v>EPA
United States
Environmental Protection
Agency
Environmental Monitoring and Support EPA-600 4-78-023
Laboratory May 1978
Research Triangle Park NC 27711
Research and Development
Survey of
Transmissometers
Used in
Conducting
Visible Emissions
Training Courses
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
SURVEY OF TRANSMISSOMETERS
USED IN CONDUCTING
VISIBLE EMISSIONS TRAINING COURSES
by
Michael C. Osborne and M. Rodney Midgett
Quality Assurance Branch
Environmental Monitoring and Support Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
MARCH 1978
-------
DISCLAIMER
This report has been reviewed by the Environmental Monitoring and
Support Laboratory, U. S. Environmental Protection Agency, and approved
for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
CONTENTS
Figures iv
Tables v
Acknowledgments . . , vi
1. Introduction 1
History 1
Purpose 2
2. Procedures 3
Data Requirements 3
Obtaining Participants 15
3. Survey Results 18
\
4. Discussion of Results 24
General 24
By Parameter 26
5. A Need for Further Study 29
Conclusions 29
A Recommendation for a More Comprehensive
Visible Emissions Study 29
References 36
m
-------
FIGURES
Number Page
1 Response Curve for Filter No. 1A 7
2 Response Curve for Filter No. 4 7
3 Response Curve for Filter No. 21 7
4 Response Curve for Filter No. 29 8
5 Response Curve for Filter No. 88A 8
6 Calibration Wand 10
7 Range of Transmissometer Responses Over
the Visible Spectrum 20
8 Good Photopic Comparison 21
9 Poor Photopic Comparison 21
iv
-------
TABLES
Number Page
1 Smoke Meter Design and Performance Specifications ... 4
2 Spectral Transmittance of Tiffen Filter N.D. 0.1 ... 11
3 Spectral Transmittance of Tiffen Filter N.D. 0.3 . . . 12
4 Spectral Transmittance of Tiffen Filter N.D. 0.6 . . . 13
5 Survey Participants 17
6 Light Source Voltage 18
7 Photopic Response Comparison 19
8 Calibration Filter Check 22
9 Zero and Span Drift 23
10 Response Time 23
11 Summary of Survey Results 25
-------
ACKNOWLEDGMENTS
The authors wish to acknowledge with appreciation the assistance of
Mr. Willie S. Lee and Mr. Floyd M. Pearce of Environmental Industries,
currently the only manufacturer of visible emissions generators.
Appreciation is also extended to Mr. Thomas M. Rose of EPA's Region IV
in Athens, Georgia, to Mr. William D. Conner of the Environmental
Sciences Research Laboratory in Research Triangle Park, North Carolina,
and to Mr. Roy M. Neulicht of the Office of Air Quality Planning and
Standards in Durham, North Carolina.
The many generator operators who assisted us in the evaluation of
their transmissometers have contributed significantly to the success of
this survey and for their time and efforts we are grateful.
VI
-------
SECTION 1
INTRODUCTION
HISTORY
Visible emissions training using a smoke generator and a trans-
missometer was initiated in 1950 by the Los Angeles County Air Pollution
Control District (1). Since that time some notable modifications and
additions have been made to this approach in training inspectors to
"read smoke", however, the instrumentation required to train inspectors
has remained practically unchanged.
The U. S. Environmental Protection Agency (EPA) specified the
procedures involved in visible emissions training with the promulgation
of Method 9 in the Federal Register on December 23, 1971 (2). According
to the Method,
"to certify as an observer, a candidate must complete
a smokereading course conducted by EPA, or equivalent."
During the early seventies, EPA through the Air Pollution Training
Institute conducted numerous "Visible Emissions Certification or Re-
certification" courses throughout the United States. However, because
of the need for recertification every 6 months and the travel re-
strictions placed on many state and local agency inspectors, several
states and cities attempted to set up equivalent courses. Many ques-
tions' quickly surfaced concerning what constituted an equivalent course.
1
-------
To clarify this point, Method 9 was revised and expanded on
November 12, 1974 (3). Included in this revision was a list of smoke
meter or transmissometer design and performance specifications. Each
state or local agency was expected to meet these established criteria.
PURPOSE
When a survey of transmissometers used in conducting visible emissions
training courses was planned, over 2 years had already passed since
the transmissometer design and performance specifications had been
published. This was sufficient time to expect all existing visible
emissions training transmissometers to be modified and operating according
to the design and operating specifications. However, discussions with
the EPA Environmental Sciences Research Lab, the Office of Air Quality
Planning and Standards, and a smoke generator manufacturer revealed that
not all of these smoke reading courses met the outlined criteria set
forth in the November 12, 1974, Federal Register.
Thus, there was a need to identify those generators that did not
meet these criteria and to determine why they did not meet these criteria.
Since there were more smoke schools in existence throughout the country
than could be surveyed economically, the decision was made to choose two
schools from each EPA Region. This would be a sufficiently large number
for evaluation and would provide data that was not regionally biased.
Such a survey promised to answer two questions. First, are the visible
emissions training transmissometers across the country accurately measuring
the opacity in the smoke generator or are some of them operating with a
built-in bias? Second, are all of the schools concerned about cal-
ibration procedures, or are some of the schools careless in performing
regular checks on their equipment? 2
-------
SECTION 2
PROCEDURES
DATA REQUIREMENTS
In planning the survey, the first decisions that had to be made
involved the identification of information required from the survey
participants and their instruments and the methodology for obtaining
this information. Since the smoke generator specifications mentioned
in the November 12, 1974, Federal Register were limited in scope to
the smoke meter or transmissometer, we concluded that our survey should
also be limited to the instrument that measured the smoke opacity. This
approach had the advantage of not requiring that each generator be
fired during the field evaluation of the transmissometer. One disadvantage
of not including parts of the generator other than the transmissometer
was that our survey would not reveal problems in maintaining a con-
sistent burn over the entire 0 to 100% opacity range. If obtaining a
consistent burn was a problem, however, the generator operator would
already be aware of it. Little would have been gained from identifying
problems that were already well known. By eliminating other components
from the evaluation, each transmissometer could be tested in the time
span of an hour provided the instrument was warmed up before the test
began. Considering that participation in this survey was on a voluntary
basis, a shortened test time was desirable.
-------
With the study limited to the transmissometer, the question then
became, what transmissometer operating characteristics were of interest?
It was decided (as shown in Table 1) to study the smoke meter design and
determine if the performance specifications enumerated in the
November 12, 1974, Federal Register were met.
TABLE 1. SMOKE METER DESIGN AND PERFORMANCE SPECIFICATIONS
Parameter Specification
a. Light source Incandescent lamp operated
at nominal rated voltage
(+ 5% nominal rated voltage)
b. Spectral response Photopic (daylight spectral
of photo cell response of the human eye)
c. Angle of view 15ฐ maximum total angle
d. Angle of projection 15ฐ maximum total angle
e. Calibration error +3% opacity, maximum
f. Zero and spandrift +1% opacity, 30 min
g. Response time 5 sec
Having identified what we should expect from the transmissometers, our
task became one of finding methods to check each of these parameters to
see whether the design or performance specifications were being met.
Light Source
Devising a method to check the light source was relatively easy.
A voltmeter could be placed across the incandescent lamp and the measured
voltage compared with the nominal rated voltage listed in the smoke
meter specifications. The voltmeter had to handle a wide range of AC and
4
-------
DC voltages because of the large variety of voltage requirements en-
countered among the instruments being surveyed. Instrument specifi-
cations were not always available to determine the nominal rated voltage
of the lamp. When specifications were not available, the generator
operator could usually provide the needed information.
Spectral Response of Photocell
Determining if the photocell would yield a photopic response
required either checking the spectral response curve of the specific
commercial photo emissive surface as listed in the literature (4) or
measuring the spectral response curve by using wavelength specific
filters. The latter approach was used for the following reasons:
Some of the photocells were old, and the spectral response may
have changed with time.
Some of the transmissometers had undergone major modific-
ations; therefore, the photocell might have been different from the one
listed in the manufacturer's specifications.
The manufacturer's spectral response curve data may have been
suspect.
Instrument specifications identifying the photocell may not
have been available.
To test for photopic response a series of five Kodak Wratten sharp
cut-off filters were used. The filters used blocked out practically all
light at, and below, a specific wavelength. By choosing five different
filters it was possible to stepwise block out light going from left to
right on the visible spectrum. The fifth filter was a visibly opaque
-------
filter which allowed only infrared light through. Using the spectral
response curves provided by Kodak (5) shown in Figures 1-5, the wavelength
was determined at which 95% of the light was blocked out. This wavelength
was then plotted against the observed transmissometer response obtained
with this filter to get a spectral response curve for each transmissometer.
This curve, generated from the set of five filters, was compared with
the standard spectral-luminosity curve for photopic vision (6). Filter
No. 1A consistently blocked out some visible light and was therefore a
poor choice for this survey. The other filters, however, performed
satisfactorily.
Angles of View and Projection
Unlike checking the photopic response, measuring the angles of view
and projection could not be readily achieved without taking the in-
strument apart. The Federal Register provided equations for calculating
both angles but to do so required measuring the length or width of
various components of the instrument. These dimensions were not readily
accessible once the generator was placed in operation. Measurement of
these angles, therefore, was omitted from the survey procedures since taking
the transmissometer apart to make these measurements could have resulted
in damage to the instrument.
-------
0.1
3. 1 t 2
o
IU
oo
V)
10
100
0.1
CO
K
8. 1
Figure 1. Response curve for filter no. 1A.
t 2
CO
111
a
100
0.1
ป''ป'>ป>::>
.i-.-.-.-.'.'.'.
Figure 2. Response curve for filter no. 4.
LLI
CO
10 5 1
100
200
300
400
700
500 600
WAVELENGTH, nanometers
Figure 3. Response curve for filter no. 21.
800
900
-------
0.1 3
t 2
100
0.1
200
300
400
700
S
tu
u
<
cc
(-
to
Ui
a
10 1
100 0
500 600
WAVELENGTH, nanometers
Figure 4. Response curve for filter no. 29.
800
- "
700
800
900
WAVELENGTH, nanometers
1000
Figure 5. Response curve for filter no. 88A.
8
900
1100
-------
Calibration Error
The calibration of the transmissometer according to the Federal
Register method was to be achieved by introducing neutral density filters
of nominal opacities of 20, 50, and 75 percent. Two choices of neutral,
density filters are readily available. One is the gelatin-type filter
commonly available from Kodak, and the other is described as a metal-film
filter available from Tiffen. Although the gelatin filter has been used
almost exclusively in calibrating smoke generators, it is not a true
neutral density filter. Over the range of 500 to 400 nanometers,
the light transmission of the gelatin filters drops almost in half (7).
This can be compared to the relatively uniform light transmittance of
the metal-film filters listed in Tables 2-4 (8). Prepared with this
information, we chose to use the three metal-film filters supplied by
Tiffen to perform the calibrations.
Three different approaches have been used to introduce the filters
into the light path. The simplest but probably the worst approach is to
remove the photocell, tape the filter to it, and put the photocell back
onto the transmissometer. This method of calibration presents the
following problems:
For those photocells that screw onto the transmissometer arm,
the potential exists for not getting the photocell screwed back into the
same position after each calibration.
Metal-film filters are usually too large to tape onto the
photocell; therefore, gelatin filters must be used.
Because of the shape of the photocell, gelatin filters are
usually bent when taped to the photocell. This could affect the opacity
9
-------
reading.
Taping the filters securely to the photocell is difficult
without placing the tape in the light path.
The second method of introducing the filter is by cutting a slot 1n
either the photocell arm or light source of the transmissometer and
lowering the filter into the light path. This method is commonly used
by generator operators who have had to modify their equipment to do
calibrations.
The best alternative method for introducing the filter is to insert
it down the stack directly into the light path. This requires a special
tool known as a calibration wand (Figure 6). The wand is long enough
to be inserted from either end of a collapsible-stack generator.
The wand has a black metal, two-prong fork at one end which fits
around the transmissometer cross-bar inside the stack and secures
the wand in place. Inside the two-prong fork is a black metal filter
support with a 1-1/2 inch diameter round hole in it. The filter support
1.5 inch (38 mm) HOLE
5ft. (1.52 meters}-
Figures. Calibration wand.
10
-------
TABLE 2. SPECTRAL TRANSMITTANCE OF TIFFEN FILTER ND 0.1
Wavelength
(ran)
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
530
540
550
560
570
580
590
600
610
620
630
640
650
660
670
680
690
700
710
720
730
740
Transmittance
(Percent)
73.1
74.0
74.5
74.7
74.8
74.9
75.2
75.4
75.7
75.9
76.0
76.0
76.2
76.3
76.4
76.6
76.8
76.5
76.7
76.8
77.0
76.6
76.6
76.7
76.6
76.7
76.6
76.4
76.4
76.3
76.2
76.0
76.0
75.9
75.6
75.7
75.6
Optical
Density
0.136
.131
.128
.127
.126
.125
.124
.122
.121
.120
.119
.119
.118
.117
.117
.116
.115
.116
.115
.115
.114
.116
.116
.115
.116
.115
.116
.117
.117
.118
.118
.119
.119
.120
.122
.121
.122
11
-------
TABLE 2. SPECTRAL TRANSMITTANCE OF TIFFEN FILTER ND 0.1 (continued)
Wavelength Transmittance Optical
(nm) (Percent) Density
750 75.2 .124
760 75.2 .124
770 75.1 .124
12
-------
TABLE 3. SPECTRAL TRANSMITTANCE OF TIFFEN FILTER ND 0.3
Wavelength
(nm)
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
530
540
550
560
570
580
590
600
610
620
630
640
650
660
670
680
690
700
710
720
730
740
Transmittance
(Percent)
46.7
47.3
47.6
47.7
47.7
47.7
47.7
47.7
47.8
47.8
47.9
47.9
47.9
48.0
48.0
48.0
48.1
48.1
48.2
48.1
48.2
48.2
48.2
48.1
48.2
48.2
48.2
48.2
48.2
48.2
48.2
48.2
48.2
48.3
48.2
48.2
48.2
Optical
Density
0.331
.325
.323
.322
.321
.322
.322
.321
.320
.320
.320
.319
.319
.319
.319
.319
.318
.317
.317
.318
.317
.317
.317
.318
.317
.317
.317
.317
.317
.317
.317
.317
.317
.317
.317
.317
.317
13
-------
TABLE 3. SPECTRAL TRANSMITTANCE OF TIFFEN FILTER ND 0.3 (continued)
Wavelength Transmittance Optical
(nm) (Percent) Density
750 48.2 .317
760 48.2 .317
770 48.2 .317
14
-------
TABLE 4. SPECTRAL TRANSMITTANCE OF TIFFEN FILTER ND 0.6
Wavelength
(nm)
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
530
540
550
560
570
580
590
600
610
620
630
640
650
660
670
680
690
700
710
720
730
740
Transmittance
(Percent)
23.6
23.7
23.7
23.6
23.5
23.4
23.3
23.1
23.0
22.8
22.7
22.6
22.4
22.3
22.1
22.1
22.0
21.9
21.8
21.7
21.7
21.5
21.4
21.4
21.3
21.3
21.2
21.1
21.0
21.0
20.9
20.8
20.8
20.7
20.7
20.6
20.5
Optical
Density
0.628
.625
.625
.627
.629
.630
.633
.636
.639
.641
.654
.646
.649
.651
.655
.656
.658
.659
.661
.663
.664
.667
.669
.670
.672
.672
.674
.676
.677
.678
.679
.681
.683
.684
.685
.685
.688
15
-------
TABLE 4. SPECTRAL TRANSMITTANCE OF TIFFEN FILTER ND 0.6 (continued)
Wavelength Transmittance Optical
(nm) (Percent) Density
750 20.5 .689
760 20.4 .690
770 20.2 .695
16
-------
is just wide enough to fit through the slit in the transmissometer
cross-bar. Due to the wide range of instrument designs encountered in
this study, eight different filter supports were prepared in increasing
size increments of 1/8 inch. With this calibration wand we were able to
introduce our metal film filters across the light path and measure the
calibration error.
Zero and Span Drift
The Federal Register specifies that zero and span drift be de-
termined only after operating the smoke generator in a normal manner for
1 hr. A problem arose, however, in defining normal operation. Some
generator operators interpreted this to mean that the instrument may be
re-zeroed and re-spanned after every new reading. Since normal operation
was not officially defined, the operators' definition could not be
disputed.
The technique used in this survey to measure zero and
span drift was to perform the other checks which have already been
mentioned and to follow this with a zero and span check. Whether
the operator had re-zeroed and re-spanned his instrument during these <
earlier checks was not taken into consideration. It is not likely,
however, that the operator was able to anticipate when the zero and span
drift were going to be checked.
17
-------
Response Time
The last test performed on the transmissometer was for response
time. A variety of instruments were being used to register the measured
opacity. They included such extremes as a simple dial gage, a strip
chart recorder, and a digital recorder. With the dial gage it was
difficult to discern opacity readings in less than 5-% increments. The
digital recorder had the problem of being barely visible in the bright
sun. In general, the strip chart recorders presented the fewest problems,
although some of the recorders gave an unacceptably slow response.
Response time was measured by a stop watch over the range from zero to
span.
OBTAINING PARTICIPANTS
Although our initial goal was to obtain two participants from each
EPA Region, this was modified during the early phases of the survey due
to restrictions in our travel budget. The two west coast Regions,
Regions 9 and TO, were eliminated from the survey because they represented
the greatest strain on travel time and money. Representatives from the
other eight Regional Offices were requested to submit a list of smoke
reading courses being scheduled in their Regions along with the name of
a contact person for each school.
The criteria used in selecting survey participants included
determining whether;
The survey could be scheduled either before or after a planned
"smoke school".
18
-------
A large number of smoke inspectors were trained annually
with that particular generator.
The design of the generator was conducive to being sur-
veyed .
The generator operator was willing to have his transmisso-
meter tested.
Each of the agencies that ultimately participated did not have to
meet all of these criteria; however, the criteria did aid in the selection
of two candidates from each Region.
Table 5 lists those agencies that were selected and did participate
in the survey. Since their participation was voluntary, each agency
has been asssigned a random code number to prevent the agency from being
identified with a specific set of test results.
19
-------
TABLE 5. SURVEY PARTICIPANTS
1. EPA Region I in Boston, Massachusetts
2. State of Connecticut
3. State of New York
4. EPA Region II in New York, New York
5. City of Pittsburgh, Pennsylvania
6. State of Virginia
*
7. EPA Region IV in Athens, Georgia
8. State of North Carolina
9. City of Chicago, Illinois
10. State of Minnesota
11. State of Arkansas
12. State of Texas
13. EPA Region VII in Kansas City, Kansas
14. State of Iowa
15. State of Colorado
16. State of Utah
20
-------
SECTION 3
SURVEY RESULTS
The survey was conducted between April 21, 1977, and August 2, 1977.
All of the tests'in the survey were performed by the principal author
using the audit tools outlined in an earlier section. Tables 6-10 and
Figures 7-9 are summaries of the tests performed on each parameter of
the transmissometers.
TABLE 6. LIGHT SOURCE VOLTAGE
Generator
Number
101
102
103
104
105
106
107
108
Rated
Voltage
12
110
110
12
12
NA
6
NA
Measured
Voltage
12.7
135
125
8.7
12
6.5
6
7.9
Generator
Number
109
no
111
112
113
114
115
116
Rated
Voltage
no
24
12
6
12
12
12.5
12
Measured
Voltage
105
15
14
4.8
13.5
13.5
14.5
13.5
21
-------
TABLE 7. PHOTOPIC RESPONSE COMPARISON
Kodak Filter No.
Generator No.
True Photopic
101
102
103
104
105
106
107
108
109
no
111
112
113
114
115
116
1A
0
10
15
12
9.5
11
5
13
15
12.5
8
12
13
12
11
9.5
14
4
5
11
11
10
8
15
14
9.5
10
12
12
35
14
14
13
10.5
11
21
Opacity, %
50
34
41.5
41
42
40
10
42
40
36
38
66.5
25
37.5
38
32
44
29
81
66
87.5
87
88
75
20
87
85
64
72
87.5
56
69
70.5
63.5
87.5
88A
100
93
100
100
99
93
32
100
100
93
95
90.5
90
92.5
93
90.5
96
22
-------
IUU
90
80
70
1 60
i
* 50
H"
1 40
o
30
20
10
o
, ._T...._
1 1-
O PHOTOPIC T
- - - _
t k .___
TRANCE OF OBSERVED RESPONSES*
_
-I
<^
ซ
ซ
O
400 450
i
-L
> '
*THIS DOES NOT INCLUDE THE PHOTOPIC
MEASUREMENTS OF GENERATOR NO. 106.
1
i
500 550 600 650 '700 75
WAVELENGTH,nm
Figure 7. Range of transmissometer responses over the visible spectrum.
23
-------
ฃ
u
100
90
80
70
60
50
40
30
20
10
ofOL
O PHOTOPIC
D GENERATOR 107
O GENERATOR 116
400
450
500 550 600
WAVELENGTH, nm
650
Figure 8. Good photopic comparison.
700
750
100
90
80
70
|
I
s.
> 50
40
30
20
10
oL-o
o
<
O PHOTOPIC
O GENERATOR NO. 111
OGENERATOR NO. 112
O
o
a
o
400 450 500 550 600
WAVELENGTH, nm
Figure 9. Poor photopic comparison.
650
-o-
8-
700
750
24
-------
TABLE 8. CALIBRATION FILTER CHECK
Generator No.
NBS-550 nm
101
102
103
104
105
106
107
108
109
no
m*
112+
113
114
115
116
* u_ ^_ ___ _
N.D.
Opacity
23.5
23
22
23.5
22
24
25
23.5
23
25
19
21.5
24
25
23.5
21.5
26
--
0.1
% Error
--
-0.5
-1.5
0.0
-1.5
+0.5
+1.5
0.0
-0.5
+1.5
-4.5
-0.5
+0.5
+1.5
0.0
-2.0
+2.5
N.D.
Opacity
51.5
50.5
49.5
51
49
52
55
52.5
51
50
42
56
44
58
54
52
50.5
54
1
0.3
% Error
-1.0
-2.0
-0.5
-2.5
+0.5
+3.5
+1.0
-0.5
-1.5
-9.5
0.0
-7.5
+6.5
+2.5
+0.5
-1.0
+2.5
N.D.
Opacity
78.5
78
78
78
76
79
82
78.5
80
75
72
81.5
83.5
81
79
79
80
i i 1. 1 i
0.6
% Error
_-
-0.5
-0.5
-0.5
-2.5
+0.5
+3.5
0.0
+1.5
-3.5
-6.5
-0.5
+5.0
+2.5
+0.5
+0.5
+1.5
*Different filters were used to check this unit because slit size was too
narrow for 2-inch metal film filters.
+This instrument had a significant drift problem. The lower opacity was
observed at the beginning of the test and the higher opacity about an
hour later.
25
-------
TABLE 9. ZERO AND SPAN DRIFT
Gen. No. Zero Drift
%
101 1/2
102 3
103 0
104 0
105 0
106 0
107 1/2
108 3
Span Drift Gen. No. Zero Drift Span Drift
% % %
1/2 109 0 0
o no o o
0 111 0 1/2
1/2 112 14 5*
1 113 0 0
0 114 0 0
0 115 1 0
1 116 1/2 1/2
*The recorder was pegged
at approximately 105% opacity.
TABLE 10. RESPONSE TIME
Generator Response
101
102
103
104
105
106
107
108
Time (sec.) Generator Response Time (sec.)
17 109 2
3 110 3
1 111 4
1 112 ' 3
4 113 3
4 114 4
3 115 5
1 116 5
26
-------
SECTION 4
DISCUSSION OF RESULTS
GENERAL
Table 11 is a summary of the transmissometers that failed to meet
the performance specifications outlined in the Federal Register. Over
one-half of the survey participants failed to meet at least one of these
specifications.
Although operator understanding and familiarity with the smoke
generator was not measured quantitatively, a subjective evaluation was
made. Of the generators that failed to meet the operating specifications,
most had operators who lacked a basic knowledge of how the equipment
worked. This was not true, however, of the operators of the units
that passed all the checks.
As one might expect, the smoke meters receiving more use over the
course of a year generally performed better than those used only
sporadically. A few of those used most regularly, however, did fail
concerning at least one parameter.
27
-------
TABLE 11. SUMMARY OF SURVEY RESULTS
M - Missing Data
F - Fail
Light
Voltage Zero/
Deviation Spectral* Response Span
Generator + - Response Calibration Time Drift Overall
101
102
103
104
105
106
107
108
109
110
in
112
113
114
115
116
F
F
F
F
M
M
F
F
F
F
F
F
F
F F
F F
F
F F F
F F
F F
F F
F F
F F F F
* If any reading exceeded photopic opacity value by + 20 percent, opacity
unit failed.
+ Did not cohsider unit to fail specifications if positive deviation of
light source voltage exceeded Federal Register specifications.
28
-------
BY PARAMETER
Light Source
Each incandescent lamp was to be operated within + 5% of the nominal
rated voltage according to the Federal Register. This requirement was
determined to be too restrictive based on available instrumentation. The
only problem that will be encountered by exceeding the nominal voltage is
a shorter life for the lamp. Although this is not desirable, if the
operator is willing to stock a sufficiently large number of replacement
lamps, no real performance failures can be expected. If, on the other
hand, the lamp is operated at lower than rated voltage, the instrument
may not give an adequate response. Four of the smoke meters surveyed were
operating with lamps at less than the nominal rated voltage. Table 6 lists
the nominal rated voltage and the measured voltage for all 16 instruments
surveyed.
Spectral Response of Photocell
In order for the photocell to be classified "photopic", the Federal
Register requires that the "spectral sensitivity of the cell closely
approximate the standard luminosity curve for photopic vision. Since
there was no definition given for "close approximation", the classi-
fication of a photocell as photopic was difficult. Certainly, some
photocells were more photopic than others as can be observed by com-
paring Figures 4 and 5. If any observed reading exceeded the photopic
opacity reading by + 20%, the transmissometer was assumed to fail
the performance specification.
29
-------
The survey disclosed one photocell that could not in any way be
classified as photopic. Generator No. 106 was found to have a photo-
cell which was clearly infrared sensitive. The effect of such a photocell
is to produce very low opacity readings for very high plume opacities.
This effect can be most dramatically observed in the production of white
(oil) plumes (9). The discovery that this photocell was being used on a
regular basis in smoke reading courses is believed to be the most
significant finding of this study.
Calibration Error
Of the 16 generators only 4 failed to meet the + 3% calibration
error specification. In all four cases, the calibration error was less
with the lower optical density filters.
Generator No. 106 was only slightly out of calibration with two of
three neutral density filters. Both filters read consistently 3-1/2%
high. Generator No. 109 was high on only the 0.6 neutral density
filter.
Generator No. 110 read consistently low with the greatest error
occurring at what should have been the mid-scale reading. Surprisingly,
the instrument calibrated perfectly if the filters were introduced into
the arm of the transmissometer. When introduced into the center of the
stack, the filters read up to 9-1/2% low. This was the only instance in
which different readings were obtained in the arm and in the stack. One
possible explanation of this difference is that light scattering by the
neutral density filters may have occurred when they were placed in the
transmissometer arm. This problem is currently being investigated.
30
-------
Generator No. 112 obtained readings that varied considerably be-
cause of a significant positive drift problem. At the beginning of the
test, the instrument recorded a value of 44% opacity for a 0.3 N.D
filter. At the end of the test when this same filter was reinserted
it read 58% opacity or 14% higher. Therefore, the observed cal-
ibration problem may have been due entirely to this large positive
drift.
Zero and Span Drift
Only three generators did not meet the + 1% zero and span drift
requirement. Two of these generators did meet the span drift require-
ment but not the zero drift requirement. The only really bad drift
was by Generator No. 112, but this was reported previously in the
discussion of calibration error.
Response Time
Generator No. 101 was the only one that failed to meet the
5-sec response time requirement; but several others were very close to
the limit. Generator No. 101's problem was with its strip chart recorder.
The recorder should be repaired or replaced immediately since using a
strip chart recorder with a 17-sec response time is a severe handicap
in teaching people to "read" smoke.
31
-------
SECTION 5
A NEED FOR FURTHER STUDY
CONCLUSIONS
Based on the findings of this survey, the following conclusions
have been made:
1. Less than half of the transmissometers surveyed met all of the
requirements set forth in the transmissometer design and performance
specifications.
2. Significant differences existed in the "phptopic comparison"
of transmissometer photocells.
3. Some operators of "smoke school" generators were not ade-
quately trained to properly calibrate their equipment.
4. The need was established for all "smoke school" transmisso-
meters to be checked at regular intervals by the appropriate EPA Regional
Office.
5. Other aspects of the visible emissions training effort such as
stability and range of smoke opacities produced should also be investigated.
A RECOMMENDATION FOR A MORE COMPREHENSIVE VISIBLE EMISSIONS STUDY
A special study should be initiated to improve the training program
for smoke inspectors. The following five-phase study should contribute
significantly in bolstering the visible emissions training effort.
32
-------
Phase 1 - State and Federal Visible Emissions Generator Survey
Before making any changes to the present program, a complete
summary of what exists and how it is being used is necessary. A survey
of all visible emissions training programs throughout the country would
provide this important background information. This survey form should
include:
Location of generator
Generator manufacturer
Date generator manufactured
Name of operator
Type of location (portable or fixed)
Checks for meeting Federal Register guidelines
Number of schools per year
Number of students trained per year
Operator education in smoke reading
Operator recommendations or complaints concerning generator design
With this information available, a national locator map and data file
could be compiled.
33
-------
Phase 2 - Evaluation of Factors Affecting Visible Emissions Training Programs
A look at normal problems that affect the successful outcome of
training efforts should reveal ways that this training may be improved.
Some of these problems are weather related. The effects of temperature
extremes, rain, snow, wind, and light conditions on observer performance
should be investigated. Specifically, the use of a 90ฐ, laminar,
elbow stack attachment in high wind conditions should be checked. Also, an
evaluation of the effect of reading smoke against a variety of back-
ground features should be tested. Since some smoke inspectors have been
known to record visible emissions readings from inside vehicles through
closed windows, the feasibility of taking indoor readings through glass
media should be studied. Finally, the design of the score sheets used
to record opacity data should be investigated.
34
-------
Phase 3 - Fuel Evaluation Study for Visible Emissions Generator
Applications
With the classification of benzene as a hazardous pollutant, other
fuels are being used to make black smoke. Although toluene has been the
recommended replacement for benzene, other new fuels or hybrid fuels may
be more applicable to visible emissions training. An evaluation of the
alternate fuels for making black or white smoke could include a look at:
Health hazards
Burning characteristics
Air flow requirements
Plume color
Particle size
Stability
Storage
Burning rate vs. opacity
Freezing point
Safety aspects
The product of this evaluation would be the specification of fuels
to be used by "smoke schools" for making black and white smoke.
35
-------
Phase 4 - Development of a Better Visible Emissions Generator
After identifying the problems that are inherent with current smoke
generators and the fuels that should be used to make smoke, the next
step is to determine ways to improve smoke generator design and operation.
Factors affecting performance that should be investigated include:
Air flow stability and flow rate
Black smoke chamber design
Fuel flow transfer system
Mechanical structure of stack and chamber
Functional layout of control panels
Power supply stability requirements
Programmable opacity meter
Fuel pump and regulator valve
Photocell selection
Light source
Readout device
Calibration procedures
36
-------
With the information obtained from this investigation, a prototype
generator can be built that would more nearly meet the needs of a visible
emission training program. Developed along with the generator would be
a comprehensive operators' manual that includes:
Operating principles
Method of operation
Maintenance requirements
Electrical schematics
Guidelines for safe use
Parts list
37
-------
Phase 5 - Training Program for Visible Emissions Generator Operators
As the new generator developed in Phase 4 is marketed and gradually
replaces current generators, a training program for generator operators
will become essential. Such a program would need to be coordinated with
existing programs in EPA's Air Pollution Training Institute. Included
in the training program should be instructions on how to conduct an
initial certification course as well as a recertification course. Also
included should be an explanation of the legal aspects that pertain to
the visible emissions operator. The operators should be taught the
requirements for:
Classroom instruction
Record keeping
Calibration
Operation
Maintenance
Safety
The student operators should be given the opportunity to demonstrate:
A general knowledge of the generator
Maintenance of class control
Operation of the visible emissions generator
Certification requirements
Generator operators should also be expected to repeat this training
program at regular intervals to insure an ongoing program to maintain
working knowledge of the visible emissions training program.
38
-------
REFERENCES
1. Griswold, S. Smith, VI. H. Parmelee, and L. H. McEwan. Training of
Air Pollution Inspectors. Paper presented at the 51st Annual
Meeting of the Air Pollution Control Association, May, 1958,
Philadelphia.
2. U. S. Environmental Protection Agency. Standards of Performance
for New Stationary Sources. Federal Register, 36:24876-24895,
(1971).
3. U. S. Environmental Protection Agency. Standards of Performance
for New Stationary Sources. Federal Register. 39:39872-39875,
(1974).
4. Willard, Hobart H., Lynne L. Merritt, and John A. Dean. Instrumental
Methods of Analysis. D. van Nostrand Company, Inc., Princeton,
New Jersey, 1965, p. 49.
5. Eastman Kodak Company. Kodak Filters for Scientific and Technical
Uses. Eastman Kodak Company, Rochester, New York, 1970.
6. Condon, E. U., and J. Odishaw. Handbook of Physics. McGraw-Hill
Co., New York, New York, 1958, Table 4.2, pp. 6-66.
7. Weast, Robert C. Handbook of Chemistry and Physics. The Chemical
Rubber Co., Cleveland, Ohio, 1972, p. E-221.
8. National Bureau of Standards. Report of Test for Luminous Transmittance
of Three Neutral Density Filters. Reference Purchase Order
No. DA-7-6277H, U. S. Department of Commerce, August, 15, 1977.
9. Conner, William D. Measurement of the Opacity and Mass Concentration
of Particulate Emissions by Transmissometry. EPA 650/2-2-74-128.
U. S. Environmental Protection Agency, 1974. (Available from
National Technical Information Service, 5285 Port Royal Road,
Springfield, Virginia 22161).
39
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
2.
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AMD SUBTITLE
Survey of Transmissometers Used in Conducting Visible
Emissions Training Courses
5. REPORT DATE
March. 1978
6. PERFORMING ORGANIZATION CODE
'. AUTHOR(S)
Michael C. Osborne and M. Rodney Midgett
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Monitoring and Support Laboratory, MD-77
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1HD621
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
TV-iannlo Payfc NT 77711
Final Report
14. SPONSORING AGENCY CODE
EPA-ORD-600
r\n\
OTE
15. SUPPLEMENTARY NOTES
To be published as an Environmental Monitoring Series Report.
16. ABSTRACT
The Quality Assurance Branch (QAB) of the Environmental Monitoring and Support
Laboratory has undertaken the task of evaluating the transmissometers that are
currently being used in visible emissions training programs. The criteria used
in the evaluation were the design and performance specifications for smoke meters
promulgated in the November 12, 1974, Federal Register.
Sixteen "smoke schools" from EPA Regions 1-8 participated in the survey which
was performed between April 21, 1977, and August 2, 1977.
Results of the survey showed that only half of the transmissometers which were
evaluated met all of the design and performance specifications. A lack of operator
understanding and familiarity with smoke generators was a contributing factor among
those units that failed to meet the requirements.
A list of recommendations are, also, included for a five-phase special study
that would attempt to improve the training program for smoke inspectors.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Opaci ty
Visible Emissions
Method 9
Transmissometers
Smoke generators
Smoke meters
visible emissions
training
3. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
45
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |