EPA-650/4-75-024-<
Environmental Monitoring Series
IDELINES
FOR QUALITY ASSURANCE PROGRAMS
FOR MOBILE SOURCE EMISSIONS
MEASUREMENT SYSTEMS:
PHASE II, HEAVY-DUTY DIESEL ENGINES -
QUALITY ASSURANCE GUIDELINES
I"
55
V
^
\
LU
C3
U.S. Environmental Protection Agency
Office of Research and l)cv
Washington, D. C. 204fiO
-------
GUIDELINE DISTRIBUTION RECORD
EPA-650/4-75-024-C "Guidelines for Quality Assurance Programs for
Mobile Source Emissions Measurement Systems: Phase II, Heavy-
Duty Diesel Engines - Quality Assurance Guidelines"
The series of documents, "Guidelines for Quality Assurance Programs tor
Mobile Source Emissions Measurement Systems," have been prepared and issued in a
revisable document-controlled format so that any future revisions or additions
may be distributed to the Guideline users. Individual copy numbers have
been assigned to each volume of the series. These numbers appear on this record
sheet and on the title page of each document. The user of these documents must
complete the "Distribution Record Card" and return it in the pre-addressed
envelope to the address shown below so that automatic distribution for future
revisions may be made directly to the user. Also, notice of any reassignment
of the documents by number and title to other individuals (by name, address,
and copy number) should be sent to the same address.
U.S. Environmental Protection Agency
Office of Mobile Source Air Pollution Control
2565 Plymouth Road
Ann Arbor, Michigan 48105
1 r* '^
ATTN: Quality Assurance. COPY No. °
(cut along dotted line)
DISTRIBUTION RECORD CARD
Guideline
User Date
Last Name First Middle Initial
Title
Address
to Send
Future Street
Revisions btreet
and
Additions
City State Zip Code
If address is a firm (fill in)
Firm
I have received copy No. of EPA-650/4-75-024-C "Guidelines for Quality
Assurance Programs for Mobile Source Emissions Measurement Systems:
Phase II, Heavy-Duty Diesel Engines - Quality Assurance Guidelines"
Please send me any revisions and additions of this volume.
-------
EPA-650/4-75-024-C
GUIDELINES
FOR QUALITY ASSURANCE PROGRAMS
FOR MOBILE SOURCE EMISSIONS
MEASUREMENT SYSTEMS:
PHASE II, HEAVY-DUTY DIESEL ENGINES •
QUALITY ASSURANCE GUIDELINES
by
Harold Wimette, Rod Pilkington, and Tom Kelly
Olson Laboratories , Inc . 1 G o
421 East Cerritos Avenue
Anaheim, California 92805
Contract No. 68-02-1740
ROAP No. 26BGC
Program Element No. 1HA327
EPA Project Officers:
R. C. Rhodes
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
IS. 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
scries. 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-
Irations of pollutants in the environment and/or the variance of pollutants
as n (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-C
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.
iii
-------
EXECUTIVE SUMMARY
Quality Assurance guidelines for heavy duty diesel 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 organizations.
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 heavy duty diesel engines 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 guide-
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 Heavy Duty
Diesel Engines 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 2-12
3 MEASUREMENT SYSTEM ANALYSIS 3-1
3.1 Applicable Federal Register Procedures 3-1
3.2 Elements of a Measurement System for Heavy Duty
Diesel Engine Emission Measurement 3-2
vii
-------
4 GUIDELINES FOR PERFORMANCE AUDITS AND MAINTENANCE
PROCEDURES 4-1
4.1 Performance Audits 4-1
4.1.1 System Performance Characteristics 4-3
4.1.2 Acceptance Criteria 4-4
4.1.3 Frequency Checks 4-4
4.1.4 Audit Equipment 4-9
4.1.5 Procedures for Independent Performance Checks . . . 4-9
4.1.6 Reporting and Corrective Action Procedures 4-11
4.2 Preventive Maintenance 4-12
4.2.1 Preventive Maintenance Procedures 4-13
4.2.2 Preventive Maintenance Action 4-16
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 5-8
5.2.2 Recording Calibration Results 5-8
5.2.3 Recording Maintenance Actions 5-15
5.2.4 Reporting Unacceptable Results 5-21
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
-viii
-------
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
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 HEAVY DUTY DIESEL ENGINES 7-1
7.1 Variables Associated with the Measurement of
Smoke Emissions 7-2
7.1.1 Dynamometer Operation 7-2
7.1.2 Diesel Engine Operation 7-3
7.1.3 Smokemeter and Recorder 7-4
7.1.4 Ambient Conditions 7-5
7.1.5 Chart Reading 7-6
7.1.6 Computer 7-8
7.2 Variables Associated with the Measurement of
Gaseous Emissions 7-8
7.2.1 Determination of Brake Horsepower 7-9
7.2.2 Measurement of Exhaust Flow 7-10
7.2.3 Determination of Exhaust Emission Concentrations. . 7-10
IX
-------
7.3 The Engine as a Source of Variability 7-13
7.4 Measurement of Variability in Emission Measurement
Systems 7-14
7.5 Quality Assurance and Test Variability 7-17
8 QUALITY ASSURANCE (ON-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-2-2
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
4-1 Daily Start-up Checksheet - Heavy Duty Diesel Engine. 4-15
4-2 Preventive Maintenance Checklist Smokemeter, Six
Months 4-17
4-3 Preventive Maintenance Checklist Laminar Flow Element 4-18
4-4 Preventive Maintenance Checklist Analysis System -
CO, NO, O , and HFID Weekly 4-19
4-5 Preventive Maintenance Checklist Analysis System,
Monthly 4-20
4-6 Preventive Maintenance Checklist G.E. Engine
Dynamometer, Monthly 4-21
4-7 Preventive Maintenance Checklist Individual
Instruments, Monthly 4-22
4-8 Maintenance Log Form 4-24
5-1 Typical Receiving Inspection Form . 5-11
5-2 Instruction for Receiving inspection Report 5-12
5-3 Calibration Tags 5-13
5-4 Calibration Control Punch Card 5-14
5-5 Calibration History Evaluation 5-16
5-6 Analyzer Curve Generation Data 5-17
5-7 Monthly Dyno Calibration Log 5-18
5-8 Gas Analysis Report 5-19
5-9 Equipment Repair Authorization 5-20
5-10 Rejection Report 5-23
5-11 Failure Analysis Report 5-24
5-12 Corrective Action Request 5-26
5-13 Corrective Action Request - Flow Chart 5-27
5-14 Performance Audit Summary Sheet 5-29
5-15 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 Smoke Opacity and Engine Speed Traces from a Federal
Smoke Test 7-7
7-2 Factors Resulting in Smoke and Unfinished Combustion. 7-15
xii
-------
LIST OF TABLES
Table No. Page
3-1 Emission Regulations for Heavy Duty Engines 3-4
3-2 Federal Emission Test Procedure - Smoke and Gaseous
Emissions 3-5
4-1 Performance Audits (Typical) - for a Heavy Duty Diesel
Emission Measurement System 4-2
4-2 Federal Register Specifications - Heavy Duty Diesel . 4-5
4-3 Methods of Monitoring Variables 4-10
4-4 Preventive Maintenance Schedule - Major Components. . 4-16
5-1 Example of Quality Plan Activity Matrix - Procurement
(Receiving - Fuels) 5-7
5-2 Example of Quality Plan Activity Matrix - Procurement
(Receiving - Gas Mixtures) 5-9
5-3 Summary of Forms Referenced in Volume I 5-10
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 Major Sources of Error in Smoke Measurements
Instrumentation 7-4
7-2 Participants' Analyses of Bottled Gases 7-18
7-3 Analyses of Engine Exhaust 7-19
7-4 Engine Test Results, Carbon Monoxide, ppm 7-20
7-5 Engine Test Results, Nitric Oxide, ppm 7-21
7-6 Summary of Smoke Test Variables and Methods used for
their Control 7-22
7-7 Summary of Gaseous Emissions Test Variables and
Methods used for their Control 7-23
xiii
-------
-------
Section: 1(HD)
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 the administration of a Quality Assurance Program for
air measurement systems resulting from the implementation of the Clean
Air Act. Standards for the emissions from light and heavy duty mobile
sources have been promulgated and procedures published for the measure-
ment of their emissions and certification. Quality assurance guidelines,
however, have not been previously specified for these mobile source
emission measurement 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 heavy
duty diesel engines (Phase II). Guidelines for the other phases are
reported in separate documents.
1.1 OBJECTIVE AND SCOPE OF GUIDELINES
The measurement system for heavy duty diesel engines 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
assurance systems to the total testing process to assure the validity
and reliability of the tests and the resulting test data.
-------
Section: 1(HD)
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 heavy duty engine 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 background, objective 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 heavy duty
diesel engines is presented in a matrix format.
1.2.4 Section 4 Guidelines for Performance Audits and
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(HD)
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 with representative forms, a description of the manuals, data
recording, and failure analyses used 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 Heavy Duty Diesel Engines
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 heavy duty diesel 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 B-2 include a glossary of terms and a list of abbreviations
commonly used in the measurement system.
-------
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 1 of 14
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 heavy duty engines.
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(HD)
Revision: 0
Date: June 1975
pa.ge 2 of 14
Laboratory Operations - Performs emission tests on heavy
duty engines 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 preventions
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.
10
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 4 of 14
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 stan-
dards from outside suppliers. It must be remembered, however, that
assignment of the responsibility for total quality assurance to a partic-
ular section does not relieve the other functional groups from perform-
ing 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
Purchasing in determining qualified suppliers and to assure quality
requirements 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 avail-
able from only a single source. An actual on-site
supplier review is usually unnecessary; however, respond-
ing 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 instru-
ments can be avoided by careful procurement document
review.
o Receiving Inspection. After the supplier is selected,
Purchasing issues a purchase order including all the
11
-------
DIRECTOR
QUALITY
ASSURANCE
PROCUREMENT
CONTROL
TEST
QUALITY
CONTROL
CORRECTIVE
ACTION
AUDIT
STANDARDS
AND
CALIBRATION
DATA
VALIDATION
DEFICIENCY
REVIEW
DATA
ANALYSIS
STATISTICAL
ANALYSIS
SYSTEMS
DEVELOPMENT
COMPUTER
OPERATIONS
ADMINISTRATIVE
SERVICES
CONTRACTS
PURCHASING
FAC LITY
SERVICES
RECORDS
MANAGEMENT
TRAINING &
CERTIFICATE
—
1
LABORATORY
OPERATIONS
TEST
OPERATIONS
VEHICLE
TEST
FUEL & GAS
STORAGE
SUPPORT
OPERATIONS
-,:.::. "i,,_
CHEMICAL
ANALYSIS
EQUIPMENT
SERVICES
TEST
SCHEDULING
^0 C3 50 ^3
iQ ft < O
(B ID H- rt
H- o'
U) OS
0 ..
Figure 2-1. FUNCTION/RESPONSIBILITY CHART
C| O M
S —
2 a
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 5 of 14
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. Fuels, pure gases, and other chemicals
2. Calibration gases
3. Filtering or gas absorbing material (i.e., dryrite,
ascarite, charcoal, etc.)
4. Fuel Flow Meters
5. Replacement parts for calibration, analytical and/or test
equipment
6. Analytical instruments or systems
7. Engine Dynamometers
8. Air Measurement Systems
9. Any sampling equipment used in the analytical process,
such as tubing, flow controllers, meters, pumps, valves,
flowmeters and sample temperature controllers
10. Computer systems
13
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 6 of 14
11. Temperature and pressure measuring and controlling items
12. Chart recorders and chart paper
13, Standards and calibration test instruments
14. Engine diagnostic equipment
15. Fuel conditioning system {Heat exchangers, surge tanks,
fuel pumps, etc., as required)
16. Special instrumentation/equipment designed in-house and
purchased from outside suppliers (i.e., engine inlet air
plenum, filter holders, exhaust systems, etc.)
17. Standard reference materials (SRM) such as NBS Certified
weights, smoke capacity filters, and calibration gas
mixtures
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 stan-
dards can be ordered without analysis or with a specified "make" toler-
ance (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 and might require purchase of NBS
standard reference material or gravimetric blend.
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
14
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 7 of 14
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 proce-
dures. 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 para-
meters must be carefully studied for revision of quality control proce-
dures. 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 stan-
dards and calibration for measuring devices. 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 emis-
sion 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.
15
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 8 of 14
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 certi-
fied gas mixtures, for emission testing. Gravimetric blends have been
prepared by the EPA for carbon monoxide, carbon dioxide, nitric oxide,
and propane mixtures.
The following standard reference gases are available from NBS:
SRM's 1665 - 1669 Propane in air
SRM's 1673 - 1675 Carbon dioxide in nitrogen
SRM's 1677 - 1681 Carbon monoxide in nitrogen
SRM's 1683 - 1687 Nitric oxide in nitrogen
Any 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.
Other instruments which require calibration are:
1. Engine diagnostic equipment
2. Temperature recorders
3. Barometers
4. Hygrometers
5. Engine dynamometers
6. Recorders or output measuring devices.
7. Smokemeters
8. Fuel flow meters
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.
16
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 9 of 14
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
recommendations are presented to the Laboratory Management. Quality
Assurance has the responsibility for measuring the effect of imple-
menting these corrective 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 exten-
sive experience in 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 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. An example of
a Data Validation network is described and illustrated in QMP No 6.6 HD,
Appendix C of this report.
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
17
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 10 of 14
(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 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 Manage-
ment should have the authority to shut down any part of the testing
system producing invalid test data until the non-conforming condition is
corrected.
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 consider-
ation 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 equipment or instruments
that are required to undergo reliability testing. These
descriptions will include specifications for both quality
and reliability.
18
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 11 of 14
4. Concise specifications for the performance of tests,
including meticulous attention to the ambient condi-
tions - 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 occur-
rence of equipment failure. There is a further discussion of failure
reporting and analysis in Section 5.
Configuration and Documentation Control in a testing labora-
tory 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
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.
19
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 12 of 14
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, purchasing,
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.
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, equipment
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.)
20
-------
Section: 2(HD)
Revision: 0
Date: June 1975
Page 13 of 14
7. Controls and maintains inventory of all parts, supplies,
and equipment used in the normal operation of the emis-
sion 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
statistical programs to assist Quality Assurance in the evaluation and
improvement of test data quality.
1. Develops and processes computer programs for the reduc-
tion of test data to provide emission results on a grams
per mile basis for carbon monoxide, hydrocarbons, carbon
dioxide and nitric oxide.
2. Maintains all test data in a data file.
3. Provides statistical analysis for Quality Assurance
requirements 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.
21
-------
Section: 2(HD)
Revision; 0
Date: June 1975
Page 14 of 14
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, the super-
vision of personnel, equipment and vehicles utilized in the performance
of emission testing and implementation of in-process quality control
checks.
Quality Management Procedure Number 2.3 in Appendix C details
the primary functions of the Test Operations and Support Operations
departments.
22
-------
-------
Section: 3(HD)
Revision: 0
Date: June 1975
Page 1 of 5
Section 3
MEASUREMENT SYSTEM ANALYSIS
A total Measurement System can be defined as an orderly
arrangement consisting of the analytical method, the test sampling
procedure, the instruments or analyzers, the supporting functions, the
management organization and the technicians or personnel involved in
performing specific functions within the system. Applying this defini-
tion to the measurement system for heavy duty diesel engine emissions,
the process is composed of:
o The test procedure defined by the Federal regulations
o The preparation of the diesel engine for the emissions
test
o The exhaust emission sample transfer and analytical
console consisting of NDIR instruments for the measure-
ment of carbon monoxide (CO) and nitrogen oxides (NO ),
a heated flame ionization instrument (HFID) for measurement
of hydrocarbons (HC), a full flow optical smokemeter
mounted near the exhaust plume for measurement of diesel
smoke.
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 quality 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:
25
-------
Section: 3(HD)
Revision: 0
Date: June 1975
Page 2 of 5
Date Vol. No. Page
1. November 15, 1972 37 221 24293-24314
2. June 28, 1973 38 124 17149-17162
3. October 31, 1973 38 209 30082
4. November 21, 1973 38 225 32138
5. November 23, 1973 38 225 32258
6. February 27, 1974 39 40 7562-7568
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 heavy duty diesel engines appear in
Table 3-1.
In addition to the Federal Register the following Society of
Automotive Engineers publications were used:
o SAE J177 - Measurement of Carbon Dioxide, Carbon Monoxide
and Oxides of Nitrogen in Diesel Exhaust
o SAE J244 - Measurement of Intake Air or Exhaust Gas Flow
of Diesel Engines
o SAE J215 - Continuous Hydrocarbon Analysis of Diesel
Emissions
o SAE J255 - Diesel Engine Smoke Measurement
3.2 ELEMENTS OF A MEASUREMENT SYSTEM FOR HEAVY DUTY DIESEL ENGINE
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 and the
related tasks the measurement system was divided into the following
operations.
o Engine-dyno preparation
o Engine-dyno checkout
o Engine test cycle preconditioning
o Smoke measurement
o Smoke emission test cycle
o Intake air measurement
o Measurement of fuel flow
-------
Section: 3(HD)
Revision: 0
Date: June 1975
Page 3 of 5
o Gaseous Emission test cycle
o Measurement of gaseous emissions
o Data collection and reduction
A summary of the emission measurement procedure for both smoke
and gaseous emissions is presented in Table 3-2. 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
intended to include every detail required for this measurement. The
information discussed in this table consists of:
o A brief description of the tasks
o Applicable Federal Register paragraphs
o Applicable EPA, Ann Arbor, test procedure numbers
o Specification and tolerances included in the Federal
Register, and from engineering practices such as the SAE
recommended practices
o Quality provisions
o Invalid test (determination)
o Corrective action required
o Training and skill level required
27
-------
Section: 3(HD)
Revision: 0
Date: June 1975
Page 4 of 5
Table 3-1. EMISSION REGULATIONS FOR HEAVY DUTY ENGINES
Subpart I - Engine smoke exhaust emission regulations for new diesel heavy
duty engines
APPLICABLE
REVISIONS* SECTION TITLE
2, 6 85.802 Definitions
85.803 Abbreviations
2 85.874-9 Test procedures
85.874-10 Diesel fuel specification
85.874-11 Dynamometer operation cycle for smoke
emission tests
85.874-12 Dynamometer and engine equipment
85.874-13 Smoke measurement system
85.874-14 Information to be recorded
2 85.874-15 Instrument checks
85.874-16 Test run
85.874-17 Chart reading
85.874-18 Calculations
Subpart J - Engine exhaust gaseous emission regulation for new diesel heavy
duty engines
2, 3 85.902 Definitions
85.903 Abbreviations
2 85.974-9 Test procedures
85.974-10 Diesel fuel specifications
85.974-11 Dynamometer procedures
85.974-12 Dynamometer and engine equipment
2 85.974-13 Sampling and analytical methods
85.974-14 Information to be recorded
85.974-15 Calibration and instrument checks
2 85.974-16 Test run
85.974-17 Chart reading
2, 3, 4 85.974-18 Calculations
2) 28 June 1973, 3) 31 October 1973, 4) 21 November 1973,
5) 23 November 1973, 6) 27 February 1974
28
-------
Table 3-2. FEDERAL EMISSION TEST PROCEDURE - SMOKE AND GASEOUS EMISSIONS
to
mCCZDUW bit TASK
A. tngine-Dyne Preparation
c
C. Engine tast cycle
preconditioning
0. seeks SMeaursmsnt. and
tsit cycL*
E. InUks six measiuwat
P. NsasuiBmsnt of Fuel Tlow
C. Gaaeous taission test
cycle
H. Hejsursewnt of HC, CO
MO and 0 in diesel
eahaust
1. B*U collection and
i eduction
The rncjinr is installed on the tost bed
*nd coupled to the dyno. Auxiliary
«(|mpmtj,t, oftlidust and inlet air
systems arc installed or. Ui« engine.
lo*d condition* to check for proper
operation, maka final adjustMnt and
determine that engine spaed, torque
and fuel flow Met Hfr. specifications.
Engine is operated at maxiau* rated
horsepower tor 10 sin. oc until
temperature and oil pressure isach
The war* engine i* run through three
sequences of idle, acceleration,
art lugging mod*a.
In order to determine emissions on
* msss basis, the KSBB of intake
sir eust M Measured aa a function
cycle. ,
In order to determine the total soss
of the BKhaust, tns mass ot fuel
•uat be measured per unit time.
Mass Ex • Haaa Mir + Mass Fuel,
The teat Is designed to determine the
bzake-spacKic emissions of hydro-
caj'bona, carbon monoxide and oat Idea of
nitrogen when an engine is oparated
through a cycle which oonaiati of three
idle modes and live power modes at each
of two speeds which span the typical
ftoasuremvflts are made of the exhaust
concentrations during the last minute
of vachmode. 03 is msMttisd only Lf
required lor correction of the HTIO
reading.
FOE complete list of required in tarna-
tion sea 85.974-15. Ambient condition
fuel and air flow, analyser output
are recorded/ validated and the data
reported in grams per brake horsepower.
EPA TP WO.
8 i. 874-1 J
85. 914-12
TP-750
TP-7SO
65.874-9
TP-731
85.874-9
85. 874 -18
TF-7M
85. 974-13 -
SPEgTFlCATIOMS S*E OR ENC, PRACT.
HSAPC Advisory Circular Ho. 22A,
•hipping instructions swt pre-
paration or h«avy duty engines for
EPA testing.
•ucb as piston slap, knocks, tap
,nt gear noises that indlcats irregu-
larities in engine operation.
Test cycle repeated until 3 consecu-
tive valid teats have been achieved.
opecity filters checked by EPA avery
6 Months.
A P accuracy of 0.5 percent of idle
flow. Leakage test to within tO.S
percent of idle flow.
tl percent of observed reading.
X
£AE J100J
Calibrate Monthly, 5-point calib.
±2 percent, weekly checks. 2 fir.
sla. warmup. span 15-95 pe-rwnt.
Sample dxyar. HC range 10-6, DOO fPHC
response 90 psrcsnt - iO sec. g
Detector and ov«a teap. cont. t^ t,
Rang* 15O-200<>C. Probe - •ultiholu,
BO percent icro*« cxhK't fit*, hi»tlv transport
time checked each
TEST 1KVAL1D
mpxoper installa-
tion reeulting in
out of tol. restric-
tion. Incorrect Hti .
spec. used.
detected. Inability
to achieve proper
Inlet and/or outlet
restrictions Mi.
Spec, not achieved
by engine or dyno,
Engine not Cully
warned pcLor ta
start of test.
Incorrect time,
load or apeed.
Escaasive Ins-
trument drift.
l«aks on change in
&t. lead out
calibration
will invalidate
tsst.
Failure to record
fuel flow during
any sods. Flow-
mater malfunction
detected .
Mode tina in **ce»»
of 10 minutes, rpm,
toiqu* or time out
of tolerance.
Span or zero shift.
Incorrect sample
time.. Incorrect
span deflection.
Inadequate or too
low tamp. cent.
Calib. overdue
Detection of error
or abnormality at
thle point requires
invalidation of test
CORRECTIVE ACTION
Correct installation
and repeat tsst.
Determine correct
Shut down engine and
notify supervisor
and Hfr. Rep. Detar-
eiins restriction
problem snd correct.
Repeat tast
Repast test
Correct inst.
problem.
Correct problem and
Repeat test
mode 1.
Correct spen and
saro shift and
check point.
Repeat test cycle
•s required. After
correcting out -of -
spec, condition.
Repeat complete
test sequence
InH'^v?"
Heavy duty engine
•cchtr.ie with
knowledge of dyno
tlon procedures.
Engine dyno operator
with working knowledge
of h*avy duty engine
operations and perfot
Ae above in B
mater •Kprtienc* or
training.
Training in us* and
installation of LFE
and m&nometera.
Instruction in opera-
tion of fuel flow
meter.
in.lructlc*. in opcr.-
tion of dyno controls.
Training in operation
of enalytical console.
Kna*Ledgc of Cilib.
mslitt. snd nlnor ad).
Training in ilata
validation, chart
r**dinq, data
inspection, and
r*di*ctlon. computer
r*guir*d.
(I)
rt <
(0 H'
» M
•fl W
fl> (0
o
O 3 »
Ul
c_i o oj
C ^
--J
Ol
-------
-------
Section: 4(HD)
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
are acceptable. The occurrence of invalid data, non-acceptable instru-
mentation 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 performance
audits include the collection of normal operation data (i.e., emission
data, 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.
A list of typical performance audits with associated checks
are given in Table 4-1.
Audits may be independent of or in conjuction with normal
quality checks. Independence can be achieved by having an operator not
normally assigned to the emission system in question perform the audit.
An alternate method is to provide a standard reference sample of unknown
33
-------
Table 4-1. PERFORMANCE AUDITS (TYPICAL) - FOR A HEAVY DUTY DIESEL EMISSION MEASUREMENT SYSTEM
AUDIT
CATEGORY
Instrumentation
Operator -
Dyno/Analytical
Equipment -
Sampling System
Inlet Air System
Equipment -
Fuel Handling
Equipment -
Auxiliary
Data Processing
CHECKS
1. Observe calibration procedures.
2. Check calibration records and tags to insure proper intervals are
followed.
3. Check maintenance records and all log books.
4. Perform or observe an emission test using test cell equipment
instruments. Check for system drift, noise, gain, response.
Check for the reproducibility of the span and sample.
1. Observe the technician performing the test. Check that the forms
are being properly filled out during the test.
2. Check conformance to test procedure requirements.
3. Ask the technician questions concerning his job to determine
that he possesses the required skill level.
1. Perform leak check.
2. Inspect filters, sample conditioning items for proper maintenance.
1. check for proper calibration and maintenance of fuel measurement
system.
2. Take sample of fuel from system for analysis. Note color of
diesel fuel.
1. Check calibration, maintenance and operation of all auxiliary
equipment located or used in the test cell such as portable
tachometer, engine diagnostic, and safety alarms and interlocks.
1. Check computer alignment with instrument meter output.
2. Check for proper data validation procedures and that proper
authorization for the test sequence is being followed.
3. Select data files at random and check all forms and records.
4. Reduce all data manually to check its accuracy.
5. Check data transfers and test engine information for correctness.
vQ ft < O
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 3 of 24
value (concentration) to the operator and request that he analyze and
report the concentration values following normal operating 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 Method (Procedure) used to perform check
o Corrective action requirements
o Recording of audit results
4.1.1 System Performance Characteristics
Factors to be considered in the establishment of a system's
quality characteristics are quality of specifications, quality of
conformance and time-oriented factors such as availability, reliability
and maintainability. Specifications must have adequacy, attainability,
and compatability with the environment of the testing laboratory.
Quality of conformance is the result of numerous variables: test equip-
ment, tools, supervision, workmanship, etc. Availability is measured by
the extent to which a user can obtain service when he wants it« Reli-
ability is classically 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 maintenance of long-life systems, by planned preventive
maintenance and unscheduled 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 (Ref. 4-3.) After
identifying the critical characteristics, the analysis should be extended
to determine a means of controlling 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).
35
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 4 of 24
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.
4.1.2 Acceptance Criteria
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 Heavy Duty Diesel Emission Measurement
System are related to the requirements of the Federal Register (see
Table 4-2). An expanded version of this table is shown as Table 3-2.
Acceptance criteria will vary somewhat from one laboratory to
another, however, they must never be more lenient than those promulgated
in the Federal Register. The above referenced Tables are not necessarily
a complete list of acceptance criteria. Each laboratory should first
list the characteristics of their particular system and refer to these
Tables for the corresponding criteria. If one cannot be found for a
particular characteristic, good engineering judgment must be used to
determine the acceptance criteria, keeping in mind the effect it may
have on overall test variability.
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 improved.
4.1.3 Frequency of Checks
Performance audit intervals are usually conducted on a non-
routine basis as required by the particular system. Frequent audits are
required with new equipment, procedures and personnel. As experience is
gained with the system, audits may be performed less frequently.
However, they are necessary for a functional quality program and failure
to perform the audits may result in higher rate of invalid tests or less
reliability of the emission data.
36
-------
Table 4-2. FEDERAL REGISTER SPECIFICATIONS
HEAVY DUTY DIESEL EMISSION MEASUREMENT
SUBPART I
Section: 4(HD)
Revision: 0
Date: June 1975
Page 5 of 24
REFERENCE
PARAGRAPH
PROCEDURE OR EQUIPMENT DESCRIPTION
SPECIFICATION OR TOLERANCE
85.802
(J).902
Definition - Zero (O) hours
1974 Model Year
Point after assembly line operations and adjust-
ments and before 1 additional operating hour has
been accumulated.
85.874-1
Smoke Exhaust Emission Standard
Note: For information purposes only,
:his does not directly relate to test-
ing specifications, however, it could
be referenced to as an engine variable
and specification.
Shall not exceed:
20 percent during engine acceleration mode
15 percent during engine lugging mode
50 percent during the peaks in either mode.
85.874-10
(J).974-10
Diesel Fuel Specifications (Testing)
(See Section 85.874-10 (c) for fuel
used for service hour accumulation).
Item
Octane
Distillation range
IBP °F
10 percent point F
50 percent point °F
90 percent point F
EP °F
Gravity, API
Total sulfur present
ASTM Test
Method No.
D613
D86
Fype 1-D Type 2-D
48-54 42-50
D287
D129 or
D2622
Hydrocarbon composition D1319
Aromatics, percent
Paraffins, Napthenes,
Olefins
Flash point, F (Min) D93
Viscosity, centistokes D415
330-390
370-430
410-480
460-520
500-560
40-44
0.05-
0.20
8-15
340-400
400-460
470-540
550-610
580-660
33-37
0.2-
0.5
27 (Min)
Remainder Remainder
120 130
1.6-2.0 2.0-3.2
85.874-11
Cycle for Smoke Emission Tests
Note: Engine must be warmed up and
preconditioned according to 85.874-
16(c). Engine may be run for more
than three cycles as no limit has
been placed on number of cycles or
the time period for completing the
cycles. Data must be reported for
three successive cycles
Idle
Time
Speed
Load
5 to 5.5 seconds
recommended low idle
minimum obtainable
Acceleration
i - 200 ± 50 RPM within 3 sec. above low idle
ii - 85-90 percent rated speed in 5 ± 1.5 sec.
linear 1 100 RPM
iii - speed of maximum rated torque or 60 percent
of rated speed (whichever is higher)
within ± 50 RPM
iv - 90-100 percent rated speed in 10 f 2 seconds
Lugging Mode
i - Max. hp ? rated speed
ii - max. torque or 60 percent of rated speed,
whichever is higher in 35 t 5 seconds -
linear rate t 100 RPM
Repeat cycle 3 times
85.874-12
(J).974-12
a) Dynamometer Specifications
NOTE: This specification is inade-
quate due to the variable range
of engines. It would be diffi-
cult to write any one specification
to cover all engine families.
Adequate characteristics to perform the test
cycle.
37
-------
Section:
Revision:
Date:
Page 6 of
4{HD)
0
June 1975
24
Table 4-2. FEDERAL REGISTER SPECIFICATIONS
HEAVY DUTY DIESEL EMISSION MEASUREMENT (Continued)
SUBPART I (Continued)
REFERENCE
PARAGRAPH
PROCEDURE OR EQUIPMENT DESCRIPTION
SPECIFICATION OR TOLERANCE
85.874-12
(J).974-12
(Cont'd)
(b) Engine Cooling System
NOTE: Engine temperature
(c) Exhaust System
(d) Engine Air Inlet System
Capable of maintaining the engine at normal opera-
ting temperature during the cycle. Non-insulated
15 t 5 feet from the exhaust mainifold, or cross
over junction. Exhaust back pressure must be within
1 0.2 inch Hg. of the upper limit at maximum rated
hp. as established by manufacturer.
Terminal 2 feet of exhaust pipe. Circular cross
section straight, cut off squarely with a
diameter in accordance with rated horsepower
Maximum rated
horsepower
Less that
101 - 200
201 - 300
301 or more
101
Exhaust Pipe
Diameter (inches)
2
3
4
5
Inlet restriction must be within ± 1 inch of water
of the upper limit for the engine operating
condition which results in maximum air flow as
established by the manufacturer.
85.874-13
Smoke Measurement System
1. Continuous reading.
2. Full-flow light obscuration meter.
3. Light beam right angle axis to plume.
4. Light source - incandescent lamp operated not
less than 15 percent of mfg. spec.
5. Beam collimated to diameter of 1.125 inches.
6. The angle of divergence of the collimated beam
shall be within 4° included angle.
7. Detector sensitivity as restricted to the visual
range and comparable to that of the human eye.
8. A collimating tube with apertures equal to
the beam diameter is attached to the detector
to within 16° included angle;
Recorder
1. Continuous, variable speed 0.5 to 8.0 inches/
min. with automatic marker or chart paper
equivalent to 1 sec. intervals. Full scale
response = 0.5 sec.
2. Recorder scale, linear, 0-100 percent capacity
with 1 percent resolution.
85.874-13
Recorder For RPM and Torque Measurement
RPM resolution - 30 RPM
Torque Resolution - 10 Ib.-ft.
85.874-13
Smoke Measurement System Installation
Distance from the optical centerline to the exhaust
pipe outlet shall be 5 t 1 inch with the full flow
of the exhaust stream centered between the source
and detector apertures and on the axis of the light
beam. Instrument stabilization time = 15 minutes.
85.874-15
Calibration of smokemeter - deviations
from nominal opacity of the filter.
NOTE: Opacity filters used for cali-
bration should be traceable to an
acknowledged reference standard such
as MBS.
Deviations greater than 1 percent shall be
corrected.
38
-------
Table 4-2. FEDERAL REGISTER SPECIFICATIONS
HEAVY DUTY DIESEL EMISSION MEASUREMENT (Continued)
SUBPART I (Continued)
Section: 4(HD)
Revision: 0
Date: June 1975
Page 7 of 24
REFERENCE
PARAGRAPH
85.874-16
85.874-16
85.874-16
PROCEDURE OR EQUIPMENT DESCRIPTION
Test Cell Ambient Conditions
Engine Preconditioning
Smoke Measurement Drift Test
Zero or Full Scale
SPECIFICATION OR TOLERANCE
Inlet Air = 68-86°F
Bar = 28.5 - 31.0 Hg
10 min 8 max. rated hp.
± 2 percent Opacity
SUBPART J
85.974-1
85.974-11
85.974-13
85.974-15
Exhaust gaseous emission standard
for 1974 model year diesel engines
Heavy duty diesel test cycle
Sampling and Analytical Methods
o Carbon Monoxide and Nitric Oxide
Measurement
o Hydrocarbon Measurement
o Intake Air Flow
Calibration and Instrument Check
HC + NO = 16 grams per brake horsepower hour
CO = 40 grams per brake horsepower hour
Percent
Mode No. Engine Speed Load
1 Low Idle 0
2 Intermediate 2
3 do 25
4 do 50
5 do 75
6 do 100
7 Low Idle 0
8 Rated 100
9 do 75
10 do 50
11 do 25
12 do 2
13 Low Idle 0
Allowable deviation during last minute of each
mode.
Speed i 50 RPM
Torque = ± 2 percent of max torque at test speed
SAE recommended practices No. J177 Sections 2.1 and
2.2
SAE recommended practices No. J215 Sections 2.1 and
2.2
SAE recommended practices No. J244
SAE-RPN J177 Section 2.3.1
SAE-RPN J215 Sections 3 and 7
39
-------
Section: 4(HD)
Revision: 0
Date; June 1975
Page 8 of 24
Table 4-2. FEDERAL REGISTER SPECIFICATIONS
HEAVY DOTY DIESEL EMISSION MEASUREMENT (Continued)
SOBPART J (Continued)
REFERENCE
PARAGRAPH
PROCEDURE OR EQUIPMENT DESCRIPTION
SPECIFICATION OR TOLERANCE
85.974-16 Test Run Conditions
NOTE: Time of modes should give some
tolerance to allow for time measurement
inaccuracies.
NOTE: Test is repeated until valid test
is run, number of tests not specified.
Inlet Air
Fuel
Bar
Warm Up
Time Each Mode
Allowable Time
Between Modes
Analysis Time
Engine Speed Load
Changes
Check and Reset
Zero and Span
End of Test
Full scale change greater than 2 percent voids
test
68-86°F
100 ± 10°F
28.5 - 31.0 in. hg.
10 min. hp - rated speed or
until T-P equilibrium
10 minutes
10 minutes
Last 5 minutes each mode
1st minute each cycle
Second idle mode (No. 7)
40
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 9 of 24
4.1.4 Audit Equipment
In most cases special equipment is not required for an audit
function in mobile source emission testing. Audits are usually per-
formed by observation, validation, calculation, checking records and
actual use of the emission equipment. An auditor may request primary
standards or measurement equipment from the calibration or standards
groups for audit purposes. Equipment, if used in the audit must have
unquestionable reliability as for example, thermometers certified by NBS
for checking ambient temperatures and thermocouples, and recently calibrated
gauges. Audit equipment required for checking a characteristic must be
clearly specified in the documented audit procedure.
4.1.5 Procedures For Independent Performance Checks
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 con-
tained in Appendix H of Reference 6-1. Various applications of these
charts are shown in Section 6.0. These checks should be made indepen-
dent 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 supplied 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 4-3 summarizes some of the more important vari-
ables and how they can be monitored. A discussion of test variability
appears in Section 7.
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 (Ref. 4-4) can
provide a basic insight into the frequency of performance audits, and
statistical sampling techniques can be used to good advantage.
41
-------
to
Table 4-3. EXAMPLE METHODS OF MONITORING VARIABLES
VARIABLE
1. Calibration Ga* Concentration
2. H.O/COj Interference
Low Range CO-NDIR
3. Zero Drift
4. Span Drift
5. System Noise
6. Temperature Variation
MONITORING METHOD
Measurement of control samples ai
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.
FREQUENCY OF CHECK
Verify concentration when initially
purchased. Audit concentrations
at monthly intervals 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.
ACCEPTANCE CRITERIA
Zero gas with 10.1 deflection.
Span gas within 10.2 deflection.
Concentration should be within
12 percent of stated value.
See Federal Register, Volume 39,
No. 101, Thursday, May 23, 1974,
85.075-20(0 (11).
Drift should not exceed 11 per-
cent of full scale.
Initial and post calibration should
agree within 11 deflection.
Noise should not exceed 11 percent
of full scale.
Should conform to specifications a
indicated by manufacturer of each
instrument checked.
t> O » W
» » fp o»
«O ft < O
16 IB H- ft
« 01 H-
M H- O
O 05
a ••
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 11 of 24
4.1.6 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. 4-5) :
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 eleminate 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
requirements.
43
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 12 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. 4-6.)
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-failures,
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.
44
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 13 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 perfor-
mance on control charts (Section 6.0) and identify the optimum mainte-
nance 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 mainte-
nance 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 cquired
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
Note: Some instruments are never turned off while they
are "in service" to avoid long warm-up time
o Check and record gain settings for HFID, CO, NO analyzers
45
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 14 of 24
o Check and record zero settings for HFID, CO,
NO analyzers
o Check working gas cylinders for correct pressures and
record pressure
o Check recorders for zero, gain, correlation with computer
and paper supply
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 pressure gauges
and record results
o Check dynamometer torque by hanging weight on torque arm
o Check smokemeter calibration noise and installation
o Check engine inlet air system
o Purge fuel measuring system and check heat exchanger
operation
o Turn on engine cooling system
o Record information on log sheet or log book
Further information concerning pretest preparation may be
found in the Test Procedures in Volume II of this report.
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.
46
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 15 of 24
HEAVY DUTY DIESEL ENGINE TEST CELL DAILY START-UP CHECKSHEET
DEPT. NO.
SHIFT
CELL
DATE
P. I.C.
ANALYTICAL INSTRUMENT CALIBRATION
^~" — '
CO
HFID
NO
Zero
LEAK CHECK
FL
MTR
OBS
X
MAG
OBS
X
RNG
1,000
5,000
2.5
500
1,000
1,500
6,000
^^
GAIN
,/'
ZERO
CYCL NO.
CONC
Recorder
Chart Sp
DVM Corr
Gain
Zero
__^_ ""~
Press.
^^--"'"
Temp.
Sample
Line
DEFL
^
HFID
Fuel
Filter
PRESS
Air
Detector
SMOKEMETER
Filter Readings
10 Percent
20 Percent
40 Percent
Zero
100 Percent
Drift
Alignment: Z~7 Lens: £7 Light: ZI7 Purge Air: £37
Weight:
Actual:
Dyno Torque Check
Inlet Air System /~7 Fuel System /"7 Engine Cooling System /~7
Comments:
Figure 4-1
47
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 16 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 4-4, along with the frequency requirement. Daily checks
such as smokemeter calibration are given in the daily start-up procedure
(4.2.1).
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 5-11, 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 5-9, 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 4-4
PREVENTIVE MAINTENANCE SCHEDULE - MAJOR COMPONENTS
MEASUREMENT SYSTEM
COMPONENT
Smokemeter
Inlet Air/Exhaust
Measurement System
Analysis System
(Includes HC, CO, 02
NO analyzers)
Dynamometer
(Includes speed and
torque meters)
Related Equipment
FREQUENCY
CHECK
6 Months
Monthly
Weekly
Monthly
Monthly
Monthly
REFERENCE
CHECKLIST
Figure 4-2
Figure 4-3
Figure 4-4
Figure 4-5
Figure 4-6
Figure 4-7
(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.
48
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 17 of 24
PREVENTIVE MAINTENANCE CHECKLIST
DEPT. NO. TRAIN DATE TECHNICIAN
EQUIPMENT: Smokemeter
PERIOD: 6 months
/ / Return optical filters to the EPA
for recalibration.
/ / Clean optical lenses and check lamp
illumination power.
/ / Check electrical connection to lamp and
photo multiplier assembly - Repair/Replace
damaged or loose wires/connectors.
/ / Calibrate and service recorder.
(Refer to Manufacturer's manual)
Supervisor Review
Figure 4-2
49
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 18 of 24
PREVENTIVE MAINTENANCE CHECKLIST
DEPT. NO. TRAIN DATE TECHNICIAN
EQUIPMENT: Inlet Air/Exhaust -Measurement System.
Laminar flow element
PERIOD: Monthly
/ / Check air filter. Clean with compressed
air or replace as required by manufacturer's
recommended practice.
/ / Leak check system according to SAE -J244
Section 2.3.1.1.
/ / Clean manometers to remove wall deposits and
add new fluid.
/ / Calibrate incline manometer with dead weight
tester or micromanometer to assure accuracy
and correct fluid specific gravity.
/ / Clean or replace all lines and check connections.
Supervisor Review
Figure 4-3
50
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 19 of 24
PREVENTIVE MAINTENANCE CHECKLIST
DEPT. NO. TRAIN DATE TECHNICIAN
EQUIPMENT: Analysis System - CO, NO, O and HFID
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.
/ / 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.
/ / HFID Burner Peak - See instrument manual of
specific manufacturer. (SAE J215 - Section
3.1.1.1.)
/ / HFID Oxygen Response - Check according to
TP-754 Section 8.0, Test Sequence 103.
Supervisor Review
Figure 4-4
51
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 20 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 and repair.
/ / Recharge polarographic oxygen sensor.
/ / Perform complete monthly calibration on all
analyzers. See Test Procedure TP-203.
Supervisor Review
Figure 4-5
52
-------
Section: 4(HD)
Revision: 0
Date: June 1975
Page 21 of 24
PREVENTIVE MAINTENANCE CHECKLIST
DEPT. NO.
TRAIN
DATE
TECHNICIAN
EQUIPMENT: G. E. Engine 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.
/ / Check oil level at dynamometer support
trunnions.
/ / Lubricate - Lubricate points on the dynamom-
eter as indicated in the manufacturer's
maintenance manual.
/ / Perform complete monthly calibrations of
speed and torque meters. Test Procedure
TP-250.
/ / 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
Figure 4-6
53
-------
Section: 4(HD)
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)
/ / Fuel Measurement System
/ / Torque and Speed Recorder (RPM)
/ / Hygrometer
/ / MV Recorders-Analytical Console
/ / Manometers (accuracy check)
/ / Pressure Gauges
Supervisor Review
Figure 4-7
54
-------
Section: 4(HD)
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
heavy duty engine 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 4-8 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.
55
-------
MAINTENANCE REPORT
DEPT. NO.
TRAIN
DATE
P.I.C
•d o w to
pi fa fl> to H- rt
.. 0) H-
to »••• o
*" 2 3
_ » ••
o ••
Mi
(O C{ O ifc
*. C -.
M
3
en
1
EQUIPMENT DESCRIPTION; 5
SERIAL NO.: 7
SUMMARY OF PROBLEM; 9
CORRECTIVE ACTION: 10
TIME/DATE EQUIPMENT ON LINE;11
MANUFACTURERS
TIME OF FAILURE: 8
CORRECTIVE ACTION PERFORMED BY: 12
WHITE/FILE CANARY/QUALITY CONTROL PINK/MANAGER
SUPERVISOR_J3_
Figure 4-8
-------
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 1 of 30
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, mainte-
nance, 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 II of this program has been prepared in
two volumes. Volume I contains the Quality Assurance Guidelines for
Heavy-Duty Engine Emissions Measurement Systems. The Quality Management
59
-------
Section: 5(HDO
Revision: 0
Date: June 1975
Page 2 of 30
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, includ-
ing the assignment of functional responsibility. The procedures are
60
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 3 of 30
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 documents outlining the testing
procedures to be used to measure the emissions from mobile sources is
the Federal Register and the recommended practices of the Society of
Automotive Engineers (SAE) (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 pn 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(HD)
Revision: 0
Date: June 1975
Page 4 of 30
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. Teat 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(HD)
Revision: 0
Date: June 1975
Page 5 of 30
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 the major activities or
operations usually included in the plan are:
o Procurement (ordering)
o Procurement (receiving)
o Calibration
o Verification and correlation
o Test Operations
Engine Preparation and Installation
Smoke Emission Test
Gaseous Emission Test
Fuel Control
63
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 6 of 30
o Data Reduction and Validation
o Preventive Maintenance
o Auditing
For each of the above activities the following items should be
considered:
o A list of the characteristics to be checked.
o A list of the acceptable level of quality requirements
established by Quality Planning (QA).
o A description of the frequency of checking each
characteristic.
o A brief description of the method, equipment or reference
standards to be used for checking each characteristic.
o Directions for the inspector to follow if the character-
istic does not comply with acceptance limits.
o A description of the type of record in which the accept/
reject data is to be reported.
Examples of a Quality Plan Activity Summary Matrix for Procure-
ment (receiving) are given in Tables 5-1 and 5-2. 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
64
-------
in
Table 5-1. EXAMPLE OF QUALITY PLAN ACTIVITY MATRIX
PROCUREMENT (RECEIVING - FUELS)
ITEM/CHARACTERISTICS
F-1. Diesel Fuel
Type 1-D
o Flash Point
o Distillation
Range
o Vendor Batch
Analysis
ACCEPTABLE
LIMITS
120± 2°F
IBP °F 330-390
10% 370-430
50% 410-480
90% 460-520
EP 500-560
Octane 48-54
Gravity
40-44° API
Total Sulphur
0.05-0.20%
HC Comp.
Arom. 8-15%
Viscosity
1.6-2.0
FREQUENCY
OF CHECK
Each Batch
Each Batch
Each Batch
METHOD
ASTM-D-93
ASTM-D-86
Analysis by
Supplier
Accompanies
Shipper
CORRECTIVE
ACTION
Reject Batch
Reject Batch
Reject Batch
RECORD/
REPORT
Receiving
Report
Receiving
Report
Receiving
Report
v a jo w
iQ ft- < O
fl> ••
(0
a
O
-------
Section: 5{HD)
Revision: 0
Date: June 1975
Page 8 of 30
o Reporting failures
o Initiating and assuring closed-loop corrective actions
o Recording audit results
o Initiating procedural or equipment change notices
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 5-3.
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.
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 5-3
shows some examples of calibration tags and a rejection tag; Figure 5-4
66
-------
Table 5-2. EXAMPLE OF QUALITY PLAN ACTIVITY MATRIX
PROCUREMENT (RECEIVING - GAS MIXTURES)
1
ITEM/CHARACTERISTICS
G-l. Span Gas - CO
o Concentration
o Cylinder
Leakage
o Hydrocarbon
Content
o Valve
ACCEPTABLE
LIMITS
±2% of
Supplier Analysis
Range 3000 ppm
2350-2650
Range 2.5%
2.0-2.4
None
20 ppm
GGA 350
FREQUENCY
OF CHECK
Each
cylinder
Each
cylinder
1 cylinder
per order
Each
METHOD
NDIR Gas
comp
Soap Sol'n
valve neck,
stem, seat.
FID
Reference
standard
Visual/
Fitting
CORRECTIVE
ACTION
Analysis-
relabel
cylinder
with average
of duplicate
Analysis
Range -
Return to
Supplier
Return to
supplier
Analyze
all cylinders
Return to
Supplier
Return to
Supplier
RECORD/
REPORT
Receiving
Log
Failure
Report
Rejection
Report
Rejection
Report
Receiving
Log
Failure
Report
Rejection
Report
t) O V W
fa fa 0> ®
ua rt < o
(0 <0 H- ft
.. 0) H-
V0 H- 0
O 3
O 3 ••
Q O in
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 10 of 30
Table 5-3. SUMMARY OF FORMS REFERENCED IN VOLUME I
TITLE OF FORM
SECTION REFERENCE
Daily Start-up Checksheet
Preventive Maintenance Checklist,
Smokemeter, 6 months
Preventive Maintenance Checklist, Monthly
Inlet Air/Exhaust Measurement System
Preventive Maintenance Checklist, Analysis
System, Monthly
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
Analyzer Curve Generation Data
Monthly Dyno Calibration Log
Gas Analysis Report
Equipment Repair Authorization
Rejection Report
Failure Analysis Report
Corrective Action Request
Performance Audit Summary Sheet
Procedure/Equipment Configuration Change
Notice
Section 4, Figure 4-1
Section 4, Figure 4-2
Section 4, Figure 4-3
Section 4, Figure 4-5
Section 4, Figure 4-6
Section 4,
Section 4,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Section 5,
Figure 4-7
Figure 4-8
Figure 5-1
Figure 5-3
Figure 5-4
Figure 5-5
Figure 5-6
Figure 5-7
Figure 5-8
Figure 5-9
Figure 5-10
Figure 5-11
Figure 5-12
Figure 5-14
Section 5, Figure 5-15
68
-------
DATE
RECEIVING INSPECTION REPORT
NVOICE NO.
Section: 5(HD)
Revision: 0
Date: June 1975
Page 11 of 30
1. Received From
2. Delivered By
3. Shipping Damage
k. Received By
5. No. Pkg./Weight
6.
PARTIAL
COMPLETE
7. Purchase Order No.
9. Shipped to Attention of.
10. Packing Slip No
8. For Department.
.11. Unpacked By.
12. Invoice — Packing Slip — Purchase Order checked for correct count
and Material Part No. :
13. Final inspection to be completed by
14. Sent for final inspection Date
15. Inspected By
16. Q.C. Inspection Plan
MATERIAL INSPECTION REPORT
Dept._
Procedure No.
17. Characteristics
Checked
Acceptable
Quality Level
Actual Measured
Conformance
18. Disposition of Material: ACCEPTED || SEND TO USER
HOLD FOR ORDER COMPLET ION Q] SEND TO STORES
Reason
REJECTED
Rej . Report No..
Distribution: 1.
Purchasing 2. Requestor
Procurement Control
Figure 5-1
69
3. Rece iving File
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 12 of 30
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
70
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 13 of 30
CALIBRATION
Dot*
Du.
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
71
Figure 5-3 CALIBRATION TAGS
-------
s-
e
Ol
^
H
0)
to
(1)
1 1 t 1 I
1 1 1
c^ CONTROL NO. "
(4)
i 1 1 l-l 1
1 1
41 MODEL *
(2)
III II
1 1 I
L 11 1
"wSl" <8t- NOMENCLATURE
(5) (6) (7)
i MM
M Tvp£ *, « MFfl is ''CYCLE »'
(8)
1 1 1
••OftON. NO"
(9)
II
-• FAC.ci
COfc
CONTROL NO.
SI1SMI888
3399393399
j.: i < i i i i i »
0
S
H
88
3 9
II 1!
NOMENCLATURE MANUFACTURE
8 8 8 8 1 1 1 1 1 1 1 8 1 1 8 1 8 1 8 1 > 1
9939333393333993999999
i) u u u u ii n n it » a n n » » n » » » » » »
111818(8
99939999
n H H v n u u ii
li
ii.
(3)
1 1 1 1 ! 1 1
1
" MANUFACTURE
(10)
1 1 1
19 CUE '*
(11
|
M (N '
(12;
i
t'*OUT7'
MODEL
8888881888
9999999399
O II fl> H- rf
« tn H'
*
Ml
O Ul
§
-J
U)
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 15 of 30
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.
Figure 5-5 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 analytical instruments, dynamometer and the gas mixtures. The
smokemeter calibration is performed daily on each test and the daily
start-up sheet is used for documentation (Figure 4-1). Procedures for
calibrating these items have been detailed in Volume II of this report.
Examples of forms presently in use are given in Figures 5-6, 5-7, and 5-
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 data, 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 4-8. 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 5-9.
73
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page-16 of 30
CALIBR&1E&N HISTORY EVALUATION
EQUIP TYPE
MFR.
MODE I _
EVALUATED BY.
TIMES FAIL
DATE
TIMES CAL..
TIMES O.T. .
-. O.T.
AVC. CAL. TIME .
PERCENT
HOURS
CONTROL NO.
I.T.
O.T.
FAIL
REMARKS
Evaluation Summary:
Corrective Action:
By
Date
Follow-up Required:
Follow-up By:.
Date ,
/4
.Remarks.
Figure 5-5
-------
TESTING SERVICES DIVISION
ANALYZER CURVE GENERATION DATA
Section: 5(HD)
Revision: 0
Date: June 197S
Page 17 of 30
DEPT NO
TRAIN
DATE
P. I.C.
ANALYZER
CYLINDER NO
RANGE
DEFLECTION
CELL LENGTH
CONCENTRATION
ATTENUATION
FLOW RATE
NEXT RUN NUMBER
NO.
i
ZERO
GAS
AIR
NITROGEN
NEW
UPDATE
COMMENTS:
Figure 5-6
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 18 of 30 MONTHLY DYNAMOMETER CALIBRATION LOG
Cell No.
Model No.
Serial No.
H.P. Max.
A. RPM
Calibrated By
Supv.
Date
Output
Calib. Equip. Type_
Control Number
Before Adjustment
After Adjustment
0
500
1,000
1,500
2,000*
2,500
^Adjustment Point
Computer Input Alignment Checked
Comment:
Failure Report Required / /
B. TORQUE
Output
Torque Arm Length
Before Adjustment
After Final Adjustment
TORQUE CALIBRATION
Zero =
Zero =
Weights I.D.
Dyno. Constant
Max.
Max.
WEIGHT
ADDED
1 0
2
3
4
5
6
7
8
9
10
TORQUE CALCULATION
WEIGHT x LENGTH
0
TORQUE ACTUAL
Computer Input Alignment Checked
Maintenance Performed
Comment:
/7 Failure Report Required /7
Figure 5-7
76
-------
Figure 5-8
Section: 5(HD)
Revision: 0
Date: June 1975
Page 19 of 30
GAS ANALYSIS REPORT
CYLINDER *
Supplier
Requestor
No. of Cylinders in <
MIXTURE REQUESTED
•
Analyst
Date
P.O.a
Project tt _.,.
Drder
Invoice it
ANALYSIS
*
•
Report Data To
Comments
77
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 20 of 30
REQUEST
Name _
Figure 5-9
Equipment Repair Authorization
Job*.
. Section
Branch
:Equipment I.D. Number
Job Description (Attach sketches needed)
Date Submitted —
Extension
Special Equipment Required
Proprietary Item: fj Yes
Craft Requested
Date Item To Be Delivered for Test
Latest Acceptable Compfetion Date
SCHEDULE - Equipment Maintenance
Date Request Rec'd
Craft(s)/Team Assigned
Equipment Repair Service Contacted
Date Time 1 am/pm
Authori zed by '
Date
Time
_am/pm
Equipment Back On Line
Date • Time
Equipment Repaired
Replaced
am/pm
Total Down Time
No. of Test Rescheduled_
Test Supervisor
Equipment Maintenance Report
Technician
Date Regan Time am/pm
Date nnrnptpt** Time am/pm
Man.Hniirc
Parts R*»plared
.
Comments
Repair -Service Report
Date R0gar> . Time
Date Complete , Time
Man.Hnurc
Parts Replaced
_am/pm
.am/pm
*
Comments
•
. t
Service Charges
WHITEi REQUESTOR'S COPY
YELLOW: PRODUCTION CONTROLLER'S COPY
PINK: LAB SECTION CHIEFS COPY
GOLD: REQUESTOR'S IN-PROCESS COPY
78
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 21 of 30
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.
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 engine inspections, test cell
correlations and other auxiliary laboratory tests. A typical Rejection
Report for use in reporting unacceptable results is shown in Figure 5-10.
This report should contain the type of result such as void test, the
unacceptable characteristic or data such as torque-RPM 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 an engine 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, engines and instruments 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 estab-
lished and corrective measures 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 5-11.
79
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 22 of 30
REJECTION REPORT
NO.
PART NUMBER
PART NAME
SUPPLIER/MFR
REJECTED
CONTRACT
QUANTITY
DATE
PURCHASE ORDER NO.
REC. REPORT NO,
ITEM NO.
DISCREPANCIES
REJECTED BY
DATE
SUPERVISOR APPROVAL DATE
Q.A. APPROVAL DATE
DISPOSITION
USE AS IS
RETURN TO SUPPLIER!
CHECK IF FAILURE ANALYSIS REQUIRED
FAILURE ANALYSIS REPORT NO
OTHER (SPECIFY)
CORRECTIVE ACTION
Q.A. APPROVAL.
DATE.
Figure 5-10
so
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 23 of 30
^
to
S3
II
S§
P« B
§2
O W
P4
o
DATE
REPORT NUMBER
FAILURE ANALYSIS REPORT
EQUIPMENT TYPE
MANUFACTURER
MOD. NO.
SER. NO.
NO. OF PAST FAILURE
REPORTS
DATE -LAST FAILURE
HOURS SINCE LAST
LOCATION FAILURE
ESTIMATED CAUSE
DETERMINATION
EXPECTED, RANDOM
ASSIGNMENT CAUSE REQUIRES LAB
INVESTIGATION
CAUSE DETERMINED OR TYPE INVESTIGATION NEEDED
TESTS PERFORMED:
FAILURE CAUSED BY:
gg
W K
^§
5
u
DISPOSITION:
/"7 REPAIR
/~J SCRAP
/"7 REPLACE
f~J OTHER
/"7 RETURN TO
MANUFACTURER
RECOMMENDED CORRECTIVE ACTION TO PREVENT RECURRENCE
ACTION COPIES TO:
RECOMMENDED BY
REVIEWED BY
X U
« H
CM
Q h
W O
U
W
O
CORRECTIVE ACTION TAKEN
RESULTS AND RECOMMENDATIONS
APPROVED
CLOSED OUT
DATE
DATE
Figure 5-11 FAILURE ANALYSIS REPORT
81
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 24 of 30
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
potential 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 5-12.
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 5-13 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.
82
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 25 of 30
Corrective Action Request
o
EH
OJ
W
a
w
o
z
w
w
2
M
O
z
w
O
EH
Z
w
w
a
CQ
H
(0
5;
o
CU
en
w
a:
1. Request initiated by
Date
Dept.
Authorization
2. Brief description of non-conformance
3. Recommended Action
4. Assigned to
Date
File No.
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
Nc
8.
Action Assigned To
Completion Date
Date
_Supervisor
Special Instructions
9. Action Completed - Date
Time
Quality Engineering Notified
Requestor Notified
10. Comments
Figure 5-12
83
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 26 of 30-
CORRECTIVE ACTION REQUEST - FLOW CHART
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.
I
Support Operations repairs Gas Analyzer.
Completes Equipment Repair Authorization.
YES
NO
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
1
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
dic.aT}nT"rt\7f*H i ^0.1100 npu f"1 A P ir» JVr*i"irtn AHs^voccoo
notes status in C.A.R. Log. Follows up on new C.A.R.
until Action Addressee has completed
approved Corrective Action.
FIGURE 5-13
84
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 27 of 30
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 5-14). 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) by flame ionization (FID) or by
an FID with a heated sampling system.
85
-------
PERFORMANCE AUDIT SUMMARY SHEET
oo
a\
AUDIT ELEMENT: Q INSTRUMENTATION Q OPERATOR £) SAMPLING SYSTEM Q DATA PROCESSING
AUDIT
DATE
TYPE OF AUDIT
AUDIT RESULT
ACC.
UNACC.
CORRECTIVE ACTION TAKEN
AUDITED
BY
vQ ft < O
ID IB H* ft
.. (a H-
NJ H- 0
03 O 3
3 ••
O »
Hi
to t| O Wl
FIGURE 5-14
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 29 of 30
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 heavy duty diesel procedure but are required by
the 1975 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 5-15. A
similar document could be used to effectively control changes in com-
puter programs in facilities which employ computer systems for testing
and computational services. Procedures for document control are given
in Appendix C of this report.
87
-------
Section: 5(HD)
Revision: 0
Date: June 1975
Page 30 of 30
Figure 5-15
PROCEDURE/EQUIPMENT CONFIGURATION CHANGE NOTICE
ORIGINATOR:
2. DATE:
3. TYPE OF CHANGE: EQUIPMENT
4. REFERENCE DOCUMENT:
PROCEDURE £7
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 fj E&D £J CHEM £J LAB l_J
HOT £J I6E £J C6M £J 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.
88
-------
-------
Section: 6(HD)
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(HD)
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, and are 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
92
-------
Section: 6(HD)
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. •••••i
UPPER CONTROL LIMIT
+ 1.0.
9—
CVS FRAME
•o a
P 01
«Q rt
(D (D
•a tn
p C< O Ox
g -
5 S
VO
-o
cc
o
oc
CENTER LINE
LU 0
-1.0-
LOWER CONTROL LIMIT
T"
2
8
T~
9
10
11
12 13
15
SAMPLE NUMBER
Figure 6-1. PROPANE INJECTION TEST - % ERROR
-------
Section: 6(HD)
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
2
/L(x-x) \
\ n-1 /
%R = -- x 100%
CV = -- (100)%
95
-------
Table 6-1. APPLICATIONS OF STATISTICAL CONTROL CHARTS IN 1975 FTP TESTING
TYPE OF CONTROL CHART
vo
a*
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
>a D » w
Pl p) (D (II
\Q rf < O
<0 a H- rt
.. to H.
00 ^} O CT>
1
-------
Section: 6 (HD)
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 eacli concentration
should be found, i.e., calculate x , s , x , s , x , s , x , s^. Plot
these values on a scatter diagram. They will 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(HD)
Revision: 0
Date: June 1975
Page 8 of 38
CASE 1 - STANDARD DEVIATION ESSENTIALLY
INDEPENDENT OF CONCENTRATION
o0.2-
H-
<
UJ
o
So.i,
to
0.0
©
23456
CONCENTRATION MEAN
8
CASE 2 - STANDARD DEVIATION INCREASES
PROPORTIONATELY WITH CONCENTRATION
0.3-,
o
£0.2-
UJ
o
(£.
<
O
0.1-
0.0
2 3 k 5 6
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(HD)
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
x = 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 (HD)
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
— R
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 = D3R.
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
-------
">,
16-
RANGE OF PAIRS OF REPLICATES
UPPER CONTROL LIMIT = 16.99
w
&> fl) (0 (D
vQ rt < n
fl>
-------
Section: 6 (HD)
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
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
UCL = D. R = 3.267 x 5.2 = 16.99
4
LCL = D R = 0 x 5.2 = 0.0
D_ and D. are multiplication factors when observations
in each subgroup - 2
102
-------
Section: 6 (HD)
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
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
UCL = D %R.
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 i
UJ
QC
LL>
30-
20
10 -
REUTIVE RANGE OF GAS MEASUREMENT CONCENTRATIONS
UPPER CONTROL LIMIT - ^3.03
o
A
/ i
/ \
\ —
\AVERAGE %R=I3.17 O
\ /
1 2
6 7 8 9 10
DAY TESTED
11 12 13 \k 15 16
v o » in
p> pi to to
iQ ft < O
fl> m P- rt
" CO H-
W Cj O
00 § "•
fl> g
VD
~a
tn
Figure 6-4. RELATIVE RANGE CONTROL CHART
-------
Section: 6(HD)
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
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(HD)
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
2J{x-x)2 1
-n^i/ •
For each sub-group of replicates, calculate the coefficient
of variation using the formula
CV = — {100}%.
x -
Calculate the average coefficient of variation using the
formula
N
when N = total number of sub-groups.
• Calculate the lower control limit using the formula
LCL = B CV
The value of B3 is obtained from Appendix A-2.
• Calculate the upper control limit using the formula
UCL = B. CV
4
The value of B4 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
k. On
UPPER CONTROL LIMIT = 3.90
z3.0H
o
<
cc
>
u_ 2.0
o
£
t i. (u H- rt
.. in P-
i_. (-•• o
-j O 3
O
U) C-i O
-------
Section: 6(HD)
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
8
.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
B = 0 and B4 = 2.089 when observations in each subgroup = 5
108
-------
Section: 6 (ID)
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
c
(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 (HD)
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
-------
UJ
o
DIFFEREN
+ 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
xQ
r \
/ \
- : '•
1 \
Q / \
A / \
/ \ • \
n / \ / \ n
L /W " '
/ o o
O
13 D 73 (ft
— Hi PI (D (D
iQ rt < O
fl> (D !-•• rt-
•• m H-
NJ H- 0
- M § »
O
i-h
OJ t-| O CTl
oo c *-•
— D a
n> D
M
LOWER CONTROL LIMIT ^
1 234 56 7 89 10
FIGURE 6-6. SIGNED DIFFERENCES CONTROL CHART
-------
Section: 6(HD)
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
-------
Section: 6(HE)
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, CO2, and NO gas analyzer performance, sampling
from past six montlvs data.
o Determining frequency of multi-point calibration, sampling
from past six month's 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(HD)
Revision: 0
D^te: 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, 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 five different. 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 demonstrate 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 (HD)
Revision: 0
Date: June 1975
Page 25 of 38
Table 6-7. ANALYSIS OF VARIANCE - ONE-WAY CLASSIFICATION
GAS CONCENTRATION MEASUREMENTS FROM THREE CARS
TEST
NUMBER
1
2
3
4
5
SUMS
SUMS OF
SQUARES
TOTAL S.S.
fao c c
EMISSION MEASUREMENTS
CAR TYPE 1
6.6
7.2
6.4
7.4
7.8
35.4
251.95
- 662.79 - -
(35. 4)2 +
CAR
6.
6.
7.
6.
6.
33.
218.
(99. 2)2
15
(33.0)
TYPE 2
6
4
0
2
8
0
20
- =6.75
2 + (30.8)
CAR TYPE 3
7.0
6.0
5.0
5.8
7.0
30.8
192.64
2 2
(99. 2r
15
= 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 (HD)
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. = (20.2)2 + (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(HD)
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 .(HD)
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 (HD)
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(HD)
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 X, is suspect
—^—
If X is suspect Most sensitive
criterion when
_ v
_
rio " x - x.
n 1
or
X - X ,
n n-1
X - X., 3 « n s 7
n 1
Y — Y
X2 Xl
X _ - X,
n-1 1
or
X - X ,
n n-1
X - X,,
n 2
8 «j n s 10
Y _ Y
X3 Xl
r = ——^———
21 X . - X.
n-1 1
or
X - X .
n n-2
X - X.
n 2
11 s n s 13
X3"X1
Xn-2 - Xl
or
X - X .
n n-2
X - X_
n 3
14 * n s 25
X. denotes either individual values or means of data sets arranged in order
£f 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 r , r ,
and r_2 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.
emissions in parts per million were as follows:
The CO
Vehicle
CO Emissions in
Parts Per Million
A
B
C
D
E
F
510
521
523
501
493
605
120
-------
Section: 6(HD)
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
r = 6 " S = 60S - 523 _82_ _
10 605 - 493 112 ~ U*
X6 " Xl
The critical value is 0.56 for or = .05 per the referenced tables.
Therefore, since the computed valued of r (0.732) is greater than
0.56, it can be concluded that P 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(HD}
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
d =
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 of the 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
to
W
MERITS
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 o jij en
la fa (D (D
iQ rt < o
n> n> P- rt
.. 01 H-
u> H- o
oj 03
oo
c_i o en
a
D
-------
Section: 6 (HD)
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,
Poisson, 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.
_ _ . , . , . cumulative frequency ,rtrt
3. For each value, calculate =—* *- 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
-------
.ection: 6(HD)
Revision: 0
Date: June 1975
Page 35 of 38 »
e
o -f- - - - - ~ -;- • — -H —
0
M
O
in
0*
:::: :::: :::: :::: :::: :::: :::: r::: =n:
ji — ~~^ ~~4 — ri — ""^r 1
:: ::::: 1 ::
«> _, " . "!! "I! "! !I /_
J
'
8 ""T = l~ """--i !;::=::a::
1 ._ 1— 1
| 1 HI || HI II II |l ijj] || |l|n
/
0> "' •"'
7
0:-""-" :r- = :?i: :::;::::::::::::::::::::
_ — _
1
* - i
_i /-- -i 1
L- _ 1
*
$ i i L _^ -L
Vs
: /
' j
l[l\ ^l \ \\\\ \[\ill\\\\lll\[^ -j—t
7 s
' ; ">.
z *
/
j. :.
:::: ::::::::::::::::::: ::::i :::=:::::::::
_;:: -".-.-.-.--.-.-.Y----------^------ ------------ 3
j
;:::::: = ::::: :..:_7L^:.. «
/
_ m
f
/
"i : :z _ : :
.. ^ _ f 1 1 — 1 — _
==!!==[=H:!=: [=[=!!===: !=!!!!|=! !===!====:»
— .2 S _ — g
J2 ^- — j ]
1 / — ' 1 H — I • 1 ' T-TT Q
i 1 ' i 1 III i 1 i 1 1 1 1 i ! i r i 1 ' 1 i 1 1 1 i 1 1 '• ' " " " *
J
- -t 1 — i ~~ H *- -i pH- - - j j i - - -H- N
1_. 1 . 1
I ± ±
:: ---'. :.:---_- -u, — )_--
:___:: : --": '--'- m
N
::::: ::: L^^JTnij 1 1 [T'||f|5
1 L 3
f» +(T /u+2tr /u +3a-
Figure 6-7. NORMAL PROBABILITY PAPER
125
-------
Section: 6 (HD)
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
-------
Table 6-10. TABULAR DESCRIPTION OF CO
EMISSION LEVELS IN PPM
Section: 6(HD)
Revision: 0
Date: June 1975
Page 37 of 38
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) x 100
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 (HE)
Revision: 0
Date: June 1975
Page 38 of 38
Sr
R-
o
0
•
8 -
"
S •
S -
in
s •
8 :
+
»
S
Ok
S
S
.1C -
!
. __ "1J
:
_j_j. . - : -j- - -i £_-___----_-
HHiyHiiinHinHiHiHyiy
— i — — j — i — — i — i — — -M— t— i -i — — i— j— — j — . — i- -
' ir
,
, J
'
( 1
1
'
jit
H"Hfl 1 1 'H H H H 1 IjHIl 1 1 1 1 1 TUff
— j — — — (_ ^ —
/ ^i
'
. - _-
7 '
> 1
L
1 )
,^ |
::«:::
j *
--i j^- 1 •lJ — ' 44- -H-l-
[
•
~ in 11 10 10 iJ. ir
._ ^ i~*^ .L
,^j j.._^^d
A - _ — -
7
J
f
A
1 I.I 4-p^— u t~- — ^-^ "S"~ ~^^
-' 1 T^-j j_ -._. 1 1 , PC - - .
_. -W ' 1 -r- • « r— * *— |- j
II /
^ 4^^ "" ""
r!
j t
:::::::::: ^:.__+^ :.._
_ . ^ 0~" • H ^ 1 — •
1 i
j 1 i
::ElEEn = ::Ej|EEE:E|EE]EiE= = = = E:
---]:— S^-H tgiH i!^-d-« j:aF]l"
1^ IT 10 in on 01
| = :|
4k
i —
\
"s
: 01
S p
-5 E
s L-
R
: E
|«?
" u
N
IA U
E
R
t S
-
-- 3
B
- _
T
:: 8
in
" 0
N
'• o
!i3
V4
Figure 6-8. NORMAL PROBABILITY PAPER (CO EMISSION LEVEL)
128
-------
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 1 of 22
Section 7
ANALYSIS OF VARIABILITY IN THE MEASUREMENT OF
EMISSIONS FROM HEAVY DUTY DIESEL ENGINES
Identification, control and quantification of variables is a
required objective in a program to assess the reliability of a measure-
ment system. The precision and accuracy of mobile source emission
measurements are dependent on the variability existing in the engine
operation and the measurement system. Although these guidelines reflect
the current state-of-the-art, EPA is conducting extensive studies to
specifically quantify the major contributions to heavy duty emissions
test variability.
Test procedures and equipment specifications will be made more
definitive and quality provisions (acceptability criteria) will be added
to improve the accuracy and precision of the total measurement system.
Interlaboratory variability by engine manufacturer had been
studied through an analysis of data obtained by the EPA from the engine
certification tests. Although the total measurement system variability
is of prime importance, it is essentially a composite of all variables
and will only be useful for predicting overall test variation between
the laboratories or test cells involved in the program. If bias should
exist in one or both systems being compared e.g., the use of a gas
deviating from the acceptable standard, the cause of the variation
between systems will not be identified.
A knowledge of specific variables significantly affecting the
data is a prerequisite for achieving a predetermined goal, improving
data reliability and detecting bias factors in the system. These varia-
bles are either determinate or indeterminate. Determinate variables may
be objectively studied, but the nature of indeterminate variables require
subjective evaluation. Indeterminate variables are usually estimated
through experience with the measurement system, engineering evaluation
of the test procedure and statistical analyses of the data.
The measurement system for heavy duty diesel engines involves
the measurement of smoke and gaseous emissions. In addition to engine
inconsistencies certain measurement system variables have been estab-
lished as prime sources of error. Efforts to reduce these variables
include the use of instruments and calibration standards having improved
precision and accuracy, and improvement in methods of sampling gaseous
emissions.
131
-------
Sections 7(HD)
Revision: 0
Date: June 1975
Page 2 of 22
The purpose of this section of the guideline is to discuss the
methods used to identify these major sources of variation, to quantify
the effect of the determinate variables and the involvement of Quality
Assurance in the effort to reduce test variables.
7.1 VARIABLES ASSOCIATED WITH THE MEASUREMENT OF SMOKE EMISSIONS
The Federal smoke emission test is designed to determine the
opacity of smoke in exhaust emissions using a continuous recording, full
flow light obscuration (opacity) meter. The engine dynamometer cycle
utilizes engine operating conditions which tend to promote smoke from
diesel engines. The smoke emitted during acceleration and lugging modes
is monitored continuously and opacity values are averaged for each half
second interval. Three identical cycles are run following a 10 minute
preconditioning mode.
The primary sources of variability in this procedure are generally
considered to be the
o Dynamometer Operation
o Diesel Engine Operation
o Smokemeter
o Recorder
o Ambient Conditions
o Chart Reading
o Computer
The contributions to total test variability from each of these
sources will vary from one laboratory to another and even from cell-to-
cell within a test facility because of the wide variations permitted in
the design of the measurement system.
Methods for controlling these variables include:
o Application of computer automation techniques
o Frequent maintenance and calibration of the smokemeter
and recorder
o Application of statistical quality control techniques as
an "early warning" system to detect impending problems.
7.1.1 Dynamometer Operations
The dynamometer speed and torque meters are calibrated at
monthly intervals. The associated recorders are checked and aligned
132
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 3 of 22
with the meters prior to each test. The RPM output is checked with a
primary standard such as a strobotac with a standard accuracy of +1
percent. The torque meter is usually calibrated using weights placed on
the torque beam. The weights should be traceable to the NBS and should
have an accuracy of +O.1 percent. The test variability associated with
the operation of the dynamometer would be difficult to assess without an
extensive program involving several operations, dynamometers and engines.
Computer controlled dynamometers have become quite popular;
however, these systems require manual intervention, with its associated
inherent variability to program the computer each time an engine is
installed in the test cell.
The test cycle is operated according to engine RPM within the
specified tolerance for a particular mode. Speed-torque traces are made
for each test and those not meeting the RPM tolerance are voided and the
test is rerun.
An assessment of this dynamometer variable could be achieved
by performing 10 consecutive tests (cycles) on a well preconditioned
engine using the same computer program or operator alternated with 10
consecutive tests during which the RPM is varied within the specified
limits during the cycle, using different operators if possible.
Comparison of the variability (S2) of the two sets of "a", "b" and "c"
factors (F.R. Ref. 85.874-18, para, (a)(b) and (c) ) would give an
objective assessment of the variability due to dynamometer operation.
Naturally this would only be valid for that particular engine-dyno
combination. The test would be repeated with different engine-dyno
combinations each time a new engine is installed. A Coefficient of
Variation control chart as described in Section 6.2.5.3 could be
established using the collected data. The analysis of variance tech-
niques also referred to in Section 6 could be applied to determine
significant differences between the various combinations.
7.1.2 Diesel Engine Operation
The preconditioning of the engine and its operation during the
test cycle may be two of the greatest contributors to test variability.
They have not received the detailed attention associated with the precon-
ditioning and testing of light duty vehicles (Ref. Phase I Report). The
prescribed test requires a ten minute preconditioning of a "warm" engine
at maximum rated horsepower. The monitoring of engine operating param-
eters such as fuel and air flow, engine and exhaust temperatures, fuel
temperature and pressure, and inlet/exhaust pressures as is done in the
gaseous test provides valuable information for the detection of engine
malfunction or change in operating characteristics. Engine variability
is associated with determination of gaseous and smoke emissions and is
discussed further in Section 7.3.
133
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 4 of 22
7.1.3
Smokemeter and Recorder
The smokemeter type and specifications are defined by the
Federal Register in paragraph 85.874-13 and in test procedure TP-751 in
Volume II. Smokemeters meeting these specifications are available from
several instrument manufacturers. They are usually supplied with a
primary set of neutral density filters having calibration traceable to
NBS. The primary set of filters are sent to the EPA, Ann Arbor Facility
annually for calibration and verification. These filters have a cali-
brated accuracy of +1 percent, traceable to EPA or NBS. It is important
to handle and clean filters carefully since any contamination on a
filter may change its opacity value.
Filters having a nominal opacity of 10, 20 and 40 percent are
used to verify the linearity of the smokemeter output prior to each
test. Zero and 100 percent points are checked before and after the
test. Changes in instrument calibration in excess of +2 percent during
the test require a readjustment, re-calibration and repetition of the
test.
The following table summarizes instrument characteristics that
are major sources of error in smoke measurement.
Table 7-1. MAJOR SOURCES OF ERROR IN SMOKE MEASUREMENT
CHARACTERISTIC
TOLERANCE
ESTIMATED COEF-
FICIENTS OF VARIANCE
Drift
Noise
Linearity
Calibration Filter
Recorder Response
Maintenance
+_2 percent
+1 percent
+2 percent
+1 percent
0.5 Sec full scale
Indeterminate
0.5 percent
0.25 percent
0.5 percent
0.25 percent
0.5
1.0
percent
percent
Total Variance =1.37 percent
The acceptable tolerances shown in the table are from the
Federal Register. Estimated coefficients of variance are listed as
examples of what might.be typically expected if these characteristics
were determined experimentally in the laboratory. The total variability
is estimated from the square root of the sum of the squares of the
individual coefficients of variance.
134
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 5 of 22
7.1.4 Ambient Conditions
The humidity barometric pressure air circulation and tempera-
ture in the test cell can affect the operation of the instruments as
well as the engine.
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 preferred
audit tool because of its portability.
Other recommended methods for reducing 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°P or better.
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 be done by data
validation or a computer, utilizing one of the control chart techniques
described in Section 6. Comparison to the reading of the previous test
on the same day would provide an additional check.
The air circulation and venting system should be controlled so
that the smoke plume and engine temperatures are not affected, and to
prevent "hot" spots or sudden changes in temperatures in the area where
the instrumentation is located.
The temperature of the cell is specified to be between 68
and 86 F (20-30 C). If tests are consistently run at one extreme the
test results will generally differ from those run at the other and of
the 18 degree range. Ideally, temperatures should be controlled to
within ±3 F of the mid point of this temperature range.
135
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 6 of 22
7.1.5 Chart Reading
Data calculations for smoke opacity are performed in the
following manner:
(a) Locate the acceleration modes and the lugging mode on the
chart. Divide each mode into one-half second intervals
beginning at the start of each mode. Determine the
average reading during each one-half second interval
except those recorded during the transitional portions of
the cycle.
(b) Locate and record the fifteen highest one-half second
readings during the acceleration mode of each cycle.
(c) Locate and record the five highest one-half second
readings during the lugging mode of each cycle.
(d) Record the three highest values of the above 20 readings.
(e) Repeat the recordings required in steps (a), (b) and (c)
for a total of three cycles required by the test procedure.
(f) Average the 45 readings by the acceleration mode and
designate the value as "a".
(g) Average the 15 readings of the lugging mode and designate
the value as "b".
(h) The nine highest readings determined by steps (d) and (e)
are averaged and designated as value "c".
In addition deterioration factors "A", "B" and "C" are also
calculated as specified in Federal Register 85.874-28 using data from
emission and durability engines for compliance purposes.
An example of the speed and opacity trace is shown in
Figure 7-1. It is apparent from this chart, that locating the intervals
manually is a difficult and tedious job. Many errors could be expected
in manual reduction of the data, consequently reduction by computer
integration is the preferred method.
Since the three cycles, theoretically, should result in the
same value, it would be more advantageous from a validation or audit
standpoint to average the cycles independently. This data would then
lend itself to the establishment of a control chart to detect cycle-to-
cycle changes. Excessive changes from one cycle to another would indi-
cate a need for analysis of the system or engine to determine cause of
changes and corrective action required.
136
-------
•
Figure 7-1. SMOKE OPACITY AND ENGINE SPEED TRACES FROM A FEDERAL SMOKE
(SOURCE: REFERENCE 7-1)
TEST
H- O
O 3
X
O
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 8 of 22
7.1.6 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 and should be periodically checked for
reliability. One proven method of checking data reduction is the use of
a previously prepared standard set of manually calculated data. This is
fed into the computer and the output is compared with the calculated
values. The same set of data could be used in cell-to-cell or labora-
tory-to-laboratory correlation studies.
Another source of error which may occur when using computer
integration of a smoke trace is the inherent electronic noise associated
with this type of instrument. The computer is usually unable to distin-
guish between noise and trace signal. Electronic filters are used to
eliminate some of the noise but these do not completely eliminate the
noise problem.
7.2 VARIABLES ASSOCIATED WITH THE MEASUREMENT OF GASEOUS EMISSIONS
The heavy duty diesel gaseous emission test begins with a warm
engine which is subjected to a prescribed sequence of engine operating
conditions using an engine dynamometer, with continuous analyses of the
exhaust gases. The test is designed to determine the brake-specific
emissions of hydrocarbons, carbon monoxide and oxides of nitrogen when
an engine is operated through a cycle. The cycle consists of three idle
modes and five power modes at each of two speeds which span the typical
operating range of diesel engines. The procedure requires the determi-
nation of the concentration of each pollutant, the exhaust flow and the
power output during each mode. The measured values are weighted and
used to calculate grams of each pollutant emitted per brake horsepower
hour.
The specified equations for the calculations are as follows:
1) Calculate the mass emissions for HC, CO, and NO in grams
per hour for each mode as follows:
HCmass = °-0132 x HcCOnc (ppm~carbonj x exh mass (lb/min)
CO' = 0.0263 x CO (ppm) x exh mass (Ib/min)
niass corxc
NO_ = 0.0432 x NO (ppm) x exh mass (Ib/min)
f. mass cone
138
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 9 of 10
2) Calculate the weighted BHP, HC , CO , and NO_
c , ,,,,.,. .mass mass 2 mass
for each mode by multiplying the value of each by the
weighting factor for that mode. THe weighting factors
are 0.20 for the average idle mode and 0.08 for all other
modes.
3) Calculate the brake specific emissions for HC, CO, and NO
as follows:
BSHC = Z (HC x WF)
mass
X (BHP x WF)
BSCO = T (CO x WF)
*• mass
X(BHP x WF)
BSNO- = £ (NO,, x WF)
2 2 mass
I (BHF x WF)'
It is apparent from these formulas that any error associated
with the measurement of the three factors involved, mass flow, pollutant
concentration, and power will have a direct effect on the brake specific
emissions. In addition as with any system which measures more than one
component, the variability will differ for each component because of the
inherent differences in the instruments and sampling characteristics.
7.2.1 Determination of Brake Horsepower
The observed brake horsepower is calculated by
bhp = RPM • Torque
5252 (Ref. 7-2)
Calibration of the speed and torque outputs are discussed in
section 7.1.1. The tolerances associated with the measurement are
±50 RPM and ±2 percent of the maximum torque at the test speed. The RPM
trace shall have a resolution within 30 RPM and the torque trace shall
have a resolution within 10 ft.-lbs. The deviation of a reading will
vary with the mode and engine horsepower. This deviation could be
appreciable for determination of horsepower on the lower end of the
scale.
A standard practice for reducing torque reading errors is to
hang a reference weight on the torque arm prior to each test. A devi-
ation of more than ±2 percent of calibrated reading indicates a need for
recalibration of the dynamometer.
139
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 10 of 22
Engine speed is cross-checked using a alternate RPM meter such
as a portable tachometer.
Another means of reducing reading errors would be the use of
an expanded range for the lower readings.
7.2.2 Measurement of Exhaust Flow
The Federal Register references the SAE Recommended Practice
No. J244 titled "The Measurement of Intake or Exhaust Flow in Diesel
Engines" for the determination of Mass flow. This referenced paper
proposes three separate metering systems and the associated equipment
required to measure diesel engine gas flows at steady state operating
conditions. Accuracy goals are established, together with procedures
and equipment necessary to obtain these goals. Of the three methods,
the EPA has selected for use the laminar flow element method which has
an NBS traceable accuracy of ±1 percent of the true flow value. A
density factor of 0.0748872 is used to convert CFM to Ibs./min. Suit-
able tolerances for this procedure are recommended in Appendix III of
the Federal Register (Ref. 7-3). A coefficient of variation of 1 per-
cent has been established for this method from Constant Volume Sampler
data (Ref. 7-4).
The mass of air thus determined is added to the mass of fuel
determined directly using a fuel mass flowmeter. The primary reference
standard for this type of meter is a "dead weight" type mass flow device
(See TP-755 and TP-756 in Volume II). Indicated accuracies for the fuel
measurement are generally ±1 percent.
7.2.3 Determination of Exhaust Emission Concentrations
Exhaust emission concentrations are determined using an analyt-
ical 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. Gravimetric stan-
dards 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). Instru-
ment 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 concen-
tration (.Reference 7-4}.
The primary sources of variability in the analytical system
are:
140
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 11 of 22
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 HFID Fuel
o HFID sample temperature
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. In the determination of the effect of error in concentration
measurement, a coefficient of variation of 1 percent of the full scale
is usually encountered in a repetitive determination on the same sample
(Ref. 7-4). However, variation between analytical systems has been
experienced as high as ±3 percent for the same sample. Correlation
values in excess of this are, however, considered to be undesirable 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 emission values.
By measuring the concentration on the lowest convenient range the accuracy
of the data is improved. Instrumentation with a capability of multiple
range selection is available and used by many laboratories, consequently
they are able to use lower ranges than those specified in the Federal
Register.
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. 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 reliabil-
ity desired when obtaining the laboratory standards and "working gases."
141
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 12 of 22
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 necessary.
Calibration curves may be checked weekly by using the primary
set of gases. However, this is a rather lengthy process involving the
use of a large number of cylinders. When correct instrument maintenance
procedures are complied with, the instrument curve shape generally
remains stable for long periods of time. Consequently a simpler process
would be to check the mid-point of the curve using a primary standard
and the span for the most frequently used ranges. These results (the
mid-point deflection) are plotted on control charts (Section 6) for each
instrument. When an "out of control" situation is detected the complete
calibration curve should be repeated and the span gas reanalyzed.
Hydrocarbon emissions from diesel engines require pre-heated
sample lines and detectors because of the low vapor pressure of the
hydrocarbons in the exhaust. An (SAE) recommended control temperature
is 160 ±10 °C (320 ±18°F). A temperature set point control of ±2°C is
desired however maintaining this throughout the sampling system for the
entire 13 mode cycle is a difficult task.
If one compares the boiling ranges of the two fuels (gasoline
75-415 F, Diesel 350-660 °F) it is readily apparent why the elevated
temperature is necessary for the detection of hydrocarbons in diesel
exhaust. Control of this elevated temperature is critical for both
sampling efficiency and detector response. In correlation studies the
temperature set point should be specified.
Another problem associated with sampling diesel exhaust is the
oxygen response of the detector. A procedure for optimizing the per-
formance and determining oxygen response is detailed in Volume II, TP-
754, and SAE J215 and J1003 (References 7-5, 7-6). This problem is not
encountered in light duty testing since the concentration of oxygen in
the sample is close to that of air, and air diluent standards are used
for calibration. These same standards are also used for heavy duty
diesel engine emission testing, but the sample O concentration ranges
from 5 to 20 percent. If the 02 correction is in excess of ±2 percent
over this range a correction factor must be established and used. If it
is in excess of 10 percent the detector or instrument should be replaced.
If the correction factor is required then the uncertainty of the CL
concentration will have an effect on the mass data.
Nitrogen oxides (N0x) are presently being detected utilizing
the NDIR instrument which is specific for NO. Nitrogen dioxide (NO ) is
not detected by this method. NO in the presence of O will oxidize to
NO2. Consequently it is important when using this method to keep the
sample transport time to a minimum by locating the analyzer reasonably
142
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 13 of 22
close to the engine and using high flow rates. Future standards will
most likely be based on the use of a chemiluminescence analyzer using a
converter similar to those presently used in light duty testing. This
would eliminate the NO - NO_ conversion problem and the interference
from other exhaust components associated with the NDIR method.
7.3 THE ENGINE AS A SOURCE OF VARIABILITY
Heavy duty diesel engines are themselves sources of vari-
ability on a test-to-test basis and during the determination of dura-
bility emission data. Emissions are known to change significantly after
a break-in period and over the period of 1,000 hours of durability
tests. Consequently deterioration factors are applied to the data
submitted to EPA for engine certification.
Engine are affected by ambient conditions and the applied
torque of the dynamometers. A reference or correlation engine should be
characterized by varying the various operating parameters within the
allowable operating ranges. A significant number of tests should be run
to characterize the emission profile.
In addition the conditioning and soak time prior to a test are
prime sources of test variability. Engine soak time must be specified
with ample running time prior to soak to assure equilibrium conditions
similar to those encountered in running the engine for 125 hours inter-
vals. The Federal test procedure requires the preconditioning of a
'"warm" engine prior to the start of the test. The definition of a
"warm" engine and a procedure for warming the engine are not specified.
Better definition of this preconditioning procedure would reduce vari-
ability and improve correlation between laboratories.
An illustration of factors which may result in diesel smoke is
presented in Figure 7-2. In most cases these same factors also affect
the gaseous emissions. Practically all of these factors as sources of
variability cannot be controlled by emission test laboratories. Engine
variables are the responsibility of the manufacturer but they do serve
as a reminder that an engine installed in a test cell must be set to
correct engine operating specifications in order to achieve a reliable
determination of the emissions.
7.4 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
143
-------
Figure 7-2
FACTORS RESUMING DT SMOKE AND UNFINISHED COMBUSTION
UNFIHISHBD
COMBUSTION
SOLID SMOXS
(SOOT)
SLUGGISH
COMBUSTION
LATE
IGNITION
FUEL PARTLY
UNZGNITED
LONG IGNITION
LAG
LATH
INJECTION
SHORT
IGNITION LAG
ABSOLUTE
OVERLOAD
SPRAY
IMPINGEMENT
LOCALLY
OVERRICH MIXTURE
INSUFFI-
CIENT
TUR-
BULENCE
EXCESSIVE
PENE-
TRATION
POOR ATOM-
IZATION
HIGH 001
PRESSION
SPRAY
[DISPERSION
HIGH TEM-
PERATURE
EARLY
INJECTION
LOW COM-
PRESSION
LOW TEM-
PERATURE
OVER
COOLED
COMB.
CHAMBER
LOW
CO-IP
RATIO
MALSHWED
COMB. CHAM
DEFECTIVE
NOZZLE
DEFECTIVE
PUMP
Souroei Schweitzer, P. H., Must Diesels Smoke? Diesel Power and Diesel Transportation March, 1947.
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 15 of 22
same engine without changes to hardware or engine 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 7-7) and General Motors (Reference 7-8) in the applications
for suspension of the 1977 Federal Emission Standards. As the emission
requirements become lower, the level of variability significantly affects
the ability to develop and certify emission control systems. Vari-
ability 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 stan-
dard 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 variability 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. In other
words, as the confidence level is increased, the confidence interval
becomes wider. Therefore, in the case of a certification engine 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 cconfident that all engines will meet the Federal
Emissions Standards.
The Coordinating Research Council has carried out a four-phase
cooperative program to evaluate techniques in measuring gaseous emis-
sions in diesel exhaust (Reference 7-9, 7-10).
Each phase involved the measurement of diesel exhaust from a
single engine either by circulating the engine among the laboratories or
by all participants measuring emissions on the same engine at the same
time. In the fourth phase of the program a multicylinder diesel gener-
ating set was circulated among 15 participating laboratories each using
the same procedure recommended by the Society of Automotive Engineers
(J215), In addition to the engine, unknown gas mixtures were also
circulated among the laboratories. Analysis for hydrocarbons were
fairly consistent within laboratories both on unknown gas mixtures and
engine exhaust, with coefficients of variation of 3 percent and 10
percent respectively. However, analysis differed substantially among
145
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 16 of 22
laboratories both on gas mixture and exhaust with coefficient of varia-
tions of 10 percent and 22 percent respectively. The data are presented
in Tables 7-2 and 7-3. There is an indication of need for further
improvement in techniques for analyzing diesel exhaust hydrocarbons, and
especially in the analysis of unknown gas mixtures. With the availability
of gas mixtures with accuracies of better than ±2 percent a coefficient
of variability in the area of 2-4 percent would be expected.
The results for CO and NO were reported for idle, mid-power
and high-power and appear in Tables 7-4 and 7-5. Coefficient of varia-
tions (CV) reported for CO were - Idle 5.7, Mid-power 9.0 and High-power
24 percent. For NO the reported CV's were-Idle 16, Mid-power 13 and
High-power 7.9. This data also indicates a need for improvement
especially in High-power CO mode.
Puther test details and data analyses are contained in the
referenced data and associated CRC reports. The development and improve-
ment of procedures for future heavy duty emissions measurements are
presently being accomplished at the EPA, Ann Arbor facility. Identifi-
cation of specific details of these test procedures and associated
instrumentation will aid in reducing the present interlaboratory test
Variability.
7.5 QUALITY ASSURANCE AND TEST VARIABILITY
Statistical methods that can be used to control test vari-
ability 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 reduction of test variabil-
ity 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. Tables 7-6 and 7-7 are summaries of the test variables and the
methods used for their control.
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.
Test facilities should continually record and assess variabil-
ities from various sources within the test system in a systematic summary.
Periodic review of variability and comparisons with data from other
facilities will indicate which sources of variability have priority for
reduction efforts.
146
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 17 of 22
Table 7-2. PARTICIPANTS' ANALYSES OF BOTTLED GASES
HYDROCARBON CONCENTRATION, ppm C
Laboratory No.*
1
2a
2b
3
4
5
6
7
8
9
10
11
12
13
14
15
16(3)
Average
Range
Standard Deviations
Pooled Within Laboratories
Of Laboratory Averages
Cylinder No. 1
Average
Cylinder No. 2
Average
283
307
308
304
328
304
319
290
228
284
253
327
282
318
292
295
99(34%)**
8(3%)**
27(9%)**
36
7
8
30
9
3
22
22
3
2128
2284
2235
2378
2390
2221
2357
2340
1613
3 2210
5 1898
8 2323
9 2175
9 2412
7 2350
2221
799(36%)**
36(2%)**
212(10%)**
Range
140
45
160
40
22
20
13
64
10
20
63
39
84
30
80
*Laboratory 2 used two analyzers, identified as (a) and (b) . Laboratory 16
is the same as Laboratory 3; it is identified separately because of the
lapse in time between the two sets of tests.
**( ) indicates % of mean value.
(Reference 7-9, 7-10)
147
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 18 of 22
Table 7-3. ANALYSES OF ENGINE EXHAUST
HYDROCARBON CONCENTRATION, ppm C
Idle Mid Power Full Power
Laboratory No. Average Range Average Range Average
1
2a
2b
3
4
5
6
7
8
9
10
11
12
13
14
15
16(3)
Average
Range
330
847
874
709
720
606
1024
696
631
552
814
753
628
754
813
494
672
701
694(99%)
24
55
90
130
135
39
150
132
135
40
126
111
95
26
48
119
83
188
468
409
413
371
398
448
390
325
300
510
485
394
432
412
266
432
391
322(82%)
61
54
68
110
130
63
123
112
90
156
190
246
125
68
78
53
142
204
343
280
319
292
560
330
340
254
246
427
381
338
398
272
247
454
334
356(106%)
40
55
23
45
14
162
116
92
103
84
69
128
90
49
15
80
78
Standard Deviations
Pooled Within Laboratories 46(7%) 57(15%) 35(11%)
Of Laboratory Averages 158(22%) 75(19%) 88(27%)
(Reference 7-9, 7-10)
148
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 19 of 22
Table 7-4. ENGINE TEST RESULTS CARBON MONOXIDE, ppm
PARTICIPANT
NUMBER
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
X
Range , ppm Hi
Lo
%
Standard Deviation,
ppm
%
IDLE*
495
563
551
498
266+
541
561
586
559
443+
565
576
561
518
601
526
550
601
495
19
31
5.7
MID-POWER*
437
505
463
435
194
410
450
494
485
501
622+
551
483
546
?92+
3§8+
480
551
410
29
43
9.0
HIGH-POWER*
759+
400
311
372
159
286
382
544
470
530
615
692+
469
90§+
1197+
375
432
615
286
76
102
24
*Average Value, 4 runs.
+Value Dropped.
(Reference 7-9, 7-10)
149
-------
Section: 7(HD)
Revision: 0
Date: June 1975
Page 20 of 22
Table 7-5. ENGINE TEST RESULTS NITRIC OXIDE, ppm
PARTICIPANT
NUMBER
1
2
3
4
6
7
8
9
10
12
13
14
15
16
X
Range, Hi
Lo
%
Standard Deviation,
ppm
%
IDLE*
_
84
95
78
109
68
75
95
70
138
86
68
73
96
83
109
68
49
13
16
MID-POWER*
_
525
660
502
554
511
475
546
463
571
538
485
387
573
522
660
387
52
66
13
HIGH-POWER*
.
1287
1365
1178
1240
1179
1044
1220
1186
1065
1150
878
645
1103
1183
1365
1044
27
94
7.9
* Average Value, 4 runs.
(Reference 7-9, 7-10)
150
-------
Table 7-6. SUMMARY OF SMOKE EMISSION TEST VARIABLES AND METHODS USED FOR THEIR CONTROL
Ul
TEST VARIABLE
Dynamometer
Engine
Smoke Meter
Drift
Noise
Linearity
Calibration
Recorder
Temperature
Barometer
Humidity
Data
Charts
Computer
Cali-
bra-
tion
X
X
X
X
X
X
X
X
X
Ref. Data
Stan- Vali-
dards dation
X X
X
X
X
X
X
X X
X X
X X
X
X
Daily
Checks
X
X
X
X
X
X
X
X
X
METHOD USED TO CONTROL TEST VARIABLES
Precon-
Monthly Hainte- Train- dition- Corre- Control
Checks nance ing Audit ing lation Charts
x x x x x
X X XX
XX XX
X X
X
X
XX X
XXX
X XX
X X XX
XX X
X X
Instru- Ehviron-
Receiv- ment mental
ing Insp. Range Control
x x
X X
X
X
X x
X x
X x
V o 50 cn
(a fu ft (D
ifl rt < 0
tt> (D H- ft
.. tn H-
M H- 0
M OS
"
NJ C-| O -J
to C -—
-------
Table 7-7. SUMMARY OF GASEOUS EMISSION TEST VARIABLES AND METHODS USED FOR THEIR CONTROL
N)
TEST VARIABLE
Brake Horse-
power
Torque
RPM
Air Flow(LPE)
Fuel Flow
Calibration
Gases
Instrument
Span Gas
Sample System
Zero Gas
HFID Fuel
Operator
Engine
METHOD USED TO CONTROL TEST VARIABLES
Call- Ref. Data Precon- Instru- Environ-
bra- Stan- Vali- Daily Monthly Mainte- Train- dition- Corre- Control Receiv- ment mental
tion dards dation Checks Checks nance ing Audit ing lation Charts ing Insp. Range Control
X X X X X X X X
XXX XX x X
xx xxx x
xxxxx xxx x
x x xxx
x xxxxx x x
xxxxx x x
xxxxx x
x x X x
X X
XX X
X XXXXX X X
•n a yo to
P* (IV Q (D
vQ ft < O
» fl> H-rt
.. (0 H-
.H-O
o
Hi
10
.
S2
(D
s
Ul
a
O
-------
-------
Section: 8(HD)
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 engine 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.
155
-------
Section: 8(HD)
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
156
-------
Section: 8(HD)
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,
157
-------
Section: 8(HD)
Revision: 0
Date: June 1975
Page 4 of 14
calibration and zero gases, fuel, etc., directly affecting
the quality of laboratory testing or results of testing.
Do the procurement orders require test reports or certifica-
tions? 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 utilization 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.
158
-------
Section: 8(HD)
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.
159
-------
Section: 8(HD)
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?
160
-------
Section: 8(HD)
Revision: 0
Date: June 1975
Page 7 of 14
E. Engine Testing
1. "Provision of applicable inspection and test documents."
Request copies of procedures covering the engine 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.
161
-------
Section: 8(HD)
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 an engine 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 an engine are adequately
documented and reported.
10. "Maintenance of accurate and complete test results and
data, with traceability to the tested engines 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 engine 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.
162
-------
Section: 8(HD)
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.
163
-------
Section: 8(HD)
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 (see attached).
164
-------
Section: 8(HD)
Revision: 0
Date: June 1975
Page 11 of 14
QUALITY ASSURANCE SYSTEM SURVEY REPORT
Survey No.
Date of Survey
Type of Survey Q] Initial
| ] Resurvey
Laboratory Name_
Street Address
City
State
zip
Code
Phone
No.
Name of person responsible for quality assurance functions at above address.
Name Title
Parent Organization:
This firm is: Independently
| | Owned & Operated
Affiliate
| | Subsidiary
|~"| Division of
Number of Personnel: Testing_
Personnel Contacted
Engineering
Quality Assurance_
Title
Survey Requested by:
Name Dept.
Div.
Date
Contract/P.O. Number_
Results of Survey^ (Zl Approved Q Conditional Approval [~~j Disapproved
Resurvey Date (For use with conditional approval only) _
REMARKS:
Survey Performed By;
Approved;
Department;
165
Date:
-------
Section; 8(JHD)
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
i
A
C
U
166
-------
Section: 8(HD)
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
ENGINE 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
167
-------
Section: 8(HD)
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
ENGINE 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 retest of instruments
and equipment reworked, repaired or modified after
testing
Maintenance of accurate and complete test results
and data, with traceability to the tested engines
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)
168
-------
Section: 9(HD)
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
171
-------
Section: 9(HD)
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 Storment, J.O. and Springer, K.J., A Surveillance Study of Smoke
from Heavy-Duty Diesel-Powered Vehicles, Environmental Protection
Agency Report No. EPA-460/3-74-003, Final Report EPA Contract
No. EHS 70-109, January, 1974.
7-2 Diesel Engine Emission Measurement Procedure, Society of
Automotive Engineers, Inc., New York SAE J1003, SAE Recommended
Practice, March 1973.
7-3 Federal Register, Volume 38, No. 124, June 28, 1973. p. 17167
7-4 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
7-5 Continuous Hydrocarbon Analysis of Diesel Emissions, Society
of Automotive Engineers, Inc., New York SAE J215, SAE Recommended
Practice, November 1970.
172
-------
Section: 9(HD)
Revision: 0
Date: June 1975
Page 3 of 3
7-6 Application for Suspension Of 1977 Motor Vehicle Exhaust Emis-
sion Standards. Ford Motor Company, Volume 1, Section III-E,
January, 1975.
7-7 General Motor's Request for Suspension of 1977 Federal Emission
Standards, Appendix 20, Volume III of III, January 10, 1975.
7-8 Wagner, T.O., Broering, L.C. and Johnson, J.H., CRC Evaluation
of Techniques for Measuring Hydrocarbons in Diesel Exhaust -
Phase IV, Society of Automotive Engineers, Inc., Warrendale,
Pennsylvania. Publication No. 750203, February, 1975.
7-9 Perez, J.M., Broering, L.C. and Johnson, J.H., Cooperative
Evaluation of Techniques for Measuring Nitric Oxide and Carbon
Monoxide (Phase IV Tests), Society of Automotive Engineers, Inc.,
Warrendale, Pennsylvania. Publication No. 750204.
8-1 Evaluation of a Contractor's Quality Program, Handbook H 50,
April 23, 1965. Department of Defense, Washington D.C.
173
-------
-------
Appendix A-l
SELECTED STATISTICAL TECHNIQUES AND NOMENCLATURE
177
-------
Appendix: A-l
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
X2 + X3
x =
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.
179
-------
Appendix: A-x (HD)
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:
Where P stands for the ordinate of the normal probability
distribution, and a and u. are the standard deviation
x x
and mean of the distribution of x values and
77= 3.14159 and e = 2.71828 are constants.
Variance - A measure of the scatter of observations
around the mean. The variance of the population and of
a sample area and s respectively, i.e.,
21 ^(x -fi)2
a ~ N 4f D
Where
-------
Appendix: ^-i(HD)
Revision: 0
Date: June 1975
Page 3 of 8
s2 =-i-7 ;xn~x>~ . 3 = l,2,3,...n
n-
2
Where s = variance of sample
x. = observed values
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
. n p (1-p)
r! (n-r) !
181
-------
Appendix: AT!(HD)
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
coma'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.
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.
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.
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)%
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 ft), 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,
182
-------
Appendix: A-1(HD)
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
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:
A = X -
fc is
n-1
v/ST
B = X +
t ,s
n-1
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.
183
-------
Appendix: A-l (HO)
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.
184
-------
Appendix: A-1(HD)
Revision: 0
Date: June 1975
Page 7 of 8
"t" 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 p . 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., A 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
3
limit for a range control chart.
D,, - Compute the 3 sigma upper control
limit for a range control chart.
4
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
1 limits for average control charts,
using (T .
A - Compute 3 sigma upper and lower control
2 limits_for average control charts,
using R.
185
-------
Appendix: A-l (HD)
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.
186
-------
Appendix A-2
CONTROL CHART
MULTIPLICATION FACTORS
187
-------
Appendix: A-l(HD)
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
FACTORS FOR CONTROL LIMITS
\ '
3.67
2.39
1.88
1.60
1.41
1.28
1.17
1.09
10 1.03
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
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
°3
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
°4
3.267
2.575
2.282
2.115
2.004
1.924
1.864
1.816
1.777
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 (HD)
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 (HD)
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(HD)
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.
Catn " 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 0 , 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 (HD)
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.
Computer 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
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 (HD)
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 points, 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.
198
-------
Appendix: B-1(HD)
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.
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.
- Bypass of excess sample flow of exhaust gases during analysis.
199
-------
Appendix: B-l (HD)
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 (HD)
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.
Grams 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 (HD)
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 (HD)
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 (HD)
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.
204
-------
Appendix: B-1cu. UU.LXIig a.
after combustion. The sum of the NO and NO concentrations in the
NO^ Analyzer- Analytical instrument used to analyze NO and NO by
chemiluminescence. The formation of NO by the reaction of NO and O
(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 .
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, CO , NO, HO.
Partial Pressure - The pressure exerted by any single gas in a mixture
of gases.
205
-------
Appendix: B-l (HD)
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.
Photo-multiplier - 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 component 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(HD)
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-l(HD)
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.
Stoichiometry - 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 (HD)
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
pres sure.
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(HD)
Revision: 0
Date: June 1975
Page 18 of 18
Vacuum 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 that 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 (HD)
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
GARB 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
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)
HOT 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
LDT Light Duty Testing
Lb. Pound(s)
Max.
Mi.
Maximum
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
0 Oxygen
0_ 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 (HD)
Revision: 0
Date: June 1975
Page 3 of 4
215
-------
Appendix: B-2 (HD)
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
(HEAVY DUTY DIESEL)
219
-------
Appendix: C(HD)
Revision: 0
Date: June 1975
Page 1 of 3
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
221
-------
Appendix: C(HD)
Revision: 0
Date: June 1975
Page 2 of 3
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 Assur-
ance 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 measure-
ment 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 heavy duty diesel engine emission measurement
procedures has been developed and is presented in Volume II of this
report.
222
-------
QUALITY MANAGEMENT PROCEDURES
Change and Revision Summary
Appendix: C(HD)
Revision: 0
Date: June 1975
Page 3 of 3
EPCN
Number
Procedure
Date
Number
Revision
Date
Procedure Title
Entered
By
223
-------
-------
QUALITY MANAGEMENT PROCEDURES
SECTION 1.0
INDEX
227
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
1.1
REVISION DATE
SUBJECT:
TABLE OF CONTENTS
ii
Q.M.P
1.1
2.1
2.2
2.3
2.4
2.5
3.1
3.2
3.3
4.1
4.2
QUALITY MANAGEMENT PROCEDURES MANUAL
Title Page
Changes and Revisions
Introduction
Section 1.0 index
Table of Contents
Section 2.0 Organization
Function/Responsibility Chart
Functional Outline - Administrative
Services
Functional Outline - Laboratory
Operations T- Heavy Duty Engine Testing
Functional Outline - Quality Assurance
Management
Functional Outline - Data Services
Section 3.0 Administration
Preparation of Quality Management
Procedures
Document Control
Quality Assurance Training Program
Section 4.0 procurement
Procurement Document Review
Receiving Inspection
Effective
Date(s)
CONCURRENCES
DATE
IMPLEMENTATION
PREPARED BY:
PAGE
OF
APPROVED BY:
DATE ISSUED:
229
-------
QMP No. 1.1
Page 2 of 2
QUALITY MANAGEMENT PROCEDURES MANUAL (Continued)
Effective
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 Heavy 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
230
-------
-------
QUALILTY MANAGEMENT PROCEDURES
SECTION 2.0
ORGANIZATION
233
-------
o
o
z
70
z
•n
to
to
01
o
>
H
m
QUALITY
ASSURANCE
MANAGEMENT
PROCUREMENT
CONTROL
STANDARDS &
CALIBRATION
TEST QUALITY
CONTROL
CORRECTIVE
ACTION
FACILITY
DIRECTOR
DATA
SERVICES
DATA
VALIDATION
STATISTICAL
ANALYSIS
SYSTEMS
DEVELOPMENT
COMPUTER
OPERATIONS
ADMIN.
SERVICES
TRAINING &
PERSONNEL
PURCHASING
FACILITY
SERVICES
DOCUMENT
CONTROL
LABORATORY
OPERATIONS
TEST
SCHEDULING
EMISSION
TESTING
FUEL S- GAS
STORAGE
to
e=
CO
m
o
o
*3
H
1
w
H
ffl
O
SB
m
5
O
c
H
O
m
Z
m
•o
3D
O
o
m
o
3D
m
CHEMICAL
ANALYSIS
O
to
to
DEFICIENCY
REVIEW
EQU|PMENT
SERVICES
m
>
o
AUDIT
to
o
-------
EPA QUALITY MANAGEMENT
QMP NO. REVISION DATE
PROCEDURE 2.2
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 FUNCTION
Purchasing 1.
2.
3.
Facility Services 4.
5.
6.
7.
Training & Personnel 8.
CONCURRENCES
PREPARED BY:
APPROVED BY:
Purchases all materials, equipment, instruments,
expendable items, office equipment, etc., which
are used by the laboratory.
Requests Quality Assurance to provide quality
requirements and approvals of purchase orders
and related specifications and drawings.
Requests Quality Assurance approval and review
of suppliers' products as required.
Establishes contracts for facility services such
as equipment maintenance and calibration.
Provides for all facility engineering requirements
such as building modifications, plumbing, electrical
wiring, heating, cooling, ventilation and general
storage.
Initiates, recommends, implements safety program
procedures and equipment to meet personnel and
building requirements in accordance with the
applicable regulations.
Controls and maintains inventory of all parts,
supplies, equipment, etc., used by the laboratory.
Maintains records of equipment on loan and surplus
equipment inventory.
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.
DATE IMPLEMENTATION
PAGE 1 OF 2
DATE ISSUED:
237
-------
QHP 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.
238
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
2.3-HD
REVISION DATE
SUBJECT:
FUNCTIONAL OUTLINE - LABORATORY OPERATIONS
- HEAVY DUTY ENGINE TESTING
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 heavy duty engines used in the
performance of these tests.
RESPONSIBILITY
Receiving Inspection
Test Scheduling
Emission Testing
Fuel & Gas Storage
Chemical Analysis
FUNCTION
1. Receives and inspects engines for testing.
2. Prepares and supervises installation of engine in
the test cell and schedules emission testing.
3. Returns each engine in its original condition to
the owner after successful completion of the emis-
sion test.
4. Conducts emission testing on vehicles according to
the government regulations and procedures outlined
in the Test Procedures Manual for Heavy-Duty
Engine Emission Measurements.
5. Measures and reports Heavy-Duty Engine gaseous
emissions, including smoke emission on diesel engines,
according to the Federal Procedures.
6. 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.
7. Completes all required forms and records necessary
for the performance of an emission test.
8. Provides for the proper storage and handling of
Heavy-Duty Engine fuel and high pressure gas
cylinders by initiating detailed procedures contain-
ing Quality Assurance checks to prevent errors such
as the use of improper fuel in the engines.
9. Performs chemical analysis as required for
receiving inspection and non-routine emission
testing.
CONCURRENCES
DATE
IMPLEMENTATION
PREPARED BY:
PAGE i OF 2
APPROVED BY:
DATE ISSUED:
239
-------
QMP NO. 2.3-HD
Page 2 of 2
RESPONSIBILITY
Chemical Analysis
(Continued)
Equipment Services
FUNCTION
10. 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.
11. 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.
12. 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.
13. Maintains records of surplus and loaned equipment
and determines disposition.
14. 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.
15. 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.
240
-------
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 1 OF 2
DATE ISSUED:
241
-------
QMP No. 2.4
Page 2 of 2
RESPONSIBILITY FUNCTION
Audit 8. 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.
9. Conducts intralaboratory and interlaboratory
correlation of emission measurement equipment to
improve the accuracy and reliability of the test
data.
242
-------
EPA QUALITY MANAGEMENT
QMP NO. REVISION DATE
PROCEDURE 2-5
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 FUNCTION
Data Validation 1 .
2.
3.
Statistical Analysis 4.
Systems 5.
Development
6.
CONCURRENCES
PREPARED BY:
APPROVED BY:
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.
Maintains all test data in a data file.
Assists Quality Assurance in monitoring all data to
verify the accuracy and reliability of emission
measurements .
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 .
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 .
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.
DATE IMPLEMENTATION
PAGE 1 OF 2
DATE ISSUED:
243
-------
QMP No. 2.5
Page 2 of 2
RESPONSIBILITY
Computer Operations
FUNCTION
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.
244
-------
-------
QUALITY MANAGEMENT PROCEDURES
SECTION 3.0
ADMINISTRATION
247
-------
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. a. 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.
b. 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
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE 1 OF 5
DATE ISSUED:
249
-------
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
T
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-X.JC X
.Section of the manual (1-7)
-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)
250
-------
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.
251
-------
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.
252
-------
QMP No. 3.1
Page 5 of 5
IV RESPONSIBILITIES AND PROCEDURES (Continued)
E. Flow Schematic - QMP
Quality Assurance
Originates All
QMPs
Reviews Comments
1.
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
£
QA
Audits For
Revision and
Conformance
253
-------
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
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE * OF 3
DATE ISSUED:
255
-------
QMP No. 3.2
Page 2 of 3
IV RESPONSIBILITIES AND PROCEDURES
A. Quality Assurance
1. 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 Control
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.
256
-------
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.
257
-------
EPA 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 also 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.
B. Training Coordinator
1. Determines total time required for training session and personnel
to attend.
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE 1 OF 2
DATE ISSUED:
259
-------
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.
260
-------
-------
QUALITY MANAGEMENT PROCEDURES
SECTION 4.0
PROCUREMENT CONTROL
263
-------
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:
265
-------
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
266
-------
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
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE i OF 3
DATE ISSUED:
267
-------
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.
268
-------
IV RESPONSIBILITIES AND PROCEDURES (Continued)
D. Procurement Control Flow Schematic
QMP No. 4.2
Page 3 of 3
Requestor.
Initiates
Purchase
Request
Quality
Assurance (QA).
Reviews PD
Supplier Audit
Purchasing.
Prepares
Procurement
Document (PD
Vendor.
Ships
Material
Receiving.
Count &
Documents
QA
Inspection
Procedure
Receiving.
Copy of
Approved PD
Purchasing.
Notifies
Vendor of
Defect
QA
Rejection
Report
Receiving
Inspection
Requestor.
Inspection
Report &
Material
QA
Material
Review
QA
Coordinates
Corrective
Action
269
-------
-------
QUALITY MANAGEMENT PROCEDURES
SECTION 5.0
STANDARDS & CALIBRATION
273
-------
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
DATE
IMPLEMENTATION
PREPARED BY:
PAGE 1 OF 8
APPROVED BY:
DATE ISSUED:
275
-------
QMP No. 5.1
Page 2 of 8
Function
Vehicle
Test
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.
276
-------
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.
277
-------
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).
278
-------
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.
279
-------
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.
280
-------
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.
281
-------
QMP NO. 5.1
Page 8 of 8
IV RESPONSIBILITIES AND PROCEDURES (Continued)
N. Flow Schematic - Equipment Calibration Control
T
Calibration
Tag
jntrol
(RC)
Le by
'
Equipment
Stores
*~
Loan Order
Copy 2
1— *
Loan Order
Copy 1
Assi
Retu
Inst
gnee
rns
rumen t
9
Calibration
Tag
i
Calib
Order
*
Equipment
Stores
Recc
Cont
'Record
-i
Calibration
Control Card
282
-------
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
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE l OF 3
DATE ISSUED:
283
-------
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.
284
-------
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
285
-------
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 1 OF 2
APPROVED BY:
DATE ISSUED:
287
-------
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.
288
-------
QMP No. 5.3
Attachment
OUT OF TOLERANCK 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%
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
5.4
REVISION DATE
SUBJECT:
CALIBRATION STANDARDS
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 1 OF 2_
DATE ISSUED:
291
-------
QMP No. 5.4
Page 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.
292
-------
-------
QUALITY MANAGEMENT PROCEDURES
SECTION 6.0
LABORATORY OPERATIONS
295
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
6.1-HD
REVISION DATE
SUBJECT:
HEAVY DUTY TEST OPERATIONS
I 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 capability
of the measurement system.
Ill SCOPE OF APPLICATION
A. All phases of Heavy-Duty Engine 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 Heavy-Duty Engine 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.
3. Schedules engine for test and subsequent return to the manufacturer.
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE 1 OF 4
DATE ISSUED:
297
-------
QMP NO. 6.1-HD
Page 2 of 4
IV RESPONSIBILITIES AND PROCEDURES (Continued)
B. Receiving Inspection
1. Receives and inspects the scheduled engine and completes the
vehicle receiving inspection documentation.
C. Data Validation
1. Verifies accuracy and satisfactory completion of engine
receiving inspection and required installation specifications
and equipment.
D. Heavy Duty Engine Testing
1. Installs engine test bed on engine dynamometer.
2. Attaches fuel lines, thermocouples, air flow devices, etc. as
required.
3. Verifies that proper amount and type of fuel is used.
4. Performs engine preconditioning checks, engine performance
checks, engine preconditioning test and completes applicable
preconditioning reports.
E. Data Validation
1. Verifies that all elements of the preconditioning require-
ments have been satisfactorily and accurately completed and
authorizes engine to next test function.
F Heavy Duty Engine Testing
1. Performs smoke test on all heavy duty diesel engines. Tests
according to procedures contained in subpart I of the
Federal Regulations and the Laboratory Test Procedure Manual.
2. Performs the 13-mode gaseous emission test cycle for all diesel
engines and the 9-mode cycle for all gasoline powered engines.
3. Performs sample analysis in accordance with prescribed pro-
cedure and documents all data, including ambient conditions
and instrument operating parameters.
4. Submits all documentation to data validation.
298
-------
QMP No. 6.1-HD
Page 3 of 4
IV RESPONSIBILITIES AND PROCEDURES (Continued)
G. Data Validation
1. Checks all data for completeness and accuracy and forwards to
Data Services for preliminary processing.
H. Quality Assurance
1. Maintains continual surveillance over the functions associated
with the performance of all phases of the heavy duty engine
emission testing program.
2. Assures proper and current calibration of instruments and
equipment used in heavy duty engine testing.
3. In the event of a test failure, whether instrument, dynamometer
operator, or engine, 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.
I. Data Validation
1. Checks all submitted data and heavy duty engine information
for completeness and accuracy.
2. Forwards all data to Data Services for final processing.
J. Heavy-Duty Testing/Shipping
1. After satisfactory completion of all tests and restoration of
the engine to the "as received" condition, completes shipping
order and returns engine to the manufacturer/owner.
299
-------
QMP No. 6.1-HD
Page 4 of 4
K. HEAVY DUTY TEST FLOW SCHEMATIC
Engine
Received
(Preliminary
Checkout)
No
Engine
Manufacturer
Engine
Dynamometer
Checkout
Engine
Preparation
and
Installation
Engine
Operational
Engine
Shipping
No
Gasoline
Engine
Gaseous
Emission Test
(9-Mode)
Q.A. Performs
Failure
Analysis
As Required
Failure
Report
Q.A. Audits
and
Analyzes
Data Files
Diesel Engine
Gaseous
Emission Test
(13-Mode)
Q.A. Peforms
Failure
Analysis As
Required
Failure
Report
Q.A. Performs
Failure
Analysis As
Required
Failure
Report
Laboratory
Operations
300
-------
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
APPROVED BY=
DATE ISSUED:
301
-------
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.
302
-------
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
Document
Control
Assigns EPCN
Distributes
Draft
Other Affected
Departments
Reviews
Draft EPCN
Quality
Assurance
Reviews
Draft EPCN
Yes
Revision
Required
J
Lab. Operations
Reviews
Comments
Document
Control
Drafts Change
and Distributes
Affected
Departments &
Manual Holders
File Revisions
/
Quality
Assurance
Audits for
Implementation
303
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
6.3-HD
REVISION DATE
SUBJECT:
TEST ENGINE FUEL CONTROL
I PURPOSE
This QMP establishes the requirements for the control of fuel used in the
performance of a Heavy Duty Engine 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. Some of
the 1975 engines may be equipped with catalyst devices which become
inoperative 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 engine.
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 additives should be avoided since this may
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)
gasoline. Type 1-D and 2-D diesel fuels used in the emission test.
B. Leaded fuel must not be used in engine equipped with catalyst devices.
PREPARED BY:
APPROVED BY:
ENCES
DAT
IMPLEMENTATION
PAGE 1 OF. 4
DATE ISSUED:
305
-------
QMP No. 6.3-HD
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.
306
-------
QMP No. 6.i-HD
Page 3 of 4
IV RESPONSIBILITIES AND PROCEDURES (Continued)
D. Chemical Analysis
1. Analyzes all received fuel for lead phosphorous and sulphur
content. Unleaded fuels shall not be released for use until lead
content is verified to be within specifications. Diesel fuels-
flash point, distillation curve, and total sulphur are commonly
determined "in-house."
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. Test Scheduling
1. Stamps fuel requirements, clearly, on all work sheets and check
lists associated with each test engine.
F. Heavy Duty Engine Testing
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 engine 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.
307
-------
QMP No. 6.3-HD
Page 4 of 4
IV RESPONSIBILITIES AND PROCEDURES (Continued)
G. Test Engine 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
No
Prod. Control
Identifies
Fuel
Requirements
HD Operations
Storage &
Color Code
Yes
HD Operations
Ensure Proper
Use of Fuel
in Engine
Q.A.
Review
308
-------
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
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE * OF 3
DATE ISSUED:
309
-------
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.
310
-------
QMP No. 6.4
Page 3 of 3
IV RESPONSIBILITIES AND PROCEDURES (Continued)
E. Typical Scheduling Flow Schematic
Test
Requester
Initiates
lest Request
Laboratory
Operations
Quality
Assurance
Production
Control
Work Order
Each Test
Production
Schedule
Shippinq &
Receiving
Yearly
(updated
monthly)
Weekly
Vehicle
Test
Daily
Test Operations
Lt. Duty
Testing
E. & D.
Testing
Heavy Duty
Testing
4J
to
4)
\
Test Validation
Tech.
EPA Rep.
Man. Rep.
-P
U)
0)
EH
•d
•H
311
-------
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:
313
-------
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.
314
-------
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.
315
-------
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.
316
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
6.6-HD
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 Heavy Duty Gasoline and Diesel Engine
Emission Tests.
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, a retest schedule.
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 Heavy Duty Engine 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. Checks inspection, and engine preparation, and engine check-out
information for completeness, authorization, and compliance to
specs.
2. Checks analyzer traces for obvious errors and compare chart values
with those logged on the Analyzer Read-out form.
3. Checks temperature, engine RPM and load traces for conformance
with prescribed cycle conditions.
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE 1 OF 3
DATE ISSUED:
317
-------
QMP No. 6.6-HD
Page 2 of 3
IV RESPONSIBILITIES AND PROCEDURES (Continued)
A. Data Validation (continued)
4. Checks for completeness of all required information on data sheet.
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 Service for preliminary analysis.
2. Obtains preliminary results and makes final check of data.
3. If errors are discovered, corrections are made and corrected
data sheet is re-routed to the Data Service for a new print-
out or calculation.
4. Checks off the remaining documentation and enters it in the
data file.
5. Requests schedule of retest if test is declared invalid.
6. Releases engine to shipping for return shipment to manufacturer and
submits valid data with authorized signature or mark to laboratory
operations for distribution.
D. Quality Assurance
1. Reviews and evaluates all heavy duty emission testing procedures
with reference to error and bias in collection, handling and
analysis of samples.
E. Quality Assurance
1. Reviews all data records at regular intervals for possible human
error such as:
a. Failure of technician to record pertinent information.
318
-------
QMP No. 6.6-HD
Page 3 of 3
IV RESPONSIBILITIES AND PROCEDURES (Continued)
E. Quality Assurance (continued)
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.
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.
F. Data Flow Schematic
See QMP 6.1-HD
319
-------
-------
QUALITY MANAGEMENT PROCEDURES
SECTION 7.0
FORMS INSTRUCTIONS
323
-------
QMP NO. REVISION DATE
EPA QUALITY MANAGEMENT PROCEDURE 7.1
SUBJECT:
1. FORM
2. FORM
2.1
3. FORM
3.1
FORMS INSTRUCTION - INSTRUMENT LOAN ORDER
NUMBER 7.1:1-31-75
USE: QMP 5.1
To provide a record of the individual and organization having custodial
responsibility for equipment requisitioned from Standards and Cali-
bration Equipment Stores.
INSTRUCTIONS
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
APPROVED BY
: PAGE JL OF i
: DATE ISSUED:
325
-------
QMP NO. 1.1
Attachment No. 1
INSTRUMENT
FORM
MAN NO.
LOAN ORDER
©
ORGN.
1
CONTROL NO.
©
DATE
KIND OF INSTRUMENT
©
WORKMAN
until it
to you.
Employee
Supervisor.
NOTE: This instrument is in your charge
is returned. |f lost, it will be charged
Keep this slip until instrument is returned.
©
(T)
FORM NO. 7.1:- 1-31-75
326
-------
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^ 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
verified as being discrepant.
CONCURRENCES
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTAT ION
PAGE 1 OF l
DATE ISSUED:
327
-------
•n
CO
to
I
Ol
(1)
1 1 1 1 1
1 1
c(ff- CONTROL NO. '"
(4)
i 1 1 1 '1
1 1
** MODEL M
(2)
1 1 1 1 1
1 1 1 1
1 1 1
"wai1 fS~ NOMENCLATURE "
(5) (6) (7)
1 1 1 1 J_i_
" TYPE *'• " MFR. »• "'CYCLE "
(8)
1 1 1.
"ORGN. HO."
(9)
1 1
*• FAC. "
s
(3)
1 1 1
it
(10)
J ' '»
18 DUE
1
1
1
MANUFACTURE
(11
1
JN
(12
1
"our"
[13
|
"REJ.
II
€iyilJlTICy GOiiNiiliiL CAi&D
CONTROL NO.
8888888888
9999999999
i • i i » i i i » 11
D
H
88
9 9
NOMENCLATURE MANUFACTURE
8888888888888888888888
8881888
8
JVi 9"9 9999999993999999999333993
D u » u ii it » n 11 n n ;i n n 11 11 n n » » n u " » >' " "i " «'
MODEL
8888
888888
9999999999
1) l| I) l| <1 ll II U il U
TYPE
888
999
'.i 11 n
MFR
888
999
u » u
:YCU
988
n 9
IU II
ORGN.
8888
9999
u n u ii
FAC
188
i 9 U
in ii
s
8
9
i)
DUE
8888
9999'
ii n n n
IN
88
9 9
u n
OUT
88
9 9
u "
REJ
88
9 9
il n
8
9
N
-------
EPA QUALITY MANAGEMENT PROCEDURE
SUBJECT:
FORMS
1. FORM
2. FORM
2.1
2.2
3. FORM
3.1
INSTRUCTION - CALIBRATION ORDER
NUMBER: 7.3: 1-31-75
USE: QMP 5.1, 5.3
QMP NO. REVISION DATE
7.3
To recall equipment for periodic calibration/maintenance.
To authorize receiving inspection
equipment.
INSTRUCTIONS
, repair or special calibration of
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
equipment .
8. Fac - Identification of the
organization having assignment 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
PREPARED BY:
APPROVED BY:
DATE IMPLEMENTATION
PAGE 1 OF 3
DATE ISSUED:
329
-------
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/To1 - 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).
330
-------
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.
331
-------
RECEIPT
COPY
Ul
u>
NJ
UP COPY
s
§
H
Ul
H1
Ul
. . * i n M ORDER rwLL«" -j__zr
CALIBRATION ORDER RECORD COPY 1
CONTROL NO. NOMENCLATURE MAMI i«rAr--n IOITD ,,~~,-, 1
(5)
DUE
SIGNATURE
REMARKS
c
COMP. BY
MAT'L COST
(6) | (7) (8) (9J) ]REPAijR£lO) 1 RECALL (if •j"!
CALIB j ORGN FAC S 1 ' f BUY IN (^ \ OTHPR jf ]o ) j
(1*0 DATE (15) ]EXT- (l6j 1 RETURN INSTRUMENT: YES [ f NO
/-1Q\ " ~ .. .. - 1 1 j_
(17^flR^
L J
/
' '
(20) j_DATE (21) ' CONDITION RECEIVED: IN/TOL./g^) OUT/TOL* (^
(2k) ; TIME EXPENDED (2=;) i NEXT SERVICING DATE ( o£\
*OUT/TOL. DETAILS (gj) " +*-"-' ~ - -----
"
O/T REVIEW
BY (28) DATE (29)
ft
O Z
"
rt •
Ul
Z
O
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
7.4
REVISION DATE
SUBJECT:
FORMS INSTRUCTION - TEST CONDITION REPORT
1. FORM NUMBER: 7.4:1-31-75.
2. FORM USE: QMP 6.1, 6.6.
2.1 The Test Condition Report (TCR) is used to record details of any
failures that occur during Mobile Source Emission Testing.
3. FORM INSTRUCTIONS:
3.1 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:
PAGE
OF
APPROVED BY:
DATE ISSUED:
333
-------
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.
334
-------
TEST CONDITION REPORT
QMP NO. 7.4
Attachment No. 1
Failure
Name
Branch
Test Type:
VV Void
(4)
(
-------
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. Effectivity - 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 i OF 2
APPROVED BY:
DATE ISSUED:
337
-------
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.
338
-------
QMP'tNO. 7.5
Attachment No. 1
w
EQUIPMENT/PROCEDURE CHANGE NOTICE
DATE
EPCN NO.
00
PAGE
OF
I. ORIGINATOR (Name)
2. PHONE EXT.
3. DATE REQUIRED
4. TYPE OF CHANGE
Q EQUIPMENT
PROCEDURE D OTHER
5. REFERENCES
6. CHANGE REQUESTED BY (Name)
7. PURPOSE OF CHANGE
8. DESCRIPTION OF CHANGE (Attach details, specifications, or drawings if necessary)
9. EFFECTIVITY (Date or Other)
10. DURATION OR EXTENT OF USE (See 9.)
Q PERMANENT D TEMPORARY _
11. AREAS OF MSAPC AFFECTED BY THIS CHANGE
D LOT D ESO D CHEM Q LAB Q QC/QA
DHDT fji&E QCSM DOATA DECTD
D OTHER
12. REVIEWS AND APPROVALS
REVIEWED BY
A.
B.
C.
0.
E.
^PKOrWROVEDrPLEASE DISCUSS" DETA
DATE
LS"~ON"REVERSE"S1
ALL REVIEWS AND APPROVALS HAVE BEEN RECEIVED
/WP DOCUMENTED
APPROVED BY *
F.
G.
H. '
I.
J.
)E)
13. QC/QA MANAGER
FINAL IMPLEMENTATION
THE PROVISIONS OF THIS EPCN ARE APPROVED AMD HEREBY
IMPLEMENTED
14. LAB BRANCH CHIEF
DATE
DATE
DATE
FORM NO. 7.5; 1-31-75
339 DISTRIBUTION: ORIGINATOR
QC/QA OFFICE
DIVISION FILES
AREAS AFFECTED
-------
QMP NO. REVISION DATE
EPA QUALITY MANAGEMENT PROCEDURE 7.6
SUBJECT:
FORMS INSTRUCTION - QMP CHANGE AND REVISION SUMMARY
1. FORM
2. FORM
2.1
3. FORM
3.1
NUMBER: 7.6:1-31-75.
USE: QMP 3.1.
To provide a summary of changes and revisions to QMP's.
INSTRUCTIONS
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
APPROVED BY
: PAGE^1 OF...1
: DATE ISSUED:
341
-------
QMP NO. 7.6
Attachment No. 1
QUALITY MANAGEMENT PROCEDURES
Change and Revision Summary
EPCN
Number
0)
Date
00
Procedure
Number
(3)
Revision
Date
(4)
Procedure Title
(5)
Entered
By
(<»)
FORM NO. 7.6: 1-31-75
342
-------
EPA QUALITY MANAGEMENT PROCEDURE
QMP NO.
7.7
REVISION DATE
SUBJECT:
FORMS INSTRUCTION - REJECTION REPORT
1. FORM NUMBER: 7.7:1-31-75
2. FORM USE: QMP 4.2
2.1 The Rejection Report is used to document, identify and withhold
discrepant materials and equipment.
3. FORM INSTRUCTION
3.1 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
PREPARED BY:
APPROVED BY:
DATE
IMPLEMENTATION
PAGE JL OF 2
DATE ISSUED:
343
-------
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.
344
-------
QMP NO. 7.7
Attachment No. 1
REJECTION REPORT
NO. 11 101
PART NUMBER PART NAME SUPPLIER/MFR
CD (2) (3)
REJECTED CONTRACT PURCHASE ORDER NO. REC . REPORT NO.
QUANTITY^
ITEM NO.
(*>
DATfTS (Q (7) (8)
DISCREPANC ES
(to)
i
j
!
i
REJECTED BY DATE SUPERVISOR APPROVAL DATE Q.A. APPROVAL DATE
00 0*> 03) O) (15) (ft)
DISPOSITION (17) CHECK IF FAILURE ANALYSIS REQUIRED <}*> •
USE AS IS
RETURN TO
FAILURE ANALYSIS REPORT NO .
SUPPL ER
OTHER (SPECIFY) CORRECTIVE ACTION (l9)
O.A. APPROVAL feo) DATE fcl)
FORM NO. 7.7s 1-31-75
345
------- |