EPA-R2-72-099
68-02-0309/4-72
IN-STACK TRANSMISSOMETER TECHNIQUES
FOR MEASURING OPACITIES OF
PARTICULATE EMISSIONS
FROM STATIONARY SOURCES
by
Carl M. Peterson, Ph.D.
M. Tomaides, Ph.D.
ENVIRONMENTAL RESEARCH CORPORATION
3725 North Dunlap Street
St. Paul, Minnesota 55112
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
National Environmental Research Center
Division of Chemistry and Physics
Research Triangle Park,
North Carolina 27711
Contract No. 68-02-0309
April 1972
-------
IN-STACK TRANSMISSOMETER TECHNIQUES
FOR MEASURING OPACITIES OF
PARTICULATE EMISSIONS
FROM STATIONARY SOURCES
by
Carl M, Peterson, Ph.D.
M. Tomaides, Ph.D.
ENVIRONMENTAL RESEARCH CORPORATION
3725 North Dunlap Street
St. Paul, Minnesota 55112
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
National Environmental Research Center
Division of Chemistry and Physics
Research Triangle Park,
North Carolina 27711
Contract No. 68-02-0309
April 1972
-------
ABSTRACT
Field studies were conducted to obtain basic research data
as a base for developing design and performance specifica-
tions for transmissometers which are to be used to measure
smoke stack plume opacities. Tests, conducted on the stack
of a pulverized coal-fired power plant, were designed to
evaluate the influence of transmissometer illumination and
light receiving angles, and transmitted light wavelength on
in-stack opacity measurements and their correlation with
the stack plume opacity.
Two specially-designed transmissometers, one having a
small fixed illumination-viewing angle design and the other
having adjustable illumination and viewing angles, were
mounted on a cylindrical 145-inch diameter steel stack to
measure the in-stack opacity, A 0.5° telephotometer was
used to determine the out-of-stack opacity of the plume as
viewed from a distant river bank.
The results show a significant dependence of measured in-
stack transmittance as a function of illumination and
receiver angles. This dependence is most pronounced at
small angles. The measured transmittance increases with
increasing illumination and viewing angle, resulting in
5 and 46 percent errors for 53 and 60° angles, respectively,
when compared to true transmittance. The dependence of
measured transmittance as a function of illumination light
wavelength was also established.
Using a small fixed illumination^viewing angle design
transmissometer, a good correlation was obtained between
plume opacity and in-stack transmittance.
This report was submitted in fulfillment of Contract
68-02-0309 under the sponsorship of the Environmental
Protection Agency,
111
-------
CONTENTS
Section Page
I CONCLUSIONS 1
II RECOMMENDATIONS 3
III INTRODUCTION 5
IV OBJECTIVES 9
V INSTRUMENTATION 11
VI LABORATORY EVALUATION OF INSTRUMENTS 21
VII FIELD FACILITIES AND CONDITIONS 33
VIII RESULTS AND DISCUSSION 45
IX ACKNOWLEDGEMENTS 53
X REFERENCES 55
XI APPENDICES 57
-------
FIGURES
Number Page
1 Schematic and Basic Dimensions of the 12
Reference Transmissometer Transmitter
2 Schematic and Basic Dimensions of the 13
Reference Transmissometer Receiver
3 Photograph of the Reference 15
Transmitter Assembly
4 Schematic of the Experimental Trans- 17
missometer Transmitter
5 Schematic of the Experimental Trans- 19
missometer Receiver
6 Linearity of the Reference Transmisso- 24
meter Readout as Determined Using
Neutral Density Light Filters
7 Linearity of the Experimental Trans- 30
missometer Readout as Determined Using
Neutral Density Light Filters
8 Experimental Stack Dimensions and 34
Port Locations
9 Experimental Steel Stack Viewed Against 36
the Concrete Stack Used as the Contrasting
Target for the Telephotometer Measure-
ment of Plume Transmittance
10 Example of the Experimental Data 39
Recorded in the Field Test
11 Example of the Experimental Data 40
Recorded in the Field Test
12 Transmittance Measurements by 3° Angle 46
of View and 1.5° Illumination Angle
Transmissometer Inside Stack and 0.5°
Angle of View Telephotoraeter Outside
Stack
VI
-------
FIGURES (continued)
Number
13 In-Stack Experimental Transmittance 48
Detected at Various Transmitter and
Receiver Angles of the Experimental
Transmissometer Normalized for Constant
True Reference In-Stack Transmittance
of 0.807.
14 Summary of Interference Filter Tests. 51
Relationship of the Experimental In-
Stack Transmittance and Transmitter
Angle for Four Interference Light
Filters at 5° Receiver Angle.
Vi i
-------
TABLES
Number Page
1 Stability of Reference Transmissometer 25
With Varying Line Voltage
2 Stability of Reference Transmissometer 26
During Startup
3 Dark Current Signal of Experimental 28
Transmissometer Receiver
4 Calculated Accuracy of Transmittance 32
Measurements
5 Data on the Plume Transmittance vs 45
In-Stack Transmittance Correlation
VI 11
-------
SECTION I
CONCLUSIONS
The major conclusions drawn from the test results are:
1. The apparent transmittance of an aerosol as measured by
a transmissometer increases with the size of the transmitter
and receiver angles of the transmissometer. This is due to
the increase in the amount of scattered light detected by
the transmissometer as these angles increase in size.
Detection of scattered light by a transmissometer represents
an error in the measurement and should be minimized as much
as possible .
2. The measured transmittance can be singly influenced by
varying either the receiver or transmitter angle and both
these angles are nearly equally influential in effecting the
measured transmittance within 0-60°.
3. A positive error of about 5 percent was noted when 5°
receiver and transmitter angles were employed for transmit-
tance measurements of a pulverized coal-fired boiler effluent.
The error increased to about 46 percent when 60° receiver
and transmitter angles were employed. The tests were con-
ducted with an experimental transmissometer using a tungsten
lamp, a photoelectric detector, and a light beam diameter
(aperture size) of approximately four inches.
4. The transmittance of the pulverized coal-fired boiler
effluent studied was light wavelength dependent. A differ-
ence from 6 to 10 percent in transmittance was measured
when employing interference filters of 0.438 to 0.656 micron.
The 0.656 micron filter (red band) consistently provided the
lowest in-stack transmittance. This result would indicate
that the particles in the effluent were sufficiently large
to exceed the first maximum of the particle extinction curve
and/or significant spectral absorption effects were present.
5. Good agreement was found between the opacity of the plume
and the in-stack transmittance of the coal-fired boiler
emission. For high transmittance conditions (greater than
70 percent) ideal agreement was observed between remote
plume transmittance measurements by telephotometry of con-
trasting targets in back of the plume and in-stack trans-
mittance measurements by a transmissometer with a 3° receiv-
ing angle, 1.5° transmitter angle, and a tungsten lamp light
-------
diameter of 0,5 inch. For lower transmittance conditions
(less than 40 percent!, the in-^stack measurement was
approximately 4 percent higher. The error may be due to
an increase in the particle size and/or multiple scattering
effects at low transmittance. In either case, the error
would probably be reduced by employing a smaller receiving
angle .
-------
SECTION II
RECOMMENDATIONS
This program was limited to field tests performed at one
facility. It was not within the scope of the program to
extensively evaluate optical properties of fly ash under
varying size distribution and concentration as related to
the opacity measurement. It is recommended, however, that
such testing be carried out.
As found during tests, the measured value of in-stack
opacity depends strongly on the transmissometer illumina-
tion and viewing angles. It may also be dependent on the
light beam diameter used for the aerosol illumination.
Tests, or at least theoretical calculations, should be
carried out to evaluate the influence of this common design
parameter.
-------
SECTION III
INTRODUCTION
Since the turn of this century, a number of theorists have
concerned themselves with fundamental studies to explain,
predict or calculate the extinction and scattering of a
beam of light passing through a suspension of fine particles
or intercepting a single particle. Coincident with such
studies has been the development of methods, techniques
and instruments to measure the magnitude of such phenomena.
As a result, it is now possible to specify with reasonable
certainty the optical conditions for extinction and scatter-
ing measurements in systems of particles of undetermined
shape, size and composition, and to anticipate what quan-
titative information about the particles can be deduced
from such measurements. Hodkinson1 has clearly and con-
cisely presented a review and critique of the noted activity,
described methods of applying the evolved principles to
achieve suitable measurements and quantitatively inter-
preted results of his and fellow investigators. Conner
and Hodkinson2 have discussed in detail the optical prop-
erties and visual effects of smoke-stack plumes.
Properly measured and interpreted optical properties of
plumes can be related to the quantity and quality of
material contained in a plume. Capable of being inter-
preted in such a manner, a measure of the optical proper-
ties can be used to indicate the effectiveness of system
processes, applied control equipment and malfunction in
controls .
One of the basic plume optical properties is its optical
transmi ttance which is defined as the ratio of light flux
which reaches a light sensitive device (eye, photocell,
etc.) when .the flux from the source passes through the
plume and when it does not. The basic equation is given
by Equation (1) :
T = = exp (-naQt)
o
-------
where
T = transmittance
F = flux of transmitted light through plume
F = maximum flux of light measured in absence
of plume in line of source and sensor
n = the number of particles per unit volume
of air in the light path of length t_
through the aerosol
a = the projected area of one of these
particles
Q = the particle extinction coefficient or
efficiency factor defined as:
_ total flux scattered and absorbed by a particle
flux geometrically incident on the particle
The particle extinction coefficient or extinction effi-
ciency factor Q depends on the particle refractive index
relative to the surrounding medium, its shape and its size
relative to the wavelength usually expressed as a = rrd/A ,
where d is the particle diameter and A is the wavelength
of light in the medium surrounding it; a is termed a
particle size parameter.
Another property of the plume is its opacity which is a
measure of the light flux attenuated by the plume that can
be calculated from the transmittance as per Equation (2):
0 = 1-T (2)
The optical density (OD) of the same plume is expressed as
the logarithm of the transmittance or opacity as per
Equation (3):
OD = -log1QT = -Iog10(l-0) (3)
-------
From the discussion and results presented by Conner and
Hodkinson,2 a remote method of measuring plume opacity by
means of contrasting targets is described. Actually, this
is a calculated plume transmittance value (_T ) obtained by
Equation (4J: P
(4)
Bs and Bs' are the luminance measured by focusing with a
telephotometer on a specific target (.sky) respectively
clear and through the plume. B^ and B^' are the luminance
obtained when focusing on a second target (hill) respect-
ively clear and through the plume. Performed in this
manner, a defineable plume transmittance or opacity is
obtained. It is this value upon which the plume opacity is
judged and compliance to most control regulations is based.
Therefore, if in-stack transmittance measurements are to be
of value for determining plume opacity, they must be relat-
able to such plume measurements. A significant difference
between plume and in-stack opacity measurements may be
realized in the fact that the effective path length may be
different. In-stack opacity readings can be reduced in
practice to the plume opacity by use of Equation (5) :
(5!
where
0 = plume opacity
0 = in-stack opacity
£ = stack exit or plume diameter
H = in-stack transmissometer path length.
-------
SECTION IV
OBJECTIVES
A number of in-stack transmissometers are commercially
available for the continuous monitoring of in-stack gas
transmittance. However, due to the absence of defined
operation and performance specifications, there has been
little standardization in design of such instruments with
regard to illuminating and receiving angle, wavelength of
illumination, sensitivity and type of receiver, and relation-
ship of output to plume opacity. Environmental Protection
Agency (EPA) officials and users have questioned the
accuracy and meaning of the data generated by such an array
of uniquely-designed transmissometers. It was, therefore,
the object of this study to answer these questions by
developing an experimental transmissometer with extensive
research flexibility and collect and interpret data from
such a unit while operating it on a coal-burning electric
power plant stack.
The experimental transmissometer employed was flexible
enough in design and the effluent stack gases were of
adequate quality to determine the following:
1) The effect of transmitter and viewing angles
on the measurement of transmittance;
2) The dependence of illuminating wavelength on
the measurement of transmittance; and
3) The relationship between in-stack transmittance
and plume opacity as a function of in-stack
path length.
This report contains a description of the transmissometers
employed in the study, the experimental test program
applied, and the results achieved.
-------
SECTION V
INSTRUMENTATION
Two in-stack transmissometers Ca reference and an experi-
mental model) were designed, developed and employed in this
study to achieve the results reported. The reference model
was designed with small fixed transmitter and receiver
angles and was used to continuously monitor the transmittance
of gases flowing from the boiler to the atmosphere. The
experimental model was designed with sufficient flexibility
to permit changes to be made in the transmitter and viewing
angles and also to accommodate various interference filters.
All other equipment used in the course of the program
activity was commercially available.
Reference Transmissometer
The reference transmissometer consists of a transmitter
and a receiver. In operation the transmitter is mounted
on one wall of a duct or stack and is the source of illum-
ination. The receiver is mounted directly opposite the
transmitter and functions to measure that portion of the
radiant energy which is transmitted through the gas flow-
ing within the confines of the stack boundaries.
Schematics of the reference transmitter and receiver are
shown in Figures 1 and 2, respectively. The basic support
and housing for each is a rigid 3-inch I.D. aluminum tube,
flat-black anodized to eliminate excessive light reflection,
Within this tube and at defined separation distances, as
shown in the figures, are located the proper lens, pin
holes and photocell. The support tube was of sufficient
internal diameter so that all internal items can be first
mounted within an independent holding fixture. At two
points (diametrically opposite) and near the circumference
of each fixture, provisions were made to accept a quarter-
inch rod and locking set-screw. Two quarter-inch rods were
then attached to the first fixture and serially extended
through each succeeding fixture. The distance between
fixtures was readily measured and held in this position by
tightening the locking set-screws. The entire optical
assembly was then positioned in the support tube as shown
in Figures 1 and 2.
11
-------
to
Lamp Voltage
Regulating
Photovoltaic Cell
Pin Hole
0.0485 in. dia.
Smoke Stop
'0.5 in. dia. Flange
Tungsten Filament Lamp
IYODA - 8V5A
Flushing
Air
Figure 1. Schematic an4 Basic Dimensions of the
Reference Transmisaometer Transmitter.
-------
Flushing Air
- Connector
Selenium Cell
IR A15M
/-Pin Hole
0.04 in
dia .
I
f
w
^,
\v
2 . 39
3. 36
13. 5
Smoke Stop
1.5 in . dia
Figure 2
Schematic and Basic Dimensions of the
Reference Transmissome ter Receiver.
rfl
H
Tl
Flange
-------
As shown in Figure 1, the transmitter optical assembly
consists of a Tiyoda 8V5A square-shaped filament lamp, two
lenses, and a pinhole of 0.0485-inch diameter. The optical
system is protected against dust deposits by introducing
clean flushing air between the first .lens and the stack. A
smoke stop is an integral part of the support tube which
terminates in a flange used for attaching the assembly to
the stack wall.
The optical assembly tube and the illumination lamp are
mounted to a rigid frame to secure optic alignment during
instrument operation or transportation. This frame also
encompasses the electronic components of the instrument and
is entirely enclosed in an aluminum housing with a removable
cover for the convenience of electronic circuitry adjust-
ment and maintenance. Complete assembly of the reference
transmissometer transmitter is shown in Figure 3.
The optical portion of the reference receiver is designed
to be assembled in the same manner as the transmitter. The
receiver consists of a single lens protected from dust
deposits by a clean air ventilated smoke stop, followed by
a pinhole and a IR A15M selenium cell surface. The
electronic system of the transmitter/receiver is located
in the transmitter housing and performs two functions: it
provides a stabilized voltage to the lamp; and conditions
the photocell output signal. The wiring diagram for this
device is included as Appendix B of this report.
The stabilized lamp voltage power supply maintains the
Tiyoda 8V5A lamp at a nominal 6.5 volts. The lamp intensity
is continuously monitored and adjusted for fluctuations
and aging by a secondary photocell which is part of a feed-
back circuitry of the power supply.
The electrical signal-conditioning circuitry consists of a
temperature-stabilized solid-state amplifier. The
electrical signal is received at the conditioning circuit
from the receiver through a shielded weatherproof cable
and, after proper conditioning, is suitable for 0-10 milli-
volt recorder operation. Electrical jacks are provided on
the transmitter housing for continuously recording the
conditioned signal.
14
-------
I
Figure 3. Photograph of the Reference
Transmitter Assembly.
15
-------
Experimental Transmissometer
The separated transmitter and receiver concept was also
applied to the experimental transmissometer design.
Several techniques for varying the transmitter and
receiver angle of transmissometers were investigated and
the final design selected is shown in Figure 4. The
experimental transmissometer contains a lamp (Tiyoda 8V5A)
which can be moved along the transmitter support tube axis,
and the light from the lamp is collected by a 5-inch dia-
meter condenser lens. The light beam angle for the lamp
located close to the condenser is about 100°; while for a
lamp distance of about 6 inches from the condenser, the
light beam is nearly parallel and of very high intfinsity.
The aluminum support tube of 3-inch I.D. and all parts of
the optical system were flat-black anodized to eliminate
excessive light reflection.
The sheath air provided to this system simultaneously
protects the condenser lens surfaces and cools the lamp.
This arrangement has the advantage of increased sheath air
temperature which prevents water vapor condensation in
front of the transmitter condenser.
Possible light beam intensity changes caused by the lamp
aging or the lamp power supply fluctuations are sensed by
a small diameter photocell located close to the lamp. The
photocell output is used as a feedback to the lamp power
supply automatic control circuitry that maintains the
light intensity constant. The photocell temperature is
stabilized to within ±0.5°F by a small wire resistor which
is located in the photocell mount. The temperature is
sensed by a thermistor which operates a Rosemount Tempera-
ture Controller STO-21501 that functions to control the
energy to the resistor.
Similarly to the reference transmitter, the experimental
transmitter is also enclosed in an aluminum housing and
provided with power and low voltage shielded cord sockets.
The transmitter lamp voltage can be conveniently checked
on two outside terminals and adjusted to the voltage desired
upon removing the housing cover. The light beam angle can
be adjusted and the angle determined on a scale readable
through a small opening on the side of the transmitter
housing. The opening is sealed with a removable cover when
16
-------
Secondary
Sheath Air
Primary
Sheath
Spacers
-Condenser
Tungsten
!'i lament
i Lamp
Light Intensity
Control Photocell
Figure 4. Schematic of; th.e Experimental
Transmissometer Transmitter.
-------
in use. As calibrated, the transmitted light angle can be
varied within near parallel zero degree angle up to 100
degrees. The lamp power supply located in the transmitter
aluminum housing is of the same design as for the reference
transmitter; however, for the experimental transmitter, a
nominal lamp voltage of 8.0 volts was employed.
The final experimental transmissometer receiver design is
shown in Figure 5. It contains a double aspheric objective
lens that collects the light from the illuminated section
of the stack at angles from about 110 degrees down to 2
degrees. The exact angle is dependent on the iris opening
that is positioned between the objective and the light
focusing lens. As a light sensing device, a photovoltaic
cell A15M is used in connection with low noise solid-state
signal pre-amplifier, which increases the-total signal
amplification of the secondary amplifier located in the
transmitter housing by 10 and 100. The photovoltaic cell
is temperature-stabilized by the Rosemount Temperature
Controller STO-21501. The objective lenses are protected
by flushing air supplied behind the smoke stop of the
receiver.
The electronic circuitry of the experimental transmissometer
is shown in Appendix B.
Auxiliary Equipment
A telephotometer, Spectra Brightness Spot Meter Model SB
(0.5° type) by Photo Research Corporation, was used to
determine the plume opacity. The technique used in this
determination was that of contrasting targets.
A set of calibrated interference filters supplied by Optic
Technology, Inc. was used to evaluate spectral behavior
of the transmissometers. The specific major peak trans-
mittance for each of five respective filters was 436, 486,
579 and 656 millimicron.
Two sources of clean air to protect the optical systems
were constructed. Each source consists of a small blower
and an absolute filter located in a metal box. Electrical
heaters with proportional thermostatic control could be
18
-------
'Focusing
Lens
Photocel1
Pre-Amplifier
Aspheric
Obj e ctive
Interference
Filter
Heater
Thermostat
Figure 5. Schematic of the Experimental
Transmissometer Receiver.
-------
installed in each assembly if h,igh temperature flushing
sheath air were required. Air flow rate for each source
was about 7 cfm.
Two Bausch & Lomb VOM->5 single channel strip chart
recorders were used for simultaneous monitoring of the
experimental and reference transmissometer signals.
20
-------
SECTION VI
LABORATORY EVALUATION OF INSTRUMENTS
Functions of the instruments were checked and the trans-
missometers calibrated in the laboratory b'efore collecting
any field experimental data. The attention was mainly
centered on the calibration of the reference and experi-
mental transmissometer output when instruments were set to
an optical path equal to the diameter of the stack. For
each transmissometer the amplifier linearity was determined
and the experimental transmitter and receiving angle scale
factors were established. The data collected during labora-
tory calibration were used to calculate the accuracy and
reproducibility of both transmissometers.
Reference Transmissometer
The reference transmissometer of 1.5° fixed transmitter
angle was designed and used throughout the experiments. The
receiver component was first operated and evaluated in the
laboratory with a 3° receiving angle but later, when in-
stack transmittance data showed a strong dependence on the
magnitude of the receiver angle, the design changes were
made to decrease the angle of view down to about 0.8°.
To determine the instrument sensitivity the entrance of the
receiver was shielded and the amplifier output read at the
so-called "dark current" condition. The output voltage as
measured with a Keithley solid-state electrometer was within
0.0002 to 0.0003 volts. This value represents the minimum
accuracy of the photovoltaic cell amplfier readout.
Based on this calibration the minimum detectable transmit-
tance can be calculated. When the receiver and transmitter
were optically aligned and the transmitter lamp operated
at 6.5 volts, the receiver signal corresponding to clean air
conditions and light path of 10 feet was 0.24 volts and
for a 30-foot distance, 0.0357 volts. Supposing the ampli-
fier output reading of 0.002 volts has an accuracy of
±0.0002 volts, that is +10 percent, then the transmittance
determined with the same accuracy would be:
21
-------
0.002 x 100
T = 0357 = percent for 30 ft distance
and
T = ' o 2400 ~ 0.83 percent for 10 ft distance.
The accuracy in measurement improves rapidly with increas-
ing trans-mittance and since the transmittance values below
twenty percent are not anticipated or considered allowable
during normal stack emission rates, the performance is
considered to be acceptable.
The voltage stability of the photocell amplifier was deter-
mined by step-wise varying the amplifier input line voltage
from 115 to 100 volts. No detectable change in amplifier
output was detected. The drift of the amplifier during
startup was less than Q.5 percent of the output reading and
was stable after one hour of amplifier operation.
Similarly, the lamp power supply was found to be stabilized
after one hour of operation. To simulate field operation
and noise pickup, the transmitter and receiver were sep-
arated a distance of 9 feet, the light beam properly
aligned, and the photocell amplifier output measured while
decreasing the input line voltage of the power supply to
87 percent of its initial value. The corresponding change
in the amplifier output was less than 0,4 percent. This
represented a combined performance of the photocell
amplifier and light source stability.
The photocell and amplifier circuit were selected and
designed to provide for linearity between illumination and
amplifier output. Two linearity checks were performed to
determine the true relationship between the photocell
illumination intensity and the amplifier output voltage.
The first check was performed by utilizing a 15-watt high
intensity bulb as a point source of light and causing this
light to be focused on the photocell from various known
distances and at various levels of intensity. The ampli-
fier output was measured and recorded at each distance
and intensity. For a linear illumination-amplifier output
to exist, the amplifier output needs to be inversely
proportional to the square of the lamp receiver distance
during constant light source intensity. This was found to
be experimentally true. When the lamp intensity was
22
-------
varied by varying the voltage to the lamp, the photocell
circuit linearity characteristics were determined for very
low photocell illumination. By this method the linearity
was shown to exist for amplifier output voltages which are
greater than 0.0002 volts.
The second photocell amplifier linearity readout check in
the laboratory consisted of separating the transmitter and
receiver a distance of approximately 12 feet (145 inches)
and inserting neutral density filters in the transmitter
light beam. The transmitter angle employed in this check
was 1.5°, and a receiver angle of view of 0.8°. Six filters
having respective neutral densities of 1.0, 0.7, 0.6, 0.4,
0.2 and 0.1 were serially inserted into the light beam
and the corresponding output voltage recorded. For each
respective neutral density filter the corresponding filter
transmittance is given as 10, 20, 25, 40, 63 and 80 percent.
The results of plotting known filter transmittance versus
reference amplifier output in terms of volts is given in
Figure 6 .
The angle of view of the receiver was determined by
assembling the receiver in a horizontal position at the
center of a circle 46 feet in diameter. The distance from
the receiver to the circumference was 23 feet and a move-
ment of 4.8 inches along the circumference represented a
one degree angle change. This formation was used in deter-
mining the angle of view for a small 1.5-inch diameter high
intensity light source which was positioned on the 23-foot
arc and focused to the receiver. The maximum photocell
output was noted and the light source was moved in both
directions along the arc until a minimum photocell output
was recorded. In the first receiver design, a movement of
7.2 inches of the source either side of the zero angle was
required to produce the minimum output reading. This
corresponded to a total movement of 14.4 inches or an angle
of view of 3 degrees. When this same procedure was
employed for the receiver after it was redesigned to reduce
the angle of view, it was found that total movement of
3.8 inches caused the receiver to drop to the minimum out-
put value. This showed that the angle of view was then
0.8 degrees .
23
-------
100
0)
u
G
NJ 4J
£> 4J
H
6
in
c
Q)
-P
50
lTr.an!smiut«r i
t * ' >
1 \-~-l- ~-|
0.1
Reference Transmissometer Output, v
0. 2
Figure 6. Linearity of the Reference Transmissometer Readout as Determined
Using Neutral Density Light Filters.
-------
Prior to installation on the stack, the reference trans-
missometer was assembled and aligned in the laboratory with
an optical path length of 145 inches to obtain a clean air
calibration of the entire assembly. This calibration was
performed in the laboratory after sunset to obtain dark
room conditions. The transmissometer operating character-
istics determined for the 3 degree receiver angle were
found to be the following during this final laboratory
calibration:
1) The collimated transmitter light beam focused
on the receiver was 2.5 inches in diameter
at the receiver.
2) Operational stability was found to be excellent
'over a line input voltage variation from 107
to 116 volts. At 104 volt input, the lamp
voltage decreases to 6.45 volts from the
nominal 6.5 volt value. The resultant decrease
in lamp intensity and possible degradation in
photocell amplifier circuit at this lower input
voltage caused a decrease in photocell output
voltage of 0.007 volts or 1.8 percent of nominal
value. Increasing the line voltage to 125 volts
caused the photocell output for the same con-
trolled lamp voltage to increase by 0.003 volts
or 0.8 percent of nominal value. The results
of this stability check with regard to line
voltage change is shown in Table 1.
3) The reference transmissometer startup and
stabilizing period is ten to fifteen minutes.
The lamp voltage and photocell output readings
recorded during the period of time from
applying a constant 116 volt input until stable
readings are recorded is shown in Table 2.
Table 1. Stability of Reference Transmissometer
With Varying Line Voltages.
Input
[volts)
104
107
116
125
Lamp
[volts)
6 .45
6.5
6 .5
6 .5
Electrometer Reading
(volts)
0 . 388
0.395
0 . 395
0 .398
25
-------
Table 2. Stability of Reference Transmissometer
During Startup.
Time from Start Lamp Electrometer Reading
(min) (volts) (volts)
0 6.3 0 .382
5 6.4 0 .390
10 6.5 0. 395
15 6.5 0.395
*
Line voltage constant to 115 volts
Transmissometer performance characteristics were also
established for the 0.8° receiver angle and with the
exception that the stable (nominal clean air) photocell
amplifier output voltage decreased to 0.195 volts, the
stability remained the same as for the 3° receiver angle.
Nearly identical percent changes in output were found for
variations in line voltage in this case.
Experimental Transmissometer
Laboratory evaluations of the experimental transmissometer
were performed to determine system stability, angle of
transmittance, angle of view, dark current photocell output,
and to calibrate the output and determine the accuracy of
the instrument.
A lamp power supply voltage of 8.0 volts is the desired
operational value and this voltage remains constant with a
line input voltage to the power supply over a range of
103 to 130 volts. Over the same range of voltages, the
amplifier output signal is constant for each fixed angle
of receiver and transmitter. This range of stability was
established by repeatedly measuring the lamp voltage and
recording the amplifier output for various incremental
settings of the power supply input voltage.
The various transmitter light beam angles are achieved by
adjusting the light source to a fixed distance from the
condenser lens. The light source is mounted in a light
tube which slides internal to the aluminum cylindrical
housing containing the condenser lens. Therefore, it was
26
-------
necessary to determine in the laboratory the position of
the light tube with respect to the lens. Once the respec-
tive distance and position was established for each desired
light beam angle, the outer surface of the light tube was
scribed and labeled. Establishment of the light tube for
each light beam angle (10° increments) was achieved in the
laboratory by placing the transmitter a fixed distance from
a plain dark wall and focusing the positioned light source/
condenser beam onto the wall. The resultant light beam
diameter on the wall was measured and, knowing the distance
from the source to the wall, the sine/cosine relationships
were employed to determine the resultant transmitter light
beam angle. In actuality, the required light beam diameters
for each desired light beam angle were marked on the wall
and the light tube adjusted until that diameter beam was
obtained. By this method the transmitter was calibrated
for zero degree and in ten degree increments up to the
maximum beam angle of 100°.
The experimental transmitter viewing angle range was estab-
lished in the same manner as that described for the refer-
ence transmissometer. By this method it was found that the
minimum viewing angle of the experimental transmissometer
was 2° and the maximum was 120°.
For various viewing angles within this range, the viewing
angle adjustment mechanism was calibrated and scaled to
provide for 10° incremental changes. Below 5° a special
pinhole plate was constructed and needs to replace the iris
to achieve the 2° angle of view.
The "dark current" reading of the photocell, preamplifier
and amplifier circuitry of the experimental transmissometer
was determined by eliminating all light to the photocell
and measuring the resultant amplifier output signal.
Readings were obtained for all three preamplifier range
settings with a Keithley 502 electrometer. The results
are shown in Table 3.
27
-------
Table 3. Dark Current Signal of
Experimental Transmissometer
Receiver.
Preamplifier Dark Current Reading
Range (volts)
XI -0 .00075±0 .00005
X10 -0.0065 ±0.0005
X100 -0.0660 iO.005
As shown by the table, for all preamplifier ranges the
"dark current" phenomena yields a negative voltage signal.
These readings are characteristic of most photocells and
the value is magnified by the increased preamplification.
However, because of its small value it was found to be
difficult to eliminate electronically and achieve a true
base zero for no illumination. Therefore, for accurate
transmittance determinations the zero point offset must be
corrected for arithmetically. In use, the measured values
for any transmittance are increased by the amount of the
absolute "dark current" amplifier reading specific for the
amplifier range selected (XI, X10, X100).
The linearity of the photocell amplifier circuitry was
determined by the same two(point light source and neutral
density filters) techniques as described for the reference
transmissometer. For the point light source technique the
quantity of light reaching the photocell was varied by
changing the distance between source and the receiver. The
intensity of light to the receiver follows the inverse
square law as given by Equation (6) :
E = KL~2 (6)
where
E = amplifier output, volts
K = constant
L = optical path between source and receiver, feet.
28
-------
The corollary to Equation (6) with the data obtained for
each of the three preamplifier ranges was:
2 3 8
E = KL ' for the XI range over values of E
from 0.0014 to 0.028
-2 19
E = KL ' for the X10 range over values of E
from 0.018 to 0.28
-2 19
E = KL " for the X100 range over values of E
from 0.031 to 2.8.
The correlation coefficient for each set of linearity
measurements were within 0.995 to 0.999. Even though the
linearity was not defined at the "dark current" values, it
is expected from the experience with the reference trans-
missometer that acceptable linearity is also achieved at
lower levels of illumination.
In the case of the neutral density filter technique to
determine photocell and related circuitry linearity, tests
were performed for 2° angle of view and for transmitter
angles from 0 to 60°. As shown in Figure 7, the linearity
is good over the range of transmitter angles from 10° to
60°. At the near 0-P transmitter angle the generated
data was not of sufficient quality to establish a good
linear relationship. The reason for this condition was
that the condenser of the light energy created a discolor-
ation with the center and edges of the light beam at this
angle of focus and ND filters used appeared to be color-
sensi tive.
A calibration of the experimental transmissometer was per-
formed to determine the experimental transmitter light
intensity at the optical distance for the transmitter and
receiver of 145 inches, and for receiver light angles of
2, 4, 5, 6.5, 8.5, and 16°. Additionally, the transmitter
light angle was varied over the range of from 0° to 100° in
10 degree increments. The receiver photocell amplifier
output corresponding to each transmitter light angle used
was measured by a Keithley 502 electrometer with no inter-
ference filters inserted in the receiver optical system,
and the procedures repeated for 0.436, 0.486, 0.534 and
0.656 micron light interference filters.
29
-------
100
OJ
o
0)
o
a
td
e
w
0)
-p
1-t
H
0.1
0.01
0.2
0.02
0.3
0.03
Experimental Transmissometer Output, v
Figure 7..
Linearity of the Experimental Transmissometer Readout as
Determined Using Neutral Density Light Filters.
-------
The illumination lamp voltage of 8.0 volts was set and
used throughout the experiments. A near linear relation-
ship between the amplifier output signal and the transmitter
light angle was found, except for the near 0° transmitter angle
For any given receiver angle this relationship can be
expressed by Equation (7):
where
E = K x e
E = amplifier output, volts
K = constant
£ = transmitter light angle, degrees
b = exponent
(7)
The results of the experimental transmissometer output
calibration are tabulated and presented in Appendix A of
this report. The data are presented for transmitter angles
from 0 to 60 degrees and receiver angles 2, 4 and 6.5
degrees. For larger receiver angles, the photocell output
signal for 6.5 degrees can be used because it was found
that the output is no longer dependent on the receiver
angle magnitude above this value.
Using the calibration data the accuracy of in-stack trans-
mittance determination can be calculated. The accuracy is
generally decreasing with increasing transmitter light beam
angle because the light intensity as seen by the photocell
also decreases. This same tendency is also seen when
interference filters are used in the receiver.
Table 4 contains the calculated accuracy for a given trans-
mittance when the transmissometer is operated at 100
degrees transmitter angle, 4 degrees receiver angle and with
various interference filters in position.
31
-------
Table 4. Calculated Accuracy of
Transmittance Measurements.
Illuminat ion
of Filter
rt Accuracy as
Transmittance Percent of
(%) Transmittance
Accuracy of
Transmittance
White
0.436
0.486
0.534
0.656
Light
ym
ym
ym
ym
0 .8
30
10
5
5
±10
±15
±5
+ 2 .5
±2 .5
±o
±4
±0
±0
+ 0
.08
.5
.5
.13
.13
Measured with 100° transmittance angle, 4° receiver angle
32
-------
SECTION VII
FIELD TESTS
Field Facilities and Conditions
The number eleven pulverized coal-burning 100 MW boiler and
associated stack located at the Northern States Power
Company Highbridge Plant in St. Paul, Minnesota was selected
and approved as the test site. Tests performed in January
of 1970 on this boiler revealed the following data as
measured downstream of the electrostatic precipitator : gas
flow of 388,000 scfm; stack gas temperature about 200°F;
average dust concentration 0.095 gms/ft^; precipitator
collection efficiency approximately 93 percent.
As shown schematically in Figure 8, this stack is of
cylindrical steel construction approximately 12 feet in
diameter and 292 feet tall. The measured internal diameter
of the stack is 145 inches and this dimension is also the
effective light path length with the transmissometers
located in position. The stack protrudes through the
boiler room roof which is 110 feet above the ground. Within
the boiler room, the boiler gases pass to a split flow
electrostatic precipitator from which two separate gas
streams exit and enter the stack just above the roof line
through two diametrically opposite but identical ducting
systems. The experimental transmissometer was positioned
on the stack 50 feet above the roof. The reference trans-
missometer was located 12 feet higher up. The 12-foot
spacing between transmissometers was determined to be
sufficient to prevent any interference between transmitters
during simultaneous operation.
Both transmitters and receivers were mounted on the stack
wall using directionally adjustable ports and special
brackets welded to the stack wall supporting the instru-
ments and locking them in the position desired, special
attention was given to locating the optical system of the
experimental transmitter and receiver close to the gas
stream flow boundaries to permit the evaluation of as large
transmitting light angles as possible.
33
-------
"Reference
Tr a n s n i s s o rn e t e r
Experimental
Transmissoneter
L4 5 " I . D .
Boiler Room
Roof
Figure 8. Experimental Stack Dimensions
and Port Locations.
34
-------
Two separate scaffoldings attached to the stack provided
a base for the transmitter and receiver operators. The
transmissometers were electrically connected with suffi-
cient lengths of shielded cabling to the readout instru-
ments located in a shelter on the boiler room roof.
During the field evaluation the boiler was energized with
a mixture of approximately 10 percent Western Montana
powdered coal and 90 percent Illinois coal. Boiler opera-
tion was steady during the periods of testing, but electro-
static precipitator rapping, performed at one-minute
intervals for about 15 seconds, caused some fluctuations
in the actual stack gas transmittance, as seen in Figures
10 and 11. The plant supervisor and engineer allowed the
investigators to vary the electrostatic precipitator's
high voltage power supply to obtain plume opacities within
limits of 4Q to 90 percent. This was helpful in obtaining
data for a wide range of the plume opacities, but evalua-
tion and interpretation of results became more difficult.
Th_is difficulty may have arisen from a change in the
particle collection efficiency of the precipitator for
various-sized particles, as the voltage was varied, causing
variations in the effluent particle size distribution.
Having no quantitative information about particle size
changes, no consideration could be given to the changed
optical properties of particulates in relation to the plume
transmittance.
The location which provided the best view of the plume and
permitted optimum opportunity for obtaining contrasting
target plume opacity measurements was a river bank located
about 700 feet southeast of stack number eleven. From this
location the plume, as it exited from the stack, was
readily visible. A concrete stack under construction at
the time extended above and to the left of stack number
eleven and served as the contrasting target for the proper
plume opacity determinations. A photograph of the two
stacks as taken from the plume viewing location is shown
in Figure 9.
35
-------
Figure 9. Experimental Steel Stack* Viewed Against the
Concrete Stack Used as the Contrasting Target
for the Teleplxotometer Measurement of Plume
Transmittance,
36
-------
Field Optical Alignment of Instruments
Each time the transmissometer was installed on the stack,
the following procedure was followed to assure proper
alignment of the instruments:
1) The transmitter light beam angle was set
to the minimum value to obtain a collimated
light beam.
2) The covers over the closed mounting ports
were opened and the transmitter was posi-
tioned in the holder.
3) The transmitter was energized and aligned
so that the light beam was observed through
the receiver port opening. Alignment is
achieved by the observer on the receiver
port side viewing the beam and directing the
individual controlling the transmitter to
perform the necessary adjustment until the
maximum light intensity is visually realized.
4) The receiver is then positioned in its holder
and slowly adjusted up-down and left-right
until the maximum photocell output signal is
establi shed.
5) When proper alignment is achieved, the
transmitter and receiver are physically
locked in position.
During the course of the field tests, each transmissometer
was removed and re-installed on the stack several times.
It was found that near identical alignment and reproduci-
bility in results could be achieved very rapidly by follow-
ing the above-noted procedure. The need for repeated in-
stallation was necessary because the demand for electrical
energy by the customers of Northern States Power Company
would not permit a shut-down of the boiler for periodic
clean air baseline photocell measurements. Furthermore,
rigid routine lens and other component inspection schedules
were maintained to eliminate the possibility of obtaining
erroneous information. The air flushing system of the
transmissometers performed properly for all components
except the experimental transmitter, where it was necessary
37
-------
to essentially expose the condenser lens to the flue gas
particles in order to obtain maximum transmitter angles.
Due to the close proximity of lens and contaminated
particles, this transmitter was removed during testing
activities every 60 minutes for particulate cleaning if
needed and examination of the lens.
Field Data Procurement
With the transmissometers properly aligned and operating,
the photocell readout from each transmissometer was
recorded on a Bausch & Lomb VOM-5 strip chart recorder.
Examples of the resultant chart recordings obtained are
shown in Figures 10 and 11, The reference transmissometer
output recording was readily related to the plume opacity
and used to determine true in-stack transmittance. The
experimental transmissometer readings were related to the
reference transmittance to evaluate the effect of the
various modes of operation.
Chart recordings of Figures 10 and 11 show only the period
of opacity fluctuations during the time electrostatic
precipitator rapping was occurring. The elapsed time is
approximately 15 to 20 seconds, and each of these periods
were followed by relatively steady opacity period of about
40 to 45 seconds duration.
For most experiments the sequence of the data taking was to
scan the transmitter light beam angles from 0 to 60 degrees
for a constant angle of view with and without interference
light filters in the receiver. The angles of view of the
experimental receiver that were tested were: 2, 3, 4, 5, 6.5,
8.5, 10, 16, 20, 30, 40, 50 and 60 degrees. Transmitter
angles of near 0, 10, 20, 40 and 60 degrees were scanned.
Weather conditions permitting, telephotometer measurements
of the out-of-stack transmittance were taken using the
technique of contrasting targets.
During some of the tests, samples were collected from the
stack through the experimental transmissometer port in order
to obtain some information on the size distribution of the
effluent particulates . An isokinetic probe, incorporating
a filter holder containing a glass fiber absolute filter
38
-------
Run ;
Transmitter:
Receiver:
Filter:
lOb - 3/22/72
10°
40°
None
EXPERIMENTAL TRANSMISSOMETER
1
i "
1
i
\ *
-j
1
1 o
. t- - - -
' V
j £~~
1
I
-' .0
[
"
-
_-
-
-
. ....
"
1
_....
-
}
L..
A
... j
' 1
:..
- --
. Ul
-
tr -
-
~ "
.
L .
_
_:_:"
--
~~
tj
L
-
-
^_ .
^1
-' -
--'
T :
t-rr-
h
LlJ
- __
-
_^r:_
r::_
^
^-1
.
3~
~^.~-T~~
-Z."
..
i
_._
i
-
-
pf
f
ir.
__
L-1
....
r
_.-__
:-:-----
T"__
K--
UT
_^
;
-
L _..
:.-.
-7
. -.
--
3 '.-.-
l_J
Time, seconds
REFERENCE TRANSMISSOMETER
Figure 1Q. Example of the Experimental Data
Recorded in the Field Test.
39
-------
Run: 5a - 3/22/72
Transmitters 0°
Receiver: 3°
Filter: 0.656ym
EXPERIMENTAL TRANSMISSOMETER
Time, seconds
REFERENCE TRANSMISSOMETER
-3-
2-
-L-O
Time, seconds
Figure n. Example of the Experimental Data
Recorded in the Field Test.
40,
-------
(.Gelman Type E) was located in the gas flow, 4 feet from the
stack wall. The probe was allowed to equilibrate to the
stack temperature and a sample was drawn through the filter
where the particulate collection occurred. The dust col-
lected was evaluated for particle size using optical and
electron microscopy.
Evaluation of Field Transmittance Data
The basic parameters measured and recorded in the field for
further evaluation were the reference and experimental trans-
missometer output signals and the telephotometer readings.
The reference and experimental transmissometer measurements
are used to calculate the in-stack transmittance and to
describe the effects of operating the experimental trans-
missometer with defined operational parameters (i.e.,
transmitter and viewing angles). The reference in-stack
transmittance is calculated by use of Equation (8) :
T = -
R E
RT
RTO
where
T = calculated in-stack reference transmittance,
dimensionle ss
RT
RTO
measured in-stack reference transmissometer
photocell output signal, volts
reference transmissometer photocell output
signal for 100 percent transmittance. Deter-
mined in the laboratory for a defined optical
path of 145 inches and for the transmitter
angle of 1.5°. For 3° receiver angle, ERTO =
+0.240 volts; for 0.8° receiver angle the
RTO
value is +0.195 volts
The experimental in-stack transmittance is calculated by
means of Equation (9):
T =
E E
ET
(9)
ETO
41
-------
where
T = experimental in-stack transmittance,
dimensionless
ET
ETO
"dark current" adjusted photocell
output, volts
"dark current" adjusted calibration
photocell output for laboratory determined
100 percent transmittance at a given
transmissometer setting, volts
EET was determined by adding the established "dark current"
reading to the measured value. The "dark current" value to
be added is dependent on the preamplifier scale used to
obtain the measurements. The proper values as determined
in the laboratory are given in Appendix A.
Different EETQ values were developed in the laboratory for
each defined receiver and transmitter angle listed and for
each interference filter employed. The established values
are given in Tables 1 through 3 of Appendix A.
In spite of relatively good short term flue gas stability,
changes in transmi ttance were detected by the reference
transmissometer during prolonged periods of data taking.
Therefore, in order to compare data from one readout period
to the next, the experimental transmittance data was
normalized by applying the proper values in Equation (10) :
norm
x T = x T
T R T R
(10)
where
norm
T
= normalized experimental transmittance, %
= measured experimental transmittance, %
= reference transmittance corresponding to
42
-------
T = average reference transmittance for the
defined array of measured experimental
transmittance values being normalized
T = T /T relative in-stack transmittance.
R E
By this method of normalization the normalized experimental
transmittance values are based on an average reference trans-
mittance for the defined array of measured experimental
transmittance values, and all values so normalized can be
compositely compared and analyzed.
The plume transmittance was calculated from Equation (4) .
43
-------
SECTION VIII
RESULTS AND DISCUSSION
Plume and Reference In-Stack Transmittance Relationship
The average values of the reference in-stack and telephoto-
meter plume transmittance are presented in Table 5 and
plotted in Figure 12. These results are for the reference
transmissometer operated with a 1.5° transmitter angle and
a 3° angle of view. As seen from results in Figure 12, a
plot of the experimental data provides a line which has a
slope just above unity. With perfect agreement, the data
should have fallen on the unit slope line. It can be seen
by the data points that for transmittance close to 100 per-
cent this is indeed realized. For lower transmittances , the
measured in-stack transmittance was slightly higher. This
could be due to the detection of scattered light by the 3°
angle of view of the in-stack transmissometer used in these
experiments.
Table 5. Data on the Plume Transmittance vs.
In-Stack Transmittance Correlation.
NOTE: Plume transmittance measured by Spectra SB 1/2
Brightness Spot Meter by means of contrasting
targets technique; angle of view 0.5°. In-stack
transmittance measured at 1.5° transmitter angle
and about 3° of the angle of view.
Out-of-Stack In-Stack
Transmittance Transmittance
84 87
80 81
78 76
70 74
69 68
53 56
48 49
47 53
41 46
36 38
Linear regression of the data: Slope = 1.049; y intercept =
-5.316; correlation coefficient = 0.99.
45
-------
100
o
u
c
(8
-P
4J
H
e
CJ
C
flj
U
IT)
4J
O
I
4J
50
100
In-Stack Transmittance, %
Figure 12.
Transmittance Measurements by 3° Angle of
View and 1.5° Illumination Angle Trans-
missometer Inside Stack and 0.5° Angle of
View Telephotometer Outside Stack,
46
-------
In-Stack Transmittance vs. Experimental Transmitter
and Receiver Angle Characteristics
As previously discussed, all of the experimental trans-
mittance data were normalized prior to studying the effect
of the various transmitter and receiver angles.
A plot of the normalized in-stack experimental transit-
tance versus the operational receiver angles is shown in
Figure 13 for various transmitter angles. The line des-
cribing the transmittance for a near 0° transmitter angle
is shown as a dashed line to indicate that it is based on
widely scattered experimental data due to the effect of
light beam discoloration caused by the condenser lens at
this small angle. However, the data clearly indicated the
same trend as shown for the larger transmitter angles.
The data shown in this plot demonstrates that the measured
transmittance is strongly dependent on the receiving and
transmitting angles. The contribution of light scattered
to the receiver from particles inside and outside the field
of the receiver view is given as the reason the experimental
transmittance is always higher than the corresponding
reference in-stack transmittance.
From Figure 13 the error due to any combination of trans-
mitter and receiver angle can be determined. For example,
the error is about 9 percent for a receiver angle of 5°
and a 10° transmitter angle.
Data similar to that presented in Figure 13 but for lower
values of average in-stack reference transmittance were
reviewed during the course of the study and similar type
curves were generated. For lower in-stack transmittance
the difference between reference transmittance and experi-
mental transmittance was greater for all combinations of
receiver and transmitter angles investigated. This phen-
omena may be due to an increase in size of particles in the
stack at the time of the lower transmittance measurements.
The lower transmittance in the stack was created by decreas-
ing the voltage in the electrostatic precipitator and,
consequently, it is very likely that a difference size
distribution of particles would penetrate the precipitator.
A changed particle size distribution flowing through the
light beam would reflect a change in the relative amount
of light scattered to the receiver. Ensor and Pilat have
shown theoretically the effect of particle size and trans-
missometer receiving angles on transmissometer measurements.
47
-------
CO
o
0)
N
(0
6
0)
o
c
E
ui
c
rfl
ra
4J
0)
cu
X
w
1.2-
1.1
1.0
0.9
0.8 -
0.7
Range of In-Stack Reference
Transmittance: 0.607-0.880
Average In-Stack Reference
Transmittance: 0.807
Interference Filter: None
Run: 1, 6, 7, 8, 9, 12
of 4/19/72
Reference Transmissometer
Viewing Angle: 0.8°
Illumination Angle: 1.5°
Experimental Transmittance
Normalized to 0.807 In-Stack
Reference Transmittance
10
20
50
60
Figure 13
30 40
Receiver Angle. °
In-Stack Experimental Transmittance Detected at
Various Transmitter and Receiver Angles of the
Experimental Transmissometer Normalized for True
Reference In-Stack Transmittance of 0.807.
-------
These results show that both the receiver and transmitter
angles of a transmissometer should be as small as practical
to restrict the detection of scattered light and the
associated error.
In-Stack Transmittance-Wavelength Characteristics
Tests with four different light interference filters,
namely 0.436, 0.486, 0.579, and 0.656 micron wavelength,
inserted in the light path before the experimental receiver
photocell were performed to define the measured trans-
mittance as a function of wavelength. The results of this
group of experiments are seen in Figure 14 where the nor-
malized experimental transmittance is plotted versus trans-
mitter angle for all four filters evaluated. These plots
show the dependence of the in-stack transmittance on wave-
length. The 0.656 micron interference filter yields the
lowest in-stack transmittance. The transmittance increases
with each incremental decrease in wavelength evaluated.
This result would indicate that spectral absorption effects
were present and/or the particulates were larger than about
one micron and sufficiently monodisperse to observe oscil-
lation of extinction efficiency with wavelength. Micro-
scopic analyses of particulate samples indicated the
number mean diameter of around 1.4 micron and geometric
standard deviation of around 1.6.
The trend for the difference between the measured experi-
mental transmittance and reference or plume transmittance
to become greater for generally decreasing transmittance
was also seen in this set of data. Also, the value of the
experimental measured transmittance is more dependent on
the magnitude of the receiver angle when filters are
employed. This phenomena results in generally larger
experimental transmittance values to be measured with
relation to the reference transmittance, and may be again
caused by differences in the effluent particle size distri-
bution influencing the light scattering characteristics.
Secondly, the rate of change in the difference between
experimental and reference transmittance is greater for
increasing receiver and transmitter angles when filters are
utilized. This phenomena may be caused by different light
scattering characteristics of the effluent, due to larger
fly ash particles present during the "interference filters"
experiments.
49
-------
d
0)
N
rH
O
S3
ID
O
C
td
4J
jj
H
s
0)
q
fd
rH
<0
P
C
Hi
e
H
1-1
X
w
1.3
1.2
1 . 1
1 .0
0.9
0. 8
Jk;i trans a4..i;rahca:;D
f -t ' -::!::;:!::: ::;
Interference Filters:
x -0.4 36ym
O - 0.486ym
A - 0 . 579ym
D - 0.65Gvim
Experimental Receiver
Angle: 5°
Out-of-Stack Transmittance:
0 .41-0.53
Reference Transmissometer
Viewing Angle: 3°
Illumination Angle: 1.5
Experimental Transmittance
Normalized to 0.807 In-
Stack Reference
Transmittance
30 40
Transmitter Angle,
60
Figure 14. Summary of Interference Filter Tests. Relationship
of the Experimental In-Stack Transmittance and
Transmitter Angle for Four Interference Light Filters
at 5° Receiver Angle.
-------
Evaluation of the transmittance data shows that the
closest agreement between in-stack and plume transmittance
measurements are achieved when either white light or an
interference filter between 0.486 and 0.579 micron is used
52
-------
SECTION IX
ACKNOWLEDGEMENTS
The work described in this report was performed by
Environmental Research Corporation (ERC), St. Paul,
Minnesota, under EPA Contract No. 68-02-0309 by a team
consisting of Dr. Carl M. Peterson, Dr. M. Tomaides, Ben
Wahi, Keith Rust, Bob McKimmy, Len Graf and Roger Johnson.
The contract was administered under the direction of the
U. S. Environmental Protection Agency (EPA), with Mr.
William D. Conner as Project Officer. The authors wish to
thank Mr. Conner for his valuable contributions and
guidance.
Also, they wish to express their appreciation to all
workers of Northern States Power Company for their kind
assistance throughout the field experiments; namely, to
Mr. Dave Williams, Patrick Gordon and Tom Anderson.
53
-------
SECTION X
REFERENCES
1. Hodkinson, J.R., "The Optical Measurement of Aerosols"
In: Aerosol Science. Edited by C.R. Davies, Chapter
X, p. 287, Academic Press, New York, New York 1966.
2. Conner, W.D. , and J.R. Hodkinson (1966) , "Observations
on the Optical Properties and Visual Effects of Smoke
Plumes", Environmental Health Service, U.S. Govt.
Printing Office, PHS Publication No.999-AP-30.
3. Ensor, D.S. and M.J. Pilat (1971), "The Effect of
Particle Size Distribution on Light Transmittance
Measurement", Am. Ind. Hyg. Assoc. J., 32, 287-292.
55
-------
SECTION XI
APPENDICES
Page
A - EXPERIMENTAL DATA 59
Figure A-l: 0.436 micron Interference Filter 64
Results
Figure A-2 : 0.486 micron Interference Filter 65
Results
Figure A-3: 0.579 micron Interference Filter 66
Results
Figure A-4 : 0.656 micron Interference Filter 67
Results
Figure A-5 : No Interference Filter Results 69
Tables A-l through A-3: Results of Experimental 61-
Transmissometer Calibration 63
Tables A-4 through A-15: In-Stack Transmis- 70-
someter Field Results 81
B - NOTES TO THE REFERENCE AND EXPERIMENTAL 83
TRANSMISSOMETER OPERATION
Figure B-l : Reference Transmissometer Wiring 86
Diagram
Figure B-2; Experimental Transmissometer 87
Wiring Diagram
57
-------
APPENDIX A
EXPERIMENTAL DATA
Results of Experimental Transmissometer Calibration
Results of laboratory experimental transmi ssometer calibra-
tion are presented. When calibrating the instrument in the
laboratory, the transmitter and receiver were optically
aligned and firmly attached in that position on two sepa-
rate heavy laboratory benches, 145 inches apart. The output
of the receiver photocell signal amplifier was monitored
with a high input impedance Keithley 602 Electrometer and
by a Bausch & Lomb strip chart recorder. For constant
setting of the receiver angle of 2, 3, 4, 5 and 6.5 degrees,
the transmitter angle was adjusted from 0 to 100 degrees,
and for each angle, the amplifier output measured. This
procedure was repeated for four different interference
light filters after taking data for the transmi ssometer
operation without any light filter in the line. The data
taken were corrected for "dark current" photocell operation
that was determined by closing the receiver inlet and
reading the amplifier output signal under this condition.
The "dark current" was -0.00075 volt for "low" preamplifier
setting on the receiver panel, and -0.00650 volt for
"medium" switch position.
To correct the amplifier output readings, the following
equation was used:
ECOR - *EM-(ED.C.>
where
E = corrected output
COR
E = amplifier output reading(was negative
M
for low photocell illumination) , volt
E = "dark current" amplifier output:
D'C" -0.00075 volt "low" preamplifier
-0.00650 volt "medium" preamplifier
59
-------
The resulting corrected photocell amplifier outputs are
tabulated for no interference filter in Table A-l;for
0.436 micron filter in Table A-2; and 0.656 micron filter
in Table A-3; and the data can readily be used for experi-
mental data evaluation or for future transmissometer
useage when the optical path length is 145 inches.
Field Determination of "Dark Current1
With the instruments optically aligned and attached to the
stack, the "dark current" amplifier output was determined
by closing the receiver inlet. Because this value depends
on the photocell temperature, this procedure was repeated
every hour during data taking process. This detected
"dark current" amplifier output was used to correct the
outputs determined for various experimental transmissometer
operational conditions. The procedure used in correcting
measured data for "dark current" conditions was identical
to that described in the transmissometer calibration section
In-Stack Transmittance-Wavelengtn Characteristics
Tests with four different light interference filters, namely
0.436, 0.486, 0.579 and 0.656 micron wavelength, inserted
in the light path before the experimental receiver photocell
were performed to define the measured transmittance as a
function 6f wavelength illumination or detection.
The results of this group of experiments are plotted in
Figures A-l, A-2, A-3 and A-4. In these figures the
relative transmittance defined in Equation (10) is plotted
against the receiver angle as a function of the various
transmitter angles. As previously defined, the larger the
relative transmittance number, the lower the measured experi-
mental transmittance or normalized values. It must be
noted that the true in-stack transmittance used in calcul-
ating the relative transmittance was measured with the 3°
receiver angle reference transmissometer. Based on the
results described in Section VIII, all transmittance values
in Figures A-l, A-2, A-3 and A-4 would have been about
5 percent lower if the 0.8 degree receiver angle reference
transmitter had been used. Such a correction would be
necessary if interference filters results and no interfer-
ence filter results are to be compared.
60
-------
Table A-l.
Results of Experimental Transmissometer
Calibration .
Transmitter
Angle
(degrees )
0
10
20
30
40
50
60
0
10
20
40
60
0
10
20
40
60
Receiver
Angle
(degrees )
2.0
11
tl
II >
n
"
M
4.0
M
It
M
It
6.5
II
II
"
II
Interference
Filter
(Vim)
None
M
n
it
ii
n
n
None
n
n
n
ti
None
ii
. n
n
n
*
Photocell
Amplifier
Output
(Corrected)
(v)
2.82075
0. 26375
0.09725
0. 06055
0 .04825
0. 04025
0.03455
3.35075
0.32075
0.12475
0.06375
0.04475
3.38075
0.33575
0.13275
0.06625
0. 04675
Data are valid for "Low" pre-amplifier setting on the
experimental receiver cover. For "Medium" setting,
multiply values by 10. Setting "High" is not
recommended.
Data corrected for the following "dark current"
amplifier outputs:
"Low" pre-amplifier setting = -0.00075 volt
"Medium" pre-anplifier setting = -O.OOG50 volt
61
-------
Table A-2.
Results of Experimental Transmissometer
Calibration.
Transmitter
Angle
(de grees)
0
10
20
30
40
50
60
0
10
20
40
60
0
10
20
40
60
Receiver
Angle
(degrees)
2.0
tt
tt
it
n
tt
it
4.0
ft
It
»
ft
6.5
It
It
n
n
Interference
Filter
(prn)
0.436
it
H
ft
11
n
it
0.436
It
It
ft
11
0.436
it
n
n
M
it
Photocell
Amplifier
Output
(Corrected)
(v)
0.08975
0.00240
0.00075
0.00043
0.00032
0.00026
0 .00022
0.11275
0.00268
0.00088
0.00045
0.00033
0.11275
0.00268
0.00088
0.00045
0.00033
Data are valid for "Low" pre-amplifier setting on the
experimental receiver cover. For "Medium" setting,
multiply values by 10. Setting "High" is not
recommended.
Data corrected for the following "dark current"
amplifier outputs:
"Low" pre-amplifier setting = -0.00075 volt
"Medium" pre-amplifier setting = -0.00650 volt
62
-------
Table A-3. Results of Experimental Transmissometer
Calibration .
Transmitter
A a g 1 e
(degrees )
0
10
20
30
40
50
60
0
10
» r\
40
60
0
10
20
40
60
Receiver
Angle
(degrees )
2.0
tl
11
II
II
II
11
4.0
II
11
II
II
6.5
ii
M
"
II
Interference
Filter
(Vira)
0.656
II
II
M
II
M
It
0.656
It
II
0 .656
I
n
I
n
*
Photocell
Amplifier
Output
(Corrected)
(v)
0.24375
0.00820
0 .00303
0.00180
0. 00140
0.00114
0.00097
0. 32075
0. 01095
0. 00200
0. 00138
0.33575
0. 01145
0.00440
0.00200
0. 00138
Data are valid for "Low" pre-amplifier setting on the
experimental receiver cover. For "Mediun" setting,
multiply values by 10. Setting "High" is not
r ecominended .
Data corrected for the following "dark current"
amplifier outputs:
"Low" pre-amplifier setting = -0.00075 volt
"Medium" pre-anpli f i.er setting = -0.00650 volt
63
-------
1.5
1.4
60
Receiver angle,
Figure A-l.
Relationship of the Relative Transmittance and
the Receiver and Transmitter Angle for 0.436nm
Interference Filter.
-------
CTv
Ul
1.5
l-'
1.3 __
w
H
V 1.2
R
tcance
H
4J
(fl
.-I
<1)
C4
I-
1.0 -
E
W)
2 o-'
H
O
CD
0.7 -
O
CTi
0.5
10
20 30 40
Receiver Am/le, °
60
Figure A-2. Relationship of the Relative Transmittance and
the Receiver and Transmitter Angle for 0.486ym
Interference Filter.
-------
w
EH
(U
u
4J
4J
H
'E
W
q
m
ti
E-i
0)
1.5 f
1.4
1.3
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
Hti
Tir an em it tor 'AngLe v:
:p:
28!
:J£r. armm ^ 11 an c_e
fin
10
20 30
Receiver Angle,
40
50
60
Figure A-3. Relationship of the Relative Transmittance and
the Receiver and Transmitter Angle for 0.579|im
Interference Filter.
-------
w
E-i
v,
P
E-"
u
c
a
H
4J
n)
1.5
1.4
1.3 .
1.2 _:
1.1 L-
1.0
0.9
0. 8
0.7
0.6 -
0.5
9.'.rM?J*S^
:-i:0:.,65
"~.'~~}~~'
Trananittqr 'Andle::
FLlt^r':- ^ :-LO>p5G
. . . . . "I . . 7
10
20 30 40
Receiver Angle, °
50
60
Figure A-4, Relationship of the Relative Transmittance and
the Receiver and Transmitter Angle for 0.656jJm
Interference Filter.
-------
The results can be summarized as follows:
1. The plots show the dependence of the in-stack trans-
mittance on wavelength as already described in
Section VIII.
2. The closest agreement between in-stack and plume
transmittance measurements are achieved when either
white light or interference filter between 0.486 and
0.579 micron are used.
3. The trend for the difference between the measured
experimental transmittance and reference or plume
transmittance to become greater for generally de-
creasing stack gas transmittance was also seen in this
set of data. The difference becomes more evident for
all cases when the data of Figure 13 of the main
report (no filter) is replotted in the same dimension-
less transmittance format as Figures A-l through A-4
Cfilter data). The data of Figure 13 of the main
report is presented in terms of the relative transmit-
tance in Figure A-5. The results shown in Figures
A-l through A-4 were obtained when the plume transmit-
tance was generally around 45 percent, whereas the
data for Figure A-5 (no filter) was obtained at 60 to
80 percent transmittance.
Tabulated Experimental Data
The representative data collected throughout the experiments
are presented for possible further evaluation in the follow-
ing tables. Tables A-4 through A-7 contain the data
obtained when no interference filters were used. Tables
A-8 and A-9 were generated when a 0.436 micron filter was
used. Data obtained while using 0.486, 0.579 and 0.656
are respectively presented in Tables A-10, A-ll, A-12,
A-13, A-14 and A-15.
68
-------
0)
O
c
0
in
O
rt
4J
H
4J
1.0
0.9
0. 8
0. 7
0.6
10
Figure A-5
20 30 '0
Receiver Angle, °
50
60
Range of In-Stack
Reference Tr ansmittance :
0.607 - 0.880
Reference Transmissometer
Angle of View :
0. 8°
Interference Light
Filters Used:
None
Run : 1, 6, 7, 8,
of 4/19/72
9,12
The Receiver and Transmitter Angle Influence on the Relative
In-Stack Transirjittance Expressed as a Ratio of the
Reference In-Stack Transmittance and of the Transmittance
Measured by the Experimental Transmissometer for Given
Receiver and Transmitter Angles.
-------
Table A-4 . In-Stack Transmissometer Field Results.
c
3
K
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
c
tanco
Re f erenc
Transmit
2
0.612
0.607
0.744
0.723
0. 744
0.744
0. 752
0.764
0. 744
0. 764
0. 793
0.785
0. 773
0.855
0. 626
0.806
r*
1 U
41 O
C U
L'xpe rime
Tran smi t
Angle, c
3
10
10
20
20
40
60
60
10
10
10
20
20
2C
40
60
60
fH
4
4J
C
Exper ime
Rcccivei
Anglo, c
4
2
2
2
2
2
2
2
4
4
4
4
4
4 !
4
4
4
a
u
c E
U Zl
Interfei
Filter,
5
N *
N
N
N
N
N
N
::
;;
ti
N'
N1
N
H
N
N
«
.w
c
Experinu
Receive]
Range
6
L *
L
L-
L
L
L
£,
. _ Z-
L
L
T^
L
L
L
L
1,
sd Zero Con-
No Aerosol
*n t al Trans .
V
4J - E -
0 C -H iJ
e> o ^ a
^ * V fit
tt *J a, -w
o -H x y
u a u o
7
0.26475
0.26475
0.09325
0.09325
0.04925
0.03555
0.03555
i,. ^217.
0.32175
0.32175
0.12575
0.12575
0.12575
0.06475
0.04575
0.04575
intal
.somoter
V
Exper inu
Transmi ;
Output ,
8
0.169
0.170
O.Q78S
0.078
0.0395
0.0286
0.0293
J.264
0.259
0.269
0.113
0.113
0.111
0.0637
0 .0443
0.0436
id
?ntal
T some ter
V
ii E ~< -
0 ~i E U
4) W W ^
U O C 2,
i4 O<
-------
Table A-5. In-Stack Transmissometer Field Results.
c
a.
1
6
6
6
6
6
e'
6
6
a
8
8
a
8
8
0
u
c
c
C -rf
V*
V
£K E<
2
0. 859
0. 826
0. 847
0.859
0.859
0.806
0.859
0.347
0.830
0.839
0. 826
0. 818
0.835
0.847
H
tt M
p o
c *J
g -4
h (fl o
0 C -H
a ra ci«
W t* <
3
10
10
20
20
40
60
60
60
10
20
20
40
60
60
r+
fa
u
c
e V
V4 ~4 0*
O CJ .H
a o tp
H a <
4
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6. ^
8.5
8. 5
8.5
8.5
8.5
8.5
V
o
c 6
« 3
CJ -
h 0
0 -^
4J iH
K Et<
5
N *
K
N
N
;;
N
N
N
:i
I!
N
N
rH
«
4J
C
£ a
u * o
e c c-
c. o c
B K C
6
L *
L
L '
L
L
L
L
L
T
L
L
L
L
L
d Zero Con-
No Aerosol
ntdl Trans.
V
*J - E -
O o V4 D
W ~1 O C.
k -> C. -U
CJ Q U O
7
0. 33875
0. 33675
0.13375
0.13375
0. 06725
0. 04775
0.04775
: "47~ 5
0.34175
0.13375
0. 1337S
0. G672S
0.04775
0.04775
ntal
s o m c t o r
V
*? J
ti 0 3
oca.
C. n *J
a H o
8
0. 317
0.303
0.134
0.137
0.0706
0.0476
0. 0514
0.05- .
0.322
0.135
0.137
O.C6S7
0.0514
0.0531
d
n tnl
some tor
V
U E -i -
O W M O
M 0 C £.
K 0, H 4J
u u t> o
9
0. 31875
0. 30975
0. 13575
0.13375
0 . 07235
0.04935
0.05315
0.05315
0. 32375
0.13675
0.13875
0.07G45
0.05315
0.05485
ntal
t ancc
w in
o c
. 0, 0
U H
10
0.948
0 . 920
1. 015
1 .037
1 . 078
1.033
1.113
1.113
0.947
1 . 022
1. 037
1 . 047
1.113
1.149
n Ic s s
t a n c c
H U -*
C *J f.
owe
E 1 H
a M £
11
0.908
0.898
O.S35
0.823
0.799
0.780
0. 772
0.761
0.877
0. 820
3.797
0.781
0.750
0.737
cd
ntal
. t a n c e
-< £ -<
n v< u:
E i c
U 0. «
Z M H
12
0.889
O.S39
0.966
0 .975
1.010
1.035
1.045
1.060
0.920
0. 984
1.012
1.033
1. 076
1.095
No filter; L - Low preaaplifier setting.
71
-------
Table A-6. In-Stack Transmissometer Field Results.
c
3
a
1
9
9
9
9
9
9
9
9
9
c
<
CJ ^J
c -H
14 n
a c
u ic
.
X
*
_i
10
44
C
0 U
6 ai
Expcr
Rccc i
Range
6
L*
L
L'
L
L
L
L
L
r
1 *
C «H 0)
O O C
on*
O K
out-.
I* O
U < rt
0 4J
a z c >
o a
E -I -
-4 6 U
oca
& « *>
X w 3
a E- o
8
0. 315
0.316
0.327
0.139
0.0769
0.0721
0 . 07 5 1
,.0534
0.0527
ti
4J
H O
« e
4J O
O C « >
o o e -H -
U --4 g 4J
a ^ « 3
^4 a c o.
>4 a « u
0 X U 3
CJ M H O
9
0. 31675
0.31775
0.32875
0.14075
0.07865
0.07385
0. 07685
0.05515
0.05445
0
r* O
ti C
4J
-------
Table A-7. In-stack Transmissometer Field Results.
«
1
1 2
12
12
12
. 12
12
12
12
12
12
12
o a
a £
o c
a H
2
0. 806
0. 797
0.335
0. 806
0.863
0.818
0 .347
0.368
0 .880
0.797
0 . 314
i-4
« U
AJ 1)
e a
£ ~<
-H E -
>w w O
o e -.
X V4 C
K r~ <
3
10
10
10
20
20
20
40
40
40
60
60
_<
a
AJ
c
E o
-« >
Vj -^ O
c c ^
X O C
a c <
4
60
60
60
60
60
60
60
60
60
60
60
o
u
C £
a 3.
o
:
;;
N
;;
;;
N
N
K
i
^H
*f
4J
C
E 0
x >
U -H a
l i r-
X O ^
w c «
6
^ *
L
L
L
L
L
L
^
L
I
C -< K
O O C
U n r;
0 H
0 U H
V4 C
0 < -1
N 0
0 u
= z = >
** » e »
U C -H -P
C 0 V4 3
^ -H C Cj
0-4X3
U O tJ O
7
0.33675
0.33675
0. 33675
0.13075
0.13075
0. 13075
0.06725
0 .06725
0. 06725
0.04795
0.04795
w
o
AJ
H O
n E
AJ O
C 0 >
E -i -
-< E -U
V CT 3
ace.
X V4 S
K E- t<
3
0.333
0.322
0.332
3.138
0.155
0.133
0.0757
O.OS16
0 . C S 1 4
0.0542
0 . 0548
ki
o
AJ
^ C
0 E
JJ 0
3 C n >
AJ £ ^1 -
U -I E AJ
'j u n 3
^ c c £.
0 X ». 3
'J t: ' O
9
0. 33975
0. 32375
0.33375
0.13975
0. 15675
0.14075
0.0"'745
0.08335
0.08315
0.05595
0.05655
o
^ u
-------
Table
In-Stack Transmissoirjeter Fieia Results.
*»
c
3
p£
1
2a
2b
2c
2d
2s
14a
14b
14e
14d
14u
15a
ISb
15c
15d
ISe
ence
ni ttance
v c
WJ <0
u fc
a. H
2
0. 562
0. 562
0. 562
0.562
0. 562
0. 5G2
0.547
0. 531
0. 541
0. 531
0. 531
0.531
0'. 531
0.531
0. 531
Iment al
mltter
o
Q. « cr>
x M c
U £-1 <
3
0
10
20
40
60
0
10
20
40
60
0
10
20
40
60
-iraental
uver
'-, °
04 U 0*
X 01 C
H K <
4
3
3
3
3
3
5
5
5
' 5
5
10
10
10
10
10
-f erence
ar, \ia (
4J rt
C -H
M fcl
5
A*
A
A
A
A
A
A
t-\
i
A
A
A
A
A
A
-imental
Lver
i
C, 0 C
K O a
U K K
6
H*
M
M
M
M
M
H
H
M
H
M
M
M
M
>cted Zero
L tion , No
sol Experimental
3 . Output , v
u rj o c
h C rt «
o o o n
O O «S E"
7
0.4918
0.0410
0.242
0. 0202
0. 0193
1. 2689
0.0509
0 . 0 2 '/ x
0. 0216
0.0202
1 . 3049
0. 0509
0.0274
0.0218
0.0203
-iraental
sraissoraeter
It, V
C' C C,
O, 1C *J
x M a
U H O
3
0.223
0.00276
-0.00656
-0. 00823
-0. 00383
O.C96
0.0107
-0. 00387
-0.00529
-0. 00762
0.721
0.0106
-0. 00276
-0.00605
-0.00689
;cted
riraental
;missometer
Jt, V
V4 CJ C Q<
U ft rO +J
O X W 3
U K H O
9
0.247
0. 02676
0. 01744
0. 01577
0.01517
0.720
0.0347
0. 02013
C . 01771
0.01638
0. 745
0.0346
0.02124
O.C1795
0.01714
rinental
sntittancc
u c
0« 11
X Li
W E-i
10
0. 502
0. 639
0. 721
0.781
0. 786
0. 567
0.682
0.743
0.820
0. 811
0. 571
0. 680
0.775
0.823
0.844
is ionic ss
tack
smittance
Q
-t e h
a M H
11
1.12
0.88
0. 78
0.74
0.71
0.99
0 .80
0. 715
0.66
0.65
0.93
0.78
0.685
0.64
0.63
ilized
riraental
saittance
E « c
U O, nj
0 X U
2 W [-
12
set ed
Tsionless
smittance
k V C
>< E *
O ~* w
o Q t.
13
1 .064
0.836
0. 741
0. 703
0. 674
0. 940
0 . 7SO
0.681
0.627
0.617
0.833
0.741
0.650
0.608
0. 598
A ~ 0.436 po interference filter; M - Medium preaaplifier setting.
-------
Table A-^9 . In-^Stack Tra,nsmis,someter Field Results
m
c
3
C£
1
IGa
ICb
IGc
16d
IGe
17a
17b
17c
17d
17e
ce
tt ance
Re f ere n
Transni
2
0.531
0. 502
0 . £31
0.531
C. 541
0.541
0.525
C. 531
0.553
0. 541
ental
tter
o
Experim
Transni
Angl e ,
3
0
10'
20
40
60
0
10
20
40
60
rH
«J
+J
c
HI U 0
Experir
Receive
Angle ,
4
20
20
20
20
10
40
40
40
40
40
rence
Urn
Intcrf e
Filter,
5
A *
A
A
A
A
7,
A
A
A
A
ental
r
Expe r ir
Receive
Range
6
M*
M
K
M
K
M
;,
.~1
M
M
ed Zero
on , No
Experimental
Output , v
U 4J O
a *< w m
^ TJ O 1^
u c u at
0 0 o ^
O O < H
.7
1 . 304 9
0.0509
0.0274
0 . 0 2 1 S
0.0203
1 . 3040
C . 0 5 »
0.02:<
0.0213
0.0203
ental
ssomoter
V
H E 4J
^ tn 3
o c a*
c. a -J
X l< P
WHO
8
0. 740
0.0127
-0. 00260
-0. 00633
-0. 00657
0. 75C
^ . 31 I
-0. 0022S
-0. 00531
-0.00629
od
icn t al
sso meter
V
U -H E 4J
a ^ w 3
n o s a,
M n< n 4J
O X M D
O W H O
9
0.772
0. 0367
0 . 0214
0.01767
0. 0174 3
C.780
0. 036C
0.02172
0. 01869
0.01771
i e n '^ a 1
ttanco
-H E
14 ffl
CJ C
o< n
X M
W H
10
0.591
0.721
0. 781
0.810
0.359
0 . 558
0.719
0.793
0. 857
0.372
on 1 o f s
-k
L 1 1 a n c e
D i m e r. s 3
In-Stac
Trail SITU
11
0. 90
0. 78
0. 68
0.65
0.63
0.90
0.73
0.67
0 . o 4
0.62
o
~l U
1 C
a 4J «
0 C JJ
S C) 4J
Norma 1
Experir
Transm
12
.ed
Lonless
L 1 1 nnce
o tr E
QJ c ta
U V C
^ t -9
0 -H V<
U Q H
13
0.855
0.741
0.646
0. G17
0. 598
0.355
0.693
0 . 6 3 C
0 . 608
0. 539
0.436 pm interference filter; K » Medium preaaplifier setting
-------
Table A-10, In-Stack Transmissometer Field Results.
er\
r
c
3
K
1
3a
3b
3c
3d
3e
' n-.
19h
I9c
19d
19e
20a
20b
20c
20d
20e
m
0 U
U 4J
c -x
0 E
M 01
a c
u v<
2
0.537
0.525
0. 541
0. 547
0.537
0.528
0.531
0. 547
0. 541
0.531
0. 537
0.531
0.525
0.531
|o.531
erinent
nsmitte
Ic, °
K \* C.
W t- <
3
0
10
20
40
60
0
10
20
40
60
0
10
20
40
60
id
erinent
eiver
le, °
x
o
erf eren
ter, (Jra
e -i
5
B *
B
B
B
B
B
B
-
a
B
B
B
B
B
B
«
icriment
eiver
igc
X o to
a ft K
e
M*
M
M
H
M
M
M
M
M
M
K
U
M
rl
«
*J
c
«J *>
0 -H *
U 0 W 4J
u a o 3
M D. C.
- X 4J
a c w a
0 O O
*J -H .-<
o -u o
o -H in 01
w -a o e
o o o v*
O U < H
7
2.8863
0. 0798
0.0356
0.0271
0. 0240
3. 3709
0.1C2f3
o . o ; 3 c
0. 0311
0,0269
3 .4692
0.1033
0.0436
0.031S
0.0273
n
to
4J
.-< o
a e
J 0
C ol >
U tt
E -I -
H e *>
W 07 3
CJ C 0,
X ti 3
WHO
8
1. 27
0.0231
0.0010
-0. 00312
-0.00532
1.65
0. 00822
0.00074
-:. 00268
1 .S3
0. 043
0.00823
0.0012
-0.0022
\4
o
*j
r-t V
a e
U 0
o e B >
0 U ul
4-1 E * -
0 * E V
&> k w 3
w o c a<
0 X K 3
O W H O
9
1,29
0. 0471
0.0250
0. 02088
0, 01868
1.67
0.0643
0.03222
0.02479
0.02132
1.85
0. 067
0.03223
0. 0252
0.0218
o
^ u
n a
U K
t: a
v u
E -H
^. E
V in
u c
x n
U H
10
0.448
0, 590
0. 702
0.770
0.778
0.497
0. 628
0.719
0.797
0. 792
0.511
0. 643
0.739
0. 792
0.79B
n «
H U
c c
K *>
O M "
'H U -H
B fl E
£ ** W
a) in c
H C V<
D u E-
11
1. 12
0.89
0.77
0.71
0.69
1 .06
0.84
0.73
0.68
0. 67
1.01
0. 32
0.71
0.67
0.66
01
~< u
*o c
rmalized
perincnt
ansmitta
0 x u
2 W H
12
0} 0)
0} U
HI C
O C 4J
0) 0 4J
AJ -r< -H
U to 6
0) C K
k « C
^ E rt
O -1 H
O Q H
13
1.064
0.845
0.731
0.674
0. 655
1 .007
0. 793
0.693
0.646
0. 636
0.959
0.779
0.674
0.636
0.627
B - 0.486 )im interference filterj M - Helium preamplifier setting.
-------
Table A-ll. In-Stack Transmissometer Field Results.
3
PC
1
21a
21b
21c
21d
21e
22a
.72b
22c
22d
22Q
u
c
0) U
Pi E"*
2
0.547
0.531
0. 547
0.531
0.531
0.469
0.469
0.469
0 . 469
0 .469
-HE*
01 C ^H
X VJ C
W H <
3
0
10
20
40
60
0
10
20
40
60
.-4
10
C
01 H O
E *
01 01 ^t
cu u o*
X flJ C
w B; «
4
20
20
20
20
20
40
0
40
40
40
0>
o
C E
01 XL
01 ~
^ OJ
01 -U
C -1
M Ix
5
D«
B
B
I)
B
B
3
B
B
B
(H
a
c
01 k4
E 01
** >
M TH 0)
01 O CF>
n. u c
X 4) rtj
U fti QJ
6
M*
M
M
M
M
M
M
.1
M
H
ti
iJ
B
01 >
E
O -H
K 0 H a
01 2 01 3
N o, a,
» X -U
T3 C W 3
01 O O
0 4J O
u -a o c
U C U 0
O O 01 fc
U U < H
7
3.4692
0.1033
0 . 0436
0.0318
0.0273
3 . 4692
1.1033
0.04.6
0 . 0313
0.0273
H
O
r-l
Cl 01
E *
H E iJ
U 01 3
O C &
Q* rt *J
X ^ 3
WHO
8
1.93
0.0454
0 . 0100
0.0015S
-0 .00099
1 . 69
"l . J 3 L
0.00563
-0 .00105
-0.0040
H
a
4J
rH U
al E
U 0
a c 01 >
01 0) u
0 -H E "
01 M 0] 3
H 0 C 0.
M Q. id *J
O X t4 3
U W £- O
9
1. 95
0 . 0694
0 .0340
0.02558
0. 02301
1 . 71
0.062
0. 02963
0.02295
0. 0200
0
^ u
a C
+J rt
C 4J
a; ^>
C -H
H E
M (A
Cl C
X W
W H
10
0. 563
0.672
0.780
O.S04
0.843
0.494
0.600
0.679
C.722
0.733
01 0)
01 U
01 C
r~\ 1)
C *J
O ^ *1
H U -H
ol rt 2
C -M W
a w c
a H H
11
0.97
0.79
0.70
0.66
0.63
0. 95
0 . 78
0.09
C .65
0 . 64
a
H O
"3 C
TJ -P fl
Cl C 4J
H C "1
-1 ~< E
12
0) o
01 U
01 C
tj C *J
o 01 e
a c a
n 6 fl
O -1 W
U O H
13
0.921
0.750
0.665
0.627
0 . 598
0. 902
0.741
0 . £ 5 5
0.617
0. 608
0.486 wm interference filter; M - Mediun preamplifier setting.
-------
Table A-12. In-St^ck Transmissometer Field Results.
tt
1
4a
4b
4c
4d
4e
25a
25b
25c
25d
2So
ot E
o; H
2
0. 566
0. 566
0.573
0.573
0. 576
0.533
0. 533
0.533
0.533
0.533
" 1 ""
2Sa I 0. 533
26b
26c
26d
26e
0.533
0.533
0.533
0.533
t
*J O
G P
** s *
W H <
3
0
10
20
40
60
0
10
20
40
60
0
10
20
40
60
,_(
at
4J
C
OHO
C c;
rt > -
U -H V
CJ 3
IM M
k 0»
c p
JJ . (
C "^
M &.
5
C *
c
c
c
c
c
c
c
c
c
c
c
c
c
c
r4
n
I*
c
U 14
E 01
-H >
v< -H a
a) a» tr
Q, L> C
X 0) 0
w o: n:
6
H *
M
M
M
M
M
M
>.
M
K
M
H
M
H
X]
C
ffl >
E
0 -< »
K O i, P
0) 2 01 P
M C. Q.
- X p
O C W 3
at o O
p -H rH
U AJ O
u -H n m
14 t) O C
h s u a
0 O 01 ^
U U «t H
7
0. 3292
0.1285
0,0530
0.0345
0.0294
6.0439
0.1634
^.0^.--
0.0414
0 . 0344
6 . 220S
0. 1701
0 . 0683
0 . 0426
0.0350
i.
Q
p
f< (U
<0 E
p O
C 01 >
C) tfi
F. -I -
-HEP
^ 01 3
O C C*
3. « P
X U 3
WHO
8
0.121
Q .0572
0. 0145
0 .00265
-0.00052
2 .75
0. 033
0.0244
0.0076
O.OC35
2.99
0 .090
0.0273
0.0090
0.00422
u
^j
(H 01
a e
P 0
a c « >
o a u
P E -i -
01 U bl 3
k o c a.
M Cl< rt AJ
0 X W 3
U U H O
9
0.145
0.0812
0. 0385
0 . 02666
0.02343
2 . 77
0.107
0 . 0484
0.031S
0.0275
3.014
0.114
0. 0513
0.0330
0.02322
0)
r-t U
13 C
4j a
C .p
Ot 4J
e -i
H E
u n
41 G
C, TO
X k(
U t.
10
0.440
0.632
0. 726
0.773
0. 799
0.459
0.635
0.720
0.763
0.799
0. 434
0.670
0. 751
0.775
0.806
en o
in u
o c
H ffl
C u
O J^ AJ
-H U -H
0 « E
C -P W
O 10 C
S 1 *
r4 C \,
Q HI fr"
11
1. 28
0. 39
0.79
L_2-'7
-------
Table A-13 . In-Stack Transmissometer Field Results.
ID
C
3.
i
27a
27b
21:
27d
27o
2 C a
28h
2Bc
23d
23o
ce
ttance
aj K
^ (ft
<*4 (13
0) Vf
B; £-
2
0. 540
0.533
0.533
0 . E 3 3
0.533
0.533
0.533
0.533
0.533
0.533
ental
tter
o
H C -
ki w CJ
U fl C>
X U C
W £- <
3
0
10
:o
'10
60
0
) 0
so
10
60
r-t
a
v>
c
V W O
-I > -
W -H 01
o< u tp
X 11 C
u « <
4
20
20
20
20
20
40
,u
40
40
40
rence
ym
^ ti
w o
*j r-t
c -*
M r^
5
C *
c
c
c
c
c
c
.c
c
c
ental
r
H >
M -rt 0)
a u c
X 41 fl
W « K
6
K*
]:
K
M
(1
M
M
!t
M
M
ed Zero
on , No
Experimental
Ou t |
-------
Table A-^14 . In-Stack Tra.nsmissometer Field Results.
CD
O
K
1
5a
5b
5c
5d
So
30a
3 Ob
30c
30d
30e
31a
31b
31c
31d
31e
o
u
c
Cl 4J
U 4J
01 E
M 0)
oi c:
a. t>
2
0.531
0. 531
0.534
0.531
0.531
0.5-00
0.516
0.506
0. 516
0. 516
0.475
0.475
0.494
0.484
0. 500
PH
a to
C V
V JJ 0
E -i
* E -
M 01 O
Cl C ^
O. flj D1
X k C
3
0
10
20
40
60
0
10
20
40
60
0
1C
20
40
60
<0
^j
c
0) U 0
E 01
H >
W -H 41
Ot 01 r-t
D. 0 0<
x a) c
U OS «C
4
3
3
3
3
3
5
5
b
5
S
10
10
10
10
10
d
u
c S
«1 3
^4
O "
*W V4
>j 01
U -rt O
Ct fij t7>
a u c
X OJ 13
A a. a.
6
M*
M
M
M
M
H
M
.-
;i
H
M
M
M
M
H
r-4
C
0) >
E
O "H
V4 O U 4J
01 SI <» 2
M a o.
- X AJ
a c u a
01 O O
4J -r^ 1-4
U 4J O
« ^i a a)
u -a o c
H C tl «
O 0 11 M
U U < H
7
4. 7159
0.1151
0. 0500
0.0339
0.0296
5.5069
0. 1499
u.Of,^
0.0407
0 . 0346
5.6679
0.1519
0.0641
0.0418
0.0353
^i
g 01
E -H -
M » 3
v c a
D. n) -u
X V< 3
WHO
8
1.54
0.035
0 . 00797
-0.00135
-0. 0029
1.88
0.0617
0.0159
0.00469
0. 00132
2.03
0. 0614
0.0194
0.00511
0.00215
K
0
4J
~f 0)
a E
U O
a c 01 >
01 a a
*J E ^(
u -rt E -u
QJ ^ W D
H « C 0.
W d <0 4J
O X V4 D
U M H 0
9
1 . 564
0.059
0.03197
0.02265
0.02110
1.904
0.0357
0.04090
0. 02369
0.02532
2.054
0.0854
0.0434
0.02911
0.02615
o
rH U
a c
*j in
C *J
0 *>
c ^
H e
01 C
O< rt
X U
M H
10
0. 332
0.512
0 .639
0.668
0.713
0. 346
0 . 572
0.648
0.705
0.732
0.362
0. 562
0.677
0.696
0.741
01 a
01 U
a) c
fH id
C 4J
O A! 4J
H O *
U « 6
C -u m
01 Ul C
E i a
rf C U
Q n H
11
1.6
1.03
0.835
0. 79
0 . 74
1.45
0 .90
0.78
0.73
0. 70
1 . 31
0.34
0.73
0.69
0.67
u
r* 0
n c
t3 *^ <3
O C *>
H a *j
M g M
-i -H E
rt U ul-
E 0 C
M a, v
0 X U
2 W H
12
n o
01 o
a c
-< <
O C *J
iJ M M
0 r, E
o c to
LJ O C
^ E fl
O -H K
o a t.
13
1 . 52
0.978
0. 793
0.75
0.703
1. 38
0.855
0.741
0.693
0. 665
1. 244
0.796
0.693
0.655
0.636
E 0.656 Hm interference filter; H « Medium preamplifier setting.
-------
Table A-15. In-Stack Tran^missometer Field Results.
CD
e
3.
i
32a
32b
32c
32d
32e>
33a
33b
33c
33d
33e
c
U 4J
01 C
<4-l (0
K I-
2
0. 500
0. 500
0. 500
0. 500
0.500
0. 462
0 . 500
0 . 500
0.516
0.484
«J 14
01 JJ 0
E -4
0. rt CT
U H <
3
0
10
20
40
60
0
10
20
40
60
r4
a
4J
C
01 M o
-4 > -
^ -H 01
01 01 -H
G, u en
X 01 C
(U Pi <
4
20
20
20
20
20
40
40
40
40
40
o
u
c e
01 ^
X4 01
01 4-»
4-1 «-4
C -^
M &-
5
E*
E
E
E
E
E
*
E
E
E
r4
(0
4J
C
H >
14 -H 01
C! 0) D1
Oi U C
X 01 d
u a: K
6
M*
H
M
K
M
M
rf
M
M
a
a
01 >
E
14 0 V4 4J
0 2 01 3
N a. a
a c u 3
01 0 O
O *J O
ai T4 w n
V4 TJ O C
K c ^ e
o o 01 M
u a < H
7
5 .6679
0.1519
0 . 0641
0. 0418
0.0353
5 .6679
0 .l^i-J
0 . 0641
0. 0418
0. 0353
'M
a
4J
^ O
a E
4J O
G (n >
o> tn
-1 E 4J
o c c.
C« fl -U
X t4 3
W f- O
8
2.19
0. 0685
0. 0205
0 . 006S6
O.C0226
2.100
0.0704
0.0211
0. 00848
0. 00189
u
01
4J
r-4 Cl
« E
U O
a c «] >
o -H E a
M 0) C p4
O X V4 3
U K f O
9
2.214
0. 0525
0. 0445
0. 03096
0.02626
2.124
0 . 0 & 4 4
0.0451
0. 03243
0.02589
01
M U
« c
C 4J
O 4J
H E
tl 0)
0) C
X h
W t-
10
0. 390
0.610
0 . 694
0.741
0. 744
0.375
0.621
0.703
C. 777
0.733
rf -i E
rO tv W
E 01 C
O X 14
Z W H
12
01 O
tn U
01 C
TT C *J
01 O 4-1
u n E
V4 01 C
0 -H V4
u a E*
13
1 . 216
0. 779
0.684
0 .636
0.536
1 . 1 6 H
0 . 7 i. 0
0.674
0.627
0.617
E - O.G56 Jim interference filter;
Mejiua preaaplificr setting.
-------
APPENDIX B
NOTES TO THE REFERENCE AND EXPERIMENTAL
TRANSMISSOMETER OPERATION
Electrical Hookup and Alignment of the Instruments
Both transmissometers operate on 115 volts A.C., 60 Hz,
connected to the transmitter housing by a twist-type
electrical plug. While operating, the transmitter and
receiver must be interconnected by a shielded cable (about
15 feet long) supplied with the instruments. The amplifier
output signal is available from a ampheral plug on the side
of the transmitter housing and can be transmitted through
about 50 feet of shielded cable to any high input impedance
voltmeter. The range of the readout for the reference
transmissometer should be from 0 to about 0.5 volt depend-
ing on the optical distance of the transmitter and receiver.
The range for the experimental transmissometer should be
from -0.001 to +7.0 volt.
After about 30 minutes warmup period, the power supply
voltage for the transmitter lamp must be adjusted to 8.0
volts on the experimental transmitter, and 6.5 volts on the
reference transmitter. The voltage should be measured on
the lamp socket of the reference transmitter that is
accessible after removing the cover. After attaching
measurement probes, the voltage must be checked with the
cover closed so as not to influence the lamp control photo-
electric circuit. Two lamp voltage terminals are provided
on an outside panel of the experimental transmitter housing
and the cover of the experimental transmitter does not have
to be opened during the lamp voltage check.
The lamp voltage can be adjusted in each instrument by
adjusting a trimmer that is labeled LAMP and which is
accessible after removing the transmitter housing cover.
No other electrical alignment except the lamp voltage is
required.
Suggested Optical Alignment
To align the instrument on the stack, the procedure des-
cribed in Section III of this report should be followed.
-------
Maintenance
No special maintenance of the reference transmissometer is
required. When the instrument is operated on a negative
pressure duct or stack, the lens protection system is
sufficient to keep them clean for several hundreds of hours
The only maintenance may be the lamp replacement which can
be easily done after opening the reference transmitter
cover.
The experimental receiver lens requires checking and clean-
ing about every 80 hours of operation and the transmitter
lens must be checked and, if necessary, cleaned once an
hour.
When the instruments are to be used on a positive pressure
duct or stack, a high pressure and possibly greater volume
of clean flushing air to the optical system must be
provided.
The lamp in the experimental transmitter is replaced by
performing the following operations:
1) Remove the instrument housing cover.
2) Disconnect the cable lead-ing from the lamp
assembly to the electronic component.
3) Slide the lamp assembly out of the support
tube housing.
4) Remove the flushing/cooling air base fitting
from the lamp assembly.
5) Remove the lamp voltage control photocell
located inside the lamp assembly tube.
6) Loosen the three set-screws at the base of
the lamp socket.
7) Turn and lift out lamp.
The new lamp may be installed and the instrument re-
assembled by following the described lamp removal process
in reverse.
84
-------
Electrical Diagrams
The electrical diagrams of the experimental and reference
transmissometers shown in Figures B-l and B-2 are enclosed
for ease in trouble-shooting in the event that electrical
malfunction occurs.
-------
/A/40O-3 (41
/zov
CO
CTv
EHVI*QNM£HTAL KESEAfTCH COKF>
SCHEMATIC
£f?C-O/
Figure B-l . Reference Transmissometer Wiring Diagram.
-------
(*m*0/J
BO9*T> I
1
fMfiOHHVn-4,. »eSfA»CH CMP
""*
'" '£-^'-TL
2>j rx- /".
'""
'
'^fS^S4L»^Syfm^' "SB 100
Figure B-2 . Experimental Transmissometer Wiring Diagram
------- |