(LG)
Revision:  0
Date:      June 1975
Page 16 of 29


temperature variability.  During a test transient pressure fluctuations
occur in the system prior to the positive displacement pump.  If they should
continue for a significant amount of time and are not properly weighted
test error would result.  When these fluctuations seem to occur with the
particular vehicle being tested, continuous monitoring of inlet pressure
should be implemented.
          Deposits on the rotor lobes may cause varying blower flow
characteristics resulting in uncertainty in the calibrated displacement
volume.
          Disproportionate sampling into bags may result from pressure
and temperature fluctuations ahead of the sample pump, introducing
errors in sample collection.  Visual monitoring of the flowmeters during
sampling will aid in decreasing this variability, which usually occurs
with increasing age of the pumps and can readily be detected through
propane injections with time intervals equivalent to the driving cycle.
Disproportionate sampling results in bag samples which do not accurately
represent the flow averaged exhaust sample.  Leaks in the system should
be avoided, because they can dilute the samples.  However, if they occur
on the pressure side of the bag sample pump, they will have no detectable
effect on the results.  Leaks in the CVS blower system would not be
significant unless they occurred after the sample probe and before the
positive displacement pump because they might affect the V .  calculation.
          Condensation in the heat exchanger can result from^highly
humidified dilution air, together with lower dilution ratios occuring
during periods of high exhaust flow rates.  This can change character-
istics of the heat exchanger (pressure drop in the heat exchanger)  and
is sometimes evident from the appearance of water droplets in the sample
bag.  Humidity control of dilution air and higher CVS flow rates would
eliminate this problem.  The result is inaccurate measurement of V .  .
          Stratification, incomplete mixing, and sample probe locations
were all considered likely sources of error in the early days of CVS
testing, but with the present design of the system these problems
seldom occur.
          The Coordinating Research Council has conducted a study on
CVS testing and issued a report (Reference 7-6).  The "Recommended
Practice" contains much useful information concerning problems associ-
ated with CVS testing, corrective action, calibration, and theory.

Computer

          Computers, with their built-in checks and reliability, are
very useful in reducing test variability.  The variety of computers
used in mobile source testing ranges from "desk top" to completely
automated systems.  Although the computer is generally more reliable
than manual operations it is not infallible, therefore, it requires
periodic reliability checks.  One proven method of checking data

-------
                                                             Section:    7(LG)
                                                             Revision:   0
                                                             Date:       June 1975
                                                             Page  17 of  29
reductions is the use of a previously prepared standard set of manually
calculated data.  This is fed into the computer and the output is com-
pared with the actual value.  This same set of data normalized for the
curves stored in the computer could be used in a cell to cell or labora-
tory to laboratory correlation study.

Sample Handling System

          The primary sources of variability in the analytical system
are the calibration gases and instruments previously mentioned.  However,
other sources of error exist in the process of transferring the sample
from the bag to the instruments.  Leaks in the system, "hang up" of
hydrocarbons in the line, incorrect or fluctuating flow rates, or
component failures could cause test error.  Preventive maintenance,
cleaning and leak checks of the system will reduce these sources of
error.  In a laboratory with more than one system a weekly cross-
check of a bag containing an exhaust sample would be a valuable tool
for reducing the variability of the analytical system.
          In the analysis of NO , the NO  in the sample must be converted
to NO before it is introduced into the Cnemiluminescent-NO instrument.
The efficiency of this conversion is an obvious source of error and
should be checked periodically.  Weekly checks are recommended, however,
experimentation has shown that daily checks improve reliability, detect
discrepancies faster, and result in fewer voided tests.  This is another
area where a data file of efficiency checks could be used to spot
potential problems.
          In summary, the total equipment variability is somewhat
indeterminable.  Estimates of various equipment variables have been
made (Reference 7-3, 7-5), but vary from one laboratory to another.
Experience and data from a particular measurement system will allow
an estimate of the variability which will most likely be found in the
range of 3 to 5 percent.  Variability in excess of this would indicate
a need for corrective action.

7.2.3     Emission Measurement Variability Contributed by the
          Operator and Driver

          The probability of occurrence of random errors in areas of
analyzer calibration and other test equipment setups is a function of
the level of training and education of the human operators.  Even with
highly skilled technicians, human errors do occur, making the test
operator a possible source of variability.
          A driver attempts to minimize the error between the actual
and required speeds with the help of the accelerator or brake application.
Rapid modulations of the accelerator introduces a dynamic state that is
likely to produce more emissions.  For example, more frequent operation

-------
 Section:    7(LG)
 Revision:   0
 Date:       June 1975
 Page 18 of 29
 of the power enrichment system in the carburetor  during an EPA test may
 result in higher CO emissions.   Therefore,  the  driver's inability to
 exactly duplicate  the  required speed or alternatively,  the difference
 between erratic and steady drivers makes the driver  an  important source
 of variability particularly for manual transmission  vehicles.   The
 driver variability has been estimated as high as  4 percent for the
 1975 procedure, but would  fluctuate from one driver  to  another.
           Data validation  plus periodic operator  and driver evaluations
 are usually  effective  in detecting and controlling these random errors.
 These  must be classified as indeterminate errors, since any attempt
 to measure them would  be biased.   However,  one  effective method  would
 be the construction of control charts from  data files on each  operator/
 driver, plotting the number of void tests due to  human  error versus
 number of  tests run by each opera tor/driver.  The number of void tests
 should be  reviewed continuously as trends or even sudden changes in
 void rate  will indicate a  need for investigation  and possible  corrective
 action.
7.3       MEASUREMENT OF VARIABILITY  IN EMISSION MEASUREMENT SYSTEMS

          Variability of the measurement  system is defined as the
inability to achieve identical test results from repeated tests on the
same vehicle without changes to hardware  or vehicle adjustments specifi-
cations.  Variability exists in .test  results to varying degrees dependent
on the type of variability, test-to-test,  cell-to-cell within a labora-
tory, or laboratory-to-laboratory.
          A discussion of the importance  of determining variability
and its effect on the automobile manufacturers has been presented by
Ford (Reference 7-5) and General Motors  (Reference 7-7) in the applica-
tions for suspension of the 1977 Federal  Emission Standards.  As the
emission requirements become lower, the level of variability signifi-
cantly affects the ability to develop and certify emission control
systems.  Variability factors are affected not only by the vehicle,
but also by the test-to-test variability.  It is important, therefore,
to determine the expected variability to  ascertain the actual levels
of exhaust emissions for certification of emission control devices.
Consideration is given in these reports not only to the "in house"
variability, but also to correlation  factors which exist between the
manufacturer's laboratory and the EPA laboratory.
          Variability in emission measurement systems is usually
expressed as the coefficient of variation which is defined as the
standard deviation  (s) divided by the mean of the results, expressed
as a percentage (CV =  =• (100) percent).   Also variability may be
defined for some confidence level, for example, to assess the vari-
ability associated with a 90 percent  confidence level, the standard
deviation times 1.645 is added to and subtracted from the mean.  For
the 95 percent confidence level, 1.96 is  used as a multiplier in a
similar calculation.
                               Ik6

-------
                                                            Section:    7(LG)
                                                            Revision:   0
                                                            Date:       June 1975
                                                            Page 19 of 29

In other words, as the confidence level is increased, the confidence
interval becomes wider.  Therefore, in the case of a certification
vehicle the higher the confidence level selected the more efficient the
emission control system must be in order to obtain the emission values
required to be statistically certain that all vehicles will meet the
Federal Emission Standard.
          Listed in Table 7-3 are Coefficients of Variation as deter-
mined by several sources, referenced in Column 1 of the table.  This
represents a sample of the actual variabilities that have occurred in
various emission laboratories.  The methods used to determine varia-
bility differ from one source to another and also will differ with the
type of engine emission control system being  .aasured.  (Reference 7-3)
For further details of the programs and type of data the listed references
should be consulted.
          The data in Table 7-3 shows generally high variability for HC
and CO, and less for NO .  The determination of CO  exhibits the lowest
variability.  However, £he range of reported variability differs greatly
depending on the source and methods used to determine or estimate the
standard deviation, and the reference standard or mean  (1975 or 1977).
          Variability also exists in the vehicle population as exhibited
in the following example developed from certification data containing 35
different vehicles as reported in the Federal Register  (Reference 7-
10).

                                   H£        CO        NO^


     Mean  X                      0.846     7.914      2.440
     Standard Deviation          0.285     3.128      0.364
     Coefficient of Variation    33.7      39.5       14.9

           It is interesting to note the variability of HC and CO is
about equal and NO  is less by about half.
           The sources of  error in  the measurement  system would be
incomplete without consideration of the vehicle itself.  If a vehicle
could be controlled sufficiently to produce  identical tail pipe emission
concentrations each time  it was tested, variability would be reduced  to
a minimum.  However, this is  not the case, therefore  the methods used
to control changes in vehicle emissions are  important.  Two of  the
major sources, the dynamometer and the driver have been previously
discussed.  Other  sources of  variability  are:

           o    Engine, carburetor  design
           o    Emission  Control System

-------
           Section:    7{LG)
           Revision:   0
           Date:       June 1975
           Page 20 of 29
                       TABLE 7-3..  SELECTED VEHICLE EMISSION TEST
                            VARIABILITY PROM SEVERAL SOURCES
SOURCE
(Ref)
EPA
(7-3 -Table 2)
EPA
(7-3)
EPA
(7-3 File 8)
AC Spark Plug
Div. of CMC
(7-8)
HAS Report
(7-9 P 15)
TEST DESCRIPTION
29 Tests 1971 Ford
Estimate of Test to Test
Variability
7 Tests 1974 Vega
Estimate of population x + s
For a single Vehicle
1975-76 Vehicles estimate
of variation
Coefficient of Variation (
HC CO NO
14.2 11.5 6.5
6 63
11.6 31.1 4.3
15.1 19.7 28.6
10-25 15-30 5-15
%)
co2
2.1
1
1.13
"•
-
Ford  (7-5)


Ford  (7-5)


General Motors
 (7-7)
EPA
(7-3 Table 3)

Honda CVCC
(7-9 p 152)
1975 Emission Level Vehicles        8.8       11.4     13.4
198 Tests

1977 Tail Pipe Emission Level      19.2       27.2      8.1
Vehicles                           18.2*      24.3*     8.1*

Estimates of the limits of          39        50       15
variability at 95% confidence
level for 1977 vehicles
1975 FTP maximum variability       37.5
at 90% confidence level
35.5     22.6
10 Tests                            8.1        5.4      4.7
8.7
                   2.5
                   *The Variability of emission
                    data with corrective action
                    determined by estimating the
                    reduction in variability due
                    to improvements in equipment
                    and procedures at the same
                    mean level
                                          148

-------
                                                            Section:   7(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 21 of 29
          o    Malfunction of engine control system component
          o    Ambient temperature
          o    Humidity
          o    Altitude and Barometer
          o    Fuel

          These variables are difficult to control.  They are discussed
extensively in the references cited in this section.  However, a brief
discussion is given to illustrate their contribution to test
error.

Engine  and Emission Control Systems

          The variability of test data for four different types of
engine  - emission control systems has been reported (Reference 7-3)
and is  depicted in Figure 7-2.  This data compares the effects of the
conventional engine control systems, diesel fuel injection engines
systems, stratified charge combustion systems and oxidation-catalyst
control systems on exhaust emission data in the separate portions of
the 1975 Federal test procedure.  The reduction of variability for
CO by proper range selection is shown for the diesel engine.

Malfunction of Engine Control System Components

          A malfunction of these components obviously causes additional
error in the test procedure although the magnitude would, of course,
be indeterminate.  Therefore, care in inspection of the vehicle and
preconditioning is important in reducing this type of error.

Ambient Conditions

          Barometric pressure, humidity, temperature, and air circu-
lation in the test site affect emission results from a vehicle, and
any variation in these parameters introduces variability.  For example,
a lower barometric pressure is accompanied by enrichment of the mixture
from the carburetor providing an ideal condition for higher CO concen-
trations.  The extent of enrichment is dependent upon sensitivity of
the carburetor.  Humidity or inlet air moisture affects air-fuel
(A/F) ratios and peak cycle temperatures in the engine, which in turn
affect emissions.  Ambient temperatures and air circulation around the
vehicle can affect fuel temperature in the carburetor and in the fuel
tank, thus affecting A/F ratios and evaporative system interaction with
carburetion, with resulting changes in emission results.  Although a
NO  humidity correction factor does exist  for the  1975 EPA procedure,
different sensitivities for HC and CO for  different emission systems
precludes the possibility of developing any correction factors
for barometric pressure, humidity, and temperature.  The ambient
conditions are therefore important sources of emission test variability
and must be controlled as rigidly as possible, within practical
limitations.

-------
Section:    7(LG)
Revision:   0
Date:        June 1975
Page 22 of 29
                                     Figure 7-2

                         Variability associated with each component measured during
                   each phase of '75 FTP for different control systems and use of propel
                   instrumentation ranges to reduce variability
                                 nc
CO	C02
Eag. Mod.

Diesel

Stratified

Catalyst


P
23 2.5
5«
223 5
•••
•••••
S^ .9
••MM
•
•M
•zm .9

ST
zza 16.7
F~
tll) __mc|« CD
La a "
••
•i
^
^ 9.1
••••
••
•••i
ZZ2Z3 5.B
• ' •
1
|
1 B4O
l
r
1 4O6
1
399

98A

I
a 4.7
1
1
3 14
••
Ml
••
223 .9
•
•
•
V

                                                                 Stf 2
                              »»JO   0  M 2030    OB   0
                                Varlatloo   Coefficient
                                           150

-------
                                                            Section:    7(LG)
                                                            Revision:   0
                                                            Date:      June 1975
                                                            Page 23 of 29


          Data that has been reported in a study of the change in emis-
sions with temperature appears in Figure 7-3.  Since the test is always
run between 68 and 86 F, effects outside this range are of little concern.
However, an allowable range of 18 degrees is a large interval and emissions
measured at the extreme ends are known to give different results.
          For example, it is the practice at some laboratories to conduct
their emission testing and diurnal soak in the same area.  The tempera-
ture range for the soak is 76 to 86 , consequently the emission tests
are run usually between 78 and 84  (note a ±3 degree variation is normal
for an average heating and cooling system) .  If this same vehicle were
to be tested in a laboratory where the average temperature was around
70 F, different emission results would be obtained.  Therefore, for
correlation purposes, tests should be run in the same temperature range,
preferably with the best temperature controls obtainable within practi-
cal and economical limits.
          The results of tests carried out in an environmental chamber,
designed to evaluate the sensitivity of exhaust emissions to barometric
pressure and humidity variations as reported by General Motors are
summarized in Table 7-4 (Reference 7-9).  For the sake of comparison,
Ford Motor Company data based on multiple regression analysis of three
vehicles tested on the FTP-H  (hot start) test cycle are also included.
This same effect would be expected when testing vehicles at high altitudes.

Fuel

          The fuel used in testing the vehicle is often overlooked as a
potential source of test error.  Test  laboratories presently have a
choice of three fuels popularly known  as 91 Octane, Indolene 30 and
Indolene Clear  (HO).  The specifications for these fuels are regulated
by the EPA, however, this does not assure that the fuel obtained from
the supplier, the tank or barrel, or fuel conditioning cart meets these
specifications.  The results  of using  leaded fuel in a catalyst vehicle
has been well publicized.  Foolproof controls must be implemented to
preclude the use of the wrong fuel.
          Other characteristics known  to have an effect on emissions are
the Reid Vapor Pressure (RVP), octane  rating and hydrocarbon composition.
The RVP affects the starting  ability of the car and the evaporative
emissions.  The RVP can be changed through improper storing, overheating
of the fuel, age, and improper handling.  The use of  "weathered" fuel
can cause starting difficulties and, therefore, fresh fuel should always
be used for emission tests.
          In view of these potential sources of test error, fuel
received in a laboratory  should be tested  for conformance to specifi-
cations and should not be released for use if the results of the test
differ from the specifications.  Storage drums  should be clearly
marked and color coded.   Special nozzles are required for catalyst
vehicles.  Care must be taken to contain each type of fuel in  separate
storage tanks, with thorough  drainage  of a tank prior to filling with
another type of fuel.  For  example,  it might be conceivable that a
                                 151

-------
                                                                                                                                 nj o ys en
                                                                                                                                 P» B> (B rt
                                                                                                                                 in ft < n
                                                                                                                                  W   H- O
                                                                                                                                  *»   OS
                                                                                                                                  N) C| o -J
                                                                                                                                  VD C   -»
                                                                                                                                    VD
                                                                                                                                    -J

                                                                                                                                    U1
111
I  '
 0)
 IU
<  »

to

Q

O
u
O
n
IE
<
o
o
X   i
                          • 4 *»IMVM CNMMH Cm I* uiw
_J

i

3
                                             CO
                                             in
                                             (9
o
u

CO
>
u

III
g
x
O

o
                                                70
                                                60
                                                SO-
                 40     GO     80     100


              TEST AMBIENT TEMPERATURE I'f I
                                          120
                                             O



                                             ff


                                             CO

                                             o
                                             10

                                             u



                                             u
                                             o
                                             o
                                             E
                                             I-
                                             o

                                             CO
                                                                                          X

                                                                                          O
                                                                   20 Production Ot 1967 197}
                                                                                                            _L
                                                       70     40     60     BO     100


                                                          TtST AMBIENT tEMPERATURE I'FI
                                                        20     40      60      80     100      170


                                                            TEST AMBIENT TEMPERATURE I F)
        FIGURE  7-3  EFFECT OF  AMBIENT  TEMPERATURE ON EXHAUST EMISSIONS DURING  THE CVS COLD-START TEST
         SOURCE: REFERENCE 7-9

-------
                       Table 7-4.  EFFECT OF  BAROMETRIC PRESSURE
                           AND HUMIDITY ON  EXHAUST EMISSIONS
                                Section:    7(LG)
                                Revision:   0
                                Date:       June 1975
                                Page  25  of  29
      Source
                            HC
Percent Change

 CO        NO
                                                             —2
                        Range  of  Study
One  inch Hg increase
  barometric pressure

  GM environmental
    chamber data           -10

  Ford data based on
    multiple regression
    analysis of three
    vehicles               -13.6

 50 grains  increase  in
  absolute humidity

  GM environmental
    chamber data           +10
 -30
 -21
+5
+2.2
+12.5
+7.7
26-30" Hg
28.7-29.51" Hg
  +25
             -1.5
            30-100 grain/
              Ib dry air
   Data based on FTP-H tests
    Source.  Reference 7-9 p. 159
                                            153

-------
 Section:    7(LG)
 Revision:  0
 Date:      June 1975
 Page 26 of 29
technician would  top up a  fuel conditioning cart containing  leaded
fuel with unleaded fuel.   Using such  a fuel mixture  in a catalyst
system would destroy the effectiveness of the catalyst.
     Using fuel of the wrong octane may cause "ping" or "knock"
in  some vehicles  which may result  in  certification test failure.
Hydrocarbon composition in part determines fuel  octane and the running
characteristics of the vehicle.  In addition, the response of the FID
can be affected by different ratios of paraffins, aromatics  and olefins.
Therefore, fuel analysis and proper handling are important in controlling
test variability.
7.4       FUEL ECONOMY MEASUREMENTS

          Fuel economy data have been  determined for the 1975 Federal
Test Procedure (FTP), referred to as the Urban driving cycle and the
recently developed Federal highway cycle.   The determination of fuel
consumption from exhaust emission measurements is based on the principle
of carbon mass balance, a common technique  used to evaluate chemical
processes.  According to this principle,  the mass of material leaving
the process must equal the mass of material entering the process *  In
the case of the 1975 FTP, the carbon of  the test fuel entering the
engine must equal the sum of the amounts of carbon contained in the
exhaust emissions.
          The equation used to calculate the fuel economy of a vehicle,
in miles per gallon (mpg), from data gathered during a Federal Emission
Test is of the following form:
mpg - -
g carbon in exhaust/mile
K (g/gal)
(A-l)
(A-2)
mpcr =
 ^   K^g HC/mile) + K2"(g co/mile) + K3  <9 CO2/mile)

where:

K- = carbon weight fraction of gasoline or unburned HC (mol. wt. C)/
      (mol. wt. HC, oc), = 0.866
                 J. • O J

K2 = carbon weight fraction of CO,  (mol. wt. C)/ mol. wt. CO), = 0.429

K  = carbon weight fraction of CO ,  (mol. wt.C)/(mol. wt. CO.), = 0.273
 J                               £                          2

g/gal = mean density of Indolene 30 test fuel = 2798

substituting:

    	0.866 (2798)    	
    ~ 0.866 (g/mile HC) + 0.429 (g/mile CO) + 0.273 (g/mile CO )

    =	2423	__^
      0.866 (g/mile HC) + 0.429 (g/mile CO) + 0.273 (g/mile CO )

-------
                                                            Section:   7(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 27 of 29


          This method of determining fuel economy depends primarily on
the measurement of CO2.   While all of the carbon in the exhaust should
be measured,  the contributions of HC and CO are only small percentages
of the total carbon content.   For example, a '75 FTP on a standard-size
vehicle may yield 800 g/mi CO2, 15 g/mi CO, and 1.5g/mi HC.  The masses
of carbon measured as HC and CO account for about 3 percent of the total
mass of carbon.  Errors as high as ±10 percent in the HC and CO values
would cause errors of less than ±0.5 percent in the fuel economy calculation.
          It is clear from this that the variability of the fuel economy
measurement is dependent on the variability of the CO2 emission.  It
has been reported (Reference 7-3) that the overall composite variability
of CO2  on the "75 FTP was 1.0 percent from test to test.  This varia-
bility can be translated directly as the variability of the fuel economy
from test to test.
          Fuel economy is an indicator of the amount of work expended by
the vehicle during the '75 FTP.  As such, the measurement is influenced
by vehicle or equipment variations which cause changes in the total
amount of work required.
          For the '75 FTP driving cycle, the dynamometer inertia setting
has a dominant influence upon the amount of work performed because of
the transient, start-and-stop characteristics of urban driving.  Since
the inertia setting is a mechanical and invariable quantity once it has
been set, its contribution to fuel economy variability is low.
          The highway driving cycle is a quasi-steady speed profile of
10.2 miles at an average speed of 48.2 mph.  Since the dynamometer power
absorption unit is calibrated and set at a constant 50 mph, the varia-
bility of fuel economy measurements on the highway driving cycle should
also be quite low.
          Most of the variables previously discussed apply to the mea-
surement of fuel economy but to a lesser degree.  Variables associated
with the CO2 measurement are of prime concern, in particular the CO2
instrument and calibration gases.  The carbon balance is a mass measure-
ment, therefore the CO2 standard should be traceable to a gravimetric
standard.
7.5       QUALITY ASSURANCE AND TEST VARIABILITY

          Statistical methods that can be used to control test
variability have been described extensively  in Section  6.  Quality
Assurance has the responsibility for controlling the  test-to-test
variability and improving data reliability.  Many studies have been
done on methods of reducing test variability.  However,  further  reduc-
tion of test variability is impractical  in many cases;  consequently
Quality Assurance should advocate the use of procedures such as  data
validation, calibrations, and maintenance, and assure that  these
procedures are being complied with.  Table 7-5 is a summary  of the
test variables and the methods used for  their control.
                                155

-------
                                    Table 7-5.   SUMMARY  OF TEST VARIABLES AND METHODS  USED FOR  THEIR CONTROL
r •-.•••
TEST VARIABLE
Vehicle
V
o
Counters
Humidity
Barometer
dP, CVS Pump
CVS-Temp
Ambient Condi-
tions
Dynamometer
Analyzer
Calibration
Gas
Zero Gas
Operator
Driver
Computer
Drivers Aid
CVS
Sample Trans-
port System
Fuel
NO Converter
X
Span or Work-
ing Gases
Dilution Air/
Pollutants
METHOD USED TO CONTROL TEST
1
Cali-
bra-
tion

X

X
X
X
X


X
X





X
X





X


Ref .
Stan-
dards



X
X




X
X

X
X













Data
Vali-
dation
X


x
Daily
Checks

X
X

X
x
X

X
X
X





X
j
X
X
X
X
X
X
X


X
x

X









y



X
Monthly
Checks



X
X

X


X


X


X

X

X
X


X


Mainte-
nance
X

X

X

X


X
X






X

X

X




Train-
ing






Audit

X






]




X
X
X
X












X
X
X
X
X
X


X

X

X

X
Precon-
dition-
ing
X




X



X
















VARIABLES
Corre-
lation
X
X

X
X
X



X
X

X
X
X
X

X

X

X
Control
Charts

X
X


X



Receiv-
ing Insp.
X








i







X



X






X
X
Instru-
ment
Range

Environ-
mental
Control
•y
i








X
1
X
X




X


X


X


X
jj













X




X
















X
1
CO
o
Ml
to
























                                                                                                                                                  o ja cn
                                                                                                                                                  B>  (5 CD
                                                                                                                                                  ft < O
                                                                                                                                                  (B  H- ft
                                                                                                                                                     0) H-
                                                                                                                                                     H- 0
                                                                                                                                                     O 3
                                                                                                                                                K> Ci O -J
                                                                                                                                                  c    •**»

                                                                                                                                                  (B    O
                                                                                                                                                  tO
                                                                                                                                                  -J
                                                                                                                                                  01
H
V/l
ON

-------
                                                            Section:    7(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  29  of  29
          Many of the precautions and checks mentioned in this section
are included in the Test procedures (Volume II).   Each test facility,
depending upon its experience and judgment should carefully review this
section to determine if some or all of the additional precautions and
checks should be introduced as routine or periodic checks into their
operational test procedures.
                                 157

-------

-------
                                                             Section:    8(LG)
                                                             Revision:   0
                                                             Date:       June 1975
                                                             Page 1  of  14
                             Section 8

             QUALITY ASSURANCE SYSTEM (ON-SITE) SURVEY
          The greatest drawback to effective quality assurance is the
failure to provide well developed quality assurance plans and proce-
dures.  Actual proof of a system's effectiveness lies in determining how
the plans and procedures are converted to the required physical action.
This can be accomplished by means of a system survey.  The evaluation of
a mobile source emission testing facility quality assurance system by
means of a survey is discussed in this section.
8.1       GENERAL REQUIREMENTS

          A Quality Assurance system survey must be able to pin-point
quality system failure problems and provide a positive system for correc-
tive action and follow-up procedures.  With effective follow-up proce-
dures, corrective action becomes the "closed loop" feature in the sys-
tems survey cycle.  To ensure that test facilities have the capabilities
of meeting quality assurance requirements they must adequately demon-
strate their acceptability during a survey and review of their manage-
ment organization, facilities, personnel, procedures and data systems.
          Surveys are usually performed by a team from quality assurance
and engineering.  If the results of the survey are related to a very
important pending or actual contract purchasing may need to be involved.
(Teams composed of personnel experienced in only certain areas, but who,
as a group, meet all the necessary qualifications, may be used.)  Surveys
can be performed by a single individual provided he has a thorough
knowledge of, and sufficient experience in, investigating and assessment
of all areas and facets of quality assurance systems, and mobile source
emission testing.
          The survey is specifically designed for a test facility con-
ducting emission tests on mobile sources.  Its use, however, is not
limited to evaluation of those laboratories conducting emission tests
for the EPA.  It may be used by any test facility for self-evaluation or
by any organization such as an emission device or vehicle manufacturer
purchasing testing services from independent testing laboratories.
          Prior to traveling to a test facility the survey team should
research the facility's quality history data, and seek pertinent infor-
mation from purchasing, test and quality engineering to determine the
facility's current status.  The survey team should hold a preliminary
meeting to discuss the survey plan.
                                 159

-------
Section:   8(LG)
Revision:  0
Date:      June 1975
Page 2 of 14
             The survey checklist should not be regarded as a panacea for
   quality system evaluation.   It is a device used to assure a systematic
   look at the important areas.  The investigator must ask for and see
   objective evidence of each aspect of items on the checklist.  The
   results of the checks and other observations may lead the investigators
   to survey an area not specifically covered in the checklist.  Comments
   concerning these other areas should be recorded in the "Remarks" section
   of the survey report to assist in the final evaluation.  The Department
   of Defense has provided a detailed handbook on the evaluation of a
   contractor's quality program, however,  many requirements are more restric-
   tive than are warranted in mobile source emission testing (Reference 8-1).
             It is sometimes said that some facilities are too small to
   have a quality assurance system.   However,  smallness is no excuse for
   lack of control.  Obviously many small facilities do not need a full
   time quality assurance representative,  or require the imposition of
   elaborate controls.  Those  conducting the survey may elect to de-emphasize
   certain areas such as procurement controls and incoming material inspection
   requirements, if previous experience with the facility, or the particular
   activities at the facility  warrant it.   In contrast, if the survey is in
   connection with the award of an important service contract,  then strict
   adherence with respect to all portions of the survey would be necessary.
   8.2       ADMINISTRATION GUIDELINE
             QUALITY ASSURANCE SYSTEM SURVEY REPORT

             The Quality Assurance System Survey Report consists of two
   sections,  (i)  a cover sheet containing general information and the end
   results of the survey, (ii)  a detailed survey checklist covering the
   various elements of a Quality Assurance system for mobile source emis-
   sion testing facilities.
             Each Survey Report should be assigned a separate identifica-
   tion number for administrative traceability.  This information is entered
   in the top right hand corner of the cover sheet, together with the date
   of survey and an indication as to whether it is the first survey or a
   re-survey.
             The cover sheet briefly describes the location of the facility
   being surveyed,  identifies who is responsible for the Quality Assurance
   functions, indicates the organizational structure of the facility, the
   proportion of personnel in testing, engineering and quality assurance
   and identifies the personnel contacted during the survey.  The investi-
   gator, should also indicate who requested the survey and the contract/P.O.
   number if applicable.
             The survey should not consist of merely asking questions.  The
   investigator should request visual proof of how the system works.  In
                                160

-------
                                                          Section:    8(LG)
                                                          Revision:   0
                                                          Date:      June 1975
                                                          Page 3 of 14
evaluating the various audit elements, three alternative decisions are
available to the investigator, (i) Acceptable  (A),  (ii) Conditional
Acceptance (C), (iii)  Unacceptable (U).  Further amplification of these
decisions can be made in the "Remarks" space on the last page of the
checklist.  The following guidelines, listed in the same sequence as the
system elements on the checklist, will assist  the investigator in
evaluating the Quality Assurance system.

A.        Organization

          1.   "Organizational authority of quality assurance."  Does
               the established system identify the organizational ele-
               ment responsible for quality assurance?  Do the personnel
               performing the quality functions have sufficient autho-
               rity, responsibility and freedom of action to identify
               and evaluate quality problems and initiate, recommend, or
               provide solutions?  Verify that there is one individual
               who has overall responsibility  for quality assurance in
               the organization.

          2.   "Documentation of quality system  requirements."  Are
               documented procedures  available and  used for all testing
               and laboratory operations which affect  quality?  Ask to
               see copies.  Are procedures  reviewed on a systematic
               basis to assure accuracy, completeness  and operator/
               analyst compliance?   Do  supervision  and quality assurance
               personnel make proper use of procedures?  Verify that
               procedures are available for all  routine operations
                (receiving,  assembly,  test,  sampling, calibration,  analy-
               sis, etc.).  Review for  current status, control, and
               availability on a  "need  to know"  basis.

           3.    "Issue of activity and audit reports to management,
                listing deficiencies  and corrective  action  taken."  Are
                reports sent to management highlighting quality problems
                and  corrective action taken  to alleviate  those problems?
               Verify that  activity, and  independent performance  audit
                reports initiated by  quality assurance are  sent to top
                level  management.  Ask to  see  a copy of the latest report.

 B.         Procurement Control

           1.    "Imposition  of quality requirements on procurement orders."
                Review ordering documents to assure that the laboratory
                includes  quality assurance and acceptance provisions for
                all  procured items such as testing services, equipment,
                               161

-------
Section:   8(LG)
Revision:  0
Date:      June 1975
Page 4 of 14
                 calibration and zero gases, gasoline, etc., directly
                 affecting the quality of laboratory testing or results of
                 testing.  Do the procurement orders require test reports
                 or certifications?  Does the person responsible for
                 quality assurance review the procurement orders?  Ask to
                 see the procurement order file.  Emphasis should be
                 placed on the selection of suppliers, pre-planning the
                 requirements from the supplier, and the maximum utiliza-
                 tion of supplier data and quality information with a
                 corresponding minimization of incoming inspection
                 requirements.

  C.        Incoming Material Inspection

            1.   "Availability of acceptance standards and procedures in
                 receiving inspection area."  Go to the receiving inspec-
                 tion area and ask to see some acceptance procedures.
                 Determine if they are current, useful and appropriate.

            2.   "Maintenance of inspection records on all items received."
                 Review receiving records to assure that inspection
                 acceptance/rejection data are being maintained for all
                 procured items directly affecting the quality of emission
                 testing or results of testing.  Does quality assurance
                 inspect supplier's material to the extent necessary upon
                 receipt?

            3.   "Segregation and identification of non-conforming supplies."
                 Verify that non-conforming fuels, chemicals, gases,
                 equipment and components are positively identified and
                 segregated in a manner which prevents contamination of
                 accepted lots.  How is it identified?  Where is it stored
                 while awaiting disposition?  Go and look at it.

            4.   "Indication of inspection status on all supplies."
                 Verify that the inspected items are stamped, tagged or
                 otherwise identified as to their acceptance/rejection.
                 Identifying stocks of fuel, chemicals, gases, etc., and
                 keeping uninspected, untested/rejected material separate
                 from that already inspected, tested/accepted must be done
                 very carefully.  The inadvertent issue of wrong or defec-
                 tive material can be disastrous.
                               162

-------
                                                         Section:   8(LG)
                                                         Revision:  0
                                                         Date:      June 1975
                                                         Page 5 of 14
          5.    "Verification of  certified fuels,  chemicals and gases  by
               chemical/physical analysis on established frequencies."
               Review chemical/physical  test reports  provided by sup-
               pliers of  fuels,  chemicals and gases to determine if
               suppliers  periodically  perform verification tests to
               validate test reports and certifications.   Review results
               of verification tests to  assure that constituents are
               correctly  stated  on test  reports and certifications and
               that they  conform with  applicable EPA  specifications.

          6.    "Verification of  performance testing,  functional testing
               and  calibration of procured equipment."  Verify that
               records are maintained  as objective evidence of perfor-
               mance and  functional testing of procured equipment and
               that the equipment has  been calibrated correctly.

          7.    "Identity  and storage of  limited life  items."  Verify
               that all limited  life items have the date of manufacture
               or receipt of the items clearly marked on their con-
               tainers.   They  should be  stored in such a manner that
               they can be used  in order of receipt and thus spend
               minimum  time in storage,  and are not to be used beyond
               their expiration  date.

          8.    "Maintenance of a system for obtaining corrective action
               from suppliers."   Verify that there is a system for
               obtaining  supplier corrective action.   Ask to see a
               recently completed request for corrective action from a
               supplier.

D.        Calibration of  Inspection and Test Equipment

          1.    "Written description of calibration system covering
               measuring  and test equipment."  Request a copy of the
               laboratory's written description of its calibration
               system and audit program for maintaining correctly
               calibrated equipment.   Emphasis should be placed upon
               maximum utilization of equipment manufacturer calibration
               methods or standard calibration methods prescribed by
               A.S.T.M.,  S.A.E.  and Federal Register procedures, rather
               than an invented method of the user.

          2.    "Provision for the calibration of measuring  and  test
               equipment at periodic  intervals."  Determine that reali-
               stic calibration intervals are assigned for  measuring and
               test equipment, and that they are  established on the
               basis of stability, purpose  and degree of  usage.
                              163

-------
Section:   8(LG)
Revision:  0
Date:      June 1975
Page 6 of 14
            3.   "Maintenance of calibration records on all measuring and
                 test equipment".  Verify that adequate calibration records
                 are maintained to identify and categorize each item of
                 measuring and test  equipment.  Adequate records would
                 include history of  the item, its accuracy, present loca-
                 tion, calibration interval and when due, calibration
                 procedures  and controls necessary, actual values of
                 latest calibration  and inventory of maintenance and
                 repair made.

            4.   "Validity of calibration decals/labels."  Verify that
                 calibration decals/labels are affixed to each item of
                 measuring and test  equipment, indicating the date of last
                 calibration, by whom,  and the date when next calibration
                 is due.  Ask to see a  master list of all equipment on the
                 calibration schedule,  select some items at random from
                 both  the receiving  inspection and testing areas, and
                 visually check the  selected items for current calibration
                 decal/label.

            5.   "Availability of  calibration traceability to NBS/EPA."
                 Select certifications  of several reference standards and
                 determine if they are  traceable to the Standard Reference
                 Materials prepared  by  the NBS, or the EPA Primary Govern-
                 mental Standards.  Do  calibration sources other than the
                 National Bureau of  Standards or a government laboratory
                 have  their  standards compared with a National standard at
                 planned intervals?   Are secondary standards or working
                 gases referenced  or analyzed against these primary
                 standards?

            6.   "Imposition of requirement  on suppliers to have a system
                 which assures accuracy of their measuring and test equip-
                 ment."  Verify that the laboratory has taken action to
                 assure the  accuracy of test and measuring equipment used
                 by its suppliers.  Are the  limits of impurities and
                 analytical  tolerances  specified by their purchase orders?
                 Are analytical methods referenced or defined?  Are cali-
                 bration methods defined such as NBS, ASTM, etc?

-------
                                                         Section:   8(LG)
                                                         Revision:  0
                                                         Date:      June 1975
                                                         Page 7 of 14
E.        Vehicle Testing
          1.    "Provision of applicable inspection and test documents."
               Request copies of procedures covering the vehicle testing
               performed by the laboratory.  Verify that the prescribed
               procedures are not in conflict with Federal Register
               requirements.  Verify that the tests are conducted in
               accordance with the written test procedures by observing
               the technicians performing the test.  Request and observe
               certain calibrations of the test equipment.  Give techni-
               ciane oral quiz using the written procedure for the
               source of items to determine his familiarity with docu-
               mented procedures.

          2.    "Availability of documented test procedures, adequate
               test equipment and appropriate work environment."  Verify
               the use of documented test procedures, the specification
               of adequate test equipment and a suitable work environ-
               ment.  Request a copy of a recently issued test procedure.

          3.    "Provision of acceptable/unacceptable criteria for each
               test measurement."  Review the inspection/test procedures
               and data recording forms for inclusion of acceptance/
               rejection criteria.

          4.    "Accomplishment of testing in accordance with test speci-
               fications and procedures."  Witness a test to determine
               if laboratory is accomplishing and reporting the testing
               in accordance with the test procedures.

          5.    "Application of corrective measures when non-compliance
               occurs."  Verify the use of a prompt, effective correc-
               tive action system.  Is there an adequate form in use for
               requesting corrective action?  Who initiates a request
               for corrective action?  Who determines the adequacy of
               corrective action?  Do corrective action statements
               include the cause of rejection and the action taken to
               prevent its recurrence?  What follow-up methods are
               employed?

          6.    "Indication of current calibration status on test equip-
               ment."  Check items of test equipment to assure that they
               have current calibration decals, stickers or tags affixed,
               and are in good working condition.  Visual check equip-
               ment for cleanliness, apparent damage and/or malfunction.
                              165

-------
Section:   8(LG)
Revision:  0
Date:      June 1975
Page 8 of 14
             7.    "Maintenance  of  controlled conditions  as  required for
                  testing sequences."   Witness a test and verify that
                  conditions  are controlled and maintained  as specified in
                  the  test specifications/procedure.

             8.    "Issue of reports to engineering on test  and inspection
                  problems or deficiencies."  Request and review a copy of
                  a recent report  issued to inform engineering on problems
                  or deficiencies  in inspection or testing.   Is there an
                  adequate form used for this report?

            -9.    "Documentation,  reinspection and retest of instruments
                  and  equipment reworked,  repaired or modified after
                  testing."  Review inspection/ test  records to verify that
                  instruments and  equipment reworked, repaired or modified
                  due  to a malfunction during a vehicle  test are rein-
                  spected and tested prior to being placed  back in service
                  again.   Review inspection/test records to verify that
                  repairs or  modifications made to a  vehicle are adequately
                  documented  and reported.

            10.    "Maintenance  of  accurate and complete  test results and
                  data,  with  traceability to the tested  vehicles and the
                  test and measuring equipment used." Review records of
                  data and test results.  The records should show evidence
                  of configuration control and clear  traceability back to
                  the  tested  vehicle and test cell equipment.
   F.         General
             1.    "Provision of qualified testing personnel  and a training
                  and certification program for personnel  involved in
                  testing."   Verify the existence of a written,  established
                  training program for personnel involved  in testing,
                  analysis,  and quality assurance.  Check  for  records  of
                  individual training history and evidence of periodic
                  personnel  testing and applicable test results.

             2.    "Maintenance of housekeeping and facilities commensurate
                  with testing requirements."  Check for evidence of poor
                  housekeeping practices and determine if  facilities are
                  commensurate with testing requirements.
                                166

-------
                                                Section:   8(LG)
                                                Revision:   0
                                                Date:      June 1975
                                                Page 9 of  14
3.    "Maintenance of a quality data reporting and analysis
     system,  with built-in validation checks for accuracy,
     precision and completeness."  Verify the existence of
     (a.)  written procedures, forms, etc., used in performing
     necessary computations, data reductions and validations,
     (b.)  an audit program to verify data accuracy  (c.) the
     application of statistical quality control chart techni-
     ques if appropriate, and (d.) the reporting of the
     quality of the data and test results to top management on
     a periodic basis.

4.    "Issue of inspection stamps, calibration decals, etc.,
     controlled by quality assurance."  Verify that quality
     assurance maintains records on the issue and control of
     inspection stamps, calibration decals, etc., including
     date of issue, reference number and recipient information.

5.    "Maintenance of a configuration control system to account
     for changes in equipment/ documents."  Verify that confi-
     guration control procedures exist and include the follow-
     ing provisions.

     (a)  Removal of all obsolete equipment/documents from
          affected departments

     (b)  Distribution of all new or revised equipment/documents
          to affected departments

     (c)  Recording of point at which changes become effective

     (d)  Maintenance of a master index to reflect  all docu-
          ment issues and revisions

     (e)  Review of documents/specifications prior  to release

     A testing laboratory would be principally concerned  with
     configuration accounting to assure that all  similar
     equipments have  the same configuration and that all
     document changes, including computer programs  have been
     recorded.
                    167

-------
 Section:   8(LG)
 Revision:   0
 Date:      June 1975
 Page 10 of 14
           6.    "Maintenance of a quality cost system."  Determine if the
                testing laboratory has developed specific cost data to
                identify prevention,  appraisal,  internal and external
                failure costs,  and the effective use of this data in
                quality management.

           7.    "Provision of reliability and preventive maintenance
                requirements."   Verify the existence of written,  estab-
                lished procedures pertaining to  reliability  and preven-
                tive maintenance.  The consideration of reliability and
                preventive maintenance in air pollution measurement is
                becoming increasingly important  due  to  the complexity and
                sophistication  of sampling,  measurement,  and automatic
                recording systems.

           Upon completion of the survey  the investigator/survey  team
evaluates  the  checklist and other observations  noted during the  survey,
and discusses  the  findings with the  interested  laboratory management
personnel  to clarify any differences as  to the  facts.   If the  survey is
of an informal nature,  the approval/disapproval recommendations  may be
dispensed  with.  For a  formal  survey,  once the  facts are established the
investigator indicates  approval,  conditional approval  or disapproval in
the appropriate box on  the cover  sheet.   If conditional approval is
granted, time  should be allowed for  correction  of noted deficiencies in
establishing the re-survey date.' The investigator  notes any specific
system weaknesses that  require  corrective action in the space  assigned
for "Remarks", also  any other comments pertaining to the survey,  signs
name in the bottom  left hand box  under "Survey  performed by".  The
completed  survey is  routed to the investigator's departmental  supervisor
for approval, prior  to  distribution.
           The  report is then sent to the surveyed laboratory with
request or suggestions  for improvement of their quality assurance/
testing program.  Usually  one member of  the survey  team is  requested to
follow-up  after the  laboratory  has communicated in  writing  that  the
suggested  changes/improvements  have  been complied with.  A  laboratory
which has  been disapproved should be allowed to request a new  survey
after a certain time period if  they  can  show that a corrective action
program has been implemented and completed.
8.3       QUALITY ASSURANCE SYSTEM SURVEY REPORT

          The report consists of a cover sheet and checklists as
illustrated on the following pages.
                                168

-------
                                                                Section:   8(LG)
                                                                Revision:  0
                                                                Date:      June 1975
                                                                Page 11 of 14
QUALITY ASSURANCE SYSTEM SURVEY REPORT            Survey No.	

                                                  Date of Survey	
                                                  Type of Survey  |  | Initial

                                                                  |  I Resurvey

Laboratory Name	

Street Address	__	
                                            Zip            Phone
City_	 State	 Code	  No. 	
Name of person responsible for quality assurance functions at above  address.

Name                         	 Title	
                                                       Parent Organization:
This firm is:     Independently
              f~"| Owned & Operated     [~~| Subsidiary

              |  | Affiliate            (~1 Division of


Number of Personnel:  Testing	 Engineering	 Quality Assurance	

     Personnel Contacted                               Title
Survey Requested by:

Name	Dep t.	pi v.	Date	

Contract/P.O. Number	

Results of Survey;    Q Approved   Q] Conditional  Approval    f~| Disapproved

Resurvey Date (For use with conditional approval  only)		.

REMARKS:
 Survey Performed By;    Approved;          Department;          Date^
                                     169

-------
    Section:    8(LG)
    Revision:   0
    Date:       June 1975
    Page 12 of 14
                 QUALITY ASSURANCE SYSTEM SURVEY CHECKLIST
                                                          Survey No.
A - Acceptable
C - Conditional Acceptance
U - Unacceptable
Survey
Element
No.
A
1
2
3
B
1
C
1
2
3
4
5
6
7
8
Requirements
ORGANIZATION
Organizational authority of quality assurance
Documentation of quality system requirements
Issue of activity and audit reports to management,
listing deficiencies and corrective action taken
PROCUREMENT CONTROL
Imposition of quality requirements on procurement
orders
INCOMING MATERIAL INSPECTION
Availability of acceptance standards and procedures
in receiving inspection area
Maintenance of inspection records on all items
received
Segregation and identification of non- conforming
supplies
Indication of inspection status on all supplies
Verification of certified fuels, chemicals and
gases by chemical/physical analysis on established
frequencies
Verification of performance testing, functional
testing and calibration of procured equipment
Identity and storage of limited life items
Maintenance of a system for obtaining corrective
action from suppliers
A












C






-





U












                                   170

-------
                                                               Section:   8(LG)
                                                               Revision:  0
                                                               Date:      June 1975
                                                               Page 13 of 14
                QUALITY ASSURANCE SYSTEM SURVEY CHECKLIST
                                                          Survey  No.
A - Acceptable
C - Conditional Acceptance
U - Unacceptable
Survey
Element
No.
D
1
2
3
4
5
6
E
1
2
3
4
5
6
Requirements
CALIBRATION OF INSPECTION AND TEST EQUIPMENT
Written description of calibration system covering
measuring and test equipment
Provision for the calibration of measuring and test
equipment at periodic intervals
Maintenance of calibration records on all measuring
and test equipment
Validity of calibration decals/labels
Availability of calibration traceability to NBS/EPA
Imposition of requirement on suppliers to have a
system which assures accuracy of their measuring and
test equipment
VEHICLE TESTING
Provision of applicable inspection and test documents
Availability of documented test procedures, adequate
test equipment and appropriate work environment
Provision of acceptable/unacceptable criteria for
each test measurement
Accomplishment of testing in accordance with test
specifications and procedures
Application of corrective measures when non-
compliance occurs
Indication of current calibration status on test
equipment
A












C












U












                                   171

-------
    Section;   8(LG)
    Revision:  0
    Date:      June 1975
    Page 14 of 14
                 QUALITY ASSURANCE SYSTEM SURVEY CHECKLIST
                                                          Survey No.
A - Acceptable
C - Conditional Acceptance
U - Unacceptable
Survey
Element
No.
E
7
8
9
10
F
1
2
3
4
5
6
7
Requirements
VEHICLE TESTING (Continued)
Maintenance of controlled conditions as required
for testing sequences
Issue of reports to engineering on test and inspection
problems or deficiencies
Documentation, reinspection and re test of instruments
and equipment reworked, repaired or modified after
testing
Maintenance of accurate and complete test results
and data, with traceability to the tested vehicles
and the test and measuring equipment used
GENERAL
Provision of qualified testing personnel and a train-
ing and certification program for personnel involved
in testing
Maintenance of housekeeping and facilities commensurate
with testing requirements
Maintenance of a quality data reporting and analysis
system, with built-in validation checks for accuracy,
precision and completeness
Issue of inspection stamps, calibration decals,
etc. , controlled by quality assurance
Maintenance of a configuration control system to
account for changes in equipment/documents
Maintenance of a quality cost system
Provision of reliability and preventive maintenance
requirements
A











C











U











  REMARKS:  (Attach additional sheets if required.   Identify
           with S/N of this report)
                                    172

-------

-------
                                                           Section:   9{LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 1 of 3
                            Section 9
                           REFERENCES
2-1       Juran,  J.M.,  Quality Control Handbook, McGraw Hill, New York
          1974.   Section 4-16.

2-2       Development of Written Test for Certification of Emission Labora-
          tory Technicians.  Environmental Protection Agency Report No.
          EPA-460/3-74-008, June 1974.

2-3       Juran,  J.M.,  Quality Control Handbook, Op. Cit., Section 5.

4-1       Quality Assurance Handbook for Air Pollution Measurement Systems,
          Environmental Protection  Agency Report  (Preliminary Draft.)
          Section 1.4.16

4-2       Ireson, W. Grant,  Reliability Handbook, McGraw Hill, New York
          1966.

4-3       Juran,  J.M.,  Quality Control Handbook, Op. Cit., Section 9-6.

4-4       Ibid,   Section 11-16 and 11-17.

4-5       Ammer,  D.S.,  Manufacturing Management and Control, Meredith
          Corporation, 1968.  Page 166.

4-6       Preventive Maintenance Inspections and Calibration Checks, Envi-
          ronmental Protection Agency, Ann Arbor, Michigan.  Internal Re-
          port -  S.D. Funk, September 1973.

5-1       Juran,  J.M.,  Quality Control Handbook, Op. Cit., Section 16.

6-1       Quality Assurance Handbook for Air Pollution Measurement Systems,
          Op.  Cit., Section Appendix H.

6-2       Grant,  E.L.,  Statistical Quality Control, McGraw Hill, 3rd
          edition, 1964, Page 112 et. seq.

6-3       Grubbs, F.E.,  The Difference Control Chart With An Example Of
          Its  Use, Industrial Quality Control, July 1966.  p.p. 22-25
                               173

-------
 Section;   9(LG)
 Revision:  0
 Date:      June 1975
 Page  2 of 3
6-4       Duncan,  A.J.,   Quality Control and Industrial Statistics,
          Richard  D.  Irwin Inc., Homewood,  Illinois.  3rd edition, 1965,
          Chapters XXIX  - XXXI.

6-5       Gionet,  P.A.,   Analysis of Variance, Society of Automotive Engi-
          neers, Inc., New York.  December, 1963.  Publication No. SP-250,
          p.p.  35-48.

6-6       Rider, P.R.,   Criteria^ for Rejection of Observations, Washing-
          ton University Studies, New Series, Science and Technology,
          No. 8, October 1933.

6-7       Dixon, W.J.,   Processing Data for Outliers, Biometrics, Vol. 9,
          No. 1.   p.p.  74-89, March 1953.

6-8       Development of a Curve Generation Procedure for Gas Analyzer
          Calibrations,  Environmental Protection Agency, Ann Arbor,
          Michigan.  Internal Report - C.D. Paulsell and D. Johnson,
          January, 1974.

6-9       King, J.R.,  Probability Charts for Decision Making, The Indus-
          trial Press.   New York 1971.

6-10      Quality Control Practices in Processing Air Pollution Samples.
          Environmental Protection Agency Report APTD-1132, March, 1973.

7-1       Automobile Exhaust Emission Surveillance, A Summary, Environ-
          mental Protection Agency Report APTD-1544, May, 1973.

7-2       Consultant Report to the Committee on Motor Vehicle Emissions on
          Emissions and Fuel Economy Test Methods and Procedures, Washington
          D.C., September 1974, Section 4.4.

7-3       Paulsell, C.D. and Kruse, R.E.,   Test Variability of Emission and
          Fuel  Economy Measurements using the 1975 Federal Test Procedure,
          Society of Automotive Engineers Inc., New York.  Publication
          No. 741035

7-4       Federal  Register, Volume 39, No.  101, May 23, 1974.  p. 18079,
          85.075-23 (c)(11)

7-5       Application for Suspension Of 1977 Motor Vehicle Exhaust Emis-
          sion Standards.  Ford Motor Company, Volume 1, Section III-E,
          January, 1975.
                                174

-------
                                                           Section:   9(LG)
                                                           Revision:  0
                                                           Date:      June 1975
                                                           Page 3 of 3
7-6       Recommended Practices for Constant Volume Sampling of Vehicle
          Exhaust, Coordinating Research Council, Inc., New York, N.Y.
          January, 1972.

7-7       General Motor's Request for Suspension of 1977 Federal Emission
          Standards. Appendix 20, Volume III of III, January 10, 1975.

7-8       Moore, M.L.,  Assurance and Control of Vehicle Emission Testing,
          Society of Automotive Engineers Inc., New York.  Publication
          No.  730534, May,  1973.

7-9       Report by the Committee on Motor Vehicle Emissions, Commission
          on Sociotechnical Systems, National Research Council, National
          Academy of Sciences, Washington D.C., November, 1974.

7-10      Certification Test Results for 1975 Model Year, Federal Register,
          Volume 40, No. 48, March 11, 1975  (11496-11534)

8-1       Evaluation of a Contractor's Quality Program, Handbook H 50,
          April 23, 1965.   Department of Defense, Washington D.C.
                               175

-------

-------
                  Appendix A-l




SELECTED STATISTICAL TECHNIQUES AND NOMENCLATURE
                       177

-------
                                                            Append ix:  A-1(LG)
                                                            Revision:  0
                                                            Date:      June  1975
                                                            Page 1 of 8
                           Appendix A-l

         SELECTED STATISTICAL TECHNIQUES AND NOMENCLATURE
          Initially an understanding of certain technical terms is
necessary in discussing statistical methodologies recommended for use.
The following concepts provide  the tools and definitions necessary to
complete statistical analyses.   (References 6-2, 6-4}

          o    Statistical Quality Control - A regulatory process
               through which actual quality performance is measured
               using quantitative, statistical methodologies.

          o    Central Tendency Measures - These measures are used to
               describe the value about which data tend to cluster.
               Examples of central tendency measures are the arithmetic
               mean, geometric  mean, mode and median.

          o    Arithmetic Mean  - This is the most frequently used measure
               of central tendency and is defined as the sum of the
               observed values  divided by the number of observations,
               i.e.,

                                               n
                      + x  + x, + . . . + x.   .  ,   J
                         2    3	2. - 3=1
                              n
               where     x. = observed performance values

                         n = number of observations
              Median - The median of a set of numbers arranged in order
              of magnitude (i.e., in an array) is the middle  value if
              there  is an odd number of values in the set,  or the
              arithmetic  mean of the two middle values if there is an
              even number of values in the set.

              Mode -  The  mode of a set of numbers is that value which
              occurs  with the greatest frequency.
                                178

-------
Appendix:  A-l(LG)
Revision:  0
Date:      June 1975
Page 2 of 8
               Random Variable - A quantity that has a definite value
               for each possible result of an experiment.   These values
               may be thought of as outcomes, e.g., instrumentation
               readings.   Although the random variable values are unknown
               prior to the  outcome of a reading, the probability that
               the random variable will take on specific values may be
               known in advance, as prescribed by a frequency distribution.

               Frequency Distribution - In summarizing data, it is
               useful to distribute data into categories and to deter-
               mine the number of individuals, e.g., measurement values,
               belonging to  each category.   A tabular arrangement of
               data by category together with the corresponding fre-
               quency with which each value occurs is called a frequency
               distribution.

               Normal Distribution - A bell-shaped distribution speci-
               fied by the function:
                              -(x-ji )2/2
-------
                                             Appendix:  A-l(LG)
                                             Revision:  0
                                             Date:      June  1975
                                             Page 3 of 8
         n
      1  >  (x.-x)
     n^TfT'  D
              _ 2
2 	^ y (x^-x)       J = l,2,3,...n
            2
Where      s  =  variance of sample
           x.  = observed values
           3
           x = sample mean

           n = number of observations in
               the  sample
 Standard Deviation  - A measure of the variation of
 individual  observations about the mean.  The unit of
 measurement for the standard deviation is the same as
 that  for the individual observations.  The standard
 deviation (equal to the square root of the variance) is
 referred to as a and s  for the population and a sample
 respectively.

 Bernoulli Trials -  Describes the conditions which must be
 met before  using the binomial distribution, which can
 establish QA acceptance criteria.  The conditions are:

 1.    Results of "trial" (e.g., selection of sample) must
      be  totally separate of any other outcome (i.e., the
      outcomes cannot be related in any way) .

 2.    Only two outcomes of the trial exist (e.g., either
      pass or fail,  heads or tails, etc.).

 3.    The probability of a given outcome of a trial must
      remain constant throughout the sequence of the
      trials.

 4.    The trials are statistically independent (i.e., the
      outcome of a given trial does not depend on that of
      another trial).

 Binomial  Distribution - A family of probability distri-
 butions  describing  the probabilities of possible experi-
mental outcomes for all possible experimental outcomes
 for all possible combinations of n trials and p, the
probability of  an outcome during a trial.  The distri-
bution is given by:
                         r ,   n-r

 ><**•»-            f(fi
                 180

-------
Appendix:   A-^l(LG)
Revision:   0
Date:      June 1975
Page 4 of  8
               Where     r = actual number of specific
                             outcomes during a sequence
                             of trials

                         n = number of trials in the
                             sequence

                         p = probability of a given out
                             come's occurrence during
                             sequence of events

               Such a distribution is important in that it forms the
               basis for much of the QA acceptance sampling theory.  It
               is possible to compute mathematically the probability
               that a lot of a given percentage defective (e.g., the
               number of automobiles above certain prescribed exhaust
               emission levels) will be accepted under a given sampling
               plan.

          o    Random Error - Inaccuracies due to small, indeterminate
               variations in a system's performance.  The distribution
               of random error is usually assumed to be normal, i.e.,
               Gaussian, with a mean equal to zero.

          o    Range - The difference between the maximum and minimum
               values for a sample of observed values.  When the number
               of observed values is small, the range is a relatively
               sensitive measure of general variability.  As the number
               of observations increases, the efficiency of the range
               (as an estimator of the standard deviation) decreases
               rapidly.

          o    Coefficient of Variation - The ratio of the standard
               deviation to the mean, also referred to as the relative
               standard deviation.   It is usually expressed as a per-
               centage and is given by:
                         CV =  -  (100)%
                              x

               Where     s = standard deviation of a sample

                         x = mean of a sample
               Confidence Levels - The probability that an  assertion  is
               correct about a characteristic of a measurement  system.

               Confidence Interval -*A statistic  (e.g., the mean x)
               is computed from the data  for a  sample.  The statistic is
               then used as a point estimate of the  population  parameter
                (e.g., the mean fi).  It is recognized that the statistic
               computed from a second sample would not be identically
               equal to that for the first sample.   Because of  this,

                               181

-------
                                            Appendix:  A-1(LG)
                                            Revision:  0
                                            Date:      June 1975
                                            Page 5 of 8
points A and B are determined such that it can be said
with a specified probability that the interval described
by A and B contains the true value of the population
parameter.

For example the probability statement for the 95 percent
confidence interval estimate of the population mean is
given by:
= 0.95
    /-    /Vl»\    - An-i
    (*  -(jFf""'\jF
Where     x = sample mean

          s = sample standard deviation

          t  . = student "t" value for n-1
                 degrees of freedom

             n = number of observations in
                 the sample

The probabilities usually associated with confidence
interval estimates are 90, 95, and 99 percent.  For a
given  sample size, the width of the confidence interval
increases as the probability increases.

Confidence Limits - The end points of the confidence
interval A and B as discussed above, whereas:
          B
Sample - A  set of objects or things from a larger set
called the  "population."  The objects or things may be
physical such as specimens for testing or they may be
data values representing physical samples, or data values
from a larger set of data values.  Unless otherwise
specified,  all samples are assumed to be random samples.

Random Samples - Samples obtained in such a manner that
all items of the lot or population have an equal chance
of being selected in the sample.

                182

-------
Appendix:   A-1(1X3)
Revision:   0
Date:      June 1975
Page 6 of  8
               Stratified Sample -  (Stratified Random  Sample)  - A sample
               of the various portions which have been obtained from
               identified subparts or subcategories  (strata) of the
               total lot or population.  Within each category  of  strata,
               the sampling would be taken randomly.   The objective  of
               taking stratified samples is to obtain  a more repre-
               sentative sample than that which would  otherwise be
               obtained by a completely random sampling.  The  idea of
               identifying the subcategories or strata is based on
               knowledge or suspicion, or precaution against differences
               existing between the strata for the characteristics of
               concern.

               Representative Sample - A sample taken  to represent the
               universe or population as accurately and precisely as
               possible.  A representative sample may  be either a
               completely random sample or a stratified sample depending
               upon the objective of the sampling and  the conceptual or
               actual population for a given situation.

               Acceptance Sampling - Sampling inspection in which
               decisions are made to accept or reject  the total popu-
               lation from which the sample is taken or for which the
               sample represents.  The science that deals with the
               procedures by which decisions to accept or reject  are
               based on the results of the sample inspection.

               Audit (General) - A random check to determine the  quality
               of operation of some function or activity.  Two types of
               audits are used in Quality Assurance:   (1) performance
               audits, and (2) system surveys.

               Performance Audit - Planned independent (duplicate)
               sample checks of actual output made on  random basis to
               arrive at a quantitative measure of the output  from all
               or part of the total system.

               System Survey - A systematic on-site qualitative review
               of facilities/ equipment, training, procedures, record-
               keeping, validation, and reporting aspects of a total
               (quality assurance) system to arrive at a measure  of  the
               capability and ability of the  (quality  assurance)  system.
               Even though each element of the system  survey is quali-
               tative in nature, the evaluation of each element and  the
               total may be quantified  (scored) on some subjective
               basis.
                               183

-------
                                             Appendix:  A-l(LG)
                                             Revision:  0
                                             Date:      June  1975
                                             Page 7 of 8
111" Distribution - A probability distribution developed
by W. S. Gosset (writing under the pseudonym "Student")
used in the computation of confidence interval estimates
when the population standard deviation is unknown.  In
such a case s (the sample standard deviation) is used as
an estimate of n .  When the sample size is small the
value of "t" for a given probability level differs
significantly from the "z " value for the normal dis-
tribution.  For example, in determining the 95 percent
confidence interval estimate of the mean when the sample
size was 10, the value of t is 2.262 whereas the value of
from the normal distribution is 1.96 (regardless of
sample size).

Control Chart Multiplication Factors -  Factors as
applied in the manual are multipliers used to calculate
statistical control limits for control charts.  They
provide a method of approximating the distribution of all
the values in the population when calculating statistical
limits.  This is necessary because the distribution of
sample values differs from the distribution of population
values.  The factors used in this manual are D , D , B ,
B., AI and A~.  Definitions of these factors and
formulae for computing them are in Reference 6-2, Appen-
dix III. Tables with the factors used for the 99 percent
confidence interval are in Appendix A-2.  The application
of each of the factors is:

D  - Compute the 3 sigma lower control
     limit for a range control chart.

D  - Compute the 3 sigma upper control
     limit for a range control chart.

B  - Compute the 3 sigma lower control
     limit for standard deviation or
     coefficient of variation control charts.

B. - Compute the 3 sigma upper control limit
     for standard deviation or coefficient
     of variation control charts.

A  - Compute 3 sigma upper and lower control
     limits for average control charts,
     using a .

A_ - Compute 3 sigma upper and lower control
     limits for average control charts,
     using R.

-------
Appendix:   A-l(LG)
Revision:   0
Date:      June 1975
Page 8 of 8
          Definitions and use of control charts will be discussed
          in other sections.

          o    Replicates - Repeated but independent tests or analyses
               of the same sample, under the same conditions.  Replicates
               may be performed to any degree, e.g., duplicates, triplicates,
               etc.

          o    Precision and Accuracy - The concepts of precision and
               accuracy must be understood in formulating control chart
               limits.  A system, e.g., instrument, will not necessarily
               display identical readings even when making measurements
               on a single sample.  Rather, the values will tend to
               scatter about a point of central tendency.  Precision is
               the ability of a system to reproduce its own levels of
               performance, e.g., measurements.  Precision is determined
               from replicate analyses.  It represents the variability
               of results among the replicate analyses.  Precision can
               be expressed in terms of standard deviation, variance, or
               range.

               Accuracy is the difference between a measurement and its
               true value.  It describes the magnitude of error in a
               measurement.  It is expressed either as a relative error,
               expressed in percentage, or in terms of units, e.g.,
               parts per million.  Usually, critical parameters in an
               analytical system should be evaluated in terms of ac-
               curacy or precision.

          o    Performance Levels - Defined, acceptable levels of
               performance.  These levels must be specified before
               evaluating the analytical performance of a system.  Some
               sources of information which could possibly affect the
               choice of performance levels are Federal Register speci-
               fications, EPA recommendations, method specifications,
               and good engineering practices.
                                 185

-------
     Appendix A-2







     CONTROL CHART




MULTIPLICATION FACTORS
          187

-------
                                                          Appendix:  A-l(LG)
                                                          Revision:  0
                                                          Date:      June 1975
                                                          Page 1 of 1
                         Appendix A-2
             CONTROL CHART MULTIPLICATION FACTORS*
Observation in
SUB-GROUP, n
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
FACTORS FOR CONTROL LIMITS
Al
3.67
2.39
1.88
1.60
1.41
1.28
1.17
1.09
1.03
0.97
0.93
0.88
0.85
0.82
0.79
0.76
0.74
0.72
0.70
0.68
0.66
1.65
0.63
0.62
A2
1.880
1.023
0.729
0.577
0.483
0.419
0.373
0.337
0.308
0.285
0.266
0.249
0.235
0.223
0.212
0.203
0.194
0.187
0.180
0.173
0.167
0.162
0.157
0.153
B3
0
0
0
0
0.030
0.118
0.185
0.230
0.284
0.321
0.354
0.382
0.406
0.428
0.448
0.466
0.482
0.497
0.510
0.523
0.534
0.545
0.555
0.565
B4
3.267
2.568
2.266
2.089
1.970
1.882
1.815
1.761
1.716
1.679
1.646
1.618
1.594
1.572
1.552
1.534
1.518
1.503
1.490
1.477
1.466
1.455
1.445
1.435
D3
0
0
0
0
0
0.076
0.136
0.184
0.223
0.256
0.284
0.308
0.329
0.348
0.364
0.379
0.392
0.404
0.414
0.425
0.434
0.443
0.452
0.459
D4
3.267
2.575
2.282
2.115
2.004
1.924
1.864
1.816
i. in
1.744
1.716
1.692
1.671
1.652
1.636
1.621
1.608
1.596
1.586
1.575
1.566
1.557
1.548
1.541
*References:  6-2 Appendix  III,  6-4 Appendix II  Table M.
                               189

-------
  Appendix B-l






GLOSSARY OF TERMS
      191

-------
                                                            Appendix:  B-l(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 1 of 18
                           Appendix B-l

                         GLOSSARY OF TERMS
Acceleration - The rate of change of velocity per unit time.  e.g.,
miles per hour per hour.

Advance (spark)  - To cause the occurrence of spark earlier in the
combustion cycle.

Air Cleaner (carburetor) - A device mounted on the carburetor through
which air must pass on its way into the carburetor air horn,  it filters
out dust particles, silences intake noise, and safeguards against back-
fire through the carburetor.

Air Guard - An air injection exhaust emission system used by American
Motors Corporation.

Air Injection - A system where pressurized air is transmitted to each
exhaust port of the engine.  Here the fresh charge of air mixes with hot
exhaust gases and promotes more complete burning of hydrocarbons and
carbon monoxide.

Air Injection Reactor - An air injection exhaust emission system using a
pump to inject air into a specially designed exhaust manifold.

Air Pump - An engine, belt driver, air pump incorporating a rotor and
three vanes.  The vanes rotate freely about an off-center pivot pin and
follow the circular-shaped chamber.  A basic component of all air injec-
tion type exhaust emission systems.

Aldehydes - Partially oxidized hydrocarbons in which oxygen atoms are
bonded to carbon atoms at the end of a molecular chain.  These gases
contribute to the formation of eye irritating materials formed in
photochemical smog.

Ambient Air - Air in the surrounding area which is used as the diluent
air by the CVS system.

Ambient Temperature - The measured temperature of the air which sur-
rounds an object.

Amplifier - A device employing vacuum tubes or transistors, which
multiplies an input signal and provides an output of greater magnitude.

Analytical System - Refers to all the components of an analyzing system
including the instruments, pumps, flow controllers, valves, lines,
output devices etc., required to perform  the exhaust analysis.

Arithmetic Mean  - A value that is computed by dividing the  sum of  a set
of terms by the  number  of terms; average  value.
                                193

-------
Appendix:  B-l(LG)
Revision:  0
Date:      June 1975
Page 2 of 18
Atom - The smallest subdivision of an element which retains the chemical
characteristics of that element.

Attenuator - A device for proportioning input signals,  i.e., to change
the span or range of an instrument by a known increment or multiple.

Audit  (general) - A methodical examination and review to determine the
quality of some function or activity.

Automatic Driver - An instrument that mechanically drives a car through
a test cycle by electromagnetically  (or engine vacuum) comparing the
speed  variations recorded on magnetic tape to the dynamometer roll
revolutions.

Backfire - An explosion in the induction of exhaust system.

Backfire Suppressor Valve - A device used in conjunction with the early
design "Thermactor" exhaust emission system.  Its primary function is to
lean-out the excessively rich fuel mixture which follows closing of the
throttle during deceleration.  Allows additional air into the induction
system whenever intake manifold vacuum increases.

Bag  -  An enclosure made of flexible  inert material  (usually teflon or
tedlar) used to store diluted samples of either emission or ambient air.

Barometric Pressure - Atmospheric force per unit area exerted at a given
point.

Binary Gas Mixture - A mixture of two gases only in a container (cylinder,
bag, etc.).  This is also referred to as a single component blend and is
not  a  double component blend which is a mixture of three gases used as a
standard for 2 different analyzers.  In calibration mixtures air is
usually regarded as a single gas.

Blowby - Name given to the high pressure gases that escape past the
engine piston rings into the crankcase during compression and power
strokes.  More pronounced on high mileage engines because of imperfect
seal of piston rings to cylinder wall.  Comprised mostly of unburned
fuel-air mixture.

Blower - See positive displacement pump.

Buoyancy - The tendency of a body to float or rise when submerged in a
liquid or gas.  The power of a liquid or gas to exert an upward force on
a body placed in  it.

Brake  Horsepower  - A unit measurement of work; e.g., amount of horse-
power  delivered to the transmission by the engine.

Calibration - Process of establishing analyzer response to a series of
known  concentrations of gases.

Calibration Curve - The points established in calibration are mathemati-
cally  treated to determine the best  fit line to form the curve.

-------
                                                            Appendix:  B-l(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 3 of 18
Calibration Gases - A set of gases of known concentrations within a
desired range, for the purpose of establishing calibration curves.  The
levels of concentration must bracket the level for which actual measure-
ments are to be made.

California Air Resources Board - Name of the official regulating body in
California which established criteria and recommends legislation for the
control of and standards for vehicle emissions.

Cam - A device that controls or alters motion.  For example, the ignition
distributor breaker cam, in rotating, causes contact points to open and
close.

Capacitor  (Condenser) - An electrical device that permits the storage of
energy.

Capillary Column - A section of tubing with very small inside diameter
used to restrict flow; FID, NOCL.

Carbon - A nonmetallic element (C) found as a constituent of petroleum
in combination with hydrogen atoms, e.g., hydrocarbons; generally measured
as ppmc by FID.

Carbon Dioxide  (CO,.) - A heavy, colorless nontoxic, noncombustible gas;
a by-product of complete combustion.

Carbon Monoxide  (CO) - A colorless, odorless, toxic, combustible gas; a
by-product of incomplete combustion.

Carburetor - A device to meter and mix air and fuel in the correct
proportion, according to the demands of the engine.

Catalytic Muffler - A muffler packed with chemicals which acts as a
catalyst in oxidizing HC and CO; promotes completion of the combustion
of HC and CO.

Centigrade - A temperature scale calibrated at O  , to the melting point
of ice, and 100  , the boiling point of water.

Centrifugal Force - The force tending to make rotating bodies move away
from the center of rotation due to inertia.

Centrifugal Advance Mechanism - A device that advances ignition  timing
with relation to engine speed.

Centrifugal Filter Fan - A filter fan mounted on  the air pump drive
shaft used to clean the air entering the air pump.

Certification - Acceptance, by the Administrator  EPA, of a vehicle type
which has met the Federal Standards  for exhaust and evaporative  emission
control.
                                 195

-------
Appendix:  B-l(LG)
Revision:  0
Date:      June 1975
Page 4 of 18

Charcoal - Treated carbonaceous material obtained by the imperfect
combustion of wood or other organic substances, used to filter or absorb
gasoline vapors.

Chassis Dynamometer - Apparatus used  for applying and measuring rolling
resistance and speed of vehicles; specifically used for exhaust emis-
sions testing by  simulating inertia and horsepower encountered during
the performance of steady and transient states of a vehicle on the road.

Check Valve - A one-way valve to prevent exhaust gas backflow into the
air pump.

Chemiluminescent  - A chemical reaction that gives off energy directly in
the form of light.

Choke Plate - A valve in the carburetor which chokes off air flow through
the carburetor air horn producing a partial vacuum at the main discharge
nozzle(s) for greater fuel delivery,  as during cranking.

Chopper - A two-segmented blade rotating at 5 revolutions per second in
order to block simultaneously, ten times  (10 x) per second, the infrared
beams generated by tungsten filaments inside the NDIR analyzer cells.

Closed System - Related to a crankcase emission system which obtains
fresh air through the carburetor air  cleaner and routes it through a
tube to the filler cap; there is no venting to the atmosphere.

Closed Throttle - Position of carburetor throttle plate at engine idle.
See also wide open throttle.

Code Number - Identification number of any exhaust emission test conducted.

Cold Start Test - A Federal test for  exhaust emissions which is performed
after a 12-hour soak period.

Cold Start "Transient" - First 505 seconds of the 1975 Federal driving
cycle.  (C.S.T.)

Cold Start "Stabilized" - Last 867 seconds of the 1975 Federal driving
cycle   (C.S.S.).

Combustion*- The  burning process which requires three basic ingredients;
fuel, oxygen, and ignition.

Computer - An electronic system capable of performing automatically a
long series of computational or logical operations on stored data, using
an appropriate sequence of stored instructions.

Comguter Program  - The complete plan  for the solution of a mathematical
problem; more specifically the complete sequence of machine instructions
and routines necessary to solve this  problem.
                                196

-------
                                                            Appendix:  B-l(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 5 of 18

Concentration - The weight or volume of one substance with respect to
the total mixture, e.g., grams per liter, parts per million.

Condenser (Cooling) - A water-filled container cooled by ice or refrigera-
tion, housing 1/4" O.D. coils of stainless steel tubing,  its purpose is
to cool the sample exhaust gas to the dew point which removes the moisture
in the sample by condensation.

Console - The structure which houses the analyzers, amplifiers, condenser,
filters, recorders, pumps, plumbing, and controls required to measure
exhaust emission gas concentrations.  Also referred to as an analytical
system.

Constant Volume Sampler (CVS)  - Sampling system in which diluent air is
combined with vehicle exhaust gases and is collected in bags for analysis.

Control Chart - A chronological, graphical comparison of actual data
quality characteristics with limits reflecting the ability to perform as
shown by past experience with the testing variables.

Control Limits - A quality control technique employing a mean (average)
and upper and lower limits.

Control System - Standard production components for the control or
reduction of exhaust and evaporative emissions.

Controlled Combustion System - Modified engine exhaust emission system
used by General Motors Corporation.

Correlation Program - A quality control application for establishing
test cell equivalence by minimizing variability.

Correlation Vehicle - Vehicle used to obtain emission test data for the
correlation program.  These vehicles are specially prepared to minimize
variability in HC, CO, and NO  levels.

Crankcase Emissions - Airborne substances emitted to the atmosphere from
any portion of the crankcase ventilation or lubrication system.

Crowd - An acceleration made at a continually increasing throttle opening.

Cubic Centimeter Displacement (C.C.)  -  The total piston displacement of
an engine obtained from piston diameter, number of pistons, and piston
stroke, calculated in cubic centimeters, 1 inch = 2.54 cm.

Cubic Inch Displacement (C.I.D.)  - Total piston displacement calculated
in cubic inches.  See cubic centimeter  displacement.

Curb Weight - Actual or manufacturers estimated weight of vehicle in
operation with standard equipment.

Cycle - A series of events that occur in a given sequence; e.g., 1 - in
an internal combustion engine, the four strokes; intake, compression,
power, exhaust,  e.g., 2 - in the Federal "Cold Start Test", the series
of transient and steady state driving modes.

                               197

-------
Appendix:  B-l(LG)
Revision:  0
Date:      June 1975
Page 6 of 18
Cycling - Oscillation from a low level to a high level characterized by
periodicity.

Dashpot - A device whose function is to slow down the closing action of
the carburetor throttle plates; aids in the reduction of rich mixtures
in the intake manifold during deceleration.

Data - Detailed information.

Deceleration - The rate of decrease of velocity per unit of time (Nega-
tive acceleration).

Deceleration Valve  (Distributor Vacuum Advance Control Valve)  -  A
device used in conjunction with the dual diaphragm vacuum advance unit
to advance timing under deceleration conditions.

Deflection - Chart recorder pen position reflecting instrument response
on a scale to a gas.

Density - The ratio of the mass of a substance to its volume;  e.g.,
Ib/ft3.

Desiccant - A chemical compound used for the extraction of moisture from
exhaust gases entering the sampling train.

Deviation - Departure from an average value or norm.

Dew Point - The temperature at which vapor, such as water, begins to
condense.

Diaphragm - A flexible membrane, made of fabric and rubber, clamped at
edges and spring loaded, used in various automotive components; pumps
and controls.

Dieseling - Auto ignition, usually applied after vehicle ignition is
shut off.

Differential Pressure - Pressure difference obtained by measuring two
separate reference pointr, e.g., manometer before and after CVS blower.

Diluent - A diluting agent such as the nitrogen or air used in prepara-
tion of standards gravimetric - of or relating to measurement by weight.

Diluent Air - Ambient air drawn into the CVS system to dilute the raw
exhaust gases.

Dilute Exhaust Gas - The combination of vehicle exhaust gases and
diluent air.

Dilution Factor - A number by which lean or rich fuel mixtures are
adjusted to a stoichiometric mixture.
                               196

-------
                                                            Appendix:  B-l(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 7 of  18
 Distributor -  The part of the ignition system which closes and opens the
 circuit to the ignition coil and distributes the resulting high voltage
 surges from the ignition coil to the proper spark plugs.

 Distributor Plate (Stationary)  - The plate in the distributor that is
 fastened to the housing and does not move.

 Distributor^ (Sub-Plate)  - The plate in the distributor that pivots on
 the stationary plate with movement of the vacuum advance.   The points
 and condenser  are usually fastened to this plate.

 Distributor Vacuum Advance Control Valve - Refer to deceleration valve.

 Diurnal Breathing Loss - Fuel evaporative emissions resulting from daily
 fluctuations in temperature to which the fuel system is exposed.

 Drift - Deviation of instruments from zero or set point after cali-
 bration.   See  cycling.

 Driver's Aid - An electronically controlled chart recorder with pre-
 traced driving schedule.   The pen deflection is directly proportional to
 the roll revolutions and therefore by accelerating and decelerating the
 vehicle the driver can maintain the pen on this driving schedule.   The
 chassis dynamometer roll revolutions are converted into electrical
 signals which  then drive the pen on the chart recorder.

 Driver Variability - Inability of a single driver to repeat a CVS  cycle
 precisely the  same way each time;  also inability among drivers to  drive
 a CVS cycle precisely the same way; variability.

 Dry Bulb Temperature - The temperature indicated when a thermometric
 device, such as a thermometer,  is inserted in an air vapor mixture
 (ambient air);  as applied to exhaust emissions testing, the temperature,
 in degrees Fahrenheit, in front of the radiator cooling fan.

* Dual-Diaphragm - A vacuum advance mechanism that attaches to the engine
 distributor to control spark timing.  One diaphragm provides normal
 ignition timing advance for starting and acceleration; the other diaphragm
 retards the spark during idle and part throttle operation.  Some engine/
 transmission applications utilize a special valve to advance timing
 during deceleration to further reduce emissions.

 Duct - A tube  or channel used in conveying air or liquid from one  point
 to another; in emission systems, a device used in the temperature  regu-
 lation of carburetor intake air in conjunction with a thermostatic valve
 and vacuum motor.

 Duct and Valve Assembly - An assembly incorporated in the air cleaner to
 regulate the temperature of carburetor intake air.

 Dump - Bypass  of excess sample flow of exhaust gases during analysis.
                                199

-------
Appendix:  B-1(LG)
Revision:  0
Date:      June 1975
Page 8 of 18
Dynamometer - An apparatus for measuring mechanical power, as of an
engine.

Dynamometer Driving Schedule - Pre-traced curves representing a specific
series of idle, acceleration, cruise, and deceleration modes, of dif-
ferent rates.

End of Line Test - Abbreviated exhaust emission analyses performed on
the vehicles at the end of the production line.

Emission - Substances emitted to the atmosphere by:  chemical reactions
between sunlight and natural organic compounds, evaporation, and com-
bustion of fuels.

Evaporate - To change from a liquid to a gas.

Event Marker - An electric switch operated ink pen on the chart recorders
used to time-orient the chart record with driving mode changes.

Exhaust Emission - Substances emitted to the atmosphere from any opening
downstream from the exhaust port of a vehicle engine; by-products of
hydrocarbon combustion.  Included are raw hydrocarbons, carbon monoxide,
carbon dioxide, oxides of nitrogen, oxygen, and particulate material.

Exhaust Gas Recirculation (EGR) - A system in which a portion of the
exhaust gases are recirculated into the intake manifold for the reduc-
tion of nitric oxide by minimizing peak combustion temperatures and
pressures.

Exhaust Manifold - The part of the engine that provides a series of
passages through which burned gases from the engine cylinders flow.

Exhaust Volume - The amount of gases emitted from the exhaust during a
CVS test;  calculated theoretically by using the blower revolutions, CO
ratio, and test time.

Fahrenheit - A temperature scale calibrated at 32 , to the melting point
of ice, and 212 ,  the boiling point of water.

False Start - An engine stall prior to turning on the driving aid.  A
situation when an engine stops immediately after starting.

Fast Idle Cam - The mechanism of the carburetor that holds the throttle
valve slightly open when the engine is cold, to provide higher engine
speed.

Filter - Pressed fiber pads or fine steel gauze set in a sampling stream
for the removal of particles from the gas before analysis.
                                200

-------
                                                            Appendix:   B-l(LG)
                                                            Revision:   0
                                                            Date:       June 1975
                                                            Page 9 of  18
Flame lonization Detector - An analytical instrument used to measure
hydrocarbon concentration.  The hydrocarbons are first broken up  into
ions by combustion in the flame.  These ions then migrate toward  elec-
trodes creating electrical current which is measured.  The amount of
current generated is directly proportional to the concentration of
hydrocarbon.

Flow Rate - Volume of gas or fluid that passes a given cross section
area per unit time; e.g., cubic feet/hour.

Force - Strength exerted against a mass to cause it to change motion or
deform.

Frequency - The rate of occurrence of an observed value of a variable.

Frequency Distribution - Graphical or tabular description of the  fre-
quency of range of values of the variable.

Fuel - Gasoline normally used in internal combustion engines for  emis-
sions testing; e.g., Indolene 30, Indolene Clear (HO).

Fuel Evaporative Emissions - Unburned fuel vapors collected in charcoal
traps from two areas, air cleaner and vehicle canister.  Part of  Federal
Certification Standards.

Fuel System - The combination of fuel tank, fuel pump, fuel lines, and
carburetor, or fuel injection components, and includes all fuel system
vents and fuel evaporative emission control systems.

Gain - Amplification of a signal.

Gain Control - Calibrated potentiometer for the adjustment of signal
amplification.  Used to set upscale calibration point while flowing a
normalizing gas.

Gas Permeable - Any material that allows gas to diffuse through its
surface.  Usually referred to in O  analyzer membrane.

Gram - Metric unit of weight equal to approximately 0.035 ounces.

Grains per mile - Unit of measurement for accumulated weight of  exhaust
emissions per vehicle mile driven on the chassis dynamometer roll.

Gravity - The gravitational attraction of the earth's mass  for  bodies at
or near its surface.

Gravimetric - Of or relating to measurement by weight.

Gross Vehicle Weight - Curb weight plus rated load.   (Emission  control
systems not required currently on engine applications for vehicles that
exceed 6000 Ib GVW).
                                201

-------
Appendix:  B-l(LG)
Revision:  0
Date:      June 1975
Page 10 of 18

Hang-up - The resultant effect of residue from sample gases collecting
on the inner surface of the gas sample train.  This effect is evident
when instruments fail to return to zero deflection with nitrogen gas
introduced into the measuring system after a test.  This term also
applies to slide wire friction on the recorders.

Heat Build - The process by which the fuel in the vehicle's tank is
heated at a prescribed rate during the diurnal breathing loss test.

Heat Exchanger - A device in the Constant Volume Sampler where cooled
air or water, circulating through a sleeve surrounding the exhaust gas
stream, absorbs heat from the gas thru maintaining an even temperature.

Hesitation - A temporary lack of response in acceleration rate.

Horsepower - Unit of work, equivalent to 550 foot-pounds per second.

Hot Idle Compensator - A thermostatically controlled carburetor valve
that opens whenever inlet air temperatures are high.  Additional air is
allowed to discharge below the throttle plates at engine idle.  This
feature improves idle stability and does not allow the rich fuel mixture
normally associated with increased fuel vaporization of a hot engine.

Hot Soak Loss - Fuel evaporative emissions collected during the first
hour immediately following the dynamometer test.

Hot Start Test - Any exhaust emissions test performed after a prescribed
engine warm-up period which follows the same sequences as a CVS test.

Humidity Factor (K) - correction factor used to adjust nitric oxide
emission values to standard humidity at 75 grains of water per pound of
dry air.

Hydrocarbons - Organic compounds containing carbon and hydrogen atoms in
numerous combinations (H C )  which occur in nature as living organisms,
crude oil, natural gas, and coal.  Excessive amounts in the atmosphere
are considered undesirable contaminants and a major contributor to air
pollution.

Idle Limiter - A device to control the amount of adjustment of idle
mixture screws, and therefore, maximum idle fuel richness of the car-
buretor.  Also aids in preventing unauthorized persons from making
overly rich idle adjustments.  The limiters are of two distinct types;
the external plastic limiter caps installed on the head of idle mixture
adjustment screws or the internal needle type located in the idle channel.

Idle Mixture Adjusting Screws - The adjusting screw that can be turned,
in or out, to lean or enrich the idle mixture.

Idle Port - The opening into the throttle body of the carburetor through
which the fuel in the idle circuit discharges.
                               202

-------
                                                            Appendix:   B-l(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  11 of  18
Idle Vent - An opening from an enclosed chamber through which air can
pass to lean out air/fuel ratio during idle conditions.

Indolene Clear - A petroleum based, lead free, fuel for vehicles used in
exhaust emissions testing.  (Ref. Federal Register, Vol. 39, No. 14,
Title 40, Part 85.)

Indolene 30 - Emissions test fuel containing 3 cc. of lead per gallon of
fuel.  (Ref. Federal Register, Vol. 39, No. 14, Title 40, Part 85.)

Inertia - A property of matter by which it remains at rest or in uniform
motion in the same straight line unless acted upon by some external
force.

Inertia Weights - Flywheels having specified weights which are connected
to the dynamometer drive roll for the purpose of simulating vehicle
inertia.

Infrared Radiation - Electromagnetic radiation from two to fifteen
microns wavelength produced in nature by black body sources.  Nearly all
chemical compounds absorb infrared radiation and can be identified by
this specific absorption.  Theory of non-dispersive infrared analyzer.

Inlet Depression - Pressure differential between the dilute exhaust
mixture entering the CVS positive displacement pump and the atmosphere.

Intake Manifold - The part of the engine that provides a series of
passages from the carburetor to the engine cylinders through which the
air fuel mixture flows.

Integrate - A method that uses the collective properties of a group of
numbers to compute a value which is representative of that group -
average value.

Inverse - Direct opposite.  When two factors are inversely related one
increases as the other decreases proportionally.

Ions - An atom or group of atoms that carries a positive or negative
electrical charge.

Kickdown - Release of the automatic choke  from high cam position on a
cold engine by increasing engine speed to  2,500 ±  100 RPM and releasing
accelerator within 3 seconds.  Deactivation of the fast idle mechanism.

Knock  (Ping) - Auto ignition that  is audible.

Lead - Tetraethyl lead added to gasoline as a lubricant and antiknock
additive.

Light Duty Vehicle - A motor vehicle designed for  the  transportation of
persons or property on a  street or highway and weighing 6000 pounds
gross vehicle weight or  less.
                                 203

-------
 Appendix:   B-l(LG)
 Revision:   0
 Date:       June 1975
 Page  12 of 18
Linear  - A relationship between two variables  such that a change in one
is accompanied  by a proportional change in the other.

Loaded  Vehicle  Weight - Vehicle curb weight of a light duty vehicle plus
300 Ibs.

Magnehelic Gauge - Pressure gauge referred to  by the manufacturers brand
name.   Commonly incorporated in NDIR consoles.

Malfunction - The act of performing improperly or a condition describing
vehicle or test equipment failure.

Manifold - A tube or pipe for conveying liquids or gases as in the
intake  of  fuel/air mixtures and the exhaust of burned gases.

Manifold Control Valve - A thermostatically operated valve in the exhaust
manifold for varying heat to intake manifold during the engine warm-up
period.

Manometer  - A glass tube, either "u" shaped or linear, filled with a
liquid  and clamped against a retainer having a graduated scale used to
measure pressure or vacuum.

Mass -  The quantity of matter in a body as measured in its relation to
inertia.

Maximum Rated Horsepower - Maximum brake horsepower output of an engine.

Micron  - A unit of length equal to 3.937 x 10   inch used in measuring
wavelengths of  light,  and particle diameter.

Modal Analysis - Summation of exhaust emission data for each specific
mode throughout a test cycle.

Mode -  Division of a test cycle into established segments which describe
the vehicle's operating state; acceleration, deceleration, cruise, and
idle conditions.  (Transient or steady states.)

Modification - A change from the original, such as engine modifications;
design  change,  component change, etc.

Modulator  - A device used to integrate two signals into one; to vary the
amplitude,  frequency,  or phase of a carrier wave or signal.

Mole -  The molecular weight of a compound expressed in grams; the number
of moles of  a compound is equal to its mass in grams, divided by the
molecular  weight.

Mole Percent -  The number of moles of a compound in a mixture divided by
the total  number of moles and multiplied by 100.

NBS Cylinder - A gas standard prepared and certified by the National
Bureau  of  Standards.
                                20k

-------
                                                            Appendix:   B-l(LG)
                                                            Revision:   0
                                                            Date:       June  1975
                                                            Page  13 of  18
Ni_trlc_0_xide - A colorless, toxic gas  (NO) formed by the oxidation of
nitrogen; also a by-product of the combustion of hydrocarbon  fuels.

Nitrogen - A colorless, tasteless, odorless, nontoxic gas  (N  ) that consti-
tutes 78 percent of the atmosphere by volume.

Nitrogen Dioxide - A brown, highly toxic gas (NO ) formed  by  the  union
of nitric oxide (NO) and oxygen (O ) or ozone (O ).

Non-Dispersive Infrared - NDIR:  gas analyzer which uses a specific
infrared wavelength for analyzing each different component; e.g., HC,
CO, CO , and NO.

Non-Dispersive Ultraviolet - NDUV:  gas analyzer which uses a specific
ultraviolet wavelength for analysis for NO .  This instrument is  in
limited use at the present time.

Normal Distribution - A frequency distribution whose graph is bell-
shaped and symmetrical.  This distribution is common in the data  from
many natural events and many measurement processes.

NO  - Refers to the oxides of nitrogen which are produced  during and
after combustion.   The sum of the NO and NO  concentrations in the
exhaust sample.

NO  Analyzer- Analytical instrument used to analyze NO and NO by
chemiluminescence.  The formation of NO  by the reaction of NO and 0
(ozone) emits light the intensity of which is directly proportional to
the concentration of NO and can be measured by a photomultiplier tube.

Nozzle - A restricted orifice or hole; the final outlet for air entering
the exhaust manifold on injector emission systems; fuel discharge point
of the carburetor main system.

Open System - Crankcase emission control system which draws air through
the oil filler opening.

Oxidation - A chemical reaction in which oxygen combines with an element
or compound to form a new compound, e.g. , the action of oxygen on iron
to form rust; the action of oxygen on hydrocarbons to form oxidized
hydrocarbons (aldehydes).

Oxides of Nitrogen - See NO .
                           X,

Oxygen - An element that is found free as a colorless, odorless, taste-
less gas constituting 20.9 percent of atmospheric air by volume; sup-
ports life and the combustion process; contributes to the  formation of
exhaust process; e.g., CO, C0_, NO, HO.
                             £»       £

Partial Pressure - The pressure exerted by any single gas  in  a mixture
of gases.
                                205

-------
Appendix:  B-l(LG)
Revision:  0
Date:      June 1975
Page 14 of 18
Part Throttle Acceleration - An acceleration made at entirely wide open
throttle (from any speed).

Photochemical Smog - Misnomer for a type of air pollution formed by the
reaction of sunlight with hydrocarbons, nitrogen oxides, and ozone.
Smog is a combination of smoke and fog which is not necessary for the
formation of photochemical air pollution.

Photomultiplier - A vacuum tube which measures light intensity and
amplifies this signal into milliamperes.

Polyurethane - Synthetic substance used in filtration materials normally
associated with filtering carburetor inlet air.

Positive Crankcase Ventilation Valve - Controls crankcase vapors dis-
charged into the engine intake system and passes them through the engine
cylinders rather than being discharged into the air.

Positive Displacement Pump - A pump, usually of the rotary vane type,
which displaces a certain volume per pump revolution.  This volume
theoretically does not vary, therefore, knowing the number of revol-
utions of the pump, the inlet depression and temperature and the cali-
brated displacement, the total volume passed through the pump can be
calculated.  This type of pump is the basis for the design for most
constant volume samplers.

Potentiometer - A three terminal, variable resistance in the analyzer
amplifier/control sections used to adjust the upscale calibration point.

Power Absorption Unit - A comporient of the chassis dynamometer for the
absorption of vehicle power.

Power Switch - Generally an on-off switch, but as applied to NDIR a
three-position rotary switch which controls the electronic circuitry;
(1) the OFF position removes power from all circuit components; (2) in
READ position, the meter indicates the output of the amplifier/control
section and this position is used for calibration and analysis; (3) in
TUNE position, the meter indicates the rms value of the half-wave recti-
fied carrier wave.

Pressure - Force applied to or distributed over a surface; measured as
force per unit area; e.g., Ibs/sq.in.  Absolute Pressure:  Measure with
respect to zero pressure.  Gauge Pressure:  Measure with respect to
atmospheric pressure,  e.g. , absolute pressure = gauge pressure + atmos-
pheric pressure.

Primary Calibration Gas - A gas having a known concentration which has
been accurately measured, usually gravimetrically.  The concentration
should be known to within ±0.5 percent.
                                206

-------
                                                            Appendix:  B-l(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 15 of 18
Probe - Stainless steel tubing which is fitted inside a test vehicle
tailpipe for the collection of exhaust gases for analysis.

Procedure - A step-by-step method of conducting a test or performance of
an operation.

Purge - An operation included in the sampling and analysis of concen-
trated exhaust gases by which a non-reactive gas such as nitrogen is
flowed through the analyzer in the reverse direction for the purpose of
driving out responsive gases.  The process by which the sample bags are
filled and evacuated with air or N_ for the purpose of removing the
sample gas.

Quality - The composite product characteristics of engineering and
manufacturing that determine the degree to which the product in use will
meet the expectations of the customer.  For testing purposes it is the
degree to which the measurement system produces emission data within
acceptable limits.

Quality Assurance - A system for integrating the quality functions of
the various groups in an organization so as to assure production and
service at the most economical levels which satisfy the quality require-
ments of the testing facility or contractor.

Quality Control - Any program or device employed to minimize sources of
variation inherent in all analytical and technical functions.  Any
procedure designated to maintain the reliability of emission test data.

Rated Speed - Speed at which manufacturers specify the maximum rated
horsepower of an engine.

Ratio - The expression of the proportional mixture of two substances,
usually expressed as a numerical relationship, such as 2:1, 10:1, etc.,
in emission systems, concern is with air-fuel mixtures.

Raw Sampling - Collection of exhaust gases for analysis at any point
between the exhaust manifold and the tailpipe.

Reactor System - Similar to an air injection system, but employing a
larger exhaust manifold having insulated walls for less heat transfer to
maintain high exhaust temperatures for continuing oxidation of exhaust
gases in the manifold.

Recorder Response Time - The time required for the chart recorder pen to
move from zero to 90-100 percent of upscale position on the introduction
of a normalizing gas to the analyzer.

Relief Valve - A pressure limiting valve located in the exhaust chamber
of the air supply pump.  Its function is to limit the air flow to the
exhaust ports when the vehicle exhaust back-pressure exceeds a pre-
determined value.
                                207

-------
Appendix:  B-1(LG)
Revision:  0
Date:      June 1975
Page 16 of 18

Retard - To delay the timing of the spark to the combustion chamber;
usually associated with spark timing mechanisms of the engine.

Road Draft Tube - A means by which the engine crankcase was ventilated
prior to the introduction of crankcase emission control systems.

Road Load - The horsepower required to drive a vehicle at zero grade and
zero wind velocity at. a constant speed to overcome rolling and wind
resistance.  The value, of HP varies with speed and vehicle weight.

Rolls - A common name for chassis dynamometer.

Rotometer - A gauge that consists of a graduated glass tube containing a
free float for measuring the flow of a fluid or gas; a flowmeter.

Running Loss - Fuel evaporative emissions resulting from an average trip
in an urban area or the simulation of such a trip.

Sampling System - The total plumbing required to obtain a representative
sample of exhaust gases for analysis.

Span - The act of introducing an end point or set adjusting point gas
into an analyzer and the response to a predetermined set point for that
gas.

Stall - Inability of an engine to continue operating at any time other
than starting (see false start).

Standard Deviation - A statistic indicating the variability of a distri-
bution, calculated by obtaining -the sum of the squares of the differences
of all values from the arithmetic mean.

Statistics - A branch of mathematics dealing with the collection,
analysis, interpretation, and presentation of numerical data.

Steady State - A condition of vehicle performance on the dynamometer
rolls in which engine speed and/or rpm remains constant during a speci-
fic test condition; e.g., "Road Load", "Idle Emissions", "HP Setting",
and "Cruise Modes".

Stoichiometric - As applied to the spark ignition engine, the ideal
air/fuel mixture for complete combustion of fuel.

S toichiome try - Applications of the laws of definite proportions and of
the conservation of matter and energy to chemical activity.

Stretchiness - A lack of anticipated response to throttle movement.

Surging - A condition of leanness resulting in short fluctuations in
engine and vehicular speed.
                                208

-------
                                                            Appendix:  B-l(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 17 of 18

System Response Time - The time interval between the introduction of
sample gas into the probe and when the chart recorder indicates the
presence of this gas.

Tachometer - An instrument for measuring engine rpm.

Tank Fuel - Fuels representative of commercial fuels which are generally
available through retail outlets.

Tank Fuel Volume - Volume of fuel in the fuel tank, prescribed to be a
percentage of the nominal tank capacity rounded to the nearest whole
U.S. gallon.

Test (emissions) - Qualitative and quantitative determinations of the
various components of exhaust gases.

Test Cell - An area specifically designed and equipped for the purpose
of qualitative and quantitative determinations of species of exhaust
gases.

Thermostat - A valve which depends on heat to control temperature by
opening or closing a damper.  In emission systems, to control hot or
cold carburetor inlet air.

Timing - The point at which a spark plug fires in relationship to the
rotation of the crankshaft and piston.

Tip-In - Vehicle response to the initial opening of the throttle.

Top Dead Center (TDC) - The highest point a piston travels in the cylinder.

Transducer - A device which is actuated by power from one system so  that
it may supply power in any other form to a second system; e.g., the
conversion of torque to an electrical signal for recording.

Train - See console,  analytical system.

Trap - A cylindrical, usually stainless steel, device located at the
bottom outlet of the condensing coils inside the ice water bath  for  the
purpose of collecting moisture from a sample gas prior to analysis.

Tune Adjustment - NDIR:  control used to tune the oscillator if meter
does not indicate the correct value when the power  switch is in  tune
position.

Uncontrolled System - A term applied to vehicles without emission
control systems.

Vacuum - A  term to describe a pressure that is  less than atmospheric
pressure.

Vacuum Advance  - A mechanism which advances ignition timing  in  relation-
ship  to engine  load conditions.  This is achieved  by using engine  vacuum.
                                209

-------
Appendix:  B-l(LG)
Revision:  0
Date:      June 1975
Page 18 of 18
ya_cuum_Control Temperature Sensing Valve - A valve that controls mani-
fold vacuum to the distributor advance mechanism under hot idle conditions.

Vane - Any flat, extended surface attached to an axis and moved by or in
air or liquids.  Part of the integral revolving portion of an air supply
pump.

Vehicle Curb Weight - The manufacturer's estimated weight of the vehicle
in operational status including standard and optional equipment and
weight of fuel at nominal tank capacity.

Ventilation - The process by which fresh air is caused to circulate, so
as to replace impure air.  Principle utilized in crankcase emission
systems.

Visual Integration Analysis - A method for visually averaging by means
of a template, raw modal deflections from test chart traces.

Wavelength - The distance between adjacent crests of the wave form in a
beam of radiation.

Weighting - A numerical coefficient assigned to a term to express its
relative importance in a frequency distribution; spec., in exhaust
emissions testing, the modal weighting factors are based on modal time
and modal exhaust volume.

Wet Bulb Temperature - The temperature, in degrees Fahrenheit, from the
passage of ambient air over a wetted surface to reach a condition of
dynamic equilibrium.  In this state, the heat transferred from the
ambient air will be equal to tha"t transferred from the surface in the
diffusing vapor; used with the dry bulb temperature to calculate rela-
tive humidity and corresponding correction factors for humidity.

Wide Open Throttle - Position of the carburetor throttle plate when the
accelerator is depressed to the maximum allowable travel.

Zero Adjust - Control used to set zero point while flowing nitrogen
through analyzers.
                                210

-------
    Appendix B-2




LIST OF ABBREVIATIONS
          211

-------
                           Appendix B-2
                                                            Appendix:   B-2(LG)
                                                            Revision:   0
                                                            Date:       June 1975
                                                            Page 1 of  4
                       LIST OF ABBREVIATIONS
The following abbreviations are representative of terms commonly used in
emissions testing.   Variations in capitalization are widespread, as no
specific rule governs their use.  Therefore, the interchangeable use of
capital or lower case letters is acceptable.
A/C       Air Conditioning
AC        Alternating Current
A.I.R.    Air Injection Reaction
AMA       Automobile Manufacturers Association
Accel.    Acceleration
Ar.       Argon
ASTM      American Society for Testing and Materials
ATDC      After Top Dead Center
BAR       Bureau of Automotive Repair  (California)
Bar       Barometric Pressure
B/F       Backfire
BHP       Brake Horsepower
BTDC      Before Top Dead Center
C         Centigrade; also Carbon
CAP       Clean Air Package
CARB      California Air Resources Board; also Carburetor
CC        Cubic Centimeter(s)
C.C.S.    Controlled Combustion System
CFH       Cubic Feet Per Hour
CFM       Cubic Feet Per Minute
CGA       Compressed gas association  (usually refers  to  a
            type of cylinder pressure regulator  connector)
CID       Cubic Inch Displacement
CL        Chemiluminscent Analyzer
CO        Carbon Monoxide
CO        Carbon Dioxide
Cone.     Concentration
CSD       Certification and Surveilliance Division
CT        Closed Throttle
Cu.In.    Cubic Inch(es)
CVS       Constant Volume  Sampler
 DC        Direct Current
 Decel.     Deceleration
                                213

-------
Appendix:  B-2(LG)
Revision:  0
Date:      June 1975
Page 2 of 4
Displ.    Displacement
Dist.     Distributor
E  & D     Evaluation  & Development
EGR       Exhaust Gas Recirculation
EP       End Point
EPA       Environmental Protection Agency
E/S       Engine Stumble
E.S.      Engine Surge
Evap.     Evaporative
 F.         Fahrenheit
 FET       Federal Emission Test
 FID       Flame lonization Detector
 FL         Full Load
 FT.       Foot/Feet
 FTP       Federal Test Procedure
 Gal.       GalIon(s)
 Gm.        Gram(s)
 GVW        Gross Vehicle Weight
 H        Hydrogen
 HC       Hydrocarbon(s)
 HDT      Heavy Duty Testing
 He       Helium
 HEW      Department of Health,  Education,  and Welfare
 Hg       Mercury
 HP       Horsepower
 HWFET    Highway Fuel Economy Test
 I         Current  (electrical)
 IBP       Initial  Boiling  Point
 ICE       Internal Combustion Engine
 IN.       Inch(es)
 Ind.-Cl   Indolene Clear;  also  Indolene-HO
 Ind.-30   Indolene 30
 ID        Internal Diameter
 K         Correction factor for Humidity
 LA-4      Federal Driving Cycle
 LOT       Light Duty Testing
 Lb.       Pound(s)
 Max.       Maximum
 Mi.        Mile

-------
Min.      Minimum;  also minute(s)
Ml.       Milliliter(s)
MPH       Miles Per Hour
mm.       Millimeter(s)
mv.       Millivolt(s)
N?        Nitrogen
NDIR      Non-Dispersive Infrared
NDUV      Non-Dispersive Ultraviolet
NO        Nitric Oxide
NO        Nitrogen Dioxide
NOCL      Nitric Oxide Chemiluminescent
NO        Oxides of Nitrogen
N/V       Ratio of wheelturns to drive shaft turns
OD        Outer Diameter
OEM       Original Equipment Manufacturer
02        Oxygen
O         Ozone
Pb        Lead
PCV       Positive Crankcase Ventilation
Pot.      Potentiometer
ppm.      Parts per million by volume
ppmC      Parts per million carbon-methane by volume
Psia.     Pounds per square inch absolute
PSI (psig.)Pounds per square inch gauge
PT        Part Throttle
PTA       Part Throttle Acceleration
PTD       Part Throttle Deceleration
QA        Quality Assurance
QC        Quality Control


R         Rankine; also resistance; also range
Rev.      Revolution
RPM       Revolution per minute
R/S       Roll Slippage
RVP       Reid Vapor Pressure
SAE       Society of Automotive Engineers
S/B       Sensitive Brakes
Sec.      Second(s)
SO        Sulphur Dioxide
SO        Sulphur Trioxide (sulphate)
S.S.      Stainless Steel
                                                            Appendix:  B-2(LG)
                                                            Revision:  0
                                                            Date:      June 1975
                                                            Page 3 of 4
                                215

-------
Appendix:  B-2(LG)
Revision:  0
Date:      June 1975
Page 4 of 4


 TDC       Top Dead  Center
 TLV       Threshold Limit Value
 TML       Tetramethyl  Lead
 V         Venturi(s)
 Vac.       Vacuum
 VIA       Visual Integration Analysis
 Vs         Versus
 VWA       Volume Weighted Ambient
 WOT      Wide  Open Throttle
 Wt.      Weight
                                 216

-------

-------
         Appendix C

QUALITY MANAGEMENT PROCEDURES

             FOR

    MOBILE SOURCE TESTING
    (LIGHT DUTY VEHICLES)

-------
                           INTRODUCTION
               QUALITY MANAGEMENT PROCEDURES MANUAL

      The Environmental Protection Agency places prime importance on the
integrity and validity of data and reports generated during Mobile
Source Emission Testing.  To achieve an optimum degree of confidence in
the ultimate results of these tests, a quality assurance program must be
integrated into the emission measurement system.  Primary goals of a
quality assurance program are improvements in the credibility and docu-
mentation of emission measurements.  The achievement of these goals
calls for quality assurance in nearly all segments of emission testing
activities, procurement control, standards and calibration, laboratory
operations and documentation control.
      This manual presents Quality Management Procedures (QMP) governing
the interrelationships between quality functions and various departments.
It is a means for assigning quality responsibilities to all key person-
nel/functions in the organization.
      The chart which appears in Section 2.0 is designed to show only
those functions requiring inclusion in a quality program.  It does not
represent any existing organizational chart either at the EPA emission
facility or other organization.  The line of authority and assignment of
quality functions will vary with the size and scope of a particular
organization.
      This manual may seem too complex and extensive to be incorporated
into a small company involved in Mobile Source Emission Testing, how-
ever, it can provide guidelines for the development of a Quality Assur-
ance Program manual.  In small testing facilities many of the functions
and responsibilities may be delegated to a single person within the
organization.  The main objective of this manual, which is the assign-
ment of responsibilities and documentation of procedures used to accomp-
lish a quality function, should be kept in mind when planning a quality
assurance program. Such a program need not be elaborate and costly to
adequately assure the validity of the data produced.
      The cost effectiveness and capability of a quality program is of
prime importance in selling the program to top management.  Therefore,
in the initial planning of an emission testing quality program the ratio
                               C-l

-------
of valid to invalid tests should be considered.  An extensive audit of past
data and testing history would be a logical starting point in planning to
reduce the number of invalid tests, decrease the overall costs of testing, and
improve the credibility of test data.
          The complete support of management is a prerequisite to an effective
quality program.  Management attitude towards the quality program will be
reflected throughout the organization.  Their failure to support a quality
function for the sake of getting a job done faster or for an apparent reduction
in cost against the advice of Quality Assurance Management will make the
program ineffective from that point on.  On the other hand. Quality Management
has the responsibility to actually demonstrate cost effectiveness and production
of valid and reliable data.  Along with the careful planning, auditing and
detailing of the program an analysis of its effectiveness as well as that of
the measurement system must be performed.
          Therefore a QMP manual is necessary to formalize and document the
quality program for ease of implementation and definition.  Constant review
and analysis of the documented program will result in changes to procedures
and assignment of responsibilities, requiring manual revisions to maintain a
viable and effective program.
          Other manuals documenting specific step by step procedures for the
performance of emission tests, maintenance, training, etc., should be developed
and utilized in the measurement system.  A test procedure manual detailing the
1975 light-duty gasoline-powered vehicles emission measurement procedures has
been developed and is presented in Volume II of this report.
                                  C-2

-------
                       QUALITY MANAGEMENT PROCEDURES

                        Change and Revision Summary
 EPCN
Number
Date
                   Procedure
                 Number
              Revision
                Date
                              Procedure Title
Entered
  By
                                     221

-------

-------
QUALITY MANAGEMENT PROCEDURES
         SECTION 1.0
            INDEX
                223

-------

EPA QUALITY
SUBJECT:

MANAGEMENT

CM
PROCEDURE

MP NO. REVISION DATE
1.1

TABLE OF CONTENTS


i
ii

Q.M.P

1.1
QUALITY

Changes
MANAGEMENT PROCEDURES MANUAL
Title Page
and Revisions


Introduction



Table of


Section 1.0 index
Contents
Effective
Date(s)


Section 2.0 Organization
2.1
2.2
Function/Responsibility Chart
Functional Outline - Administrative
Services
2.3
Functional Outline - Laboratory
Operations
2.4
Functional Outline - Quality Assurance
Management
2.5
Functional Outline - Data Services
Section 3.0 Administration
3.1
Preparation of Quality Management
Procedures
3.2
3.3
Document
Control

Quality Assurance Training Program
Section 4.0 procurement
4.1
4.2

Procurement Document Review
Receiving Inspection
CONCURRENCES
DATE
PREPARED BY:
APPROVED BY:
IMPLEMENTATION
PAGE i OF 2
DATE ISSUED:

-------
                                                                 QMP NO. 1.1
                                                                 Page 2 of 2
QUALITY MANAGEMENT PROCEDURES MANUAL (Continued)

                                                                      Effective
Q.M.P                                                                  Date(s)


                   Section 5.0  Standards & Calibration

5.1                 Equipment Calibration & Cycle Control
5.2                 Calibration Instruction Document
                      Maintenance
5.3                 Calibration Intervals
5.4                 Calibration Standards
                       Section 6.0  Test Operations

6.1                 Light Duty Test Operations
6.2                 Coordination and Implementation of
                      Equipment or Procedures Change
                      Notices
6.3                 Test Vehicle Fuel Control
6.4                 Scheduling
6.5                 Testing Facility Support Services
6.6                 Data Validation
                      Section 7.0  Forms Instruction

7.1                 Instrument Loan Order
7.2                 Calibration Control Card
7.3                 Calibration Order
7.4                 Test Condition Report
7.5                 Equipment/Procedure Change Notice
7.6                 QMP Change Summary
7.7                 Rejection Report
                                      226

-------

-------
QUALILTY MANAGEMENT PROCEDURES
          SECTION 2.0
         ORGANIZATION
              227

-------
  •o
  TJ
  30
  O
  CD
  •<
73
m
o

-<
         O
         O
         o
         m
         in
oo
  m
  CO
  to
  c
  m
  o
      £7)
                 PROCUREMENT
                 CONTROL
                STANDARDS &
                CALIBRATION
                TEST QUALITY
                CONTROL
                 CORRECTIVE
                 ACTION
            DEFICIENCY
            REVIEW
                  AUDIT
                                             FACILITY
                                             DIRECTOR
                                DATA
                                VALIDATION
                                STATISTICAL
                                ANALYSIS
                                SYSTEMS
                                DEVELOPMENT
                                COMPUTER
                                OPERATIONS
 TRAINING &
 PERSONNEL
PURCHASING
 FACILITY
 SERVICES
 DOCUMENT
 CONTROL
                                        CO
                                        c
                                        CD
                       m
                       2


QUALITY
ASSURANCE
MANAGEMENT

—


DATA
SERVICES




ADMIN.
SERVICES

—

1
LABOR
OPERA

TEST
SCHEDULING,
EMISSION
TESTING
                                                                                              9
                                                                                              H
                                                                                              8
en
H
w
H
f
H
                                        8
                                                                                            o
                                                                                            m
                                                                                            X
                                                                                            m
•o
a
O
o
m
o
a
m
FUEL 5- GAS
STORAGE
CHEMICAL
ANALYSIS
                    EQUIPMENT
                    SERVICES
                                                                                               CO
                                                                                               o
                                                                                                    o
                                                                                                    J>

-------
 EPA QUALITY MANAGEMENT  PROCEDURE
                         QMP NO.
                             2.2
                                                                    REVISION  DATE
 SUBJECT:
        FUNCTIONAL OUTLINE - ADMINISTRATIVE SERVICES
                                    SUMMARY

    The Administrative Services performs all the necessary peripheral functions
    required by the laboratory such as purchasing, facility engineering,  equipment
    management, training,  forms and document control.
    RESPONSIBILITY
    Purchasing
    Facility Services
    Training & Personnel
                   FUNCTION

1.   Purchases all materials, equipment, instruments,
     expendable  items, office equipment, etc., which
     are used by the laboratory.

2.   Requests Quality Assurance to provide quality
     requirements and approvals of purchase orders
     and related specifications and drawings.

3.   Requests Quality Assurance approval and review
     of suppliers' products as required.

4.   Establishes contracts for facility services such
     as equipment maintenance and calibration.

5.   Provides for all facility engineering requirements
     such  as building modifications, plumbing, electrical
     wiring, heating, cooling, ventilation and general
     storage.

6.   Initiates,  recommends, implements safety program
     procedures  and equipment to meet personnel and
     building requirements in accordance with the
     applicable  regulations.

7.   Controls and maintains inventory of all parts,
     supplies, equipment, etc., used by the laboratory.
     Maintains records of equipment on loan and surplus
     equipment inventory.

8.   Maintains personnel records and provides for
     personnel requirements of the laboratory by issuing,
     advertising and posting job descriptions of avail-
     able  openings.  Conducts preliminary interviews and
     schedules interviews with the appropriate department
     supervisors or manager.
               CONCURRENCES
                     DATE
IMPLEMENTATION
PREPARED BY
                                                          PAGE  i  OF  2
APPROVED BY:
                              DATE  ISSUED:

-------
                                                                 QMP No. 2.2
                                                                 Page 2 of 2
RESPONSIBILITY                               FUNCTION

Training & Personnel (continued)

                         9.   Conducts training and orientation programs for
                              new employees.  Provides facilities and
                              support for technician training and evaluation
                              programs.

Document Control        10.   Issues and controls procedures and equipment design
                              documentation and revisions and provides for the
                              timely revisions of procedures manuals used in
                              the laboratory.
                                      230

-------
QMP NO. REVISION DATE
EPA QUALITY MANAGEMENT PROCEDURE 2.3
SUBJECT:
FUNCTIONAL OUTLINE - LABORATORY OPERATONS

SUMMARY
Laboratory Operations is responsible for the daily operations of the test
facility. It has the responsibility for the performance, calibration, mainte-
nance and analytical requirements necessary to perform the emission tests and
is responsible for the personnel, equipment and vehicles used in the performance
of these tests.
RESPONSIBILITY FUNCTION
Test Scheduling
Emission Testing
Fuel & Gas Storage
Chemical Analysis

1. Receives, inspects and schedules vehicles for
testing. Returns vehicle in its original condition
to the owner after successful completion of the
emission test.
2. Conducts emission testing on vehicles according to
the government regulations and procedures outlined
in the Test Procedures Manual for Light-Duty
Vehicle Emission Measurement Facilities.
3. Measures and reports vehicle gaseous emissions and
fuel economy according to the Federal Procedures .
4. Performs non-routine emission tests as requested
by other divisions. Test procedures for non-
routine tests shall be documented and approved
by Quality Assurance and the Laboratory.
S. Completes all required forms and records necessary
for the performance of an emission test.
6. Provides for the proper storage and handling of fuel
and gases by initiating detailed procedures contain-
ing Quality Assurance checks to prevent errors such
as the use of improper fuel in the vehicles .
7 . Performs chemical analysis as required for
receiving inspection and non-routine emission
testing.
8. Performs analysis and reports results of all cali-
bration gases used by the facility and other
facilities requesting this service. Analysis is
traceable to gravimetric standards by not more
than one generation.
CONCURRENCES DATE IMPLEMENTATION
PREPARED BY:
APPROVED BY:
PAGE X OF 2
DATE ISSUED:
231

-------
                                                                 QMP No. 2.3
                                                                 Page 2 of 2
RESPONSIBILITY

Chemical Analysis
(Continued)
Equipment Services
                     FUNCTION

 9.   Prepares gravimetric binary gas mixtures to be
      used as laboratory primary standards and main-
      tains the standards inventory to assure adequate
      availability of such standards.

10.   Designs, fabricates, inspects parts, equipment
      and instrument systems requested by a Job Order
      accompanied by appropriate approved drawings
      issued by Production Control.  Reports completion
      of Job Order to Production Control.

11.   Maintains records of surplus and loaned equipment
      and determines disposition.

12.   Provides for periodic calibration of all instru-
      ments and equipment used in the test facility
      to assure the accuracy and reliability of the
      test data.  Reports data and records of calibra-
      tion to Quality Assurance.

13.   Performs maintenance of all instruments and
      equipment on an "as needed" or periodic basis.
      Performs preventive maintenance on equipment to
      assure trouble-free operation and avoid major
      equipment malfunctions.
                                     232

-------
EPA QUALITY MANAGEMENT
QMP NO. REVISION DATE
PROCEDURE 2.4
SUBJECT:
FUNCTIONAL OUTLINE - QUALITY ASSURANCE MANAGEMENT

SUMMARY
Quality Assurance has the overall responsibility for ensuring adherence to
quality and reliability standards throughout all phases of mobile source emission
testing and related facility operations.
RESPONSIBILITY FUNCTION
Quality Assurance 1.
Management
Procurement Control 2.
3.
Standards and 4.
Calibration
Test Quality Control 5.
Corrective Action 6.
Deficiency Review 7.

CONCURRENCES
PREPARED BY:
APPROVED BY:
Formulates, recommends, and implements Quality
Management Procedures, Quality Planning and Quality
Cost programs consistent with management objectives
and mobile source emission measurement requirements .
Performs source inspection of suppliers as required
for quality control of procured material and
services.
Monitors, plans and performs required inspection
and test of all incoming materials and equipment
to be used in the mobile source emission test
operations. Rejects those items not meeting
specifications and maintains records denoting
acceptance or rejection of incoming materials
and equipment.
Directs and coordinates the system for controlling
the accuracy of measurement through the calibration/
maintenance and control of all standards and
measurement test equipment.
Monitors all mobile source emission operations and
verifies the authenticity of the resultant data and
reports . Develops and maintains inspection plans
and implements quality control programs.
Establishes and coordinates a systematic and timely
"closed loop" mechanism for feedback of the
unsatisfactory conditions to those responsible for
corrective action, with follow-up until completion
of satisfactory corrective action.
Conducts reviews of unsatisfactory conditions to
determine the cause and makes recommendations
for correcting the situation.
DATE IMPLEMENTATION
PAGE l OF 2
DATE ISSUED:
233

-------
                                   QMP NO.  2.4
                                   Page  2 of  2
RESPONSIBILITY
Audit
                FUNCTIQN

Conducts independent random checks of data,
.personnel,  equipment and test cell log books  to
assure that proper procedures are being followed,
calibration and maintenance intervals are being
observed, and to judge for  effectiveness of
training programs.

Conducts intralaboratory and inter laboratory
correlation of emission measurement equipment to
improve the accuracy and reliability of the test
data.

-------
EPA QUALITY MANAGEMENT  PROCEDURE
                        QMP NO.
                             2.5
                                                                   REVISION  DATE
SUBJECT:
        FUNCTIONAL OUTLINE - DATA SERVICES
                                    SUMMARY

   Data Services is  responsible for the development of  computer programs for data
   reduction.   Processes, monitors and validates test related data to ensure the
   accuracy and reliability of the emission measurements.   Maintains data files of
   test results and  provides statistical programs to assist Quality Assurance in the
   monitoring  of test data accuracy.
   RESPONSIBILITY
   Data Validation
   Statistical Analysis
    Systems
    Development
                   FUNCTION

1.    Performs data validation according to formalized
     procedures and informs Test Operations and
     Quality Assurance of invalid tests.  Notifies
     Production Control to reschedule vehicle.  Initiates
     corrective action and failure reports when necessary
     to reduce the number of invalid tests.

2.    Maintains all test data in a data file.

3.    Assists Quality Assurance in monitoring all data to
     verify the accuracy and reliability of emission
     measurements.

4.    Provides statistical analysis for Quality Assurance
     requirements such as determination of acceptable
     test parameter limits, preparation of control
     charts, reduction of correlation data and cost
     analysis.

5.    Assists Quality Assurance and Laboratory Operations
     in providing for computer programs with mathema-
     tically correct formulas for the reduction of data
     for non-routine test programs, revision of emission
     data programs, and other computer programming
     requirements.

6.    Assists Quality Assurance in developing and imple-
     menting correlation and audit programs to assure
     the reliability of the data on a "cell to cell"
     basis and/or other laboratories performing mobile
     source emission testing.
               CONCURRENCES
                      DATE
IMPLEMENTATION
 PREPARED  BY:
                               PAGE  1   OF
 APPROVED  BY:
                               DATE  ISSUED:
                                       "235"

-------
                                                                QMP No.  2.5
                                                                Page  2 of  2
RESPONSIBILITY                               FUNCTION

Computer Operations      7.    Assists in the development of computer programs,
                              processes computer programs  for the reduction of
                              test data to provide emission results on a gram
                              per mile basis for carbon monoxide, hydrocarbons,
                              carbon dioxide and nitric oxide.  Provides results
                              for fuel economy on a mile per gallon basis.

                         8.    Processes computer programs  for calibration data,
                              maintains calibration data file, and computes
                              instrument calibration  curves.  Informs Quality
                              Assurance and Test Operations when calibration
                              and maintenance has not been performed according
                              to prescribed intervals.

                         9.    Maintains the calibration gas cylinder inventory
                              by number, type of standard  and receiving analysis
                              concentration.  Maintains and processes all data
                              related to the primary  gas standards such as the
                              NBS-SRM gases and/or those analyzed by the EPA.
                                    236

-------

-------
QUALITY MANAGEMENT PROCEDURES
         SECTION 3.0
       ADMINISTRATION
              237

-------
EPA  QUALITY  MANAGEMENT PROCEDURE
                              QMP NO.
                                  3.1
                                                                    REVISION  DATE
SUBJECT:
        PREPARATION OF QUALITY MANAGEMENT PROCEDURES
   I    PURPOSE

        This procedure defines  the  formal documentation of Quality Management
        Procedures (QMP).

   II   BACKGROUND

        A.   The Quality Management Procedures are written to reflect the
             Organization's policy  concerning the administrative/functional aspects
             of a Quality  Assurance Program and the interrelationship of these
             functions/responsibilities.

        B.   The Quality Management Procedures provide the instructions required to
             implement a Quality  Assurance Program.  They define the purpose, back-
             ground and scope of  application of the procedure and, in addition,  show
             the assignment of  functional responsibility for performing the procedure.

   Ill  SCOPE OF APPLICATION

        A.   QMPs are generated by  Quality Assurance in order to document the
             procedures and the assignment of responsibilities of all quality
             related functions  within  the mobile source emission measurement system.

        B.   The Quality Management Procedures are prepared by Quality Assurance
             and distributed by Document Control.  Basically these procedures are
             divided into:
             1.
Changes and Revisions  - QMP Form No. 7.6 on which all
distributed revisions  to  the manual are recorded and
inserted in the manual by the manager/supervisor.

Introduction - contains a description of the purpose and
objectives of the manual  and the general philosophy of
its preparation along  with the organizational policy for
its use.
             2.   Section One - Index -  Lists Table of Contents.

             3.   Section Two - Organization -  contains the function/responsibility
                  chart and the function outlines.  This chart is designed to show
                  the required functions and responsibilities of a Quality Assurance
                  Program but not necessarily their interrelationship which can
                  only be done for a specific organizational structure.  The func-
                  tions of each major department are outlined using the "play script"
                  format.
               CONCURRENCES
                           DATE
IMPLEMENTATION
PREPARED  BY
                                    PAGE   l OF  5
APPROVED  BY:
                                    DATE  ISSUED:

-------
                                                                QMP  No.  3.1
                                                                Page 2 of  5
III  SCOPE OF APPLICATION (Continued)

          4.    Section Three - Administration - contains procedures  generally
               related to the quality functions and responsibilities of  admin-
               istrative services.

          5.    Section Four - Procurement Control - contains procedures  directly
               related to the quality function and responsibilities  of purchasing
               and receiving of  equipment and materials  used in  the  laboratory.

          6.    Section Five - Standards and Calibrations - describes quality
               procedures and functions applicable to  equipment  service  and
               metrology.

          7.    Section Six - Laboratory Operations - details the quality proce-
               dures and responsibilities related to the operation of the mobile
               source emission testing laboratory.

          8.    Section Seven - Forms Instruction - describes the procedure for
               completion of forms  required by the QMPs.  Forms  will be  numbered
               as follows:
               7.1:  1-14-75
               I
                                  ,Section of  the manual containing form instruc-
                                   tions

                                  ,Sequential  number assigned to each form when
                                   first  issued

                                  .Effectivity/Revision date
    C.   The decimal  system  is used  for  numbering each procedure in the manual
          according  to the section  in which it appears as follows:

          QMP-JC.X: x

                   		Section  of  the manual  (1-7)
—A •
                                  -Sequential QMP Procedure number within
                                  a section.   (1-9)

                                  .Reserved.  Used only if more than 9 procedures
                                  are to be included in a section.  (1-9)
                                   2kO

-------
                                                                 QMP No. 3.1
                                                                 Page 3 of 5
III  SCOPE OF APPLICATION (Continued)

     D.   The format to be followed for each procedure is described as follows:

          1.    Section Two - contains  the function/responsibility chart and the
               functional outlines. Functional outlines are prepared in "play
               script" format,  i.e., the group or department responsibility
               for the outlined function is indicated in the left margin.

          2.    Sections Three through  Seven - Quality Management Procedures
               (QMP)  - follow the format:

               I     PURPOSE - briefly  describes the purpose or objective of
                    the Procedure.

               II   BACKGROUND - generally describes the reason or need for
                    the procedure in addition to any pertinent historical
                    information.

               Ill  SCOPE OF APPLICATION - defines the areas of the measurement
                    systems affected or involved in the particular procedure and
                    specific effectivity such as a particular emission program
                    or period of time  are included.

               IV   RESPONSIBILITIES AND PROCEDURES - describes the duties in
                    detail for every function involved in the procedure, by
                    order of importance and sequentially, if possible.  See
                    sample below for numbering system.

                    IV   RESPONSIBILITIES AND PROCEDURES
                         A.   Quality Assurance

                              1.
                              2.
                              3.
                                   a.
                                   b.
                                   c.

                    In addition to the described duties, this section will
                    usually contain a flow schematic showing the interrelation-
                    ship of functions and responsibilities and/or the documenta-
                    tion distribution.
                                      241

-------
                                                                 QMP No. 3.1
                                                                 Page 4 of 5
IV   RESPONSIBILITIES AND PROCEDURES

     A.    Quality Assurance

          1.   Prepares detailed procedures in rough draft, assigns numbers
               to new procedures and/or revisions and designates distribution
               list prior to routing to Document Control.

          2.   Coordinates any variance between draft QMP and actual practices
               reported by a manager and/or supervisor and sends revised
               draft QMP to Document Control for final draft preparation and
               distribution.

          3.   Maintains a master file of active and historical procedures
               and associated documents issued.

     B.    Document Control

          1.   Distributes draft copies of procedures to management and
               supervisors for review and comment.

          2.   Distributes approved copies of procedures and/or manuals to
               management and supervisors requiring copies for frequent use
               in performance of their normal duties.

          3.   Maintains records of location of each manual or procedures and
               the person responsible for their update.

     C.    Department Manager/Supervisor

          1.   Reviews and comments on draft copies of procedures.

          2.   Maintains a manual in his area and becomes familiar  with the
               contents of all procedures with responsibilities related to
               his particular function.

          3.   Records all new or revised QMPs inserted in manuals  on QMP
               Form No. 7.6 which appears as the first page of the  manual.

          4.   Observes and utilizes applicable procedures and responsibilities
               assigned to his function by the Quality Management Procedure
               (QMPs).

          5.   Initiates an Equipment and Procedures Change Notice  (QMP
               Form No. 7.5)  to inform Quality Assurance of any variances
               between QMPs and any applicable engineering documents and/or
               any observed errors in contents.

-------
                                                                QMP No. 3.1
                                                                Page 5 of 5
IV   RESPONSIBILITIES  AND PROCEDURES  (Continued)
     E.   Flow Schematic - QMP
          Quality Assurance
           Originates All
               QMPs
          Reviews Comments
           QA
     Prepares Draft
     Assigns No.  &
      Distribution
Document Control
Finalizes Draft
and Distributes
   For Review
                             Yes
           Document Control
            Prepares Final
          QMP and Distributes
             to All Manual
               Holders
No
Manager/Supervisor

      Reviews
       Draft
Manager/Supervisor
Inserts QMP and
Logs Required
Revision on
Form 7 . 6
f

QA
Audits For
Revision and
Conformance
                                     243

-------
EPA QUALITY  MANAGEMENT PROCEDURE
   QMP NO.
        3.2
                                                                    REVISION  DATE
SUBJECT:
             DOCUMENT CONTROL
   I    PURPOSE

        This procedure provides  a method for issuing, revising,  and controlling the
        documentation of manuals, forms and other records used in a mobile source
        emission measurement system.

   II   BACKGROUND

        A.   The responsibilities and procedure for preparing, numbering,  imple-
             menting, and revising of forms and procedures used in the measurement
             systems must be clearly defined since timely response to the  changing
             requirements of the system is of utmost importance.  The maintenance of
             forms and manuals in a current status requires prompt submission and
             processing of change notices and resulting revisions, and effective
             control of publication and distribution of documentation to prevent
             obsolescence.

        B.   A master file of all procedures, forms, and subsequent revisions showing
             effective dates should be maintained for future reference.

   Ill  SCOPE OF APPLICATION

        A.   Manual Control - Any manual produced by a department or function
             within the test facility shall be submitted to Document Control, in
             draft form, for identification, completion, filing and distribution.
             Manuals specifically covered by this procedure are:

             1.   Quality Management Procedures
             2.   Training
             3.   Test Procedures
             4.   Maintenance
             5.   Administrative or Management Policies

             All subsequent authorized revisions of the contents of these  manuals
             shall be submitted  to Document Control for distribution to.the
             manual holders.

        B.   Equipment and Procedure Change Notices  (EPCN) - All EPCN's shall be
             submitted to Document Control for assignment of a  file number and
             distribution.

        C.   Forms, blueprints,  equipment specifications and schematics used in the
             measurement system  shall be submitted to Document Control for assign-
             ment of a document  identification number and when necessary prepara-
             tion and distribution of a form instruction.
               CONCURRENCES
DATE
     IMPLEMENTATION
PREPARED  BY:
PAGE
                   OF
APPROVED  BY:
          DATE  ISSUED:
                                         245

-------
                                                                 QMP No. 3.2
                                                                 Page 2 of 3
IV   RESPONSIBILITIES AND PROCEDURES

     A.   Quality Assurance

          li   Originates and revises the contents of the Quality Management
               Procedures Manual as required.

          2.   Audits manuals, EPCN's and forms being used by the Laboratory
               on a regular schedule for proper identification and format,
               current revisions, and adequate maintenance of document files.

          3.   Reports results of audits to Manager/Supervisor, coordinates
               and monitors corrective action when necessary.

          4.   Approves all new or proposed revisions of forms prior to
               publication.

          5.   Approves format and distribution list of all laboratory note-
               books/log books.

     B.    Document Controj^

          1.   Coordinates and distributes "Review and Comment" draft copies,
               obtains approval and release of final draft documents.

          2.   Issues, subject to department management approval,  manuals,
               copies of procedures and forms to employees requiring copies
               for frequent use in performance of their normal duties.

          3.   Maintains control of procedure manual masters and manual
               distribution list.

          4.   Assigns form reference numbers to all forms.

          5.   Maintains master file of all forms.

          6.   Distributes approved copies of records forms to management,
               other  agencies  and  testing laboratories requiring copies for
               use in performance  of their normal duties.

          7.   Distributes laboratory notebooks/logbooks by sequential
               number for use  in the laboratory, and provides instructions on
               the format to be followed in making entries.   Maintains  a file
               recording the name  of the person (s)  responsible for the  note-
               book/logbook, location and other applicable information.
                                     246

-------
                                                                 QMP No.  3.2
                                                                 Page 3 of 3
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     C.   Manual Holders

          1.   Submit requests for forms,  manuals,  EPCN's and log books to
               Document Control.   Submit proposed procedures, forms, etc. ,
               in draft form to Quality Assurance and Document Control with
               appropriate "Review and Comment" distribution list.

          2.   Maintain manual in current status in accordance with distri-
               buted change notice.

          3.   Submit any change request on an EPCN (QMP Form 7.5) to
               Laboratory Operations.

          4.   Returns manual to Document Control when no longer required, or
               when terminating employment.

-------
ERA QUALITY MANAGEMENT  PROCEDURE
                                            QMP NO.
                                                 3.3
                                                                   REVISION DATE
SUBJECT:
        FORMAL TRAINING PROGRAMS
   I    PURPOSE

        This QMP establishes guidelines for preparation and presentation of formal
        Quality Assurance  training programs.

   II   BACKGROUND

        Quality Assurance  will provide formal training  on relevant Quality Assurance
        topics to personnel from Quality Assurance and  interfacing organizations
        such as Engineering and Test Laboratories personnel.  Training programs will
        be scheduled  and certifications issued upon completion.

   Ill  SCOPE OF APPLICATION

        Quality Assurance  training programs should reinforce  the recognition of the
        importance of quality in each individual's efforts in addition to specific
        job or subject matter training.  The programs should  be directed not only
        to Quality Assurance personnel, but alsd to personnel in interfacing
        organizations.

   IV   RESPONSIBILITIES AND PROCEDURES

        A.   Quality  Assurance Manager/Supervisor

             1.   Selects  a subject of interest for a training program, to consist
                  of  a training session or series of sessions.  Examples of such
                  subjects are statistical quality control, configuration control,
                  sampling plans, special process requirements, Federal Register
                  requirements, etc.  Discusses subject matter and training
                  requirements with training coordinator.

             2.   Maintains records of certificate holders when periodic recerti-
                  fication is required.

             3.   Notifies affected supervisor prior to expiration dates for
                  certificate holders under his supervision,  and arranges with
                  test-coordinator to conduct examinations to verify continued
                  proficiency of individuals requiring  recertification.

             Training Coordinator

             1.   Determines total time required for training session and personnel
                  to  attend.
B.
               CONCURRENCES
                                         DATE
IMPLEMENTATION
 PREPARED  BY:
                                                   PAGE   1  OF
APPROVED  BY:
                                                   DATE  ISSUED:

-------
                                                                QMP No.  3.3
                                                                Page  2 of  2
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     B.   Training Coordinator (Continued)

          2.   Assigns personnel to prepare lesson plan and give presentations.

          3.   Obtain concurrence from management of  other organizations for
               attendance of their personnel.

          4.   Establishes date, time and location schedule.

          5.   Notifies attendees and supervision of  the topic and training
               program schedule 2 weeks in advance.

          6.   Follows up with scheduled attendees a  week prior to training
               session to verify their availability for time established.
               Notifies supervisor in case of  conflict.

          7.   Issues a certificate to attendees  on successful completion of
               program.

          8.   Arranges recertification training  and  examination as required.

     C.   Personnel Assigned to Give Presentation

          1.   Prepares lesson plan and presentation  to cover the subject in
               the time allotted.

          2.   Prepares a summary of the training session to be issued to
               attendees as  an outline.

          3.   Reviews lesson plan presentation and summary with Quality
               Assurance Manager/Supervisor.

          4.   Presents training material to the  attendees at the scheduled
               sessions.
                                     250

-------

-------
QUALITY MANAGEMENT PROCEDURES
         SECTION 4.0
     PROCUREMENT CONTROL
              251

-------
EPA QUALITY  MANAGEMENT PROCEDURE
    QMP NO.
        4.1
                                                                    REVISION  DATE
SUBJECT:
        PROCUREMENT DOCUMENT REVIEW
    I    PURPOSE

        This procedure establishes the  requirements for review of procurement
        documentation to ensure the inclusion of quality and reliability provisions.

    II   BACKGROUND

        Procurement documents are used  for  the purchase of materials,  supplies  and
        services used in implementing mobile source emission testing.   To  maintain
        a high level of quality throughout  the program it is essential that these
        documents be reviewed for the inclusion of necessary quality and reliability
        requirements.

    Ill  SCOPE OF APPLICATION

        A.   All procurement documents  for  material, equipment or services will be
             subject to review and approval by Quality Assurance prior to  release
             and placement.

        B.   Certain items, procured on a routine basis and as determined  by mutual
             agreement between procurement  and QA, may be purchased without QA
             review and approval.

        C.   Programs will be reviewed  to determine the scope of the quality and
             reliability requirements applicable to the contracts and the  associated
             procurement activities.

    IV   RESPONSIBILITIES AND PROCEDURES

        A.   Purchasing

             1.   Routes all procurement documents for the purchase of material,
                  supplies and services used in implementing emission testing to
                  Quality Assurance for review and application of quality  and
                  reliability provisions.

        B.   Quality Assurance

             1.   Reviews each procurement  document to determine applicable quality
                  requirements and coordinates with other divisions/departments as
                  necessary to assure consideration of all quality and reliability
                  interests.

             2.   Applies quality and reliability requirements to procurement
                  documents.
               CONCURRENCES
DATE
     IMPLEMENTATION
PREPARED  BY:
PAGE
                   OF
APPROVED BY:
         DATE  ISSUED:
                                        253

-------
                                                                 QMP  No.  4.1
                                                                 Page 2 of  2
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     B.   Quality Assurance (Continued)

          3.    Enters evidence of Quality Assurance approval on procurement
               documents and returns them to purchasing.

          4.    Maintains records of procurement document  review activities.

     C.   Procurement Requirements

          The following quality requirements are generally applicable  to  all
          procurement actions.

          1.    Approved Suppliers - Quality Assurance  approval  of procurement
               sources required.

          2.    Source Inspection - Source inspection shall be required when
               (a) the necessary inspection and test equipment  or required
               environment is not available at the  test facility,  (b)  articles
               being procured are at a  level of assembly  which  precludes  veri-
               fication of quality upon receipt or  (c) in-process controls
               have such an effect on the quality of the  article that  the
               quality cannot be determined by inspection or tests of  the
               completed articles.

          3.    Physical/Chemical Test Reports  - All procured raw materials
               shall be accompanied by  physical/chemical  test reports  which
               establish conformance to the applicable specification requirements,

          4.    Age Control - Articles for which acceptability is limited  by
               maximum age shall be clearly identified with a manufacture date
               and expiration date.

          5.    Packaging and Shipping Instructions  - Special packaging, preserva-
               tion or shipping instructions that may  be  applicable.   Special
               attention to this item is required when drop shipments  or  hazar-
               dous materials are involved.

          6.    Inspection and Test Data - Requirements for submission  of
               inspection and/or test records  with  procured articles.

          7.    Certificates of Compliance - Supplier certifications of con-
               formance with specification requirements.

          8.    Serialization/Identification -  Requirements for  serialization
               and identification of materials or equipment.

     D.   Procurement Flow Schematic

          See Procedure 4.2

-------
EPA QUALITY  MANAGEMENT PROCEDURE
   QMP NO.
        4.2
                                                                    REVISION DATE
SUBJECT:
        RECEIVING INSPECTION
   I    PURPOSE

        This QMP describes the methods used for the inspection and test of all
        procured material, parts and equipment (hereinafter referred to as
        "material")  upon receipt from the supplier.

   II   BACKGROUND

        Purchased material used in the test facility should be subjected to inspection
        when first received from the supplier to assure that  it meets purchase order
        specifications  and that non-acceptable material is precluded from use in the
        measurement system.

   Ill  SCOPE OF APPLICATION

        A.   All procured materials which influence or are intended for use in mobile
             source emission testing shall be inspected and tested as necessary
             to verify  their conformity to purchase order specifications and any
             program requirements.

        B.   Certain material such as calibration gas mixtures, and analytical
             instruments require special receiving inspection procedures which are
             prepared and issued by Quality Assurance.

   IV   RESPONSIBILITIES AND PROCEDURES

        A.   Receiving  (Material)

             1.   Checks shipment for count and completeness, prepares Receiving
                  Report and collects all pertinent documentation.

             2.   Moves all materials and paperwork to Receiving Inspection.

        B.   Receiving  Inspection

             1.   Inspects incoming materials in accordance with established
                  priorities so that an effective flow of material is assured.

             2.   Checks the purchase order for special requirements and assures
                  that  all of the purchase stipulations have  been complied with.

                  a.   If source inspection is a requirement, verifies that parts
                       and documentation are properly identified and accepted by
                       source inspector.
               CONCURRENCES
DATE
IMPLEMENTATION
PREPARED BY:
         PAGEi    OF 3
APPROVED BY:
         DATE  ISSUED:
                                        255

-------
                                                                 QMP No. 4.2
                                                                 Page 2 of 3
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     B.   Receiving Inspection (Continued)

          2. (Continued)

               b.   If test data, laboratory reports or certifications of
                    compliance are required, verifies that the appropriate
                    documents have been submitted and that they provide
                    satisfactory evidence of conformance to specification
                    requirements.

               c.   Raw materials will be accompanied by test reports and/or
                    physical and chemical analysis reports which will be
                    checked against the applicable material specifications for
                    verification of material quality.  Such data must be
                    positively identified to correlate with the raw material
                    submitted.

               d.   Materials that are subject to quality degradation with age
                     (limited life items) shall be identified with a tag or
                    stamp indicating the manufacturing date and the expiration
                    date for issue or use.

          3.   Performs inspection and test operations to verify conformity with
               applicable specification and purchase order requirements.

               a.   Material that has successfully met all applicable receiving
                    inspection criteria shall be identified by applying evidence
                    of acceptance to all material and paperwork.

               b.   Material that fails to meet any portion of the applicable
                    receiving inspection criteria shall be rejected on a
                    Rejection Report and forwarded to Quality Assurance for
                    disposition.

     C.   Quality Assurance

          1.   Reviews the Rejection Report and coordinates with other
               organizations as required to establish final disposition of
               the rejected material.

          2.   Advises Purchasing of the disposition of the rejected material.

          3.   Conducts an analysis of any discrepancies/failures that may occur
               on a first order basis or when indicated by inspection reports.
                                      256

-------
                                                                QMP No. 4.2
                                                                Page 3 of 3
IV   RESPONSIBILITIES AND PROCEDURES  (Continued)

     D.    Procurement Control Flow Schematic
                                  Requestor.
                                  Initiates
                                  Purchase
                                  Request
Quality
Assurance (QA).
Reviews PD
Supplier Audit
Purchasing.
Prepares
Procurement
Document (PD]
                                      I
->
                                                           Receiving .
                                                           Copy of
                                                           Approved PD
Vendor .
Ships
Material


s
J
Purchasing .
Notifies
Vendor of
Defect


                                  Receiving.
                                  Count S
                                  Documents
              QA
          Inspection
          Procedure
                         Receiving
                         Inspection
             QA
          Rejection
          Report
                                       Yes
                         Requestor.
                         Inspection
                         Report &
                         Material
             QA
         Material
         Review
                            QA
                        Coordinates
                        Corrective
                        Action
                                    257

-------

-------
QUALITY MANAGEMENT PROCEDURES
         SECTION 5.0
   STANDARDS S, CALIBRATION
             259

-------
EPA QUALITY  MANAGEMENT PROCEDURE
                         QMP NO.
                             5.1
                                                                    REVISION DATE
SUBJECT:
            EQUIPMENT CALIBRATION AND  CYCLE CONTROL
   I    PURPOSE

        This QMP establishes a system that provides for the assignment, account-
        ability and the initial and periodic  calibration of all instruments and
        equipment involved in the performance of mobile source emission testing.

   II   BACKGROUND

        A.   The accuracy and adequacy of all equipment used to measure, test or
             inspect physical or technical aspects of the vehicle or emissions are
             assured by initial and periodic  inspection and calibration of this
             equipment.

        B.   The establishment of equipment controls for calibration purposes re-
             quires a knowledge of equipment  status, usage, location, and the
             identification of personnel responsible for the equipment.

        C.   Temporary borrowers of equipment must be indoctrinated regarding their
             responsibilities for the equipment, specially with regard to cali-
             bration status and return of  equipment after use.

   Ill  SCOPE OF APPLICATION
        A.   The equipment items requiring initial  and periodic calibration for
             light duty testing are listed below.
        Function

        Receiving &
        Inspection
        Vehicle
        Preparation
        Vehicle
        Test
          Equipment

1.   Tach, dwell,  RPM Equipment

2.   Idle exhaust  CO/HC meters

3.   Platform scale  for vehicle weight

4.   Thermocouples

5.   Temperature Recorders

6.   Driver's Aid

7.   Constant Volume Sampler

     a.    Positive displacement pump

     b.    Temperature probe & controller
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE


IMPLEMENTATION
PAGE 1 OF 8
DATE ISSUED:

-------
Function

Vehicle
Test
                                                            QMP No. 5.1
                                                            Page 2 of 8
III  SCOPE OF APPLICATION (Continued)

                                   Equipment

                              c.   Pressure measuring device

                              d.   Counters - Pump RPM

                         8.   Analytical System

                              a.   Hydrocarbon Analyzer

                              b.   Carbon Monoxide Analyzer

                              c.   Carbon Dioxide Analyzer

                              d.   Nitric Oxide Analyzer

                              e.   Recorders and/or Digital Voltmeter

                              f.   Gas Mixtures

                         9.   Dynamometer

                        10.   Barometer

                        11.   Hygrometer or Psychrometer

          Other auxiliary equipment used by a particular laboratory may be sub-
          jected to calibration at the discretion of Quality Assurance.

     B.   All equipment in the standards and calibration accountability control
          system shall be subject to this procedure.

     C.   Measurement standards used for calibration purposes shall be traceable
          to the National Bureau of Standards (NBS) when possible.

     D.   Calibration gas mixtures used as primary standards shall be traceable
          to the EPA gravimetric standards and/or the NBS Standard Reference
          Material.

     E.   Each organization using the above listed equipment shall be responsi-
          ble for assuring that instruments are not used beyond the "calibration
          due" date and for notifying Quality Assurance when inaccuracies or
          malfunctions occur.
                                 262

-------
                                                                 QMP No. 5.1
                                                                 Page 3 of 8
III  SCOPE OF APPLICATION (Continued)

     F.    Calibration control of the instruments in the standards and cali-
          bration accountability control system is accomplished through the
          exclusive use of three documents.

          Form No. QMP 7.1 "Loan Order".  Identifies the location and person
          responsible for the equipment.  This is not used for "surplus" equip-
          ment only but has the primary objective of showing the location of
          all equipment in the system.

          Form No. QMP 7.2 "Calibration Control Card".  A keypunch card is used
          by computer operations to identify when a calibration becomes due.
          The information on this card  is filled in by Records Control each time
          it is informed of any change  in equipment status.

          Form No. QMP 7.3 "Calibration Order".  A multi-copy form issued by
          Computer Operations one week  in advance of a calibration due date.

          Instructions for filling out  these forms appears in Section 7.0 of the
          QMP Manual.

IV   RESPONSIBILITIES AND PROCEDURES

     A.    Equipment Services

          a.   Performs inspection and  calibration of new equipment upon receipt
               to determine conformance with applicable requirements.

          b.   Assures that each piece  of equipment is identified with a con-
               trol number for accountability and periodic calibration control.

          c.   Affixes a distinctive label or tag to each piece of equipment
               reflecting date of last  calibration, by whom it was calibrated
               and date when it is due  for recalibration.

          d.   Initiates a record in the instrument maintenance log book noting
               the accomplishment and results of each calibration performed.

          e.   Transmits to Records Control a Calibration Control Card  (Form No.
               QMP 7.2) containing equipment control number, description, loca-
               tion and recalibration date information for accountability and
               calibration control records.
                                     263

-------
                                                                 QMP No.  5.1
                                                                 Page 4 of 8
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     B.   Equipment Services (Equipment Stores)

          1.   Stores equipment carefully to prevent damage, corrosion or
               contamination.

          2.   Issues equipment only upon receipt of a properly completed
               Instrument Loan Order (QMP Form No. 7.1)  and distributes
               copies as follows:

               a.   Copy number 3 to assignee

               b.   Copy number 2 to file

               c    Copy number 1 to records control

          3.   Files copy number 1 in control number order when returned by
               records control.

          4.   When equipment is returned as no longer needed by the assignee,
               stamps #1 and #2 copies as "received", gives #1 copy to assignee
               and forwards the #2 copy to Standards and Calibration (Records
               Control).

          5.   Maintains available inventory file by control number and status.

     C.   Standards and Calibration (Record Control)

          1.   Submits initial Calibration Control Card to Computer Operations.

          2.   Upon receipt of Instrument Loan Order copy noting loan or
               return of equipment, enters location changes on a new Calibration
               Control Card for equipment affected and forwards to Computer
               Operations.

          3.   Destroys the #2 copy after processing location change.

     D.   Computer Operations

          1.   Sorts calibration control data file weekly for items due for
               recalibration the following week.

          2.   Prints calibration orders (QMP Form No. 7.3) for recall of
               items for recalibration and  delivers to  Standards  and Cali-
               bration  (Records Control).
                                     26k

-------
                                                                 QMP No.  5.1
                                                                 Page 5 of 8
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     E.    Standards and Calibration (Records Control)

          1.   Remove the follow-up copy of the calibration order(s)  and for-
               ward to Standards and Calibration (Equipment Stores).

          2.   Forward the remaining copies (3)  of the calibration orders(s)
               to the using organization(s).

     F.    Using Organization

          1.   Completes signature, extension (telephone)  and indicates in the
               appropriate "yes-no" blocks on the calibration order whether or
               not the equipment is to be returned to the user after calibration.

          2.   Removes and retains the receipt copy of the calibration order.

          3.   Attaches the remaining two copies of the calibration order to the
               equipment and returns equipment to Standards and Calibration
               (Equipment Stores).

     G.    Standards and Calibration (Equipment Stores)

          1.   Files follow-up copies of the calibration order according to due
               date.  Notify Quality Assurance when calibration is past due.

          2.   Removes the follow-up copy of calibration order from file when
               equipment with traveler and record copies of calibration order
               is received for scheduled recalibration.

          3.   If a replacement item is furnished from Equipment Stores note
               this information in the "Remarks" block on the traveler and
               record copies of the calibration order.

          4.   If "no" return block is checked or replacement item furnished,
               process instrument loan order as in IV B.

     H.   Equipment Services (Calibration and Maintenance Technician)

          1.   Calibrate equipment per applicable instructions and affix cali-
               bration label or tag with stamp and date entries  to the equipment.

          2.   Record results of calibration on the traveler and record copies
               of the calibration order.

          3.   Discard follow-up copy of calibration  order, forward  record  copy
               to Standards and Calibration (Records  Control) and forward
               traveler copy with equipment to Equipment stores.


                                      265

-------
                                                                 QMP No. 5.1
                                                                 Page 6 of 8
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     I.   Standards and Calibration (Records Control)

          1.   Notes the changes on a calibration control card from information
               on the record copy of the calibration order.

          2.   Forwards the calibration control card with changes to Computer
               Operations and files record copy of the calibration order in
               the equipment history file.

               NOTE:  Refer calibration orders indicating "out of tolerance"
               information to  assigned personnel for evaluation prior to filing.
               Assigned personnel review the history file of instruments found
               to be "out of tolerance" to identify critical and chronic con-
               ditions peculiar to the instrument or common to all instruments of
               that type.

               a.   Corrective action must be taken to prevent recurrence of "out
                    of tolerance" conditions; i.e., reduce the calibration inter-
                    val for the item(s) affected, revise the calibration checklist
                    affected to improve the method of calibration, utilize more
                    accurate standards and/or include preventive maintenance
                    requirements, dispose of the items affected, etc.

               b.   Record completion of review and evaluation by entries in
                    applicable blocks of the calibration order.

     J.   Computer Operations

          1.   Keypunch and print new calibration control card incorporating
               changes.

          2.   Return old and new calibration control cards to Standards and
               Calibration (Records Control).

     K.   Standards and Calibration (Equipment Stores)

          1.   Return calibrated equipment with the traveler copy of the cali-
               bration order to the user.

               NOTE:  If the calibration order indicates that a return instru-
               ment was not required or a replacement was furnished, place the
               equipment in Stores.  Forwards the traveler copy of the calibra-
               tion order to the user in those instances where the equipment
               affected was found to be out of tolerance.
                                      266

-------
                                                                 QMP No. 5.1
                                                                 Page 7 of 8
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     L.    Quality Assurance

          1.   Conducts follow-up of equipment not returned for calibration by
               due date, places an "out of service" tag on equipment, and main-
               tains follow-up to assure that equipment is not used again until
               re-calibrated.

          2.   Maintains surveillance on an audit basis of the proper calibration
               status of test and measuring equipment.

     M.    Equipment Services

          1.   Repair

               a.   Performs repair/recalibration of equipment submitted due
                    to failure or damage in use.

               b.   Initiates and files record of repair/recalibration accomplished.

               c.   Transmits to Data Processing  the new due date for recalibration.

               d.   Returns the repaired/recalibrated equipment to the submitting
                    organization, if return was requested.
                                    267

-------
                                                                        QMP No.  5.1
                                                                        Page 8 of 8
IV   RESPONSIBILITIES AND PROCEDURES  (Continued)
     N.    Flow Schematic -  Equipment Calibration Control
                 Records Control
                         (RC>
                 Master File by
                 Control No.
                Calibration
                Order
                     Record

-------
EPA QUALITY MANAGEMENT  PROCEDURE
   QMP NO.
       5.2
                                                                    REVISION DATE
SUBJECT:
             CALIBRATION  INSTRUCTION DOCUMENT MAINTENANCE
   I    PURPOSE

        This QMP specifies  the procedure for the acquisition,  use  and maintenance of
        calibration instruction documents.

   II   BACKGROUND

        A.   Documentation  of the calibration procedures used  by standards and cali-
             bration is necessary to assure that the correct calibration procedure
             is used for a  particular instrument, and to provide a reference  source
             when the calibration of an instrument is questioned.

        B.   Uncontrolled calibration documents have a tendency to disappear  or be
             unavailable when needed.  It is therefore, of utmost  importance  that
             these documents be kept in a central controlled file.

   Ill  SCOPE OF APPLICATION

        A.   All documents, i.e., manufacturer manuals, calibration checklists, pro-
             cedures providing methodology for calibrating or  repairing specific types
             of equipment shall be subject to this procedure.

   IV   RESPONSIBILITIES AND PROCEDURES

        A.   Standards and  Calibration

             1.   Provides  adequate information for calibration and repair of equip-
                  ment submitted  for initial calibration.

             2.   Obtain manuals  or procedures needed from reliable source(s); i.e.,
                  manufacturer, government agency, professional society, etc.

             3.   Initiates Calibration Checklists/Reports defining specific  scope
                  and method of calibration and maintenance when practical or
                  necessary.

             4.   Establishes and maintains files of all documents.

             5.   Checks out and  utilizes applicable documents to conduct  calibration
                  or repair of equipment.

             6.   Returns documents to file immediately upon completing calibration
                  or repair of equipment.
               CONCURRENCES
DATE
IMPLEMENTATION
 PREPARED  BY:
          PAGE  1  OF  3
 APPROVED  BY:
          DATE  ISSUED:
                                        269

-------
                                                                 QMP No.  5.2
                                                                 Page 2 of 3
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     B.   Equipment Services

          1.    Forwards all information bulletins, calibration and maintenance
               manuals etc., to standards and calibration for review and filing
               upon receipt of a "new-order" instrument.

     C.   Quality Assurance

          1.    Assists standards and calibration in review and preparation of
               calibration procedures, forms, etc.

          2.    Maintains surveillance on an audit basis to assure correct use
               of calibration documents and that proper control procedures are
               being maintained.
                                      270

-------
                                                                QMP No. 5.2
                                                                Page 3 of 3
IV   RESPONSIBILITIES  AND PROCEDURES

     D.    Flow Schematic - Calibration Instruction Document Maintenance
          Equipment Services

          Submits all pertinent informa-
          tion received from the manufacturer
                                  Calibration Document
                                  File

                                  Maintained by Standards
                                  and Calibration
          Standards & Calibration/Quality Assurance

          Reviews and prepares/obtains  calibration
          instructions when instrument  is first
          submitted for calibration
          Instrument
          Submitted
          For
          Calibration
                                   Calibration Technician
Checks out required documents
for instrument being cali-
brated, and returns to file
after use
                                     271

-------
EPA QUALITY MANAGEMENT  PROCEDURE
   QMP NO.
        5.3
                                                                    REVISION DATE
SUBJECT:
             CALIBRATION  INTERVALS
   I    PURPOSE

        This QMP describes  the procedure for establishing realistic  calibration
        intervals to maintain prescribed accuracy of all measuring and  test
        equipment.

   II   BACKGROUND

        A frequency distribution chart for determining calibration intervals  is
        presented in Attachment No. 1, which may be used to adjust calibration
        intervals of a specific model after the evaluation of a minimum of one year's
        calibration results or twenty (20) calibration results, whichever occurs
        first.

   Ill  SCOPE OF APPLICATION

        All measuring and test equipment used in conjunction with emission testing
        operations shall  be subjected to the requirements of this procedure.

   IV   RESPONSIBILITIES  AND PROCEDURES

        A.   Standards and  Calibration

             1.   Establishes the calibration interval for each piece of equipment
                  (based  upon its stability, reliability, usage and  calibration
                  history of identical or similar equipment), at the time it  is sub-
                  mitted  for initial calibration.

             2.   Enters  the calibration interval in the applicable  block of  the
                  Calibration Order  (QMP Form No. 7.3) before forwarding with the
                  equipment to Calibration for initial acceptance.

             3.   Transmits the calibration interval information for each piece of
                  equipment to Computer Operations using the Calibration Control
                  Card (QMP Form No. 7.2).

             4.   Periodically  (minimum each 12 months) evaluates the calibration
                  history of equipment, by manufacturer and model,  to determine if
                  an adjustment of calibration interval is needed.

                  a.   Utilize the attached chart based upon the percent of  times
                       equipment has been out of tolerance when submitted  for
                       scheduled recalibration.
               CONCURRENCES
DATE
     IMPLEMENTATION
 PREPARED  BY:
PAGE
                   OF
 APPROVED  BY:
         DATE  ISSUED:
                                        273

-------
                                                                 QMP No.  5.3
                                                                 Page 2 of 2
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     A.   Standards and Calibration (continued)

              'b.    Adjust calibration interval when needed and transmit informa-
                    tion to Data Processing to change affected Calibration
                    Control Card(s) (QMP Form No. 7.2).

                    NOTE:  The calibration interval for any item can be extended
                    only when its history for the past year shows zero (0)
                    "times out of tolerance".

          5.   Initiates action to dispose of or replace individual pieces of
               equipment with a history of poor reliability or uneconomical
               maintenance cost.
                                      21k

-------
                                                                QMP No. 5.3
                                                                Attachment
                  OUT  OF TOLERANCE FREQUENCY DISTRIBUTION CHART
*This chart  shall  be  used  for adjusting the
 calibration interval of a specific model of
 test equipment  only  after the  evaluation of
 a minimum of one  year's calibration results
 or twenty (20)  calibration results, which
 ever occur  first.
                                                                   20-50%
                 12
                                                                   10-20^
                                                                  0-10^
         •I.
         a
         -
         i
         53
         .
         -
         v
                                                                Times  out  of tolerance*
                         1     2     3     k     6     9

                             Adjusted Interval  In Months

Example of Chart  Usage:

1.  Current Interval  -  six  (6)  months

2.  Percent Out of Tolerance  -  eighteen  percent
                                                              12
 3-  Follow  six (6) month current interval line to intersection of
    (10-20^) line.  Read down vertical line to adjusted  interval  of
    four  (h] months.  This establishes a new base interval  of four
    (k) months.
                                        275

-------
EPA QUALITY  MANAGEMENT PROCEDURE
QMP NO.
     5.4
                                                                    REVISION  DATE
SUBJECT:
             CALIBRATION STANDARDS
   I    PURPOSE

        This QMP prescribes the  requirements for establishing and controlling
        standards used to determine the accuracy of emission test systems.

   II   BACKGROUND

        A.   The continuous validity of test variables is dependent upon sequential
             comparisons of equipment accuracy with known standards of progressively
             higher orders of precision.

        B.   As a goal, standards  used to calibrate other equipment are to have ac-
             curacies of at least  4 times better than that of equipment to be
             calibrated.

        C.   Definitions:  The nomenclature and definitions used among emission
             laboratories varies widely, therefore, the standards discussed in this
             procedure are defined below:

             Measuring and Test  Equipment - Measuring and sensing devices used to
             establish specifications or determine the acceptability of processes
             or data.

             Transfer Standards  -  Measuring and sensing devices having accuracies dir-
             ectly traceable to  Reference Standards.

             Reference Standards - Measuring and sensing devices having the highest
             order of accuracy in  the calibration system.

   Ill  PROCEDURE

        A.   Standards and Calibration

             1.   Establish Reference and Transfer Standards having accuracy,
                  stability and  range which are compatible with test specification
                  requirements.

             2.   Establish and  maintain intervals for recalibration of Reference and
                  Transfer Standards based upon stability, reliability, intended
                  usage and calibration history of the equipment.

             3.   Use Reference  and Transfer Standards in an atmosphere controlled as
                  necessary, to  assure accuracy of measurements and to prevent con-
                  tamination or  corrosion.
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE


IMPLEMENTATION
PAGE i OF 2
DATE ISSUED:
                                        277

-------
                                                                 QMP No.  5.4
                                                                      2 of 2
III  PROCEDURE (Continued)

     A.    Standards and Calibration (continued)

          4.    Maintain records certifying that the calibrations of Reference
               Standards are traceable to the National Bureau of Standards,  or
               have been derived from accepted values of physical constants,
               or have been derived by ratio type of self-calibration techniques.

-------

-------
QUALITY MANAGEMENT PROCEDURES
         SECTION 6.0
    LABORATORY OPERATIONS
              279

-------
EPA QUALITY  MANAGEMENT PROCEDURE
   QMP NO.
       6.1
                                                                    REVISION  DATE
SUBJECT:
             LIGHT DUTY TEST OPERATIONS
        PURPOSE
        This QMP establishes  the functions to be performed during mobile source
        emission testing  to ensure the quality and validity of the data generated
        during the test.

   II   BACKGROUND

        A.   Specific detailed procedures for performing emission tests are outlined
             in the Test  Procedure Manual.  It is also necessary to outline  the
             responsibilities and interrelationships of Test Operations and  Quality
             Assurance by generating a QMP.

        B.   Certain quality  functions are necessary to ensure the precision and
             accuracy of  the  data generated by the measurement system.  Quality
             Assurance has the responsibility of determining that these functions
             achieve the  desired level of precision and accuracy within the
             measurement  system.

   Ill  SCOPE OF APPLICATION

        A.   All phases of vehicle emission testing shall be subject to definitive
             Quality Assurance provisions on both a scheduled and audit basis.

        B.   Data generated during each  sequential test phase shall be documented and
             validated prior  to start-up of next test sequence.

        C.   Any deficiencies encountered during the testing operations shall be fully
             documented,  investigated and corrected to preclude their reoccurrence.

   IV   RESPONSIBILITIES  AND  PROCEDURES

        A.   Test Operations

             1.   Prepares, implements and revises the Test Procedure Manual, which
                  details the procedures to be used in light-duty vehicle emission
                  testing.

             2.   Assures that  the procedures are being correctly followed and that
                  the technician has the required skill and knowledge  to perform
                  his assigned  tasks, by implementing evaluation and training
                  programs.

        B.   Production Control

             1.   Schedules vehicle  for  test.
               CONCURRENCES
DATE
IMPLEMENTATION
 PREPARED BY:
          PAGE  1  OF	5
 APPROVED BY'-
          DATE  ISSUED:
                                         28l

-------
                                                                 QMP No.  6.1
                                                                 Page 2 of 5
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     B.   Production Control (continued)

          2.   .Receives and inspects the scheduled vehicle and completes the
               vehicle receiving inspection documentation.

     C.   Data Validation

          1.    Verifies accuracy and satisfactory completion of vehicle
               receiving inspection.

     D.   Vehicle Test

          1.    Verifies that proper amount and type of fuel is used.

          2.    Performs vehicle preconditioning checks, vehicle performance
               checks, vehicle preconditioning test and completes applicable
               portion of driver's preconditioning report.

     E.   Data Validation

          1.    Verifies that all elements of the preconditioning require-
               ments have been satisfactorily and accurately completed and
               authorizes vehicle to proceed to next test function.

     F.   Vehicle Test

          1.    Verifies the correct type and amount of fuel is being used in
               test performance.

          2.    Performs evaporative test preparation and conducts a diurnal
               evaporation test according to prescribed test procedure.

          3.    Records all test results on proper form and submits them to
               data validation.

     G.   Data Validation

          1.    Verifies that all elements of the diurnal evaporation test
               have been satisfactorily and accurately completed.  Author-
               izes vehicle to proceed to next test function.
                                      282

-------
                                                                 QMP No.  6.1
                                                                 Page 3 of 5
IV   RESPONSIBILITIES AND PROCEDURES  (Continued)

     H.    Vehicle Test

          1.    Performs dynamometer preparation and constant volume sampler
               (CVS)  set up according to prescribed procedures.

          2.    Performs vehicle driving schedule within specified speed-
               time limits.

          3.    Performs sample analysis in accordance with prescribed pro-
               cedure and documents all data,  including ambient  conditions
               and instrument operating parameters.

          4.    Completes the hot soak evaporative  loss test and  submits all
               documentation to data  validation.

     I.    Data Validation

          1.    Checks all data for completeness and accuracy and forwards to
               Data Services for preliminary processing.

     J.    Vehicle Test

          1.    Performs Federal Highway Fuel Economy test to determine fuel
               consumption in miles per gallon and forwards all  documenta-
               tion to data validation.

     K.    Quality Assurance

          1.    Maintains continual surveillance over the functions associated
               with the performance of all phases  of the light-duty vehicle
               emission testing program.

          2.    Assures proper and current calibration of instruments and
               equipment used in vehicle testing.

          3.    In the event of a test failure, whether instrument, driver, or
               vehicle, prepares a Test Condition Report to describe the
               nature of the problem, and coordinates with other organizations
               to assure that expedient corrective action is taken.

          4.    Performs audits and correlation studies of test operations
               to ensure the reliability and accuracy of the data.

          5.    Submits reports of  data and failure analyses to Management
               and Laboratory Operations.
                                      283

-------
                                                                 QMP No.  6.1
                                                                 Page 4  of 5
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     L.   Data Validation

          1.    Checks all submitted data and vehicle information for com-
               pleteness and accuracy.

          2.    Forwards all data to Data Services for final processing.

     M.   Production Control

          1.    After satisfactory completion of all tests and restoration of
               the test vehicle to the  "as received" condition,  completes
               shipping order and returns vehicle to the manufacturer/owner.
                                     284

-------
                                                                 QMP No. 6.1
                                                                 Page 5 of 5
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     N.   Light Duty Vehicle Test Flow Schematic
Production
Control
Receives &
Inspects Vehicle
                                                         Yes
Exhaust
Emission
Test
Yes
                  No
           Yes
Evaporative
Emission Test

Te
Repo
'
_.
st
rt
r
Quality Assurance
Performs Failure
Analysis as
Needed
i




Data
Service
I
f J YPS

Quality Assurance
Audits and Performs
Analysis of Data
Files
1

*
Vehicle
Returned To
Manufacturer

Laboratory
Operations
T
                                     285

-------
EPA QUALITY  MANAGEMENT  PROCEDURE
   QMP NO.
       6.2
                                                                    REVISION  DATE
SUBJECT:    COORDINATION AND IMPLEMENTATION OF EQUIPMENT
             OR PROCEDURE CHANGE NOTICES
   I    PURPOSE

        This Quality Management Procedure  (QMP) describes the procedure for origina-
        ting, coordinating,  and implementing changes in practices or equipment
        specified in the Testing Procedures  (TP) used in the Mobile Source Emission
        Measurement Program (MSEMP).

   II   BACKGROUND

        A.   Quality in the MSEMP is  dependent upon strict adherence to prescribed
             procedures and equipment configuration as defined in the Test Pro-
             cedures and the Federal  Register.

        B.   This QMP provides three  functions -

             1.   A means to revise or improve prescribed procedures, with documented
                  control of any such changes.

             2.   A mechanism to control changes made to equipment used in the test
                  facility.

             3.   A formal method of  introducing Federal Register revisions
                  into the EPA Test Procedures.

   Ill  SCOPE OF APPLICATION

        A.   The general scope of application of the equipment and procedures
             change notice (EPCN) is  the area of Laboratory Operations including
             Test Operations, Support Operations, and Test Scheduling.  An EPCN
             may be originated by any department manager or team leader.

        B.   The EPCN may also be originated by Quality Assurance or other functional
             groups.

        C.   All EPCNs are to be implemented using QMP Form No. 7.5 shown in
             Section 7.5 of  the forms instruction.

   IV   RESPONSIBILITIES AND PROCEDURES

        A.   Originator of EPCN

             1.   Drafts the EPCN using QMP Form No. 7.5 and submits it to Laboratory
                  Operations.

             2.   Maintains a copy of the  original draft EPCN for record and follow-up
               CONCURRENCES
DATE
     IMPLEMENTATION
PREPARED  BY:
PAGE
                   OF  3
APPROVED  BY:
         DATE  ISSUED:
                                         257

-------
                                                                 QMP No. 6.2
                                                                 Page 2 of 3
IV   RESPONSIBILITIES AND PROCEDURES  (Continued)

     A.   Originator of EPCN  (continued)

          3.   Makes major revisions  to the draft EPCN as requested by reviewers
               and resubmits  to  Laboratory Operations.

     B.   Laboratory Operations

          1.   Determines the areas affected  and indicates the distribution
               of the draft EPCN.

          2.   Submits draft  EPCN  to  Document Control for assignment of EPCN
               number and distribution to the affected departments for review
               and comment.

          3.   Reviews comments  on the draft  EPCN and makes decision to re-
               turn to the originator for revision or determines that the
               draft EPCN should be implemented.

          4.   Determines effective date(s) of change implementation and main-
               tains EPCN file.

          5.   Obtains required  approvals necessary for implementation of EPCN.

          6.   Approved EPCN  is  forwarded to  Document Control for formal
               implementation of change.

     C.   Quality Assurance

          1.   Reviews draft  EPCN  for incorporation of quality provisions and
               acceptance criteria, and adequate equipment specifications
               and blue prints and/or schematic diagrams in procedural/
               equipment changes specified in EPCN.

          2.   Forwards draft EPCN to Laboratory Operations with recommendations.

          3.   Performs procedure  or  equipment audit to assure implementation
               by the effective  date, and incorporation of adequate quality
               provisions in  revised  documents.

     D.   Document Control

          1.   Assigns EPCN number and distributes draft EPCN to reviewers as
               prescribed by  Laboratory Operations and maintains the originals
               in a file by numerical sequence.
                                       288

-------
                                                                 QMP No.  6.2
                                                                 Page 3 of 3
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     D.   Document Control (continued)

          2.   Implements required changes to prescribed procedures/equipment
               configuration and maintains appropriate records of these
               changes .

          3.   Distributes revised procedures together with copy of EPCN
               authorizing change, to all Procedure Manual holders.

          4.   Distributes copies of EPCN's affecting equipment configuration
               changes together with revised documentation to departments/
               personnel affected by the change.

     E.   EPCN Reviewer
          1.   Comments on, approves or disapproves of the EPCN.   Suggests appro-
               priate revisions and returns to Laboratory Operations in a
               timely manner.

     F.   EPCN Flow Schematic
   Originator
   Drafts EPCN
Lab. Operations
Reviews and
Assigns
Distribution
                      Yes
Document
Control
Assigns EPCN
Distributes
                                 Other Affected
                                 Departments
                                  Reviews
                                  Draft EPCN
                                        L
               Quality
               Assurance
                Reviews
                Draft EPCN
                    4
Lab . Operations
Reviews
Comments
                             Document
                             Control
                             Drafts Change
                             and Distributes
Affected
Departments &
Manual Holders
File Revisions
t

Quality
Assurance
Audits for
Implementation
                                       289

-------
EPA QUALITY  MANAGEMENT PROCEDURE
   QMP NO.
      6.3
                                                                    REVISION  DATE
SUBJECT:
             TEST VEHICLE FUEL CONTROL
   I    PURPOSE

        This QMP establishes the requirements for the control of fuel used in the
        performance of a Vehicle Emission Test.

   II   BACKGROUND

        A.   Fuels used in the performance of a test must meet Federal Register speci-
             fications.  The composition and characteristics of the fuel used for an
             emission test can affect the data and make a test invalid.  Many of
             the 1975 vehicles are equipped with catalyst devices which become inopera-
             tive if fuels containing lead additives are used.  Therefore, it is of
             utmost importance that responsibilities are designated for the controlled
             use of fuel, and that procedures and equipment are designed to prevent
             the use of incorrect fuel in an emission test vehicle.

        B.   Storage and handling of these fuels must be controlled since such speci-
             fications as the Reid vapor pressure can change during storage or
             transfer of the fuel.  Contamination of fuels with undesirable com-
             ponents such as lead or diesel fuel should be avoided since this would
             have detrimental effects on engines and emissions control systems.
             Storage tanks cannot be pumped "dry" so there is always some residual
             fuel left in the tank, and frequently water and sludge collect in the
             bottom of the tanks.  Historical records of fuel storage facilities must
             be kept up to date, tanks should be used only for fuels of the same
             specifications and periodic examination of storage tanks must be
             conducted.

        C.   Procurement control of the fuels used in the testing facility is also
             critical and the responsibilities and procedures for purchasing,
             receiving, and inspecting fuels must be detailed.

   Ill  SCOPE OF APPLICATION

        A.   This procedure generally applies to any fuel used in the test facility
             but specifically to the leaded  (Indolene 30) and unleaded  (Indolene HO)
             fuels used in the emission test.

        B.   Leaded fuel must not be used  in vehicles equipped with catalyst devices.
                CONCURRENCES
DATE
     IMPLEMENTATION
 PREPARED BY:
PAGE
                   OF
APPROVED BY:
         DATE  ISSUED:

-------
                                                                 QMP No.  6.3
                                                                 Page 2 of 4
IV   RESPONSIBILITIES AND PROCEDURES

     A.   Purchasing

          1.   Indicates Federal Register fuel specifications on purchase order
               and obtains approval of Quality Assurance.   Requests batch analysis
               from the vendor.  Specifies ASTM or other method for analysis of
               specified characteristics.

          2.   Fuels other than those required by the Federal Register or not
               clearly specified, such as the fuels used for durability or emis-
               sion data vehicles, must be clearly specified and approved by
               Quality Assurance.

     B.   Quality Assurance

          1.   Reviews procurement documents for all fuel used in the test facility.

          2.   Specifies, approves ASTM or other methods for analysis of fuel
               characteristics.

          3.   Develops, details and implements procedures for fuel inspection
               and monitoring programs.

          4.   Receives copies of all fuel analysis reports and releases fuel
               that meets specifications to the testing operations.

          5.   Reports any discrepancies found in the fuel specifications
               analyses to Purchasing and supplier.  Determines final dis-
               position of the fuel and assures that it has not been and will
               not be used in any test vehicle.

          6.   Coordinates corrective action with Purchasing and Test Operations
               when necessary to ensure uninterrupted availability of correct
               test fuels.

     C.   Receiving

          1.   Checks batch number against batch analysis.  Checks batch
               analysis to confirm compliance with purchase order specifications.

          2.   Obtains sample of fuel from all bulk shipments in rinsed one
               gallon container and forwards to chemical analysis.
                                     292

-------
                                                                 QMP No.  6.3
                                                                 Page 3 of 4
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     D.   Chemical Analysis

          1.   Analyzes all received fuel for lead and phosphorous content.
               Unleaded fuels shall not be released for use until lead content
               is verified to be within specifications.

          2.   Determines that fuel meets specifications either by analysis
               "in-house" or by independent test laboratory.

          3.   Prepares detailed laboratory procedures for "in-house" fuel
               analysis and obtains Quality Assurance approval.

          4.   Removes fuel sample from bulk storage containers monthly for
               analysis, and reports data to Laboratory Operations and
               Quality Assurance.

          5.   Monitors all fuel storage areas for proper environmental control.

     E.   Production Control

          1.   Stamps fuel requirements, clearly, on all work sheets and check
               lists associated with each test vehicle.

     F.   Vehicle Test

          1.   Verifies the type of fuel required and the nozzle configuration
               used for each type of fuel.  Place appropriate color coding on
               pumps, containers, fuel conditioning equipment, bulk fuel lines
               and vehicles to clearly identify the correct fuel.

          2.   Trains technicians in the proper handling, storage, transferring
               of fuels and color coding of vehicles to ensure vehicle fuel
               requirements are met.

          3.   Maintains vehicle fueling logs including the vehicle identifi-
               cation number, type of fuel, number of gallons dispensed, and
               signature of technician and witness.
                                      293

-------
                                                                 QMP No.  6.3
                                                                 Page 4 of 4
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     G.   Test Vehicle Fuel Control Flow Schematic
                   Purchasing
                   Initiates
                   P.O.
                   Supplier
                   Ships Fuel
                   with
                   Bulk Analysis
 Q.A.
 Procurement
 Document
 Review
                   Receiving
                   Doc. Review
                   Samples Fuel
Q.A.
Review
Fuel Analysis
Reports
                   Chem. Analysis
                   Fuel Sample
Prod. Control
Identifies
Fuel
Requirements


Vehicle Test
Storage &
Color Code


                                                                    Yes
                  Vehicle  Test
                  Ensure Proper
                  Use of Fuel
                  in Vehicle
Q.A.
Review
                                     294

-------
EPA QUALITY  MANAGEMENT PROCEDURE
   QMP NO.
       6.4
                                                                    REVISION DATE
SUBJECT:
           TEST VEHICLE SCHEDULING
   I    PURPOSE

        This procedure establishes the requirements attendant to the scheduling
        of vehicle testing operations.

   II   BACKGROUND

        A.   The orderly and timely performance of mobile  source emission testing
             dictates the need for identifying the responsibilities and procedures
             for scheduling these tests.

        B.   Scheduling and timely reporting of the projected testing load to Vehicle
             Test is necessary for organization and planning of future test
             requirements.

   Ill  SCOPE OF APPLICATION

        A.   All requests for emission test and retests must be submitted with proper
             authorization to Production Control.

        B.   Production Control notifies Laboratory Operations and Vehicle Test of
             the test schedule on a daily, weekly and monthly basis.

   IV   RESPONSIBILITIES AND PROCEDURES

        A.   Production Control

             1.   Upon receipt of a Test Request from the  Certification and/or other
                  divisions:

                  a.  Determines the type(s) of test required.

                  b.  Determines equipment and facility availability.

                  c.  Establishes priority based on test  program requirements.

             2.   Schedules the test and sends notification of date and time to the
                  requester.

             3.   Prepares a weekly test schedule summary  for submission to Labora-
                  tory operations a week prior to the scheduled testing, and requires
                  notification of concurrence with the test schedule no later than
                  the last working day of that week.
               CONCURRENCES
DATE
     IMPLEMENTATION
PREPARED  BY:
PAGE
                  OF
APPROVED  BY:
         DATE ISSUED:
                                        295

-------
                                                                 QMP No.  6.4
                                                                 Page 2 of 3
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     A.   Production Control (continued)

          4.   Submits daily test schedules to Test Operations on the day pre-
               ceding scheduled tests.

          5.   Prepares yearly projections (updated monthly)  for Testing Opera-
               tions and the Laboratory.

          6.   Receives test vehicles and ships them (after notification of
               test validity) upon authorization obtained on the Receiving and
               Shipping order.

          7.   Schedules vehicles for retest at the earliest possible date,
               when original test is declared invalid and an authorized request
               for retest is received.

     B.   Vehicle Test

          1.   Submits authorized requests for retest of vehicles invalidated
               for any reason together with a description of test priority
               requirements.

          2.   Informs Production Control of scheduled "down time" of test
               cells and immediately informs Production Control of unscheduled
               "down time"  and the expected start up date and time.  Keeps
               Production Control informed of test cell status on a daily basis.

     C.   Test Operations

          1.   Authorizes,  submits and monitors the projected test scheduling
               on a monthly basis.  Submits projected annual test loads and
               develops and implements plan for meeting these requirements.

     D.   Quality Assurance

          1.   Assists Production Control and Laboratory Operations in
               developing efficient programs for meeting future commitments,
               with the implementation of specific quality requirements
               where necessary.
                                      296

-------
                                                                  QMP No.  6.4
                                                                  Page  3 of  3
 IV   RESPONSIBILITIES AND PROCEDURES (Continued)
      E.    Typical Scheduling Flow Schematic
  Test
  Requester
  Initiates
Test Request
 Laboratory
 Operations
                   Quality
                   Assurance
      Production
       Control
       Work Order
       Each Test
                                    Production
                                    Schedule
               Shipping &
                Receiving
            Yearly
         (updated
         monthly)
  Weekly
                                   Vehicle
                                   Test
Daily
                        Test  Operations
            Lt.  Duty
            Testing
  E. & D.
  Testing
Heavy Duty
Testing
                                                                    -P
                                                                    (0
                                                                    0)
                                                                    •d
                                                                    n)

                                                                    H
                          Test Validation
                 Tech.
EPA Rep.
                                              Man. Rep.
-p
to
0)
                                                   T)
                                                   •H
                                                   r-i
                                                   rt
                                                                       No
                                      29?

-------
EPA QUALITY  MANAGEMENT PROCEDURE
                                                     QMP NO.
                                                         6.5
                  REVISION DATE
SUBJECT:
             TEST FACILITY SUPPORT SERVICES
   I    PURPOSE

        This QMP outlines the responsibilities and procedures for providing support
        services to vehicle testing such as chemical analysis, equipment engineering,
        instrument services, correlation and maintenance and craft services.

   II   BACKGROUND

        A.   Extended "down time"  created by outside supplier service delays cannot
             be tolerated and the  expense of maintaining duplicate sets of equipment
             is prohibitive in many cases.  It is essential therefore, that a
             measurement system should be designed to be self sufficient in supplying
             support services for  equipment and instruments used in the system.

        B.   Responsibilities and  procedures used for the support groups are pro-
             bably the most varied and least defined in measurement systems.  This
             QMP describes support services generally as they exist at a typical
             government test facility. Development of support service responsi-
             bilities and procedures will depend largely on the ability and desire
             of the Testing Laboratory to invest in support equipment, testing
             equipment and calibration standards and, in addition, the availability
             and skill of the personnel in the support group.

   Ill  SCOPE OF APPLICATION

        A.   These responsibilities and procedures generally apply to groups not
             directly involved in  testing a vehicle but in maintaining, repairing,
             calibrating and correlating equipment and instruments used in the
             facility.  It applies also to groups performing any functions, test or
             analysis not specifically required by the federal test procedure but
             necessary for the particular program or organization.

        B.   Any service related to the instrumentation, equipment, fuels, or gases
             used in performance of mobile source emission testing is subject to
             evaluation by Quality Assurance.

        C.   All maintenance must  be authorized by the vehicle test management.

        D.   Support services may  not perform unscheduled services without
             authorized work order issued by Production Control.
               CONCURRENCES
DATE
IMPLEMENTATION
PREPARED  BY:
         PAGE l   OF  4
APPROVED  BY:
         DATE  ISSUED:

-------
                                                                 QMP No.  6.5
                                                                 Page 2 of 4
IV   RESPONSIBILITIES AND PROCEDURES

     A.   Chemical Analysis

          1.   Generates binary gravimetric gas mixtures to be used as pri-
               mary standards.

          2.   Determines purity of gases used in generation of laboratory
               standards.

          3.   Determines lead concentration in test fuels.

          4.   Determines sulfates by wet chemistry.

          5.   Blends and stores propane used for CVS Tracer gas injections.

          6.   Performs required calibration of barometers and hygrometers.

     B.   Quality Assurance

          1.   Performs audit of gravimetric gas mixtures to assure analysis
               output validity.

          2.   Evaluates incoming gases to assure that the desired purity
               standards are maintained.

     C.   Equipment Engineering

          1.   Maintains inventory of all equipment and instrumentation
               utilized by the Laboratory operations.

          2.   Issues and controls use of measurement-related equipment.

          3.   Designs prototype measurement and analytical systems for
               special contract requirements.

          4.   Coordinates with Craft Services during production of design
               equipment.

     D.   Quality Assurance

          1.   Coordinates with Equipment Management/Design to  assure equip-
               ment required for  the Measurement System meets contract
               specifications.

          2.   Checks design specifications to assure the desired  results will
               be attained.
                                      300

-------
                                                                 QMP No.  6.5
                                                                 Page 3 of 4
IV   RESPONSIBILITIES AND PROCEDURES (Continued)
     E.   Instrument Services
          1.   Performs periodic calibration of chart recorders, temperature
               recorders and support electronic instrumentation.

          2.   Performs routine, preventive and emergency maintenance on
               electronic equipment.

          3.   Maintains control of  all instrument and equipment manuals re-
               quired for the measurement equipment.

          4.   Maintains complete file of instrument failures and corrective
               action.

     F.   Quality Assurance

          1.   Reviews and evaluates calibration procedures with reference
               to data collection and analysis.

          2.   Reviews maintenance procedures, frequency of repair to assure
               timely, efficient repairs are accomplished, and verifies the
               implementation of corrective action where applicable.

          3.   Reviews manual control file to determine that a complete informa-
               tion file exists.

     G.   Correlation and Maintenance

          1.   Performs periodic calibration of the CVS, chassis dynamometers
               and gas analyzers.

          2.   Provides gas analysis inspection of incoming gases.

          3.   Performs CVS Tracer verification, dynamometer calibration
               verification and NO  efficiency checks.

          4.   Updates log books and completes calibration tags correctly.

     H.   Quality Assurance

          1.   Reviews and evaluates data collection and curve analysis
               techniques in calibration procedures to assure the integrity
               of the operation practices.

          2.   Validates verification checks performed on the measurement
               system.
                                     301

-------
                                                                 QMP No. 6.5
                                                                 Page 4 of 4
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     I.   Craft Services

          1.   Performs equipment modification required to meet current
               operation requirements.

          2.   Produces prototype systems under the direction of Equipment
               Management/Design.

     J.   Quality Assurance

          1.   Evaluates modification requirements and witnesses functional
               operation tests.
                                      302

-------
EPA QUALITY  MANAGEMENT  PROCEDURE
QMP NO.
    6.6
                                                                     REVISION  DATE
SUBJECT:
            DATA VALIDATION
   I    PURPOSE

        This QMP establishes the criteria to be followed in the  evaluation of  raw
        measurement data generated from the Federal Test Procedure  (FTP) and Highway
        Fuel Economy Test (HWFET).

   II   BACKGROUND

        A.   Experience indicates that data and information generated during mobile
             source emission testing is subject to error.   All raw  data must be
             checked by personnel familiar with the procedure but not directly
             involved in the performance of a test.  The validation procedure  must be
             performed expeditiously, as the test must be validated prior to vehicle
             release, or in the event of an invalid test,  Production Control must
             schedule a retest.

        B.   Data validation may be done manually or automatically  by computers
             programmed to detect omissions or suspect data.

   Ill  SCOPE OF APPLICATION

        A.   All data generated from the various phases of Vehicle  Emission Tests
             shall be validated according to prescribed procedures.

        B.   Any unusual values discovered during the evaluation will be fully docu-
             mented and examined for validity prior to a rejection  decision.

        C.   Data validation is concerned with the accuracy, precision and complete-
             ness of the data, however, this function should not be considered as,
             or take the place of a data audit by Quality Assurance.

        D.   Data validation also assists in the preparation and distribution  of the
             forms used in the test facility.

   IV   RESPONSIBILITIES AND PROCEDURES

        A.   Data Validation

             1.   Records test number and manufacturer's inspection data on CVS
                  Data Sheet.

             2.   Distributes daily test schedule to appropriate sections.

             3.   Following the FTP, receives all traces and forms  pertaining  to
                  the test.
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE


IMPLEMENTATION
PAGE 1 OF 4
DATE ISSUED:
                                          303

-------
                                                                 QMP No.  6.6
                                                                 Page 2 of 4
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     A.   Data Validation (continued)

          4.   Checks analyzer traces for obvious errors and compares chart
               values with those logged on the Analyzer Read-out form.

          5.   Transcribes concentration values on to CVS Data Sheet and
               checks all entries for spurious values.

     B.   Quality Assurance

          1.   Performs audit of test data to assure the required data is
               complete.

          2.   Evaluates spurious values discovered by Data Validation to
               determine if the data is acceptable.

          3.   In the event of test failures, prepares a Test Condition Report
               which describes the reasons for rejection and coordinates with
               other organizations to affect corrective action.

     C.   Data Validation

          1.   Submits data to the Data Branch for preliminary analysis.

          2.   Obtains preliminary results and CVS Data Sheet from Certification
               Branch representative and makes final check of data.

          3.   If errors are discovered, corrections are made and corrected
               data sheet is re-routed to the Data Branch for a new print-
               out.

          4.   Checks off the remaining documentation and enters it in the
               vehicle file.

     D.   Quality Assurance

          1.   Reviews and evaluates all vehicle emission testing procedures
               with reference to error and bias in collection, handling and
               analysis of samples.

     E.   Data Validation

          1.   Receives HWFET results from the CVS operator.

          2.   Checks analyzer traces for errors and enters concentration values
               on to the HWFET CVS Data Sheet.
                                      304

-------
                                                                 QMP No.  6.6
                                                                 Page 3 of 4
IV   RESPONSIBILITIES AND PROCEDURES (Continued)

     E.    Data Validation

          3.   Submits HWFET results to the Data Branch for preliminary analysis.

          4.   Analyzes the supporting data for out of limit conditions and
               enters it into the vehicle file.

     F.    Quality Assurance

          1.   In conjunction with the Certification Branch representative, ex-
               amines the driver's trace, preliminary results and supporting data
               for spurious values.

          2.   If a test failure is discovered, caused by either instrumentation,
               human, or vehicle error, submits request for retest to Production
               Control.

     G.    Data Validation

          1.   Packages all documentation and preliminary results from the FTP
               and HWFET (if applicable) in file envelopes and the complete data
               file is sent to the Data Branch for final processing.

          2.   Following final processing, checks results for errors and makes
               necessary corrections.

          3.   Marks "official values" on the print-outs and delivers two
               copies plus the blue CVS Data Sheet to the Certification Branch.

     H.    Quality Assurance

          1.   Reviews all data records at regular intervals for possible human
               error such as:

               a.   Failure of technician to record pertinent information.

               b.   Errors in reading an instrument.

               c.   Errors in calculating results.

               d.   Errors in transposing data from one form to another.

               e.   Errors in keypunching data.

               f.   Errors in computer tape handling, programming and print-outs.
                                      305

-------
                                                                 QMP No. 6.6
                                                                 Page 4 of 4
IV   RESPONSIBILITIES AND PROCEDURES  (Continued)

     H.   Quality Assurance  (continued)

          2.   Utilizes statistical sampling and control chart techniques when-
               ever they can be applied advantageously in data verification.

          3.   Assures corrective action is implemented to prevent recurring
               errors in data recording and analysis.

     I.   Data Flow Schematic

          See QMP 6.1
                                     306

-------

-------
QUALITY MANAGEMENT PROCEDURES








         SECTION 7.0







     FORMS INSTRUCTIONS
               307

-------
EPA QUALITY  MANAGEMENT  PROCEDURE
   QMP NO.
        7.1
                                                                   REVISION  DATE
SUBJECT:
             FORMS  INSTRUCTION - INSTRUMENT LOAN ORDER
   1.   FORM NUMBER 7.1:1-31-75

   2.   FORM USE:     QMP 5.1

        2.1  To provide a record of the individual and organization having  custodial
             responsibility for equipment requisitioned from Standards and  Cali-
             bration Equipment Stores.

   3.   FORM INSTRUCTIONS

        3.1  The paragraph numbers listed below coincide with the numerals  in  the
             blocks in Attachment No. 1.

             1.   Man No. - Employee number of  the individual requisitioning the
                  equipment.

             2.   Control No. - The control number affixed to the equipment being
                  requisitioned; i.e., ACL 81352, ORD 15421, etc.
                                                /
             3.   Orgn. - The organization number of the individual requisitioning the
                  equipment.

             4.   Date - The date the equipment is requisitioned.

             5.   Kind of Equipment - Nomenclature, Mfr. and Model No. of equipment
                  borrowed; i.e., Counter, H-P  522B, etc.

             6.   Employee - Signature of the individual requisitioning the equipment.

             7.   Supervisor - Signature of the requisitioning individual's
                  supervisor.
               CONCURRENCES
DATE
IMPLEMENTATION
PREPARED BY:
         PAGE
    OF
APPROVED BY:
         DATE ISSUED:
                                        309

-------
                                         QMP NO. 7.1
                                         Attachment No. 1
                «s>*-
INSTRUMENT
FORM
MAN NO.
ORGN

LOAN ORDER j
^.^ CONTROL NO.
DATE
KIND OF INSTRUMENT
©
WORKMAN
until it
to you.
Employee
Supervisor.

NOTE: This instrument is in your charge
is returned. If lost, it will be charged
Keep this slip until instrument is returned.
©
©

•\
&
**•
***
d
«t~
.— •
•— '
***
FORM NO. 7.1:  1-31-75
                     310

-------
EPA  QUALITY  MANAGEMENT  PROCEDURE
QMP NO.
    7.2
                                                                    REVISION DATE
SUBJECT:
           FORMS INSTRUCTION -  CALIBRATION CONTROL CARD
  1.    FORM NUMBER:   7.2:  1-31-75

  2.    FORM USE:   QMP 5.1, 5.3

       2.1  The calibration control card is used by Records Control and Computer
            Operations to automatically scan the file for instrument calibration
            requirements.  The information is completed by Records Control  and a
            new card issued by Computer Operations each time a change in calibration
            status or location is determined.

  2.    FORMS INSTRUCTION

       3.1  The paragraph numbers listed below coincide with the numerals in the
            blocks in Attachment No. 1.

            1.   Control Number - The number(s) assigned by standards and Calibra-
                 tion to equipment  to be used in emission testing.

            2.   Nomenclature - Equipment type, name; i.e., "Power Supply,"
                 "Cap Decade," etc.

            3.   Manufacturer - Manufacturer's name; i.e.,  "Gen Radio," "Gen Elect.,"
                 etc.

            4.   Model  - Equipment  model  number;  i.e.,  "320A," "CDA5," etc.

            5.   Type - 3 digit code number  specifying  equipment  type.

            6.   Mfr. - 3 digit number  specifying manufacturer.

            7.   Cycle - Interval (days)  for recalibration  (cycle period);
                 i.e.,  "60", "90",  "120",  etc.

            8.   Orgn.  No.  - Identification  of the organization having custody of
                 the equipment.

            9.   Fac - Identification of  the facility  where equipment  is located.

             10.  Due - Date the equipment is due for recalibration/maintenance.

             11.   In. - Number of times instrument was  found to be within acceptable
                  tolerance limits when recalibrated.

             12.   Out. - Number of times instrument was found to be out of  tolerance.

             13.   Rej. - Number of times instrument was rejected when in use and
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE


IMPLEMENTATION
PAGE l OF l
DATE ISSUED:
                                         311

-------
U)
           -J

           k>
           U)

(1)
1 I 1 1 1
1 1 1
of- CONTROL «0. "
(4)
i f 1 1 -1 1
1 1

-------
EPA QUALITY MANAGEMENT  PROCEDURE
    QMP NO.
        7.3
                                                                    REVISION  DATE
SUBJECT:
        FORMS INSTRUCTION -  CALIBRATION ORDER
   1.   FORM NUMBER:   7.3:   1-31-75

   2.   FORM USE:  QMP 5.1,  5.3

        2.1  To recall equipment  for periodic calibration/maintenance.

        2.2  To authorize receiving inspection, repair or special  calibration of
             equipment.

   3.   FORM INSTRUCTIONS

        3.1  The paragraph numbers listed below coincide with the  numerals in the
             blocks in Attachment No. 1.

             1.   Control Number  - The number(s)  assigned by Standards and
                  Calibration to  equipment to be used in emission  testing.

             2.   Nomenclature -  Equipment type,  name; i.e., "Power Supply," "Cap
                  Decade," etc.

             3.   Mfr. - Manufacturer's name; i.e., "Gen Radio," "Gen Elect.," etc.

             4.   Model - Equipment model number; i.e., "320A,"  "CDA5," etc.

             5.   Due - Date the  equipment is due for recalibration/maintenance.

             6.   Cal - Interval  (days) for recalibration (cycle period); i.e.,
                  "60," "90," "120," etc.

             7.   Orgn - Identification of the organization having assignment of the
                  equipment.

             8.   Fac - Identification of the facility where equipment is located.

             9.   Status - Code number of the equipment status (determined by
                  organization).

             10.  Repair - X entered when order is for repairing an item which
                  failed when in  use.

             11.  Recall - X entered when order is for periodic  recalibration/
                  maintenance of  equipment.

             12.  Buy-in - X entered when order is for receiving inspection of
                  equipment.
               CONCURRENCES
DATE
IMPLEMENTATION
PREPARED BY:
         PAGE 1   OF  3
APPROVED BY:
                                                          DATE ISSUED:
                                        313

-------
                                                                 QMP No. 7.3
                                                                 Page 2 of 3
3.   FORM INSTRUCTIONS (Continued)

          13.  Other - entered when order is for services other than 10,
               11 or 12 above.

          14.  Sign - Signature of the individual returning the equipment
               (Recall Order) or  initiating the order for "Repair," "Buy-in,"
               or "Other."

          15.  Date - Date of signature.

          16.  Ext - Telephone extension of individual returning equipment,
               or initiating order.

          17.  Yes - X entered by individual returning equipment if he wishes
               it back after recalibration.

          18.  No - X entered by  individual returning equipment if he does not
               want it back after recalibration.

          19.  Remarks - Special  instructions pertaining to orders for "Repair,"
               "Buy-in" or "Other" when necessary or information that a replace-
               ment item has been furnished.  Enter "See calibration report"
               when such a report is generated to support the calibration job.

          20.  Comp By - Man number of technician completing the work specified
               by the order.

          21.  Date - Date work is completed by the technician.

          22.  In/Tol - X entered if equipment was found to be in tolerance
               during scheduled recalibration.

          23.  Out/To 1 - X entered if equipment was found to be out of tolerance
               during scheduled recalibration.

          24.  MTL Cost - Cost (to nearest tenth of a dollar) of parts used to
               repair equipment;  i.e., "10.5," "1.3," etc.

          25.  Time Exp - Time spent by technician to complete work.

          26.  Next Servicing Date - Date the equipment will require recali-
               bration (month, date and year).

-------
                                                                 QMP No. 7.3
                                                                 Page 3 of 3
3.    FORM INSTRUCTIONS (Continued)

          27.   Out/Tol Details - Specific function(s)  involved,  variables
               data defining discrepancies and rework  performed to correct the
               condition(s).

          28.   O/T Reviewed by - Signature of individual evaluating out/tol
               details and related records to isolate  chronic or critical
               conditions.

          29.   Date - Date of O/T Review.
                                      315

-------
(JO
              •v]
              •

              Ul
               I
              Ul
              H

              ^J
              Ul
CALIBRATION OR DDR RECORD COP'*
	 " 	 fl> (?) ! (3) 00
CONTROL NO. NOMENCLATURE:. MANUFACTURER MODEI
(5) (6) j (7) (8) (9D iRBPAiRXio) |
DUE CALIB j ORGN FAC S | ' 1 BUY IN ( 1? ) 1
SIGNATURE (l4) DATE (l5) i EXT (l6) ' RETUKN INSTRUMENT
REMARKS (19)
O.

. -
COMP. BY (20) j DATE (2l) ' CONDITION RfCttVEC IN/TOL/p
M AT L COST (2l|.) i TIME EXPENDED (2S) , NEXT SERVICING PATF (gg)
*OUT/TOL. DETAILS (27)
— — - 	 	 	 	 • — - ----- _.....

O/T REVIEW BY (28) DATE (2Q)

RECALL (]•)
OTHbF< f]^
YES [ ; N
. 117)
g) OUT/TOL
                                                                                                                                                                                                                           Z
                                                                                                                                                                                                                           0

-------

QMP NO. REVISION DATE
EPA QUALITY MANAGEMENT PROCEDURE 7.4
SUBJECT:

FORMS INSTRUCTION - TEST CONDITION REPORT
1. FORM
2. FORM
2.1

3. FORM
3.1




















NUMBER: 7.4:1-31-75.
USE: QMP 6.1, 6.6.
The Test Condition Report (TCR) is used to record details of any
failures that occur during Mobile Source Emission Testing.
INSTRUCTIONS:
The paragraph numbers listed below coincide with the numerals in
the blocks in Attachment No. 1.
1. Failure - X entered if a test failure has occurred.
2. Void - X entered if a test has been voided.
3. Retest Requested - X entered if a retest has been requested.
4. Name - Name of person originating TCR.
5. Date Submitted - Date TCR was issued.
6. Branch - Identify organization to which originator of TCR
reports .
7. Section - Identify section/unit to which originator of
TCR belongs.
8. Extension - Telephone extension of originator of TCR.
9. Test Type - X entered in appropriate block to indicate type
of test, i.e., Light Duty (LD) , Medium Duty (MD) , Heavy
Duty (HD) , or Other Tests.
10. Manufacturer - Vehicle manufacturer's name, e.g.. Ford, GM, etc.
11. Identification Number - Vehicle Identification Number.
12. Date - Date failure or voided test occurred.
13. Time - Time of day that failure or voided test occurred.
14. Operator - Name of operator performing test.
15. -Equipment Involved in Failure - X entered in appropriate box(es).
CONCURRENCES DATE IMPLEMENTATION
PREPARED BY
APPROVED BY
: PAGE * OF 2
' DATE ISSUED:
317

-------
                                                                 QMP No. 7.4
                                                                 Page 2 of 2
3.   FORM INSTRUCTIONS (continued)

          16.  Failure Description - Originator writes in a complete description
               of the failure or condition which caused the test to be voided.

          17.  Void Point - Identify sequence in test at which failure or
               void condition aborted the test.

          18.  Hours Lost - Record time taken to run test up to void point
               (include preparation time).

          19.  Corrective Action Taken - Specify corrective action measures
               taken to preclude recurrence of voided test/failure.

          20.  Signature - Originator of TCR signs.
                                      318

-------
                            TEST CONDITION REPORT
Failure    y/

Name       (.4)
Void
                                                 Retest Requested

                                             Date Submitted
           HC
           CVS
           CVS Counter
           Dynamometer
        	NO      	

Drivers Trace            	

  	Cold Start

  	Other  (Specify)
Failure Description
                                                                 2

                                                               FID
                                                                     QMP NO. 7.4
                                                                     Attachment No. 1
Branch
Test Type:
(Q Section
ft) LD
Manufacturer (jo)
Date
(ll) Time (r£)
ft) Extension
MD HD
Identification Number
Operator
to
Other
CM)
04)
Equipment Involved in Failure: (jS)

	Analysis System          	Recorder          	Bags

                            CO               NO               CO
Temp.
                                                                   Hot Start
Void Point
Corrective Action Taken   (J9)
              Hours Lost
              (Include Prep Time)
 FORM NO. 7.4:  1-31-75
                                                           (20)
                                                      Signature
                                      319

-------
EPA QUALITY  MANAGEMENT PROCEDURE
   QMP NO.
      7.5
                                                                    REVISION DATE
SUBJECT:
            FORMS INSTRUCTION - EQUIPMENT/PROCEDURE CHANGE NOTICE
   1.   FORM NUMBER:   7.5:1-31-75.

   2.   FORM USE:   QMP 3.2,  6.2.

        2.1  The Equipment/Procedure Change Notice is used to document and
             implement changes in practices or equipment specified in Test
             Procedures (TP)  and Quality Management Procedures (QMP) used
             in the Mobile Source Emission Measurement Program.

   3.   FORM INSTRUCTIONS

        3.1  The paragraph numbers listed below coincide with the numerals  in
             Attachment No.  1.

             1.   Originator - Name of person originating EPCN.

             2.   Phone Ext.  - Phone extension of originator.

             3.   Date Required - The date EPCN is needed.

             4.   Type of Change - "Equipment" X entered if equipment change,
                  "Procedure" X entered if procedure change, "Other" X entered
                  for any other type of change.

             5.   References -  Identify referenced procedures, specifications, etc.

             6.   Change Requested By - Identify person requesting change.

             7.   Purpose of Change - Specify reason for change.

             8.   Description of Change - Describe change, and attach details,
                  specification, or drawings if necessary.

             9.   Effactivity - Effective date or serial number, etc., as
                  determined by Laboratory Operations.

             10.  Duration  or Extent of Use - "Permanent" X if change is per-
                  manent, "Temporary" X if temporary change only and indicate
                  date effectivity expires.

             11.  Areas Affected by Change - Indicate areas affected by  change
                  by marking X  in appropriate boxes.

             12.  Reviews and Approvals - Reviewers/Approvers sign and enter
                  date of review/approval.
               CONCURRENCES
DATE
IMPLEMENTATION
 PREPARED  BY:
         PAGE
    OF  2
 APPROVED  BY:
         DATE  ISSUED:

-------
                                                                 QMP No. 7.5
                                                                 Page 2 of 2
3.   FORM INSTRUCTIONS (continued)

          13.  QC/QA Manager - signifies approval and date approved.

          14. . Lab Branch Chief - Signifies approval and date approved.

          15.  Date - Date EPCN is initiated.

          16.  EPCN No. - Reference number assigned by Document Control.

          17.  Page 	of	-Page number.
                                       322

-------
                                                  QMP-tNO. 7.5
                                 Attachment No.  1
 EQUIPMENT/PROCEDURE   CHANGE  NOTICE
                                                             DATE
                                   EPCN NO.
                                    C*)
PAGE
                                  OF
1.  ORIGINATOR (Name)
                                          2. PHONE EXT.
                   3. DATE REQUIRED
                                          4. TYPE OF CHANGE
                                              Q EQUIPMENT
                      Q PROCEDURE     D OTHER
S.  REFERENCES
6. CHANGE REQUESTED BY (Name)
7. pURPOSl Ot
8. DESCRIPTION OF CHANGE (Attach details, specifications, or drawings if  necessary)
 9. EFFECTIVITY (Date or Other)
            10. DURATION OR EXTENT OF USE (See 9.)
            	D PERMANENT      D TEMPORARY —
 11. AREAS OF KSAPC AFFECTED BY THIS CHANGE
      D LOT      D "D      D CHEM     Q LAB      E QC/QA
      DHDT      ni*E      D«M      DOATA     QECTD
      D OTHER 	
                       DCSD
                                       12. REVIEWS AND APPROVALS
            REVIEWED BY
 A.
DATE
APPROVED BY
            F.
                                                                                              DATE
 B.
            G.
 C.
            H. '
 D.
 E.
 "'TirNOOPPROVED, PLEASE DISCUSS DETAILS ON'REVERSE SIDE)
 ALL REVIEWS AND APPROVALS HAVE BEEN RECEIVED
    DOCUMENTED
            13.  QC/QA MANAGER
                                                                                          DATE
FlfiAL IMPLEMENTATION
THE PROVISIONS
IMPLEMENTED
OF THIS EPCN ARE APPROVED
AND
HEREBY
14. LAB BRANCH
CHIEF
DATE
      FORM SO.  7.5;  1-31-75
     323    DISTRIBUTION   ORIGINATOR
                          QC/QA OFFICE
                  DIVISION FILES
                  AREAS AFFECTED

-------
EPA QUALITY MANAGEMENT  PROCEDURE
   QMP NO.
      7.6
                                                                  REVISION  DATE
SUBJECT:
         FORMS  INSTRUCTION - QMP CHANGE AND REVISION SUMMARY
   1.   FORM NUMBER:  7.6:1-31-75.

   2.   FORM USE:  QMP 3.1.

        2.1  To provide a summary of changes and revisions  to QMP's.

   3.   FORM INSTRUCTIONS

        3.1  The paragraph numbers listed below coincide with the numerals
             in the blocks in Attachment No. 1.

             1.  EPCN Number - The Equipment/Procedure Change Notice file
                 number assigned by Document Control on QMP Form 7.5, which
                 accompanies all changes and revisions to  QMP's.

             2.  Date - Date as indicated on EPCN.

             3.  Procedure Number - QMP number of affected procedure.

             4.  Procedure Revision Date - Revision Date shown  in QMP.

             5.  Procedure Title - Subject Title of QMP.

             6.  Entered By - Name of QMP manual holder recording change.
               CONCURRENCES
DATE
     IMPLEMENTATION
 PREPARED  BY:
PAGE
                  OF
 APPROVED  BY:
         DATE  ISSUED:
                                       325

-------
                                                              QMP NO. 7.6
                                                              Attachment No. 1
                        QUALITY MANAGEMENT PROCEDURES


                         Change and Revision Summary
EPCN
Number
0)
Date
(*>
Procedure
Number
(3)
Revision
Date
(4-)
Procedure Title

C*>
Entered
By
CO
FORM NO. 7.6:  1-31-75
                                      326

-------
QMP NO. REVISION DATE
EPA QUALITY MANAGEMENT PROCEDURE 7.7
SUBJECT:
1. FORM
2. FORM
2.1
3. FORM
3.1












FORMS INSTRUCTION - REJECTION REPORT
NUMBER: 7.7:1-31-75
USE: QMP 4.2
The Rejection Report is used to document, identify and withhold
discrepant materials and equipment.
INSTRUCTION
The paragraph numbers listed below coincide with the numerals in
the blocks in Attachment No. 1.
1. Part Number - Enter drawing/specification number of item being
rejected.
2. Part Name - Noun description of item being rejected.
3. Supplier/Manufacturer - Name of supplier/manufacturer if
procured item.
4. Quantity Rejected - Number of items being rejected.
5. Date Rejected - Date items were rejected.
6. Contract - Contract number or code.
7. Purchase Order Number - Purchase Order identification number,
if procured item.
8. Receiving Report-Number - Receiving Report identification
number if procured item.
9. Item Number - Item number of discrepancies starting with
one (1) and progressing sequentially. Do not list more than
one type of discrepancy under one item number.
10. Discrepancies - Describe discrepancies, itemizing by type
of discrepancies.
11. Rejected by - Name of person inspecting and rejecting
discrepant materials/equipment.
12. Date - Date inspector signs off in (11) .
CONCURRENCES DATE IMPLEMENTATION
PREPARED BY
APPROVED BY
: PAGEj 	 OF 	 2_
: DATE ISSUED:

-------
                                                                 QMP No. 7.7
                                                                 Page 2 of 2
3.   FORM INSTRUCTION  (continued)

          13.  Supervisor Approval -  Inspector's  supervisor reviews rejection
               and signifies approval by  signing  here.

          14.  Date -  Date supervisor signs off in  (13).

          15.  Quality Assurance Approval - Quality Assurance reviews rejection
               and signifies approval by  signing  here.

          16.  Date -  Date Quality Assurance signs off  in  (15) .

          17.  Disposition - When designated authority  or material review
               board determines disposition of rejected material/equipment,
               check appropriate box  (if  "other"  specify type of disposition
               made).

          18.  Failure Analysis Required  - Check  box if a formal failure
               analysis  report is required, and upon receipt of the report
               enter identifying reference number.

          19.  Corrective Action - Statement of corrective action taken to
               preclude  recurrence of discrepancy.

          20.  Quality Assurance Approval - Quality Assurance Manager/Supervisor
               reviews Rejection Report and Corrective  Action statement and
               signifies approval by  signing here.

          21.  Date  -  Date Quality Assurance Manager/Supervisor signs off.
                                       328

-------
                                                QMP NO. 7.7
                                                Attachment No. 1
                            REJECTION REPORT
                                                NO. 11 101
 PART NUMBER
      (0
      PART NAME
                U)
                          SUPPLIER/MFR
                                       (3)
 REJECTED
QUANTITY^)
      CONTRACT

     	(fe)
            PURCHASE  ORDER NO.

            	(7)
     REC. REPORT  NO

     	(8)
  ITEM NO.
DISCREPANCIES
                           (to)
REJECTED BY
    00
   DATE
     0*)
SUPERVISOR APPROVAL   DATE
        05)           0*)
Q.A. APPROVAL  DATE     .
     (15)	Ob)
DISPOSITION
    (17)
USE AS  IS

RETURN TO SUPPLIER!
       CHECK  IF FAILURE  ANALYSIS REQUIRED  0*)

       FAILURE ANALYSIS  REPORT  NO	
OTHER  (SPECIFY)
             CORRECTIVE ACTION   (l9>
                             Q.A.  APPROVAL.
                                                             DATE  fcl)
 FORM NO. 7.7:  1-31-75
                              329

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
I. REPORT NO.  •
 EPA-650/4-75-024-a
                             2.
                                                           3. RECIPIENT'S ACCESSION-NO.
», TITLE AND SUBTITLE  '
 Guidelines for QA Programs for Mobile Source Emissions
 Measurement Systems  - Phase I-, Light Duty  Gasoline-
 Powered Vehicles - QA Guidelines
             B. REPORT DATE
                June 1975
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 Rod Pilkington, Tom  Kelly and Harold Wimette
                                                           8. PERFORMING ORGANIZATION REPORT NO.
i. PERFORMING ORGANIZATION NAME AND ADDRESS
 Olson Laboratories,  Inc.
 421 East Cerritos Avenue
 Anaheim, California   92805
             10. PROGRAM ELEMENT NO.
                1HA327
             11. CONTRACT/GRANT NO.
                68-02-1740
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA,  NERC, QAEML, QAS
 Research Triangle Park
 North Carolina   27711
             13. TYPE OF REPORT AND PERIOD COVERED
                Final
             14. SPONSORING AGENCY CODE
is. SUPPLEMENTARY NOTES  Thig report is one of  two volumes for Light Duty Gasoline-Powered
 Vehicles (Phase  I).   Other volumes are to  be issued for Phase  II Heavy Duty Diesel
     npR.. Phage ITT  Light-. Dul-y nipgpl -PowF>rpd \7phi rl pg r anri Phase IV Hpavy nnf-y riagnli
     TRACT           •        "                                                  ~
     es.

 Quality Assurance  guidelines for Light Duty Gasoline-Powered Mobile Source
 Emissions Measurement Systems are presented with the concept of a total
 Quality Assurance  System.  The guidelines  apply to Quality Assurance principles and
 techniques  in the areas of procurement, standards and calibration, test quality
 control, data Validation and corrective action.  Model Quality Management
 Procedures are presented to describe the relationships and responsibilities of
 the various organizational elements in accomplishing the quality functions.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                              COSATI Field/Group
 Quality Control
 Quality Assurance
 Quantitative Analysis
 Gas Analysis
 Emissions - Exhaust Gases
 Compliance Testing
 Air Pollution
 Mobile  Source Emission
 Testing
13H
14D
O7D
13B
18. DISTRIBUTION STATEMENT
 Unlimited
                                               19. SECURITY CLASS (ThisReport)

                                                        i f i ort	
                                                                          21. NO OF PAGES
20. SECURITY CLASS (Thispage)
  Unclassified
                                                                          22. PRICE
EPA Form 2220-1 (9-73)

-------