United States
Environmental Protection
Agency
Office of Mobile Sources
Emission Control Technology Division
2565 Plymouth Road
Ann Arbor, Michigan 48105
EPA 460/3-85-009b
September 1985
Air
c/EPA
Outdoor Smog Chamber Experiments:
Reactivity of Methanol Exhaust
Part II: Quality Assurance and Data
Processing System Description
-------
Outdoor Smog Chamber Experiments:
Reactivity of Methanol Exhaust
Part II: Quality Assurance and
Data Processing System Description
H. E. Jeffries, K. G. Sexton,
R. M. Kamens, M. S. Holleman
Department of Environmental Sciences
and Engineering
School of Public Health
University of North Carolina
Chapel Hill, N.C. 27514
Prepared under Subcontract with
Southwest Research Institute
Contract No. 68-03-3162
Work Assignment 30
EPA Project Officer: Craig A. Harvey
Technical Representative: Penny M. Carey
Prepared for
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Mobile Sources
Emission Control Technology Division
2565 Plymouth Road
Ann Arbor, MI 48105
September 1985
-------
DISCLAIMER
This report has been reviewed by the Emission Control Technology Division, U. S. Environ-
mental Protection Agency, and approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or recommendation for use.
-------
Abstract
This report describes the Quality Assurance and Data Processing procedures and systems used
at the UNC Outdoor Smog Chamber Facility. The primary product of research conducted at this
facility is information in the form of measurements of reactants and products in photochemical
systems and measurements of the critical parameters that influence the chemical transformations
system.
Generating useful data begins with understanding the goals of the project and the special
needs and concerns of conducting a successful smog chamber operation. The system components
are designed to collect, transfer, process, and report accurate, high resolution data without loss
or distortion. The system components in the Quality Assurance and Data Processing system are:
people, hardware, software, checklists, and data bases.
Quality assurance checks are made at every level of the program. Pressurized gas tank and liquid
mixtures were used to establish experimental conditions of HC assuring consistency throughout
the program. Several NBS traceable standards and liquid injections into the chamber used for
calibration have been intercompared and show good agreement. Resulting calibration data indicates
good instrument stability during the program.
The document describes the calibration techniques and data processing procedures, and the
processing of experimental data. A description and examples of the format of the final documentation
and data file are given. The last chapter summarizes the Quality Assurance steps and important
accuracy and precision aspects of the data quality.
in
-------
IV
-------
Contents
1 Overall QA Steps .1
Introduction 1
Clearly Established Goals 1
Facilities and Analytical Support Requirements 2
Planning of Activities 2
Calibrations 2
Characterization Runs 2
Special Mixtures 5
Special Techniques 5
Reviewing and Scheduling Methods 5
Priority of Runs 5
Use of NOAA Weather Radio 5
Feedback Scheduling 6
Site Operation 6
Lab Air Conditioner 6
Smog Chamber 7
Chamber Sampling Manifold 7
Chamber AC Drying System . 7
Ethylene Converter 7
Mass Balance 7
Supplies 8
Check Lists 8
Setup Checkout Lists 8
Operator Checkout List 8
Run Checkout List 8
Run Packup List 8
Preventative Maintenance 9
Characterization Experiments 9
Types of Characterization Experiments 9
Frequency 10
-------
2 System Components 11
Location of Work 11
Experimental Site 11
Data Processing Location 11
Personnel 12
Advisory Group, AG 12
Administrative Research Assistant, AA 12
Project Coordinator, PC 12
Site Operator, SO 12
Peak Picker, PP 12
Computer Technician, CT 12
Computers Used 12
DEC PDP-11/40 13
DEC LSI-11/23 13
IBM PC . . 13
DEC VAX-11/780 13
Computer Programs Used 14
Description of General Purpose Programs 14
Description of Specific Programs 14
Description of Data File Formats and Contents 15
"U" file 15
"G" file 15
"P" files 16
"C" files 16
"K" files 17
"R" files 17
"Q" files 17
"A" files 17
"E" files 17
ACDTR.COM file 17
Description of Databases 17
Description of Database Procedures 18
List of all DTR Procedures by Name and Function 18
Worksheets/Databases 19
3 Calibration 21
Approach 21
vi
-------
NBS-Traceable Sources 21
NO and O3 Sources 21
Hydrocarbon Sources 22
Transfer Standards Created from Primary Sources 23
Primary Source Comparison 23
NO and O3 Sources 23
Hydrocarbon Sources 27
Transfer Calibration to working "AUTOCAL" Sources 27
NOx and O3 Calibration Sources 27
Manual Gas Phase Titration 27
Hydrocarbon Sources 31
Routine NOx and O3 Calibrations Methods 40
Automatic Source Sampling 40
Routine HC Calibrations 40
Automatic Source Sampling 41
Manual Precision Liquid Injections 41
Meteorological Sensor Calibrations 42
Solar/UV radiation Calibrations 42
Temperature Calibrations 42
Dewpoint Calibrations 42
Flowrate Measurements - Calibration 43
Time Measurement - Clock Calibration 43
Calibration Documentation and Processing Databases 43
Paper Forms 43
Electronic Forms and Other Documentation 44
Calibration Data Processing Status Database (fALEUTRY 48
Calibration Databases 48
Security of Calibration Databases 52
Calibration Data Processing 53
NOX and O3 53
HC Processing 54
4 Experimental Data 71
vii
-------
Experimental Results Documentation 71
Data 71
Forms 73
Plots 77
Run Data Processing Status Databases 77
Description of Data Processing Status Database 89
Description 89
Use of Data Processing Status Database 89
Digital Voltmeter Data Processing . . . ' 91
Strip Chart Data Processing 95
Documentation Processing 99
General Documentation Form 100
General Description and Purpose of Experiment 100
Initial Conditions 100
Meteorological Conditions 102
Data Times, Data Exceptions, Special Problems and Concerns 102
Quality Assessment 102
Documentation File Production 104
Final Segmented File Production 105
Security of Data 107
Inventory 107
Retention of Data 107
Backup of Data 107
5 Segmented Data File 108
General Description 108
File Formats 109
ANSI Tape File Format 112
Tape Contents 113
Final Data Plots made from SegFile 113
Vlll
-------
6 Summary of QA 124
Summary of Calibration and Data QA 124
List of QA Steps 125
Before Runs 125
After Run-Experiment Logging 125
During Calibration 126
During DVM Data Processing 127
During GO Data Processing 128
During Documentation Stages 129
After SegFile Made 130
APPENDICES
A Site Checklists 132
Checklist For 0900 EDT Site Operator 132
Outside Check 132
TSR and UV zero check 135
GC Check 135
Checklist for Leaving the Site Set Up for a Run 137
Computer 137
Formaldehyde 137
NOx and O3 Instruments 137
TSR and UV Chart 138
Charts 138
Carle I 139
Carle II and Carle III 139
Sigma 10 Integrator 140
Dew Point Meter 141
ATC and Varian 141
CO Meter 141
Air Generators 141
General 142
Early Morning Checkout 144
Run Pack-Up Checklist 146
B HC Tank Calibration Source Certification 147
C Official Calibration Sources 168
D CALANA Plots and Analysis Reports 178
ix
-------
Tables
1. Methanol Target Experiments 3
2. NO Tank Calibration Sources 22
3. HC Tank and Liquid Calibration Sources 22
4. OCS from Injections in Chamber 24
5. Species ID Numbers and Properties 26
6. Comparison of Primary NO Calibration Sources 29
7. Intercomparisons Among Sources 39
8. Comparison of Primary HC Calibration Sources 39
9. Processing System for Instrument Auto-Calibration Data 49
10. CALSUM Example 50
11. NOX and 03 Calibrations for 1984 56
12. Processing System for Instrument Auto-Calibration Data 61
13. Summary of CALANA statistics 67
14. Record Description for RUNENTRY Database System 81
15. RUNENTRY Processing Summary for September 14, 1981 82
16. Processing System for DVM Data 93
17. Processing System for Instrument Data 97
18. Documentation Steps 100
19. Final Segmented File Production Steps 106
20. Example Segmented Data File 110
21. VAX/VMS ANSI VOLl Label 120
22. First File Header Label (HDRl) Fields 121
23. Second File Header Label (HDR2) Fields 122
24. Third File Header Label (HDR3) Fields 123
-------
Figures
1. Comparison of Primary NO Calibration Sources 28
2. NOX and O3 Cal Processing Flowchart. 32
3. LOTUS Spreadsheet for Manual Cal Processing 33
4. LOTUS Analysis Spreadsheet of Manual Cal 35
5. LOTUS Analysis Spreadsheet Manual Cal Plots 37
6. Official Calibration Source Form 45
7. Cal Pick Instruction Form 47
8. AUTOCAL Processing Flowchart 55
9. HC Autocal Processing Flowchart 60
10. CALANA Example Figures 64
11. Calextr Examples 69
12. Runsheet 74
13. Instrument Checklist Form 75
14. Run Folder Inventory Checksheet Form 76
15. Data Processing Instruction Form 78
16. Runentry Welcome and Main Menu Screens 83
17. RUNENTRY Date and Display Selection Screens 84
18. RUNENTRY General, Documentation, and File Status Screens 85
19. RUNENTRY DVM and Picked Instrument Status Screens 86
20. PC Report Example 87
21. DVM Processing Flowchart (Computers and Files) 92
22. Instrument Processing Flowchart (Computers and Files) 96
23. Documentation Processing Flowchart (Computers and Files) 101
24. Basic Layout of a VAX/VMS ANSI Labeled Volume 114
25. Single File/Single Volume Configuration 115
26. Multifile/Multivolume Configuration 116
27. VAX/VMS ANSI VOL1 Label Format 116
28. HDRl Label Format 117
29. HDR2 Label Format 118
30. Blocked Fixed-Length Records 119
31. Variable-Length Records 119
XI
-------
Acknowledgements
This project was assisted by several university staff and many students. Kenneth Crossen,
Systems Programmer, wrote the data collection computer code, the data transfer computer code and
many utility programs. Thomas Morris, Computer Programmer, maintained the site data collection
computer code and hardware. Randy Goodman, Research Technician, renovated the laboratory
bench and instrument support systems, the chamber/laboratory manifold, the smog chamber and
the chamber drying system. The site operations were managed by John Suedbeck.
The following students performed data processing tasks: Lynn Clark, Jeffrey Hoffner, Charles
McDowell, Jennifer Jeffreys, Greg Yates, and Cindy Stock. The following students performed ex-
periments at the site: Jeff Arnold, David Benham, Lisa McQuay, and Joe Simmeonssen.
xn
-------
Overall QA Steps
Introduction
There are a number of general activities that contribute to overall quality in the
experimental data. These include:
having clearly established experimental program goals;
establishing facilities and analytical support requirements;
planning activities for experimental and quality assurance programs required to
achieve goals;
reviewing and scheduling activities;
maintaining basic Research Site Facility;
using checklists;
using preventative maintenance;
performing special test experiments routinely.
Each of these will be discussed further.
Clearly Established Goals
The most basic part of a QA program is to assure that efforts are properly designed
and directed towards clearly defined goals. This assures that the data obtained will
satisfy the needs of the program. The approach and design of the experimental
program and the selection of experiments are as important as the accuracy and
-------
Overall QA Steps Facilities and Analytical Support Requirements
precision aspects of the final data. A range of experiments conducted with appro-
priate chemical conditions must be identified. Minimum data requirements must
be determined beforehand to assure that the results will be useful to achieve the
program goals.
The experimental conditions of interest must be identified and prioritized. The
number of experiments to be performed are decided based upon program goals,
approach, design and resources.
Once an experimental program is designed, a table of Prioritized Target Condi-
tions can be made to guide the experimental program. An example table is shown
in Table 1.
Specific experimental conditions of species (and sources) and amounts (cham-
ber concentrations; milligrams of solids, microliters of liquids, or seconds at specific
flowrates of gases) are calculated by the project coordinator and recorded on the
Run Sheet used to conduct the experiment to assure that the proper experiment is
conducted. This form will be discussed further in the chapter entitled "Experimen-
tal Data Processing".
Facilities and Analytical Support Requirements
The facilities and analytical support requirements must be identified. Special mod-
ifications or implementation of new analytical methods consume resources. These
activities could potentially hinder the attainment of the project goals if problems are
encountered. Facilities and analytical capabilities are described in two Appendices
attached to each Final Report.
Planning of Activities
Calibrations
All instruments must be calibrated frequently to assure accuracy and to determine
precision. Along with scheduling the experimental program, scheduling of calibra-
tions are required to assure that the data can be processed with minimal uncertainty.
Calibration procedures will be discussed in the next chapter.
Characterization Runs
Several types of smog chamber experiments are generally conducted to characterize
aspects which can effect interpretation of experimental results and aid modelers.
-------
Table 1. Methanol Fuel Reactivity Experiments
(AM experiments have 0.35 ppm NOX)
Num
1
2
3
4
5
6
TYPE
Reduction, 33%
mid-ratio
Substitution
mid-ratio
normal HCHO
Substitution
mid-ratio
low HCHO
Substitution
mid-ratio
high HCHO
Reduction, 33%
low-ratio
Substitution
low-ratio
normal HCHO
REACTANTS
First Side
3.00 ppmC SynAuto
3.00 ppmC SynAuto
3.00 ppmC SynAuto
3.00 ppmC SynAuto
1.000 ppmC SynAuto
1.000 ppmC SynAuto
REACTANTS
Second Side
2.00 ppmC SynAulo
2.00 ppmC SynAuto
0.89 ppmC MeOH
0.01 ppmC MeNO2
O.lOppmC HCHO
2.00 ppmC SynAuto
0.99 ppmC MeOH
0.01 ppmC MeN02
2.00 ppmC SynAuto
0.79 ppmC MeOH
0.01 ppmC MeNO2
0.20 ppmC HCHO
0.666 ppmC SynAuto
0.666 ppmC SynAuto
0.300 ppmC MeOH
0.003 ppmC MeN02
0.030 ppmC HCHO
PURPOSE
To determine the effect of 33% reduction in HC in an auto-
ez/iaust-like environment at atypical HC-to-NOx ratio. Expect
30% reduction in ozone maximum.
To determine the reactivity of the most likely methanol fuel
exhaust in an auJo-ez/if!t<«>-like environment at a typical HC-
to-NOx ratio. Expect 20% reduction in ozone maximum.
To determine the reactivity of the lowest reactivity methanol
fuel exhaust in an «u(o-ez/iaust-like environment at a typical
HC-to-NOx ratio. Expect 30% reduction in ozone maximum.
To determine the reactivity of a highly reactive methanol fuel
exhaust in an outo-ex/ioust-like environment at a typical HC-
to-NOx ratio. Expect less than 10% reduction in ozone maxi-
mum.
To determine the effect of 33% reduction of HC in an auto-
ei/iaust-like environment at a low HC-to-NOx ratio. Expect a
large reduction in ozone maximum.
To determine the reactivity of the most likely methanol fuel
exhaust in an auto-ezJioust-like environment at a low HC-to-
NOx ratio. Expect a large reduction in ozone maximum.
-------
Table 1, cont'd. Mcthanol Fuel Reactivity Experiments
(All experiments have 0.35 ppm NOX)
Num
7
8
9
10
It
12
TYPE
Reduction, 33%
in id-ratio
Substitution
mid-ratio
normal HCHO
Substitution
mid-ratio
high HCHO
Substitution
low-ratio
normal HCHO
Substitution
high-ratio
normal HCHO
Chemistry
REACTANTS
First Side
3.00 ppmC SynUrban
3.00 ppmC SynUrban
3.00 ppmC SynUrban
1.000 ppmC SynUrban
6.00 ppmC SynAuto
l.OOppmC HCHO
REACTANTS
Second Side
2.00 ppmC SynUrban
2.00 ppmC SynUrban
0.89 ppmC McOH
0.01 ppmC MeN02
O.lOppmC HCHO
2.00 ppmC SynUrban
0.79 ppniG McOH
0.01 ppmC MeNO2
i_ 0.20j>pjnC HCHO
0.6GC ppmG SynUrban
0.300 ppmG McOH
0.003 ppmG MeNO2
0.030 ppmG HCHO
4.00 ppmC SynAuto
1.78 ppmC MeOH
0.02 ppmC MeN02
0.20 ppmC HCHO
l.OOppmC HCHO
1.00 ppmG MeOH
PURPOSE
To determine the effect of 33% reduction in HC in an urban-
like environment at a typical HC-lo-NOx ratio. Expect 30%
reduction in ozone maximum.
To determine the reactivity of the most likely methanol fuel
exhaust in an ur6an-likc environment at a typical HC-lo-NOx
ratio. Expect 20% reduction in ozone maximum.
To determine the reactivity of a highly reactive methanol fuel
exhaust in an ur&an-like environment at a typical HC-lo-NOx
ratio. Expect 10% reduction in ozone maximum.
To determine the reactivity of a typically reactive melhanol
fuel exhaust in an ur/ian-like environment at, a low HC-lo-NOx
ralio. Expect large reduction in ozone maximum.
To determine the reactivity of the most likely methanol fuel
exhaust in an aufo-ei/iousf-like environment at a high HC-to-
NOx ratio. Expect 20% reduction in ozone maximum.
To illustrate the chemistry of methanol in a highly reactive
environment.
-------
Priority of Runs Overall QA Steps
These are performed throughout the programs. A discussion of chamber background
reactivity was given in Jeffries and Sexton12.
Special Mixtures
The experiments often utilize synthetic mixtures. These mixtures require extensive
experimental support. For example, previous experiments may compare the reactiv-
ity of these mixtures with real auto exhaust or with other synthetic mixtures which
are further compared with simpler photochemical systems. This previous experi-
mental experience is an additional QA step which assures reference to well-studied
photochemical systems.
These mixtures are either ordered from scientific gas suppliers or are mixed
and prepared on site. In either case, the resulting mixtures are compared with
NBS-traceable calibration sources.
Special Techniques
Some project goals require the development of special techniques such as the pro-
duction of a particular dilution rate profile or the use of highly labile compounds
such as methyl nitrite. To use methyl nitrite requires that the synthesis, storage
and use (injection into chamber) techniques be developed and implemented.
Reviewing and Scheduling Methods
To maximize the quality of program results, reviewing and careful scheduling are
needed throughout the program. Important considerations are the priority of the
experiments, status of the smog chamber site facilities, weather, and previous ex-
perimental results. Inherent in this process is a continuous review of experimental
results to assure that target conditions are being attained and experimental condi-
tions were satisfactory.
Priority of Runs
The results of experiments are affected by the initial concentration conditions, even
for the same composition of ozone precursors. For most studies, several experiments
are desired. Higher priorities are assigned to those experiments which will yield the
most information for the more average conditions. Later experiments are used
to establish the range of conditions or to establish more basic results to assist in
interpretation.
-------
Overall QA Steps Use of NOAA Weather Radio
Use of NOAA Weather Radio
Outdoor Smog Chambers are subjected to real weather situations. This leads to
difficulties in scheduling runs. To provide up-to-the-minute information needed to
schedule runs, we use a weather radio to receive the airport NOAA weather radio
station. Weather reports are updated every hour and forecasts are made every 3
minutes. The most desirable conditions are clear, sunny, dry and warm.
When clear weather is forecast, the site operator is contacted by telephone. He
gives a status report of the site facilities and instrumentation. The next highest
priority experiment which can be conducted considering the current status of the
facilities is then scheduled.
When cloudy weather is forecast ("partly cloudy") with a low probability of
rain, a characterization experiment is scheduled unless a set of calibrations are
needed and the long range forecast is for good weather.
When rainy or "mostly cloudy" weather is forecast, a "CAL DAY" is scheduled.
Feedback Scheduling
Experimental results are reviewed frequently. Experiments are repeated or new
experiments are designed and scheduled depending on the conditions or outcome of
experiments already conducted.
Site Operation
A great deal of QA occurs in the routine operation of the smog chamber site.
Procedures more basic than calibrations are required to assure useful and complete
data. The laboratory itself must be maintained. Adequate supplies must be in
place, all documentation must be complete, instruments must be in good operating
order, support equipment must be checked-out, and the chamber must be cleaned
and leak-checked.
Lab Air Conditioner
The laboratory air conditioner maintains a consistent environment needed for the
operation of instruments and computers in the typical 95°F, high humidity, summer
conditions in North Carolina. Lab temperature is monitored continuously by the
computer data acquisition system and is reported in the final distributed exper-
imental results segmented data file. The lab temperature is maintained at 75°F.
The air conditioner is checked at the beginning of every run season by a professional
-------
Ethylene Convertor Overall QA Steps
service. The filter is changed every three weeks during most of the season and every
two weeks during the warmest part of the season.
Smog Chamber
The smog chamber itself requires routine maintenance. It is frequently checked
for holes and leaks, which are repaired. Occasionally insects find their way into
the chambers and must be removed. The floor is swept and sponged with distilled
water when needed. The automated venting doors and air drying ductwork is
checked daily. Leaks are often discovered by analysis (least-squares exponential fit:
see section on computer programs below) of the inert tracer data collected during
every run on both sides of the dual chamber.
Chamber Sampling Manifold
The sampling manifold is routinely checked for breaks. It is warmed with electric
heat tapes the entire length from the chamber to the fans inside the laboratory.
Minor leaks can be discovered by analysis of the inert tracer data collected during
every run on both sides of the dual chamber. The manifold is checked daily for
breaks, inoperating heating lines by touch, and by appearance of condensation
(warm chamber air into a cooler laboratory).
Chamber AC Drying System
An automated chamber air drying system in place under the chamber is used before
experiments to prevent condensation which might form during the cool morning
hours after venting the chambers. A Freon check and general maintenance checkout
is performed at the beginning of every run season. Testing is performed early in the
season to measure the dewpoint depression capability. Routine monitoring of daily
performance (for each experiment) is the most common QA procedure. Dewpoint
is continuously monitored. Condensation is checked by the morning operator.
Ethylene Convertor
The ozone monitor utilizes ethylene. The excess is routed through a heated con-
vertor which oxidizes it. Untreated excess ethylene can be a severe contamination
problem. To check for leaks in the ethylene treatment system, two GC-FID systems
(Carles 1 and 2) monitor for this possible contamination in background chamber
air analyses before every experiment.
-------
Overall QA Steps Mass Balance
Mass Balance
A high precision Mettler balance is used for weighing chemicals used for reagents
and paraformaldehyde used for injection of formaldehyde into the chamber.
Supplies
An inventory is performed at the beginning of every run season to determine supplies
which must be ordered long in advance of the run season. Special mixtures of HC
and other precursors must also be ordered. More common supplies are tracked and
stocked routinely. During the run season, the use of expendable supplies is carefully
monitored and reordered when necessary.
Check Lists
Many details must be checked to assure that a smog chamber experiment will be
successful. Several checklists are used before, during, and after experiments.
Setup Checkout Lists
To make sure that things are ready for an experiment the next day, a setup checkout
list is used. All supplies (gases, etc.), computer and instrument switch settings, and
injection gases are checked; syringes and liquid chemicals are set out; the run sheet
is rilled out; venting is checked; the manifold is checked for leaks, etc. This checklist
is shown in Appendix A.
Operator Checkout List
For the early morning operator responsible for initiating the experiment, another
checklist is used. This is somewhat redundant to the setup list to check extremely
important details that might have been overlooked by the setup operator. In addi-
tion, check items include details about proper operation of instruments, manifold,
and chamber, injections of chemical precursors, and recording documentation about
details of experiment. This list is also shown in Appendix A.
Run Checkout List
Another checkout list is used by site personnel that arrive at mid-morning. Check
items include checking on status of instruments, stripchart recorders, computer,
chamber, etc. to assure successful continuation of the experiment and data acquisi-
tion. This list is also given in Appendix A.
-------
Types of Characterization Experiments Overall QA Steps
Run Packup List
This list is primarily an inventory checklist to insure that all data is properly packed
up, documented, and returned to the School of Public Health data processing office.
This list is included in Appendix A.
Preventative Maintenance
To assure proper operation of all instruments, the smog chamber, support facilities,
instrument stability (reflected in "cal factor variation"), and maximum "up-time,"
preventative maintenance is routinely performed.
Characterization Experiments
As mentioned above, characterization experiments are performed routinely. These
experiments are used by modelers to determine the importance of chamber artifact
processes. Background, reactivity, surface processes (such as off-gasing of NOX,
HCHO), decay of 03 and NOX are of primary interest. Previous experiments conducted
over several years have been studied to characterize these and other aspects of
chamber behavior. Benchmark experiments are also routinely performed. These
include propylene and NOX experiments.
Types of Characterization Experiments
Background
This experiment shows how much O3 is formed when no chemical precursors are
injected into the chamber except for the rural air used in venting and any material
which may remain on the chamber walls after venting.
O3 Decay
O3 is injected into otherwise clean air and allowed to decay. The rate is determined.
This experiment is usually performed at night. Some modelers include this decay
process although the rate is very low.
JVO Oxidation, Daytime
NO is injected into clean chamber air (nothing else injected) to detect sources of
radicals. The rate of oxidation is observed. The magnitude of and types chamber
radical sources required to achieve the observed rate can be estimated by modeling.
To modify the test, CO can be added to one side to help determine the type of
-------
Overall QA Steps Frequency
radical source. The CO changes the balance of HO and HO2 radicals without adding
any radicals to the system itself.
NO Decay, Nighttime
This experiment is conducted like the one described above except that no sunlight
is involved.
Acetaldehyde and CO
Matched injections of acetaldehyde (usually 1 ppmC) are made into each chamber.
One side has CO injected (usually 50 ppm). This experiment forms PAN which can
aid in detecting any wall NOX off-gasses. The CO changes the balance of HO and
H02 radicals without adding any radicals to the system itself.
Formaldehyde and CO
Matched injections of formaldehyde are made into each chamber. One side has
CO injected (usually 50 ppm). Formaldehyde,photolyzes to produce HO, and CO.
Without injected NOX, only the background NOX in the rural air or from the chamber
walls can participate in this system. Any HO produced can react with formaldehyde
or with the additional CO added to one side. The CO changes the balance of HO
and HO2 radicals without adding any radicals to the system itself.
Propylene and NOx
Matched conditions of propylene and NOX have been used for years as a bench-
marking experiment. Experience has shown that experiments conducted on the
same day a year apart can be duplicated if the sunlight quality is good. Demon-
strating matched results from the two chambers is another reason for conducting
matched condition experiments.
Frequency
These experiments should be conducted routinely. However, they are not conducted
as frequently when the program falls behind schedule because of poor weather.
These experiments have been studied and the range of magnitude of these chamber
processes is well known.
10
-------
System Components
This section lists and describes the system components. Major types of tasks and
responsibilities have been organized and assigned to different people. Computers,
instruments and other hardware are used to generate and process data. Many com-
puter programs (software) are used. Policies and procedures have been developed
to assure proper and consistent treatment of the data. Databases have been created
to organize and to aid in retrieving and processing the large amounts of calibration
data.
Location of Work
The research work occurs in two different locations. The experiments are conducted
at the Smog Chamber Site located in a rural area with relatively clean background
air, and the data is processed on the University campus.
Experimental Site
The site is located in Chatham County, North Carolina, approximately 32 kilo-
meters from the University of North Carolina at Chapel Hill, and approximately
10 kilometers from the small town of Pittsboro. Chatham County is one of the
most rural, least industralized counties in North Carolina and is heavily wooded.
The background concentrations of NOX and NMHC are usually less than 5 ppbC and
less than 80 ppbC. More importantly, the air exhibits very low reactivity in the
chamber.
11
-------
System Components Data Processing Location
Data Processing Location
The data are processed in the School of Public Health on the campus of UNC, where
all computers and data processing facilities are easily accessible.
Personnel
A number of people with different tasks contribute to the process of conducting
smog chamber experiments and processing the results. The names and roles of
these people" are listed below.
Advisory Group, AG
An advisory group, consisting of the principal investigator, the co-principal inves-
tigator, the project coordinator, and the administrative research assistant meet
frequently to review the progress of each project.
Administrative Research Assistant, A A
The administrative research assistant is responsible for logging the experiments and
calibrations, ordering supplies, and managing the budget.
Project Coordinator, PC
The project coordinator oversees the experimental program and the processing of
the data.
Site Operator, SO
Site operators are responsible for maintenance, calibration, and operation of the
research site. They work closely with the PC to determine which experiments should
be conducted next or if time should be scheduled for calibration or maintenance.
Peak Picker, PP
The "peak pickers" process most of the data under the guidance of the PC.
Computer Technician, CT
The computer technicians are responsible for unpacking the data from the smog
chamber data acquisition system and for making tapes of segmented data files and
a backup of all data processing files.
12
-------
DEC VAX-11/780 System Components
Computers Used
Several computers are used in the program. They help conduct experiments, process
data, plot data, analyze results, and aid in writing.
DEC PDP-11/40
The DEC PDP-il/40 is used to control the experiments and perform the data acqui-
sition for much of the data. This computer writes data to 8 inch flexible diskettes
for transfer to the data processing office.
DEC LSI-11/23
Two DEC LSI-11/23 computers are used to process most of the experimental and
calibration data. Most programs are written in PASCAL. A major component is the
DIGITIZER which permits rapid cind accurate conversion of stripchart data to com-
puter data files. Each computer has at least one 8 inch floppy drive and a 10
Mb hard disk. These computers can communicate to the IBM PC and to the DEC
VAX-11/780.
IBM PC
The IBM PC is used with LOTUS 123 to implement many spreadsheets and databases
discussed throughout this report. It is also used for word processing. Some PASCAL
programs are also used. The IBM PC uses 51/4 inch floppies and can "upload" and
"download" data from the DEC VAX-n/780.
DEC VAX-11/780
A DEC VAX-11/780 mainframe computer is used to maintain several large databases
discussed in this report. The VAX system also makes plots and prints reports using a
large, high resolution laser printer. Modeling input decks are set up and submitted
from the VAX to be run on the university's IBM 4381 mainframe computer. "Printouts"
and "punched" output are returned to the VAX as standard ASCII files. The VAX can
also read and write to 8 inch floppies and industry standard magnetic tape.
13
-------
System Components Computer Programs Used
Computer Programs Used
Description of General Purpose Programs
LOTUS 123
LOTUS 123 is an electronic spreadsheet, database, and graphic program that runs on
the IBM PC. The spreadsheet in 123 has 256 columns and 2096 rows; each column
can be a different width and each cell can hold a text label, a numerical value, or a
formula that includes references to other cells. When new data is entered into cells,
the program automatically updates the values of all the formulas. Ranges of cells
can be selected to be graphed. Graphs include line, bar, and pie; up to 6 lines can
be plotted on a graph. In addition, the spreadsheet contents can be treated like
a database table in which the rows are records and the columns are fields in the
record. Records can be selected, deleted, copied, and sorted to an output location
on the spreadsheet. The whole spreadsheet or selected parts of the spreadsheet can
be printed to a printer or file. Data can also be 'imported' from other programs or
files.
DATATRIEVE
DATATRIEVE is a data base query, update, and report writing system running on
the VAX-ll. DATATRIEVE is used to maintain calibration and data processing sta-
tus databases and serves a tool to coordinate effort among the PC, PP, and CT. In
DATATRIEVE, the PC can ask the system to print a list of all runs that had all in-
strument data "picked" but did not have calibration factors determined for those
instruments. From this list, the PC determines which calibration factors must be
produced to complete the processing of the runs. Typical run tracking data bases
grow to be 500,000 characters during a run processing season. The total size of
data managed by DATATRIEVE is more than 20 million characters.
Description of Specific Programs
UNPACK converts packed binary DVM data from the chamber site's DA system to
ASCII format ("U files").
FASTDV produces temporary ("P" file) files of raw DVM data for making plots
for preliminary QA procedures, and strips NOX and O3 calibration data from DVM
data file.
DIGPIK allows the digitizer to be used to convert experimental peak height data
for all species for a given instrument and writes data to ("P" file) files on LSlii/23.
14
-------
"U" file System Components
It also is used to digitize calibration data and produce calibration response ("R")
files.
PLOP1C plots raw ("P" file) experimental data from both DVM or stripcharts.
ACALAV automatically produces average NOX and Os instrument calibration
data for zeros and spans stripped from the DVM data with FASTDV.
CALFAC calculates calibration factors from digitized calibration data and the
official calibration source file.
CALLOK lists and plots calibration data for a given instrument and species
combination.
CALANA performs statistical analysis on calibration data for any combination
of instrument and species.
DVMFIX applies calibration factors to DVM ("U" file) data and outputs concen-
tration and physical measurement ("G") files.
CALCON applies calibration factors to digitized ("P" file) data from stripcharts
and outputs concentration ("C" files).
PLOCON plots concentration ("C" file) experimental data from stripcharts.
HCANAL reads time/concentration data files of HC data and computes compo-
sition analysis.
CSTAR reads tracer data files and computes chamber dilution rate by fitting
with exponential least-squares method.
Description of Data File Formats and Contents
«U" file
The primary data file originating from the Data Acquisition system is "IT'npacked
after transport from the chamber site to the data processing office. This file con-
tains "raw" data from the NOX and Os monitors, chamber, ambient and laboratory
temperature, total solar and ultraviolet light, and dewpoint monitors. The data
is taken every minute alternating between the two sides. Calibration data from
automatic calibrations of the NOX and Os monitors is also contained in this file.
15
-------
System Components "G" file
"G" file
The final data file processed from the primary "IT file is referred to as the "G"as
Chamber file. This file contains corrected (adjusted with calibration factors) data.
The final data is for four minute intervals alternating between the two sides.
The file is comprised of two parts: documentation and data. The documentation
identifies the experiment. Next information pertaining to the processing of the data
is given:
source of calibration factors;
data last processed;
explanation of data identifiers (labels);
explanation of units;
explanation of data alteration.
Finally the actual calibration factors are listed. Calibration factors are specified
for the beginning and ending of periods of valid data.
The data are preceded by a "header" card identifying the columns of data. Data
are listed in the same order as the header card. An example of this type of file is
shown in Chapter 6.
"P" files
The data files of "P"icked data from other individual instruments which are not
connected to the Data Acquisition System are produced and processed separately
with more manual effort (e.(/.measuring peak heights from gas chromatographs).
This file is comprised of two parts: documentation and data. The documentation
indicates:
the source of the data;
the data processor ("peakpicker");
attenuation;
units;
species identification.
The data follows with one data point per line: the format allows for easy editing
and correction during processing.
16
-------
ACDTR.COM file System Components
"C" files
These are the files of final corrected "C"oncentration data processed from the "P"
files. The format of these files are identical to the data portion of the UG" files. An
example of this type of file is shown in Chapter 6.
"K" files
For every "C" file there is a matching "K" file containing documentation pertaining
to the source and processing of the data. In addition to the documentation originally
in the "P" file is the maximum amplitude and concentration of each species in the
file and the calibration factor (and source) used in processing the data. An example
of this type of file is shown in Chapter 6.
«R" files
Calibration "R"esponse data to official calibration sources other than from the
NOX and 03 monitors are stored in these files, which are produced from the same
program which produces the "P" files (DIGPIK). They have very similar format and
documentation but also include the DCS identification number.
"Q" files
The resulting calibration factors from comparing the "R" files to the Official Cali-
bration Sources are stored in the "Q" files along with all associated documentation
from the "R" file. This file also contains DATATRIEVE commands which will store the
data in the calibration DATATRIEVE database HCCAL.
"A" files
Calibration response data from the "A"utomatic calibrations of the NOX and 0$
monitors are produced with FASTDV from the data stored in the "U" file.
"E" files
The "A" files often contain data which represent a transition reading between the
chamber air and the calibration source. This unwanted data is "E"dited from the
file.
ACDTR.COM file
The "E" files are processed (averaged) to a single value with a program called
ACALAV which is output to a file Ceilled ACDTR.COM which also contains DATATRIEVE
commands to store the data in the DATATRIEVE database AUTOCAL.
17
-------
System Components Description of Databases
Description of Databases
Several databases are used in processing the data and calibration data. The fol-
lowing are created with DATATRIEVE on the VAX-11/780. These are described in more
detail in Chapter 3.
RUNENTRY is the primary database identifying the experiment and tracking
the processing of the data from each instrument used.
CAL.ENTRY is the database for tracking the calibration data processing.
CAL.SOURCES is the database of the "O"fficial "Calibration "S"ources ocs)
including all supporting information and corresponding identification number.
HC.CAL is the database of the final processed calibration factors resulting from
calibrations performed on the analytical instruments producing data which is not
collected from the Data Acquisition System.
AUTO.CAL is the database of the calibration data from "auto"matically per-
formed calibrations of the NOX and Os monitors.
RUN.CAL.USED is the database of the calibration factors actually used in
processing the manually processed ("P" file) data.
SPECIES is the database of official species names and corresponding identica-
tion numbers used in data processing to assure consistency and proper spelling and
identification.
Description of Database Procedures
List of all DTR Procedures by Name and Function
Various procedures are used with the DATATRIEVE databases to assist in accessing and
using the data and producing reports or files to be used by other data processing
computer programs.
PPREPORTS is a VAX command procedure which creates a task report for Peak
Pickers (data processors running DIGPIK using DATATRIEVE and the RUNENTRY data base.
The report lists each data processing step presently needing PP action, such as
instrument stripcharts needing to be digitized and plotted.
18
-------
Worksheets/Databases System Components
PCREPORTS is a VAX command procedure which creates a task report for
Project Coordinators using DATATRIEVE and the RUliEHTRY data base. The report lists
each data processing step presently needing PC action, such as various QA steps.
CTREPORTS is a VAX command procedure which creates a task report for
Computer Technicians using DATATRIEVE and the RUtiEl.'TRY data base. The report lists
each data processing step presently needing CT action, such as moving data to the
VAX or vice versa.
PRINT-SPECIES prints a list of the official species names and corresponding
identification number.
PRINT.CAL-SOURCES prints a list of the Official Calibration Sources includ-
ing name, serial number, source date, validation date and personnel, manufacturers
and the validated concentrations.
PRINT.OCS-FILE-FOR-LSI produces a file of the official calibration sources to
be used by another program (CALFAC) used to calculate calibration factors from
the calibration response data.
PRINT.CALS-AND-FACTORS calculates and produces a report of the cali-
bration factors calculated from the automatic calibration data from the NOX and O3
monitors from the AUTO.CAL database.
PRINT-HC-CAL.FACTORS produces a report of the calibration data for all
the other instruments other than NOX and O3monitors from the HC.CAL database.
PRINT-RUN-CAL.USED- produces a report of the calibration factors used by
instrument, species, and date from the RUll.CAL.USED database.
PRINT.CALENTRY.SUM.ON.FILE produces a summary report of all cali-
bration data processing steps completed by date, by instrument from the CAL.EIITRY
database.
Worksheets/Databases
Several worksheets implemented with the LOTUS 123 software on the IBM PC are
used. These are backed-up on 5.25 inch floppy diskettes. Each of these will be
discussed in detail in Chapter 3.
OCSICyy - Injections into the smog chamber are used as calibration sources.
This worksheet/database calculates the chamber concentration from the amount
19
-------
System Components Worksheets/Databases
injected, the chamber temperature, the ideal gas law, and compound physical data
from a lookup table in the worksheet. A different worksheet is maintained for each
year (yy).
CAL - All hydrocarbon calibration sources are compared and "validated" in
this worksheet/database. It is listed in Appendix B.
MANmmddyy - "Man"ually performed gas phase titration calibrations for the
NOX and ©3 monitors are processed in this worksheet. One worksheet is produced
for each calibration identified by date (month, day, year).
MANCALyy - The calibration factors resulting from the MANmmddyy work-
sheets is analyzed in this worksheet.
20
-------
Calibration
Approach
All instruments are calibrated at a minimum of two reference levels: a "zero" and a
"span" point. The latter is generally near the expected maximum concentration in
the experiments. "Zero" air samples are obtained by sampling from a zero air gen-
erator or by sampling from a filter specific for the species being monitored. "Span"
values are generated from several sources that will be described below. During the
programs, different "span" levels are used. These different levels assure that linear-
ity can be checked in final QA procedures, to be discussed in later sections. Several
different calibration standard sources, including NBS traceable sources and sources
from outside groups, are utilized and compared. Calibrations for most instruments
are performed immediately before an experiment. NOX and 03 monitors are cali-
brated twice daily. These calibration data are used to adjust the data collected in
final data processing steps after the calibration data itself has been processed and
reviewed in the context of the entire calibration database for the season.
NBS-Traceable Sources
NO and O3 Sources
Two techniques are used to calibrate the chemiluminescent NOX and 03 monitors.
The primary technique is the EPA gas-phase titration method3 which utilizes a high
concentration NO source (NBS-traceable) as the primary standard. Comparison of
this primary source with other sources is performed regularly. In one recent study,
three of these high concentration sources were compared and found to agree within
a few percent of each other. Low concentration sources of NO and O3 are also used
21
-------
Calibration
Hydrocarbon Sources
to perform span checks more frequently and automatically (AUTOCAL source). Low
concentration (<1 ppm) pressurized gas cylinders are used for NO sources, and a
generator built into the Os monitor is used for the O3 AUTOCAL source. These are
calibrated and used as transfer standards. Table 2 lists the calibration sources used
for NOX.
Table 2. NO Tank Calibration Sources
Name
Hanuf.
Manf. ID UtIC ID
DCS ID
NO 46.0 ppm Matheson SX-13759 UNCAB2113 94
MO 52.4 ppm Airco 1122590 UIJCBC9045 169
HO 52.6 ppm Scott BAL177 UNCAB0348 170
HO 0.823 ppm CC-15860 4
HO 0.288 ppm 171
MO 0.665 ppm Airco
CC-15780 U11CAB2478 172
The OCS ID heading will be explained below.
Hydrocarbon Sources
Commercially prepared NBS traceable calibration sources are also used for hydro-
carbons. Five different tanks purchased by UNC were used. In addition, two other
tanks were borrowed from the Research Triangle Institute (RTI) who is involved
in a multiyear intercomparison of commercial hydrocarbon calibration sources. A
low concentration tank prepared from diluting a small portion of one of the com-
mercially prepared high concentration tanks was calibrated as a transfer standard
for automatic calibration. These calibration sources are listed in Table 3. Another
source of hydrocarbon standards was prepared in the chambers by injecting liquid
hydrocarbons. These are listed in Table 4. The identities and properties of the
species calibrated this way are given in Table 5. All sources were compared and
results are described below.
22
-------
NO and O3 Sources Calibration
Table 3. HC Tank and Liquid Calibration Sources
IIEV.' AIRCO Airco
LOV: HW (HC) Scott
LOV.' MW (LC) U11C
HIGH MW Scott
RTI- ethylene
RTI- propylene
82HCCALMIX
NEV/SCOTTCALTANK
I.IQ STD 8/14
X-3362
AAL11557
1IA
AAL11551
2H
3A
XA-1336
NA
U1ICAB2467
U11CAB2136
HA
UNCAB2139
112
109
111
108
167
168
1
110
123
Transfer Standards Created from Primary Sources
For NOX, O;j, and hydrocarbons, several sources traceable to NBS are used. These
are high concentration tanks for stability and accuracy which require dilution or
instrument attenuation adjustment when used. Accurate liquid injections into the
smog chamber are also considered as primary standards. Several low concentra-
tion NO tanks were purchased as a source for automatic calibrations ("AUTOCAL"
sources). A low concentration HC tank was prepared by diluting a portion of the
LOW MOLECULAR WEIGHT (high concentration) calibration tank into an empty
tank with nitrogen to be used as an AUTOCAL source. AUTOCAL sources do not require a
morning operator to be present to change the attenuation settings and are not sub-
ject to possible errors from dilution. These calibration sources were intercompared.
The RTI tanks were considered absolute because of their extensive comparative
analysis history. In the case of the hydrocarbon sources, some judgements resulted
in small adjustments to the "validated" concentrations to make all the UNC sources
more consistent. The intercomparison procedures result in calibration of the AUTOCAL
sources.
Primary Source Comparison
NO and O3 Sources
This test is conducted to compare the manufacturers stated analyzed gas concen-
trations of the NO tanks. Nitrogen oxide tanks are available for use in the EPA Gas
Phase Titration method (see description in "Routine NOX and Os Calibrations").
Three tanks with NO concentrations of 52.6, 52.4, and 46 ppm (see 2) were diluted
with clean air (zero air generator) to concentrations of approximately 0.2, 0.6 and
23
-------
Table 4.
OFFICIAL CALIBRATION SOURCES FROM INJECTIONS INTO SHOS CHAHBER
LAST UPDATE: 13-JUN-B5
CHAHBER VOLUME 150000 (L) STP VOL: 22.414 (L)
DCS OCS SOURCE
ID t NANE DATE
95
94
97
98
99
100
101
102
103
104
105
106
107
115
116
117
118
118
118
118
118
118
119
120
123
123
123
123
123
123
123
124
125
126
127
128
129
130
131
i/25
4/23
6/27
6/26
6/26
6/19
6/19
7/10
7/10
7/11
7/11
7/27
7/27
7/31
6/28
6/28
7/16
7/16
7/16
7/16
7/16
7/16
7/27
7/27
3/14
8/14
8/14
8/14
8/14
8/14
8/14
6/27
7/13
7/13
SPEC
ID t
68
55
55
68
71
55
68
139
139
139
139
139
139
139
55
68
40
55
68
71
67
89
82
40
19
28
42
82
55
68
71
68
139
139
205
205
205
205
205
C
t
8
7
7
8
8
7
8
1
1
I
1
1
1
I
7
8
6
7
8
a
8
a
8
6
5
6
7
a
7
8
a
a
i
i
i
i
i
i
i
DENSITY UK
0.864
0.867
0.867
0.864
0.897
0.867
0.864
0.815
0.815
0.815
0.815
0.815
0.815
0.815
0.867
0.864
0.878
0.367
0.864
0.897
0.861
0.906
0.692
0.878
0.626
0.653
0.695
0.692
0.867
0.864
0.897
0.364
0.315
0.815
0.791
0.791
0.791
0.791
0.791
106.1
92.14
92.14
106.1
106.1
92.14
106.1
30.05
30.05
30.05
30.05
30.05
30.05
30.05
92.14
106.1
78.11
92.14
106.1
106.1
106.1
104.1
114.2
78.11
72.15
86.18
100.2
114.2
92.14
106.1
106.1
106.1
30.05
30.05
32.04
32.04
32.04
32.04
32.04
UL
or 6
283.000
556.000
463.000
283.000
243.000
370.500
282.900
.0.110
0.210
0.192
0.192
0.192
0.038
0.192
556.000
283.000
416.000
420.000
372.000
373.000
274.000
276.000
238.000
188.600
0.212
0.212
224.000
224.000
80.100
74.700
196.700
T
F
70.0
70
70
71
71
70
70
70
.0
.0
.8
.8
.0
.0
.0
70.0
70.0
70.0
70.0
70.0
70.0
77.3
77.3
85.
85.
85.
85.
85.
85.
85.
70.
68.
68.
70.
74.
74.
75.
0
0
0
0
0
0
0
0
0
0
4
0
0
5
73.0
PURITY
100Z
1001
1001
100!
1001
100Z
100Z
921
92Z
92Z
92Z
92Z
92Z
92Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
92Z
92Z
100Z
100Z
100Z
100Z
100Z
CONC
UL
2.9679
5.8955
4.9093
2.9780
2.6530
3.9285
2.9668
NA
NA
NA
NA
NA
NA
NA
5.9767
3.0083
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
o.oooo
2.9875
3.1609
2.9897
2.9925
2.9876
2.9765
2.6630
1.9779
NA
NA
0.8914
0.8974
0.3209
0.3001
0.7866
CONC
S
NA
NA
NA
NA
NA
NA
NA
0.5421
1.0350
0.9463
0.9463
0.9463
0.1893
0.9463
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
1.0384
1.0384
NA
NA
NA
NA
NA
24
-------
Table 4, cont.
OCS DCS SOURCE
ID 1 NAME DATE
132
133
134 '
135
134
137
138
140
143
144
147
148
159
162
162
162
162
163
163
163
163
164
165
16S
165
165
165
165
166
166
166
166
166
10/7
10/7
10/9
10/9
10/16
10/16
8/25
9/1
9/3
9/17
9/25
9/5
5/14/85
5/14/85
5/14/85
5/14/85
5/17/85
5/17/85
5/17/85
5/17/85
5/20/85
5/22/85
5/22/85
. 5/22/85
5/22/85
5/22/85
5/22/85
5/24/85
5/24/85
5/24/85
5/24/85
5/24/85
SPEC
ID 1
205
139
139
139
139
139
139
205
205
205
205
139
139
55
40
68
82
55
40
68
82
139
82
40
55
68
71
78
82
40
55
68
71
C
i
t
1
I
I
'1
1
1
1
I
1
I
1
1
7
6
8
8
7
6
8
8
1
a
6
7
8
3
9
8
6
7
8
8
DENSITY IW
0.791
0.815
0.815
0.815
0.815
0.815
0.815
0.7914
0.7914
0.7914
0.7914
0.315
0.815
0.867
0.378
0.864
0.692
0.367
0.373
0.864
0.692
0.815
0.692
0.378
0.367
0.864
0.897
0.876
0.692
0.878
0.367
0.364
0.897
UL
or 6
32.04 65.500
30.05 0.192
30.05
30.05
30.05
30.05
30.05
32.04
32.04
32.04
32.04
30.05
30.05
92.14
78.11
106.1
114.2
92.14
78.11
106.1
114.2
30.05
114.2
78.11
92.14
106.1
106.1
120.2
114.2
78.11
92.14
106.1
106.1
0.097
0.192
0.192
0.048
0.096
74.000
249.000
66.900
151.000
0.192
0.192
92.000
45.000
185.000
240.000
100.000
50.000
200.000
300.000
0.201
100.000
150.000
200.000
400.000
500.000
500.000
100.000
150.000
200.000
400.000
500.000
T F
74.5
54.0
54.0
55.0
55.0
61.0
61.0
66.2
59.0
66.7
45.0.
62.0.
73.0
36.0
86.0
86.0
36.0
72.0
72.0
72.0
72.0
82.0
30.0
80.0
80.0
80.0
80.0
80.0
65.0
65.0
65.0
65.0
65.0
PURITY
100!
921
921
921
921
921
921
100Z
100Z
1001
1001
921
921
100Z
100Z
1001
100Z
100Z
100Z
100Z
100Z
95Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
100Z
CONC
UL
0.2627
NA
NA
NA
NA
NA
NA
0.2921
0.9695.
0.2644
0.5721
NA
NA
1.0050
0.5033
1.9988
1.9290
1.0643
0.5449
2.1054
2.3494
NA
0.7949
1.6593
2.1607
4.2741
5.5431
5.3791
0.7728
1.6132
2.1006
4.1553
5.3889
CQNC
6
NA
0.9177
0.4636
0.9195
0.9195
0.2325
0.4651
NA
NA
NA
NA
0.9320
0.9606
NA
NA
NA
NA
NA
NA
NA
NA
1.0461
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
25
-------
Calibration NO and O3 Sources
Table 5. Species ID numbers and Properties
ID NAHE I DENS NU PURITY
19 N-PENTANE 5 0.624 72.15 1001
28
40
42
55
60
67
68
71
82
89
139
2-HETHYIP
BENZENE
2,3-DI«ET
TOLUENE
N-OXTANE
P-XYLENE
H-XYLENE
Q-XYLENE
2,2,4-TRI
STYRENE
HCHO
6
6
7
7
8
8
8
8
8
8
1
0.
0.
0.
0.
653
878
695
867
0.703
0.
0.
0.
0.
0.
0.
861
864
897
692
906
815
86.18
78.11
100.2
92.14
114.2
106.1
106.1
106.1
114.2
104.1
30.05
IOOZ
IOOZ
IOOZ
IOOZ
IOOZ
IOOZ
IOOZ
IOOZ
IOOZ
IOOZ
92Z
205 HEOH 1 0.7914 32.04 IOOZ
26
-------
Manual Gas Phase Titration Calibration
0.8 ppm. The exact calculated concentrations were determined by flow measure-
ments with soap-bubble meters for the air and NO. The measured concentrations
were plotted against the calculated concentration. All data were fit to a straight
line with the least-squares method. The data is plotted in Figure 1 and is shown in
Table 6. Agreement between the sources is excellent. The values assigned to these
tanks determine the absolute accuracy of our reported NOX and O3 data.
Hydrocarbon Sources
Because there is much less of a distinction between primary and working sources for
hydrocarbons than for IMOX and 03 sources, the comparison of primary hydrocarbon
sources will be discussed below in the next section on transfer standards.
Transfer Calibration to working "AUTOCAL" Sources
NOx .and O3 Calibration Sources
Experience and analysis of past calibration data indicates that calibration factors
estimated from the results of the "manually" performed gas-phase titration method
vary more than the calibration factors obtained from the automatic span checks
using low concentration sources. But when the manual values are averaged, they
yield the same results. The low concentration sources are convenient, but require
careful cheiriicterization, since their concentration decreases slightly during the run
season. Our experience and analysis also shows that the instruments are more stable
if the zero and span settings are not adjusted. Therefore, the following strategy is
followed to maximize the quality of the data from these instruments.
The NOX and O$ instruments are not adjusted after the initial start-up at the
beginning of the experimental season. Manual gas-phase titration calibrations are
performed to estimate calibration factors for the instruments throughout the exper-
imental season. The low concentration sources are used twice a day. The responses
are plotted for the entire run season to establish a decay rate for the low concentra-
tion sources. The manual calibration factors establish the concentrations of these
low concentration sources at the beginning of the season. The calibrated responses
from the low concentration sources are then used to calibrate the instruments. The
flowchart illustrating the procedures associated with the transfer of NOX and O3
calibrations is shown in Figure 2.
Manual Gas Phase Titration
The manual Gas Phase Titration calibrations are performed occasionally through-
out the season as discussed above. The technique is the EPA reference method as
27
-------
Calibration
Manual Gas Phase Titration
COMPARISON OF PRIMARY NO CAL SOURCES
I
8
1
. 0
Q 1984
0.4
0.6
0.8
CALCULATED CONCENTRATION (TANK LABEL)
+ 1983 o Aeroaol Linear Fit
Figure 1. Comparison of Primary NO Calibration Sources
28
-------
Manual Gas Phase Titration Calibration
Table 6. Comparison of Primary NO Calibration Sources
CALC. ESTIMATE MEASURED DELTA PERCEi.'T
TANK cone. cone. couc. cone.
1
1984
52.4 ppm
2
1983
46 ppm
3
Aerosol Project
52.6 ppm
ave 2.3%
0.197
0.392
0.803
0.202
0.422
0.886
0.192
0.377
0.786
0.213
0.418
0.849
0.218
0.449
0.936
0.208
0.402
0.831
0.218
0.423
0.855
0.229
0.419
0.930
0.215
0.403
0.840
0.005
0.005
0.006
0.011
-0.030
-0 . 006
0.007
0.001
0.009
2.3%
1.2*
0.7%
4.8%
6.7%
0.7%
3.4%
0.2%
1.0%
29
-------
Calibration Manual Gas Phase Titration
described in the CFR3.
This calibration technique is based upon the rapid gas phase reaction between
NO and O3 to produce stoichiometric quantities of NO2. The quantitative nature of
this reaction is such that when the NO concentration is known (primary calibration
standard), the concentration of N02 and O3 can be determined. O3 is added to excess
NO in a dynamic calibration system, and the NO/NOX/NO2 analyzer is used as an
indicator of changes in NO concentration. Upon the addition of O3, the decrease
in NO concentration observed on the calibrated NO channel is equivalent to the
concentration of N02 produced and 03 consumed.
Calibration Factors from Primary Sources
Each "manual" calibration is processed with the aid of an electronic form imple-
mented with LOTUS 123 on an IBM PC called MAHCALSH.MS in Figure 2 (see figure Fig-
ure 3. Information is transfered from the paper form rilled in during the gas phase
titration calibration procedure. The results of each manual calibration are inspected
for notes of problems that were encountered, such as odd or unstable readings. The
electronic form assures that the calculations are performed correctly. The com-
pleted form is printed out and the file is saved on diskette and backed up to another
diskette, thus resulting in at least four sources of the calibration data.
Another LOTUS 123 electronic form and database is used to collect the final cal-
ibration factors for all calibrations (called MANCAL.WKS in Figure 2). Output
from this worksheet is shown in Figure 4. The data are fitted with a straight line
using a least-squares method. The data and the straight line fit are plotted as a
function of date; these are shown in Figure 5. The plots are used to visually inspect
the data for trends and large variations. If necessary, the individual calculation
electronic forms can be inspected for errors in transfering data. The original paper
form can also be examined. If there are problems, the data in the database may be
invalidated.
If no trend is observed, the calibration factors are simply averaged for the season
to estimate the true calibration factor for the instrument. Analysis of manual
calibrations performed over many years by many people show variation by human
error but usually no trend. If a trend were observed, the calibration factor for a
particular time of interest would be estimated by a least-squares estimate through
the calibration data points.
Calibration of AUTOCAL Sources - Transfer Standards
The response and zero data from the AUTOCAL procedures (NO, NOX, and 03) are
30
-------
Hydrocarbon Sources Calibration
plotted and listed using a DATATRIEVE database. The first QA step is to investigate
the large variations or odd values. Sometimes a calibration was not conducted and
regular experimental data was processed as a calibration. A regression is calculated
to estimate the decay rate and to estimate the intercept (correct response) at the
beginning of the season or start of a new span check tank. The average (or if
a trend was observed, the appropriate time specific) manual calibration factor is
used to calculate the true concentrations of the AUTOCAL calibration sources from the
responses at these times (see bottom of Figure 4and discussion of routine NOX and
03 calibrations for final report of AUTOCAL calibration factors).
Hydrocarbon Sources
A procedure similar to that performed for the different NO and 0$ calibration sources
was performed for the hydrocarbon calibration sources. The comparison was far
more complex because of the large number of species and different sources utilized.
Five commercially prepared tanks purchased by UNC were compared and analyzed.
Two additional well-characterized (several years history) tanks which are part of
an on-going study of commercial HC calibration sources were borrowed from RTI.
A low concentration tank was prepared from one of the purchased tanks. These
were also compared. Liquid injections into the smog chamber used as calibration
sources were also compared with these tank sources. The low concentration tank
was calibrated as a working AUTOCAL source. All of the tanks, however, were used
occassionally for calibration. A list of the sources that were compared is given in
Table 7.
The borrowed RTI tanks were considered to be an absolute standard. The
values of the other calibration sources were "analyzed" based on the RTI tanks as
the primary standard. In most cases the final analyzed values for all compounds for
all sources were within a few percent of the manufacturers reported concentrations
and the liquid injected sources from the UNC chamber. Only one compound (2-
methyl-l,3-butadiene ) was outside of the manufacturers stated uncertainty. This
compound is not important to the present research program.
The intercomparisons of the HC sources was performed with a LOTUS 123 spread-
sheet on an IBM PC. The spreadsheet is listed in Appendix B.
Table 8 shows the agreement with the manufacturers analysis after "validating"
the tanks with the RTI NBS-traceable tanks (see Table 7). For each tank, the aver-
age and standard deviation of the ratio of the "validated" to manufacturers stated
concentration for each species in the tank are listed. The range of agreement is 0.94
to 0.99 with a mean of 0.963 and a relative standard deviation of 1.8 percent; the
31
-------
Calibration
Hydrocarbon Sources
LOTUS PRINT
IBM* PC
>Ct;
AUTQCAL
PROCESSING
CALANA
RESPONSE
DATA
INrriAL_AUTOCAL_VALUE
RATE_OF_CHANGE
(AUTOCALyy)
Figure 2. NOX and O3 Cal Processing Flowchart
32
-------
Hydrocarbon Sources
Calibration
NO-03 Calibration Data Sheet
DATE 8/6/SS
NAME Suedbeck
Air Flow:
volume 0.483 time 0.221 0.223
0.222
0.223 0.221
air flaw
avs 0.22200
2.18468
NO Cal Tank Flowttank and 03 turned on
volume 0.01 time 0.361 0.361
_30_ min be-fore flow rate measurements)
0.359 0.360 0.361 av: 0.3604O
NO flow
Predicted NO: tnk cone 52.453
NO Flow
0.02775
0.02775
0.65783
2.21243
NO + Air Flow
ZERO READING !ZERO POT SETTING
!
be-fore/ a-fter! !be-fore/ after!
NO (1) -0.015 ! 214
0.01254
SPAN READING !SPAN POT SETTING
N02 (!) -0.009
NOX (1) -0.013
03 (2) 0.001
33
1S1
0
! enter NA if no adjustment made
(1) zero air with H20 (pura-fil)
03 SLIDE OUT AND TITRATION BEGINS!
NO during titration 0.2133
(2)
(2)
(2)
be-fora/ a-fter!
0.5S6
-0.002
0.642
! be-fora/
! 604
!
i
i
572
714
230
after!
to pot settings
(2) with NO added
delta 0.3724
N02 during titration
0.4112
delta 0.4130
Ozone during titration 0.0665 (measure at outlet)
Ozone with NO TURNED OFF 0.4153
(wait for stable value)
(measure at inlet)
0.4162
(measure at outlet)
Air Flow
2.1S46S
0.41009
2.21243
NO « Air Flow
0.98746
If difference between inlet and outlet is more than a few ppb, wait longer.
Ozone (delta corrected for NO dilution): delta 0.34363
03 rotoball setting 3 Ethylene pres. 15 02 pres. 20
Figure 3. LOTUS Spreadsheet for Manual Cal Processing
33
-------
Calibration
Hydrocarbon Sources
SPAN CALIBRATION FACTORS
NO 1.0944 - predicted NO / (NO span - NO zero)
N02 0.9869
NOX
03
1.0079
(delta NO x NO span cal factor) / delta NQ2
(predicted NO + ((N02 span - N02 zero) * N02 cal factor)) /
(NOx span - NOx zero)
1.1861 - (delta NO x NO span cal factor) / delta 03
IF instrument adjustments ara to be made - these are desired span values
ADJUSTMENT ORDER: 1) ZEROS, 2) NO SPAN, 3) NOX SPAN, 4) N02 SPAN, 3) 03 SPAN
NO span value
NQx span value
N02 span value
03 span value
predicted NO + NO zero
- (predicted NO + ((N02 span - N02 zero) * N02 correction))
NOx zero
» (delta NO x NO span correction) + N02 zero
(should be 1.0 now) (should be 0.0)
= (((03 value during titration x 03 span correction) +
(delta NO :< NO span correction)) x DCF) + 03 zero
DCF
(should be 1.0 now)
NO -i- Air Flow
2.21243 !
2.1S468 !
Air Flow !
(should be 0.0)
1.01270
03 rotoball setting
Time that 03 instrument adjustments made or NA
Time that NOx instrument adjustments made or NA
»*»»»»»****»»»»***»»*»**#*****»**»**»»**»»*»»»***<
AUTO ZERO AUTO SPAN DATA
TANK AB2473
SPAN CORRECTED
»»»»»»»**»»»»»
AUTO SPAN & CQRR
TANK
SPAN CORRECTED
NO (1) -0.0202
NQ2 (1) -0.0068
NOX (1) -0.0217
03 (1) 0.00088
(2) 0.60751 0.68706
(2) 0.00023 0.00693
' 0.69400
(2) 0.66927 0.69631
(3) 0.10S82 0.12446
(2)
(2)
(2)
(3)
(1) regular auto zero (2) be sura to check for excess flaw rate
(3) regular auto span gonerator
Figure 3, cont. LOTUS Spreadsheet for Manual Cal Processing.
34
-------
Hydrocarbon Sources
Calibration
MANUAL CALIBRATIONS
2l-Hay-35 - last update
52.4523 ppi NO Tank
Cal
Date
02-Jul-84
10-Jul-84
li-Jul-84
lS-Jul-84
lB-Jul-84
18-Jul-84
29-Aug-84
day zero
02-Jul-84
0
8
14
16
16
16
58
real
0.7482
0.7693
0.78355
0.7835
NO
delta
response
0.7710
0.8360
0.8540
0.8430
NOx
real/
response
0.9704
0.9208
0.9175
0.9515
0.9615
0.9443
0.9294
real
delta
real/
real
response response
0.7521
0.7707
0.7836
0.7835
0
0
0
.8170
.8810
.9000
0.8840
0.9205
0.8747
0.8706
0.8863
03
delta
response
0.2640 0.2344
0.2385 0.2403
0.2340 0.2520
0.2172
0.2387
real/
response
1.1260
0.9926
0.9284
0.9100
LEAST SQUARES FOR SPAN NO
N
1
1
1
1
1
1
1
7
I
0
a
14
16
16
16
58
Y
0.97042801
0.92080143
0.91750585
0.9515
0.9615
0.9443
0.92941874
m
0
8
14
16
16
16
58
128
AVE CAL:
XBAR:
N*Y
0.97042801
0.92080143
0.91750585
0.9515
0.9615
0.9443
0.92941874
6.59545404
0.94220772
18.2857142
YDEV2
0.00079638
0.00045822
0.00061018
0.00008634
0.00037219
0.00000437
0.00016355
0.00249127
est y
0.9491
0.9461
0.9438
0.9431
0.9431
0.9431
0.9273
SO:
12
o-o
64 0
196 0
256 0
256 0
256 0
3364 0
4392 6
0,
Y2
.94173053
.84787528
.84181699
.90535225
.92448225
.89170249
.86381919
.21677899
.01886519
KY
0
7.36641148
12.3450819
15.224
15.384
15.1088
53.9062870
119.334580
SLOPE:
INTER:
R2:
rel std dev
1.13
-1.34
-1.39
0.45
0.98
0.07
0.11
-3.744E-04
0.94905347
0.11541280
LEAST SQUARES FOR SPAN NOx
N
1
1
1
1
4
X
0
a
14
58
*
Y
0.92050183
0.87474460
0.87061111
0.88631221
NtX
0
8
14
58
80
AVE CAL:
IBAR:
N*Y
0.92050183
0.87474460
0.87061111
0.88631221
3.55216977
0.88804244
20
YDEV2
0.00047114
0.00455127
0.00512607
0.00312430
0.01327279
est
0.
0.
0.
0.
Y
8925
8907
8894
8796
SD:
12
0 0.
64 0.
196 0.
3364 0.
3624 3.
0.
Y2
84732363
76517812
75796370
78554934
15601481
05760381
XY
0
6.99795636
12.1885555
51.4061085
70.5926210
rel std dev
0.49
-0.28
-0.33
0.12
SLOPE: -2.227E-04
INTER: 0.89249673
R2: *
0.06530589
Figure 4. LOTUS Analysis Spreadsheet of Manual Cal
35
-------
Calibration
Hydrocarbon Sources
LEAST SQUARES FOR SPAN 03
N I 1
N*»
NtY
YDEV2 est y
Y2
JY rel std dev
1
1
1
1
4
AVE CAL:
NO
0.94221
NOx
0.88804
03
0.98926
0 1.12601851
-8 0.99243298
14 0.92837585
58 0.91000502
AVE CAL:
XBAR;
Auto Tank 1 : 6/21 to 9/1
responsel CAL VALUE
0.30515 0.28751
0.!
responsel CAL VALUE
0.32207 0.29401
Auto Cal Gen : 5/30 to 11/4
response CAL VALUE
0.12218 0.12087
0 1.12401851 0.03378640
8 0.99263298 0.00254270
14 0.92837585 0.00019132
58 0.91000502 0.00103701
80 3.95703238 0.03755744
0.98925809
20.0000 .
decay*
-4.485E-05
531 agreeient
decay*
-5.304E-05
decay*
-7.179E-05
1.0428 0 1.26791769 0 0.86
1.0214 64 0.98532023 7.94106384 -0.30
1.0053 196 0.86188173 12.9972620 -0.79
0.8875 3364 0.82810915 52.7802916 0.23
3624 3.94322881 73.7186174
SO: 0.09689872 SLOPE: -2.479E-03
INTER: 1.04283546
R2: 0.50605026
Auto Tank 2 : 9/1 to 10/16
response2 CAL VALUE decay*
0.70602 0.66522 -5.687E-04
-0.423: agreement
response2 CAL VALUE decay*
0.75224 0.66802 -3.688E-04
responsel -response value for 21-JUN-84 determined froi least squares regression of response data
response2 -response value for l-SEP-84 detertined froi least squares regression of response data
decay* -froi least squares regression of response data
CAL VALUE -VALUE ENTERED INTO AUTOCAL DATABASE TO GENERATE DAILY CAL FACTORS
Figure 4, cont. LOTUS Analysis Spreadsheet Manual Cal Plots.
36
-------
MANUAL CALIBRATION - SPAN - NO
MANUAL CALIBRATION - SPAN - NOx
PS
u
k.
X
8
jj
o
u
I
1.4 -
1.3 -
1.2 -
1.1 -
1 -
(
0.0 -
O.8 -
0.7 -
o.e -
o.s -\
^ < 1
^-a a-
PS
o
B
f
2
s
B
o
u
1
o 20 40 eo
DAYS
D data + lln »f
l.S -
1.4 -
1.3 -
1J8-
1.1 -
1 -
o.o -!
0.8 -
0.7 -
o.e -
A K
>- ^^
0 2O 40 61
DAYS
a data * lln r«f
MANUAL CALIBRATION - SPAN - 03
o.s
Figure i. LOTUS Analysis Spreadsheet, of Manual Cal Plots
-------
u
I
MANUAL CALIBRATION - SPAN - NO
1.0
1.4 -
1.3-
IJt-
1.1 -
1 -
0.0 -
0.8 -
0.7-
0.0
0.0
leas
I
20
data
40
DAYS
I
00
80
O
s
o
u
I
09
MANUAL CALIBRATION - SPAN - NO2
ieoo
1.4 -
1.9 -
1.2-
1.1 -
1 -
o.o -
o.a -
'0.7 -
0.6-
0.9
20
Un r«(
d»U
40
DAYS
00
BO
Un i
kt
X
O
u
1
MANUAL CALIBRATION - SPAN - NOx
1.0
1.4 -
1.3-
1.2-
1.1 -
1 -
0.0 -
0.0-
0.7 -
o.e
0.0
leas
20
a data
40
DAYS
eo
80
a
H
U
O
u
0.
01
MANUAL CALIBRATION - SPAN - 03
1.9
1.4 -
1.3 -
1.2 -
1.1 -
1 -
0.0-
0.8 -
0.7 -
o.e -
0.9
less
20
Un
data
40
DAYS
t Un r»f
OO
eo
Figrire 5, cont. LOTUS Analysis Spreadsheet of Manual Cal.
-------
Hydrocarbon Sources Calibration
largest individual relative deviation from the average is 2.5 percent. The agreement
between the standard tanks and the RT1 tanks is very good. The absolute accuracy
of the hydrocarbon calibrations depends directly upon the concentrations assigned
to these tanks.
Table 7. Intercomparisons Among Sources
Source Source Date
RTI ethylene AIRCO 8/1-2
RT1 propylene AIRCO 8/1-2
AIRCO LOV/MV (Hi Cone) 7/18
AIRCO HUM (Hi Cone) 7/18
AIRCO LIQ STD 8/14 8/14
AIRCO HIMV; (Hi Cone) 8/14
LIQ STD 8/14 (C/area) HIMVJ (Hi Cone) 8/14
LIQ STD 8/14 (C/inch) HIMV! (Hi Cone) 8/14
LOVJMV: (Hi Cone) LOV.'MVJ (Lo Cone) 7/13
LOV/HV; (Hi Cone) HIMV.' (Hi Cone) 5/30
LOV/MV! (Hi Cone) 82HCCALMIX 5/30
LOV/MW (Hi Cone) SCOTTTAIIK 10/5
HIM!;; (Hi Cone) SCOTTTAIIK 10/5
LOV/1W (Lo Cone) SCOTTTAIIK 10/5
LOVJMV (Lo Cone) SCOTTTAIIK 8/17
AIRCO SCOTTTAIJK 8/17
39
-------
Calibration Routine NOx and O3 Calibrations Methods
Table 8. Comparison of Primary HC Calibration Sources
DCS ID SOURCE ANALYZED/ DEV REL DEV ABS REL DEV
MANUFACTURER
avg std
0.8% 0.87.
2.3% 2.3%
-0.7% 0.7%
-2.5% 2.5%
ave 0.963 0.077 1.6%
std 0.017 0.038 0.8%
rel std 1.8%
1
108
109
112
82HC
HMWHC
L1WHC
AIRCO
0.971
0.985
0.956
0.939
0.126
0.047
0.034
0.101
0.008
0.022
-0.007
-0.024
Routine NOx and O3 Calibrations Methods
This section describes the types of calibration methods routinely performed. The
section "Calibration Data Processing" describes how the data are processed, in-
cluding the QA steps. The calibration reports are included or refer to appendices
of calibration reports.
Automatic Source Sampling
The transfer technique to characterize the AUTOCAL sources for NOX and Os was de-
scribed above. Before and after every experiment, these sources plus the AUTOZEROs
are used to calibrate the instruments by scheduling 30 minutes to sample from the
span sources and from the zero sources by computer control (0200-0230 EDT, 1800-
1830 EDT for AUTOZERO; 0230-0300 EDT, 1830-1900 EDT for AUTOSPAli). The two low
concentration NO tanks used for AUTOCAL sources are listed in Table 2. The generator
built into the Os monitor is used for the 03 AUTOCAL source.
DATATRIEVE is used to maintain an AUTOCAL database. In this database, the de-
cay rates and the initial concentrations at the beginning of the season are used to
estimate the concentration of the span sources for each AUTOCAL performed. A data-
base procedure PRUIT.CALSJVIID-FACTORS computes calibration factors from the AUTOCAL
response and the computed tank concentration.
40
-------
Calibration Routine HC Calibration?
Routine HC Calibrations
Automatic Source Sampling
Routine HC calibrations were performed automatically before every experiment.
The GCs are activated several hours before injections and are allowed to produce
several chromatograms. A timer switches on the calibration source tank connected
to a calibration manifold at approximately 0300 EDT. The flowrate is adjusted to
provide an excess which is vented. The timer also switches a three-way valve to
change the GCs from the chamber sample manifold to the calibration source man-
ifold. The "low concentration" tank listed in Table 3 and discussed above was the
source most commonly used for an AUTOCAL, however other tanks listed in the table
were sometimes used. This would normally be when the presence of a morning site
operator was required many hours before sunrise. Sometimes calibrations were per-
formed after an experiment when a wider range of calibration species were desired.
The site operator indicated on the instrument checklist that a. calibration was per-
formed. When the run is logged in later in the data processing office, these notes
of calibrations performed will be logged into the calibration database (discussed
below).
Manual Precision Liquid Injections
More extensive calibrations were performed on days that were poor for performing
an experiment (poor sun or rain forecast). On these "CAL" days, several tanks
would be used for calibrating. More importantly, liquid standards could be injected
into the unused chamber for calibrating compounds with vapor pressures too great
to be stored in pressurized cylinders (e.g.l, 2,4-trimethylbenzene). These calibration
data are stored in a "CAL" folder. The presence of the "CAL" folder and a calendar
entry indicating a "CAL DAY" will trigger the processing of a calibration that was
not performed during an experiment.
Liquid injections of reasonably large amounts (100's of /*!) used to produce initial
injections for a smog chamber experiment were usually processed as calibration
sources also. The project coordinator decides if the injections are appropriate as a
calibration source and then logs in the calibration.
Injection Calibration Procedures
The chamber is usually vented more than 10 hours before injections are made.
Hamilton "Standard Microliter " gas chromatography syringes are used to inject
known volumes (in 10's to 100's //I) range) into the smog chamber. Stated accu-
racy is 1% of the syringe volume. The temperature is recorded for concentration
41
-------
Calibration Meteorological Sensor Calibrations
calculations. All data are recorded on the stripchart or by a special note on the RUN
SHEET or IliSTRUUEl.'T CHECKLIST.
Injection calibration data are transfered to the Official Calibration Source form
(OCS), then to the electronic OCSIC84 form which performs ideal gas law calcu-
lations, using the amount injected and the chamber volume. The electronic form
automatically looks up density, carbon number, MW etc.A.1] calculations are thus
recorded. Errors (from improper data transfer or calculation) are therefore mini-
mized.
The determination of chamber volume was by a "strapping" technique. The di-
mensions of the chamber were carefully measured and used to calculate the volume.
Several people have independently made these measurements with less than 1.5%
variation in volume.
Meteorological Sensor Calibrations
Solar/UV radiation Calibrations
Two sets of both the Eppley Black and White Pyranometer (Total Solar Radiation)
and Eppley Ultra-Violet Radiometer (Ultraviolet) sensors are used. One set is
used continuously and the other set is reserved as the References. Both sets are
calibrated by Eppley. The continuous sensors are sent back to Eppley annually
for recalibration. At Epply, the total solar radiation sensor is compared with the
Eppley group of reference standards at approximately one half a solar constant.
The UV sensor is calibrated by comparison with an Eppley standard of spectral
irradiance lamp.
Temperature Calibrations
The temperature thermistors are checked against an ice bath and a water bath near
normal high afternoon temperature, measured with a general laboratory mercury
thermometer.
Dewpoint Calibrations
There is no direct calibration technique for the dewpoint monitor. The sensor is a
high precision point thermometer sensor. The electronics of the instrument must
be frequently adjusted to balance the sensor bridge. The detector must also be
cleaned routinely. The dewpoint meter is checked by monitoring the saturated air
pulled through a water bubbler of known temperature. Response to dry air (air
42
-------
Paper Forms Calibration
pulled through Drierite) tests for proper cooling circuit operation needed to reach
low dewpoints.
Flowrate Measurements - Calibration
Accurate flowrates are required for the manual gas phase titration method for cali-
brating NOX and 03 and for making injections into the smog chamber. The primary
standards and routine flowrate measuring devices are a series of 10. 100, 1000, and
2000 milliliter soap bubble flowmeters. The volume of the soap bubble tubes was
determined by measuring water volume in NBS calibrated graduated cylinders. A
four place digital stopwatch is used for timing.
Time Measurement - Clock Calibration
All times reported are Eastern Daylight Time, even during winter. The main system
clock is housed within the computer. Time is set at the beginning of the run season
against standard time reported at the local weather bureau at the Regional Airport.
This time is verified several times during the year. Stripcharts are marked at the
beginning and throughout the experiment with time read from the computer clock.
Calibration Documentation and Processing Databases
A formal system has been developed to document and track the processing of the
large calibration database. This is important to a QA program to assure that
all data is processed and utilized. Various paper and electronic forms, calendars,
notebooks, and computer databcises are utilized. Forms help assure consistency
and completeness. Electronic forms allow for ease of organization, manipulation,
retrieval, and analysis. Some of the form fields are actually computed internal to
the form.
Paper Forms
Run and Calibration Calendars
Calendars at both the site and the office in the School of Public Health are used to
record days on which experiments were conducted. These entries also indicate days
on which only calibrations were performed.
Run Folder Inventory Check Lists
Instrument Checklists are filled out for each experiment. These lists identify the
calibrations that were performed (see Appendix A).
43
-------
Calibration Electronic Forms and Other Documentation
Stripcharts
Stripcharts from an experiment identify the calibrations that were performed, the
calibration source, the chart attenuation, the chamber temperature, and the number
of microliters that were injected.
Manual NOX/03 Calibration Form
Paper forms are used to record manually performed calibrations. These are filled out
at the site and transported to the data processing office where they are transfered
to electronic form (see below) and stored.
Official Calibration Source Form
All information, including the calibration source concentration used for a calibra-
tion, is recorded on Official Calibration Source (OCS) forms (see Figure 6). Infor-
mation concerned with several stages of processing and several different programs,
worksheets, and databases is recorded. The calibration information comes from sev-
eral sources including Stripcharts and attached notes. This form has an id number.
identification information, amounts and temperature, and official validated concen-
tration. This information is further stored in a computer database described below
which generates a file for processing calibration factors.
Calibration Pick Instructions
A paper form is used to give and record processing instructions to Data Proces-
sors for the raw calibration response data. The location of the calibration data is
recorded and the corresponding calibration sources used are identified. Notes can
be made about processing status and special problems and processing instructions.
Processing status is recorded on this form and concurrently in CALEllTRY. a computer
database for tracking processing status (see below). Figure 7 shows this form.
Electronic Forms and Other Documentation
COMMAND FILE
Command files are created at the smog chamber site on the computer used to
control the experiment. Commands issued to perform auto cals can be used later
to document the exact time the instruments are calibrated. Other instruments not
directly controlled yet auto calibrated are identified in remarks in the command file.
44
-------
Electronic Forms and Other Documentation
Calibration
OfRcal Calibration Source Form
Who is filling out this form:
Date form filled out:
SOURCE INFORMATION.
INSTRUCTIONS:
ID NUMBER:The id number is the next sequential calibration source
number.
DESCRIPTION: examples are: 82SCOTHC, or STDMIX2INJ. Please
print.
SOURCE DATE: the date tank was acquired or the date of in-
jection.
SERIAL NUMBER: the tank serial number or the initials of the
injector. Please print. /
VALIDATION OF SOURCE:
VALIDATION DATE The date that the source was approved.
INITIALS The person doing the approval.
ID NUMBER
Three digits.
DESCRIPTION Ten chars.
SOURCE DATE dd-mmm-yy
SERIAL NUMBERTen chars.
Figure 6. Official Calibration Source Form
45
-------
Calibration Electronic Forms and Other Documentation
SOURCE VALUE :: one page per species.Page of
SPECIES NAME Full name
SPECIES NUMBER Three digits
STATED CONG. 9.9999
ACTUAL CONG. 9.9999
CONG. UNITS e.g. ppmC :
VALIDATION DATE dd-mmm-yj
INITIALS
COMMENTS:
Figure 6, cont. Official Calibration Source Form.
46
-------
Electronic Forms and Other Documentation
Calibration
CAL PROCESSING INSTRUCTION FORM
CAL DATE INSTRUMENT
CAL INFO IN: RUN FOLDFR OR CAL FOLDER
CAL SOURCE -NAME -10 NUMBER
DIGITIZE (OR -PICK) CAL STARTING (TIME) . ENDS
M E R 0 E UITH
COMPOUND NAME ID * SIDE ATTN
(AND UNITS) NOTES
DATE INIT
DIGPIKR
CALFAC
RTTOWAX
TO OTR
of
Figure 7. Cal Pick Instruction Form
47
-------
Calibration ' Calibration Data Processing Status Database (CALEHTRY)
REMARKS FILE from Computer DA system
Another file created during the experiment by the computer is the remarks file
which often has comments about calibration events, especially if deviations from
the normal procedures are made.
Manual NOx/Os Calibration Form
Information from a paper form filled out during a manual NOx/Os calibration is
manually transferred to electronic form (in LOTUS 123 spreadsheet, see description of
general purpose programs below) where the calculations are performed automati-
cally.
Official Calibration Source
A computer database of the Official Calibration Sources also recorded on paper form
is maintained on a DEC VAX. This database can be summarized by a database
procedure and output in a form used by another program CALFAC to process
calibration factors when compared to the response data from a calibration.
Injected Conditions as Official Calibration Sources
Information from paper forms (primarily from the OCS paper form) of injected
liquids and solids for calibration is input, and processed in electronic form (in LOTUS
123 spreadsheet) where calculations are performed automatically (OCSIC). The
spreadsheet looks up the molecular weight, density, and carbon number using the
species id number (from the Official Species List), temperature at time of injection,
the chamber volume, and the ideal gas law. It was shown earlier as Table 4.
Calibration Data Processing Status Database (CALEHTRY)
A calibration data processing status database called CALENTRY is maintained on the
VAX-il/780 using the DATATRIEVE database system. The AA or PC logs in the existence
of calibration data for a given instrument and day. The PC and PP update CALE1JTRY
as the various tasks are performed. DATATRIEVE report procedures are used to assist
in updating the database, and generating status reports. The PC and PP use the
status reports to guide the processing of the calibration data. Table 9 shows the
CALEHTRY tasks flowchart and the events explained. Table 10 shows an example of
the CALSUM report.
Calibration Databases
Several calibration databases are maintained. Some double as spreadsheets to per-
48
-------
Calibration Databases
Calibration
Table 9
Processing System for Instrument Auto-Calibration Data
tot inst cal analysis
\
!CPI!
\
\
-
\
IOCS!
\
\
---(Q2V)--(DTR)--(C2L)----!CQA!
Step
!sss! == stages determined by Project coordinator (PC)
== stages determined by Peak Pickers (PP)
(sss) == stages determined by Computer Techs (CT)
CALENTRY stages
Meaning
PC
PP
PC
PP
CT
CT
CT
PP
PC
PC
PC
CPI
DPR
DCS
CFR
Q2V
DTR
C2L
CLR
CQA
NTB
BAD
CAL pick Inst
Digpik run
DCS ok
CALFAC run
Qfile to vax
DTR file updated
Cals to LSI
CALA11A run
Cal QA
Hot to be proc
Stop processing
Charts marked, cal status sheet in folder
Cal data digitized and R-file exists
Official Cal Source exists in DTR
Cal Factor computed, Q-file created
Q-file moved to VAX
Cal Data added to DTR data base
DTR CAL file loaded on LSI-11
Cal Factors analyzed
Cal Factor Quality Assurance
Not to be processed
Something is wrong with this run
49
-------
Calibration Calibration Database?
Table 10. CALSUM Example
Count of calibrations that have reached stated stages of processing
. . ' for METH
Total calibrations
265
******** HC Instrument data ************
CPI DPR OCS CFR Q2V DTK C2L CLR CQA C2S NTB BAD
181 181 181 181 179 179 0 0 144 0 53 3
Sum of CPI and NTB Is 234
CAL PROCESSING STATUS 6-May-1985
Page 5
CAL Calibration
DATE Instrument Steps
Carle III GC CAL pick Inst
0 Igp 1 k run
CALFAC run
13-Jul-1984
Carle I GC CAL pick Inst
D Igp 1 k run
CALFAC run
Carle II GC CAL pick Inst
D Igp 1 k run
CALFAC run
Carle III GC CAL pick Inst
D Igp Ik run
CALFAC run
CEA Formaldehyde CAL pick Inst
D Igp Ik run
CALFAC run
14-Jul-1984
Carle I GC CAL pick Inst
D Igp 1 k run
CALFAC run
Carle II GC CAL pick Inst
D Igp 1 k run
CALFAC run
Carle III GC CAL pick Inst
DIgp I k run
CALFAC run
16-Jul-1984
Carle I GC CAL pick Inst
D Igp 1 k run
CALFAC run
Carle II GC CAL pick Inst
D igp 1 k run
CALFAC run
Carle III GC Not to be proc
50
-------
Calibration Databases Calibration
form calculations. The databases serve several functions. Database commands and
procedures can sort by instrument, and species, make plots of cal-factors vs time,
perform regression and correlation calculations, convert, to a common attenuation
or range, output reports, output files for other programs, and in general allow us
to organize, retrieve and update information easily.
HC Tank Calibration Source Certification Database ("CAL")
Many different calibration sources were used for hydrocarbons. Some are "cal" tanks
and others are analyzed by the manufacturer with a stated uncertainty. Liquids
injected into the chamber are another calibration source. All "cal" sources were
intercompared to check for consistency. Data was collected in both peak height and
integrator area, and organized in a LOTUS 123 spreadsheet database. Calculations of
the intercomparison were then performed (see Appendix B).
Official Calibration Sources (OCS) Database
The Official Calibration Sources (ocs) Database were mentioned earlier in the elec-
tronic forms section. The database utilizes DATATRIEVE on the VAX-11/780. The data-
base appears as a collection of forms on terminal screens which makes updating
easier. Procedures have been written to generate reports (see Appendix C) and a
file of the calibration source data required by the program CALFAC used to calculate
the calibration factors.
Gas Phase Titration Database ("MANCAL" LOTUS 123 )
The resulting calculations (calibration correction factors) processed from the raw
calibration zero and response data are summarized and analyzed in another data-
base/spreadsheet using LOTUS 123. Plots of calibration data vs time with regressions
and correlation are produced easily. Trends and variation are easily recognized.
Reports are quickly produced.
NOX and Oz Auto-calibration Database: AUTOCAL
The NOX and O3 AUTOCAL data are maintained in a separate DATATRIEVE database.
Instrument calibration zeros and responses to the AUTOCAL sources are recorded.
Report procedures compare the corrected responses to the known calibration source
concentration and generate and list the calibration factors, sorted by date and time
of day (autocals for NOX and Os are usually performed twice a day).
51
-------
Calibration Security of Calibration Databases
Hydrocarbon Auto-Calibration Database
Another DATATRIEVE database similar to the NOX and Os AUTOCAL database is main-
tained for hydrocarbon calibration data. The database is more complex because of
the many more instruments which are calibrated. The database is also much larger.
A calibration factor is recorded for each species. Attenuation is also recorded. In-
teractive report procedures can retrieve and sort calibration data by instrument,
species, and date, and convert calibration data to a common attenuation. This
report is the primary source of final calibration data used to process data. Another
procedure outputs a file which can be read by another custom program (CALANA)
which statistically analyzes calibration data for a given species instrument combi-
nation and plots the calibration history for the run season.
Calibration Factors Used Database
r
Another DATATRIEVE database is used to record the calibration factors actually used
to process the final experimental time-concentration data. Some interpretation and
choice must be made when the statistics and the various sources of calibration fac-
tors do not agree. The PC makes a judgement based on all calibration information
and chooses a calibration factor. The chosen calibration factor is kept in a separate
data base. The resulting calibration factor database is maintained as a final source
of information for making a choice in a calibration factor. We believe that except
for rare occurences, calibration factors for most instruments do not vary much. This
database is consulted to see if the calibration factor chosen varies much from those
used on data from days (experiments) before and after the current experiment being
processed.
Security of Calibration Databases
There are three basic approaches to data security: inventory, retention of original
data in restricted areas, and maintaining several backup copies of all data in different
locations. All data are under lock at night and on weekends.
Inventory
At the site during an experiment, the calibrations that were performed are indicated
on a Instrument Checklist Form. Afterwards, before anyone is allowed to work with
the run folder contents, a run folder inventory is performed which documents all
data and forms. This assures that no data or documentation are lost.
52
-------
NOX and 03 Calibration
Retention of Data
All original data in both paper and electronic files are retained. Paper format data
(all stripcharts, forms) are not allowed out of the School of Public Health office.
Backup of Data
Multiple copies of data are maintained in several formats, in several locations.
Backup copies of raw and final data on floppies on IBM PC are kept in several loca-
tions. LOTUS 123 worksheet files are backed up in two locations. A hardcopy (print-
out) of worksheet formulas using a spreadsheet AUDITOR program is kept. Backup
copies of raw and final data on LSI-11 computers are maintained in several locations,
including: raw data and processed data sets backed up on two diskettes, processing
programs (sources) backed up on two diskettes, and whole image backup of LSI
hard disk. Backup copies of final data are maintained on the VAX and are further
backed-up on tapes of VAX files. There are also frequent printouts of final calibration
data in notebooks. The school operates a daily, weekly, and monthly backup of VAX
files with off-site vault storage for ten years for all VAX data.
Calibration Data Processing
This section describes and illustrates the actual data processing procedures. Pro-
cessing techniques are discussed separately for NOx/Os and HC data, because of
the different data and processing requirements and the different type of calibration
sources. The databases are also maintained separately. Another important distinc-
tion is the way the different types of calibration data are transported from the smog
chamber to the data processing office in the School of Public Health.
NOX and O3
Manual Calibration Processing for Characterization of AUTOCAL Sources
The calibration sources used for the NOX and Os monitors were discussed above. Also
discussed was the relationship between the manual and auto calibrations performed.
The manually performed gas phase titration calibrations are used to calibrate the
auto calibration sources.
AUTOCAL Data Processing
AUTOCAL data for zeros and spans is obtained twice daily. The PDF/11-40 computer
at the site initiates the calibration from a COMMAND file written by a site operator
(30 minutes zero; 30 minutes span). The calibration data is recorded along with the
experimental results by the comptiter DA system (one value every minute). Therefore
53
-------
Calibration HC Processing
the calibration data is transported along with the electronic data. Figure 8 shows a
flowchart of the autocalibration data processing procedures. These tasks are mostly
performed by the CT.
A computer program was written (FASTDV) to strip the calibration data from the
DVM data file (see Figure 8). One file is made for every AUTOCAL performed. There are
usually two autocals per day. The files are named MMDDYA.DVn where n is 1st
and 2nd calibration performed. These are called "A-files". The files are backed-up
on floppies. Each file is edited if necessary to eliminate readings taken in transition
to the calibration value. These edited files are named MMDDYE.DVn ("E files").
These are also backed-up on floppies. They are then appended together into a
file called AUTCAL.NEW. This file is appended to AUTCAL.DVM which is the
"OFFICIAL" file of appended "E files", which is also backed-up on floppies.
The AUTCAL. HEV.' file is further processed. It is averaged to a single value for both
zero and span with the program called ACALAV on the LSl-ll/23. A listing of the
processed averaged data is made and stored in the NOX/O3 FASTPLOT notebook. This
program generates a file called ACDTR.COH which contains the processed averaged
calibration data with additional DATATRIEVE commands. The file is transferred to
the VAX and activated as a command procedure which causes the data to be auto-
matically entered into the DATATRIEVE AUTOCAL database. The database also contains
the characterized concentration values of the AUTOCAL sources (see discussion above).
A report procedure in DATATRIEVE called PRIHT.CALSJUJD-FACTORS reports the original
calibration data by date and time and reports the calculated correction factors.
Table 11 shows the final AUTOCAL calibration factors for 1984.
HC Processing
Processing of hydrocarbon calibration data is more complicated than processing
NOX and Os calibration data. Each gas chromatograph monitors dozens, potentially
hundreds, of different compounds. Each species used in an experiment requires a
calibration factor for each instrument. Not every species is calibrated explicitly
before every experiment. Several different calibration sources are used. Resulting
calibration data must be reviewed before assigning a value for final data processing.
An explicit calibration source for a compound monitored in an experiment may
not exist and other techniques must be used to establish an appropriate calibration
factor. The different types of calibration sources and the documentation, forms,
and data bases associated with this processing were described above. This section
discusses in more detail the procedures of processing the calibration data.
54
-------
HC Processing
Calibration
.: a '
;S;n8>:f3,
XU' file
DISK j mmddyA.DVn
EDIT
mmddyE .DVn
APPEND
SLU^i AUTCAL.NEV
21^-J AUTCAL.DVM
v
ACALAV
I
CAL SOURCE \
VALUES
dANCAL
t
stat
rept
/
\
71
c
C
DISK
+
ESTTI:
DISK
J ^
>
J
^
ALANA
^^*
^Crn^ACDTR.COM
L/IJ3IV
JL
AX^», , DTR (AUTOCALyy)
REPORT
LISTING
Figure 8. AUTOCAL Processing Flowchart.
55
-------
Table 11. NOX AND 03 CALIBRATIONS FOR 1964
16-M»y-198S
DATE NOX ZEPO NO ZERO NOZ ZERO O3 ZERO NOX SPAN NO SPAM 03 SPAN NOX PESP NO RESP 03 *ESP NO CAL NOX CAL O3 CAL
o>
Jun23 -0.0026
JunZS -0.00Z6
JunZS -0.08Z6
Jun?6 0.001Z
JunZ6 -0.0014
Jun27 -0.0013
JunZ7 -0.00Z5
Jun28 -0.0020
JunZB -O.t'019
Jul'JC -0.0002
Jul07 -0.0014
Jul07 -fl.0012
Jul09 -0.0009
Ju)09 -B.0025
Jul 10 -0.0013
JullO -0.0021
Oulll -0.17016
Julll -0.0026
JulZ0 -0.0011
JulZI -0.0003
JulZl -0.0006
JuIZZ -0.e0.37
Jul22 -0.O009
Jul24 -0.0004
JulZ4 -0. H033
JulZS -0.PB08
Jul25 -0.0819
Jul2G -B.O010
Jul26 -0.1016
JulZB 0.0B10
Jul28 0.0023
Jul29 B.H00B
Jul29 -0.P003
Jul3l O.I'OOI
Auo,02 -0.W030
AutjeZ -0.P02I
Au.j03 -0.O018
AugC3 -0.0030
Aug04 -0.C042
AuqLM -0.0832
Auo«5 -0.C039
A.jolTS -0.0034
Aug06 -B.0042
Aug06 -B.1'040
Aug07 -O.f'057
Au'i07 -0.1/034
Augl'B -0.0062
Aug!>8 -0.C038
Au-jlG -0.0008
0.0001
-0.P004
-0.0006
0.0025
-0.0008
0.0005
-0.0009
0.0000
-0.0006
0.0008
3.B01B
0.BO00
0.0013
0.0002
0 . 00 1 0
fl. 0004
0.0008
-0.00.03
0.0007
0.0010
0.0028
0.0013
0.0004
0.0013
-0.0009
0.0039
B . 0 0 'J 1
c.oeod
0.*0u2
e . 03 1 2
0.0012
0.0017
0.0013
0.J08;.'8
-0.0C'05
-0.0010
0.0092
-0.C809
-0.08Z0
-0.0069
-0.0023
-0.0012
-0.0021
-H.CUZ0
-0.00.il
-0.0018
-0.0036
-3.0021
0. «0<71
-0.0015
-0.0010
-0.0004
-0.0004
0.0004
-0.0006
-0.O004
-0..90 10
0.0002
0.0003
-0.0012
-0.0001
-0.0005
-0..0011
-0.0007
-0.3007
-0.0008
-0.B005
0.0002
0.0005
0.0009
0.0005
0.0306
0.0008
0.B003
B.iraee
0 . 0005
0.0005
o.aoes
0..'I018
0.0034
0.0005
0.9068
0.0016
O . Mf 008
0..00B8
0.0004
0./JB04
0.UC02
0.0003
0.0008
0.0005
O.O007
B.0B04
0.0806
0.0004
0.01)04
0.0003
0.0015
-0.0148
0.0008
0 . 00 1 I
0.0008
0.0022
0.0007
0.0B12
0.0007
0 . 00 1 9
0.0002
0.0007
0.0010
0.0007
0.0009
0.0006
0.001 1
0.0006
0.0010
0.0907
0.0096
0.0010
0.0008
0.0012
0.0007
0.0007
0.0006
0.0009
0.0007
0 . 00 1 6
-0.0021
-0.01)08
-0.001 1
-0.0014
0.0006
0.0006
0.0013
0.C007
0.0009
0. 0110 7
0.00U8
0.0008
0.0089
0.0007
0.0088
0.0007
0.0015
0.0007
0.0008
0.0006
0.3259
-9.9999
0.3200
0.3267
0.3190
0.3256
0.3207
0.3248
-9.9999
0.3140
0.3138
0.3105
0.3193
0.3130
0.3163
0.31 19
0.3145
0.3060
0.3205
0.3224
0.3200
0.3229
-9.9999
0.3220
0 . 3 1 3«
0.3200
-9.9999
-9.9999
0.3100
0.3219
0.3237
0.3239
0.3244
0.3193
0.3424
0.3342
0.3249
0.31G3
0.3150
-9.9999
-9.9999
0.3153
0.3141
0.3122
0.3101
0.3111
0.3006
0.3111
0.3230
0.3101
-9 . 9999
0.3061
0.31 16
0.3962
0.3121
B.3t''9
0.3090
-9.9999
0.2988
0.2985
0.2954
B. 31/39
0.2'JBE
0.3H16
0.2274
0.2994
0.2?28
0.3061
0.3'j'77
0.3061
0.3S/1
-9.9999
0.3(171
0.2937
0.3053
- n . 9 11 2 9
-3.9999
0.3030
a. 31! 6 6
0.3079
0.3K94
0.3H93
0.3f.MB
0.3164
0.3155
O. 2/782
.0.3" 14
O.P'ign
-9.9'.!99
-9.9399
0 . 30 1 2
0.2991
a.Z'lUZ
0.2'jr,o
)1.2'.I75
U.2950
0.2973
0.3071
0.1139
-9.9999.
B. 12!i3
0. 1234
0. 1273
0. 1291'
0. 1261
II. 1275
e. 1284
0. 1220
0. 1263
0. 1225
0. 1ZI.-1
0. 1255
0. 1264
a . 1 1 4 r.
0 . 1 1 '. (I
0.1116
0. 11 1 1
0.1142
U.I 146
0. 1175
0.1174
0. 1204
0. 1 H,S
11. 1 IUG
0. 1 1 tip
0 . 1 1 8H
B. 1 173
0. 1 168
H. 1163
il. 1189
0. 1 103
0 . 1 Z 1 fi
0.1212
0 . 1 2«i'
U. I21B
0. 1 IU7
e . i r. i
0. 172
0. 12G
0. 144.
0. 091.
0. 16i:
0 . 148
n . i?;:
0. IG4
H . 1 8 I
0. 12.03
0.3235
-3.9973
0.3226
e.3?ss
0.3204
0.3269
0.3232
0.32GB
-9.9980
0 . 3 1 50
0.3152
0.3117
0.3202
0.3155
0.3176
0.3140
0.3161
0.3086
0.3216
0.3227
0.3214
0.3236
-9 . 9990
0.3Z24
0.3163
0.3?')8
-9 . 9(.1;I0
-9.9'.'B9
0.3196
0.3219
0.3214
0.3239
0.3247
O.3192
0.3454
0.33G3
0.3?fi7
0.3193
0.3?00
-9.9967
-9.9960
0.3107
0.3103
0.3IG2
B.3158
fl.3145
0.3140
0.3119
0.3230
0.3100
-9 . 9995
0.3067
0.3091
0.3070
0.3116
0.3078
0.309B
-9.9993
0.Z9B0
0.2975
0.2954
0.3026
0.2986
0.3006
0.2970
0.2986
0.2931
0.3054
0.3067
0.3053
0.3058
0.3058
0.2996
0.3044
*******
0.302B
0.3054
0.3067
0.3077
0.3000
0.3032
0.3169
0.3165
0.30B0
0.3023
0.3019
-9.9990
-9.9976
0.3024
0.3012
0.3002
0.2991
0.2993
0.29B6
0.2994
0.3070
0.1287
0. 1242
0.1276
0.1251
0.1283
0.1249
0.1268
0.1265
0.1226
0.1256
0.1215
0.1257
0.1246
0.1258
0.1135
0.1142
0.1106
0. 1 104
0. 1 136
0. 1 136
0.1167
0.1162
0.1197
0.1148
0.1180
0.1151
0.1181
0.1157
0. 1189
0.1171
0. IZ00
0.1197
0.1212
0.1206
0.1187
0.1211
0.1178
0.1154
0.1164
0.11 18
0.1135
0.1089
0.1154
0.1141
0.1157
0.1157
0.1173
0.1197
0.9268
-0 . 0Z87
0.9364
0.9290
0.9354
0.9214
0.9328
0.9266
-0.0Z87
0.9620
0.9634
0.9703
0.9469
0.9595
0.9530
0.9645
0.9592
0.9772
0.9364
0.9323
0.9366
0.9349
-0.0286
0.9345
0.9539
0.9387
-0.0286
-0.0286
0.9435
0.9351
0.9312
0.9280
B.9271
0.9414
0.9004
0.9016
0.92G3
0.9437
0.9448
-0.0285
-0.0285
0.9431
0.9467
0.9499
0.9532
0.9525
0.9546
0.9521
0.9272
0.8746
-0.0Z87
0.8903
0.88Z2
0.8962
0.8783
0.8883
0.8784
-0.0287
0.9100
0.9093
0.0195
0.8948
0.9081
0.9026
0.9123
0.9061
0.9ZB1
0.8892
0.8861
0.8896
0.8834
-0.0286
0.8864
0.9035
0.8907
-0.0286
-0.0Z86
0.8939
0.8900
0.8886
0.8816
0.8794
0. 8942
0.0261
0.8415
0.8733
"0.8935
0.8914
-0.0285
-0.0ZB5
0.8949
0.8958
0.9018
0.9028
0.9065
0.9055
0.9052
0.8791
0.92.15
-0.0119
0.9SSB
0.9298
0.9484
0.9Z4Z
0.9493
0.9345
0.9367
0.9619
0.9383
0.9700
0.9364
0.9447
0.9351
.0364
.0295
.0630
.05U0
.0286
.02G6
.0006
.0049
0.9744
I .0163
0.9078
1 .0127
0.9863
1 .0068
fl-9705
0.9935
0.96t'9
0.9714
0.9SOZ
0.9617
0.9771
0.9572
0. 9049
1 .1/030
P.995Z
1 .0355
1 .02IVO
1 .06i.'4
1 .00/6
1 .0134
0.9?94
0.99P7
0.91)51
0. 96H6
-------
Table 11, COnt. NOX AND 03 CALIBRATIONS FOR 1984
ie-May-1985
DATE I NOX ZERO NO ZERO N02 ZERO O3 ZERO NOX SPAN NO SPAN O3 SPAN NOX RESP NO RESP 03 RESP NO CAL NOX CAL O3 CAL
AuglO -0.0040
AuglO -0.004S
Au3l9 -0.0047
Augl9 -0.0035
Aug20 -0.0048
Aug20 -0.0047
Aug22 -0.0045
Aug22 -0.0012
Aug23 -0.0036
Aug2S 0.0050
Aug25 0.0020
Aug-'7 0.0016
Aug28 -0.11010
Aug29 -0.0022
Aug29 -0.0007
Sep01 -0.P0I4
Sep01 -0.IJ014
Sep02 -0.0028
Sc-p02 -0.0037
Sep02 -0.P027
Sep03 -0.J.042
Sep06 0.0055
Sep06 0.0061
SepOO 0.0047
Sep08 -0.0016
Sc-p09 -0.0016
Sep£l9 -0.0027
Sepl6 -0.0033
Sepl6 -0.0047
Si-'pl7 -0.0038
Sepl7 -0.0042
Sop 18 -0.0037
Sop I 8 -0. 110 4 4
Sup 19 -0.1)040
Supl9 -0.0030
Sop21 -0.0052
Stp21 -0.0027
S,ip25 -0.0052
Sep2S -0.0001
Oct.03 -0.8028
Oct05 -0.0037
OctOS -0.0036
OctOS -0.0050
Oct06 -9.9999
Oct07 -0.0054
Oct07 -0.0054
Oct09 -0.0040
Oct.10 -0.0045
Oclll -0.0051
-0.0027
-0.0625
-0.0027
-0.0020
-0.0027
-0.0031
-0.0034
-0.0023
-0.0026
0.0000
-0.0010
-0.0014
-0.0618
-0.0026
-0.0018
-0.0022
-0.0022
-0.0028
-0.0031
-0.01)25
-0.0021
0.0003
0.0017
a . 00.45
-0.0013
-0.0056
-B.0014
-0.0023
-0.0029
-0.0023
-0.0030
-0.0026
-0.0032
-0.00.11
-0.0631
-0.0034
-0.0025
-0.0036
-0.0038
-0 . 0028
-0.0033
-0.0041
-0.0016
-9.9999
-0.0045
-0.0065
-0.0043
-0.0055
-0.0048
0 . 00 1 1
0.0004
0.0002
0.0009
0.0006
0.0010
0.0014
0.0030
0.0016
0.0060
0.0050
0.0044
0.0031
0.0026
0.0027
0.0031
0.0031
0.0028
0.0018
0 . 00 1 9
-0.0001
0.0073
-9.9999
0 .O059
0.0015
0.0009
0.0006
0.001 1
0.O004
0.0600
0.0009
0.0005
0.0010
0.0005
0.0020
0.1)006
0.0022
0.0002
0.0014
0 . 0020
0.0020
0.0026
0 . 00 1 9
-9 . 9999
0.001 1
0 . 00 1 9
0.0018
0.0025
0.0019
0.0007
0.0007
0.0008
0.0015
0.0008
0.0010
0.0008
0.0608
0.0008
0 . 00 1 0
0.0010
0.0023
0.0008
0.00OB
0.0008
0.0017
0.11017
0.0008
0.0007
0.0007
0.0009
0.0038
0.0022
-9.9999
0.0021
0.0007
0.0010
0.0031
0.0013
0.0007
0.0011
0.00>I8
0.6008
0.0007
0.0014
0.0008
0.0019
0.0007
0.0009
0 . 000B
0.003B
0.0007
0.0007
-9.9999
0.00O7
0.0009
0.0007
0.0020
0.0407
0.3164
0.3172
0.3140
0.3139
0.314Q
0.3147
0.3183
0.3209
0.3197
0.3140
0.3155
0.3165
0.3144
0.3143
0.3129
0.3153
0.7439
0.7377
0.7377
0.72G9
0. 7390
0.7447
0.7447
0 . 7 5 'JO
0. 76ul
11.7626
0. 753G
M.750D
0.7520
0.75U1
0.7505
0.7541
0.7479
0.7552
0.7453
0. 731'J
0.7273
0.7271
-9.9999
-9.9999
0.7290
0.7257
0.730C
0 . 739C
0.7420
0.7414
0.7448
0.7400
e. 7364
0.3019
0.2099
0.2999
0.2906
0.3002
0.3fi04
0.3H37
0.3049
0.3044
0.2945
0.2965
0.2992
3.2933
0.2900
0.2972
0.2997
0. 7u6u
0.6920
0.6920
0.6U13
0.6049
0.b'.)69
0.6069
0.7140
0.7149
O.7iri6
0.7C.-32
11 . 7U I 0
0.71)25
0. 7JS83
B.G9'JS
0.7050
S.6U7JO
0.7JI4O
0.69 JO
0.G771
0.6706
0.6718
0.G691
-9.9999
0.6742
0.6723
0.6760
0 . 6U 7 2
0.6904
0.6301
0.6910
0.6bS0
0.6034
0. 1 167
0. 1 162
0. 1 17'.,
0. 1 183
0. 1 192
0. 1 ICG
0.1 20f;
0. 1201
6. 1265
0.121V.
0. 12E1/
0. 1255
0. 124G
0. 12SG
0 . 1 1 fj n
0. 1 17;.'
B. I7Z
e. 172
0. 172
0. 139
0. 173
0. 193
tl . 19',:
0 . 1 'J 2
0. 130
a . 194
a. ion
0. 1270
o. i2'ji;
O . 1 2 7 a
0 . 1 2 C li
0.127V.
0.1 2SC
II. 12 ('.'.'
0.1213
0 . 1 1 a 7
tl. 1 1U3
K. 1 165
0.1181
-9.999 9
0. 1122
0.1 1 OS
0. 1 1 13
0 . 10G3
0.1076
0. 1073
0. 1074
0. 107S
I). IDbl
0.3204
0.3217
0.3187
0.3174
0.3196
0.3194
0.3228
0.3221
0.3233
0.3090
0.3135
0.3149
0.3154
0.3165
0.3136
0.3167
S.7453
0.7405
0.7414
0.7296
0.7432
0.7392
0.7386
- 0.7543
0.7617
0.7642
0.7E63
0.7533
0.7576
0.7G19
0.7547
0.7578
0.7523
0.7592
0.7433
0.73G7
0.7300
0.7323
-9.9S98
-9.9971
0.7327
0.7293
0.7356
0.7474
0.7468
0.7496
0.7445
0.741S
0.3046
0.3024
0.3026
0.3006
0.3029
0.3035
0.3071
0.3072
0.3070
0.2945
0.2975
0.3006
0.3001
0.3006
0.2990
0.3019
£.7082
0.6948
0.5951
0.6838
0.6970
0.6966
0.6952
0.7135
0.7162
0.7202
0.710G
0.7033
0.7054
0.7106
0.7025
0.7076
0.7002
0.7071
0.G961
0.6805
0.6731
0.6754 -
0.6729
-9.9971
0.6775
0.6764
0.6806
0.6949
0.6946
0.6953
0.6905
0.6882
0. 1 160
0. 1155
0.1167
0.1168
0.1184
0.1175
0.1200
0.1193
0.1257
0.1205
0. 1240
0. 1232
0. 1240
0.1238
0. 1 177
0. 1 155
S.IIS5
0. 1164
0. 1165
0. 1 182
0. 1 164
0. 1 185
0.1171
0.1118
0. 1 187
0.1145
0. 1239
0. 1245
0.1271
0.1255
0. 1267
0. 1250
0. 12C2
0. 1 199
0. 1 179
0. 1 164
0. 158
0. 172
0. 114
0. 102
0. 106
0. 1069
0. 1064
0. 1067
0. 1055
0. 1054
0.9342
0.9410
0.9462
0.9465
0.9391
0.9373
0.9260
0.9257
0.9261
0.9GS1
0.9553
0.9452
0.9466
0.9440
0.9499
0.9403
S.9383
0.9563
0.9559
0.9717
0.9526
0.951 1
0.9530
0.9273
0.9238
0.S18U
0.S304
0.9354
0.9326
0.9252
0.9358
0.9284
0.9382
0.9284
0.9431
0.9B33
0.9739
0.9678
0.9714
-0 .0650
0.9579
0.9595
0.9529
0 .0607
0.9326
0.9330
0.9307
0.9365
0.9389
0.8881
0.8845
0.8927
0.8964
0.8901
0.8906
0.8809
6.8828
0.8794
0.9193
0.9056
0.9022
0.9007
0.8974
0.9057
0.8963
0.8557
0.9009
0.8998
0.9144
0.8970
0.9000
0.9007
0.8807
0.8722
0.8687
0.8778
0.8769
0.8710
0.8664
0.8747
0.8705
0.8768
0.8683
0.880'J
0.8935
0.9017
0.8963
-0.0G56
-0.0653
0.8894
0.8936
0.8853
0.060G
0.8707
0.8714
0.8669
0.8722
0.87S1
0.990J
0.9943
O.90J4
6.9826
0.96C7
ft . 9 7 C 1
0.9b4b
0.9602
0.9 1ft 7
0.94C3
0.9220
O.92C9
0.92C3
0.9212
it . 9 1> <* $
M.985G
S.98C&
0.9773
0.97CS
tf.9624
0.97&7
0.9576
0.960(1
0 .01 1 2
1.6137
0.9541
0 .9891
0.9UI.J
a.9JL7
0.8&C6
0.89/9
0.88L8
0.901.9
0.8913
0.931.5
0.95:13
0.9b!,G
0.96UI
0.9DC6
-0.0112
0.9939
. 0 1 JU it
.00C5
. 6 1 1 U
.041.7
.CUF.G
.0413
. oc :' 4
. o '_ :. a
-------
Table 11, cont. NOX AND 03 CALIBRATIONS FOR 1984
16-M.y-I985
-------
HC Processing Calibration
Figure 9 shows the flowchart for HC calibration data processing with CALEIITRY
events indicated. Table 12 shows the RU!!E!JTRY events flowchart.
Identification of Calibration Chromatogram and Calibration Source
If one or more forms of documentation indicate that a calibration(s) was performed,
then the first step is to identify the data and calibration source(s). If the calibration
was performed during an experiment, it was logged in CALENTRY when the experiment
was logged in RUMENTRY. Otherwise, the run calendars indicate a calibration event
and the presence of a new calibration folder. If the calibration occured during an
experiment, the run folder INSTRUMENT INVENTORY CHECKLIST indicates the instruments
calibrated, otherwise, the stripcharts present in the calibration folder indicate which
instruments were calibrated. Calibration sources are identified on the stripcharts.
Issue Calibration Pick Instructions
Calibration processing (pick) instructions are given by the PC via the CAL PICK IN-
STRUCTION form. The form is stored in the CAL PICK INSTRUCTION NOTEBOOK by date.
The fact that pick instructions are given is entered into CALENTRY. Several QA steps
occur at this point. Documentation is checked for reasonableness and completeness
(sources, chart settings, instrument settings). The chromatograms are inspected for
reasonableness (match standard example chromatograms with source for conditions
specified).
Official Calibration Sources/ OCS Report
The raw data can be processed (picked), but before final processing can occur, the
calibration source data must be validated and entered into the OFFICIAL CALIBRATION
SOURCE (ocs) DATATRIEVE database and the OCSFIL.DAT updated. The forms used for
recording the OCS data were discussed above.
Several sources are used throughout the season. These were listed above. The
validation procedure was also reported. The first step is to check the report from
the ocs database to see if the source has been entered. If the DCS does not exist then
it must be created or at least validated. The details of processing single (used once)
OCS are discussed next. The DCS report is listed in Appendix C.
Create OCS for Single Injection
These ocs are created in the smog chamber. The amounts of HC that are injected
are documented along with the chamber temperature in the calibration folders or
on the RUN SHEET. A LOTUS 123 database/spreadsheet ocsic (described above) is used
59
-------
Calibration
HC Processing
CAL
FOLDER
V
A
RUNFqLDERJ
DTR (OCS)
DIGPIK
mmddyRiii
CALFAC
*
eUSf IjQ
DISK J
mmddyQ.iii
DTR (HCCAL)
I
HCCAL84.DAT
CALANA
I
T^nr
_*
stat
rept
Figure 9. HC Autocal Processing Flowchart.
60
-------
HC Processing
Calibration
Table 12.
Processing System for Instrument Auto-Calibration Data
tot inst cal analysis
\
!CPI!
\
\
-
\
IOCS!
\
\
---(Q2V)--(DTR)(C2L)----!CQA!
Step
!sss! == stages determined by Project coordinator (PC)
== stages determined by Peak Pickers (PP)
(sss) == stages determined by Computer Techs (CT)
CALEUTRY stages
Meaning
PC
PP
PC
PP
CT
CT
CT
PP
PC
PC
PC
CPI
DPR
DCS
CFR
Q2V
DTR
C2L
CLR
CQA
NTB
BAD
CAL pick Inst
Digpik run
DCS ok
CALFAC run
Qfile to vax
DTR file updated
Cals to LSI
CALAUA run
Cal QA
Not to be proc
Stop processing
Charts marked, cal status sheet in folder
Cal data digitized and R-file exists
Official Cal Source exists in DTR
Cal Factor computed, Q-file created
Q-file moved to VAX
Cal Data added to DTR data base
DTR CAL file loaded on LSI-11
Cal Factors analyzed
Cal Factor Quality Assurance
Not to be processed
Something is wrong with this run
61
-------
Calibration HC Processing
to process this information. The species id number is used to look up all the
necessary species specific information (density, molecular weight, carbon number,
etc.) and the program calculates and records the ideal concentration. Table 4 lists
the OCSIC report. All DCS information is recorded on DCS forms and recorded in
the DCS notebook as described above. The validated ocs concentrations are entered
into the DATATRIEVE database CAL^SOURCESJMS. A database report procedure is used to
update the OCSFIL.DAT which is transfered to the LSI to await final processing. The
fact that the ocs exists is entered into the DATATRIEVE CALEIITRY database.
Run DIGPIK on calibration Data to Make R-files
Once calibration processing instructions have been given, raw data are picked using
a program on the LSI (DIGPIK) where the data are digitized and a response file
(R-file, MMDDYR.iit) is produced. The files are backed-up (copied) onto floppies.
Run CALFAC to Calculate Calibration Factors
Once the response data has been picked and the ocs data has been updated to OCS-
FIL.DAT, the calibration factor can be calculated. A program on the LSI CALFAC is run
which reads the internal documentation in the response file R-f lie and compares the
data with the appropriate ocs to generate the concentration per response unit cali-
bration factor. A QA step is performed here. The calibration factors are inspected
and reviewed in context with the existing database with the program CALLOK which
plots or lists the data vs. Julian date with a linear least-squares fit. The program
identifies calibration data which varies by more than a specified percentage. The
generated calibration data is output in files ("Q"-files) specific for the instrument
and the day calibrated. These files are backed-up on floppy,
The Q-files are transferred and updated to the DATATRIEVE HC_CAL database if the
calibration data seems reasonable. A DATATRIEVE report procedure PRIHT-HC.CAL.FACTORS,
is run to generate an updated calibration list sorted by instrument, species, and
date. All data are converted to a single specified attenuation for easy inspection.
Another DATATRIEVE procedure, PRIMT.FILE_CALA1IA, is used to generate another file
which is then transfered to the LSI-ll/23 computer to be used by a program called
CALAtIA which performs a statistical and graphical analysis of the calibration data-
base.
We expect a certain amount of noise or variation based on experience. Usu-
ally the calibration factors are very stable with little or no trend demonstrated.
Variations are found, however, and some are real indications of instrument change.
Discrepancies that are more than a few percent are investigated. Figure 10 shows
62 .
-------
HC Processing Calibration
example CALANA plots and statistical reports. Appendix D contains a larger collection
of these plots and statistics.
The statistical report first lists the data range searched in the database. The
number of data points found and included for the analysis are listed. The slope
and intercept of the straight line fit through the data are listed with the standard
deviation and relative standard deviation. The correlation coefficient is also listed.
The minimum, maximum and mean of the data are shown with the standard devi-
ation and relative standard deviation. Lastly the equation of the straight line fit is
shown. Table 13 shows a compilation of the statistics from this set.
This analysis shows that little or no trend was observed in the data. If the line
was perfectly horizontal, the correlation factor would be 0.0. Table 13 shows the
means, standard deviations, and relative standard deviations for the more complete
data set shown in Appendix D. The average relative standard deviation is just under
10 percent with a standard deviation of 2.2 percent. This suggests the uncertainty
in the data would be about 10% if the calibration factors were used straight from
the initial processing and no additional information and QA techniques were used
to check and tune the calibration data. Because specific calibration factors are used
and because the data are subjected to further examination and adjustment, the
uncertainty in the final data is less than 10%.
Valid deviations can be caused by many factors, including variation in the in-
strument operating characteristics. Calibration factors specific to the current in-
strument status should be used rather than some average value. Non-valid sources
of calibration data deviation include improper processing. Many QA measures are
taken to prevent these errors. These measures will be discussed further.
There are several explanations for variation in calibration data. Insufficient
sample flow will result in very low responses and consequently very large calibration
factors. Sometimes the instrument operating conditions change (e.g. carrier flow)
which changes the peak width arid consequently the calibration factor. Incorrectly
recorded attenuation will, of course, cause generated calibration data to be off by a
factor of multiples of 2 or 10. The Perkin Elmer Sigma 10 integrator which yields
areas not affected by the GC attenuation settings, can be compared with areas
of calibrations performed on other days to determine if this type of problem has
occurred; the attenuation can be corrected in the "R" file and reprocessed.
Estimation of Calibration Factors for Infrequently Calibrated HC
The technique discussed above requires an explicit calibration source for the com-
63
-------
Calibration
HC Processing
0.40
0.30
I I I I I I I I I I i| I I I I I I I I i I I I f I I I I I I I I
Carle I GC !
PROPYLENE 1
PPMC/IN Atten = 2.00
£ s.2»k
0.10
0
,
I IV
29-AUG
/ ^\
0.<
-J0.30
_j g>
O
0.10
' IS0 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320'
Julian Date, 1984
I*
For the species PROPYLENE and the instrument Carle I
GC:
Beginning Caldate:
Day:
Ending Caldate:
Day:
Upper Cal Factor Limit
Lower Cal Factor Limit
30-MAY-34
150
16-NOV-S4
320
0. 40000
0 . OOOOO
The number o-f data points
The number included was
The slope o-f this fit is
with a std dev o-f
The intercept is
with a std dev o-f
The std dev of the fit is
The correlation coeff is
62
62
1 . 3522654E-004
6.6911662E-OO5
1.6694184E-001
1.5341250E-002
1. 30799 19E-002
0.25246
rel . dev = 49.5 7.
rel . dev = 9.2 7.
PE = 0.01205
The min, mean, max are 0.1304O, 0.19760, 0.25009
The standard deviation is 0.01353 rel. dev = 9.4 7.
Cal = 0.00014 X (Julian Date - 150) + 0.18723
Figure 10. CALANA Example Figures.
64
-------
HC Processing
Calibration
O
LL.
ITJ
?.
0.
A
*'
0.
i i | i | i | i | i
i
i
25J-
i
i
i
2»|-
h
I
lL '
'jJ3i-MAY
J-
i
<
| 1 | 1 j 1 j 1 | 1 | i | 1 :' 1 | 1 ! 1 |
Carie II GC
ETHYLENE
FFMC/IN Atten = 2.00
f*
M ' l'
._ .... II
iS-jLL « v\. . ti
* ^a^^3>r_ _\R» ^ ^.-> -^- - -
1 i '
-
-^
^^
^~
~
A *^A
?.25
.. ,-..
" Q|
fl 1R "^
iij
i j
-------
Calibration
HC Processing
I I
Carle III GC
ISOPENTANE
FPMC/IN Atten = 2.00
i O
1 ID
1 S"
0.10
1
L L_L I I ! I I !. I I I M I ! I I I I I I I I I I ! ! ! I ! I I
160 170 130 190 206 210 220 230 240 250 260 270 280 290 300 310 320'
Julian Date, 1984-
00
For the species ISOPENTANE and the instrument Carle III GO;
Beginning Caldate:
Day:
Ending Caldate:
Day:
Upper CalFactor Limit
Lower CalFactor Limit
3O-MAY-S4
150
16-NQV-84
320
0.40OOO
O.OOOOO
The number o-f data points 60
The number included was 60
The slope of this fit is -1.222656E-004
with a std dev of 5.6O03271E-005
The intercept is 2.2209272E-001
with a std dev of 1.273873BE-002
The std dev of the fit is 1.6698649E-002
The correlation coeff is -0.27557
rel. dev = -45.3
rel . dev = 5.7 "/.
PE = 0.01113
The min, mean, max are
The standard deviation is
0.15936, 0.19463, 0.23074
0.01722 rel. dev = 3.3 '/.
Cal = -0.00012 X (Julian Date - 150) + 0.20375
Figure 10, cont. CALANA Example Figures.
66
-------
HC Processing
Calibration
Table 13. Summary of CALANA Statistics
Carle
Carle
Carle
I
II
III
propylene
isopentane
ethyl ene
ethane
propylene
propane
1-butene
n-butane
cis-2-butene
trans-2-butene
1 ,3-butadi ene
isopentane
n-pentane
benzene
toluene
m-xylene
o-xylene
methanol
ave
std dev
MIN
0. 130
0. 140
0.074
0.013
0.369
0.428
0.074
O.OS3
0.070
0.091
0.090
0. 160
0.228
0.080
0. 114
0. 187
0.256
O.EJ35
MAX
0.250
0. 196
0. 198
0.231
0.976
1.059
MEAN STD DEV REL STD DEV
0.198 0.019 9.47.
0.158 0.010 6.37.
0.121
0.147
0.353
0.607
0.009
0.007
0.054
0.060
7.67.
3.07.
9.8%
9.97.
0.074
0.053
0.070
0.091
0.090
0. 160
0.228
0.080
0. 114
0. 187
0.256
O.EJ35
0.130
0.131
0.127
0.132
0. 137
0.231
0.347
0. 106
0. 167
0.269
0.318
1.206
0.100
0.102
0.105
0. 115
0.116
0. 195
0.289
0.090
0. 137
0.225
0.281
1.050
0.011 .
0.016
0.015
0.011
0.0.12
0.017
0.028
0.009
0.014
0.027
0.024
0.143
1 1 . 57.
1 1 . 77.
14.07.
9.27.
9 . 97.
8.87.
9.87.
10.37.
10.07.
12.07.
8.77.
13.77.
9.97.
2. 27.
67
-------
Calibration HC Processing
pound being monitored to generate a calibration factor. Sometimes calibration data
is sparse or the species to be processed has no explicit calibration source. There are
several techniques for establishing appropriate calibration factors. These are often
used together.
When sufficient supporting calibration data exists, a plot of the calibration
factors of well calibrated compounds measured on the same column, as a function
of retention time, can be least-squares fit to an exponential function. A LOTUS
123 spreadsheet called CALEXTR is used for this. Figure 11 shows plots for the two
columns in the Carle III GC. Our experience has been that good fits are obtained.
The calibration factor for the uncalibrated compound can be estimated by locating
its retention time on the curve. The ratios of the frequently calibrated compounds
to the less frequently calibrated compounds can be determined for future needs. If
integrator data is available, a check is done using integrator areas ratioed to the
estimated concentrations of all the compounds.
If the compound requiring a calibration factor is part of a well characterized
mixture (the source), the concentrations of the other compounds can be used to
determine the concentration of the compound in question by ratios. The calibration
factor can then be calculated.
If reliable integrator data is available, the area count can be calibrated with
any known calibration sources, and the corrected area used to determine the con-
centration of the uncalibrated compound. The concentration can then be used to
calculate the calibration factor for the picked data.
If good calibration data exists for all species at some time, ratios of less fre-
quently calibrated species to frequently calibrated species can be determined. These
ratios can be used to calculate calibration factors for times when the calibration
data are less complete.
Final Calibration Factor Calculations
To process data, calibration factors can be taken from the HCCAL DATATRIEVE report
for explicit calibrations or extrapolated from calibration data for other species.
Calibrations may not have been performed the day of an experiment for all species.
Interpolation may be needed between the calibrations for a species done right before
and after the day in question.
The discussion to this point has been concerned with the processing of calibra-
tion data and using calibration data to estimate calibration factors for compounds
68
-------
HC Processing
Calibration
0.30 -
0.28 -
0.28 -
0.24 -
0.22 -
0.20 -
0.1B -
" -
0.14 -
0.12 -
0.10 -
o.oe -
0.06 -
LIGHT COMPOUNDS
Carlo m
0.7
0.8 -
0.3 -
0.4 -
0.3 -
0.2 -
0.1 -
a eml d«t»
20 40
RETENTION TIME
+ exp: r2 - 0.984
SO
HEAVY COMPOUNDS COLUMN
Ctrla
O eal
RETENTION TIME
o «xp: r2 » 0.98
Figure 11. Calextr Examples.
69
-------
Calibration HC Processing
not explicitly calibrated. Calibration factors need additional checking and perhaps
adjustment sometimes before they are applied to process data. These checking pro-
cedures are actually QA steps performed in experimental data processing but are
definitely part of calibration data processing.
Once calibration factors are calculated, they are applied to the raw experimental
data. The resulting concentration data should fit with the known target conditions,
the amounts injected, and the known resulting conditions of other similar experi-
ments. The data should also yield the same composition as determined by analytical
data for other similar experiments. This should occur with calibration factors used
which are at least similar to those used to process experiments conducted before
and after the experiment in question (see discussion of RUN_CAL.USED). Ratios of ini-
tial conditions should be somewhat consistent with integrator areas when peaks are
reasonably well defined and integrateable. The calibrated integrator area should be
consistent with peak height derived concentration data for large peaks with good
resolution and clean base lines.
70
-------
Experimental Data
This chapter presents the procedures, tools, and techniques for processing the ex-
perimental data. Sections will discuss documentation, data processing tracking
systems, and how data are actually processed.
Experimental Results Documentation
Documentation of experimental conditions and results is performed at every level of
processing to assure that no data or other useful information is lost and that all data
can be processed. Methods have been developed to allow for redundant documen-
tation and to provide the smog chamber operators several different opportunities
to document the experiment.
Data
Data are documented several ways reflecting the different types of data acquisition
techniques. Data is acquired and recorded by both computer and stripchart. Print-
outs are also produced by the computer during and after an experiment and provide
additional documentation.
Stripcharts
Almost every source of data is recorded on stripchart, providing a backup for the
computer DA system in case of failure. For instruments that produce data not
suitable for the computer DA system, stripchart recorders are the main data acqui-
sition device. All stripcharts are documented with rundate, instrument, time, side,
71
-------
Experimental Data . Data
attenuation, chart speed, full scale deflection, and calibration information. Docu-
mentation of stripcharts is requested by the setup checklist and performed by the
setup operator. Changes of instrument settings or conditions are noted directly
on the stripchart during the experiment. Time and side is recorded several times
throughout the experiment, as requested by all of the checklists discussed above. A
pack-up inventory checklist assures that all stripcharts are collected and stored in
the runfolder. Stripchart data is obtained for the following instruments:
Os monitor
NOX monitor
Carle 1 GC-FID for HC
Carle 2 GC-FID for HC :
Carle 3 GC-FID for HC
Formaldehyde monitor
Dewpoint monitor
Varian GC (PAN, tracers, nitrates, biacetyl)
PE900GC
TSR (total solar radiation)/UV sensors
Printouts
Printouts produced during and after an experiment also document the experimental
data. These are retained in each runfolder:
PRINT PROFILEcomputer data dump during run
Experiment COMMAND FILE listinglist of scheduled injections and chamber con-
trol commands
Experiment BAKTRN FILE listinglist of actual injections and chamber control
commands (both scheduled and manually activated)
Operator's REMARKS FILE- comment file
Computer Data
There are two types of computer data generated by this data processing system:
computer acquired data during the experiment and files produced from stripchart
recorded data after the experiment in data processing. Documentation for computer
data exists within the data files produced themselves. Time, instrument, side, and
any intermediate conversion factors used to store the data is recorded automatically.
The remarks file contains comments entered by the site operators before, during,
72
-------
Forms Experimental Data
and after the experiment directly into the computer, which are automatically doc-
umented with time entered. The data files produced from stripchart data after the
experiment contain all of the documentation recorded on the stripcharts and the
name of the people involved in processing the data.
Forms
Several paper forms are used to help document the experimental conditions and
resulting data. Several of these are filled out at the smog chamber site as the
experiment is set up and executed.
Run Form
The run form is a major documenting tool. It is used to record all the information
needed to determine what experiment was attempted, what the chamber conditions
were, time the experiment starts and ends, and important notes about the exper-
iment actually conducted. The form is partially filled out before the experiment
to officially record the desired target conditions including sources and amounts of
chemical species. The morning operator takes instructions from this form to initiate
the experiment, and records important comments about chamber conditions, devi-
ations from target conditions, and any problems. This form is shown in Figure 12.
Instrument Checklist Form
The instrument checklist is filled out the morning of the experiment and is attached
to the run form. It is used to indicate which instruments are operational and
were used, any calibrations performed, and any problems with the instruments
themselves. This form is shown in Figure 13.
Hun Folder Inventory Checksheet Form
This form is used to inventory the run folders contents when the runfolder is first
logged into the data processing system. It is a primary QA step used in conjunc-
tion with the instrument checklist form to determine what data exists, including
documentation. Login and filling out of this form occurs before anyone is allowed
to work with the contents. This form is shown in Figure 14.
Data Processing Instruction Form
The data processing instruction form is used to collect and convey important in-
formation about the data collected on stripcharts during an experiment. It is an
intermidiate documenting tool used to indicate start and stop times of valid data,
calibration data (not to be confused and picked as run data), species to pick and
73
-------
Experimental Data
Forms
Run Date
Operator Horning _
Experiment Number
Command File(s)
RUN SHEET
Project
Operator Set-up
Operator Afternoon
from Table
Tank File Date
Run Description and Purpose
Special Tasks, Modifications
ACTUAL INJECTION CONDITIONS AND ESTIMATED INITIAL CONCENTRATIONS
Dilution Tracers ID Blue, ul Red, ul
Blue
Red
Species ID stat Concen. secs,ul,g dynara stat Concen. sec,ul,g dynan
dynam Concen. ul,liter ml/min dynam Concen. ul,liter ml/m;
N0_
N02
Total.HC_
Total.NOx
Drying Performed? Red
Dew Point(sunrise) Red
Delta Time
.F mv Time..
Blue Delta Time .
__ Blue F mv Time
Horning Chamber Condensation Observed ? (Y/N) Red: inside
Blue: inside
Vent Doors Checked (initials): Horning
CHARTS LABEL (initials): Morning
Special Problems, Concerns, or Comments
End of run
End of run
outside
outside
DVM Data Recorded (TIMES) to ; . Experiment starts (TIME)
Experimental Conditions established at (TIME)
Run rating (circle one): useless
Figure 12. Runsheet
borderline fair good great unknc
74
-------
forms Experimental Data
Instrument Checklist Run Date
Data
Not Acqui- Note
Instrument Cal Cal Opera- Opera- Not sition
Today TIME(S) tional tional Used QA
(y/n) (check) (check) (check) (init.) (num)
(type)
ATC
Carle I
Carle II
Carle III
CEA Formaldehyde
CO meter
Dew Point
DNPB Aldehydes
Nitric Acid
NOx Bendix
Ozone Bendix
PE 900 (chamber)
Saltzman
Temp
TSR
UV
Varian aerosol
Varian gas
PAN
Nitrates
Tracer
Other
Other
Other
Other
* Notes
Figure 13. Instrument Checklist Form
75
-------
Experimental Data
Forms
RUN FOLDER INVENTOR* CHECKSHEET INIT:
PROJECT: AUTO DODG REAC WOOD DATA: AERO
PRESENT ITEM COMMENTS
RUN
DATE: ._
GAS
WOOD
Run sheet
. Instrument Checklist
. ATC stripchart*
, Cade I stripchart*
Carle II stripchart*
Carle III stripchart*
CO meter stripchart*
Dewpoint stripchart
FORM stripchart*
NO stripchart
N02 stripchart
03 stripchart
PE900 stripchart*
PRINT PRO printout
.R CMANDR printout
.R DATRAN printout
.R.MERGE printout
Sigma 10 printout
TSR stripchart
DV stripchart
Varian stripchart*
DNPH aid stripchart*
OTHER:
OTHER:
OTHER:
OTHER:
Figure 14. Run Folder Inventory Checksheet Form
76
-------
Plots Experimental Data
problems in general with the data. Calibration factors and any calculations or notes
are indicated on this form. The information on this form eventually ends up in the
final segmented file used for distribution. This form is shown in Figure 15.
DVMFIX Data Sheet Form
This form is similar to the Data Processing Instruction Form used with stripchart
data. It is used to record and convey important information required to properly
process the data recorded with the DVM computer acquisition system. Calibration
factors, periods of valid data, and general documentation are indicated here. The
information on this form eventually ends up in the final segmented file used for
distribution. This form is shown in Appendix E.
Plots
Plots are a form of documentation as well as a method of presenting data. They
indicate when calibrations were performed, when injections were made, and when
the chamber is vented. They often indicate power failures, computer failures, or in
general problems, and general quality of data.
Raw Data Plots ("P"- and "U"-file Plots)
These are plots of raw uncorrected and at first unQAed data. "P"-file plots show raw
"picked" (digitized data). "U" file plots show the raw DVM data, often including the
AUTOCAL calibrations and venting. Both types of plots indicate completeness of data.
ratios of conditions, profiles and general reasonableness of data. Discontinuities
may indicate problems during picking.
Processed Data (G- and C-files QA plots)
These plots of the final corrected data show immediately any problems or errors
which might have occured in processing. Only valid experimental data should have
been processed. Concentrations should be close to target conditions. Light, dew-
point, and temperature data should be reasonable in magnitude.
Tracer Data Plots (CSTAR)
Inert tracers are usually injected into each chamber half to monitor dilution during
the experiment. Raw data are read and processed by a computer program CSTAR
which fits the data to an exponential function with the least-squares method. The
slope is the dilution rate which should be used by modelers. This data is processed
for every run; chamber leaks can be detected quickly.
77
-------
Experimental Data Plots
DATA PROCESSING INSTRUCTION FORM PS of
RUN DATE INSTRUMENT
RUN HCANAL <*/n> ON FILES BETUEEN : AND
COMPOUND SIDE CAL FAC X ATTN CAL FACTOR CALCULATIONS/DOC INITIAL/
(AND UNITS) PATE
Cal Source Reports Used - Date _ ...
Cal Source Reports Used Date
Cal Source Reports Used __ Date __
RECORD ANY SPECIAL NOTES OR CALCULATIONS ON BACK OF THIS FORM PLEASE !
Figure 15. Data Processing Instruction Form
78
-------
Run Data Processing Status Databases Experimental Data
Run Data Processing Status Databases
Description of Lab/Site Calenders
Calendars have been discussed above. They are used as a primary documenting
method. Experiments, calibrations, and some instrument or chamber work are
indicated on the calendars at the site and the data processing office first. These are
used to guide the data processing effort.
Description ofRUNEHTRY database
A DATATRIEVE database maintained on the VAX is used to track all data processing
and record the experimental conditions. The processing status of a single exper-
iment or a group of experiments can be quickly assessed. It is used with report
procedures to guide processing tasks. Tasks are listed by the priority given by a
project coordinator.
The RUNEi.'TRY system is a PL/I program that runs on the VAX-n/780 computer
in the SPH. It provides on-line, interactive access to the database of run processing
status. The program uses the VAX "electronic" Forms Management System to sup-
port a number of "forms" on a video display unit (terminal). The user can fill out
the form, and the data are either called up from the database, or are stored in the
database for future reference.
Table 14 is the individual record description for an entry in the database. In
addition to the general information about the run conditions, the status of the
processing steps for digital voltmeter data, for documentation preparation, for the
processing of each instrument that had valid data, and for the progress toward
creating the final file are stored in a record for each run.
t
There are a possible total of 110 checkpoint steps in processing a run. The
events that are entered will be listed in sections following. The record provides for
nine instruments, accounting for most of the checkpoints.
Single page printouts (132 columns wide) can be created for each run. An ex-
ample of a printout is shown in Table 15. This shows a nearly completely processed
run that had five instruments.
Figure 16, Figure 17, Figure 18, and Figure 19 are "hardcopies" of the terminal
screen showing the "electronic" forms. These are designed to be used on terminal
that have video attributes such as reverse fields (shown in the figures as dark areas),
bold letters (not shown in figures), blinking areas and letters for attention. Each
79
-------
Experimental Data Run Data Processing Status Databases
form also has on-line HELP for each field, and if more help is needed, the screens
have full-screen help forms with such information as legal key values and HC species
numbers.
Several report procedures have been developed to help guide the data processing.
These look up the status in RUMEHTRY for specific tasks for each "type" of personnel:
PCREPORTS. PPREPORTS. CTREPORTS. These were explained in Chapter 2. An example of
the PCREPORT is shown in Figure 20.
80
-------
Run Data Processing Status Databases Experimental Data
Table 14. Record Description for RUNENTRY Database System.
01 RDN_REC.
02 GENERAL.
03 RDN_DATE USAGE IS DATE
EDIT_STRING IS DD-MMM-YYYY.
03 CONDITIONS OCCURS 2 TIMES.
06 SIDE PIC X(4) .
06 NOXCONC PIC 9.999.
06 HC OCCURS 3 TIMES.
09 HCCONC PIC 99.99
MISSING VALUE IS 0.00
EDIT.STRING IS Z9.00?" ".
09 SPECIES PIC 999.
03 PROJECT PIC X(4) .
03 QUALITY PIC 9.
03 PRIORITY PIC 9.
02 DETAIL.
03 DVMDATA.
06 ITEM OCCURS 10 TIMES.
09 DVEVENT PIC X(3)
DEFAULT IS "non"
VALID IP DVEVENT IN DVEVENT_TAB.
09 DVEDATE USAGE IS DATE
EDIT_STRING IS DD-MMM-YY.
09 DVINIT PIC X(3).
03 DOCUMENTATION.
06 ITEM OCCURS 5 TIMES.
09 DOEVENT PIC X(3)
DEFAULT IS "non"
VALID IF DOEVENT IN DOEVENT_TAB.
09 DOEDATE USAGE IS DATE
EDIT_STRING IS DD-MMM-YY.
09 DOINIT PIC X(3) .
03 FINAL_FILE.
06 ITEM OCCURS 5 TIMES.
09 FFEVENT PIC X(3)
DEFAULT IS "non"
VALID IF FFEVENT IN FFEVENT_TAB.
09 FFDATE USAGE IS DATE
EDIT_STRING IS DD-MMM-YY.
09 FFINIT PIC X(3) .
03 PICKDATA.
06 TOTAL.SPECIES PIC 99.
06 MUM PIC 9
MISSING VALUE IS 0.
06 INST OCCURS 0 TO 9 TIMES DEPENDING ON NUM.
09 ID PIC X(3).
09 ITEM OCCURS 10 TIMES.
11 PIEVENT PIC X(3)
DEFAULT IS "non"
VALID IF PIEVENT IN PIEVENTJTAB.
11 PIDATE USAGE IS DATE
EDIT_STRING IS DD-MMM-YY.
11 PIINIT PIC X(3).
81
-------
Table 15.
7)
3
ft
n
»
a
D
v
IIUNENTRY Processing Summary for September 14, 1081
Run Detei U-SEP-
Priorityi t
HCCONC SPECIE
Z.72 191
««.7i i/a
ii.ii tni
HAHE
Uncut*
Ii2,4-trInethylbonienc
-COMPLETED OVM EVENTS-
oo
to
URU Unpick pan run IZ-OCT-I98I MOS
V2T At tip* era.ted MOS
UPC Riw Plot cr««td RFP
PDA Plot OA conpltd KCS
CfS C.I f.ct deter* 29-OEC-I98I KGS
0«U OVHriX pg* run 7-JAN-I98J Kit
COA Cone OA conpltd'- 7-JAN-I983 KCS
R2M Reldy to nerge J-JAH-I98J HEJ
COMPLETED INSTRUMENT EVENTS
lnitru*enti CIG Ctrlo I CC
PID Pick Initrc d.t I8-MAY-I9B2 KCS
F«U PICK pg* run 27-OCT-I982 Jill
rPT Flit plot creet 27-OCT-I9B2 JIH
POA fit pit OA coup 2H-OCT-I982 KCS
CfO C.I feci deter* 2S-HAY-I9B2 KGS
CRU CAICOII pgn run 27-OCT-I982 JLH
HRU HCANAL pg* run 27-OCT-I982 JIH
COA Coop Chi.hi OA 27-OCT-I98Z KCS
f2V Flltl tr to VAX 7-JAH-I983 HFP
V2U »«.dy to ».r9. S-JAH-1983 H£J
COHPIETED INSTRUMENT EVENTS
Initrtoanti V2C V.rl.n GC Igxl
PID Pick Initrc det IS-MAY-1982 KGS
Ml) PICK P9« run 7-APR-I982 JIH
FPT f.il plot cre.t 7-APH-I982 JIH
POA Fit pit OA coup KCS
CFD C.I f.c. deter* 2I-HAV-I982 KCS
CRU CAICON P9« run I6-OCI-I96Z JIH
COA Coup Chklhl OA- 28-OCI-I962 KCS
F2V rilei tr to VAX 7-JAH-I983 RFP
VZU needy to Berge S-JAH-1983 HEJ
COMPLETED DOC EVENTS
SUM Run Siinry conpl 29-OCf-l382 KGS
D2V Doe tr.n to VAX 7-JAN-I981 KGS
V2T AL tope ereited 9-JAN-I963 Off
T2D Tape trn to dik 9-JAH-I903 IIEJ
DOA Doe OA eonpletd 9-JAN-I983 HEJ
COHPIETED INSTRUMENT EVENTS
Inftruxenti C2G C.rle II CC
PID Pick Initrc det I8-HAY-I982 KGS
PRU TICK pg» run S-APR-1982 JLH
FPT F.It plot cre.t S-APR-1982 JIH
POA Fit pit OA comp 2B-OCT-I982 KCS
CFD C«l f*c> deter* 2S-MAY-I982 KGS
CRU CAICON P9« run e-SEP-1982 COM
HRU HCANAl pg* run 8-SEP-I982 CDH
COA Comp Cnrlht OA 28-OCT-I992 KGS
F2V Fllei tr to VAX 7-JAH-I983 RFP
V2U Reidy to Kerge 9-JAN-I9B3 IIEJ
COHPIETED INSTRUMENT EVENTS
Initrunenti COG Beckn.n C8B0 ICOI
IAD Stop procenlng I8-MAY-I982 KGS
COMPLETED FINAL FILE EVENTS
MR) 3 lectl eierged e.j/\N-l9B3 HEJ
U2T NL tipe created K-JAN-I9B] HEJ
COMPLETED INSTRUMENT EVENTS
Initruvienti FOG CEA Form. Idchyie
PIO Pick Initrc det II-HAY-I9B2 KCS
PRU PICK pg* run
FPT Flit plot cre.t
PpA Fit pit OA conp
CFD C.I Feel doter*
CRU CAICOII pg* run
COA Coup Chkiht OA
F2V File! tr to VAX
V2U R»dy to *erge
7-APR-I982 JIH
7-APR-I982 JLH
KGS
l-JUl-1982 KCS
I-SEP-I9B2 KCS
28-OCI-I982 KCS
7-JAN-I9B3 RFP
(-JAN-I9B3 HEJ
COMPLETED INSTRUMENT EVENTJ-
"0
i
o
(n
c»
V
c
-------
Run Data Processing Status Databases Experimental Data
SMOG CHAMBER
RUN TRACIN6 PROGRAM
ENTER NAME: Harvey
ENTER DATA YEAR: 85
MAIN MENU
1 search -For a RUN using the run date
2 display run record
3 print run record
4 update information (but not run date)
5 delete records
ft create run records
7 change run date of existing run
B print -from list file
9 ex i t
ENTER CHOICE (1-9): AND PRESS
Press PF2 for Help.
Figure 16. Runentry Welcome and Main Menu Screens
83
-------
Experimental Data Run Data Processing Status Databases
SELECT A RUN
ENTER THE DESIRED RUN DATE: 05-aug-19a4 AND PRESS (RETURN).
(A run did not hive to occur on this date, but it helps.)
If the fdialing Run.Date is not the sate as the one entered,
then a run Kith the requested date does not exist.
Run.Date 5-AUS-1984 Project NETH
Red Hydrocarbon Blue Hydrocarbon
SYNTHETIC AUTO EIHAUST SYNTHETIC AUTO EIHAUST
ARQHATIC HU AROMATIC HIS
FORMALDEHYDE FORMALDEHYDE
IS THIS THE CORRECT RUN ? .y
(Y) yes, establish as current run.date
(L) no, leave search and return to tain tenu
SELECTION OF DISPLAY
1 shot GENERAL run intonation
2 sho» INSTRUMENT run information
3 Shan DVN run intonation
4 shod DOCUMENTATION and FINAL FILE intonation
5 return to ijin lenu
ENTER CHOICE (1,2,3,4, or 5): AND HIT (RETURN)
Figure 17. RUNENTRY Date and Display Selection Screens
84
-------
Run Data Processing Status Databases
Experimental Data
KEY
0 V N DATA STATUS
Run Date: 5-AU6-1994
EVENT
DATE
INITIALS
URU
ACS
RFC
U2V
PQA
V2T
CFS
DRU
CQA
nan
Unpack pga run
Auto cal stripd
Ran Plot creatd
Data trn to VAX
Plot SA coapltd
Van to tape
Cal facs detera
DVI1FIX pgi run
Cone SA coapltd
Not yet coapltd
9-AU6-1984
9-AUB-1984
9-AU6-1984
9-AUB-1984
4-SEP-1984
9-AUB-1984
30-HAY-1985
30-HAY-1985
30-MAY-1985
JKJ
JKJ
JKJ
JKJ
K6S
JKJ
KGS
JLH
KBS
press RETURN-key to continue
PICKED INSTRUMENT STATUS
Run Date: 5-AU6-1984
Instruoent selected: C2B
INSTRUMENT ID AND DESCRIPTION KEY
EVENT
DATE INITIALS
AT6
C16
C2B
C3S
FOB
V26
ATC electron ca
Carle I BC
Carle II BC
Carle III GC
CEA Foraaldehyd
Varian 2 BC (ga
PID
PRU
FPT
PQA
CFD
CRU
non
non
non
non
Pick instrc det
PICK pga run
Fast plot creat
Fst pit QA coap
Cal facs detera
CALCON pga run
Not yet coapltd
Not yet coapltd
Not yet coapltd
Not yet coapltd
20-AUB-1984
20-AUB-1984
20-AUB-1984
5-JUN-1985
5-JUN-19B5
5-JUN-1985
KGS
JLH
JLH
KBS
KBS
JLH
Enter Y for another instrument, ENTER-key to continue
Figure 18. RUNENTRY General, Documentation, and File Status Screens
85
-------
Experimental Data
Run Data Processing Status Databases
GENERAL INFORMATION
RUN DATE 3-AUB-1984 PROJECT HETH QUALITY 9 PRIORITY 9
SIDE RED NO! 0.350 HC 00.41 SP 403 SYNTHETIC AUTO EIHAU
HC 00.59 SP 698 ARQHATIC nil'
HC 00.02 SP 139 FQRHALDEHYDE
SIDE BLUE NOX 0.350 HC 00.27 SP 603 SYNTHETIC AUTO EIHAU
HC 00.33 SP 498 AROMATIC I1II
HC 00.01 SP 139 FORMALDEHYDE
SUGARY OF PROCESSING STATUS
DVHDATA 9 PICKED DATA 4 of & DOCUMENTATION 0 FINAL FILE 0
press RETURN-key to continue
DOCUMENTATION AND FINAL FILE INFORMATION
Run Date: 5-AU6-1984
DOCUMENTATION:
KEY
EVENT
DATE INITIALS
non Not yet coipltd
non Not yet coipltd
non Not yet coipltd
non Not yet coipltd
non Not yet coipltd
FINAL FILE: KEY
press RETURN-key to continue
EVENT
nan Not yet coipltd
non Not yet coipltd
nan Not yet caipltd
non Not yet coipltd
non Nat yet coipltd
DATE INITIALS
Figure 19. RUNENTRY DVM and Picked Instrument Status Screens
86
-------
Run Data Processing Status Databases Experimental Data
Figure 20. PCreport Example
Count of runs that have reached stated stages of processing
for ALLRUN84
Total runs Total Inst
63 312
id******* DVM data ************
URU RFC ACS MAC U2V V2T PQA CFS DRU CQA R2M NTB BAD UDV
58 58 58 0 57 57 54 38 37 37 0 2 1 3
Sum of URU, NTB. and NDV is 63
######## Inst data ############
PID PRU FPT PQA CFD CRU HRU CQA F2V V2U NTB BAD
190 190 190 116 104 104 0 17 0 0 135 6
Sum of PID and NTB is 325
######## Documentation ############
SUM D2V DQA
772
######## Final File ############
MRS ALT FQA
000
Number of P-files should be : 190
Number of C-files should be : 104
Number of K-files should be : 104
87
-------
Experimental Data Run Data Processing Status Databases
Figure 20(cont.). PCreport Example
ALLR runs/instuments needing Picking Instructions Determined (PID).
This is an HIST first processing stage report.
PRIORITY >» 8 <«
26-Jun-84 Humber of inst = 8
PCG PE 900 GC
(chamber)
PRIORITY >» 5 <«
25-Jun-84 Number of inst = 8
PCG PE 900 GC
(chamber)
ALLR runs/instuments needing Cal Factors Determined (CFD).
The following instruments have ( PID ) Pick inst det completed
but not ( CFD ) Cal fac det completed.
RU!!
DATE Instruments Date done Person
PRIORITY >» 9 <«
25-Jul-84 Number of inst = 5
V2G Varian GC (gas)
PID Pick inst det 6-Aug-84 KGS
PRIORITY »> 5 <«
25-Jun-84 Number of inst = 8
COG Beckman 6800 (CO)
PID Pick inst det 31-Jul-84 JA
V2G Varian GC (gas)
PID Pick inst det 31-Jul-84 JA
88
-------
Use of Data Processing Status Database Experimental Data
Description of Data Processing Status Database
This section describes the data processing status database and how it is used.
Description
The main parts of the data processing status database are the calendars and the
RUNENTRY DATATRIEVE system. RUUEHTRY maintains status information for each experi-
ment by date. The general categories are:
GENERAL
o experimental chemical conditions for both sides,
o quality and priority index,
o number of experiments
INSTRUMENT
o pick instructions,
o cal factors given,
o qa performed,
o final processing steps performed;
. DVM
o unpacking performed,
o cal factors given,
o QA steps performed,
o final processing performed;
DOCUMENTATION
o final documentation processed and written,
o moved to VAX for final segfile formation.
Use of Data Processing Status Database
This section describes how the database is used from the time the experiment is
conducted to final processing.
Log Experiment
When an experiment is conducted (attempted), the conditions and general qual-
ity are indicated on both calendars (site and office). This information is usually
obtained by telephone conversations between the project coordinators and site per-
89
-------
Experimental Data Use of Data Processing Status Database
sonnel within a few hours after the start of the experiment in order to plan for
the next days activities. Another late afternoon telephone conversation follows to
determine the outcome of the experiment and to finalize the next, days plans. If the
experiment is thought to be successful or might provide any the useful information,
it is considered to exist. The absence of major negative comments such as "failure"
written on the calendar, triggers the data processing procedure. The experiment
is packed-up and inventoried if it is not considered a "failure" for transport to the
dataprocessing office.
From the-telephone conversation mentioned above, the experiments success is
determined. If the experiment is considered worth archiving, the actual experi-
ment conditions and general quality are recorded on the calendars. Otherwise the
attempted experiment and the problems associated are noted on the calendar.
Data processing personel expect both the runfolder and data floppies to be
transported to the school within a few days after the experiment. Special locations
in the office are set aside for incoming data. The calendars and these areas are
checked daily. If data is late, the site personel are contacted and questioned. When
either the run folder or data floppies arrive at the office, the experiment is entered
into the RUHEHTRY database.
The contents of the runfolder are first inventoried before any data processing
occurs. The instrument checklist completed during the experiment is compared
with the folder contents. Missing items are investigated.
Preliminary Review
Preliminary review usually begins the day or day after an experiment is conducted
by a PC. Conversations with the site operators indicate resulting experimental condi-
tions and outcome and any problems which might have occured. The run folder con-
tents are inspected. Instrument operation behavior, chamber background(proper
venting, drying or absence of condensation), temperature and dewpoint, sunlight,
initial conditions, and reasonableness of results are determined. Basic quality and
processing priority is determined. Completeness for data and documentation is de-
termined. The calendars and the RUUEHTRY database are updated. Besides a QA
step for the experiment itself, the success of meeting the experimental plan goals is
checked. A variation in the experiment might be useful if documented but would
result in rescheduling of the target experiment.
Processing Instructions and Information
RUHENTRY is used to track data processing by recording of processing events. Auto-
matic status reports can be generated. Report procedures are used to guide and
90
-------
Digital Voltmeter Data Processing Experimental Data
assign processing tasks: PCREPORTS, PPREPORTS. and CTREPORTS. The reports are listed
by priority (0-9) as determined by the PC.
PCREPORTS tells the project coordinators that picking instructions are needed,
or picked data exists and needs QA and calibration factors, or final data exists
and needs QA. CTREPORTS indicates to the computer technicians that DVM data needs
unpacking and archival, and AUTOCAL data needs to be stripped. PPREPORTS tells the
peakpickers that pick instructions have been given, or that picked data has been
QAed and calibration factors exist for final processing.
Each worker updates RUlJEHTRY as each step is performed. The presence of each
new step completed generates instructions to the next appropriate worker to perform
the next processing task when a new report is generated. When all processing steps
are completed, the segmented file is assembled, and the final initial experimental
conditions and quality are updated in the RUNENTRY GENERAL section. This data
is input to the RUNSUMMARY database which is used to sort, organize, and analyze the
total smog chamber experimental database.
Digital Voltmeter Data Processing
The Digital Voltmeter Data (DVM) from the Computer DA system is usually processed
easily once the calibration database is established. The processing flowchart is
indicated in Figure 21. The RUtiEHTRY events flowchart is shown in Table 16.
The computer technician (CT) "unpacks" the data from the two floppies which
arrive from the research site. Several files generated by the site DA system are copied
to the LSI-11 harddisk: the main DVM file (mmddy.DVM), the remarks file (mmddy.REM),
the command file (mmddy.CMD), the adjustments file (mmddy.ADJ), and the file of actual
computer commands (mmddy.OPS). Printouts of all of these files except the mmddy.DVM
file are made and stored in the RUNFOLDER. A computer program (UNPACK) is run which
produces a "U" file, which is the same raw data but in ASCII format and is the
principle raw data file for the smog chamber experiment. This file contains 24 hours
of data taken every minute for every instrument connected to the DVM. The "U" file
is back-up onto two sets of floppies, and to two types of magnetic tape used on
different computers: the VAX-780 and the University's IBM 4381.
Several programs are used to process data from the "U-file". FASTDV is run
several times: to produce "P" files for plotting the raw data ("fastplots") from the
experiment and to strip the AUTOCAL data (makes "A" file). The processing of the
calibration data was discussed earlier. PLOPIC is used to make the plots from the
"P" files for the NOX/O3 plot and the TSR/UV, temperature, and dewpoint plot.
Theses plots are stored both in the RUUFOLDER and the FASTDV notebook. All of the
91
-------
Experimental Data
Digital Voltmeter Data Processing
mmddy.DVM, REM,
SPG, CMD
UNPACK
i
DVMFKf-
mmddyU.DVM
FASTDV-
CAL
FACTORS
e^LSIlQ
DISK I
PLOCONf
I
"OFFICIAL'
Figure 21. DVM Processing Flowchart (Computers and Files)
92
-------
Digital Voltmeter Data Processing
Experimental Data
Table 16.
Processing System for DVM Data
cal processing
/ \
tot runs (CFS)
\ / \
V (ACS) \
\ / \
(URU) (RFC) !PQA!---!CQA!--!R2M!
\ /
(UTV) /
\ /
(V2T)
!sss! == stages determined by Project coordinator (PC)
== stages determined by Peak Pickers (PP)
(sss) == stages determined by Computer Techs (CT)
Pers
DVM data processing steps
Step Meaning
CT
CT
CT
CT
PC
CT
PC
CT
PC
CT
PC
PC
CT
URU
U2V
V2T
RFC
PQA
ACS
CFS
DRU
CQA
R2M
NTB
BAD
NDV
Unpack run
Data to VAX
to archive tape
Raw Plot done
Plot QA
AutoCals strip
Cals done
DVMFIX run
Cone QA
Ready 2 merge
Not to be proc
Stop processing
No DVM data
Site floppy data expanded to ASCII Ufiles
Ufiles moved from LSI disk to VAX
Data moved from VAX to tape
Plot of voltages to examine data
Initial quality check
Cals separated for processing
Calibration factors determined
Voltages changed to concentrations
Concentration Quality Assurance
Merge process on VAX can be run
Not to be processed
Something is wrong with this data
93
-------
Experimental Data Digital Voltmeter Data Processing
tasks discussed so far are done by the CT. The CT update RUNENTRY as these tasks are
performed.
When the FASTDV plots are made and RUIJEIITRY has been updated so, the PCREPORT
will direct the PC to perform a QA on the raw data. The plot is compared with
the original stripcharts. If data is missing, the documentation (RUNSHEETS,
REMARKS FILES, INSTRUMENT CHECKLIST, CALENDARS) is investigated
for mention of power or computer failures, or instrument malfunction. The data
may need to be retransfered from the site. Missing data may have to digitized from
the stripcharts. Once the raw data is QAed, the PC updates RUliEMTRY that the raw
data has been unpacked successfully (PQA) and calibration factors can be applied
(calibration factors need to be determined).
Calibration factors are applied to raw DVM data with a program called DVMFIX
which runs on the LSI computer. The program also controls which time periods of
raw data are processed. Bad data, or cal data is eliminated in this way. Calibration
factors are obtained from a list resulting from calibration factor processing discussed
above. Interpolation may be required. Usually however NOX and Oz monitors are
calibrated twice a day, before and after each experiment. The calibration factors
can be used directly.
The PC fills out a DVMFIX sheet indicating the calibration factors to apply to the
data, the periods of valid data to process, and brief documentation. The calibration
factors are obtained from the AUTOCAL report as described above. The PC updates
RUNEliTRY that calibration factors have been determined. The information on the
DVMFIX sheet is transfered to a file on the LSI-ll computer (mmddy.CRD) usually by
a PP or CT. The "U" file is moved back to the harddisk. This might require the
assistance of the CT if these files are not still on floppies and need to be copied off
of one of the magnetic tapes. The program DVMFIX is used to apply the calibration
data in the mmddy.CRD file to the raw data, producing a new file of corrected data for
only valid experimental data for every fourth minute (mmddyG.DVM "G-file"). DVMFIX is
usally run by the person who creates the mmddy. CRD file. The "G-files" are backed-up
on floppies. RUNEMTRY is updated that DVMFIX has been run and final QA is need by the
PC.
The final corrected data is listed and plotted. Concentration data for NO and
NC>2 must equal NOX concentrations within a few ppb throughout the experiment.
"Zeros" are checked. Initial conditions are compared with the target conditions.
Dewpoint, and temperature data are checked for consistency and reasonableness
(dewpoint data cannot be higher than chamber temperature). Profiles and maxi-
mum values of light are compared with other seasons for resonableness. Data not
to be processed or missing should have "-9.9999" indicated. If there are problems,
the PC will re-issue processing instructions, deleting from RUNEIITRY that DVMFIX has
94
-------
Strip Chart Data Processing Experimental Data
been run. This will cause the PPREPORTS and CTREPQRTS to request that the data be
reprocessed. The process is repeated until the QA criteria are satisfied.
Strip Chart Data Processing
Processing strip chart data is a little more complex than processing DVM data. More
data is represented, more calibration factors are required, and there are more sources
of error. Basically the data are identified, the conditions of the instruments are
identified, the data is picked (digitized), plotted, and QAed, calibration factors are
determined and applied, and the final data is listed and plotted for final QA. The
processing flowchart is indicated in Figure 22. The RUMEKTRY events flowchart is
shown in Table 17.
Data processing starts with RUNENTRY indicating to the PC in the PCREPORT that pick
instructions are needed for the instruments used in an experiment. Data processing
starts with the documentation. The RUHSHEET and RUNENTRY indicate which species
where injected for the experiment. The instruments which require processing are
identified. The stripcharts are investigated for proper documentation: attenuation,
chart full scale setting and chart speed, time, and chamber side sampled. The
chromatograms are inspected for proper appearance (operation). The background
chromatograms are inspected for contamination. Major problems are noted. If the
data is valid but not useful the t.'TB is indicated in RUllEliTRY. If the data is bad, BAD
is indicated. No further action will be taken.
If the instrument seems to be free of problems and data are considered useful,
then pick instructions are given. Calibration data are identified and noted as not
to be processed by the peakpicker (PP). The compounds monitored and their peaks
are identified and labeled. The entire stripchart is inspected for reasonable appear-
ance and operation and data. Problems are identified and noted. The peakpicker
can indicate in the data that data are missing (off-scale, power failure, instrument
adjustment, etc.). The start and end times of valid data are identified. The com-
pounds to be picked are listed on the DATA PROCESSING INSTRUCTION FORM with any
additional instructions. RUNENTRY is updated with PID (pick instructions determined)
by the PC. This indicates in the PPCREPORT to the PP that data is ready to be picked.
The PP obtains the run folder and determines from the DATA PROCESSING INSTRUC-
TION FORM which compounds are to be picked. The stripchart is removed and spread
out over the digitizer pad. Using the instructions and peak identification on the
stripchart the data is digitized using the program DIGPIK run on the LSl-ll computer
to produce a file named mmddyP.iit ("P" file). The files are backed-up on floppies
and magnetic tape. The data is then plotted with a program called PLOPIC on the
LSI-ll with GIGI graphics. A hardcopy is stored in the RUtJFOLDER. The PP updates
95
-------
Experimental Data
Strip Chart Data Processing
DP
inst
t
RUNFqLDER
stripchart
PP
Inat
DIGPIK
mmddyP.iit
HCANAL
or
LOTUS 123
mmddyC.iit
mmddyK.iit
PLOCON
Figure 22. Instrument Processing Flowchart (Computers and Files)
96
-------
Strip Chart Data Processing
Experimental Data
Table 17.
Processing System for Instrument Data
! tlTB !
/
tot inst __ cal processing.
\ / \
!PID! !CFD!
__
\
\
\
!CQA! ---- (F2V) ---- (V2U)
\
---- --IPQA!
!sss! == stages determined by Project coordinator (PC)
== stages determined by Peak Pickers (PP)
(sss) == stages determined by Computer Techs (CT)
Pers
PC
PC
PC
RUHENTRY stages
Meaning
PID : Pick inst det
NTB : Not to be proc
CFD : Cal fac det
PP
PP
PC
PP
PP
PC
CT
CT
PRU :
FPT :
PQA :
CRU :
HRU :
CQA :
F2V :
V2U :
Pick run
FPlot done
FPlot QA
CALCOU run
HCAMAL run
Comp CS QA
File 2 VAX
Ready 2 merge
PC
BAD : Stop processing
Charts marked, Inst status sheet in folder
Ho futher processing marked on Inst Status
sheet
Cal factors for each comp entered on Inst
status sheet in folder
Data has been digitized and P-file exists
P-file has been plotted and plot is in folder
Fast plot has been marked OK (no bad data
points in P-file)
C and K files exist on floppies
HC analysis printout in run folder
Instrument QA completed on C and K files
Initial Conditions updated
C and K files moved to VAX
C files merged together, K files merged
together
Something is wrong with this data
97
-------
Experimental Data Strip Chart Data Processing
RUIIENTRY that the data has been picked (PRU: pick run) and a "fastplot" has been
made (FPT). This indicates in the PCREPORT to the PC that raw data has been picked
and requires QA.
The "fastplot" is inspected and compared with the original stripchart. Appro-
priate chemical behavior is noted. The stripchart data is compared to assure that
the data is picked correctly: time and side for chromatograms and amplitude of
peaks. When possible, picked data is inspected for consistency: similar compounds
on the different sides should show similar behavior, and the least reactive side should
match the picked data side indicating less or less reactive material. If additional or
correctional picking is required, the PC deletes the PRU and FPT entries, and changes
the date for the PID instructions to the current date. This will indicate to the PP that
picking is required. Comments on the DATA PROCESSING INSTRUCTION FORM will indicate
the new work required with notes added by the PC during the last QA procedure.
Once the correction PP work is performed, RUNENTRY will be updated with PRU again
which will indicate raw data QA procedures are needed again to the PC in the next
PCREPORT. If the data passes the QA tests, the PC enters PQA to RUNE1ITRY which will
indicate to the PC in future PCREPORTs that calibration factors are.needed for further
data processing.
Once the picked data is QAed, calibration factors need to be determined for
final data processing: conversion .of raw picked data ("P files") to concentration
data ("C" files). The techniques for determining appropriate calibration factors
was discussed above. The calibration factors are indicated on the DATA PROCESSING
INSTRUCTION FORM. The PC indicates that calibration factors have been determined by
entering into RUIIENTRY CFD (calibration factors determined). This will indicate to
the PP that raw data is ready to be converted to concentration data in the next.
PPREPORT.
A program (CALCOM) which runs on the LSI-11 computer is used to apply the
calibration factors to the raw data ("P"-files), and to document which factors were
used, who determined them, who is running the program and when, the full names
of the compounds (abbreviations are used to identify data fields), and the maxi-
mum amplitudes of the raw and final concentration data. Two separate files are
produced, the actual concentration data (mmddyC.lit, "C" file) and the documenta-
tion (mmddyK.lit, "K" file). The resulting files are listed, and the data is plotted for
final QA with PLOCOH, a plotting program using GIGI graphics and the LSI-H. These
are retained in the run folder. The PP indicates that the CALCOH program was run by
entering CRU. This will indicate to the PC that concentration files are ready for final
QA in the next PCREPORT.
The QA procedure for the final concentration data varies with the complexity
of the experiment and the chemical species. The first checks however are to look
98
-------
Documentation Processing Experimental Data
at the documentation file ("K" file) and the DATA PROCESSING INSTRUCTION FORM to be
sure that the desired calibration factor(s) was actually applied. The PLOCON plot is
inspected to determine if the resulting data is reasonable and consistent with the
other experimental data and the indicated target conditions on the RUN SHEET. If the
concentrations of the initial conditions are far from the target conditions, a reason-
able explanation must be found. Often the RUN SHEET, REMARKS FILE printout, or the
INSTRUMENT CHECKLIST will indicate problems which could explain the discrepancies.
Usually resulting data is not more than 25 percent from the target conditions.
If the concentration data is for a standard HC mixture, a compositional analysis
should conform well with the known composition. An option of the CALCON program
is to produce files which are readable by a computer program (HCAHAL) for com-
positional analysis. In some cases where several experiments are performed with
a similar mixture, the concentration data for the initial conditions for all the ex-
periments are entered into a LOTUS spreadsheet for relational composition analysis.
Calbration factors should result in data which when analyzed yields a consistent
composition if indeed a similar mixture was used throughout the program. QA pro-
cedures may indicate that the calibration factors are not reasonable or that the raw
data is invalid. See the discussion above for determination of calibration factors.
Invalid data may require special processing, or a decision that the data is not pro-
cessable (BAD). Once QA procedures are successfully performed. RUHEiiTRY is updated
with CQA (concentration data QA performed). The "C" and "K" files are backed-up
on floppies. If new factors are needed, the CRU is deleted and the CFD date is updated.
New factors are indicated with explanations on the DATA PROCESSING FORM.
Once the final QA procedures are completed, the "C" and "K" files are trans-
fered from the LSI-H to the VAX (RUNEHTRY F2V) to await merging with the DVM data
and the documentation, to form the final segmented data file for distribution. The
"P", "C", and "K" files are backed up on VAX magnetic tape. The RUM.CAL.USED
database is updated with the calibration factors used in data processing. Initial
conditions are updated in RUNENTRY in the general section and RUHSUMMARY.
Documentation Processing
Documentation is associated with all phases of the experiment; recorded on many
sources. Much is needed only to process the data. Some documentation data
however is processed to be included with the final data files. Calibration factor
documentation is processed as the data is processed ("K" file, see discussion above).
Some other documentation however is needed, which increases the value of the
resulting final data.
A general documentation file is produced in addition to the concentraion data
99
-------
Experimental Data General Documentation Form
files (mmddyy.DOC). It becomes the first part of the final segmented data file. Its
purpose is to convey all information with the data file itself, required to describe
the experiment, and data which might be useful to photochemical modelers. The
processing flowchart is indicated in Figure 23. The RUllEiJTRY events flowchart is
shown in Table 18.
18
Documentation Steps
DOCUMENTATIO!! data processing steps
Pers Step Meaning
PC SUM : Run sum done DVM, Instrument, and Raw Quality Combined.
CT D2V : Doc to VAX Documentation transferred to VAX
PC DQA : Doc QA Documentation Quality Assurance
General Documentation Form
A five page paper form called General Documentation is used throughout all stages of
the data processing. All documentation needed to be included in the final general
documentation file is recorded on this form as data are processed. In addition the
form includes checklists and formulas used to arrive at a general quality index that is
specified in the final documentation file. An example form is included in Appendix
E.
General Description and Purpose of Experiment
The first part of the general documentation file is the experimental run date which
is the primary identifier of an experiment. A general description of the experi-
ment follows. The purpose of the experiment is included if it is not clear from the
description. Results are sometimes added if a specific comparison was being made.
Initial Conditions
The initial conditions of the experiment are listed next. Concentrations for each
100
-------
Initial Conditions
Experimental Data
A
RUNFQLDER
'C'ffle
't
'G'file
EDIT
1
PLOTS
PLOTS
PLOCON CSTAR
mmddyy.DOC
Figure 23. Documentation Processing Flowchart (Computers and Files)
101
-------
Experimental Data Meteorological Conditions
species for each side are listed. These values are obtained from the "G" and "C"
files, usually at sunrise. If the initial conditions are from data taken after sunrise
then the data time is reported. Initial conditions are updated in RUHEliTRY in the
general section.
Meteorological Conditions
Next the general meteorologial conditions (solar radiation and temperature) are
indicated. Detailed data from IIOAA reports and TSR stripcharts are the basis for a
general ranking value computed for both before and afternoon.
TSR data compared against RDLJ data
Total solar radiation data is compared with the HOAA data collected at the nearby
RDU airport. The data profile for the experiment recorded at the site is inspected
for "holes". A 1 to 10 grading scale is used to rate the experiments solar radiaion.
The holes are counted and used to assess the grade value for both morning and
afternoon portions (the morning sun is thought to be more important).
Temperature compared against RDU airport data
The temperature at sunrise and at the maximum is compared with the RDU airport
data for consistency and listed.
Data Times, Data Exceptions, Special Problems and Concerns
The times that the initial conditions are established, sunrise time and when the ex-
periment ends are given. Data exception are listed. Special problems and concerns
are listed.
Quality Assessment
An overall quality assessment grading scale has been devised to help compare the
large number of experiments. The index ranges from 0 to 10 with 10 being a perfect
experiment. Although the assessment of the quality index is somewhat subjective,
many characteristics can be quantified relatively. There are five categories which
are each weighted equally.
Solar Radiation
For the earlier years of the dataset sun light was the principle factor used for rep-
resenting run quality. Starting in 1981 more formal methods were developed: a
formula which would grade the sunlight quality more consistently but with the
102
-------
Quality Assessment Experimental Data
same general techniques used earlier: a) the percent of possible solar radiation
measured for the day. b) the number "holes or dips" in the radiation profile, and
c) morning fog or haze. The earlier sunlight profile of the day is usually considered
more import than the post solar noon portion. Therefore the first half of the day is
weighted 70of the total quality index. Meteorological data obtained at a local major
airport (RDU) lists percent of possible sun data which has been found to correlate
well with observed solar profiles and is used to establish the base index value: 90
percent is converted to a quality index of "9". Then "holes" (25 percent or more
attenuation of expected intensity) are divided and counted into three classes: 15,
30 and 60 minute duration. Index units are subtracted for each "hole" weighed by
duration class.
Target Conditions Established
There are three general areas of concern: a) whether target conditions were ob-
tained, b) whether the internal (one side consideration) and external ratios (two
sides compared) for each key condition parameter were obtained, and c) whether
the experimental conditions are quantifiable or verifiable.
The base sub-index is set at "10". One index unit is deducted for each 20
percent error from the absolute target condition desired. Very often with the dual
smog chambers it is the relative ratios of the initial component concentrations that
is more important. There are two types of relative comparisons which are made: the
ratios of components to each other within a mixture and the'ratios of components in
two different mixtures (side by side comparison). If a mixture is used, the relative
fractions of components often needs to be carefully established. Very often the
initial condition of interest is a matched HC to NOx ratio. One index point is
deleted for each 5 percent error from the target relative composition.
Usually initial conditions are verifiable analytically. When an analytical tech-
niques fails or a sample is not taken before the beginning of an experiment, other
techniques may have to be used to establish or estimate the initial conditions. Di-
rect injection of neat liquids or gases are often used as a calibration technique and
therefore can be used as a second best estimate of initial conditions. If a component
of a commonly used mixture is not directly analytically monitored for a particular
experiment, its concentration can be estimated from the concentrations of the other
mixture components and relative composition data. However these techniques are
less than ideal and two index units are deleted from this quality category for each
compound not verified by an analytical technique.
103
-------
Experimental Data Quality Assessment
Analytical Support
Analytical support is not always available for all species for all runs: techniques
sometimes fail during an experiment. The base index is set at "10" and index points
are deleted weighed by importance of the species or physical data not measured.
Chamber Conditions
Chamber conditions can have a major impact on experimental results. Important
parameters include: a) condensation on walls, b) dilution rate, and c) hours of
venting before the experiment. The base index is set at "10".
Condensation observed on chamber walls is noted by the operator starting the
experiment. One index point is deleted if condensation is observed on both cham-
bers. Chamber driers do not always operate well and sometimes one side may be
dried better than the other resulting in condensation in one side only. Two index
units are deleted for this situation.
Dilution rate is usually monitored by use of inert tracers such as CC14. A
exponential least-squares fit analysis yields the dilution rate (necessary for proper
modeling) for each chamber. The quality index is deleted by one if the dilution
rates differ by more than 20 percent. Additional points are deleted proportionally
for dilution rates exceeding the normal dilution rate of 1.5 * 10-4 min-1. The index
is deleted by two if the dilution rate is not quantifiable (incomplete of poor tracer
data). The quality index is deleted by one if the dilution rate is high or irregular
from windy conditions. Dilution rates are normally well matched and less than 10
percent per ten hours.
Two hours of venting will eliminate observable concentrations of chamber pol-
lutants. Regular experimental protocol requires that each run by preceded with at
least six hours of venting. Three index points are deleted if less venting time is
used.
Overall ValueNeed
Other major aspects of an experiment are overall usefulness and value. Many exper-
iments are conducted to satisfy questions requiring immediate attention or special
effort. Often these experiments are difficult to conduct; the experimental quality
may be less than desired but nevertheless contributes usefull information which
may be otherwise impossible to obtain. These experiments may be supportive of
other experiments. This category is used to adjust the quality index to reflect these
experimental needs.
104
-------
Final Segmented File Production Experimental Data
Documentation File Production
When all of the documentation information has been collected and the GEIiERAL DOC-
UMENTATION form has been filled out, the PC updates RU!.'E!iTRY with the SUM event. This
will indicate to the CT that documentation information is ready to be moved to the
VAXll/780. The CT uses a template file and a text editor. When the documentation
file is produced, the CT updates RUNEIITRY with the D2V event. This indicates to the
PC in the next PCREPORT that the documentation file is ready for final QA. The file is
listed and stored in the RUHFOLDER.
The listed documentation file is compared with the GENERAL DOCUMENTATION form
to be sure that at least the information was correctly transfered. The documentation
information is checked one last time with general experimental results: plots of the
experimental data are compared with the initial conditions for consistency of relative
reactivity. The stated initial conditions and the data plots should be checked for
agreement. The quality indices for sunlight should be compared with the light plots
for reasonableness. The general run quality index should be checked for overall
reasonableness having read the documentation and looked at the data plots. If
it is low, the reasons should be apparent in the documentation (e.g.missing data,
chamber condensation present, or poor sunlight).
When the QA criteria are satisfied, the PC updates RUHEI.'TRY with the DQA event.
The GENERAL section of RUNEHTRY is updated with the final initial conditions and
the quality index value. This is usually the last step before producing the final
segmented file.
Final Segmented File Production
When all data processing tasks are completed, the files are moved to the VAXll/780.
The files are merged together with a procedure called FIHALMRG which follows a pre-
scribed order. An ASCII labeled tape is made. Plots are made from the segmented
file as a QA step: if the format is incorrect the plotting programs will probably
not be able to read the file. The file is listed and inspected. The RUl.'El.'TRY event
flowchart is shown in Table 19.
105
-------
Experimental Data
Final Segmented File Production
Table 19.
Final Segmented File Prodxiction Steps
FINAL FILE data processing steps
Pers Step Meaning
CT
CT
CT
PC
PC
PC
MRS
.ALT
FPC
FQA
MTB
BAD
Merged 3 sect
Files to tape
Final plot done
Final QA
Not to be proc
Stop processing
Merge pieces to final file
Ascii labeled tape made
Correct cone plot, made
Final Quality Assurance
Something is wrong v.'ith this
data
106
-------
Backup of Data Experimental Data
Security of Data
There are three basic activities to enhance data security: inventory, retention of
original data to restricted areas, and maintaining several backup copies of all data
in different locations. All data are under lock at night.
Inventory
Before anyone is allowed to work with the run folder contents, a run folder inventory
is performed.which documents all data and forms. This helps assure that no data
or documentation is lost.
Retention of Data
All original data, both paper and electronic files are retained. Paper formated data
(all stripcharts, forms) are not allowed out of the office.
Backup of Data
Multiple copies of data are maintained in several formats in several locations.
Backup copies of raw and final data on floppies on the IBM PC are kept in several lo-
cations. Lotus worksheet files are backed up in two locations. Hardcopy (printouts)
of worksheet formulas using a spreadsheet AUDITOR is kept. Backup copies of raw
and final data on LSlli/23 computers are maintained in several locations, including:
raw data and processed data sets backed up on two diskettes, processing programs
(sources) backed up on two diskettes, and whole image harddisk backup of LSlll/23.
Backup copies of final data are maintained on the VAX and further backed-up on
tapes of VAX files. Frequent printouts of of data in all stages of processing ("P"-,
"C"-, and "K"-files) and plots are kept in runfolders and in notebooks.
107
-------
Segmented Data File
This section describes the final segmented data file used for distribution of the
experimental results.
General Description
All of the data for a run are stored in a single file. This includes: documenta-
tion including the calibration information; the continuous instrument data, every
4-minutes, alternating on chamber sides in physical units; and the gas chromato-
graphic instrument data given in concentration units for each side of the chamber
for each instrument approximately every 30 minutes.
108
-------
File Formats
Segmented Data File
File Formats
No record exceeds 80 columns (data is in "card image") form. The general lay out
of the file is:
DOCUMENTATION segment
General documentation section
88888 Documentation section terminator>
first instrument documentation ''K-file'' section
88888 Documentation section terminator>
second instrument documentation ''K-file'' section
88888 Documentation section terminator>
88888
last instrument documentation section
88888
DVH data documentation section
99999
DVM Data, up to 3 ''cards1' per time
99999
First instrument data, up to 3 ''cards'
99999
Second instrument data
99999
per time
99999
An example segmented data file is shown in Table 20.
109
-------
Segmented Data File File Formats
Table 20. Example Segmented Data File
GE1JERAL DOCUMENTATION
RUNDATE: OCTOBER 04, 1983
RUNTYPE: AUTO
RU1J DESCRIPTION: COMPARISON OF REACTIVITY OF EXHAUST
FROM DIRECT INJECTION FROM DODGE CHARGER IN HIGH IDLE
WITH SYNTHETIC AUTOEXHAUST.
RESULTS: TV/0 SYSTEMS RESULTED IN SIMILAR REACTIVITY
INITIAL CONDITIONS: BLUE RED
DODGE CHARGER 0.0 2.587
SYNTHETIC EXHAUST NMHC 2.190 0.0
NO 0.214 0.215
M02 0.037 0.039
88888
04-OCT-83 GENERATED ON 19-MAR-84
SY:OC043K.C1G .C1G
PICKED DATA ENTERED BY JEFFREY HOFFNER
CALIBRATION FACTORS APPLIED BY CHARLES
TOLUENE 2.912000E-01 PPMC/IN 2 KGS
ETHYLENE 9.020000E-02 PPMC.IN 2 KGS
NAME ABBREVIATIONS - SAME ORDER AS IN DATA
TOLUENE IS TOLUENE MAX A: 1.5610 MAX CON: 0.4546
ETHYLENE IS ETHYLENE MAX A: 4.9164 MAX CON: 0.4435
110
-------
File Formats
Segmented Data File
Table 20, cont. Example Segmented Data File
88888
04-OCT-83 GENERATED 0!i 13-MAR-84
SY:OC043K.C2G .C2G
PICKED DATA EilTERED BY JEFFREY HOFFNER
CALIBRATION FACTORS APPLIED BY JEFFREY HOFFNER
ETHANE 1.498000E-01 PPMC/IN 2
PROPYLE1IE 7.307000E-01 PPMC/I11 2
KGS
KGS
NAME ABBREVIATIONS - SAME ORDER AS III DATA
ETHANE IS ETHANE
PROPYLE1IE IS PROPYLEUE
MAX A: 0.2014 MAX COIJ: 0.0302
MAX A: 0.1311 MAX CON: 0.0958
88888
88888
USER DOCUMENTATION FOR RUN 831004
CALIBRATION FACTORS USED:
SPECIE
TIME INTERVAL BEGINNING GAIN ENDING BEGINNING OFFS ENDING
GAIN SLOPE GAIN OFFSET SLOPE OFFSET
HR HR-1 HR-1
HR
03G
2.500 15.500
18.000 6.000
1.01380 0.00000 1.01380 0.00000 0.00000 0.00000
0.00000 0.00000 0.00000 -9.99999 0.00000 -9.99999
END OF PROGRAM DOCUMENTATION
99999
YYMMDDHHMM U S 03G
LTMP TSR UV
8310040500 1 7R 0.0008
58.6810 -0.0046 -0.2439
8310040504 1 3B 0.0196
21.6880 -0.0062 -0.3430
HOG
0.0864
-0.0022
111
NOXG
H02G
DPG
CTMP
0.1138 0.0237 50.0450 56.8300
-0.0007 0.0001 56.6496 -60.5560
-------
Segmented Data File
ANSI Tape File Format
Table 20. cont. Example Segmented Data File
8310041704 1 3B 0.5787 -0.0012 0.0513 0'.0507 72.7266 -60.5120
20.5120 0.1592 11.7073
8310041708 1 7R 0.5508 -0.0013 0.0578 0.0573 64.7658 83.3620
81.3040 0.1376 10.2439
99999
YYMMDDHHMM USER SIDE GENERATED Oil 19-MAR-84
2.4,4 TRI TOLUE11E ETHYLEHE N-BUTAHE TRAIJS-2-B ISOPE1ITAII H-PE11TA11E ACETYLEliE
8310040625 1 B
0.2266 0.4546 0.4435 0.0877 0.0366 0.1339 0.0000 0.0905
8310040655 1 R
0.1868 0.2580 0.2248 -999.0000 0.0459 0.1195 0.0634 0.2332
8310041455 1 R
0.1439 0.1573 0.0890 -999.0000 0.0000 0.0828
8310041525 1 B
0.1821 0.2514 0.1300 0.0688 0.0000 0.0921
99999
YYMMDDHHMM USER SIDE GENERATED 01! 13-MAR-84
ETHYLEHE ETHAI.'E PROPYLENE
8310040625 1 B
0.4522 0.0093 0.0952
8310040655 1 R
0.4097 0.0238 0.0958
0.0340 0.1672
0.0000 0.0000
8310041525 1 B
0.1603 0.0086 0.0000
8310041555 1 R
0.2675 0.0260 0.0000
99999
ANSI Tape File Format
The segmented data files are written one file after the other onto 1/2 inch wide
industry standard magnetic tape at a density of 1600 bits per inch.
112
-------
Final Data Plots made from SegFile Segmented Data File
We have adopted the American National Institute of Standards (ANSI) standard
labeled tape format. This format is supported by a wide variety of computers
including most mini-computers, and IBM. An advantage of a labeled tape is the
"directories" of the tape contents can easily be created without having to list the
data.
Individual records of the files are collected together into a "block" of records
before being written to the tape. They can appear on the tape in one or two forms:
fixed length, blocked records, or variable length, blocked records. The default value
is fixed length blocked records with a record size of 80 bytes and a block size of 2000
bytes. For variable length records, the records are between 5 bytes and 84 bytes,
and the block size is 2048 bytes.
Figure 24, Figure 25, Figure 26, Figure 27, Figure 28, Figure 29, Figure 30,
Figure 31, and Table 21, Table 22, Table 23, and Table 24 explain the details of the
ANSI tape format as an aid in reading the tape. Volumn identifiers on the ANSI
tape (in the VOLl label field are UNC001 to UNC010 depending upon the release.
File identifiers on the ANSI tape (in the HDRl label format) are YYMMDD.SEG
for the year, month, day of the run.
Tape Contents
Final Data Plots made from SegFile
Although all processing of data and documentation is subjected to many QA pro-
cedures, the final SegFile is inspected and tested. The file is listed and inspected
for order and completeness of the SegFile components. This is also a last chance
look at the general documentation and format of the data files.
The final product is intended to be read by computer. The final QA test is
using a computer program to read the SegFile and produce plots of the data to
demonstrate that the file is correct and to examine the data visually.
113
-------
Segmented Data File
Final Data Plots made from SejFile
Beginning-of-Tape
Marker (EOT)
Volume Label
(VOL1)
File Header Labels
(HDRt HDR2. HDR3)
Tape Mark (TM)
File Section
Tape Mark (TM)
File Trailer Labels
(EOF1. EOF2. EOF3.
EOV1. EOV2. EOV3)
Tape Mark
Tape Mark
Scratch Tape
End of Tape
Marker (EOT)
Marks beginning of writeable area 'on a volume
Identifies the volume
Describes and delimits each file
Separates header labels from file section
Contains user data. Data in volumes interchanged to
non-VAX/VMS systems must be ASCII "a" characters.
Data m volumes interchanged to VAX/VMS systems can
be in binary form.
Separates file section from trailer labels
Describes and delimits files. When a volume is contin-
ued. EOV labels are written instead of EOF labels. EOF3
or EOV3 labels are written only when a HDR3 label is
written.
Indicates the logical end-of-volume. Two conseculive
tape marks are always written after the trailer labels of
the last file on a volume.
Incficates tape that is blank or that has not yet been
overwritten. Scratch tape can exist between the logical
end of volume and the EOT marker.
Marks beginning of the end of the writeable area.
Figure 24. Basic Layout of a VAX/VMS ANSI Labeled Volume
114
-------
Final Data Plots made from SegFile
Segmented Data File
Labels and Components Supported by VAX/VMS
Symbol
Meaning
Symbol
Meaning
BOT
EOF1
EOF2
EOF3
EOT
EOV1
EOV2
Beginning-of-tape
marker
First end-of-file
label
Second end-of-file
label
Third end-of-file
label
End-of-tape marker
First end-of-volume
label
Second end-of-volume
label
EOV3
HDR1
HDR2
HDR3
VOL1
TM
TM TM
Third end-of-volume
label
First header label
Second header label
Third header label
Volume label
Tape mark
Double tape mark
indicates an empty
file section or
the logical
end-of-volume
*
BOT
VOL1
HDR1
HDR2
HDR3
TM
N
First File
TM
EOF1
EOF2
EOF3
TM
TM
Scratch ^"
Tape
Figure 25. Single File/Single Volume Configuration
115
-------
Segmented Data File
Final Data Plots made from SegFile
Continuation Volume
BOT
VOL1
HDR1
HDR2
HDR3
TM
^N
Second Section of Second File
TM
EOF1
EOF2
EOF3
TM
HDR1
HDR2
HDR3
*s
TM
Third File
TM
EOF1
EOF2
EOF3
TM
TM
Scratch
Tape
EOT
ScratcrA
Tape I
Multifile/Multivolume Configuration
Label Identifier Volume Identifier
1 345 » 10 11 12
VOL
'/GEORGE-
Label Number Accessibility
Owner Identifier
38 I 505152
Reserved
Reserved
37
7980
Figure 26.
DIGITAL Standard Version
User Accessible Field
Label Standard Version
2K-3SO-8I
VAX/VMS ANSI VOL1 Label Format
116
-------
Final Data Plots made from SegFile
Segmented Data File
Label Identifier
11345
File Identifier
File-Set Identifier
File-Sequence
Number
Creation Date
21 22
2728 31 32 i 35 36 39404142
47
HDR
GEORGE
0001
0001
Label Number
File-Section
Number
Generation
Number
\
Generation-Version
Number
Expiration Date Block Count System Code
48 r 535455 i 6061
Reserved
73 74
80
000000
DECFILE11A
Accessibility
User accessible field
Figure 28. HDRl Label Format
117
-------
Segmented Data File
Final Data Plots made from SegFile
Label Block Record
Identifier Length Length
1 13456
10 11
151617
36 37
Label
Number
Record
Format
System-Dependent
Form Control
System-Dependent
38
50 5152
Reserved
80
00
Buffer Offset
User Accessible Field
User Accessible Subfield
Figure 29. HDR2 Label Format
118
-------
Final Data Plots made from SegFile
Segmented Data File
REC
50 Bytes
REC
50 Bytes
REC
50 Bytes
REC
50 Bytes
REC
50 Bytes
BLOCK
300 Bytes
Blocked Fixed-Length Records
Record Size = 54 Bytes
Record Size = 112 Bytes
Variable-Length Records
Figure 30. Blocked Fixed-Length Records
REC
50 Bytes
ZK-353-81
RCW
54
DATA
50 Bytes
RCW
112
DATA
108 Bytes
Pad
Characters
14 Bytes
BLOCK
180 Bytes
ZK-354-81
119
-------
Segmented Data File
Final Data Plots made from SegFile
Table 21. VAX/VMS ANSI VOLl Label
Character
Position
Field
(length in bytes)
Contents
1-3
4
5-10
11 .
12 - 37
38 - 50
51
52 - 79
80
"Label Identifier (3)
Label Number (1)
Volume Identifier (6)
Accessibility (1)
Reserved (26)
Owner Identifier (13)
DIGITAL Standard
Version 1
Reserved (28)
Label Scandard
Version 1
Alphabetic characters VOL
Numeric character 1
Volume label consists of ASCII "a"
characters.
Volume accessibility; provides
compatibility with some
non-VAX/VMS systems. A space, the
VAX/VMS default, indicates no
restrictions. To write any ASCII
"a" character in this field, use
/LABEL=VOLUME_ACCESSIBILITY wi th
the INITIALIZE command. Any
character but a space indicates
the /OVERRIDE qualifier must be
used with the INITIALIZE and MOUNT
commands.
Spaces
Volume ownership set by the
INITIALIZE/PROTECTION command.
The contents of this field are
used for volume protection
Numeric character 1
Spaces
Numeric character 3
120
-------
Final Data Plots made from SegFile
Segmented Data File
Table 22. First File Header Label (HDRl) Fields
Character
Position
Field
(length in bytes)
Contents
1-3
4
5-21
22 - 27
28 - 31
32 - 35
36 - 39
40 - 41
42 - 47
48 - 53
54
Label Identifier (3)
Label Number (1)
File Identifier (17)
File-Set Identifier (6)
File-Section Number (4)
File-Sequence Number (4)
Generation Number (4)
Generation-Version (2)
Number
Creation Date (6)
Expiration Date (6)
Accessibility (1)
55 - 60 Block Count (6)
Alphabetic characters HDR
Numeric character 1
A user-supplied file name and file
type
Same as the file-set identifier of
the first file on the first
volume, whether single or
multivolume configuration
Numeric characters starting at
0001 and incrementing by 1 for
each additional volume with
respect to the first volume on
which the file begins
File, number within the volume set
for .this file; consists of
numeric characters, starting at
0001 that indicate the position of
this file with respect to the
first file of the set
Numeric characters that indicate
the unique generation of a file
Numeric characters that indicate
the version of a particular
generation of a file
System stores the date in the
Julian format (IVYDDD)1; the
default is the current date
User specified Julian date
(SYYDDD)* or default is the
creation date, indicating file
'expires immediately
File accessibility; provides
compatibility with some
non-DIGITAL systems. A space
(used by DIGITAL systems)
indicates no restrictions. Any
character but a space indicates
the /OVERRIDE qualifier must be
used at mount time for the user to
access this file.
Always 000000 for the HDRl label
1. The number sign (I) in the Julian
format represents a space.
(continued on next page)
-------
Segmented Data File
Final Data Plots made from SegFile
Table 22. (Cone.): First File Reader Label (HDR1) Fields
Character
Position
Plaid
(length in bytes)
Contents
61 - 73
System Code (13)
74 - 80
Reserved (7)
Identifies the file system that
created the file. DEC, the
3-character constant, occupies
positions 61 through 63, followed
by the name of the file system;
DECFILE112 indicates VAX/VMS
Version 1.6 and earlier, and
DECFILE11A indicates VAX/VMS
Version 2.0 and later.
Spaces
Table 23. Second File Header Label (HDR2) Fields
Character
Position
Field
(length In bytes)
Contents
Label Identifier (3)
Label Number (1)
Record Format (1)
6-10
11 - 15
16
Block Length (5)
Record Length (5)
System-Dependent (1)
Alphabetic characters HDR
Numeric character 2
Character definition:
F fixed-length
D variable-length
The S for spanned record format is
returned as an undefined format
when processed by VAX/VMS 1
Five numeric characters specifying
the maximum number of characters
per block
Numeric characters indicating the
record length for fixed-length
records or the maximum record
length for variable-length records
In VAX/VMS Version 2.1 and later
versions, this field contains a
space indicating the VAX-11 RMS
attributes are in the HDR3 label
For VAX/VMS Version 2.0 and
previous versions, this field does
not contain a space but contains
the first byte of the VAX-11 RMS
attributes, Indicating the VAX-11
RMS attributes are in the HDR2
label
1. To process undefined records properly, the user must know what the
original format of the records was. Only logical I/O can be used to
process undefined record formats.
122
(continued on next page)
-------
Final Data Plots made from SegFile
Segmented Data File
Table 23. (Cant.): Second File Header Label (HDR2) Fields
Character
Position
Field
(length in bytes)
Contents
17 - 36
37
38 - 50
51 - 52
53 - 80
System-Dependent (20)
Form Control (1)
System-Dependent (13)
Buffer Offset (2)
Reserved (28)
Spaces available for future use.
For VAX/VMS Version 2.0 and
earlier, this field contains the
VAX-11 RMS attribute in binary
format
Defines the carriage control
applied to the records within a
file, as follows:
A First byte of record
contains FORTRAN control
characters
M Record contains all form
control information
Space 'Line feed/carriage return
will be inserted between
records (default)
Spaces available for future use.
For VAX/VMS Version 2.0 and
earlier, this field contains the
VAX-11 RMS attributes in binary
format
The numeric characters 00
Spaces
Table 24. Third File Header Label (HDR3) Fields
Character
Position
Field
(length in bytes)
Contents
1-3
4
5-68
69 - 80
Label Identifier (3)
Label Number (1)
VAX-11 RMS Attributes (64)
System-Dependent (12)
Alphabetic characters HDR
Numeric character 3
Files-11 record attributes that
override information in fields of
the HDR2 label
Spaces
123
-------
Summary of QA
Summary of Calibration and Data QA
3 NBS traceable standards for NO were used in this project
the three standards agreed at 3 concentration levels to within an average of 2.3
percent of each other
8 HC tanks and precision liquid chamber injections were used as calibration
sources in this project; two of the HC tanks were NBS tanks from an RTI multi-
year comparison study
the four principle HC tanks, validated with the RTI tanks and liquid chamber
injections, agreed to within an average of 1.6 percent of each other; and agreed
to within an average of 3.7 (std dev 1.8 percent) percent to the manufacturers
specified concentration values
HC calibration factors show very little or no trend in instrument sensitivity
during the experimental season; the relative standard deviation from the mean
calibration factors for the principle species used in this project ranged from 5
to 14 percent with an average of 9.9 percent (std dev 2.2 percent).
computer spreadsheets were used to adjust calibration factors (within observed
"noise" range) to maintain consistency of experimental initial conditions com-
position produced from fixed composition (pressurized gas tank and liquid mix-
ture) HC sources used in this project.
two pressurized tanks of HC mixtures were used to produce the experimental
initial conditions of the lower molecular weight compounds assuring consistent
composition and initial concentration throughout the experiment series
124
-------
After Run-Experiment Logging Summary of QA
one of the principle HC source tanks for the experimental program (UNCMIX,
a component of the synthetic urban mixture) is certified to 2 percent
a liquid mixture of the higher molecular weight compounds was prepared and
used throughout the program permitting one large neat injection to be made
with 1 percent syringes to establish the experimental initial conditions: assur-
ing consistent composition and initial concentration throughout the experiment
series
a multi-computer-based data acquisition and data processing system incorpo-
rated into a QA program which monitors every step of data generation and
processing is utilized to minimize risk of human e.rror and maximize accuracy
and overall data quality.
List of QA Steps
This section lists all specific QA steps performed.
Before Runs
review progress and schedule of experiments from work plan:
o do experiments to date make sense? are they consistent?
c do they indicate new experiments or other immediate needs?
o are data for these experiments good enough or must they be rescheduled?
check site status and weather report
match highest priority experiment with immediate site capabilities
fill out run sheets and instrument checklists
use site checkout lists
perform instrument checkout and calibration
After Run-Experiment Logging
PC phone call to site and update calendars:
o verify general quality of run and conditions
o reschedule?
use "pack-up" inventory checklist at site
check calendar for transfering data to floppies at site
125
-------
Summary of QA During Calibration
perform runfolder inventory
c verify proper and complete documentation from site
c verify complete data from site
RUllEllTRY - log experiment; store folder in office bookcase by date
calibration folders - store in office bookcase by date
CALEilTRY - log any calibration performed
calendar, run folder/RUilEUTRY, fastplot in notebook check:
c were.runfolder and computer data transfered to data processing office?
o successfully ?
During Calibration
planning:
c what species will be measured: what calibration sources needed?
c frequency?
o technique?
c documentation required?
c procedure for transfering calibration data to dpo?
identify standards and perform cross comparison:
o check standards consistency .
CALEllTRY - calibration log and processing status database:
o during run log-in: check instrument checklist for calibrations performed or
check for presence of calibration folder:
o log calibration data in calentry
routinely check CALE.'iTRY:
o check that all have been given processing instructions
check documentation QA while giving processing (pick) instructions:
o verify proper and complete documentation from site
inspect calibration data stripchart for appropriate appearance
o mark peak ids and chromatograms to pick
routinely check CALENTRY:
o check that all have been digpicked
126
-------
During DVM Data Processing Summary of QA
after DIGPIK inspect picked raw data on computer - compare with stripchart data
routinely check CALEliTRY:
o check that all are run through CALFAC
run CALLOK:
o check report lists and plots to display results for representative compounds
for early QA to identify any problems
check DTR HCCAL database:
o generate HCCAL report and scan entire report;
o observe large variations and problems
run CALAMA: final check and statistical analysis
during determination of calibration factors for data processing: does calibration
result in meaningful data?
During DVM Data Processing
calendars and CTREPORT inspection:
o are there data files missing?
o new files to be unpacked?
documentation comparison:
o run sheet, instrument checklist, remarks file: what were target conditions?
o any problems recorded?
inspect stripchart:
o were target conditions achieved (within calibration uncertainty)?
o documentation complete? recorder settings, side, time for all pens? -\
o does data make sense? are there unexplained problems? missing or noisy
data?
FASTDV plot and comparison with stripchart data
AUTOCAL strip (see cal)
AUTOCAL report:
o did AUTOCAL data get processed?
o any problems or large variation?
run DVMFIX:
127
-------
Summary of QA During GC Data Processing
o check that DVMFIX NOX concentration equals concentration of NO plus NOo all
during experiment
o check for good zeros
QA PLOCOli plots:
o can plot file (format good)
o data plot agrees with original stripchart
o was data processed properly? correct times, sides, and general magnitude?
c were holes processed properly?
o dewpoint data higher than temperature?
o does temperature data check with RDU airport?
During GC Data Processing
PCREPORT from RU!,'E!!TRY:
o instruments needing pick instructions correct?
documentation comparison:
c inspect run sheet, instrument checklist, remarks file:
o what were target conditions?
o any problems recorded?
inspect stripchart
o instrument documentation complete? recorder settings, instrument atten.
sides, time?
o does data make sense? see run sheet:
o is there missing or noisy data?
o is data good? worth processing? bad?
o identify calibrations and mark not to be processed as run data
o identify compounds, pick start and end time, bad data to be skipped, base-
lines for peaks
compare raw data stripchart and picked computer file:
o is raw (picked) data picked correctly?
PLOPic plots (of raw data)
o does data look reasonable? no side switching, attenuation switching prob-
lems?
128
-------
During Documentation Stages Summary of QA
c sides id correct?
after CALCOI::
o perform checks on C- and K-files and plots
calibration factors and RUHCALUSED:
c are calibration factors used consistent with others, or if not are they ex-
plainable?
compare K-file with data processing instruction form
o are the calibration factors used shown in K-file the same as specified in data
processing instruction form?
o are names of species correct?
PLOC011 plots:
o can plot be made (file format correct)?
o do plots seem reasonable?
HCANAL or LOTUS:
o if HC data for mixture, is composition consistent with known composition?
run sheet target:
c is resulting data consistent with target experimental conditions? or expected
initial conditions?
During Documentation Stages
compare documentation:
o inspect run sheet, instrument checklist, remarks file, calendars:
> what were target conditions? any inconsistencies? may need to consult
site operators (see run sheet)
> any problems recorded?
o special conditions, modifications, or operations performed?
> missing data? species not measured?
o how many hours of venting?
> drying performed? condensation observed in morning?
inspect stripcharts for notes of injection times or amounts
check climate data, site temperature and light data:
inspect data files and plots:
129
-------
Summary of QA After SegFile Made
o how complete is data?
o explanations available for lost data?
c reason to communicate this to modelers?
K-files check:
o documentation complete?
o any species co-eluting requiring documentation editing?
compare original experiment list, general documentation paper form and run
ranking formulas, and computer file:
o is information consistent or explainable?
o was experiment close to target? if not why not?
o is paper form with general documentation and formulas reduced and trans-
fered correctly to computer file (general doc for segfile)?
After SegFile Made
verify file format and contents:
o by listing file and performing a visual inspection of the contents
c and by plotting data from file and inspecting plots
130
-------
References
Smog Chamber Background Reactivity, Parti: The Influence of NOX from Chamber Walls on
Kinetic Computer Models for Air Pollution, submitted for publicntion
Smog Chamber Background Reactivity, Part2: The Influence of Radicals from Chamber Walls
on Kinetic Computer Models for Air Pollution, submitted for publication
Code of Federal Regulations, Title 40, part 50, subchapter C, Appendix F, pages 568-572.
131
-------
Site Checklists
Checklist For 0900 EDT Site Operator.
As you leave car walk by the red barn and pick up keys. A note pad for listing
problems detected in checkout should be used.
Outside Check
Walk towards smog chamber on service-gas-tank-house-side of lab. Look at lab
door. Are there any messages?
Listen for Air generator pump runningpull back on metal plates protecting
pump and inspect:
o are pumps upright?
o do they sound "right"?
Quick check five service gas tanks (assuming BECKMAN THC is not opera-
tional): He, H2 for Carles, 02 for NOX meter, ethylene for 03 meter, and Ar/CH4
for Varian PAN instrument (generally need minimum of 100 Ibs in tank to get
through the day). While in the service tank house, check flood lamps mounted
on wood chamber platform; if they are on, turn them off.
Continue walking to BLUE chamber - walk to where manifold goes under cham-
ber. Check service tank N2 for AC driers air circulation valve actuators: sec-
ondary meter should show between 15 and 20 Ibs and NOT higher or it will
destroy pneumatic valves; note total Ibs in tank and replace (later in the morn-
ing) if below 100 Ibs.
132
-------
Outside Check Site Checklists
Touch heat tapes on BLUE Glass manifold: they should be warm. Inspect for
condensation. Look along length of manifold including under chamber for large
breaks or condensation. Visually check BLUE circulation fan over injection
manifold in chamber: it should be running and turning smoothly. ,
Walk to opposite side of chamber and observe venting doors (Make mental note
of doors for comparison with run sheet which will be read shortly.
o If there is no run then chamber is probably venting; all doors should be
open and exhaust fan should be on.
o If a "run" is on-going then in most cases all doors will be shut. Check RED
exhaust door and make sure it is closed completely and door motor arm is
positioned exactly like BLUE door motor arm.
o If auto driers (AC system) is operational (one of the circulating fans will be
on) then an intake door will be open.
c If the dehumidifiers were used for drying they should have been removed
before injections. If they have not. make note on the runsheet and call PC
to determine if the experiment can be continued.
Visually inspect for condensation both inside and outside the chamber on both
sides (RED and BLUE). RuB! hand over a large area of one of the side (not
end) panels. Note any large drops of water or puddles on the floor of chamber.
Note position of center teflon wall separating the chambers. Does it seem to
hang more to one side than the other? Are all the fans running? Finally look
for general problems such as mud (footprints), bugs, objects (tools, tape, paper
towels, etc), in the chamber. Are the glass return and sample manifold intact?
If the dehumidifiers were recently removed, were the service panels replaced and
seated properly?
Continue walking around the chamber to the other side back to the injection
house and complete inspection of glass (RED) manifold. Feel heat tapes and x
inspect for condensation, breaks along length including under chamber. Is the
RED thermistor laying on the ground?
Go to BLUE side door in injection tank house. Observe if door is open or closed.
It should be closed. Open door to inspect tanks, manifold, and solenoid plugs.
Do not touch anything except the RED and BLUE injection lines! Do not turn
off any tanks or touch any regulators. You will be coming out again shortly.
Check that, the RED and BLUE 1/4 inch Teflon injection lines are connected
to the glass manifold. Check to see that they are tightly connected. Check to
see if ALL computer controlled AC power boxes for solenoids have a plug in it
(not including the light brown plastic box used for the WOOD project). They
133
-------
Site Checklist? _ Outside Check
should also each have a light bulb.
Close door and go to the other side: open door. Do any of the wires leading over
the door catch? Quickly inspect this side of the injection shed for problems.
Don't touch anything yet. Don't close any tanks yet. Is there a small wrench
for turning off the NO; tank later? Close the door.
Go inside the lab on the same side of building. Look up as you enter and notice
if the outside floods are on. If they are turn them off with the switch just inside
the door as you enter.
Turn to the right as you enter and head for the NOX and 03 strip chart. Look and
observe if NO and N02 (or possibly 03) have been injected. Notice approximate
concentrations. Are conditions matched (generally)?
Head for computer console (decwriter). Is the computer and DATCOL up? Are
there error messages? Is there a print profile? If not, can you get the time of
day from the computer? (Type TIME and return.)
Are ALL switches in AUTO? Or is there a note saying some should be in
MANUAL? Is the 4 minute clock in AUTO?
Turn around and look for a filled out run sheet for today's run. What was the run
today? Look at the command file printout if there is one. What was injected?
What tanks were used? Did anything have to be injected or otherwise done
manually? Was someone here this morning? See if morning operators name is
filled in. Are there any notes on the desk?
Does NOX and Oa stripchart make sense compared to information on run sheet?
Walk towards Carle GCs. Were injections made? Make quick inspection of
chromatograms. Did anything show?
Go outside to the injection shed (Carle side). Consider the run sheet, command
file, NOX and 03 stripcharts, and GC stripcharts. If injections were scheduled but
failed, inspect tanks to see if they were turned on and connected to appropriate
solenoids. Were they plugged in? R/B injection lines connected to the glass
return manifold? If resulting injections seemed low, check flow rate. (Hastings
mass fiowmeter reads 8.7 for 10.0 1pm; check with soap bubble flowmeter with
Once you are finished with the injection shed, turn off all tanks with the main
valve. Do not turn off tanks with the regulator. Use a small wrench to close
the NOo tank valve.
Enter the lab again on the NOX and Og side. Go to the DVM and HP clock.
Is the DVM scanning? If not, issue immediate command get time from the
computer by typing TIME and return. Does the HP clock match? (If not set it
134
-------
TSR and UV zero check Site Checklists
later). Check 4 minute side light. Go over quickly to NOX and Os instruments.
Check NOX and 03 pressure (20) and vacuum pump pressure (25.5). Check
Os ethylene (15) and rotoball setting. Mark stripchart with current side and
time. Does it look like the 4 minute R/B switching valve is operating? Perfect
matched injections are almost impossible. If it looks like a perfect match, check
the valve and the computer 4 minute switch again. Check that stripchart is
marked properly for run date, instrument, stripchart settings, attenuation, etc.
Is the HCHO instrument running? Should it be? Is the sample pump on? Is the
rotoball at 4? Is the solutions pump on? Is it drawing solution? Are bubbles
running through mixing/reacting chamber (below solutions pump). Check de-
gasser chamber (on right side). It should not be filling up with liquid (a little
bit at the bottom is ok). Check each sample line. Look at the HCHO stripchart.
Does the signal trace look "right"? Check strip chart zero by pushing down
black zero button: it should not be negative. If it is adjust it to about 10% of
the chartscale. Adjust HCHO zero if necessary and make note on stripchart.
Check waste container. Is it too full? Check that stripchart is marked properly
for run date, instrument, stripchart settings, attenuation etc. Check dewpoint
instrument stripchart. Check dewpoint stripchart zero. Mark it as zero check,
and mark time and side. Is DP pump on? Is signal trace reasonable? IF there
is a difference in dewpoint between the two sides the signal should be a square
wave. If it shows a slow response, clean the detector and rebalance after letting
it operate for a couple of minutes. If it looks 0. K. , check balance. Walk to
other side and move switch to test, wait for it to stabilize and adjust balance
knob. Turn knob back to operate. Check that stripchart is marked properly for
run date, instrument, stripchart settings, attenuation, etc.
TSR and UV zero check
Mark time.
Check manifold on this side (RED). Start at the NOX and O3 instrument end.
(While you are at this end check zero air supply used for injections and be sure it is
off.) Look for open fittings; missing caps or dangling sample lines. Feel heat tape;
it should be warm. Look for condensation. Check zero air supply near stripchart
for DP and HCHO. Move towards the other side of the bench. Check manifold fans
to see that they are operating as you go around to the other side. Perform manifold
check on this side walking to the end where the CARLEs are. Come back to the
manifold fans. Check zero air supply. 100 Ibs in and 60 Ibs out. Check Tektronix
terminal and clear screen by hitting page.
135
-------
Site Checklists GC Check
GC Check
Check switch 6 on AUTO/MANUAL switch panel: is switch 6 on (Carles valve min-
ders power)? Walk down towards Carles. On the way check the Varian stripchart.
Are there injection peaks? H2O peaks? Tracer peaks? Is the sample pump on?
Check that stripchart is marked properly for run date, instrument, stripchart set-
tings, attenuation, etc. Check bucking current.
Are Carle valve minders in auto/repeat and sequencing? Do chromatograms
look right? Air peaks? Look at example notebook if necessary. Are the GC sync?
Do the chromatograms look right considering what was injected and the nature of
today's run? Look at background chromatograms. Contamination? Ethylene? Was
an autocal scheduled and were the chromatograms made? Peaks on scale for initial
injection? Are they still on? Does it look like the baseline might drift off scale?
Adjust with GC zero, not stripchart zero. Look at the service gas regulators. Do
the readings match the numbers on the wall? Is the sample pump on? Is the 30
min RED/BLUE clock on and working? Is the flame lit? Reasonable attenuation
getting? Are Sigma integrator reports being made?
Considering what was injected for the run:
o Target match condition? Are they matched?
o If difference, are the sides switching? Are both sets of chromatograms on scale
(no peaks off scale)?
o If different amounts of same composition are ratios of peaks correct?
o Mark chromatograms with time and side. Try to mark time of sample injec-
tions.
o Try to mark when 30 min clock changes sides and whether B to R or R to B.
o Check that stripcharts are marked properly for run date, instrument, stripchart
settings, attenuation, etc.
Check lab AC. Is it set at 72? Is fan in ON position (not AUTO)? When was
the filter changed last? (It should be changed every three weeks.)
Enter PRINT PRO command if one is not printing already. Don't wait until
later.
136
-------
Formaldehyde
Checklist for Leaving the Site Set Up for a Run
Computer
Is the system up and ready?
There has to be a day created on the disk for today in order for the system to
run tomorrow.
Create a day on the disk with tomorrow's date using the program "NEWDAY".
If necessary, create one for today also.
Create a command file for tomorrow using the Program "CMANDR". Once you
are satisfied, store the command file by hitting "W. Now run the command file
thru "DATRAN" and merge ("EMERGE") it into tomorrow's DATCOL-OPS
file.
4
Run DATCOL, wait 1 minute and you should see a "*". which will tell you that
DATCOL is up; hit and you should get another *. If there was
a run today, key in "BACK". This will print out the commands for the current
day and tell you if the restart commands to vent the chambers (Set 1,2,3,4) at
1900 are still in there; otherwise the chambers won't vent!!! If there was not
a run today, issue an immediate command IMM and Set 1,2,3,4; to
open the chamber doors.
Are all front panel switches in auto not manual position; DVM on auto, rear,
remote.
Is the 4 minute clock switched to auto and not locked on one side?
Check decwriter paper. Need at least 10 pages.
Formaldehyde
Run on water for 15 minutes by placing all three reagent tubes in water.
Loosen all four tubes on parastatic pump.
Shut off two switches on rear of instrument starting from the right as you face
the switches, leave left most switch on. This supplies charging "juice" to the
instrument batteries.
Make sure there is enough TCM stock reagent prepared (without sulfite) so that
the person making the run in the morning just has to add 0.25 g sulfite to 200
ml TMC.
137
-------
NOx and O3 Instruments
NOx and O3 Instruments
Are the 3-way valves for the O3 and NOx meters plugged into the 4-minute
power strip?
Is there a good connection between the NOx and O3 meter intakes and the
round Teflon filters?
Are the front panel settings of the O3 and NOx meters correct?
c The NOx meter should be in ambient mode. NO, NO2. and NOx should be
set to "1 ppm full scale. The oxygen pressure should be set at 20, and the
valve knob should be set to NO-NO2-NOx. Vacuum gauge on NOx pump
should read 26" Hg.
o The O3 meter should be set to 2 ppm and full scale. The ethylene pressure
should be on 15. The time constant on the instrument should be set to 10
sec. The top of the + ethylene flow rotoball should be set at 30, and the
mode selector should be set on ambient.
o Strip Chart Recorder:
t> Pens on; on "record"
[> Chart "on"
> Voltages 1, 1, 1/2 for NO, NO2, O3
i> Chart speed 3 cm/hr and "onr
c Autocal Tank:
> Tank should be turned on, 5 psi showing on low side of regulator, and
on/off needle valve on regulator turned on.
i> Flow should have been previously adjusted so that meter will have
enough sample when span tank is turned on in the morning (rotoball is
positive showing excess flow when meter is sampling from this appara-
tus.
TSR and UV Chart
Have the TSR and UV pens been properly zeroed?
Charts
Check that sufficient paper remains Red line will appear on paper when the roll
is ending.
138
-------
Carle II nnd Carle III
Record the date, type of injection, attenuation, chart speed and scale of each
instrument on the charts.
Check that pens are full of ink and pens down.
Check that power to charts works from AC switch no. 16 on blue computer
AC switch panel. Activate no. 16 to manual and observe deflection on charts,
return these to "auto" position.
Carle I
Is the G.C. up and running?
o Check pressure gauges on the wall behind the instrument H2 = 28.5 psi.
He = 63 psi, air = 13.5 psi.
o Is the flame lit? (check by placing mirror or chromed surface over the
FID. It should fog H2 pressure to 35 psi. hit ignite button; you should
see positive response on strip chart (strip chart has to be on) after 5
minutes. Return H2 to 28.5 psi. Watch the gauge go down and make
sure it stabilizes at 28.5 psi.
o Has the proper attenuation been dialed into Carle I? (X4 for mix. X8
for propylene and ethylene. everything else determined by the nature of
the cal and the injection.
Has the valve minder for Carle I been put on light no. 5 and a drum setting
of 29?
c To check this, activate switch no. 6 on the computer.
o If the valveminder is not on light 5, switch the valveminder knob from
"standby" to "sequence" at 10 second intervals until light no. 5 is on.
While the valve minder know is in standby, rotate the timing drum un-til
the face setting is 29. CHECK FLAME AGAIN. The valve minder knob
should still be on standby at this time. Continue with Carle II and Carle
III.
Carle II and Carle III
Is GC up and running?
o Check pressure gauges on wall. Carle II, He = 38 psi. H2 = 27.5 psi,
Air = 12.5 psi. Carle III, He = 64 psi, H2 = 28 psi, Air = 13.5 psi.
o Check to see if flame is lit (place mirror over the FID).
139
-------
Sigma 10 Integrator
c Check for proper attenuation (normally X4).
Have the valve minders for Carle II and III been placed on light 8 and a
drum setting of 59?
o To check this activate switch 6 on the computer.
o If light 8 is not on. switch valve minder knob from "repeat" to "manual"
and activate "event advance'7, toggle switch until light 8 is on. WAIT
10 SEC. BETWEEN EACH ADVANCE SWITCH. Make sure that you
see light no. 1 before you advance to light no. 8. Set the drum at 59.
CHECK FLAME AGAIN. Return to the computer blue switch panel
and switch no. 6 to "auto". NOW GO BACK TO THE CARLES AND
SWITCH CARLE I FROM STANDBY TO AUTO.
. ALSO TURN THE VALVE MINDER ON CARLE II AND III TO RE-
PEAT. DO NOT LEAVE IN SINGLE OR OFF!!!
Sigma 10 Integrator
If we want to collect integrator data on the Sigma 10. Carle I. II, III must be
linked to the Sigma by a "set up" procedure.
SIGMA (your key strokes)
Response SET UP KEY EIJTER
EXAMPLE:
(1.2,4) 1 Enter
method?
1(2,4) Enter
mode Enter
Sample ID 1 Enter
Carle (no. ) enter
Ptr 1 Enter
Note that: Carle 1 - Inst 1, Method 1., Carle II = Inst 2. Method 2, Carle III
= Inst 4, Method 4. PE 900 = Inst 3, Method 41.
If PE 900 is to be used, is it set up? Instrument 3 Method 4.
Is there enough paper in the printer?
c There are 199 sheets total, so if the printer has less than 20 sheets left.
change the paper.
o Void plotter unless you want to use it. it will otherwise use a lot of paper.
140
-------
Air Generators
Clear out old SIGMA files by using the following keystroke sequence: Shift*F6
enter 200 enter or Shift *F Enter Enter. Hit the "Stat" key after this is done.
You should have at least 1000 free blocks.
Turn the Tekronics terminal off (this is the second channel to the Sigma).
Dew Point Meter
Has the dew point meter been balanced and returned to operate? Set to "Test"
and balance indicator to center of scale.
Has the detector been cleaned within a week? Using isopropyl alcohol and Q-tip
on mirror in detector in rear of instrument.
Is the dew point meter chart been properly zeroed?
Is the 3-way valve for the dew point meter plugged into the 4-minute power
strip?
ATC and Varian
Check standing current in morning and afternoon.
Set the ATC and Varian injection clock right before injection.
Is the ATC and Varian injection clock plugged into charts?
Check to see if the ATC and Varian sampling 3-way valve is plugged into the
30 min red/blue clock and if the Teflon lines from the 3-way valve are attached
to the red and blue sampling manifolds.
Is the correct attenuation dialed into the Varian (atten x*)?
Is the correct attenuation dialed into the ATC (X26)?
CO Meter
Is the sample line on the gas chambers?
Turn off service air tank Back on in morning.
Are front panel switches in correct position?
Air Generators
Must have 100 psi on input gauge, and 60 psi on output gauge.
141
-------
General
General
(CHECK INJECTION TANK FLOW WITH BUBBLE METER EACH DAY AND
SET TO 10 L/M1N).
Check tanks for tomorrow's injections.
o Go to the injection house and remove the line for the tank to be used from
its control outlet.
o Plug this line into the white power cord and read the tank flow on the mass
flow meter.
c If the flow rate is low or high adjust it with the low side pressure regulator
valve. All tanks should have a flow of 10 1/min including NO.
c Plug the tank line back into the control outlet for that tank.
Check the following tanks in the service gas house: O3 ethylene NOx oxygen.
Carle I He and H2, Carle II He and H2, CO3 and H2; Varian: ArCH4.
Liquid Injections
c Determine the number of ul that have to be injected from the computer
listed table on the wall in back of the formaldehyde instrument. This table
tells you the number of u] that you have to inject into the chamber to give
1 ppmC. This has been computed for three different temperatures 60 F,
70 F. and 80 F. Selected the number ul to be injected from the list which
will approximate tomorrow's morning temperature. Clearly write down the
number of ul to be injected on the run sheet.
o Locate the liquids to be injected and set them out in the work space of the
hood. Find the liquid injection apparatus and clean the needed syringes so
that the operation will not have to spend time doing this tomorrow. Place
these on a clean paper towel by the injection ports of the return manifold.
o If you have to calculate the no. of ul to get 1 ppmC of a liquid into the
chamber use the following formula:
Has the run sheet been filled out?
Have the charts been marked with day. month, year, instrument, chart speed.
full scale chart voltage, and attenuation and range where applicable?
Check for loose or missing caps on the manifolds.
Check the chamber recirculation fans.
Check all the red blue 3 way valves connected to the GCs to make sure they are
plugged into the 30-min power outlet.
Final check
142
-------
General
o Check all auto/manual switches-computer and cal box-to make sure they
are in auto position.
o Check DVM to make sure it is in remote not local mode. Especially 4 minute
clock; it should be in auto.
o Check the red exhaust door on the chambers to see if it is the same as the
blue exhaust door.
o Make sure all charts are on and plugged into "charts" strip.
o Check Carle 1 to be sure its valveminder is in auto and that Carle II and III
valve minders are in repeat.
o Check high Cone. NOx calibration tank to be sure it is off.
o Check the Cal tank for the Carles to see that it is open.
o Check injection tanks to seer that they are open and that the flow is correct.
o Check the reagent levels for the HCHO instrument; also check the waste
flask to see that it can't overfill.
o Shut off water to toilet.
c Make sure the chambers doors will open at 19:00. and vent, by a Datcol
Restart command, or by an immediate command to Datcol to set 1.2.3.4:
before you leave the site.
143
-------
Early Morning Checkout
Early Morning Checkout
Did the command file run? If not. see HELP LIST on end of rack over the DEC
writer. Booting instructions are at other end of the rack over the disc.
Is the DVM scanning? If not, issue an IMM to stop GASDVM; then, another
IMM to start GASDVM. If DVM still is not scanning, switch from "Remote"
to "Local". "Step" through to end of series (channel 60). return to "Remote"
and issue an IMM to start GASDVM.
Did the chambers vent? Check this by reading the 03 strip chart-background
should be almost completely free of O3. If the chambers have not vented and
enough time remains before sunrise-2.5 hours-manually vent chambers until
just before sunrise. Close the chambers and then make all injections manually.
Make CEA formaldehyde solutions.
Read the run sheet and DEC printout of the day's command file to determine
target concentrations.
Find manual injection species and amounts and all apparatus .necessary to make
injections.
Check chamber condensation and all vent doors (especially red exhaust door).
Go to the injection house at the time of the injections to see if the lights confirm
that the proper tanks are being injected. A list of tank concentrations, species,
and switch numbers is on the end of the rack above the DEC writer.
Check all charts.
o Charts on at charts.
o Pens down and full.
o Charts fully labeled.
o Sufficient chart paper for the day. x
Check manifolds to see that all caps are on and that each instrument is connected
to both the red and blue manifolds.
See that 4 minute computer clock is in AUTO and not locked manually to either
side.
Check status of SIGMA 10.
o "/STAT"; all Carles and PE 900 (if used) should be set up with the corre-
sponding methods.
t> Carle 1 Method 1
> Carle 2 Method 2
144
-------
Early Morning Checkout
i> Carle 3 Method 4
> PE 900 Method 31
o Check that sufficient paper remains for the day (about 75 sheets).
Start CEA formaldehyde.
o On sample line
o Advanced past auto 0
o All pump lines connected and useable and pulling solutions.
o Sample pump on.
Turn on air tank to the Beckman 6800 and turn on the instrument if it is to be
used.
Perform manual injections before 05:00 at the latest.
Check standing current on the Varian: it should read about 600-9 x64 amps.
Check injection concentrations and relate to target amounts by measuring peak
heights and using the provided concentration conversion factors. See attached
chromatograms.
o NOX: additive value from strip chart or from DVM printout.
o Carle I,II.Ill: see attached chromatograms and respective values.
c Adjust concentrations of NOX or HC species, if necessary, to achieve desired
amounts,
Change attenuation on the Carles after the auto cal; generally, attenuation
should be x2 or x4.
. Issue a PRINT PROFILE of the DVM data.
After successful injection turn off injection tanks and successful auto HC span
turn off HCspan tank.
145
-------
Run Pack-Up Checklist
Run Pack-Up Checklist
Before a run folder is taken to school, check to see if the following items are in
it:
1. Carle I and II stripchart
2. Carle III stripchart
3. Sigma 10 printout
4. Varian stripchart
5. Dewpoint stripchart
6. Beckman stripchart
7. PE900 stripchart
8. DNPH integrator Printout and stripchart
9. TSR and UV stripchart
10. Formaldehyde stripchart
11. NOX and 03 stripchart
12. Run Sheet
13. Instrument Checklist
14. Profile Printout
15. Copy of command file used
16. Copy of BAKTRN of the day's OPS file
If one of these is missing note reason on the instrument checklist.
146
-------
HC Cal Sources Certification
147
-------
APPENDIX B.
COMPARISON OF HYDROCARBON CALIBRATION SOURCES
RTI TANKS
ave USED
regression estimate
Last Update:
PRIMARY STANDARD #1 AND #2
Species manuf.
cone.
propylene 14.820
27-Apr-78 14.580
22-May-78 14.820
23-May-78 14.340
08-Aug-78 14.940
14-May-80 14.790
03-Oan-84 14.400
14.645
02-Aug-84 14.449
13-Oun-85
Specles manuf.
cone.
ethylene 20.000
27-Apr-78 19.400
25-May-78 19.200
10-Aug-78 19.800
0G-May-80 16.800
1'J-Oul-Gl 20.000
29-May-84 19.000
19.480
02-Aug-84 19.304
Calibration of Carles 1,2,3 with RTI tanks » SIGMA 10 Integrator ppmC/area
Assumption - these species are representative of average carbon response efficiency.
Did use May 5 1980 ethylene value.
Choice between regression value or average analysis value: used average.
UNC use: 14.645
UHC use: 19.480
Carle I
Ave
al lave
8/2/84 Sigma 10
44.6
44.5
44.5
44.8
ppmc/area
0.328
0.329
0.329
0.327
slgma 10 ppmc/area
5G.3 0.346
57.1 0.341
57.2 0.341
0.328
0.343
0.335
8/1/84 slgma 10 ppmc/area
44.8 0.327
0.327
-------
Carle II
8/2/84 sigma 10 ppmc/area
19.1 0.767
19 0.771
19.2 0.763
19.6
19.2
19. 1
18.9
0.747
0.763
0.767
0.775
slgma 10
24
24.4
24.8
24.8
24.5
ppmc/area
7 0.789
0.798
0.785
0.785
0.795
Ave
al lave
0.765
ave
8/1/84 slgma 10 ppmc/area
17.4 0.842
19.2 0.763
19.3 0.759
20.5 0.714
19.3 0.759
0.767
0.778
sigma 10
25.31
24.7
24.7
25.2
0.774
0.791
ppmc/area
0.770
0.789
0.789
0.773
0.780
Carle III
8/2/84 slgma 10 ppmc/area
4.15 3.529
4.15 3.529
4.14 3.537
Ave
17
14
3.512
3.537
3.529
8/1/84 slgma 10 ppmc/area
3.95 3.708
4.03 3.634
4.44 3.298
Ave
3.547
-------
Atrco Tank - Determination of Concentration based
Date
Carle I
8/2 - RTI
ave
8/1 - RTI
ave
toluene ethylene
area cone. area
ave ppmC/are 0.335
28.90
29.90
29.40 9.86
ave ppmC/are 0.327
31.20
31 .50
31.35 10.25
on RTI tank - analyzed same day
trans-2-butene
area
21.30
21 .50
21 .40 7.18
22.70
22.70
22.70 7.42
1 sopentane
area
28.00
28.00
28.00 9.39
29.70
29.80
29.75 9.73
propane
area
. 2 2 . 00
22.00
22.00
23.60
23.60
23.60
propylene
area
24.80
24.90
7.38 24.85 8
26.20
25.90
7.71 2G.05 8
.34
.52
tn
O
Carle II
8/2 - RTI ave ppmC/are 0.778
8/1 - RTI ave ppmC/are 0.774
16.20
14.50
14.20
14.70
ave 14.90 11.53
Carle III
8/2 - RTI ave ppmC/are 3.529
3.22
3.26
ave 3.24' 11.43
8/1 - RTI ave ppmC/are 3.547
3.30
3.23
3.24
ave 3.26 11.55
4.75
5.08
4.03
4.75
4.SI
4.87
3.69
3.77
1.94 2.52
1.96 2.55
1.95 6.88 2.54
1 .84
1 .90
1 .86
1 .87
2.42
2.47
2.44
6.62 2.44
8.95
9.74 7.57 10.80 8.40
10.60
10.30
9.66
9.64
10.05
11 ,
1 1 .
IS.
I 1
60
40
90
00
7.78 11 .23
8.68
8.67
THE AVERAGE -Alrco tank 10.92 3.73
THE STD. DEV 0.72 0.04
(these values are listed below - Alrco Tank Final)
7.03
0.30
9.18
0.41
7.61
0. 15
8.48
0. 13
-------
CALIBRATION OF CARLES WITH AIRCO FOR ANALYSIS OF LOW AND HIGH MW HIGH CONC TANK - 7/18
"> <" Indicates values not used In calculations
Date
toluene
area
ethylene trans-2-butene
cone. area cone. area cone.
10.92 3.73 7.03
Isopentane propane propylene
area cone. area cone. area cone.
9.18 7.61 8.48
Carle I
7/18- Cal with Alrco standard
ave
Cal today
ave Carle
Carle II
7/18- Cal
ave
Cal today
ppmC/area
I ppmC/area
with Alrco
>
ppmC/area
26.7
32.1
29.40
0.372 23.7 15. 1 <
11.5
10.5
11.8
11 .27
0.676
26.6
26.5
2G.55
0.320
11.8
11.9
11.6
11 .9
11 .80
0.719
ave Carle II ppmC/area 0.693
Carle III
7/18- Cal with Alrco standard
> 4.17 < NOT USED
3.31
3.26
3.25
3.27
ave
Cal today ppmC/area
3.27
3.337 <
NOT USED
1 .92
1 .82
1 .82
1 .82
1 .97
1 .87
3.757
3.94
2.37
2.41
2.38
2.40
2.39
3.842
4 . 26 propane &
4 . 27 propy1ene
4.27
4.27
4.39
4.292
3 .750
-------
ave Carle III ppmC/are 3.783
Date toluene
area
Carle I
8/14- Cal with Alrco standard
ethylene trans-2-butene tsopentane propane propylene
cone. area cone. area cone. area cone. area cone. area cone.
10.92 3.73 7.03 9.18 7.61 8.48
ave
Cal today ppmC/area
26.3
27.1
26.70
0.409 < NOT USED
20.6 27.2
20.9 27.4
20.75 27.30
0.339 0.336
21 .5
21 .3
21.40
0.356
24.5
23.8
24.15
0.351
ave Carle I ppmC/area 0.345
Carle II
8/14- Cal with Alrco standard
in
to
ave
Cal today ppmC/area
13.5
13.9
13.70
0.797
4.7
4.62
4.66
0.800
9.58
9.31
9.45
0.806
10.5
10.3
10.40
0.816
ave Carle II ppmC/area 0.805
Carle III
8/14- Cal with Alrco standard
ave
Cal today ppmC/area
3.29
3.26
3.28
3.336
1 .95
1 .97
1 .96
3.584
2.57
2.57
2.57
3.573
4.65
4.63
4.64
3.469
propane &
ave Carle III ppmC/are 3.490
-------
Liquid Injection Cals same day!
SIGMA 10 AREA CALS
Date
tol uene
area
m-xylene o-xylene
cone. area cone.
2.99
2.98
area
pentane 2-me-pentane d litr 1 -me-pentane
cone. area
2.66
cone. area
2.99
cone. area cone.
3.16 5.98
Carle I
8/14
ave
Cal
ave
- Cal with LIQUID
today ppmC/area
Carle I ppmC/area
standards
8
8
8
8
0.
SS.
.06
.17
.06
. 10
369
367
8
7
7
7
0.
aro 0.
.39
.82
.37
.86
379
378 par
P/A=
7.46
6.8
6.4
6.89
#.387
0.356
0.94
8
8
8
8
0.
.82
.52
.51
.62
347
NOTE ARO CALS
8.47
8.3
9.14
8.64
0.366
HIGHER THAN PAR
17
17.1
16.6
16.90
0.354
Carle II
8/14- Cal with LIQUID standards
ave
Cal today ppmC/area
4.42
4.22
4.32
0.692
4.33
4.6
4.47
0.667
3.72
4.06
3.89
0.685
ave Carle II ppmC/area 0.681
Carle III
8/14- Cal with LIQUID standards
ave
Cal today ppmC/area
0.989
0.983
0.986
3.030
0.917
0.931
0.917 "
0.922
3.230
0.797
0.800
0.784
0.794
3.355
0.812
0.787
0.787
0.795
3.757
ave Carle III ppmC/are 3.343 NOTE DIFFERENCE BETWEEN AROS AND ALKANE
std dev 0.2654
-------
PEAK HEIGHT CALS
Date
toluene
ht In
m-xylene o-xylene pentane 2-me-pentane
cone. ht In cone. ht tn cone. ht In cone. ht in cone.
2.99 2.98 2.66 2.99 3.16
Carle I
8/14- Cal with LIQUID standards
ave
Cal today ppmC/ln
6.8
7
6.86
6.89
0.434
3.4
3.4
3.3
3.37
0.884
2.4
2.4
2.4
2.40
1 .110
8.4
8.4
8.36
8.39
0.356
6
5.8
5.8
5.87
0.539
Carle II
8/14- Cal with LIQUID standards
ave
Cal today ppmC/ln
6.1
6.1
S. 10
0.490
3.2
3.2
3.20
0.930
2.3
2.3
2.30
1 .158
Carle III
8/14- Cal
with LIQUID standards
ave
Cal
today ppmC/f n
13.2
13.2
13.20
0.226
7.3
7.3
7.30
0.40B
5.2
5.2
5.20
0.512
-------
Ul
en
Low MW
Date
Carle
7/18-
ave
Date
Car le
7/18-
ave
Date
Carle
7/18-
ave
Date
Carle
7/18-
ave
Date
Carle
7/18-
ave
HI Cone Tank - Determination of
octane
area cone.
I
AIRCO ave ppmC/a 0.313
288.60
300.10
294.35 92.08
pentane
area cone.
I
AIRCO ave ppmC/a 0.313
63.80
63.80
63.80 19.96
octane
area cone.
II
AIRCO ave ppmC/a 0.693
129.30
129.30 89.56
octane
area cone.
Ill
AIRCO ave ppmC/a 3.783
28.20
28.30
28.25 106.87
1 sopentane
area cone.
Ill
AIRCO ave ppmC/a 3.783
4.88
5.98
5.43 20.54
Concentration
based on Alrco Cals - analyzed
ethy lene/ethanbutanes-butenes
area area
35.50
38.10
36.80 11.51
274.90
277. 10
276.00 86.34
tsopentane
area
56.50
56.80
56.65 17.72
same day
propane propylene
area area
66.00 66.90
67.00 67.30
66.50 20.80 67.10 20.99
2-me-l , 3-butad lene
area
132.40
133.90
133.15 41.65
ethyl ene
area
13.40
13.30
13.35 9.25
1-butene
area
4.91
5.10
5.01 18.93
pentane
area
7.12
8.55
7.84 29.64
ethane
area
13.30
14.40
13.85 9.59
n-butane
area
2.71
2.83
2.77 10.48
propane/al lenepropy lene
area area
47.60
47.10
47.35 32.80
c 1s-2-butene
area
4.98
5.09
5.04 19.05
27.20
27.20 18.84
trans-butene 1 , 3-butad lene
area area
- 5.12 4.51
5.21 4.63
5.17 19.54 4.57 17.29
propane/proplene/allene
area
14.80
16.50
15.65 59.20
-------
Low MW HI CONC Tank as source for cal for determination of LOMW
Date
7/13
Carle I
ave
Cal Today ppmC/area
Date
7/13
Carle I
ave
Cal Today ppmC/area
Carle I ave ppmC/area
Date
7/13
»_*
Cn
°> Carle II
ave
Cal Today ppmC/area
octane
area cone.
96.17
319.70
299.70
309.70
0.31 ??
pentane
area cone.
19.96
65.40
65.00
65.20
0.306
0.315
octane
area cone.
96.17
148.30
-------
Low MW
Date
Carle
7/13-
ave
Date
Carle
7/13-
ave
Date
Carle
7/13-
ave
Date
Carle
7/13-
ave
Date
Carle
7/13-
ave
Lo Cone Tank - Determination of
octane
area cone.
I
LMWHI ave ppmC/a 0.315
19.62
19.78
19.70 6.21
pentane
area cone.
I
LMWHI ave ppmC/a 0.315
4.38
4.33
4!36 1.37
octane
area cone.
II
LMWHI ave ppmC/a 0.720
0.00 0.00
octane
area cone.
Ill
LMWHI ave ppmC/a 3.954
1 .81
1 .99
1.90 7.51
1 sopentane
area cone.
Ill
LMWHI ave ppmC/a 3.954
0.31
0.31
0.31 1.22
Concentration
based on HI cone
ethylene/ethanbutanes-butenes
area area
0 . 00 0 . 00
17.27
17.28
17.28 5.44
Low MW Cals - analyzed same
(sopentane propane
area area
3.88 ' 4.31
3.83 4.07
3.86 1.21 4.19 1.
day
propylene
area
4.41
4.85
32 4.63 1.46
2-me-l , 3-butad lene
area
8.21
8.21 2.59
ethylene
area
1.87
1.87 1.35
1-butene
area
0.33
0.33
0.33 1.30
pentane
area
0 . 00 0 . 00
ethane
area
1.19
1.19 0.86
n-butane
area
0. 18
0. 18-
0.18 0.72
propane/al lenepropylene
area area
3.45 2.67
3.45 2.48 2.67 1.
cls-2-butene trans-butene
area area
0.33 0.34
0.33 1.32 0.34 1.
92
1 , 3-butad lene
area
0.30
0.30
36 0.30 1.18
propane/proplene/al lene
area
0.99
0.99
0.99 3.91
-------
Low MW Tank as source
Date
5/30
Carle I
ave
Cal Today ppmC/area
Date
5/30
Carle I
ave
Cal Today ppmC/area
Carle I ave ppmC/area
Date
5/30
Carle II
ave
Cal Today ppmC/area
for cal
octane
area
329.00
336. 10
332.55
0.29
pentane
area
65.10
65.20
65.15
0.306
0.307
octane
area
129.70
129.70
129.70
0.741
for determination of
HMW Cal Tank
came day
et hy lone/ etna nbutenes-butenes Isopentane
cone . area
96.17 18.
56.70
56.50
56. G0
0.33
area
84
281 .40
280.50
200.95
0.30
area
05.29 17
58.30
57.70
58.00
0.31
propane propylene
area area
.72 20.80 20.99
70.30 68.40
','fS.ZS 60.20
70.25 e;-i.30
0.30 0.31
2-me-l ,3-butad)ene
cone. area
19.96 41.
131.90
131.80
131 -85
0.316
ethylene
cone. area
96. 17 9.
12.80
12.60
12.70
0.728
65
ethane
area
25
13.40
13.20
13.30
0.721
propane/a 1
area
9i59 35
48. 10
1 enepropy lene
area
.51 20.99
28.80
40.70
-------
Cn
(O
i Date Isopentane pentane
j 5/30 area cone. area
17.72 19.96
Carle III
4.30 4.71
4.32 4.72
ave 4.31 4.72
Cal Today ppmC/area 4.11 4.23
Car.le III ave ppmC/are 4.09
High MW Tank
Date
Carle I
7/18- AIRCO
ave
Date
Carle II
7/18- AIRCO
ave
Date
Carle III
7/18- AIRCO
ave
- Determination of
toluene
area
ave ppmC/a 0.313
60.50
60.50
tol uene
area
ave ppmC/a 0.693
26.40
26.40
tol uene
area
ave ppmC/a 3.783
6. 12
6.12
propane/proplene/al lene
area
56.50
13.90
13.90
13.90
4.06
Concentration based on Alrco Ca.l
ethylbenzene
cone. area cone.
63.50
18.93 63.50 19.86
ethylbenzene
cone . area
25.80
18.29 25.80 17.87
ethylbenzene
cone. area cone.
6.42
23.15 6.42 24.29
m-xy lene
area
61 .90
61.90
m-xy lene
area
29.60
29. G0
m-xy lene
area
6.80
6.80
s - analyzed same day
o-xylene 1-pentene
cone. area cone. area cone.
60.80 49.60
1.9.36 60.80 19.02 49.60 15.52
o-xylene benzene
area area
27 .00 24 .00
20.50 27.00 18.70 24.00 16.62
o-xylene 1-pentene benzene
cone. area cone. area cone. area
6.37 7.21 5.74
25.72 6.37 24.10 7.21 27.27 5.74 21.71
NOT USED
-------
Date Isobutane
area
Carle III
7/18- AIRCO ave ppmC/a 3.783
eyelohexane
cone. area
ave
Date
Carle I
8/14- AIRCO
ave
Date
Carle II
8/14- AIRCO
ave
3.03
3.03
tol uene
area
ave ppmC/a 0.345
54.84
55.94
55.39
tol uene
area
ave ppmC/a 0.805
24.42
24.63
24.53
9.G4
11.46 9.64 36.47
ethy Ibenzene
cone. area cone.
59.31
61 .65
19.14 60.48 20.89
ethylbenzene
cone. area
25.19
25.77
19.74 25.48 20.51
m-xylene
area
57.65
60.26
58.96
m-xy lene
area
30.66
31.12
30.89
o-xylene
cone. area cone.
55.37
58.22
20.37 56.80 19.62
o-xylene
area
25.87
26.31
24.86 26.09 21.00
1-pentene
area cone .
44.68
44.88
44.78 15.47
acetylene benzene
area area
20.61 23.00
21.70 23.10
21.16 17.03 23.05 18.55
Date toluene
area
Carle III
8/14- AIRCO ave ppmC/a 3.490
ave
Date
6.73
6.82
6.78
Isobutane
area
Carle III
8/14- AIRCO ave ppmC/a 3.490
ave
Date
3.37
3.38
3.37
toluene
area
ethylbenzene m-xylene
o-xylene
1-pentene
benzene
7.29 7.08
7.45 7.08
7.37 25.73 7.08 24.70
7.07
7.20
23.65 7.14 24.91
cyclohexane
cone. area
11 .77
ethylbenzene m-xylene o-xylene 1-pentene
cone. area cone. area cone. area cone. area cone.
4.30 6.36
4.31 6.43
4.30 15.02 6.40 22.32
-------
Carle I
8/14- LIQUID ave ppmC/ 0.367
54.84
55.94
ave 55.39
Date toluene
area
Carle II
8/14- LIQUID ave ppmC/ 0.681
24.42
24.63
ave 24.53
59.31
61.65
20.32 60.48 22.19
ethylbenzene m-xylene
cone. area
25.19
25.77
16.70 25.48 17.35
57.65
60.26
58.96
'lene
area
30.66
31. 12
30.09
55.37
58.22
21.63 56.80 20.84
o-xylene
area
25.87
26.31
21.04 26.09 17.77
44.68
44.88
44.78
acety 1 ene
area
20.61
21 .70
21.16
16.43
benzene
area
23.00
23. 10
14.41 23.05 15.70
Date toluene
area
Carle III
8/14- LIQUID ave ppmC/ 3.343
6.73
6.82
ave 6.78
Date ' Isobutane
area
Carle III
8/14- LIQUID ave ppmC/ 3.343
3.37
3.38
ave 3.37
ethylbenzene m-xylene o-xylene 1-pentene benzene
cone. area cone. area cone. area cone. area cone. area
7.07
7.20
22.65 7.14 23.86
cyclohexane
cone. area
11.27
7.29 7.08
7.45 7.08
7.37 24.65 7.08 23.66
4.30 G.36
4.31 6.43
4.30 14.38 C.40 21.38
Date
toluene
ph x64
Carle I
8/14-LIQ av ppmC/1n x4 0.434
attn correction
ave
16
2.95
3.00
2.98
m-xylene o-xylene
cone, ph x64 cone. ph In cone.
0.88
1.55
1 .60
20.65 1.58 22.28
1.11
1 .27
1 .30
1.29 22.81
-------
Date toluene
ph 1 n
Carle II
8/14-LIQ av ppmC/ln x4 0.490
m-xylene o-xylene
cone. ph in cone. ph in cone.
0.93
1 . 16
attn correction
16
ave
Date
Carle III
8/14-LIQ av
2.47
2.50
2.49
toluene
ph In
ppmC/ln x4 0.226
1 .80
19.48 1.80 26.79
m-xy lene
cone. ph In cone.
0.41
1.22
i.2b
1.24 22.88
o-xylene
ph in cone.
0.51
attn correction 16
ave
Date
Carle I
5/30- LMWHC
ave
Date
Carle I
5/30- LMWHC
ave
5.72
5.62
5.67
tol uene
area
ave ppmC/a 0.307
66.80
66.71
66.76
isobutane
area
ave ppmC/a 0.307
29.60
30.70
30. 15
3.72
3.80
20.54 3.76 24.53
ethylbenzene
cone. area cone.
66.50
67.70
20.50 67.10 20.61
cone .
9.26
2.80
2.85
2.83 23.15
m-xylene o-xylene 1-pentene acetylene
area cone. area cone. area cone. area cone.
38.20 NA 48.60 42.80
42.00 NA 48.60 43.70
40.10 12.31 NA NA 48.60 14.93 43.25 13.28
-------
Date toluene
area
Carle II
5/30- LMWHC ave ppmC/a 0.732
25.40
25.70
ave 25.55
ethylbenzene m-xylene
cone. area area
26.00
26.40
18.69 26.20 19.17
o-xylene
area
29.50 27.10
30.80 28.20
30.15 22.06 27.65
acetylene
area
20.23
23.80
24.60
24.20
benzene
area
23.40
23.70
17.70 23.55 17.23
Date
Carle III
5/30- LMWHC ave ppmC/a
ave
toluene
area
4.094
5.83
5.80
5.82
Date Isobutane
area
Carle III
5/30- LMWHC ave ppmC/a 4.094
ethylbenzene m-xylene o-xylene 'l-pentene benzene
cone. area cone. area cone. area cone. area cone. area
6.08
6.17
23.81 6.13 25.07
cyclohexane
cone. area
6.24
G.32
G.28 25.71
5.97 3.63 5.48
6.05 3.61 5.47
6.01 24.60 3.62 14.82 5.48 22.41
ave
ethylene
propy lene
propane
tol uene
trans-2-butene
Isopentane
2
2
2
.84
.85
.85
Alrco Tank
tnanuf. UNC-
4.80
9
8
9
6
9
.15
.61
.94
.92
.85
11
.65
Dave
3.65
8
7
10
6
9
.34
.47
.72
.91
.03
4
5
5
DCS
UNC
3
8
7
10
7
9
.67
.53
. 10
20.
88
112 UNC/
- Catnanuf .
.73 0.78
.48
.61
.92
.03
.18
0.
0.
1 .
1.
0.
0.
93
88
10
02
93
94
-------
Low MW Tank HC
HIGH CONC DCS 109
manuf. UNC-Dave U-Dave/Spread/ Spread
Carle I
Carle II
Carle ICarle III
ethylene
ethane
propy lene
propane
a 1 lene
1 -butene
butane
c is-2-butene
trans-2-butene
1 , 3-butad lene
10.0
10.JBT
21.9
21 .8
15.0
19.6
10.8
20.0
19.6
20.0
manur .
9.32 0.93
9.84 0.98
20.70
20.50
17.60
18.80
10.40
18.90
19.40
17.20
0.95
0.94
1 .17
0.96
0.96
0.95
0.99
0.86
Isopentane 18.1 17.40 0.96
pentane 20.0 19.70 0.99
2-methyl-l ,3-butadlene 17.1 41.40 2.42
octane 104.2 104.00 1.00
ave 0.97
Low MW Tank
HIGH CONC
manuf. UNC-Spd UNC/
manuf .
ethylene
ethane
propylene
propane
al lene
1 -butene
butane
c 1 s-2-butene
trans-2-butene
1 , 3-butad lene
t sopentane
pentane
2-methyl-l ,3-butadlene
octane
not Including 2-me-l,3
10.0
10.0
21 .9
21.8
15.0
19.6
10.8
20.0
19.6
20.0
18.1
20.0
17.1
104.2
-butad lene
9.25
9.59
20.99
20.80
14.70
18.93
10.48
19.05
19.54
17.29
17.72
19.96
41 .65
96.17
ave
0.92
0.96
0.96
0.95
0.98
0.97
0.97
0.95
1 .00
0.86
0.98
1.00
2.44
0.92
0.96
nianuT .
0.92
0.96
0.96
0.95
0.98
0.97
0.97
0.95
1.00
0.86
onoeT.
9.25
9.59
2^.99
2ii . iiJtl
! ' . 7.0'
20.99
20.80 sum >
1 G ii" }
10!48 >
19.05 sum )
19.54 >
17.29 >
0.98 17.72
1.00 19.96
2.44 41.65
0.92 96.17
0.96
LOW CONC
THC UNC
x 0.073
0.675
0.700
1 .532
1 .519
1.073
1 .382
0.765
1.390
1 . 426
1 .262
1 .294
1 .457
3.041
7.020
0.073
Dave
0.634
0.644
1 .350
1 .310
.240
.300
0.713
.310
.350
. 180
1 .210
1 .340
2.520
6 .000
low/hi
0.069
0.067
0.064
0.063
0.084
0.069
0.068
0.069
0.069
0.068
0.068
0.067
0.061
0.071
0.068
86.34
17.72
19.96
41 .65
92.08
DCS 111
UNC
Spread
0.629
0.628
1 .458
1 .320
. 148
.305
0.716
.317
.356
.184
1.218
1.372
2.586
6.859
9.25
9.59
18.84
}
32.80 sum > 59.20 58.1
89.56
low/hi low/hi/
{ 18.93 18.9
{ 15. <3 10.4
85.29 l-j.JJS 10.2
{ 19.54 19.3
{ 17.29 17.6
20.54 17.5
29.64 19.3
106.87
Carle II ph der
avg low/hi 5/30 7/13
0.068
0.065
0.069
0.063
0.078
0.069
0.063
0.069
0.069
0.068
0.069
0.069
0.062
0.071
0.069
0.99
0.95
1.01
0.93
.14
.01
.00
.01
.01
.00
1 .00
1.00
0.91
1 .04
1 .00
-0.01 0.667 0.592
-0.05 0.610 0.645
0.01
-0.07
0.14
0.01
.00
0.01
0.01
.00
.00
.00
-0.09
0.04
sd
0.05
avg
0.629
0.628
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
-------
High MW Tank
Carle I
Carle II
Carle III
o>
acetylene
Isobutane
1-pentene
benzene*
toluene*
ethylbenzene*
m-xylene*
o-xylene*
* AVE NOT USED
toluene
m-xylene
o-xylene
benzene*
toluene*
ethylbenzene*
m-xylene*
o-xylene*
acetylene
t sobutane
1-pentene
benzene**
toluene*
ethylbenzene**
m-xylene*
o-xylene*
Sprd av
15.49
11.08
15.22
19.49
20.46
21 .38
22.88
21.29
manuf .
21.10
22.30
22.40
5/30 7/18 8/14-T
13.28
9.26
14.93 15.52 15.47
20.50 18.93 19.14
20.61 19.86 20.89
12.31 19.36 20.37
NA 19.02 19.62
ave/ Meas .
Dave manuf . ave
20.40 0.96 20.22
23.50 1.05 23.41
22.50 1.02 22.95
8/14-L 5/30 7/18 8/14-T 8/14-L 5/30 7/18
17.704
11.646 11.46
16.43 14.819
17.228 16.62 18.55 15.70 22.413 21.71
20.32 18.691 18.29 19.74 16.70 23.805 23.15
22.19 19.167 17.87 20.51' 17.35 25.074 24.29
21.63 22.057 20.50 24.86 21.04 25.709 25.72
20.84 20.228 18.70 21.00 17.77 24.603 24.10
Carle I Carle ICarle III
20.65 19.48 20.54
22.20 26.79 24.53 *CARLE II M-XYLENE NOT USED
22.81 22.88 23.15
8/14-T 8/14-L
11.77 11.27
15.02 14.38
22.32 21.38
23.65 22.65
24.91 23.86
25.73 24.65
24.70 23.66
Car IIIMeas. ave/Car III *
ave
21.96
23.32
24.53
25.45
24.27
ave res
High MW
manuf .
14.60
12.00
15.20
20.00
21.10
23.80
22.30
22.40
Car III ave res.
20.00
0.87
22.34
0.92
0.95
0.91
Tank
UNC RTI-DavSprd av
RTI-Dave manuf. manuf.
12.90 0.88 1.06
11.30 0.94 0.92
14.60 0.96 1.00
20.00 1.00 1.00
20.40 0.97 0.96
22.50 0.95 0.94
23.50 1.05 1.05
22.50 1.00 1.02
0.97 0.99
,
DCS 108 Carle III
Sprd ave Mo's 83
15.49
11.08 12.20
15.22 15.50
20.00 22.70
20.22 23.90
22.34 24.30
23.41 25.30
22.95. 23.50
-------
80.45 24.71
1sopentane
area cone.
41 .40
41 .60
41 .50
12.74
Carle I cal 5/30
ppmC/area =
0.3071
Carle I cal 5/30
ppmC/area =
0.3071
Carle II cal 5/30
ppmC/area =
0.7316
Carle II cal 5/30
ppmC/area =
0.7316
Carle III cal 5/30
ppmC/area =
4.0938
Isopentane
area cone.
Carle III cal 5/30 3.03
ppmC/area = 3.01
4.0938
3.02 12.36
D1 methylpentane n-Octane Ethylene
area cone. area cone. area
79.10 117.80 8.50
81.80 130.G0 8.52
n-Butane cls-butene
cone. area cone. area cone.
26.30 23.90
26.40 24.10
7.37
124. 15 38. 13
2-me-l -butene
area cone.
39.90
40.40
40.15 12.33
n-Octane
area cone .
45. G0
48.80
47.20 34,53
n-Octane
area cone.
10.40
11.20
10. C!0 44.21
2-mc- 1 -butene
area cone.
3.21
3.22
0.51
propane
a i- ea
19.40
19.70
19.55
Ethy lene
area
3.43
3.43
3.43
propane
area
5.98
6.00
5.99
2.61
cone .
6.00
cone .
2.51
cone .
' 4.38
26.35 8.09
propy lene
area cont.
18.20
18.40
18.30 5.62
propy lene
area cone.
6.06
6.02
6.04 4.42
n-Dutane
area cone .
1 .90
1 .89
1.90 7.76
24.00
c 1 s-butene
area
1 .78
1 .77
1 .78
propane/propylene
area cone.
2.85
2.83
cone.
7.27
3.22 13.16
2.84 11.63
-------
d 1 methyl pen tane
n-octane
ethylene
n-butane
c 1 s-2-butene
t sopentane
2 -me- 1-butene
propane
propy lene
propylene+propane
propylene- RTI
ethylene- RTI
N
1
1
1
1
1
1
6
N
1
1
1
0
1
1
manuf .
28.700
27.680
2.680
8.000
6.800
1 9 . 500
15.450
6.060
5.430
11 .49
X
1
26
27
104
749
2078
X
1
29
106
740
1180
2225
DCS #1
19.190
20.223
2.718
7.979
8.000
12.971
16.300
6.119
6.120
12.239
Y
14.58
14.82
14.34
14.94
14.79
14.4
Y
19.40
19.20
19.80
16.00
20.00
1 9 . 00
Car I
24.706
38.127
2.613
8.092
7.370
12.745
12.330
6.004
5.620
11.623
N*X
1
26
27
104
749
2078
2985
YBAIi :
XBAIl:
N*X
1
29
106
0
1100
2225
Car II
34.530
2.509
4.382
4.419
N*Y
14.58
14.82
14.34
14.94
14.79
14 .40
87.87
14.645
497.5
N*Y
19.40
19.20
19.00
0.00
Zii . 00
19.00
Car III
44.213
7.758
7.267
12.363
13.162
-------
Official Calibration Sources
168
-------
OFFICIAL CALIBRATION SOU1CSS
ID
NVM
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
37
37
37
37
37
37
37
37
37
33
39
40
41
42
43
44
45
46
47
43
49
50
51
52
53
53
53
53
53
53
OSSC
TOLUENEINO
TOLUENEINO
TOLUENEINO
MXYLENEINO'
TOLUENEINO
TOLUENEINO
124TMBINOB
124TMBINOR
MXYLENEINO
MXYLENEINJ
OXYLENEINO
OXYLENEINO
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
TOLUENEINO
ISOPENTIMO
124TM3INO
TOLUENEINO
ISOPENTIrlO
124TMBINO
TOLUENEINO
OXYLENEINO
METOLINO
224TMPINO
ISOPENINO
OXYLENEINO
TOLUENEINO
DETHKETINO
DETHKETINO
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
SDATE
18-Auq-1933
2S-A;j!J-1983
22-Aug-1983
22-Aug-1983
6-Oct-1983
6-Oct-1983
6-Oct-1983
6-Oct-1983
6-Oct-1933
S-Oct-1933
6-Oct-1983
6-Oct-1983
23-Ou1-1983
23-Oul-1983
23-Oul-1933
23-Oul-1933
23-Ou 1-1-J83
23-Oul-1983
23-Ou1-1933
23-Oui-1983
23-Oui-1983
23-Oui-19Q3
25-Ou1-1983
25-Ou1-1933
25-Oul-1983
25-Oul-1983
25-Ou1-1933
25-Ou1-1983
8-Aug-1983
8-Aug-1983
9-Aug-1983
19-Aug-1983
19-Aug-1983
l-Aug-1983
l-Aug-1983
20-Oun-1983
20-Oun-1983
9-Sep-1983
9-Sep-1983
9-Sep-1983
9-Sep-1983
9-Sep-1983
9-Sep-1983
SER
MUM
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
TOLUENE
TOLUENE
TOLUENE
M-XYLENE
TOLUENE
TOLUENE
1,2,4-TRIMETHYLBENZE
NE
1 ,2,4-TRIMETHYLBENZE
NE
M-XYLENE
M-XYLENE
0-XYLENE
0-XYLENE
2-METHYLPENTANE
2 ,'3-DIMETMYLPENTANE
METHYLCYCLOHEXANE
TOLUENE
ETHYLBENZEME
M-XYLENE
N-PROPYLBEMZENE
2,2,4-TRIHETHYLPENTA
NE
TERT-BUTYLBENZENE
1 ,3,5-TRIMETMYLBENZG
NE
TOLUENE
ISOPENTANE
1 ,2,4-TRIMETHYLBENZE
NE
TOLUENE
ISOPENTANE
1 ,2,4-TRIMETHYLBENZE
NE
TOLUENE
0-XYLENE
M-ETHYLTOLUENE
2,2,4-TRIMETHYLPENTA
NE
ISOPENTANE
0-XYLENE
TOLUENE
3-PENTANONE
3-PENTANONE
BENZENE
P-XYLENE
0-XYLENE
1,2,4-TRIMETHYLBENZE
NE
SEC-BUTYLBENZENE
M-ETHYLTOLUENE
25-?-
Pag-
STAT-1D
cone
0.795
0.788
ft cTf - ;
.- /
0 . j ,: ..,
0.504
0.381
0.381
0.233
0.233
0. 180
0.180
0.111
0.111
1 .000
1 .000
1 .000
1 . 000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
0.601
0. 166
0.369
0.601
0. 166
0.369
9. 125
6.083
4.000
4.053
4.059
2.699
4.553
1 .010
1.010
1 .000
1 .000
1 .000
1 .000
1 .000
1.000
p-1985
?
ACT'JAL
COMC
0.795
0.788
U . 00.3
ft ~t rt <~t
'.' . '. '
Ili .SIS;
0.381
0.381
0.233
0.233
0. 180
0. 180
0. Ill
0.111
1 .000
1 .000
1 . 000
1 . 000
1 .000
1 . 000
1 . 000
1 . 000
1 .000
1 . 000
0.601
0. 1Gb
0.369
0.6:31
0. 135
0.369
9. 125
6.083
4.000
4.053
4.059
2.699
4.529
1 .010
1 .010
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
170
-------
OFFICIAL CALIBRATION SOURCES
ID
NUM
53
53
53
54
54
55
55
55
55
55
55
55
55
55
56
56
56
56
56
56
56
56
56
57
57
57
57
57
57
57
57
57
58
58
58
58
58
58
58
58
58
59
59
59
DESC
PE900CLMXS
PE900CLMXS
PE900CLMXS
NEW83COCH4
NEW83COCH4-
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLi''XS
PE900CLMXS
PE900CLHXS
PE900CLMX5
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE90CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMX4
PE900CLMXS
PE900CLMXS
SDATE
9-Sep-1983
9-Sep-1983
9-Sep-1983
30-Sep-1933
30-Sep-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
3-Oct-1983
4-Sep-1983
4-Sep-1983
4-Sep-1983
4-Sep-1983
4-Sep-l983
4-Sep-1933
4-Sep-1983
4-Sep-1933
4-Sep-1333
24-Jun-1983
24-Oun-1983
24-Oun-1983
24-Oun-1983
24-Oun-1983
24-Jun-19S3
24-Jun-1983
24-Oun-1983
24-Oun-1983
30-Oun-1983
30-Oun-1983
30-Oun-1983
30-Oun-1983
30-Oun-1983
30-Oun-1983
30-Oun-1983
30-Oun-1983
30-Jun-1983
3-Jun-1983
3-Oun-1983
3-Jun-1983
SER
NUM
ISOPROPYLBENZENE
1 ,2,3-TRIMETHYLBENZE
NE
BUTYLBENZENE
METHANE
CO
BENZENE
P-XYLENE
0-XYLENE
1 ,2,4-TRIMETHYLBENZE
NE
SEC-BUTYLBENZENE
M-ETHYLTOLUENE
ISOPROPYLBENZENE
1 ,2,3-TRIMETHYLBENZE
N.E
' .BUTYLBENZENE
BE-NZENE
P-XYLENE
0-XYLENE
1 ,2,4-TRIMETHYLBENZE
NE
SEC-3UTYLBENZENE
M-ETHYLTOLUENE
ISOPROPYLBEHZENE
1,2,3-TRIMETHYLBENZE
NE
BUTYLBENZENE
BENZENE
P-XYLENE
0-XYLENE
1,2,4-TRIMETHYLBENZE
NE
SEC-BUTYLBENZENE
M-ETHYLTOLUENE
ISOPROPYLBENZENE
1,2,3-TRIMETHYLBENZE
NE
BUTYLBENZENE
BENZENE
P-XYLENE
0-XYLENE
1,2,4-TRIMETHYLBENZE
NE
SEC-BUTYLBENZENE
M-ETHYLTOLUENE
ISOPROPYLBENZENE
1,2,3-TRIMETHYLBENZE
NE
BUTYLBENZENE
2-METHYLPENTANE
2,3-DIMETHYLPENTANE
METHYLCYCLOHEXANE
25-Sep
Page 3
STATED
CONC
1 .000
1 .000
1 .000
4.934
5.114
1 .000
1 .000
1 .000
1 .000
1.000
1.000
1.000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 . 000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1.000
1 .000
1.000
1 .000
1 .000
-1S35
ACTUAL
CONC
1 .000
1 .000
1 .000
4 .934
5.114
1 .000
1 .000
1 . 000
1 .000
1.000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 . 000
1 . 000
1 . 000
1 . 00;J
1 . 000
1 . 00:J
1 . 000
1 . 000
1 . 000
1 .000
1 .000
1 ..'J0'J
1 . 000
1 .000
1 .000
1 . 000
1 . 000
1 . 000
1 . 000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1.000
171
-------
OFFICIAL CALIBRATION SOURCES
25-Sep-1985
Pa-jo 4
ID
NUM
59
59
59
59
59
59
59
60
60
50
60
60
60
60
60
60
61
61
61
61
1J1
61
61
61
61
62
62
62
62
62
62
62
62
62
62
63
63
63
63
63
63
63
63
DESC
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXC
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMXS
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMXS
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMXS
PE900CLMX5
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
SDATE
3-Oun-l983
3-Oun-1983
3-Oun-1983
3-Oun-1933
3-Oun-1983
3-Jun-1983
3-Oun-1983
6-Ou1-1933
6-Oul-1933
6-Ou1-1983
6-Oul-1983
6-Oul-1983
6-Oul-1983
6-Ju1-1983
6-Oul-1933
6-OuT-1983
18-0ul-1933
18-Jul-1983
18-Jul-1983
13-Ou 1-1983
lS-Oul-1933
13-Ou 1-1983
13-Oul- 1983
13--0ul-1933
18-Jul-1983
18-Cul-1983
18-0ul-1983
lS-Oul-1983
18-0ul-1983
18-0ul-1983
13-0ul-1983
18-Jul-1983
18-0ul-1983
18-0ul-1983
18-0ul-1983
23-Oul-1983
23-Oul-1983
23-Ou1-1983
23-Ju1-1983
23-Oul-1983
23-Oul-1983
23-Oul-1983
23-Jul-1983
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
MA
NA
MA
NA
NA
SER
NUM
STATED ACTUAL
COMC CONC
TOLUENE
ETHYLBENZENE
M-XYLENE
N-PROPYLBENZENE
2,2.4-TRIMETHYLPENTA
NE
TERT-BUTYLBENZENE
1 .3.S-TRIMETHYLBENZE
NE
BENZENE
P-XYLENE
0-XYLENE
1 ,2,4-TRIMETHYLBENZE
NE
SEC-BUTYLGENZENE
M-ETHYLTOLUENE
ISOPROPYLBEMZENE
1,2,3-TRIMETHYLBENZE
NE
BUTYLEENZEME
BENZENE
P-XYLENE
0-XYLENE
1 , 2,4-TRIMETHYLBEHZE
NE
SEC-BUTYLGEMZENE
M-ETMYLTOLUGNE
ISOPROPYLBEMZENE
1 ,2,3-TRIMETHYLBENZE
NE
BUTYLBENZENH
2-METHYLPEMTANE
2 ,3-DIMETHYLPENTAMG
METHYLCYCLOHEXANE
TOLUENE
ETHYLBENZENE
M-XYLENE
N-PROPYLBENZENE
2,2,4-TRIMETHYLPENTA
NE
TERT-BUTYLBENZENE
1 ,3,5-TRIMETHYLBENZE
NE
BENZENE
P-XYLENE
0-XYLENE
1,2,4-TRIMETHYLBENZE
NE .
SEC-BUTYLBENZENE
M-ETHYLTOLUENE
ISOPROPYLBENZENE
1 ,2,3-TRIMETHYLBENZE
NE
1 .000
1 . 000
1 .000
1 .000
1 . 000
1 .000
1 .000
:NZE
:NZE
i .000
i .000
1 .000
1 . 000
i .000
i . 000
1 . 000
1 . 000
1 .000
1 . 000
1 . 000
1 .000
! . 000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 . 00,'J
1 . 000
1 .000
1 . 000
1 . 000
1 . 000
1 .000
1 .000
1 . 000
1 .000
1 .000
1.000
1 . 000 .
1 .000
1 . 000
1 . 005
1 . 000
1 . 00'J
1 . 00.7
1 . 00.'?
1 . 006'
1 . 00.'?
1 . 00.0
1 . 00/7
1 . 00,0
1 . 00/1
1 . 000
1 . 000
1 .000
1 .000
1 .000
1 .000
1 . 000
1 .005
1 . 000
1 . 000
1 .000
1 .000
1 . 000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 . 000
1 .000
1 .000
1 .000
1 . 000
172
-------
OFFICIAL CALIBRATION SOURCES
25-Sep-1985
Page 5
ID
NUM
G3
64
64
64
64
64
64
64
64
64
64
65
65
65
65
65
65
65
65
55
66
57
63
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
DESC
PE900CLMX5
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX4
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
PE900CLMX5
HCOHINO
HCOHIMO
HCOHINJ
HCOHINO
HCOHINO
HCOHINO
HCOHINO
HCOHINO
HCOHINO
HCOHINO
N02ASPAN
N02ASPAN
N02ASPAN
N02ASPAN
no2aspan
no2aspan
no2aspan
no2aspan
no2aspan
no2aspan
no2aspan
no2aspan
no2aspan
no2aspan
noZaspan
no2aspan
noZaspan
N02ASPAN
N046PPMCAL
SDATE
23-Oul-1983
23-Oul-1983
23-Oul-1983
23-Oul-1933
23-Oul-1983
23-Oul-1983
' 23-Oul-1983
23-Oul-1983
23-Oul-1983
23-OU1-1983
23-Oul-1983
2-Aug-1983
2-Aug-1983
2-Aug-1983
2-Aug-1983
2-Aug-1983
2-Aug-1933
2-Aug-1383
2-Aug-1983
2-Aug-1983
6-Oun-1983
lS-Oun-1982
24-Nov-1983
18-0un-1983
12-0u1-1983
12-Aug-1983
23-Oul-1983
29-Oul-1983
13-Aug-1983
19-Sep-1983
10-0un-1983
18-0un-1983
13-0ul-1983
17-0u1-1983
18-0ul-1983
20-Jul-1983
21-0ul-1983
26-Oul-1983
29-Oul-1983
31-0ul-1983
4-Aug-1983
13-Aug-1983
5-Sep-1983
7-Sep-1983
18-Sep-1983
23-Sep-1983
2-Oct-1983
14-0ct-1983
22-Oun-1934
SER
NUM
STATED ACTUAL
CONC CONC
SX13759
BUTYLBENZENE
2-METHYIPGNTANE
2,3-DIMETHYLPENTANE
METHYLCVCLOHEXANE
TOLUENE
ETMYLBENZEME
M-XYLENE
N-PROPYLBENZENE
2,2,4-TRIMETHYLPENTA
NE
TERT-BUTYL8ENZENE
1 ,3,S-TRIMETHYLBENZE
NE
BENZENE
P-XYLENE
0-XYLENE
1,2,4-TRIMETHYLBENZE
NE
SEC-BUTYLBENZENE
M-ETHYLTOLUENE
ISOPROPYLEENZENE
1 ,2,3-TRIH.ETHYLBEHZE
NE
BUTYLBENZENG
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
NO.
1 .000
1 .000
1 .000
1 . 000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1.000
1 .000
1 .000
1.000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .072
1 .000
1 .024
1 .041
1 .029
1 .032
1 .047
1 .026
1.016
1 .026
0.244
0.152
0.037
0.071
0.055
0.081
0.061
0.025
0.074
0.059
0.036
0.074
0. 105
0.096
0. 114
0. 100
0.200
0.110
46.000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1 .000
1.000
1 .000
1 .000
1 . 00.0
1 . 000
1 .072
1 .2133
1 .024
1 .04 1
1 .029
0.000
1 .047
1 .025
1 . :J 1 5
1 .02G
0.24J
0. 152
0.037
0.071
0.055
0.051
0.061
0.025
0.074
0.059
0.036
0.074
0.105
0.096
0. 114
0. 100
0.200
0. 1 10
52.600
173
-------
ID
NUM
95
96
97
98
99
100
101
102
103
104
105
106
107
108
108
108
108
108
108
103
103
109
109
109
109
109
109
109
109
109
109
109
109
109
109
110
110
1 10
110
110
110
110
1 10
111
111
111
111
111
111
111
111
111
DESC
MXYLENEINO
TOLUENEIMO
TOLUENE
M-XYLENE
0-XYLENE
TOLUENE
M-XYLENE
FORMALDEHY
FORMALDEHY
HCHOINOB
HCHOINOR
HCHOINJRAF
HCHOINORMO
HIMWHICONC
HIMWHICONC
HIMWHICONC
HIMWHICONC
HIMWHICONC
HIMWHICONC
HIMWHICOMC
HiMWHICOMC
LOMWHICOMC
LCMWHICOMC
LOMWHICOMC
LOMWHICO;!-:
LOMWHICOfiO
LOMWHICO;IC
LOMWHICOMC
LOMWHICCfIC
LOMWHICOMC
LOMWHICOMC
LOMWHICOMC
LOMWHICOMC
LOMWHICOMC
LOMWHICOMC
SCOTT84
SCOTT84
SCOTT84
SCOTT84
SCOTT84
SCOTT84
SCOTT84
SCOTT84
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
LOMWLOCONC
SDATE
25-Oun-
25-Jun-
27-Oun-
26-Jun-
26-Oun-
19-Jun-
19-Oun-
10-Oul-
10-Oul-
1 1-Oul-
11-Jul-
27-Jul-
27-Jul-
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
NA
NA
NA
NA
OFFICIAL CALIBRATION SOURCES
SER
NUM
M-XYLENE
TOLUENE
TOLUENE
M-XYLENE
0-XYLENE
TOLUENE
M-XYLENE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
ACETYLENE
ISOBUTANE
1-PENTENE
BENZENE
TOLUENE
ETHYLBGNZENG
M-XYLENE
0-XYLENE
ETHYLENE
ETHANE
PROPYLEME
PROPANE
PROPADIENE
1-BUTEME
N-BUTANE
1 , 3-BUTADIElNE
TRAN3-2-BUTENE
CIS-2-BUTGME
ISOPENTANE
N-PENTANE
N-OCTANE
2-METHYL-l,3-BUTADIE
NE
ETHYLENE
PROPYLENE
PROPANE
CIS-2-BUTGHG
ISOPENTANE
N-PENTANE
TOLUENE
2-METHYL-2-BUTENE
ETHYLENE
ETHANE
PROPYLENE
PROPANE
PROPADIENG
1-BUTENE
N-BUTANE
1,3-BUTADIENE
TRANS-2-BUTENE
25-Sep
Pag* S
STATED
COMC
3.000
6.000
5 . ;700
3.000
3.000
4 . 000
3.000
0.500
1 .060
1 . 000
1 .000
1 .000
0.200
1 4 . 600
1 2 . 000
15. 175
19.956
21 .056
23.734
22.272
22.432
10.004
1 :3 . 000
21 .9.06
2! . 3'34
15. .000
19 .500
10.300
20 . 000
20 . 000
2.0 . 000
13. 135
19.956
* * ***;!
17.055
0.600
0.900
0.900
1 .000
1 .250
1 .250
2.300
1 .250
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
-1985
ACTUAL
CONC
2.963
5.895
4 .903
2.973
2.653
3.929
2.967
0.542
1 .035
0.946
0.946
0.946
0. 189
15.490
1 1 .080
1 5 . 200
19.910
20.300
22.250
23.59.'J
22.95.J
9.25/J
9.59/T
2 0.99. 'J
Z.V.SQ'J
1 4 . 7.0.'J
1 3 . 33.'J
15.48.-;
17.29/J
19 . 540
19 .050
17 . 7C/J
19 . 96,'j
96 . 17''
41 .65.J
0.711
1.123
0..972
1.11 .'!
1 . 143
1 .347
3.250
1 .390
0.629
0.628
1 .458
1 .320
1 . 148
1 .305
0.716
1 . 184
1 .356
174
-------
OFFICIAL CALIBRATION SOURCES
ID
HUM
111
1 i 1
111
111
1 1 1
112
112
112
112
112
112
112
113
114
115
116
117
113
113
113
113
1 13
113
1 19
120
121
122
123
123
123
123
123
123
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
DESC
LOMWLOCONC
LOMWLOCONC
LOMWLOCOMC
LOMWLOCONC
LOMWLOCOMC.
NEWAIRC084
NEWAIRC084
NEWAIRC084
NEWAIRC084
NEWAIRC084
NEWAIRC084
NEWAIRC034
PROPYL741
TOLUENE211
HCHOINO
TOLUENEINO
MXYLENEINO
AROMINO
AROMINO
AROMINO
AROMIMO
ARCH I HO
AROMIMJ
TMPENTiNO
BENZENE I NO
RTIETHYL
RTIPROPYL
HIMWLIQIMJ
HIMWLIQINO
HIMWLIQIMJ
HIMWLIQIMJ
HIMWLIQIilJ
HIMWLIQiriJ
HIMWLIQIMO
M-XYLENE
HCHOINO
HCHOINO
MEOHINO
MEOHINO
MEOHINO
MEOHINO
MEOHINO
MEOHINO
HCHOINO
HCHOINO
HCHOINO
HCHOINO
HCHOINO
HCHOINO
85COCH4CAL
SDATE
SER
NUM
N415664
31-0u1-1984
28-Oun-1984
28-Oun-1934
16-0ul-1984
lS-Oui-1984
16-0ul-1984
lS-Oul-1984
16-0u1-1934
lG-Jul-1984
27-Ou 1-1980
27-Ou1-1984
14-Aug-1934
14-Aug-1934
14-Aug-1984
14-Aug-1984
14-Aug-19S4
14-Aug-1984
14-Aug-1984
27-Oun-1984
13-0ul-1984
13-0ul-1984
24-Oul-1984
3-Aug-1984
4-Aug-1984
7-Aug-1984
8-Aug-1984
9-Aug-1984
7-Oct-1984
7-Oct-1984
9-Oct-1984
9-Oct-1984
16-0ct-1984
16-0ct-1984
3-Apr-1985
NA
NA
NA
NA
NA
NA
NA
MA
NA
NA
NA
DFB
DFB
DFB
DFB
DFB
DFB
DFB
LMQ
LMQ
LMQ
OBS
DFB
ORA
DFB
ORA
LMQ
NA
NA
NA
NA
NA
NA
AAL-
15140
CIS-2-BUTENE
ISOPENTANE
N-PENTANE
N-OCTANE
2-METHYL-l,3-BUTADIE
NE
ETHYLENE
PROPYLENE
PROPANE
TRANS-2-BUTENE
ISOPENTANE
TOLUENE
PROPYLENE/PROPANE
PROPYLENE
TOLUENE
FORMALDEHYDE
TOLUENE
M-XYLENE
BENZENE
TOLUENE
P-XYLENE
M-XYLENE
0-XYLENE
STYRENE
2,2.4-TRIMETHYLPENTA
NE
BENZENE
ETHYLENE
PROPYLENE
N-PENTANE
2-METHYLPENTANE
2,3-DIMETMYLPENTANE
TOLUENE
M-XYLENE
0-XYLENE
2,2,4-TRIMETHYLPENTA
NE
M-XYLENE
FORMALDEHYDE
FORMALDEHYDE
METHANOL
METHANOL
METHANOL
METHANOL
METHANOL
METHANOL
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
FORMALDEHYDE
METHANE
2S-Sep-1905
Page 7
STATED
CONC
0.000
3 . 000
0 . 000
0.000
0.000
4.800
9. 150
8.610
6.320
9.850
9.940
17 .760
7.410
14.770
1 .000
6.000
3 . 000
2.000
2 . 000
2 . 000
2 . 000
2 . 000
2.000
1 . 000
1 . 000
1 9 . 000
1 4 . 400
3 . 000
3.000
3 . 000
3 . 000
3 . 000
2.600
3 . 000.
2.000
1 .000
1 .000
0.900
0.900
0.300
0.300
0.790
0.263
1 .000
0.500
1 .000
1 .000
0.250
0.500
1 .907
ACTUAL
CONC
1.317
1.213
1 .372
6.859
2.586
3. 730
8.480
7.610
7.030
9.180
10.920
16.090
7.410
14.770
0.946
5.977
3.009
2 . 000
2 . 000
2 . 00.J
2 . 00.'J
2 . 00.U
2 .00'J
1 . 00J
1 . 000
19.43,'J
14 .650
2.383
3. 161
.99:7
. 93 a
.975
.663
2.992
1 .973
1 !033
1 .033
0.891
0.397
0.321
0.300
0.787
0.263
0.918
0.464
0.920
0.920
0.232
0.465
1 .907
175
-------
OFFICIAL CALIBRATION 30V.CES
Pag , 3
ID
NUM
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
150
151
152
152
162
152
163
163
163
163
164
165
165
165
165
165
165
166
166
166
166
166
166
167
168
169
170
DESC
85COCH4CAL
MEOHINO
ACETALDINO
ACETALDINO
MEOHINO
MEOHINO
ACETALDINO
ACETALDINO
MEOHINO
HCHOINO
N02ASPAN
N02ASPAN
N02ASPAN
N02ASPAN
N02ASPAN
N02ASPAN
N02ASPAN
N02ASPAN
M02ASPAM
M02ASPAM
HCHOINO
ACETINO
ACETINO
AROMINO
A KOMI NO
AROMINO
AROMINO
AROMINO
AROMINO
AROMINO
AROMINO
HCHOIMO
AROMINO
AROMINO
AROMINO
AROMINO
AROMINO
AORMINO
AROMINO
AROMINO
AROMINO
AROMINO
AROMINO
RTIPROPYLE
RTIETHYLEN
AIRCONO
SCOTTNO
SDATE
3-Apr-1935
25-Aug-1934
27-Aug-1984
27-Aug-1984
l-Sep-1984
3-Sep-1984
16-Sep-1984
16-Sep-1984
17-Sep-1984
25-Sep-1984
20-Aug-1984
20-Aug-1984
18-Sep-1934
18-Sep-1984
3-Oct-1984
3-Oct-1984
4-Oct-1984
1 l-Oct-1984
30-Aug-1984
6-Oul-1984
S-Sep-1985
lS-Oct-1984
15-0ct-1984
14-May-1985
14-Mc:y-1985
14-Mav-1935
14-Mciy-1385
17-May-1985
17-May-1985
17-May-1985
17-May-1935
20-May-1935
22-May-1985
22-May-1985
22-May-1935
22-May-1985-
22-May-1985
22-May-1985
24-May-1985
24-May-1985
24-May-1985
24-May-1985
24-May-1985
3-Oun-1985
3-Oun-1985
SER
NUM
AAL-15140
ORA
KGS
KGS
KGS
LMQ
DFB
NA
NA
NA
NA
NA
NA
NA
NA
NA
MA
MA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
1122590
BAL177
CO
METHANOL
ACETALDEHYDE
ACETALDEHYDE
METHANOL
METHANOL
ACETALDEHYDE
ACETALDEHYDE
METHANOL
FORMALDEHYDE
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
PAN
FORMALDEHYDE
ACETALDEHYOE
ACETALDEHYOil
BENZENE
TOLUENE
M-XYLENE
Z ,2 ,4-TRIMETHYLPENTA
NE
BENZENE
TOLUENE
M-XYLENE
2, 2,4-TRIMETHYLPENTA
NE
FORMALDEHYDE
BENZENE
TOLUENE
M-XYLENE
0-XYLENE
1 ,2,4-TRIMETHYLBENZE
NE
2, 2,4-TRIMETHYLPENTA
NE
BENZENE
TOLUENE
M-XYLENE
0-XYLENE
2,2 ,4-TRIMETHYLP.ENTA
NE
PROPYLENE
ETHYLENE
NO
NO
STATED
CONC
3.950
0 . 300
1 . /J0.CT
i . .0'.:::;
1 .000
0.266
1 .000
1 .000
0.600
1 . 000
0. 160
S3. 153
0. 191
0. 131
S3. 166
S3. 165
0. 164
0. 166
0. 169
0.207
1 . 000
1 . 000
1 . 000 -
0.500
1 . 000
2 . 000
2 . 005
0.500
1 . 00.GT
2 . 000
3.000
1 .000
1 .500
2.000
4.000
5.000
5.000
1 .000
S3 .000
1 .500
2.000
4.000
5.000
1 .000
14.400
1 9 . 000
52.400
52.600
ACTUAL
CONC
3.950
0.292
! . 000
:. . 00.J
0.970
0.264
1 . 000
1 .000
0.572
0.932
0. 160
0. 153
0. 191
0. 131
0. 166
0.165
0. 164
0.166
0. 169
0.207
0 . 9 b '-
1 . 000
1 . 00.-J
0. 45X»
1 .005
1.99 ':!
1 . 1 L "
0 . 5 a 5
1 . J6 »
2. 105
2.349
1 .046
1 .659
2. 161
4.274
5.543
5-379
0.795
0.000
1 .613
2. 101
4.155
5.389
0.773
14 .645
19.480
52.450
52.600
176
-------
OFFICIAL CALIBRATION SOURCES 25-S<5p-1985
Page 9
ID SER STATED ACTUAL
NUM DESC SDATE NUM CONC CONC
171 LOWNOAUTO NO 0.000 0.28"
172 LOWNOAUT02 NO 0.693 0.665
177
-------
CALANA Plots and Reports
178
-------
. 40]
V i
l2 0.20^;
"TO h
GC
Carle I
PROPYLENE
FPMC/IN Mttan = 2.30
24-J1
0.30
a OQI i I i I i I i I ! I i I i i i I i ! -. I ! ! i I i ! i I i t i 1 i i ft *,
100 160 170 180 190 200 210 220 230 240 250 260 270 280 2SO 300 310 320'
Julian Date, 1984
For the species F'RDPYLENL and the instrument. Carle I
GC:
Beg i nn i nq Cai date s
Day::
E n d i n g C a 1 d a t a:
Day:
Upper CalFactor Limit
Lower CalFactor Limit
30-MAY-84
150
16-NOV-S4
3 2 0
0.40000
0.OOOOO
The number of data points 62
The number included was 62
The slope of this fit is 1.3522654E-004
with a std dev of 6.6911662E-005
The intercept is 1.66941S4E-001
with a std dev of 1.5341250E-002
The std dev of the fit is 1.8079919E-002
The correlation coeff is
O.25246
rel. dev = 49.5 X
rel . dev = 9. 2 7.
PE = 0.01205
The min, mean, max are 0.13040, 0.19760, 0.25009
The standard deviation is O.01S53 rel. dev =9.4 X
Cal = 0.00014 X (Julian Date - 150) + O. 18723
...press RETURN-key when ready...
179
-------
£_
-»>
IB
Li_
9
10
0
""-! i M 1 i 1 ' I i i ' 1 i I i 1 i 1 M I I i ! i ! i ! i
h Carle I GC
«.27L ISOPENTANE
i PFMC/IN At ten = 2.50
^»
i
3.21r- 04-AUG
h X
^ . J _^, j» ' ^| **"S a»^"»'^~1*X^
" <- ^^r«^B[jjf * ^^ ^\JU "***
j_ T^8»
j
i
L
1
0.051-
(-
ft -J i ! i ! : i i ! i ! i I i I i ! i ! '. i i ! ! ! i I i ! i
!-i i ! j
i
I
i
i
i
-"i
i
i
i
~ -- !
i
i
i
i
]
"|
ill:!
--
«,27
a i< O
3"-1aj
O.ieT1
Hi
ft
c*-
0.11 °
^
0.05
rt Ht1!
160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 32&
Julian Date, 1354
For the species ISOPENTANE and the instrument Carle I GC;
E-ieginning Caldate: 30-MAY-84
Day: 150
Ending Caldate: ' 16-NOV--S4
Day: 320
Up p e r C alFactcsr L i m i t 0.. 3 2 00 0
L!.:; w a r C a 1 F a c t o r L i m i t 0. 0 0 0 0 0
The number of data points 61
The number included was 61
The slope of this fit is -6.274S75E-O05
with a stc! dev of 3.64SOO4OE-OO5 rel . dev = -53.1 :
The' i intercept is 1. 7246965E-001
with a std dev of S.4102O93E-OO3 rel. dev = 4.9 A
The std dev of the fit is 9.72SS922E-OO3 , PE = 0.00649
The correlation coeff is -0.21852
The min, mean, max are O.I3965, O.I5816, 0.1962O
The standard deviation is 0.009S9 rel. dev - 6
Cal = -O.OOOO6 X (Julian Date - 150) -i- 0.16306
180
-------
t-
o
-l->
k-
m
"n i i i i i i
i
0.25J-
r
i
i
i
t
*l
i
1
! I 1 i i I ) i i
m
I
Ai- ,,..
£V~JUL
I | i. | l | , i i ! i | i i i |- ,
Carle II GC
ETHANE
FFMC/IH At ten = 2.00
a
f 26-AUft
JVA f^\
** / ^
' ' ' !
1
J
I
-|
i
i
i
!
~|
V OV
0.25
0.20°
^
-flU
<-*-
. . . <-t
h
J.05
l I i ! l ! i I ! ! i ! i ! i ! i ! ! ! i ! i ! i ! l i l ; ! ! i ift ^^
160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320
Julian Date, 1984
For the species ETHANE and the instrument Carle II SC:
Beginning Caldate:
Day:
E nd ing Ca1da t e:
Day;
Li p p e r C a 1 F a c t o r Li m i t
Lower C a 1 F a c: t o r L i m i t
30-MAY-B4
150
16-NOV-84
320
0.18000
O.OOOOO
The number of data points 35
The number included was 32
The slope of this fit is 5.6237375E-005
with a std dev of 4.3453789E-005
The i n ter c ep t i s 1.34691S8E-OO1
with a std dev of 9.2956360E-O03
The std dev of the fit is 7.19S2245E-QQ3
The correlation coeff is 0.22995
rel . dev = 77.3 7.
rel . dev = 6.9 7.
PE - O.OG480
The min, mean, max are
The standard deviation is
013445, 0.14661, 0.23050
0.00728 rel. dev - 5.0 7.
Cal = O.OG006 X (Julian Date - 150) + 0.14313
181
-------
1.3 P
1.2 j-
^ < r~
£_ "'* j-
o im9 LT
+^ 0.9 '(-
CJ i
OJ 0.3 {
L1- 0.7 h-
rj, 0.5 gl^JA
0.4 r^
0.3 P
9.2 h-
0.1 r
« * r\ \
: \ i | 1 | i i 1 | 1 | 1 j i | ! j i j 1 | ! j 1 1 1 j . 1 |
Carle II GC
PROPANE
FFMC/IN Atten = 2.00
Af
4j
/ir
»t!
- ^_^ ; jj
.' ~ ^s^"^ ~^~~^~~~~*-^. T i.
- -^J^*-^-=^^*=*- - - ~«- ^» ^
₯^
i 1 ! ! 1 1 I 1 1 1 ! ! I 1 1 1 ! 1 ! ! ! ! 1 ! ! i i i ! i
yi.4
"3 i-3
-jl. 2
i 4 .1
-H
i < ,!>
-10.3
!0. 3
^1.7
^ J A ^
^ w. b
q 0.5
i 0.4
i 0.3
j
-d a a
O
il>
Tl
IU
C+-
~i
160 170 180 190 200 210 220 230 240 250 260 270 260 230 300 310 32?
Julian Data, 1984
For the species PROF'ANE and the instrument Carle II GC:
Beginning Caldate:
Ending Caldate:
Day:
Day;
Up per C a1F a c t or Limit
Lower CalFactor Limit
30-MAY-84
150
16-NQV-84
320
0.70000
0.00000
The number of data points 24
The number included was 19
The slope of this fit is 9.3254754E-005
with a std dev of 3.484381&E-G04
The intercept is 5.B478932E-001
with a std dev of 8.2456137E-002
The std dev of the fit is 6.1473938E-OG2
The correlation coeff is 0.06478
rel. dev = 373.6
rel. dev = 14.1 "
F'E = 0.04O98
The min, mean, max are 0.42750, 0.60653, 1.O5930
The standard deviation is 0.05987 rel. dev = 9.9 "/.
Cal = 0.00009 X (Julian Date - 150) + 0'. 59878
182
-------
-'- |_ I j I i I
i>4 E"
1.3 ]-
i.2 '
i.iP
^! H~
^"~ 1 A 1
o l-B tr
** 0.3 h~
V t
nj ».8 £
^ 9.7 P
^^ 9.S r- ^^^
^ ' jpl-MAY
v » *r i~"
0.1 P
fl A I~! ! ! ! !
| 1 j 1 | 1 i i j I | i | 1 1 t ] 1 i I ; 1 i 1 j ;
Carle II GC
PROPYLENE
FrnC/Ii'i ntten = 2.50
*
A
,'!
''«
t Tl }
k ] -^ «i\^? _*"' k*,S *^»« \ J! _ 'ii ^
fc»^g -^ ^i yy ~§^^ ~ *~= - *af^ }
'* * *
! ! ! ! i 1 ! ! ! ! ! ! ! ! i 1 ! 1 ! ! ! ! ! ! ! !
i ' i
H
j
~>
1
H
1
i
H
H
i i~i
1.4
1.3
i.'^3
1 1 -
i .>.
'
C.'3
0.8
e!s
«.fi
J'.3
O.i
tui
Julian Date, 1984
For the species PROPYLENE: and the instrument. Carls II GC;
B e g i n n i n g C a 1 d a t. e 3
Day;
E n d i n g C a 1 d a t e:
Day:
Upper CalFactor Limit
Lower CalFactor Limit
3O-l"!AY-34
150
16-NGV-84
320
0.70000
0.00000
The number o-f data points 63
The number included was 37
The slope o-f this -fit is 3. 2677424E-OQ4
with a std dev o-f 2.03396S5E-004
The intercept is 4.8OBG472E-OOI
with a std dev of 4.5272312E-002
The std dev of the fit is 5.36267Q6E-OG2
The correlation coeff is O.21172
rel. dev = 62.2 ",
rel . dev = 9.4 "/.
PE = 0.03575
The min, mean, max are?
The standard deviation i
0.36388, 0.55264, 0.9756'
0.05433 rel. dev = 9.3 7.
Cal = 0.00033 X (Julian Date - 150) + 0.52982
...press RETURN-key when ready...
183
-------
I f^ t ^* -^
j- L-arie il
4- ETHYLENE
;;-
a^
v^
(v
o_
L?
i
i
a . 21?''
i
9.15;
i
t- A^-jf^d
t ^s-^r-*i
"'"'fci-MAY
i
J^j /Xgfv
~ *
i-
fl .UJ i ! ! ! ! ! i !
FPHC/IN' Attan = 2.?0 i
i
rr-jj A '^i '*(
j! ~|
ia-JUL » i\. » : a IR
» '^ -^ . ' \ ' ; 'j '
T « »^"^> * / > ) ( 1
! 9.1'i1
i'
i
- >
-------
v« ^U
A Tf,
S-
O
%
T~
\J
iTJ
I j » £ *
ID
O
0.19
A flrt
i i i ( j i j i | j i j i 1 i | i i l | i j i i i i i ! i i , | .
i Carie III GC
i ISOPENTANE
i FPMC/IN At ten = 2.00
r~ ~
1
t
h 12-3EP
i
fljp */^__
[i As-W^JMSjjW^ ,/*»'' v » -
r-
_
1 1 ! 1 1 '! i 1 ! ! 1 i I I 1 ! 1 1 1 ! 1 1 ! ! 1 1 ! i ! J ! ! !
j w. +t»
i
]
1
i ii(-r.-5,
rw"o
j m
i i
i
1 0.2'jZ1
1
-J
0.10
A eiti
150 I6d 170 130 19U 20e 210 220 230 240 250 260 270 280 290 300 310 320
Julian Data, 1984
For the species ISOPENTANE and the instrument Carle III GC;
Beginning C a 1 d a t e: 3 O - M A Y -- 8 4
Day: 150
Ending C a 1 d a t. e: 16 - N 0 V - S 4
Day: 32O
Up p er C a1F ac tar Limit 040000
Lower C a1F ac t or Li mi t 0.00O0O
The number of data paints
The number included was
The slope of this fit is -1.222&56E-004
with a std dev of 5.6003271E-G05 rel . dev = -45.3 7.
The intercept is 2.2209272E-O01
with a std dev of 1.273873SE-O02 rel. dev = 5.7 7.
The std dev of the fit is 1.6698649E-002 PE = 0.01113
The correlation coeff is -0.27557
The min, mean, max are 0.15986, O.19468, 0.23O74
The standard deviation is 0.01722 rel. dev = 8.8 7.
Cal = -0.00012 X (Julian Date - 150) + 0.20375
...press RETURN-key when ready...
185
-------
p- ;
i
r~
i
FFHC/IN At tan =2.00
T .A. * 1 1
i MINC.
A 4K
i v" *"w
^
"*-*-**»-&* ~. A <* '
^ -^Sr^-<:- --.__ r-"ui
o
i
'. I I I I I I I I I I . I I I I ! I I I i | i I ) i i 1 I I I I ! I
A #141 - < - 1 1 - 1 I - > - 1 - 1 I - 1 I - 1 - 1 I - 1 i - 1 - 1 - 1 - 1 I I - 1 - i_j - , - 1 - 1 - ; - 1 - 1 - 1 - 1 - 1
150 1M 170 180 130 200 210 220 230 240 250 260 270 280 290 300 310 32
Julian Date, 13S4
For the species N-BUTANE and the instrument Carle III GC:
Beginning Caldate; 30-~lv!AY-84
Day: ISO
Ending Caldate: 16-NfJV-S4
Day: 320
U p p e r C a 11- a c t Q r L. i m i t. 0. 2 0 0 0 0
L o w e r Ca1F actor Limit 0.0O000
The number of data points 40
The number included was 40
Th e slop e of this f i t is -1.50192 5E-004
with a std dev of 5.7705246E-005 rel. dev = -3S.4 :
The intercept is 1. 3697557E--001
with a std dev of 1. 24O1 192E-002 rel. dev = 9.1 7.
The std dev of the fit is 1.145S466E-002 PE = 0.00764-
The correlation coeff is -0.38897
The min, mean, max are 0.OS147, 0.10504, 0.13079
The standard deviation is 0.0122S rel. dev = 11.7 7.
Cal = -O.OO015 X (Julian Date - 150) + 0.11445
186
-------
r
0.20r-
i l I i i i i i i i i i i i i i I i i i i i
Carle III GC
1,3-BUTADIENE
FPMC/IN At ten = 2.00
. 20
O
01
3.15
"H
tu
f)
O
i
208 210 220 230 240 250 260 279 230 290 300 310 320'
Julian Date, 1984
.0.00
For the species 1,3-BUTADIENE and the instrument Carle III GC
B e q i n n i n g C a 1 d a i: e:
Day:
E n d i n g C a 1 cl a t s s
Day:
U p p e r C a 1 F a c t G r L. i m i t
Lower C a1Factor Limit
30-MAY-84
150
16-NDV-S4
320
0.. 2500O
OOOOOO
The number o-f data points 38
The number included was 38
The slope of this -fit is -1.54757OE-O04
with a std dev o-f 5. 7985O70E-005
The intercept is 1.4937310E-001
with a std dev o-f 1.2600050E-002
The std dev of the fit is 1.0673171E-OG2
The correlation coeff is 0.40642
rel . dev = ~-37.b
rel . dev = 8.4 "/.
PE - O. 00712
The min, mean, may, are 0.09043, 0.11606, 0.13717
The standard deviation is 0.01152 rel. dev = 9.9 "/.
Cal
-0.00015 X (Julian Date
150) H- 0. 12616
187
-------
iL.
O
+*
V
flj
LL.
fU
O
w"-vj ' 1 i I i i i i
i
i
1
1
A {»
a, 1-1
I
i
;
j
I"1 % -* _ ^>w_m_ -ft
/» < ftl ! ~ "^ *>
f
j
r
i
i
i
9.35|-
I
F
., .J 1 ! ! ! ! 1 i i !
i 1 j ! i 1 | i i 1 i i { 1 j 1 i i 1
Carle III GC
1-BUTENE
FFHC/IN At ten = 2.00
28-AUG
*
A
; ^
V*»-» > '
i*TT\ 1 \ o
i---ii---f^%-x^..._
Y\r^* .\ ---
* "%
! i 1 i ! ' ! i ! ! ! ! i i ! i ! i 1
i 1 i ! ' j
1
J
~~i
i
i
i
I
~j
j
i
j
4
1
i
i
1
1
H
i
,:,!,!
V.L.V
5.15
«.ie
3.'?5
A rtrt
o
Julian Date, 1984
For the species 1-BUTENE and the. instrument Carle III GC:
Beginning Caidats: 30MAY84
Day: 150
Ending Caldate: 16--NOV-34
Day: 320
Upper Gal Factor Limit 0.20OOO
Lower CalFactor Limit O.OOOOO
The number of data paints 38
The number included was 38
The slope of this fit is -1.132031E-004
with a std dev of 5.9781O57E-005 rel. dev = -50.6 "
The intercept is 1.2497382E-001
with a std dev of 1.2990315E-002 rel. dev = 10.4 "/.
The std dev of the fit is 1. 10O3754E--O02 PE = 0.00734 .
The correlation coeff is -0.31299
The min, mean, max are 0.07366, 0.09953, 0.1299S
The standard deviation is O.O1143 rel. dev = 11.5 7.
Cal = -0.00012 X (Julian Date - 150) + O.10724
...press RETURN-key when ready...
188
-------
c.\)
1 S
1 . w
O
i i : I i 1 i i i 1 i 1 i 1 i i ; j : ;
Carl
_
1 i i i i i i 1 j 1
e III GC
METHANOL
FFMC/IN
7,.
"*--»,.
Atten = 2.00
"S r ^^ -^-«i
^
-------
0.49J
i ! '' I i i i ! ' I
-------
9.3«r
9.25f
o *-2«r
i | i i i i i , i i i | ! ! i | i ;
Carle III GC
TOLUENE
PPMC/IN Atten = 2.50
A ia
-! $ 15 '
i Uli3lD
i I i ! ! ! i ! i i i ! ! ! I ! i ! I '
?£d 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320'
Julian Data, 1984
For the species TOLUENE and the instrument. Carle III GCs
B e a i n n i n g C a 1 d a t e s
Ending Caldata:
Days
Day:
Upper Cal Factor Limit
L a w e r C a 1 F a c t o r L i m i t
3 0 - M A Y - 8 4
150
16-NOV-34
320
0.30000
0 . 0 0 O 0 0
The n Limber of data points 25
The number included was 25
The slope of this fit is 9. 45263OSE-007
with a std dev of 7. 34S4704E-005
The intercept is 1 . 3672349E-001
with a std dev of 1 . 65254 08E--OO2
The std dev of the fit is 1 . 39802 13E-002
The correlation coeff is 0.00268
rel . dev = 7774.0
rel. dev = 12.1 %
PE = 0.00932
The min, mean, max are 0.11350, 0.13693, 0.16721
The standard deviation is 0.01369 rel. dev = 10.0 7.
Cal = 0.00000 X (Julian Date - 150) + 0.13687
... press 'RETURN-key when ready...
191
-------
VltlV
0.15
O
U
iTj
0
i ( 1 | 1 | ! | 1 | I | 1 | 1 | ! ! 1 i 1 i i i 1 | ! i 1 ! ! i ! |
Carie III GC \
BENZENE i
FFMC/IN Attan =2.00 !
~i
i
_ _j
i
1
i
i
A. 15
O
01
-"HI
b
~5
-
,\/fcl_ i I ___i _t_. ^ __ ^ ^ _ . . . ^ ^1
fls0 160 170 180 190 200 210 220 230 240 250 260 270 260 29* 300 3l0 3£%'V
Julian Data, 1984
For the species BENZENE and the instrument. Carle III GC:
B e g i n n i n g C a 1 d a t e:
Day:
Endi ng Caldate:
Day:
Li p p e i" Cal F a c t c r L i m i t
L o w e r C a 1 F a c t o r Li rn i t
30-MAY-84
150
16-NQV-84
320
0.20000
0.00000
The number of data points 7
The number included was 7
The slope of this fit is 1.05&4688E-O05
with a std dev of 9. 76I59602E-O05
The intercept is 8.7323S74E-O02
with a std dev of 2.150454IE-002
The std dev of the fit is 1.OOS1S72E-OQ2
The correlation coeff is O,04832
ral . dev .-- 924."
rel. dev = 24.6
PE = 0.00672
The min, mean, max are 0.08022, 0.08961, O
The standard deviation is 0.00921 rel. dev = 10.3 7.
Cal = 0.00001 X (Julian Date - 150) H- 0.08891
...press RETURN-key when ready...
192
-------
«.e«r
i j i
i , i ! i i i ; i I ; i i
Carle III GC
N-PENTANE
PFMC/IN Attan = 2.00
I !
>>.S3
i
0.50
Hi
3»J
<-*-
'-}
ttfli ; i i i i i i ; ; i ; i i i i i ! i I ! I i ! ! ! ! i ! I I ! ! M ft
'io0 160 170 180 190 200 210 220 220 240 250 260 270 280 290 300 310 320'
Julian Date, 1984
For the species N-PENTANE and the instrument Carle III GC:
Beginning Caldate:
Day:
Ending Caldate:
Day:
Upper CaiFactor Limit
Lower CalFactor Limit
30-I1AY-B4
150
16-NDV-84
320
0» & 0 0 0 0
0.OOOOO
The number o-f data points 45
The number included was 45
The slope o-f this -fit is -2.873216E-O04
with a std dev of 1.0329445E-004
The intercept is 3.5367520E-001
with a std dev of 2.372S269E-O02
The std dev of the fit is 2.63080S6E-O02
The correlation coeff is -0.39051
rel dev 36.O
rel . dev = 6.7 7.
F'E = O. 01754
The rnin, mean, max are 0.22760, 0.28S5S, 0.34680
The standard deviation is 0.02325 rel. dev = 9.S "/.
Cal ~ -0.00029 X (Julian Date - ISO) + 0.31053
...press RETURN-key when ready...
193
-------
i 1 i
! ! I
I I i i i I I I i I
Carle III GC
0.20
t-
o
tl °-15
(J
1 1
' « w.iw
_ -v _ _ f^. A^ y%"^^*
j^__ ^ >^1% Wv^/^r^v
ft
1 r 1 i ! ; ! i ! ! 1 ) I ! i r ! I i I ! ! !
= 2.00 -JO. 20
~"
ito-GCT
n 1
f
{
N^lT--
^^4
!
H
!
! ! I ! i i '. ! I !
0
Ul
0.15
~n
Ul
- ,-rt
tf.iw <-<-
0
~s
0,05
ti l^iA
160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310
Julian Date, 1984
For the species CIS-2-BUTENE and the instrument Carle III BC:
Beginning Caldate:
Day:
Ending Caldate:
Day:
Upper CalFactor Limit
Lowe r C a IF a c: t o r Li m i i:
The number of data points
The number included was
30-MAY-S4
150
16-NOV--84
320
0.35000
0.00000
45
45
The slope of this -fit is -2. 1BB447E-Q04
with a std dev o-f 4. S220399E-005
The intercept is 1.5340966E-001
with a std dev of 1.03059SOE-OO2
The std dev of the fit is 1.2219133E-OQ2
The correlation coeff is -0.56910
rel . dev = -22. 0 7.
rel . dev = 7.O 7.
F'E = 0.00815
The min, mean, max are O.O7010, O.105O7, 0.12707
The standard deviation is 0.01469 rel. dev = 14.0 7.
Cal = -0,00022 X (Julian Date - 150) + 0.12058
...press RETURN-key when ready...
194
-------
0.25r
«J
Carle III GC
TRANS-2-BUTENE
0.15
0.05
Oi OlOl
PPMC/IN Attan =2.00 H
i
-H
i
-1
1
J
1
1
1
H
< 1 I 1 l ! I 1 < i i 1 ! ! i ! i ! i 1 i ! ! I ! ! ! ! i ! ! ! t i
0.20
O
111
0.15
UJ
b
0.05
rt.ftw
160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320'
Julian Date, 1984
For the species TRANS-2-BUTENE and the instrument Carle III GC:
Beginning Caldate:
Day:
Ending Caldate:
Day:
Upper C a1F actor Limit
L o w e r C a 1 F a c t or- L i m i t
30-!VIAY~84
150
16-NQV-84
320
0.30000
0.OOOOO
The number of data points bO
The number included was 50
The slope of this fit is -1.273S05E-004
with a std dev of 4.6032130E-005
The intercept is 1.4263S30E-001
with a std dev of 1.0106748E-OO2
The std dev of the fit is 9.9038982E-003
The correlation coeff is -0.37O92
rel. dev = 36.1
rel . dev = 7. 1 7.
PE = 0.00660
The min, mean, max are 0.09145, O.I 1494, O.I3217
The standard deviation is 0.01056 rel. dev = 9.2 7.
Cal = -O.OOO13 X (Julian Date - 150) + 0.12353
...press RETURN-key when ready...
195
-------
£_
O
flj
|,
OJ
o
Wl/0j i i ! I ; i i i
(
i
0.66f~
h
i
0.50r~
h
0.40f-
f
i
w.3wi ^^^
i
i ! ; * i > i ; : ' i i i
Carl
0-XY
FFMC/IN
A __^^~^
i i i s i ; i i . i
e III GC
LENE
Atten = 2.30
"*~
"^«
1 i ' j
~i
»-H
i
1
i
i
-J
i
!
i
i
1
1
t
i
! ! 1
1
t
1 1
i i
i
~n
OJ
160 170 180 190 200 210 220 230 240 250 260 270 280 290 3'<)0 310 32$'
Julian Data, 1984
For the species 0-XYLENE and the instrument. Carle III GC:
Beginning Caldate:
End ing Caldate:
Day:
Day:
L) p p a r C a 1 F a c t o r Limit
Lciwer Cal Factor Limit
3O-MAY-84
150
16-NOV-84
320
0.. 70000
O,. 00000
The number of data points 7
The number included was 7
The slope o-F this fit is 1.3699588E-004
with a std dev of 2.3682240E-004
The intercept is 2.5194653E-001
with a std dev of 5.121S12SE-OO2
The std dev of the fit is 2.5398701E-OO2
The correlation coeff is 0.25046
r e 1 . dev -- 172. 9
rel . dev = 20.3 7.
F'E = 0.01727
The min, mean, max are
0.25603,
0.231O3,
0. 317
The standard deviation is 0.02442 rel. dev ~ 8.7 7.
Cal - 0.00014 X (Julian Date - 150) + 0.27250
196
-------
f
o
+>
CJ
nj
1 1
LJ_
i4
01
U»
j
h-
i
I
A.50J
s
l~
i
3.43J-
i
i_
i
a. 3a|-
.
i
3.20J-
i
0.10f
1
i
lA flAl ! 1
j ; j i ; i j i j i i i i i i ; ] i | i j i i i i i i i i i ;
CEA Formal deny cie
FORMALDEHYDE
FPMC/IN At ten = i.00
1
(\ _
|.\ ^' ^
, ' \ ^***^^^~^. J* \
? !^ ^ \
\
f
. *
t ! j 1 1 i 1 ! 1 1 i ! ! ! 1 1 1 1 ! ! 1 1 ! i 1 1 1 ! 1 ! I
i i
i
i
i
i
i
i
^
i
i
i
i
i
i
i
i
1
i
~1
3.51?
.-"
0.40^
j
9.30JT
j^i
^r*-
0.20^
S.10
A Oil*
<5d 16» 170 180 199 200 210 220 230 240 250 260 270 280 290 300 310 320'
Julian Date, 19S4
Far the species FORMALDEHYDE and the instrument CEA Formal d eh yd,;-
Beginning Caldate: 30-MAY--S4
Day: 150
Ending Caldate;: 16--NQV--84
Day: 32O
Upper- Ca 1 Fractor' Li mi t 0.6000O
Lower C a 1 F a c: t o r L. i m i i: 0. 0 000 0
The number o-f data points 17
'T h e n u m b e r i n c 1 u dad w a s 17
The slope o-f this fit is -2.2B2711E-004
with a std dev of 4. 7236958E-004 rel . clev = -2O6.9 "/.
The intercept is 3. 1092736E--001
with a std dev o-f 1. 132S004E-001 rel. dev = 36-. 4 7.
The std dev of the fit is 7.9710921E-002 PE = 0.05314 N
The correlation coe-ff is -0.12381
The min, mean, max are 0.12160, 0.25699, 0.39930
The standard deviation is 0.07773 rel. dev = 30.3 7.
Cal = -0.00023 X (Julian Date - 150) + 0.27669
...press RETURN-key when ready...
197
------- |