EPA-600/2-76-089b
May 1976
Environmental Protection Technology Series
INDUSTRIAL FUGITIVE
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield. Virginia 22161.
-------
EPA-600/2-76-089b
May 1976
TECHNICAL MANUAL
FOR THE MEASUREMENT OF FUGITIVE EMISSIONS:
ROOF MONITOR SAMPLING METHOD
FOR INDUSTRIAL FUGITIVE EMISSIONS
by
R.E. Kenson and P. T. Bartlett
TRC--The Research Corporation of New England
125 Silas Deane Highway
Weathersfield, Connecticut 06109
Contract No. 68-02-2110
ROAP No. 21AUY-095
Program Element No. 1AB015
EPA Project Officer: Robert M. Statnick
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
TABLE OF CONTENTS
rv
SECTION
1.0
2.0
2.1
2.2
2.3
3.0
2.1.1
2.1.2
2.1.3
2.2.1
2.2.2
2.3.1
2.3.2
3.1
3.
3.
3.
3.
3.
3.
3.
2
3.
3.
3
4
3.
3.
3.
3.
5
3.
3.
3.
3.
6
3.
3.
3.
1.1
1.2
2.1
2.2
4.1
4.2
4.3
4.4
5.1
5.2
5.3
5.4
6.1
6.2
6.3
3.7
4.0
4.1
4.2
4.3
4.4
APPENDIX
OBJECTIVE
INTRODUCTION
Categories of Fugitive Emissions
Quasi-stack Sampling Method . .
Upwind-Downwind Sampling Method
Roof Monitor Sampling Method
Selection of Sampling Method . .
Selection Criteria
Criteria Application
Sampling Strategies
Survey Measurement Systems . .
Detailed Measurement Systems
TEST STRATEGIES .............
Pretest Survey ............
Information to be Obtained .....
Report Organization .........
Test Plan ...............
Purpose of a Test Plan .......
Test Plan Organization .......
Roof Monitor Sampling Strategies . . .
Survey Roof Monitor Sampling Strategy .
Sampling Equipment .........
Sampling Systems Design .......
Sampling Techniques .........
Data Reduction ...........
Detailed Roof Monitor Sampling Strategy
Sampling Equipment .........
Sampling System Design .......
Sampling Techniques .........
Data Reduction/Data Analysis ....
Tracer Tests .............
Tracers and Samplers
PAGE
1
2
2
2
3
3
4
4
6
9
9
10
11
11
11
12
12
12
14
16
16
17
18
21
27
27
29
30
31
32
32
33
Tracer Sampling System JDesign 34
Tracer Sampling and Data Analysis 34
Quality Assurance
ESTIMATED COSTS AND TIME REQUIREMENTS
Manpower
Other Direct Costs . .
Elapsed-Time Requirements
Cost Effectiveness . .
35
38
38
38
42
42
APPLICATION OF THE ROOF MONITORING SAMPLING METHOD
TO AN ELECTRICAL ARC FURNACE INSTALLATION
iii
-------
LIST OF TABLES
TABLE
2-1
3-1
3-2
3-3
3-4
4-1
4-2
4-3
Typical Industrial Fugitive Emissions Sources . .
Measured by the Roof Monitor Sampling Method
Pre-Test Survey Information to be Obtained for
Application of Fugitive Emission Sampling Methods
Matrix of Possible Combinations of Key Test
Parameters
Elements of Conceptual Systems for a Roof Monitor
Sampling Program as Applied to Specific Types of
Fugitive Emission Sources
Range of Applicability of Common Velocity Measure-
ment Devices for Roof Monitor Sampling
Conditions Assumed for Estimating Costs and Time
Requirements for Roof Monitor Fugitive Emissions
Sampling Programs
Estimated Manpower Requirements for Roof Monitor
Fugitive Emissions Sampling Programs
Estimated Costs Other Than Manpower for Roof . .
Monitor Fugitive Emissions Sampling Programs
PAGE
5
13
22
23
26
39
40
41
LIST OF FIGURES
FIGURE
3-1
3-2
4-1
4-2
Electric Arc Furnace Operation; Roof Monitor
Showing Sampling/Mounting Configuration
Roof Ventilator Sampling Configuration
Elapsed-Time Estimates for Roof Monitor Fugitive
Emissions Sampling Programs
Cost-Effectiveness of Roof Monitor Fugitive
Emissions Sampling Programs
PAGE
19
20
43
44
IV
-------
1.0 OBJECTIVE
The objective of this technical manual is to present a guide for
the utilization of the Roof Monitor Sampling Method in the measurement
of fugitive emissions. Criteria for the selection of the most applicable
measurement method and discussions of general information gathering and
planning activities are presented. Roof Monitor sampling strategies and
equipment are described and sampling system design, sampling techniques,
and data reduction are discussed.
Manpower requirements and time estimates for typical applications
of the method are presented for programs designed for overall and speci-
fic emissions measurements.
The application of the outlined procedures to the measurement of
fugitive emissions from an electric arc furnace steel making plant is
presented as an appendix.
-1-
-------
2.0 INTRODUCTION
Pollutants emitted into the ambient air from an industrial plant
or other site generally fall into one of two types. The first type is
released into the air through stacks or similar devices designed to
direct and control the flow of the emissions. These emissions may be
readily measured by universally-recognized standard stack sampling tech-
niques. The second type is released into the air without control of
flow or direction. These fugitive emissions usually cannot be measured
using existing standard techniques.
The development of reliable, generally applicable measurement pro-
cedures is a necessary prerequisite to the development of strategies for
the control of fugitive emissions. This document describes some pro-
cedures for the measurement of fugitive emissions using the roof monitor
measurement method described in Section 2.1.3 below.
2.1 Categories of Fugitive Emissions
Fugitive emissions emanate from such a wide variety of circumstances
that it is not particularly meaningful to attempt to categorize them
either in terms of the processes or mechanisms that generate them or the
geometry of the emission points. A more useful approach is to categorize
fugitive emissions in terms of the methods for their measurement. Three
basic methods exist—quasi-stack sampling, roof monitor sampling, and
upwind-downwind sampling. Each is described in general terms below.
2.1.1 Quasi-stack Sampling Method
In this method, the fugitive emissions are captured in a temporarily
installed hood or enclosure and vented to an exhaust duct or stack of
-2-
-------
regular cross-sectional area. Emissions are then measured in the ex-
haust duct using standard stack sampling or similar well recognized
methods. This approach is necessarily restricted to those sources of
emissions that are isolable and physically arranged so as to permit
the installation of a temporary hood or enclosure that will not inter-
fere with plant operations or alter the character of the process or
the emissions.
2.1.2 Upwind-Downwind Sampling Method
This method is utilized to measure the fugitive emissions from
sources typically covering large areas that cannot be temporarily hood-
ed and are not enclosed in a structure allowing the use of the roof
monitor method. Such sources include material handling and storage
operations, waste dumps, and industrial processes in which the emissions
are spread over large areas or are periodic in nature.
The upwind-downwind method quantifies the emissions from such
sources as the difference between the pollutant concentrations measured
in the ambient air approaching (upwind) and leaving (downwind) the
source site. It may also be utilized in combination with mathematical
models and tracer tests to define the contributions to total measured
emissions of specific sources among a group of sources.
2.1.3 Roof Monitor Sampling Method
This method is used to measure the fugitive emissions entering
the ambient air from building or other enclosure openings such as roof
monitors, doors, and windows from enclosed sources too numerous or un-
-3-
-------
wieldy to permit the installation of temporary hooding. Sampling is,
in general, limited to a mixture of all uncontrolled emission sources
within the enclosure and requires the ability to make low air velocity
measurements and mass balances of small quantities of materials across
the surfaces of the openings.
These features are embodied in the typical industrial sources and
their emitted pollutants contained in Table 2-1.
The roof monitor method quantifies the emissions from such sources
as the average mass flux of emissions from buildings or enclosure openings
over the time period of measurement. The flux is obtained from air and
pollutant material balances across the openings. Tracer tests may also
be used in combination with it to define the contributions of individual
sources.
2. 2 Selection of Sampling Method
The initial step in the measurement and documentation of fugitive
emissions at an industrial site is the selection of the sampling method
to be employed. Although it is impossible to enumerate all the combina-
tions of influencing factors that might be encountered in a specific
situation, careful consideration of the following general criteria should
result in the selection of the most effective sampling method.
2.2.1 Selection Criteria
The selection criteria listed below are grouped into three general
classifications common to all fugitive emissions measurement methods.
The criteria are intended to provide only representative examples and
should not be considered a complete listing of influencing factors.
-4-
-------
TABLE 2-1
TYPICAL INDUSTRIAL FUGITIVE EMISSIONS SOURCES
MEASURED BY THE ROOF MONITOR SAMPLING METHOD
Industry
Iron & Steel Foun-
dries
Electric Furnace
Steel
Primary Aluminum
Primary Copper
Tires & Rubber
Phosphate Fertili-
zer
Lime
Primary Steel
Graphite, and
Carbide Pro-
duction
Source
Furnace or Cupola
Charging
Melting
Mold Pouring
Charging
General Operations
Carbon Plant
Potroom
Alumina Calcining
Cryolite Recovery
Converter House
Reverberatory Fur-
nace
Roaster Operations
Curing Press Room
Cement House
General Ventila-
tion
General Ventila-
tion
Blast Furnace
Cast House
BOF Operations
Open Hearth
Operations
Arc Furnace
Operation
Particulate
Emissions
Fume, Carbon Dustj
Smoke (Oil)
Fume, Dust
Dust
Metallic Fumes,
Carbon Dust
Metallic Fumes,
Dust
Tars, Carbon Dust
Tars, Carbon &
Aluminum Dust,
Flourides
Alumina Dust
Carbon & Alumina
Dust, Flourides
Fume, Silica
Fume
Fume
Organic Partic-
ulate
Dust
Dust, Flourides
Dust
Metallic Fumes
Metallic Fumes,
Carbon Dust
Metallic Fumes
Carbon Dust,
Silica Fume
Gas and
Vapor Emissions
CO, HC, S02
CO, S02
CO, HC, PNA, Odor
CO
CO
CO, HC, S02
CO, HC, S02, HF
S02
S02
S02
HC, Odor
HC, Odor
S02, HF
-
CO, H2S, S02
CO
CO
CO, Odor
-5-
-------
2.2.1.1 Site Criteria
Source Isolability. Can the emissions be measured separately
from emissions from other sources? Can the source be enclosed?
Source Location. Is the source indoors or out? Does location
permit access of measuring equipment?
Meteorological Conditions. Will wind conditions or precipita-
tion interfere with measurements? Will rain or snow on ground
effect dust levels?
2.2.1.2 Process Criteria
Number and Size of Sources. Are emissions from a single, well
defined location or many scattered locations? Is source small
enough to hood?
Homogeneity of Emissions. Are emissions the same type every-
where at the site? Are reactive effects between different
emissions involved?
Continuity of Process. Will emissions be produced long enough
to obtain meaningful samples?
Effects of Measurements. Will installation of measuring equip-
ment alter the process or the emissions? Will measurements
interfere with production?
2.2.1.3 Pollutant Criteria
Nature of Emissions. Are measurements of particles, gases,
liquids required? Are emissions hazardous?
Emission Generation Rate. Are enough emissions produced to
provide measurable samples in reasonable sampling time?
Emission Dilution. Will transport air reduce emission con-
centration below measurable levels?
2.2.2 Criteria Application
The application of the selection criteria listed in Section 2.2.1
to each of the fugitive emissions measurement methods defined in Section
2.1 is described in general terms in this Section.
-6-
-------
2.2.2.1 Quasi-stack Method
Effective use of the quasi-stack method requires that the source of
emissions be isolable and that an enclosure can be installed capable of
capturing emissions without interference with plant operations. The lo-
cation of the source alone is not normally a factor. Meteorological
conditions usually need be considered only if they directly affect the
sampling.
The quasi-stack method is usually restricted to a single source
and must be limited to two or three small sources that can be effec-
tively enclosed to duct their total emissions to a single sampling point.
The process may be cyclic in nature if any one cycle is of sufficient
duration to provide a representative sample. The possible effects of
the measurement on the process or emissions is of special significance
in this method. In many cases, enclosing a portion of a process in
order to capture its emissions can alter that portion of the process
by changing its temperature profile or affecting flow rates. Emission
may be similarly altered by reaction with components of the ambient air
drawn into the sampling ducts. While these effects are not necessarily
limiting in the selection of the method, they must be considered in de-
signing the test program and could influence the method selection by
increasing complexity and costs.
The quasi-stack method is useful for virtually all types of emis-
sions and is least affected by the emission generation rate of the
process. Dilution of the pollutants of concern is of little consequence
since it can usually be controlled in the design of the sampling system.
-------
2.2.2.2 Roof Monitor Method
Practical utilization of the roof monitor method demands that the
source of emissions be enclosed in a structure with a limited number of
openings to the atmosphere. Measurements may usually be made only of
the total of all emissions sources within the structure. Meteorological
conditions normally need not be considered in selecting this method.
The number of sources and the mixture of emissions is relatively
unimportant since the measurements usually include only the total emis-
sions. The processes involved may be discontinuous as long as a repre-
sentative combination of the worst grouping may be included in a sam-
pling. Measurements will normally have no effect on the processes or
emissions.
The roof monitor method, usually dependent on or at least influ-
enced by gravity in the transmission of emissions, may not be useful
for the measurement of larger particulates and heavy gases which may
settle within the enclosure being sampled. Emissions generation rates
must be high enough to provide pollutant concentrations of measurable
magnitude after dilution in the enclosed volume of the structure.
2.2.2.3 Upwind-Downwind Method
The upwind-downwind method, generally utilized where neither of
the other methods may be successfully employed, is not influenced by
the number or location of the Emission sources except as they influence
the locating of sampling devices. In most cases, only the total con-
tribution to the ambient atmosphere of all sources within a sampling
area may be measured. The method is strongly influenced by meteoro-
logical conditions, requiring a wind consistent in direction and ve-
locity throughout the sampling period as well as conditions of temper-
-8-
-------
ature, humidity and ground moisture representative of normal ambient
conditions.
The emissions measured by the upwind-downwind method may be the
total contribution from a single source or from a mixture of many sources
in a large area. Continuity of the emissions is generally of little
consequence since the magnitude of the ambient air volume concerned is
large enough to provide a smoothing effect to any circle emissions.
The measurements have no effect on the emissions or processes involved.
Most airborne pollutants can be measured by the upwind-downwind
method. Generation rates must be high enough to provide measurable
concentrations at the sampling locations after dilution with the am-
bient air. Settling rates of the larger particulates require that the
sampling system be carefully designed to ensure that a representative
pollutant cloud is included.
2.3 Sampling Strategies
Fugitive emissions measurements may, in general, be separated into
two classes or levels depending upon the degree of accuracy desired.
Survey measurement systems are designed to screen emissions and to
provide gross measurements of a number of process influents and efflu-
ents; detailed systems are designed to isolate, identify accurately,
and quantify individual contaminant constituents.
2.3.1 Survey Measurement Systems
Survey measurement systems employ recognized standard-or state-
of-the-art measurement techniques to screen the total emissions from a
site or source and determine whether any of the emission constituents
-9-
-------
should be considered for more detailed investigation. They generally
utilize the simplest available arrangement of instrumentation and pro-
cedures in a relatively brief sampling program, usually without pro-
visions for sample replication, to provide order-of-magnitude type data,
embodying a factor of 2 to 5 in accuracy range with respect to actual
emissions.
2.3.2 Detailed Measurement Systems
Detailed measurement systems are used in instances where survey
measurements or equivalent data indicate that a specific emission con-
stituent may be present in a concentration worthy of concern. Detailed
systems provide more precise identification and quantification of spe-
cific constituents by utilizing the latest state-of-the-art measure-
ment instrumentation and procedures in carefully designed sampling pro-
grams. Detailed systems are also utilized to provide emission data over
a range of process operating conditions or ambient meteorological in-
fluences. Basic accuracy of detailed measurements is in the order of
+ 10 to + 50 percent of actual emissions.
-10-
-------
30. TEST STRATEGIES
This section describes the approaches that may be taken to success-
fully complete a testing program utilizing the roof monitor sampling
method described in Section 2.1. It details the information required
to plan the program, describes the organization of the test plan, spe-
cifies the types of sampling equipment to be used, establishes criteria
for the sampling system design, and outlines basic data reduction methods.
3.1 Pretest Survey
After the measurement method to be utilized in documenting the fugi-
tive emissions at a particular site has been established using the cri-
teria of Section 2.2, a pretest survey of the site should be conducted
by the program planners. The pretest survey should result in an infor-
mal, internal report containing all the information necessary for the
preparation of a test plan and the design of the sampling system by the
testing organization.
This section provides guidelines for conducting a pretest survey
and preparing a pretest survey report.
3.1.1 Information to be Obtained
In order to design a system effectively and plan for the on-site
sampling of fugitive emissions, a good general knowledge is required of
the plant layout, process chemistry and flow, surrounding environment,
and prevailing meteorological conditions. Particular characteristics
of the site relative to the needs of the owner, the products involved,
the space and manpower skills available, emission control equipment in-
-11-
-------
stalled, and the safety and health procedures observed, will also Influ-
ence the sampling system design and plan. Work flow patterns and sched-
ules that may result in periodic changes in the nature or quantity of
emissions or that indicate periods for the most effective and least dis-
ruptive sampling must also be considered. Most of this information can
only be obtained by a survey at the site. Table 3-1 outlines some of
the specific information to be obtained. Additional information will
be suggested,by considerations of the particular on-site situation.
3.1.2 Report Organization
The informal, internal pretest survey report must contain all the
pertinent information gathered during and prior to the site study. A
summary of all communications relative to the test program should be
included in the report along with detailed descriptions of the plant
layout, process, and operations as outlined in Table 3-1. The report
should also incorporate drawings, diagrams, maps, photographs, meteo-
rological records, and literature references that will be helpful in
planning the test program.
3.2 Test Plan
3.2.1 Purpose of a Test Plan
Measurement programs are very demanding in terms of the scheduling
and completion of many preparatory tasks, observations at sometimes
widely separated locations, instrument checks to verify measurement va-
lidity, etc. It is therefore essential that all of the experiment de-
sign and planning be done prior to the start of the measurement program
-12-
-------
TABLE 3-1
PRE-TEST SURVEY INFORMATION TO BE OBTAINED
FOR APPLICATION OF FUGITIVE EMISSION SAMPLING METHODS
Plant
Layout
Process
Drawings:
Building Layout and Plan View of Potential Study Areas
Building Side Elevations to Identify Obstructions and
Structure Available to Support Test Setup
Work Flow Diagrams
Locations of Suitable Sampling Sites
Physical Layout Measurements to Supplement Drawings
Work Space Required at Potential Sampling Sites
Process Flow Diagram with Fugitive Emission Points
Identified
General Description of Process Chemistry
General Description of Process Operations Including
Initial Estimate of Fugitive Emissions
Drawings of Equipment or Segments of Processes Where
Fugitive Emissions are to be Measured
Photographs (if permitted) of Process Area Where
: Fugitive Emissions are to be Measured
Names, Extensions, Locations of Process Foremen and
'. Supervisors Where Tests are to be Conducted
Operations
Location of Available Services (Power Outlets, Main-
tenance and Plant Engineering Personnel, Labora-
tories, etc.)
Local Vendors Who Can Fabricate and Supply Test System
Components
Shift Schedules
Location of Operations Records (combine with process
operation information)
Health and Safety Considerations
Other
Access Routes to the Areas Where Test Equipment/Instru-
mentation Will Be Located
Names, Extensions, Locations of Plant Security and
Safety Supervisors
Regional Meteorological Summaries
-13-
-------
in the form of a detailed test plan. The preparation of such a plan
enables the investigator to "pre-think" effectively and cross-check all
of the details of the design and operation of a measurement program
prior to the commitment of manpower and resources. The plan then also
serves as the guide for the actual performance of the work. The test
plan provides a formal specification of the equipment and procedures re-
quired to satisfy the objectives of the measurement program. It is
based on the information collected in the informal pretest survey re-
port and describes the most effective sampling equipment, procedures,
and timetables consistent with the program objectives and site charac-
teristics.
3.2.2 Test Plan Organization
The test plan should contain specific information in each of the
topical areas indicated below:
Background
The introductory paragraph containing the pertinent infor-
mation leading to the need to conduct the measurement program
and a short description of the information required to answer
that need.
Objective
A concise statement of the problem addressed by the test
program and a brief description of the program's planned method
for its solution.
Approach
A description of the measurement scheme and data reduc-
tion methodology employed in the program with a discussion of
how each will answer the needs identified in the background
statement. .
-14-
-------
Instrumentation/Equipment/Facilities
A description of the instrumentation arrays to be used to
collect the samples and meteorological data identified in the
approach description. The number and frequency of samples to
be taken and the sampling array resolution should be described.
A detailed description of the equipment to be employed
and its purpose.
A description of the facilities required to operate the
measurement program, including work space, electrical power,
support from plant personnel, special construction, etc.
Schedule
A detailed chronology of a typical set of measurements, or
a test, and the overall schedule of events from the planning
stage through the completion ofthe test program report.
Limitations
A definition of the conditions under which the measurement
project is to be conducted. If, for example, successful tests
can be conducted only during occurrences of certain source opera-
tions, those favorable limits should be stated.
Analysis Method
A description of the methods which will be used to analyze
the samples collected and the resultant data, e.g., statistical
or case analysis, and critical aspects of that method.
Report Requirements
A draft outline of the report on the analysis of the data
to be collected along with definitions indicating the purpose
of the report and the audience it is to be directed to.
Quality Assurance
The test plan should address itself to the development of
a quality assurance program as outlined in Section 3.7. This
QA program should be an integral part of the measurement pro-
gram and be incorporated as a portion of the test plan either
directly or by reference.
Responsibilities
A list of persons who are responsible for each phase of
the measurement program, as defined in the schedule, both for
the testing organization and for the plant site.
-15-
-------
3.3 Roof Monitor Sampling Strategies
The roof monitor sampling method, as described in Section 2.183,
is used to quantify emissions released into the internal atmosphere of
the buildings or enclosures that contain the process equipment and which
are then ventilated to the external atmosphere as fugitive emissions. The
roof monitor sampling method may be utilized to measure the fugitive
emissions from almost any process that ventilates through building open-
ings such as doors, windows, or any of a wide variety of roof ventilators,
where the ventilation is either gravity dependent or fan driven.
The measurements made include that of the gas flow through the open-
ing either by direct measurement or by calculation (of the gas velocity)
from physical parameters (pressure drop, thermal conductivity), the
cross-sectional area of the opening, and the particulate and gaseous emis-
sion concentrations in the flowing gas. These measurements or calculations
provide the data necessary to determine the total flux of the fugitive
emissions from all sources operating within the enclosure or from selected
sources, depending on processing sequences or cycles. Since ventilation
rates, especially when gravity driven, can vary, the mass emission rates
so measured are averages over the emission concentration and velocity
measurement period. (Sections 3.4 and 3.5 describe the equipment used
for sampling, the criteria for sampling system design, sampling techniques,
and data reduction procedures for respectively, survey and detailed
roof monitor sampling programs).
3.4 Survey Roof Monitor Sampling Strategy
A survey measurement system, as defined in Section 2,3, is designed
to provide gross measurements of emissions to determine whether any
-16-
-------
constituents should be considered for more detailed investigations. A
survey roof monitor measurement system in its simplest form utilizes
one or two hi-vol type samplers set up to sample the openings by which
the fugitive emissions exit the building or enclosure and an equal num-
ber of hot wire or rotating vane anemometers for determining the gas
velocity exiting the openings. The weight of particulates/volume of
sample air collected and the average velocity across the openings are
combined with the measured area of the opening to calculate the emission
rate of the source. Grab samples of gaseous emissions may be taken at
the same time as the particulate samples and the emission rate calculated
in the same manner. Size distribution of the particulates may also be
obtained simultaneously from a variety of methods.
3.4.1 Sampling Equipment
Pollutants that may be measured by the roof monitor technique are
limited to those that can be airborne sufficiently to exit the enclosure
or structure through the vent openings, i.e., particulates and gases. The
gross measurement requirements for survey sampling of particulates are
best satisfied by high volume filter impaction devices to provide data
on the average emission rate, particle size distribution, and particle
composition. Particle charge transfer or piezoelectric mass monitoring
devices may be utilized for continuous or semi-continuous sampling of
intermittent emission sources where peak levels must be defined.
Gaseous emissions in survey programs are usually grab-sampled for
laboratory analysis using any of a wide variety of evacuated sampling
vessels. Continuous or semi-continuous sampling of specific gases may
-17-
-------
be accomplished using such devices as, for example, continuous monitor
flame ionization detectors (for hydrocarbons) and automated West-Gaeke
bubblers/impingers (for sulfur dioxide). Figures 3-1 and 3-2 show
typical setups utilized for roof monitor/ventilator sampling for fugi-
tive emissions.
3.4.2 Sampling Systems Design
The number and location of devices used to collect samples are
extremely important to the successful completion of a survey roof
monitor sampling program, especially since the program is designed for
minimum cost and provides for no replication of samples. The design of
the sampling system is influenced by such factors as source complexity
and size, physical location and size of the vent openings, variability
of the mass rate and temperature of the emissions, as well as the
homogeneity of the emissions. Most situations will, in general, fit
into some combination of the following parameters:
Source - Sources may be either homogeneous, emitting a single type
of mixture of pollutants from each and every emission location, or
heterogeneous, emitting different types or mixtures of pollutants
from different locations. The resultant pollutant emission "cloud"
("cloud" being used to describe the fugitive emission plume bound-
aries) from a homogeneous source will be homogeneous. The pollutant
as a result of mixing by suitably directed or turbulent enclosure/
structure air flow, homogeneous. The physical size of a source will
determine the extent of the pollutant emission "cloud" and may in-
fluence its homogeneity. The proximity of sources within the en-
closure/structure will also determine the extent of the "cloud" and
its homogeneity.
Emission Character - The time duration of the emissions may limit
the effective sampling time. Sources which have a short time cycle
(<10-15 minutes) may require different sampling methods than those
of a one-hour or more time scale. The temperatures of the emissions
will also effect sampling. Excessive temperatures may limit the
sampling time for the emissions. If temperatures cycle excessively,
instrumentation which can quickly adjust to this cycle would be
required.
-------
Togas
analyzers
^s— Pulley
Arm
Gaseous emission
sample line
/_
Power
line
Hi-Vol
Detail B
Fig. 3-1. Electric arc furnace operation; roof monitor showing
sampling/mounting configuration.
-19-
-------
Roof or wall
ventilator
Exhaust
fan
Gaseous emission
sample line
Gas analyzer(s)
Fig. 3-2a. Roof or wall ventilator sampling configuration (with or
without fan).
Particle charge
transfer monitoring
system
Gaseous emission
monitoring system
Fig. 3-2b. Roof ventilator sampling configuration.
-20-
-------
Site Accessibility - If the site is not readily accessible, continuous
monitoring equipment, which is usually higher in cost and also in
complexity of arrays, might be required to measure the fugitive
emissions. If standard hi-vols are used, extra samplers would need
to be located in the roof monitor to conserve the number of times
the sampling site has to be accessed to recover samples. Remote
timing equipment and remote recording would be required also.
Emission Cycle - If the emission cycle is short, continuous monitor-
ing equipment may be required. If not, multiple samples may need to
be taken on the same filter. In this case, a remote timing and
recording equipment would be required.
Table 3-3 outlines elements of conceptual systems for roof monitor
sampling programs. These elements are keyed to the numbers on the Matrix
of Table 3-2, and they correspond to the appropriate system elements need-
ed to measure fugitive emissions for that matrix entry. Each matrix
entry corresponds to a specific combination of factors which make up a
particular roof monitor sampling program for a specific source.
3.4.3 Sampling Techniques
Sampling must be scheduled and carefully designed to ensure that
data representative of the emission conditions of concern are obtained.
Effective scheduling demands that sufficient knowledge of operations
and process conditions be obtained to determine proper starting times
and durations for samplings. The primary concern of the sampling design
is that sufficient amounts of the various pollutants are collected to
provide meaningful measurements.
Each of the various sample collection and analysis methods has an
associated lower limit of detection, typically expressed in terms of
micrograms of captured solid material and either micorgrams per cubic
meter or parts per million in air of gases. Samples taken must provide at
-21-
-------
TABLE 3-2
MATRIX OF POSSIBLE COMBINATIONS OF KEY TEST PARAMETERS
i
NJ
Combination
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Source
Homogeneity
Homogeneous
Homogeneous
Homogeneous
Homogeneous
Homogeneous
Homogeneous
Homogeneous
Homogeneous
Heterogeneous
Heterogeneous
Heterogeneous
Heterogeneous
Heterogeneous
Heterogeneous
Heterogeneous
Heterogeneous
Emissions
Point
Geometry
Simple
Complex
Simple
Complex
Simple
Complex
Simple
Complex
Simple
Complex
Simple
Complex
Simple
Complex
Simple
Complex
Site
Accesibility
Easy
Difficult
Difficult
Easy
Easy
Difficult
Difficult
Easy
Easy
Difficult
Difficult
Easy
Easy
Difficult
Difficult
Easy
Emission
Cycle
Short
Long
Short
Long
Long
Short
Long
Short
Short
Long
Short
Long
Long
Short
Long
Short
Suitable
System
Elements
(1),(4) (1) ... etc.
,_v Numbers refer
to conceptual
(4) system elements
for a roof moni-
tor sampling
(1) program most
suitable for a
given matrix
(1) , (4) element, as de-
,_,. scribed in Table
C ' 3-2.
(4)
(6,)(5)
(4)
(6), (5)
(2)
(5)
(4)
(5)
-------
5.
6.
TABLE 3-3
ELEMENTS OF CONCEPTUAL SYSTEMS FOR A
ROOF MONITOR SAMPLING PROGRAM AS APPLIED TO
SPECIFIC TYPES OF FUGITIVE EMISSION SOURCES*
One Hi-Vol Sampler
One Rotating Vane Anemometer
One Cascade Impactor
Two Hi-Vol Samplers
Two Rotating Vane Anemometers
Two Cascade Impactors
One Hi-Vol Sampler
One Rotating Vane Anemometer
One Cascade Impactor
One Portable Anemometer (Vane
or Hot Wire
One Respirable Dust Monitor
One Continuous Particulate
Monitor
One Rotating Vane Anemometer
One Cascade Impactor
One Continuous Particulate
Monitor
One Rotating Vane Anemometer
One Cascade Impactor
One Portable Anemometer
One Respirable Dust Monitor
Two Hi-Vol Samplers
Two Rotating Vane Anemometers
Two Cascade Impactors
One Portable Anemometer
One Respirable Dust Monitor
Fixed Station
In Monitor
Fixed Station
In Monitor
Fixed Station
In Monitor
Manual Traverse
of Doors & Windows
Movable Across and
Down Roof Monitor
Movable Across and
Down Roof Monitor
Manual Traverse of
Doors & Windows
Fixed Station
In Monitor
Manual Traverse of
Doors & Windows
*A11 gaseous sampling done using grab samples for
laboratory analysis.
-23-
-------
least these minimum amounts of the pollutants to be quantified. The mass (M)
of a pollutant collected is the product of the concentration of the pollu-
tant in the air (x) and the volume of air sampled (V) , thus,
M (micrograms) = ^ (micrograms/cubic meter) x V (cubic meters) .
To ensure that a sufficient amount of pollutant is collected, an ade-
quately large volume of air must be passed through such samplers as
particle filters or gas absorbing trains for a specific but uncontrolla-
ble concentration. The volume of air (V) is the product of its flow
rate (F) and the sampling time (T) , or,
V (cubic meters) = F (cubic meters/minute) x T (minutes).
Since the sampling time is most often dictated by the test conditions,
the only control available to an experimenter is the sampling flow rate.
A preliminary estimate of the required flow rate for any sampling loca-
tion may be made if an estimate or rough measurement of the concentration
expected is available. The substitution and rearrangement of terms in
the above equations yields Equation 3-1:
F (cubic meters/minute) = M (micrograms/x (micrograms/cubic meter)
x T (minutes) . (3-1)
This equation permits the calculation of the minimum acceptable flow
rate for a required sample size. Flow rates should generally be adjusted
upward by a factor of at least 1.5 to compensate for likely inaccuracies
in estimates of concentration.
-24-
-------
Grab samples of gaseous pollutants provide for no means of pollutant
6
sample quantity control except in terms of the volume of the sample.
Care should be taken, therefore, to correlate the sample size with the
requirements of the selected analysis method.
The location of samplers is also important in obtaining representative
data. Where the emissions are known to exit the roof monitor or vent in
a homogeneous pollutant "cloud", one sampler can be used. However, where
the pollutant "cloud" is not known to be homogeneous or is definitely
heterogeneous, samplers should be located at 25-100 ft intervals.
In addition, unless approximations can be made based upon relative
flowrates, a sampler must be located at each separate roof monitor or
vent location on the building/enclosure. This can be simplified if in-
spection of the site indicates that some of these vents are only minor
sources of the fugitive emissions.
A critical concern in development of the mass emission rates from
roof monitor fugitive emission tests is the accuracy of the flow measure-
ments required to change air quality measurements into mass emissions.
The basic equation is:
Mass Rate (micrograms/minute) = M (micrograms)/T (minutes) =
X (micrograms/cubic meter) x F (cubic meters/minute)
Where x is known quite accurately, F is the overriding error limit
for fugitive emissions measurements. F can be obtained from:
F (cubic meters/minute) = A (square meters) x U (meters/second)
Preliminary estimates of the linear velocity (V) can be obtained
by use of a hand hot wire anemometer with a digital or scale read-
out. These will serve to determine what method of velocity measurement
-25-
-------
TABLE 3-4
RANGE OF APPLICABILITY OF COMMON VELOCITY
MEASUREMENT DEVICES FOR ROOF MONITOR SAMPLING
Device
Hot Wire
Anemometer*
Rotating Vane
Anemometer
Pitot Tube
Calibrated
Magnehelic
Gauge**
Flow Range
10-8000 fpm
100-6000 fpm )
50-6000 fpm J
500-6000 fpm
2000-10,000 fpm
Accuracy
Fair
Fair at Low fpm J
Good at High fpml
Good
Good
Usable
Temp . Range
0-225°F
0-150°F Mechanical
0-200°F Electric
0-2000°F***
0-200°F
*Cannot be used for sources with significant steam or water content.
**Although accurate has very narrow range of flow measurement and must
be calibrated for opening used.
***Water cooled for high temperatures.
-26-
-------
will be the most accurate. Temperature readings should also be taken
to determine the most suitable instrument. Table 3-4 summarizes data
on the four instruments which would be most suitable, which are:
1. Hot Wire Anemometers
2. Rotating Vane Anemometers
3. Pitot Tubes
4. Magnehelic Gauges (after calibration)
The method chosen must take into account:
1. Compatibility with chosen sampling site conditions,
2. Compatibility with desired error limits of tests.
3.4.4 Data Reduction
When the sampling program has been completed and the samples have
been analyzed to yield average pollutant concentrations in micrograms of
particulate matter or parts per million of gases in the pollutant emis-
sion "cloud", the source strength must be calculated. As previously
mentioned, this requires the multiplication of these values by the
cross sectional area of the opening and the average linear velocity
across that opening. This must be done for every significant roof monitor
or vent in the building/enclosure studied to establish the process fugitive
emission rate in grams per second, or other appropriate mass emission rate
units.
3.5 Detailed Roof Monitor Sampling Strategy
A detailed measurement system is designed to more precisely identify
and quantify specific pollutants that a survey measurement or equivalent
data indicate as a possible problem area. A detailed system is necessarily
more complex than a survey system in terms of equipment, system design,
-27-
-------
sampling techniques and data reduction. It requires a much larger invest-
ment in terms of equipment time and manpower and yields data detailed and
dependable enough for direct action towards achieving emission control.
Detailed systems in general employ sampling networks to measure the
concentration and distribution of specific pollutants within the pollutant
emission "cloud". The detailed measurements of pollutant distribution and
emission rate variation replace the averaging techniques or the assumptions
of representativeness of the sampling done in survey sampling systems.
Detailed systems are frequently employed to compare the emissions at different
process or operating conditions to determine which conditions dictate the
need for emission control.
The data provided by the sampling network are processed in conjunction
with detailed studies of the volumetric flow rate of the emissions from
the roof monitor or vents to determine mass emission rates from the fugitive
sources.
The complexity of a detailed system is largely determined by the
basic accuracy desired; increasing accuracy demands more measurements
either in the number of locations measured or in the number of measure-
ments made at each location, or both. Most detailed systems will require
a network of sets of instrumentation located across the plane of the
opening to make simultaneous measurements since the usually lower con-
centrations of specific emissions preclude the use of traversing tech-
niques with inherently short sampling durations, or assumptions regard-
ing the distribution of emissions in the flow through the opening.
-28-
-------
Identification and quantification of a specific fugitive emission
from an enclosed source may involve measurements at more than one build-
ing opening if the flow through the separate openings is of comparable
magnitude and the openings are situated to result in selectivity in the
character or quantity of the emission being vented. This could occur,
for example, when a roof monitor and a floor level door or window both
vent emissions from a variety of sources within a building. Lighter
gaseous emissions and smaller particulates would be expected to vent
through the monitor, while the heavier gases and larger particulates
would tend to settle and vent through the lower opening. If either of
the openings is situated to vent all or most of the emissions from a
specific source, resulting in a different type of emission for the two
openings, the detailed measurement system might require different types
of instrumentation at each location, thus adding to the system complex-
ity.
3.5.1 Sampling Equipment
The pollutants to be characterized by a detailed roof monitor sam-
pling system fall into the same two basic classes—airborne particulates
and gases—as those measured by survey systems. Detailed sampling and
analysis equipment is generally selected to obtain continuous or semi-
continuous measurements of specific pollutants rather than grab-sampled
overall measurement.
Particulate samples are collected using filter impaction, piezo-
electric, and size selective or adhesive impaction techniques. Gases
—29—
-------
are sampled and analyzed using flame ionization detectors, bubbler/im-
pinger trains, non-dispersive infrared or ultraviolet monitors, flame
photometry, and other techniques specific to individual gaseous pollu-
tants.
The selection of suitable sampling equipment should be influenced
by such considerations as portability, power requirements, detection
limits and ease of control.
3,5.2 Sampling System Design
The basic criteria reviewed in Section 3.4.2 for the design of a
survey sampling system are generally applicable to the design of a de-
tailed system. The need for replacement of survey assumptions as to
pollutant distribution with actual measured values, however, most fre-
quently requires the design of a sampling network that will provide
samples of a distribution at various distances along the width of the
source in both the horizontal and vertical directions. Sampler locations
may generally be determined in the same manner as those for a survey systems
except that they must be capable of finer analysis of pollutant distri-
bution. For detailed measurements, each location must make provision for
sampling across the section of the pollutant emission "cloud" horizontally
and/or vertically. Horizontal distributions over the length of the roof
monitor may be measured by adding a number of samplers (usually at least
two) at either side of the survey sampler location at distances estimated
to yield significantly different pollutant concentrations. Vertical dis-
tributions as well as horizontal distributions across the width of the
roof monitor are best determined by traversing with the samplers or their
probe devices.
-30-
-------
General rules which might be applied to system design are as
follows:
1. If emissions are reasonably homogeneous, sampler locations
along the horizontal length of the roof monitor should be
25-50 ft apart maximum. If heterogeneous, they should be
10-20 ft apart.
2. Vertical distances greater than 10-20 ft in roof monitor open-
ings would require either vertically tiered samplers or travers-
ing arrangements.
3. Traversing across the width of a roof monitor or setting up a
network in that width can be employed to sample emissions before
they leave the roof monitor. In cases where external accessi-
bility is a problem, this can be used to obtain representative
samples without leaving the building.
4. If any significant emissions (> 10%) are presumed to exit the
enclosure/structure by other than the roof monitor, that vent
or exit should have its own sampler system.
5. Where a minor (< 10%) amount of emissions are presumed to exit
the enclosure/structure by other than the roof monitor, some
estimate of this should be obtained using a portable and simpli-
fied sampler system (survey type). There can be many such
openings and caution should be applied to avoid excess expendi-
ture of time/money for tests of such minor sources.
3.5.3 Sampling Techniques
In order to obtain representative results of detailed quality, sam-
pling techniques must:
1. Differentiate the peak emissions from the average fugitive
emissions of a process. Online continuous readout devices are
preferable in these cases.
2. Determine the horizontal and vertical distribution of pollutants
within the emission "cloud". Multiple online continuous readout
devices as well as traversing are preferable in these cases.
3. Differentiate specific components of the emissions, preferably
those of highest hazard/toxicity to humans. Single component
continuous online monitors or detailed laboratory analysis of
collected samples of particulates, gases or liquids are preferred.
-31-
-------
The specific techniques which might be employed vary. However,
the selection criteria should include:
1. Portability
2. Power Requirements
3. Detection Limits
4. Response Time
5. Ease of Control (remote or close at hand)
3.5.4 Data Reduction/Data Analysis
After the analyses for pollutants are completed, the required cal-
culations are made for emission concentrations, including calculations
for the mean and standard deviation. Statistical differences between
test methods can be obtained and confirmed by conducting various statis-
tical significance procedures such as the "t" and "f" tests on the mean
and standard deviation values for the various test methods. A tabula-
tion of the statistical analysis results can then be made and related
to the process conditions at the time of the tests. Finally, the inves-
tigator can determine whether there is a correlation between the emission
results by test method and the process conditions.
3.6 Tracer Tests
Complex sources, consisting of several different sources with similar
or very different emission rate patterns, can be the cause of the fugitive
emissions from the roof monitor of a structure or enclosure. Emission
measurements at the roof monitor of complex sources must be related back
to a specific source to determine what is the most significant cause of
figutive emissions. Tracers can be released at specific rates at the location
of the source to be studies for specific time periods. Knowledge of this,
-32-
-------
as well as what sampler caught this tracer and in what concentration,
can serve to differentiate each source's contribution to the fugitive
emissions.
3.6,1 Tracers and Samplers
Both particulate and gaseous atmospheric tracers are in general
use. The most commonly used particulate tracers are zinc sulfide and
sodium fluorescein (uranine dye). The primary gaseous tracer is sulfur
hexafluoride (SFg).
Zinc sulfide is a particulate material which can be obtained in
narrow size ranges to closely match the size of the pollutant of con-
cern. The material is best introduced into the atmosphere in dry form
by a blower type disseminator although it can be accomplished by
spraying from an aqueous slurry solution. The zinc sulfide fluoresces
a distinctive color under ultraviolet light which provides a specific
and rapid means of identification and quantification of the tracer in
the samples.
Sodium fluorescein is a soluble fluorescing particulate material.
It is normally spray disseminated from an aqueous slurry solution to
produce a particulate airborne plume, the size distribution of which
can be predetermined by the spraying apparatus. Sodium fluorescein
can be uniquely identified by colorimeter assessment.
Sulfur hexafluoride is a gas which can be readily obtained in
ordinary gas cylinders. Sulfur hexafluoride can be disseminated by
metering directly from the gas cylinder through a flow meter to the
atmosphere. The amount disseminated can be determined by careful flow
metering and/or weight differentiation of the gas cylinder.
-33-
-------
Particulate tracers are usually sampled with filter impaction de-
vices or, for particles over 10 microns in diameter, the more easily
used and somewhat less accurate Rotorod sampler which collects particles
on an adhesive-coated U- or H-shaped rod which is rotated in the am-
bient air by a battery-driven electric motor.
Sulfur hexafluoride gaseous samples are collected for laboratory
gas chromatograph analysis in non-reactive bags of such materials as
Mylar.
3.6.2 Tracer Sampling System Design
All of the design guidelines presented in 3.4.2 and 3.5.2 may be
applied to the design of a tracer sampling system as site conditions
dictate. Their application is, in general, simplified since the source
strength may be controlled to provide measurable tracer concentrations
at readily accessible sampling locations.
A single ambient sampler will usually be sufficient to establish
that no significant amount of the tracer material is present in the am-
bient atmosphere approaching the source, enclosure or structure.
3,6.3 Tracer Sampling and Data Analysis
The methods introduced in Sections 3.4.3 and 3.5.3 for determining
sampler design and location are fully applicable to tracer sampling.
Like design guidelines, they may be more easily applied because of
the control of source strength available.
The analysis of the data is also simplified since the source strength
is known and no back-calculation is required.
-34-
-------
3.7 Quality Assurance
The basic reason for quality assurance on a measurement program is
to insure that the validity of the data collected can be verified. This
requires that a quality assurance program be an integral part of the
measurement program from beginning to end. This section outlines the
quality assurance requirements of a sampling program in terms of several
basic criteria points, The criteria are listed below with a brief ex-
planation of the requirements in each area. Not all of the criteria
will be applicable in all fugitive emission measurement cases.
1. Introduction
Describe the project organization, giving details of the
lines of management and quality assurance responsibility.
2. Quality Assurance Program
Describe the objective and scope of the quality assurance
program,
3. Design Control
Document regulatory design requirements and standards ap-
plicable to the measurement program as procedures and specifi-
cations.
4. Procurement Document Control
Verify that all regulatory and program design specifications
accompany procurement documents (such as purchase orders).
5. Instructions, Procedures, Drawings
Prescribe all activities that affect the quality of the
work performed by written procedures. These procedures must
include acceptance criteria for determining that these activ-
ities are accomplished,
6. Document Control
Ensure that the writing, issuance, and revision of proce-
dures which prescribe measurement program activities affecting
quality are documented and that these procedures are distributed
to and used at the location where the measurement program is
carried out.
-35-
-------
7. Control of Purchase Material, Equipment, and Services
Establish procedures to ensure that purchased material con-
forms to the procurement specifications and provide verification
of conformance.
8. Identification and Control of Materials, Parts, and Components
Uniquely Identify all materials, parts, and components that
significantly contribute to program quality for traceability
and to prevent the use of incorrect or defective materials,
parts, or components.
9. Control of Special Processes
Ensure that special processes are controlled and accomplished
by qualified personnel using qualified procedures.
10. Inspection
Perform periodic inspections where necessary on activities
affecting the quality of work. These inspections must be or-
ganized and conducted to assure detailed acceptability of pro-
gram conponents.
11, Test Control
Specify all testing required to demonstrate that applicable
systems and components perform satisfactorily. Specify that
the testing done and documented according to written proce-
dures, by qualified personnel, with adequate test equipment
according to acceptance criteria.
12. Control of Measuring and Test Equipment
Ensure that all testing equipment is controlled to avoid
unauthorized use and that test equipment is calibrated and
adjusted at stated frequencies. An inventory of all test
equipment must be maintained and each piece of test equipment
labeled with the date of calibration and date of next calibra-
tion.
13. Handling, Storage, and Shipping
Ensure that equipment and material receiving, handling,
storage, and shipping follow manufacturer's recommendations
to prevent damage and deterioration. Verification and docu-
mentation that established procedures are followed is required.
14. Inspection, Test, and Operating Status
Label all equipment subject to required inspections and
tests so that the status of inspection and test is readily
apparent. Maintain an inventory of such inspections and oper-
ating status.
-36-
-------
15. Non-conforming Parts and Materials
Establish a system that will prevent the inadvertent use
of equipment or materials that do not conform to requirements.
16. Corrective Action
Establish a system to ensure that conditions adversely af-
fecting the quality of program operations are identified, cor-
rected, and commented on; and that preventive actions are taken
to preclude recurrence.
17. Quality Assurance Records
Maintain program records necessary to provide proof of
accomplishment of quality affecting activities of the measure-
ment program. Records include operating logs, test and in-
spection results, and personnel qualifications.
18. Aud it s
Conduct audits to evaluate the effectiveness of the mea-
surement program and quality assurance program to assure that
performance criteria are being met.
-37-
-------
4.0 ESTIMATED COSTS AND TIME REQUIREMENTS
Table 4-1 presents a listing of the conditions assumed for estimat-
ing the costs and time requirements of roof monitoring fugitive emis-
sions sampling programs using the methodology described in this document.
Four programs are listed, representing minimum and more typical levels
of effort for each of the survey and detailed programs defined in Sections
3.4 and 3.5, respectively. The combinations of conditions for each pro-
gram are generally representative of ideal cases for each level and may
not be encountered in actual practice. They do, however, illustrate the
range of effort and costs that may be expected in the application of the
roof monitor technique.
4.1 Manpower
Table 4-2 presents estimates of manpower requirements for each of
the sampling programs listed in Table 4-1. Man-hours for each of the
three general levels of Senior Engineer/Scientist, Engineer/Scientist,
and Junior Engineer/Scientist are estimated for the general task areas
outlined in this document and for additional separable tasks. Clerical
man-hours are estimated as a total for each program. Total man-hour
requirements are approximately 400 man-hours for minimum effort and
750 man-hours for typical effort in survey programs , and 1600 man-hours
for minimum effort and 2800 man-hours for typical effort in detailed
programs.
4.2 Other Direct Costs
Table 4-3 estimates for equipment purchases, rentals, calibration,
and repairs; on-site construction of towers and platforms; shipping and
-38-
-------
TABLE 4-1
CONDITIONS ASSUMED FOR ESTIMATING COSTS AND TIME
REQUIREMENTS FOR ROOF MONITOR FUGITIVE EMISSIONS
SAMPLING PROGRAMS
1
Parameter
Building
Openings
Emissions
Schedule
Air Flow At
Opening
Sampling
Locations
Sampling
Frequency
Estimated
Basic Accur-
acy
_.
Survey Programs
Minimum
Effort
1 Roof
(Small)
Constant
Steady
1
Traverse
Once
+ 400%
Typical
Effort
1 Roof
(Large)
Cyclic
Cyclic
4
Fixed
Once
+ 150%
Detailed Programs
Minimum
Effort
1 Roof
(Large)
Constant
Steady
4
Fixed
Typical
Effort
1 Roof,
1 Window
Cyclic,
Mixed
Cyclic
12/Opening
Fixed
(
4 Times 10 Times
+ 50%
+ 20%
Small *\* 50' long monitor
Large ^ 200' long monitor
-39-
-------
TABLE 4-2
ESTIMATED MANPOWER REQUIREMENTS FOR ROOF MONITOR
FUGITIVE EMISSIONS SAMPLING PROGRAMS
Estimates in Man-Hours
Task
Pretest Survey
Test Plan Preparation
Equipment Acquisition
Field Set-Up
Field Study
Sample Analysis
Data Analysis
Report Preparation
Totals
Engineer/Scientist Total
Clerical
Grand Total
Survey Programs
Minimum Effort
Senior
Engr/Sci
4
4
0
0
20
0
0
12
40
Engr/
Sci
8
12
0
16
40
20
20
32
148
368
40
408 j
Junior
Engr/
Tech
0
0
12
24
40
40
40
24
180
Typical Effort
Senior
Engr/Sci
4
4
0
8
40
0
8
24
88
Engr/
Sci
8
12
8
64
80
20
20
72
284
704
60
764
Junior
Engr/
Tech
0
0
20
30
80
80
80
40
332
Detailed Programs
Minimum Effort
Senior
Engr/Sci
8
8
0
8
120
4
16
44
204
i
1
1
i
I
Engr/
Sci
16
24
16
64
240
40
40
ino
540
1448
120
1568
Junior
Engr/
Tech
0
0
40
40
240
160
160
64
704
Typical Effort
Senior
Engr/Sci
12
12
0
24
240
16
32
80
416
Engr/
Sci
24
32
16
128
480
80
80
200
1040
2688
180
2868
Junior
Engr/
Tech
0
24
80
128
480
200
200
120
1232
I
-p-
o
-------
TABLE 4-3
ESTIMATED COSTS OTHER THAN MANPOWER FOR ROOF MONITOR
FUGITIVE EMISSIONS SAMPLING PROGRAMS
Cost Item
Equipment
Instrument Purchase
Calibration
Repairs
Platforms, Etc., Construction
Shipping
Vehicle Rentals
Communications
Miscellaneous Field Costs
TOTAL
Survey Programs
Minimum
Effort
$1000
50
100
200
200
200
50
50
$1850
Typical
Effort
$2000
100
150
500
400
500
100
100
$3850
Detailed Programs
Minimum
Effort
$3000
200
250
600
500
800
200
200
$5750
Typical
Effort
$12000
800
600
3000
800
1200
600
800
$19800
-------
on-site communications for each of the listed programs. Total costs are
approximately $1,900 for minimum effort and $3,900 for typical effort In
survey programs and $5,800 for minimum effort and $20,000 for typical
effort in detailed programs.
4.3 Elap sed-Time Requirements
Figure 4-1 presents elapsed-time estimates for each of the listed
programs broken down into the task areas indicated in the manpower es-
timates of Table 4-2. Total program durations are approximately 12
weeks for minimum effort and 19 weeks for typical effort in survey pro-
grams and 22 weeks for minimum effort and 33 weeks for typical effort
in detailed programs.
4.4 Cost Effectiveness
Figure 4-2 presents curves of the estimated cost effectiveness of
the roof monitor technique, drawn through points calculated for. the
four listed programs. Costs for each program were calculated at $30
per labor hour, $40 per man day subsistence for field work for the man-
power estimates of Table 4-2, plus the other direct costs estimated in
Table 4-3.
-42-
-------
Ul
Task
Pretest
survey
Test plan
preparation
Equipment
acquisition
Field
set-up
Field
study
Sample
analysis
Data
analysis
Report
preparation
Weeks
0 5 10 15 20 25 30 35
I i i i i i i i i i i i i I I i i i i i i i i i i i i i i i i i i i i i
7A
Simple survey program
Complex survey program
Simple detailed program
Complex detailed program
I 1 1 I I I 1 I I I 1 I 1 I I I I 1 I I I I I I I I I 1 I I I I I I 1
0 5 10 15 20 25 30 35
Weeks
Fig. 4-1. Elapsed-time estimates for roof monitor fugitive
emissions sampling programs.
-------
500
400
Survey program
300
2
o
o
10
a
CD
200-
100
Detailed program
150
Costs in thousands of dollars
Fig. 4-2. Cost-effectiveness of roof monitor fugitive
emissions sampling programs.
-44-
-------
APPENDIX A
APPLICATION OF THE ROOF MONITORING SAMPLING METHOD
TO AN ELECTRICAL ARC FURNACE INSTALLATION
-45-
-------
A.1.0 INTRODUCTION
This appendix presents an application of the roof monitor fugitive
emissions measurement system selection and design criteria to an electric
furnace steelmaking shop. The criteria for the selection of the method
and the design procedures for both survey and detailed sampling systems
as presented in Sections 3.4 and 3.5 of this document are discussed.
A.2.0 BACKGROUND INFORMATION
The following information relative to the operation of an electric
arc furnace was utilized to determine the sources and expected types of
fugitive emissions that might be encountered in the measurement programs.
Figure A-l describes the use of the electric furnace in steelmaking and
shows potential emission sources.
Sources of emissions at a typical electric arc furnace installation
could include:
o Charging of scrap to the hot furnace.
o Leaks of hooding and/or electrode holes during melting.
o Normal emissions from scrap melting.
o Charging of limestone and flux to the melt.
o Charging of alloying elements to the melt.
o Tapping and pouring hot metal to the ladle.
o Tapping and pouring slag into the slag ladle.
o Transfer of hot metal within the electric furnace shop.
Both gaseous (CO, I^S, S02, etc.) and particulate (iron, limestone,
carbon, etc.) emissions are given off by these emission sources and
would require quantification in any fugitive emission test program.
Emissions from each of these sources can be potentially controlled by
collection in a variety of hoods as illustrated in Figures A-2 and A-3,
and transfer through ductwork to a remotely located baghouse. A typ-
ical state-of-the-art ventilation system for a three furnace shop is
-46-
-------
-IT
-------
Canopy hood
exhaust duct
Charging
bucket
Fig. A-2 - Electric arc furnace-capture system for emissions.
-48-
-------
Roof monitor
II
Closed
roof
To fabric filter
or scrubber
Fig. A-3 Electric arc furnace-fugitive emission control.
-------
sketched in Figure A-4. These captured emissions can be readily iden-
tified and quantified utilizing duct-type sampling systems and methods.
Some portion of the emission from each source, however, escapes
collection by the ventilation system and is carried out of the building
via a roof monitor. These emissions are predominately those which occur
when the furnace roof is removed and therefore the directly connected
duct system must swing away either with or independent of the roof.
Charging emissions are of that type, and latest designs for electric
furnace shops use canopy hoods to reduce the released emissions which
escape into the general shop areas. These uncaptured charging emissions
are the most significant source of fugitive emissions from electric
furnace steelmaking. Tapping and pouring emissions as well as hot metal
transfer and transport emissions should not be ignored in the pre-test
survey. Visual observation of the emission sources can aid in evaluat-
ing their significance as fugitive sources.
The EPA estimates for uncontrolled emissions, as published in the
Office of Air Programs Publication AP-42, Compilation of Air Pollutant
Emission Factors, are 9.2 Ibs/ton metal charged without oxygen lance and
11 Ibs/ton metal with oxygen lancing. Assuming 90 percent of the emissions
are captured by control equipment, 0.9 to 1.1 Ibs/ton metal charged could
be transmitted to the atmosphere as fugitive emissions. The potential
fugitive emissions from the roof monitor of a four furnace steelmaking
operation with 100 ton capacity furnaces operating a three shift 24 hour
cycle with 4 melts/day/furnace would therefore be 1,440 - 1,760 Ibs/day
of particulates, plus significant amounts of carbon monoxide, sulfur
gases and other emissions.
-50-
-------
Building evacuation (BE) system, closed roof.
Fabric
filter
Canopy hood (CH), closed roof.
Building t \
monitor
Clean air
Exhaust g
Canopy hood (CH), open roof.
Fig. A-4 Electric arc furnace-charging/tapping fugitive emission
control.
-51-
-------
A,3.0 SURVEY MEASUREMENT SYSTEM
To determine the total plant contribution of particulates to the
atmosphere, measurement must be made of the emissions from the roof
monitor over a typical melt cycle from a single furnace. The results
of this test can be extrapolated to estimate the total emissions over
a 24 hour cycle of the entire electric furnace shop. Visual observations
can aid in selection of the roof monitor location to ensure representative-
ness of the particulate emissions collected.
A.4.0 SAMPLER LOCATION
A typical sampler location is shown in Figure A-5. By visual ob-
servation within and outside the electric furnace shop a location which
is within the "cloud" of fugitive emissions from a specific furnace can
also aid in answering the questions:
o Is the particulate emission rate (as measured by opacity) of that
furnace typical of the entire group of furnaces?
o Is the sampler location in the main flow path of the particulate
"cloud"?
o How does the variance of particulate emissions with time affect
the sampler location?
o How long a sampling period is required to obtain a representative
melt cycle's particulate emissions?
A fixed location high-volume type of particulate sampler similar
to that shown in Figure 3-1 would be used with a recording anemometer.
The average flow rate of air through the roof monitor opening may be
calculated as:
T
F = A/ dV
O T
-52-
-------
Fugitive emission measurement station
in roof monitor for Furnace //2
Electrical and
sample lines
" Electric Tfj
furnace
H2
Ground level
test station
Fig. A- 5. Typical survey program site to determine the fugitive emissions from an electric furnace
shop using a roof monitor technique.
Fugitive emission measurement stations
in roof monitor for each furnace
Electrical and
sample lines
test station
F ; •• . A-6 Typical detailed program site to determine the fugitive emissions from an electric furnace
shop using a roof monitor technique.
-53-
-------
where F = average air volume flow rate, cubic meters/minute
V = air velocity, meters/minute
A = roof monitor open area, square meters
T = test duration, minutes.
V, A and T are all directly measured values.
The particulate matter collected must be sufficient for measurement.
For a high volume sampler of 18 cubic feet per minute, a desired sample
weight would be 100 micrograms with a 60 minute minimum sampling time.
The required concentration of particulate in the existing air would,
therefore, be:
-)( = 10"1* Cgm)/0.5 (m3/min) x 60 (minutes)
X = 3.3 x 10~6 (gm/m3)
This would be readily achieved if the particulate plume had a 10% or
greater opacity.
Samples are therefore taken over a one hour or larger period and
the volume of air passes through the sampler determined. Multiplication
of the collected mass, by the average air flow through the roof monitor
divided by the air flow through the sampler divided by the time period
will give an estimate of the average emission rate in mass/time period
for the total electric furnace shop in that time period. Section 3.4.3
details the calculations and how to estimate the sampling time periods.
A.5.0 DETAILED MEASUREMENT SYSTEM
To determine the total electric furnace shop emissions with some
accuracy, measurements across the roof monitor of the emissions from all
-54-
-------
of the furnaces. Figure A-6 shows such a setup for the roof monitor of
a four furnace electric furnace shop. The samplers are similar to those
shown in Figure 3-1. In addition, if canopy hoods are used to capture
some charging and tapping emissions, they may be sampled by use of a set-
up such as shown in Figure A-7.
The roof monitor sampling system must be designed to identify and
quantify the electric arc furnace installation fugitive emissions by
accurately measuring the air flow rate through the roof monitor while
collecting samples of the emissions. The air flow rate will be deter-
mined by measuring the velocity of the air at a number of locations
across the vertical plane of the monitor opening using hot-wire or ro-
tating vane anemometers.
Sampling instruments for the measurement of the emissions will re-
quire at a minimum analyses for:
o Carbon monoxide
o Total suspended particulates
o Particulate size distribution
Preferable analysis methods are:
Carbon monoxide - non-dispersive infrared
Total particulates - Hi-Vol or Fiberglas filters plus
particulate charge count mass monitor
Particulate distribution - Andersen Samplers or equivalent
The specific operations whose individual contributions to the total
electric furnace shop fugitive emissions which can be differentiated
include:
-55-
-------
Particulate Measurement Devices
1KOR EPA CASCADE
IMPACTOR
Canopy hoo
exhaust duct
HC and CO line
Instruments
Fig. A-7. Illustration of test set-up for measuring fugitive
emissions from an electric arc furnace canopy hood.
-56-
V^^^ISjis^JgSSE;
-------
o Charging of the hot furnace
o Melting operations
o Tapping and pouring
The use of continuous monitoring instrumentation permits the correlation
of emission rate with the process operation to which it belongs. By
monitoring the emissions for extended periods of time, meaningful average
as well as instantaneous individual emission rates can thereby be obtained,
Calibration of continuous traces with known concentration standards, both
gaseous and particulate, is required to do this effectively.
A program designed to do this would include:
o Continuous monitoring on a 24 hour basis of particulates and
gases
o Collection of filterable particulate matter after each total
melt cycle in the furnace below each sampler
o Continuous recording of anemometer traces on a 24 hour basis
o Daily calibration of continuous monitors by comparison against
reference standards. Calibration gases would be used for gaseous
monitors and the high volume filter catch and that of the backup
filter in the particle charge count mass monitor for particulate
monitors.
Additional data on the emission rates of certain specific pollutants
could also be obtained by use of:
o Flame photometer continuous monitoring of sulfur gases
o EPA Method 5 trains with condensible trains and organic emission
absorber tubes to batch analyze for organics, especially carcinogens
o Membrane type filters for collection and batch chemical/morpholog-
ical analysis of specific inorganic particulate constituents such
as toxic metals and free silica.
-57-
-------
These should be at the discretion of the investigator, since they con-
tribute more than their proportionate share to the manpower time and
money investment in the fugitive emission sampling program.
A typical 4-6 week program would involve 24 hour tests on a four
furnace shop, thus potentially acquiring 24 total melt cycles/day or 480
to 720 sets of data. Because of potential problems of equipment break-
down in the hot and dirty environment in which they are used, as well
as the use of a 12 hour test shift (to allow use of a single well trained
test crew) gives us a potential of 120 to 180 actual data sets. Each
can be broken down into subsets of:
o Furnace tested
o Type and amount of charge used
o Type and amount of fluxes and/or additives used
o Portion of operating cycle involved (charge, melt, pour)
o Data reliability and completeness
Emission factors for each part of the electric furnace melt cycle
can be determined in addition to the average emission rate as determined
for the survey test program. We can break down the collected mass of
particulate and the flow rate as follows:
FI = flow rate for charge part of cycle
Mj = mass collected for charge part of cycle
Fg = flow rate for melt part of cycle
M£ = mass collected for melt part of cycle
FS = flow rate for tap/pour part of cycle
M3 = mass collected for tap/pour part of cycle
-58-
-------
The on-line mass monitors will be required for this. Calculations can
be done as in Section 3.4.3 of each individual mass rate of emission of
participates from parts of the cycle. Similar analysis can be done for
the gaseous emissions when continuous monitors are used. The result of
this program would be very detailed knowledge of the fugitive emissions
from a typical electric furnace melt cycle.
An additional tool to be used where better definition of exact
emission sources and rates is needed is the use of in-plant tracers to
simulate the sources. Gases such as SFs (sulfur hexaflouride) or (flo-
rescent dye particulates) can be released at specific points and at mea-
sured rates inside the electric furnace shop to simulate fugitive sources.
These tracers are collected at the roof monitor and from the collection
efficiency and concentration of collected tracer, a more accurate picture
of fugitive source locations and mass rates can be determined.
-59-
-------
TECHNICAL REPORT DATA
(Please read Insiructions on the reverse before completing)
1. REPORT NO.
EPA- 600/2 -76-089b
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Technical Manual for the Measurement of Fugitive
Emissions: Roof Monitor Sampling Method for
Industrial Fugitive Emissions
5. REPORT DATE
May 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
R. E. Kenson and P. T. Bartlett
9. PERFORMING OROANIZATION NAME AND ADDRESS
TRC--The Research Corporation of New England
125 Silas Deane Highway
Wethersfield, Connecticut 06109
10. PROGRAM ELEMENT NO.
1AB015; ROAP 21AUY-095
11. CONTRACT/GRANT NO.
68-02-2110
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 6/75-3/76
14. SPONSORING AGENCY CODE
EPA-ORD
^.SUPPLEMENTARY NOTES prOject officer for this technical manual is Robert M. Statnick,
Mail Drop 62, Ext 2557.
16. ABSTRACT
The technical manual presents fundamental considerations that are required
in using the Roof Monitor Sampling Method to measure fugitive emissions. Criteria
for selecting the most applicable measurement method and discussions of general
information gathering and planning activities are presented. Roof Monitor sampling
strategies and equipment are described, and sampling system design, sampling
techniques, and data reduction are discussed. Manpower requirements and time
estimates for typical applications of the method are presented for programs designed
for overall and specific emissions measurements. The application of the outlined
procedures to the measurement of fugitive emissions from an electric-arc furnace
steelmaking plant is presented as an appendix.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air Pollution Steel Plants
Industrial Processes
Measurement
Sampling
Estimating
Electric Arc Furnaces
Air Pollution Control
Stationary Sources
Fugitive Emissions
Roof Monitor Sampling
13B
13 H
14B
13A
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)'
Unclassified
21. NO. OF PAGES
64
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-60-
------- |